CN120691301A - A coal mine relay protection setting calculation method and system based on topology identification - Google Patents

Abstract

The present application provides a coal mine relay protection setting calculation method and system based on topology identification, which is applied to the field of relay protection technology. By monitoring the changes in the power grid topology in real time and intelligently allocating and executing calculation tasks based on these changes, the limitations of underground communication and computing resources are overcome, and the problem of how to efficiently and accurately perform relay protection setting calculations in the complex environment of underground coal mines is solved. The application has the advantages of adapting to the complex environment and distributed computing requirements of underground coal mines and improving the timeliness and accuracy of relay protection setting calculations.

Description

Coal mine relay protection setting calculation method and system based on topology identification

Technical Field

The application relates to the technical field of relay protection, in particular to a coal mine relay protection setting calculation method and system based on topology identification.

Background

The underground power supply system of the coal mine is a complex network composed of a ground substation, an underground central substation, an area substation, a mobile substation, a feed switch, cables and various electric equipment. In order to ensure the safety of underground operation, a perfect relay protection device must be configured. The core function of the relay protection device is that the action parameter, namely the setting parameter, is adopted. The setting parameters need to be accurately calculated and set according to the current connection mode, namely the topological structure, of the power grid. The acquisition of the topology of the power grid relies on the real-time monitoring of the switching states of the various switching devices in the system.

However, the underground coal mine environment is extremely severe, geological conditions are complex and changeable, so that the underground communication bandwidth is generally limited, and the communication link may be unstable and easily interfered. To overcome this challenge, a distributed computing architecture was introduced into a coal mine downhole power supply system that increased the system response speed, reduced excessive reliance on ground communication links, and improved overall system reliability. However, there are significant differences in the complexity of the grid structure and the frequency of topology changes in different areas downhole, and the downhole distributed computing units are limited by factors such as volume, power consumption, etc., which are generally relatively limited in computing power and storage capacity.

Therefore, under the distributed architecture, how to intelligently coordinate different computing units in the environments of limited communication, distributed computing resources and uneven capability, reasonably allocate topology identification and relay protection setting computing tasks, and ensure timeliness and consistency of computing results becomes a technical problem to be solved urgently.

Disclosure of Invention

In view of the shortcomings of the prior art, the application provides a coal mine relay protection setting calculation method and system based on topology identification, which have the advantages of adapting to the underground complex environment of a coal mine and distributed calculation requirements and improving the timeliness and accuracy of relay protection setting calculation.

In a first aspect, a method for calculating relay protection setting of a coal mine based on topology identification, the method includes the steps of:

s1, collecting state information of switching equipment in all areas under a well in real time, and comparing the current state with the state at the last moment to detect whether the switching state changes or not;

S2, when the switch state changes, generating a reduced change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp, and transmitting the reduced change information packet to a ground center computing unit or an adjacent underground area computing unit;

s3, receiving the reduced change information packet, and updating a power grid topology model according to the reduced change information packet;

S4, cooperatively distributing relay protection setting calculation tasks according to the area, the complexity of the calculation tasks, the current load and the communication state of each calculation unit, which are related to the updating of the topology model;

And S5, controlling each computing unit to execute the distributed relay protection setting computing task, synchronizing and checking the computing results among the related computing units, and transmitting the synchronized and checked relay protection setting parameters to the corresponding relay protection devices.

According to the coal mine relay protection setting calculation method based on topology identification, through monitoring the topology change of the power grid in real time and intelligently distributing and executing calculation tasks based on the change, the limitation of underground communication and calculation resources is overcome, the problem of how to efficiently and accurately perform relay protection setting calculation in a complex underground coal mine environment is solved, and the method has the advantages of adapting to the complex underground coal mine environment and distributed calculation requirements and improving timeliness and accuracy of relay protection setting calculation.

Further, step S2 includes:

s21, collecting state data of the switch equipment in a preset time window, and calculating state change frequency;

S22, comparing the state change frequency with a jitter threshold value, and judging whether the switch state change is jittered;

And S23, when the state change frequency is lower than the jitter threshold value, judging that the switch state change is true change, generating a reduced change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp, and transmitting the reduced change information packet to a ground center computing unit or an adjacent underground region computing unit.

According to the coal mine relay protection setting calculation method based on topology identification, jitter judgment on switch state change is added before a simplified change information packet is generated, invalid state change information generated by contact jitter is effectively filtered, generation and transmission of the invalid information packet are reduced, communication efficiency of a system and processing efficiency of a calculation unit are improved, and reliability of the system is enhanced.

Further, step S3 includes:

s31, receiving the simplified change information packet, requesting retransmission to a transmitting end if the receiving fails, and recording an error log if the retransmission times exceed a preset threshold;

S32, if the receiving is successful, analyzing the simplified change information packet, if the analyzing is failed, discarding the data packet and recording an error log, if the analyzing is successful, extracting the switch equipment identifier, the new state value and the change occurrence time stamp,

S33, positioning the corresponding switch equipment in the power grid topology model through a binary search algorithm according to the switch equipment identification, and if the corresponding equipment is not found, recording an error log and terminating the update;

S34, if the corresponding equipment is found, updating the state of the corresponding switching equipment to a new state value, and recording an update operation log at the same time;

and S35, carrying out time synchronization on the power grid topological model according to the time stamp of the change, and recording the time difference before and after synchronization so as to update the power grid topological model.

According to the coal mine relay protection setting calculation method based on topology identification, the specific implementation mode of the step S3 is defined in detail, so that accuracy of a power grid topology model is ensured, and a reliable basis is provided for subsequent relay protection setting calculation.

Further, step S35 includes:

S351, acquiring time stamps of the change occurrence corresponding to all the switching devices in the power grid topology model, and constructing a time stamp set;

s352, calculating to obtain theoretical transmission delay according to the physical distance between the switch equipment corresponding to each time stamp in the time stamp set and the receiving end and the preset signal transmission speed;

S353, estimating the actual transmission delay by adopting a Kalman filtering algorithm and taking the timestamp set and the theoretical transmission delay as inputs;

And S354, calculating the time difference of the power grid topology model before and after correction according to the estimated actual transmission delay so as to update the power grid topology model.

The coal mine relay protection setting calculation method based on topology identification provided by the application has the advantages that the specific implementation mode of performing time synchronization on the power grid topology model according to the change occurrence time stamp in step S35 is defined in detail, and the problem of how to perform time synchronization more accurately and ensure the time consistency of the power grid topology model under the conditions of severe underground coal mine communication environment and uncertain transmission delay is solved.

Further, step S354 includes:

S3541, calculating to obtain a corrected timestamp according to the estimated actual transmission delay and an original timestamp in the state information of the corresponding switch equipment in the power grid topology model by using a formula of corrected timestamp=original timestamp+actual transmission delay;

S3542, replacing an original timestamp in the state information of the corresponding switch equipment in the power grid topology model with the corrected timestamp;

S3543, calculating the absolute value of the difference value between the corrected timestamp and the original timestamp as the time difference of the switch equipment;

S3544, counting time differences of all switching devices in the power grid topology model, and calculating an average value of the time differences to serve as an average time difference of the power grid topology model before and after correction.

Further, step S4 includes:

s41, determining a topology model updating area, extracting an affected switching equipment identifier and a line segment identifier, and forming area information;

S42, determining the type of a calculation task of relay protection setting required according to the area information, and evaluating the complexity of the calculation task of the affected equipment and the complexity of the task of the circuit to form task description;

s43, monitoring the resource occupation of each computing unit, and obtaining the current load of each computing unit;

S44, evaluating the communication quality between adjacent computing units to obtain communication state information;

S45, integrating the area information, the task description, enabling the current load information and the communication state of each computing unit, and determining an allocation scheme of a relay protection setting computing task by adopting an improved ant colony algorithm with the aim of minimizing computing time.

Further, step S45 includes:

s451, constructing a communication quality evaluation matrix, a calculation unit load matrix and a task allocation fitness function;

s452, initializing ant colony, wherein each ant represents a task allocation scheme, and selecting the next computing unit by adopting a roulette selection strategy according to a communication quality evaluation matrix, a computing unit load matrix and a task allocation fitness function by the ants to construct the task allocation scheme;

S453, determining the concentration of the pheromone according to the fitness function value of the task allocation scheme, and determining the task allocation scheme of each calculation unit according to the task allocation scheme with the highest concentration of the pheromone.

Further, step S453 includes:

S4531, calculating the task load balance degree of each calculation unit in the task allocation scheme represented by each ant, wherein the calculation formula is that the load balance degree=1- (the load standard deviation of each calculation unit/the load average value of each calculation unit);

S4532, calculating a comprehensive evaluation index of the task allocation scheme according to the fitness function value and the load balance of the task allocation scheme, wherein the calculation formula is that the comprehensive evaluation index=alpha is the fitness function value+beta (1-load balance), wherein alpha and beta are weight coefficients, and alpha+beta=1;

S4533, updating a calculation formula of the pheromone according to the comprehensive evaluation index, wherein the calculation formula of the pheromone comprises the concentration= (1-volatilization coefficient) of the pheromone, the current concentration of the pheromone and the concentration increment of the pheromone;

S4534, selecting a task allocation scheme with the highest pheromone concentration by adopting a roulette selection strategy according to the updated pheromone concentration so as to determine the task allocation scheme of each calculation unit.

Further, step S5 includes:

s51, controlling each computing unit to read original setting parameters from the corresponding relay protection device, and determining the setting parameter category to be updated corresponding to the type of the distributed computing task;

S52, calculating new setting parameters corresponding to the setting parameter categories to be updated in parallel according to the power grid topology model;

s53, packaging a calculation result comprising a calculation unit identifier, a device identifier, a setting parameter category and a new setting parameter value into a result data packet according to a preset format;

S54, sending the result data packet to a corresponding check node, checking the result data packet by the check node according to a preset rule, requesting retransmission if the check fails, and recording if the check fails and exceeds a retransmission threshold;

and S55, if all the check nodes finish checking and pass, issuing the new setting parameters to the corresponding relay protection devices.

In a second aspect, a system for coal mine relay protection setting calculation based on topology identification is applied to the steps of the method described in any one of the above, and the system includes:

the acquisition module acquires the state information of the switch equipment in each underground area in real time, and compares the current state with the last time state to detect whether the switch state changes or not;

The transmission module is used for generating a reduced change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp when the switch state changes, and transmitting the reduced change information packet to a ground center calculation unit or an adjacent underground area calculation unit;

The updating module is used for receiving the simplified change information packet and updating the power grid topology model according to the simplified change information packet;

The task distribution module is used for cooperatively distributing relay protection setting calculation tasks according to the area, the calculation task complexity, the current load and the communication state of each calculation unit which are related to the topology model update;

And the execution issuing module is used for controlling each computing unit to execute the distributed relay protection setting computing task, synchronizing and checking the computing result among the related computing units, and issuing the synchronized and checked relay protection setting parameters to the corresponding relay protection devices.

The method and the system for calculating the relay protection setting of the coal mine based on the topology identification have the advantages that the topology change of the power grid is monitored in real time, and the calculation tasks are intelligently distributed and executed based on the change, so that the limitation of underground communication and calculation resources is overcome, the problem of how to efficiently and accurately perform the relay protection setting calculation in the underground complex environment of the coal mine is solved, the requirement of the underground complex environment and the distributed calculation of the coal mine is met, and the timeliness and the accuracy of the relay protection setting calculation are improved.

Drawings

Fig. 1 is a flowchart of a method for calculating relay protection setting of a coal mine based on topology identification.

Fig. 2 is a block diagram of a coal mine relay protection setting calculation system based on topology identification.

Fig. 3 is a framework diagram of a coal mine relay protection setting computing system based on topology identification.

The number of the device comprises a collection module 201, a conveying module 202, an updating module 203, a task allocation module 204 and an execution issuing module 205.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present application.

It should be noted that like reference numerals and letters refer to like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present application, the terms "first, second", etc. are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.

Referring to fig. 1, a method for calculating relay protection setting of a coal mine based on topology identification includes the steps:

s1, collecting state information of switching equipment in all areas under a well in real time, and comparing the current state with the state at the last moment to detect whether the switching state changes or not;

s2, when the switch state changes, generating a simplified change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp, and transmitting the simplified change information packet to a ground center computing unit or an adjacent underground area computing unit;

s3, receiving the simplified change information packet, and updating the power grid topology model according to the simplified change information packet;

S4, cooperatively distributing relay protection setting calculation tasks according to the area, the complexity of the calculation tasks, the current load and the communication state of each calculation unit, which are related to the updating of the topology model;

And S5, controlling each computing unit to execute the distributed relay protection setting computing task, synchronizing and checking the computing results among the related computing units, and transmitting the synchronized and checked relay protection setting parameters to the corresponding relay protection devices.

The state information of the switch equipment in each underground area can be acquired in real time through a sensor arranged near the switch equipment, and the sensor transmits the state signals to the computing unit through an industrial bus or a network. The switch equipment state information refers to current switching state data of each switch equipment in the underground power supply system.

A reduced change packet refers to a data structure containing only the necessary information of the switching device where the change of state occurs, such as a device identifier, a new state value (split or combined) and a time stamp of the change occurrence.

The cooperative allocation of relay protection setting calculation tasks refers to intelligently determining which calculation tasks are allocated to which one or more calculation units to execute according to various factors including the range involved in the topology change of the power grid, the complexity of the required calculation tasks, the current calculation resource occupation condition of each distributed calculation unit and the communication condition among the calculation units.

The method realizes intelligent collaborative allocation and execution of relay protection setting calculation tasks based on real-time topology change through a series of steps. Specifically:

The system continuously collects state information of underground switch equipment in real time, and compares the current state with the state at the previous moment to accurately detect whether the topology of the power grid changes. Upon detecting a change in the switch state, the system immediately generates a reduced change packet containing only the identity of the changing switch device, the new state value, and the time stamp of the change occurrence, thereby greatly reducing the amount of data that needs to be transmitted.

This reduced packet is flexibly delivered to a surface-centric computing unit or nearby neighboring downhole region computing units, distributing communication pressure using a distributed architecture. And the computing unit receiving the simplified information packet rapidly updates the power grid topology model maintained by the computing unit according to the information in the simplified information packet, so as to ensure that the model is consistent with the actual power grid state.

Based on the updated real-time topology model, the system further comprehensively evaluates the area range related to the current topology change, the complexity of relay protection setting calculation tasks caused by the area range, the current calculation resource load condition of each distributed calculation unit and the communication state among the calculation units.

According to the multidimensional information, the system intelligently distributes relay protection setting calculation tasks by adopting a cooperative strategy, and distributes different calculation subtasks to the calculation units which are most suitable for execution so as to optimize the overall calculation efficiency. And the calculation units of the distributed tasks execute the respective calculation tasks in parallel or sequentially to obtain new relay protection setting parameters. In order to ensure the accuracy of the calculation results, the calculation results are synchronized and checked between the relevant calculation units. Only the relay protection setting parameters which are confirmed to be correct through synchronization and verification can be finally issued to the corresponding relay protection devices, so that the devices can act correctly according to the latest power grid topological structure, and the safety of a power supply system is ensured.

Through the scheme, the method and the device effectively solve the problems of relay protection setting calculation in the environments of limited underground coal mine communication, scattered calculation resources and uneven capacity. The real-time acquisition and simplified information packet transmission mechanism obviously reduces the dependence on limited communication bandwidth and improves the perception speed of topology change. The utilization efficiency of underground decentralized computing resources is optimized and the timeliness of overall computation is improved based on a real-time topology model and by comprehensively considering a cooperative task allocation strategy of multidimensional factors. And the accuracy and consistency of relay protection setting calculation results are guaranteed by a distributed execution and synchronous verification mechanism. Finally, the setting parameters can be timely and accurately issued to the relay protection device, and the safe operation level of the underground power supply system is improved.

Further, step S2 includes:

S21, collecting state data of the switching equipment in a preset time window, and calculating state change frequency;

s22, comparing the state change frequency with a jitter threshold value, and judging whether the switch state change is jittered;

S23, when the state change frequency is lower than the jitter threshold value, judging that the switch state change is true change, generating a reduced change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp, and transmitting the reduced change information packet to a ground center calculation unit or an adjacent underground area calculation unit.

The preset time window is a time period for observing and analyzing the state change of the switching device, and the length of the preset time window can be set according to the actual application scene, the type of the switching device and the expected jitter characteristic.

The state data refers to a state information sequence of the switching device collected in a preset time window, and generally includes on-off state values of the switch at different moments and corresponding collection time stamps.

The state change frequency refers to the quantized representation of the number of changes of the state of the switching device or the state change rate in a preset time window, and can be calculated by dividing the number of times the statistical state value is flipped by the length of the time window.

The jitter threshold is a preset upper frequency limit for distinguishing the actual state change from the contact jitter, and when the calculated state change frequency exceeds the threshold, the contact jitter is considered to occur, and when the calculated state change frequency is lower than or equal to the threshold, the calculated state change frequency is considered to be an actual and stable state change, and the threshold can be empirically set or determined through experiments according to the electrical characteristics of the switch equipment, the interference level of the field environment and the requirement on the response speed of the system.

For example, in one particular embodiment, the preset time window may be set to 50 milliseconds. When a change in the state of a particular switching device is detected, the system will begin to monitor the device for state data for the next 50 milliseconds. It is assumed that the state of the switch is rapidly switched 5 times between on and off within the 50 milliseconds. The system calculates the frequency of the state change to be 5 times/50 milliseconds. If the preset jitter threshold is set to 3 times/50 milliseconds, the system judges that the state change is the contact jitter at the time because the calculated frequency 5 is higher than the threshold 3, and does not generate a simplified change information packet and does not carry out conveying. If the other switching device changes state only once steadily (e.g., changes from on to off and remains in the on state within the window) within 50 milliseconds after detecting the state change, the calculated state change frequency is 1/50 milliseconds. Because the frequency 1 is lower than the threshold 3, the system judges that the change is true, then generates a simplified change information packet containing the switch equipment identifier, the new state value (combination) and the change occurrence time stamp, and sends the information packet out.

Further, step S3 includes:

s31, receiving the simplified change information packet, requesting retransmission to a transmitting end if the receiving fails, and recording an error log if the retransmission times exceed a preset threshold;

S32, if the receiving is successful, analyzing the simplified change information packet, if the analyzing is failed, discarding the data packet and recording an error log, if the analyzing is successful, extracting the switch equipment identifier, the new state value and the change occurrence time stamp,

S33, positioning corresponding switch equipment in a power grid topology model through a binary search algorithm according to the switch equipment identification, and if the corresponding equipment is not found, recording an error log and terminating the update;

s34, if the corresponding equipment is found, updating the state of the corresponding switching equipment into a new state value, and simultaneously recording an update operation log;

and S35, carrying out time synchronization on the power grid topological model according to the time stamp of the change occurrence, and recording the time difference before and after synchronization so as to update the power grid topological model.

Specifically, in the process of receiving the reduced change information packet, a mechanism for requesting retransmission is introduced, and the mechanism can actively request the transmitting end to retransmit when the initial data packet receiving fails, so that the success rate and the reliability of data transmission are improved. Meanwhile, a preset threshold value of retransmission times is set, and an error log is recorded when the preset threshold value is exceeded, so that the stability of a communication link is monitored, and a record is left when a data packet finally fails to be received, and the subsequent problem investigation and analysis are facilitated.

In addition, when the received simplified change information packet is analyzed, the verification of the analysis result is increased, if the analysis process is wrong, the data packet is discarded and an error log is recorded, so that only the data packet with correct format and valid content can be used for updating the topology model, and the error update caused by data damage or format inconsistency is avoided.

When the corresponding switch equipment is positioned in the power grid topology model, a binary search algorithm is adopted, the algorithm is an efficient search method, is suitable for ordered data sets, can rapidly position target equipment by continuously narrowing the search range, and particularly can remarkably improve the search efficiency and reduce the time consumption of updating operation when the power grid is large in scale and the equipment number is large. After the equipment state is updated, the update operation log is recorded, and the log can contain information such as update time, equipment identification, old state, new state, timestamp in a data packet and the like, so that a detailed history record is provided for subsequent topology change tracing, state rollback or fault analysis.

Finally, the time synchronization is carried out on the power grid topological model according to the time stamp of the change, the time stamp reflects the actual time point of the equipment state change, and by utilizing the time stamp, the state of the topological model can be calibrated in time or the time of different equipment states can be aligned at the receiving end, so that the power grid topological model is ensured to reflect the state under a relatively consistent time slice, which is very important for the application (such as relay protection setting calculation) needing to calculate based on accurate and synchronous topology. Recording the time difference before and after synchronization is helpful to evaluate the time synchronization accuracy of the system and the delay condition of data transmission.

Further, step S35 includes:

s351, obtaining change occurrence time stamps corresponding to all switching devices in a power grid topology model, and constructing a time stamp set;

S352, calculating to obtain theoretical transmission delay according to the physical distance between the switch equipment corresponding to each time stamp in the time stamp set and the receiving end and the preset signal transmission speed;

S353, estimating the actual transmission delay by adopting a Kalman filtering algorithm and taking a time stamp set and theoretical transmission delay as inputs;

S354, calculating the time difference of the power grid topology model before and after correction according to the estimated actual transmission delay so as to update the power grid topology model.

The time stamp set refers to a data set formed by collecting time stamps of changes associated with the states of all current switching equipment in the power grid topology model.

The theoretical transmission delay refers to the time required for information transmission in an ideal state calculated based on the physical distance between the information transmitting end (switching device) and the receiving end (calculation unit) and the propagation speed of the signal in the transmission medium.

The actual transmission delay refers to the actual time that it takes for the information to be sent from the switching device to its state change information to the computing unit.

The time difference between the grid topology model before and after correction refers to the difference between the device state timestamp and the original timestamp of the whole or part of the device state timestamp of the grid topology model after correction.

The time difference between the grid topology model before and after correction refers to the difference between the device state timestamp and the original timestamp of the whole or part of the device state timestamp of the grid topology model after correction.

In a specific embodiment, all the switching devices currently included in the power grid topology model may be traversed first, change occurrence time stamps associated with each device state may be extracted, and the time stamps may be collected to form a time stamp set. Then, for each time stamp in the time stamp set, the position information of the switching device corresponding to the time stamp in the preset power grid physical layout diagram and the position information of the computing unit receiving the information can be queried, and the physical distance between the two can be calculated. Meanwhile, a preset signal transmission speed can be set according to the type of the underground communication cable or the propagation characteristics of the wireless signal. The theoretical transmission delay of each piece of switch equipment information transmitted to the receiving end can be calculated by dividing the physical distance by the preset signal transmission speed. The set of constructed timestamps (which contains the original timestamp and the time of receipt) and the calculated theoretical transmission delay are then fed into a kalman filter as inputs.

The kalman filter may build a state space model where the state variables may include the actual transmission delay and its rate of change and the measured variables may be the apparent delay obtained by subtracting the original timestamp from the time of receipt. The filter integrates theoretical delay information and apparent delay measurement values through the prediction and updating processes, and more accurate actual transmission delay is estimated. Finally, according to the actual transmission delay estimated by the Kalman filter, the difference between the corrected time stamp and the original time stamp of each switch equipment state can be calculated, and the difference is recorded for subsequent analysis or used as part of information of model update. For example, the corrected timestamp may be substituted for the original timestamp, or the time difference may be stored as an additional attribute of the device state.

Specifically, the mathematical model of the kalman filter includes:

① State space model:

state vector: where τ _k is the actual transmission delay at time k; Is the rate of change of the actual transmission delay at time k.

The state transition equation x _k+1＝Ax_k+w_k, where a is the system matrix describing the evolution of the state vector x _k with the sampling period,T is the sampling period and w _k is the process noise.

② Measurement model:

measurement vector z _k＝τ_k+v_k, where v _k is measurement noise.

③ Kalman filtering algorithm:

Predictive state vector: Wherein, the Indicating the predicted transmission delay and its rate of change.

The prediction error covariance is P _k+1|k＝AP_k+1|kA^T +Q, wherein P _k+1|k describes uncertainty of the predicted transmission delay and the change rate thereof, and Q is a process noise covariance matrix and describes uncertainty of the actual transmission delay and the change rate thereof.

Kalman gain, K _k+1＝P_k+1|kH^T(HP_k+1|kH^T+R)^-1, a measurement matrix H representing the mapping of the state vector to the measurement of the transmission delay, a measurement noise variance R describing the uncertainty of the measurement, K _k+1 being the Kalman gain for adjusting the predicted state vector.

Updating the state: Updating state vectors Representing the transmission delay and its rate of change corrected in combination with the actual measurement value z _k+1, the actual measurement value z _k+1 representing the value of the transmission delay measured by the sensor.Representing the propagation delay estimate calculated based on the predicted state vector.

The updated error covariance is P _k+1|k+1＝(I-K_k+1H)P_k+1|k, where P _k+1|k+1 represents the uncertainty of the updated transmission delay and its rate of change, and I is the identity matrix.

The above parameters together constitute a specific implementation of the kalman filter algorithm in the actual propagation delay estimation.

Further, step S354 includes:

S3541, calculating to obtain a corrected timestamp according to the estimated actual transmission delay and an original timestamp in the state information of the corresponding switch equipment in the power grid topology model by using a formula of corrected timestamp=original timestamp+actual transmission delay;

S3542, replacing an original timestamp in the state information of the corresponding switch equipment in the power grid topology model with a corrected timestamp;

S3543, calculating the absolute value of the difference value between the corrected timestamp and the original timestamp as the time difference of the switch equipment;

S3544, counting time differences of all switching devices in the power grid topology model, and calculating an average value of the time differences to serve as an average time difference of the power grid topology model before and after correction.

The scheme describes in detail how the estimated actual transmission delay is used to calculate the time difference between the grid topology model before and after correction and update the model. Specifically:

In step S3541, a modified timestamp is calculated using an explicit addition formula. This step uses the estimated delay to correct the original timestamp to more accurately reflect the actual time of occurrence of the state change, providing the underlying data for subsequent time synchronization.

The replacing operation in step S3542 directly updates the time information in the topology model, so as to ensure that the time stamp of each device state in the model is calibrated, and improve the time consistency of the topology model.

Step S3543 quantifies the amount of time stamp correction for a single switching device reflecting how much time synchronization affects the device' S time stamp.

Step S3544 provides an overall index to measure the improvement degree of the time consistency of the whole power grid topology model in the time synchronization operation by calculating the average value of the time differences of all the devices, so that the system can evaluate the time synchronization effect and provide basis for subsequent system optimization or performance monitoring.

The scheme is combined with the method, and further provides specific operations of accurate time stamp correction and quantitative time synchronization effect evaluation by using estimated delay under the basic framework of topology acquisition, transportation, topology updating, distribution calculation and issuing execution of the method, and on the basis of refined topology updating, time synchronization and delay estimation, so that the time information of the whole power grid topology model is more accurate and reliable, and a reliable data basis is provided for subsequent relay protection setting calculation.

Further, step S4 includes:

s41, determining a topology model updating area, extracting an affected switching equipment identifier and a line segment identifier, and forming area information;

S42, determining the type of a calculation task of relay protection setting required according to the region information, and evaluating the complexity of the calculation task of the affected equipment and the complexity of the task of the circuit to form task description;

s43, monitoring the resource occupation of each computing unit, and obtaining the current load of each computing unit;

S44, evaluating the communication quality between adjacent computing units to obtain communication state information;

S45, integrating the area information and the task description, enabling each computing unit to carry out current load information and communication state, and determining an allocation scheme of the relay protection setting computing task by adopting an improved ant colony algorithm and aiming at minimizing computing time.

The regional information refers to defining a power grid range needing relay protection setting calculation according to the change of a power grid topology model.

Task descriptions refer to the specific types of relay protection setting calculation tasks that need to be performed and the calculation resources or time evaluations required to complete these tasks.

The current load of the computing units refers to the resource usage condition of each computing unit in the distributed computing system at a certain moment, such as CPU utilization rate, memory occupancy rate and the like.

The communication status information refers to the quality of communication links between adjacent computing units in the distributed computing system, such as communication delay, bandwidth, packet loss rate, etc., which may be represented by a communication quality assessment matrix or a set of link indicators.

The improved ant colony algorithm is an optimization algorithm based on group intelligence, searches for an optimal solution by simulating the behavior of ants for finding paths, and can be realized by adopting a pheromone updating rule, a heuristic function and a selection strategy which are customized for task allocation problems.

According to the relay protection setting calculation task collaborative distribution method, firstly, the affected area is accurately identified according to the update of the power grid topology model, so that the specific task range needing setting calculation is determined. And then, aiming at the affected equipment and circuits, clearly setting the type of the relay protection setting calculation task to be executed, and evaluating the calculation complexity of each task to form detailed task description.

Meanwhile, the system acquires the resource occupation condition of each computing unit in the distributed computing environment in real time, knows the current load state of each computing unit, evaluates the communication quality among the computing units and acquires the communication state information. After the multi-dimensional information such as the area information, the task description, the load of the computing unit, the communication state and the like is acquired, the information is taken as input, and an improved ant colony algorithm is adopted to carry out task allocation decision

. The improved ant colony algorithm can simulate the actions of ants releasing pheromones on paths and selecting paths according to the concentration of the pheromones, and searches an optimal allocation scheme capable of minimizing the total calculation time of all tasks under various constraint conditions such as task complexity, calculation unit processing capacity, current load, communication overhead and the like through an iterative optimization process.

The distribution strategy comprehensively considers various factors and adopts an intelligent optimization algorithm, so that challenges of computing resource dispersion, capability non-uniformity and limited communication in a coal mine underground distributed environment can be effectively met, and problems of computing unit overload, task delay or communication bottleneck possibly caused by a simple distribution strategy are avoided. By combining the power grid topology model updating step, task allocation is ensured to be carried out based on the latest power grid state, so that timeliness and accuracy of relay protection setting calculation are improved, and a reliable basis is provided for subsequent parameter issuing.

Further, step S45 includes:

s451, constructing a communication quality evaluation matrix, a calculation unit load matrix and a task allocation fitness function;

s452, initializing ant colony, wherein each ant represents a task allocation scheme, and selecting the next computing unit by adopting a roulette selection strategy according to a communication quality evaluation matrix, a computing unit load matrix and a task allocation fitness function by the ants to construct the task allocation scheme;

S453, the pheromone concentration is determined according to the fitness function value of the task allocation scheme, and the task allocation scheme of each calculation unit is determined according to the task allocation scheme with the highest pheromone concentration.

The communication quality evaluation matrix refers to a matrix for quantifying the condition of communication links between different computing units, and can be implemented by adopting a numerical matrix calculated based on indexes such as packet loss rate, delay, bandwidth and the like. Specifically, communication quality evaluation matrixWhere C _ij is the communication quality between the computing unit i and the computing unit j.Where Delay _ij represents the communication Delay between computing unit i and computing unit j, loss _ij represents the packet Loss rate between computing unit i and computing unit j, and Bandwidth _ij represents the Bandwidth between unit i and computing unit j. And the alpha, beta and gamma weight coefficients are used for adjusting the relative importance of different communication indexes. The weights can be adjusted by technicians according to actual application scenes so as to reflect the influence degree of different communication indexes on the communication quality.

The computing unit load matrix refers to a matrix reflecting the current computing resource occupation condition of each computing unit, and can be realized by adopting a numerical matrix calculated based on indexes such as CPU utilization rate, memory occupancy rate, task queue length and the like. Specifically, a unit load matrix is calculatedWhere L _i is the current load of compute unit i, L _i＝w₁·CPU_i+w₂·Memory_i+w₃·Queue_i, where CPU _i is the current CPU utilization of compute unit i, memory _i is the current Memory occupancy of compute unit i, and Queue _i is the current task Queue length of compute unit i. w ₁、w₂、w₃ is a weight coefficient, which can be set by the skilled person according to the actual situation.

The task allocation fitness function refers to a function for evaluating the quality of a specific task allocation scheme, and can be implemented by adopting a weighted combination of factors such as calculation time, communication overhead, load balancing and the like. Specifically, f (B) =λ ₁·TotalTime(B)+λ₂·CommCost(B)+λ₃ · LoadBalance (B), where B is a task allocation scheme.

Total Time (B) is the total computation time under task allocation scheme B:

Where T _ib is the execution time of task B on computing unit i, B _ib is whether task B is assigned to computing unit i,1 indicates assignment, and 0 indicates non-assignment. n is the total number of computing units and m is the total number of tasks.

CommCost (B) is the total communication overhead under task allocation scheme B:

Where D _jb denotes the data amount of task B, C _ij denotes the communication quality between computing unit i and computing unit j, B _ib denotes whether task B is allocated to computing unit i, and B _jb denotes whether task B is allocated to computing unit j.

LoadBalance (B) is a load balancing metric under task allocation scheme B:

where L _i is the current load of the computing unit i; an average load for all computing units.

Lambda ₁、λ₂、λ₃ is a weight coefficient that can be set by the skilled person according to the actual situation.

The complex task allocation problem is converted into a computable model by constructing a communication quality evaluation matrix, a calculation unit load matrix and a task allocation fitness function. By utilizing the global searching capability of the ant colony algorithm, a plurality of possible task allocation schemes can be explored. Probability is introduced through a roulette selection strategy, and the situation that the local evaluation value is higher is avoided. Through the accumulation and volatilization mechanism of pheromones, the algorithm can learn and strengthen the distribution scheme with higher evaluation value, and gradually converges to be close to the optimal solution. Finally, a task allocation scheme can be generated that effectively balances the computational load and communication overhead, thereby minimizing the total time of relay protection setting calculations. The efficiency and reliability of relay protection setting calculation in the coal mine underground environment with limited communication, distributed calculation resources and uneven capacity are improved.

Further, step S453 includes:

S4531, calculating the task load balance degree of each calculation unit in the task allocation scheme represented by each ant, wherein the calculation formula is that the load balance degree=1- (the load standard deviation of each calculation unit/the load average value of each calculation unit);

S4532, calculating a comprehensive evaluation index of the task allocation scheme according to the fitness function value and the load balance of the task allocation scheme, wherein the calculation formula is that the comprehensive evaluation index=alpha is the fitness function value+beta (1-load balance), wherein alpha and beta are weight coefficients, and alpha+beta=1;

S4533, updating a calculation formula of the pheromone according to the comprehensive evaluation index, wherein the calculation formula of the pheromone comprises the concentration= (1-volatilization coefficient) of the pheromone, the current concentration of the pheromone and the concentration increment of the pheromone;

S4534, selecting a task allocation scheme with the highest pheromone concentration by adopting a roulette selection strategy according to the updated pheromone concentration so as to determine the task allocation scheme of each calculation unit.

The task load balance degree refers to an index for measuring the uniformity of the distribution of tasks among the distributed computing units.

The fitness function value refers to a single objective function value for evaluating the merits of a task allocation scheme, and can be expressed by calculating an index of total time, total energy consumption, or total cost for completing all tasks.

The comprehensive evaluation index is a composite index which combines a plurality of evaluation dimensions (such as fitness function values and load balancing degrees) to perform overall evaluation on the task allocation scheme.

The pheromone concentration refers to a numerical value which indicates the probability of a certain path or scheme being selected in an ant colony algorithm, and the numerical value can be updated in a numerical value accumulation and volatilization mode. The volatility coefficient refers to a parameter that simulates the dissipation of pheromones over time in an ant colony algorithm, and can be expressed by a constant between 0 and 1. The pheromone concentration increment refers to the amount of the pheromone added on the corresponding path after one scheme is evaluated in the ant colony algorithm, and can be determined according to the comprehensive evaluation index or other performance indexes of the scheme. Roulette selection strategy refers to a probability-based selection method in which the probability of each option being selected is proportional to its corresponding weight (e.g., pheromone concentration), which can be implemented by constructing a probability distribution and randomly sampling.

The above technical solution is adopted in the present application, because in the distributed computing environment, only pursuing a single optimization objective (such as minimizing time) may cause overload of some computing units, while other computing units are idle, which not only affects the overall efficiency of the system, but also may cause instability of the system.

By introducing consideration of the load balance degree of each calculation unit when evaluating the task allocation scheme and combining the load balance degree with the original fitness function value to form a comprehensive evaluation index, the advantages and disadvantages of one scheme can be comprehensively measured. Specifically, calculating the load balance quantifies the uniformity of resource allocation, and the comprehensive evaluation index balances the optimization target and the resource utilization efficiency. Updating the pheromone based on this comprehensive evaluation index allows those schemes that perform better in terms of both optimization objectives and load balancing to leave stronger signals on the pheromone track, and thus be more easily selected in subsequent iterations.

This mechanism enables the ant colony algorithm to search and converge to a task allocation scheme that not only meets the original optimization objective, but also achieves more reasonable load distribution among the computing units. Compared with a scheme which only evaluates and selects according to a single fitness function value, the method can effectively avoid the problem of local overload and improve the overall performance and stability of the distributed computing system. The mode of integrating load balancing as an important evaluation dimension into an ant colony algorithm pheromone updating and selecting mechanism ensures that the determining process of a task allocation scheme is more intelligent and robust, and can be better suitable for complex and changeable computing environments in coal mine.

Further, step S5 includes:

s51, controlling each computing unit to read original setting parameters from the corresponding relay protection device, and determining the setting parameter category to be updated corresponding to the type of the distributed computing task;

S52, calculating new setting parameters corresponding to the setting parameter categories to be updated in parallel according to the power grid topology model;

s53, packaging a calculation result comprising a calculation unit identifier, a device identifier, a setting parameter category and a new setting parameter value into a result data packet according to a preset format;

S54, sending the result data packet to a corresponding check node, checking the result data packet by the check node according to a preset rule, requesting retransmission if the check fails, and recording if the retransmission exceeds a retransmission threshold;

and S55, if all the check nodes finish checking and pass, issuing new setting parameters to the corresponding relay protection devices.

The scheme provides a set of detailed execution and verification flow aiming at the problems of reliably executing relay protection setting calculation tasks, verifying calculation results and safely and underground setting parameters in a distributed and communication-limited underground coal mine environment.

Step S1 this ensures that the subsequent calculations are based on the current actual state of the device and the required update targets, improving the pertinence and effectiveness of the calculations.

Then, in step S52, the new parameters are ensured to be suitable for the current network structure by calculation according to the power grid topology model, which is the basis of correct setting, and the capability of the distributed calculation unit is utilized in parallel calculation, so that the calculation efficiency is improved, and the time required for setting is shortened.

The standardized packaging mode in step S53 facilitates transmission, synchronization and subsequent processing of the result data, and improves standardization and efficiency of data management.

Further, step S54 introduces check nodes and performs check according to a preset rule, so that errors possibly occurring in the calculation process or anomalies in the data transmission can be effectively found, which is a key link for ensuring the correctness of the calculation result.

When the verification fails, the mechanism for requesting retransmission provides fault tolerance, improves the reliability of data transmission, and records error logs when the retransmission threshold value is exceeded, thereby facilitating the fault diagnosis and processing of the system.

Finally, in step S55, it is provided that the calculated new setting parameters are issued to the corresponding relay protection device only if all relevant check nodes have completed the check and the result has passed.

Through the strict winding piece as a final safety guarantee, the protection device ensures that only correct parameters verified by multiple parties can be applied to the protection device, greatly reduces the risk of protection misoperation or refusal operation caused by the incorrect parameters, and improves the safety and reliability of the whole power supply system.

Referring to fig. 2 and 3, a system for setting and calculating relay protection of a coal mine based on topology identification is applied to the steps of any one of the above methods, and the system includes:

The acquisition module 201 is used for acquiring state information of the switch equipment in each underground area in real time, and comparing the current state with the last time state to detect whether the switch state changes or not;

The transmission module 202 is used for generating a reduced change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp when the switch state changes, and transmitting the reduced change information packet to a ground center calculation unit or an adjacent underground area calculation unit;

the updating module 203 is used for receiving the simplified change information packet and updating the power grid topology model according to the simplified change information packet;

The task allocation module 204 is used for cooperatively allocating relay protection setting calculation tasks according to the area related to the topology model update, the complexity of the calculation tasks, the current load and the communication state of each calculation unit;

And the execution issuing module 205 is used for controlling each computing unit to execute the distributed relay protection setting computing task, synchronizing and checking the computing result among the related computing units, and issuing the synchronized and checked relay protection setting parameters to the corresponding relay protection devices.

The acquisition module 201 refers to a unit responsible for acquiring operation state information of a field device, and may be implemented by using an Intelligent Electronic Device (IED), a Remote Terminal Unit (RTU), or various sensors.

The transport module 202 is a unit responsible for formatting and transmitting the collected state change information to the processing unit, and may be implemented by using a communication interface, a network protocol stack, or a data transmission service.

The update module 203 is a unit responsible for maintaining the latest state of the power grid topology model according to the received state change information, and may be implemented by using a database management system, an in-memory data structure or a topology analysis algorithm.

The task allocation 204 module refers to a unit responsible for determining specific computing tasks assumed by each computing unit according to the current system state and computing requirements, and may be implemented by using a scheduling algorithm, a resource manager, or a distributed coordination framework.

The execution issuing module 205 refers to a unit responsible for triggering a computing unit to perform tasks and apply the results of the computation to the field devices, which may be implemented using a computing task manager, a result synchronization mechanism, or a parameter issuing interface.

The system of the application embodies each functional link in the relay protection setting calculation method into an executable system component through modularized design. The acquisition module 201 continuously monitors the status of the downhole switching device, and triggers the delivery module upon detection of a status change. The transmission module 202 generates a simplified data packet according to the change information, intelligently selects a transmission path, and can send the data packet to a ground center or a nearby underground computing unit, which is helpful for reducing data transmission burden and improving efficiency in a communication-limited underground environment. After receiving the information packet, the updating module 203 is responsible for updating the power grid topology model maintained in the system in time, so as to ensure that the model is consistent with the actual power grid state. Based on the updated topology model and the evaluation of the complexity of the calculation task, the load of each calculation unit and the communication state, the task allocation module 204 cooperatively schedules the underground scattered calculation resources, and reasonably allocates the setting calculation task to different calculation units for execution. The execution issuing module 205 controls the computing unit to execute tasks in parallel, ensures the accuracy and consistency of the computing result through a synchronization and verification mechanism, and finally reliably issues the calculated new setting parameters to the corresponding relay protection device. By the cooperative working mode, the system provides a stable and efficient operation platform for complex method steps, so that the method can be effectively implemented in a severe underground environment.

In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.

The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and variations will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims (10)

1. The coal mine relay protection setting calculation method based on topology identification is characterized by comprising the following steps:

s1, collecting state information of switching equipment in all areas under a well in real time, and comparing the current state with the state at the last moment to detect whether the switching state changes or not;

S2, when the switch state changes, generating a reduced change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp, and transmitting the reduced change information packet to a ground center computing unit or an adjacent underground area computing unit;

s3, receiving the reduced change information packet, and updating a power grid topology model according to the reduced change information packet;

S4, cooperatively distributing relay protection setting calculation tasks according to the area, the complexity of the calculation tasks, the current load and the communication state of each calculation unit, which are related to the updating of the topology model;

And S5, controlling each computing unit to execute the distributed relay protection setting computing task, synchronizing and checking the computing results among the related computing units, and transmitting the synchronized and checked relay protection setting parameters to the corresponding relay protection devices.

2. The method for calculating relay protection setting of a coal mine based on topology identification according to claim 1, wherein step S2 comprises:

s21, collecting state data of the switch equipment in a preset time window, and calculating state change frequency;

S22, comparing the state change frequency with a jitter threshold value, and judging whether the switch state change is jittered;

And S23, when the state change frequency is lower than the jitter threshold value, judging that the switch state change is true change, generating a reduced change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp, and transmitting the reduced change information packet to a ground center computing unit or an adjacent underground region computing unit.

3. The method for calculating relay protection setting of a coal mine based on topology identification according to claim 1, wherein the step S3 comprises:

s31, receiving the simplified change information packet, requesting retransmission to a transmitting end if the receiving fails, and recording an error log if the retransmission times exceed a preset threshold;

S32, if the receiving is successful, analyzing the simplified change information packet, if the analyzing is failed, discarding the data packet and recording an error log, if the analyzing is successful, extracting the switch equipment identifier, the new state value and the change occurrence time stamp,

S33, positioning the corresponding switch equipment in the power grid topology model through a binary search algorithm according to the switch equipment identification, and if the corresponding equipment is not found, recording an error log and terminating the update;

S34, if the corresponding equipment is found, updating the state of the corresponding switching equipment to a new state value, and recording an update operation log at the same time;

and S35, carrying out time synchronization on the power grid topological model according to the time stamp of the change, and recording the time difference before and after synchronization so as to update the power grid topological model.

4. The method for calculating relay protection settings of a coal mine based on topology identification as recited in claim 3, wherein the step S35 includes:

S351, acquiring time stamps of the change occurrence corresponding to all the switching devices in the power grid topology model, and constructing a time stamp set;

s352, calculating to obtain theoretical transmission delay according to the physical distance between the switch equipment corresponding to each time stamp in the time stamp set and the receiving end and the preset signal transmission speed;

S353, estimating the actual transmission delay by adopting a Kalman filtering algorithm and taking the timestamp set and the theoretical transmission delay as inputs;

And S354, calculating the time difference of the power grid topology model before and after correction according to the estimated actual transmission delay so as to update the power grid topology model.

5. The method for calculating relay protection settings of a coal mine based on topology identification of claim 4, wherein step S354 comprises:

S3541, calculating to obtain a corrected timestamp according to the estimated actual transmission delay and an original timestamp in the state information of the corresponding switch equipment in the power grid topology model by using a formula of corrected timestamp=original timestamp+actual transmission delay;

S3542, replacing an original timestamp in the state information of the corresponding switch equipment in the power grid topology model with the corrected timestamp;

S3543, calculating the absolute value of the difference value between the corrected timestamp and the original timestamp as the time difference of the switch equipment;

S3544, counting time differences of all switching devices in the power grid topology model, and calculating an average value of the time differences to serve as an average time difference of the power grid topology model before and after correction.

6. The method for calculating relay protection setting of a coal mine based on topology identification according to claim 1, wherein step S4 comprises:

s41, determining a topology model updating area, extracting an affected switching equipment identifier and a line segment identifier, and forming area information;

S42, determining the type of a calculation task of relay protection setting required according to the area information, and evaluating the complexity of the calculation task of the affected equipment and the complexity of the task of the circuit to form task description;

s43, monitoring the resource occupation of each computing unit, and obtaining the current load of each computing unit;

S44, evaluating the communication quality between adjacent computing units to obtain communication state information;

S45, integrating the area information, the task description, enabling the current load information and the communication state of each computing unit, and determining an allocation scheme of a relay protection setting computing task by adopting an improved ant colony algorithm with the aim of minimizing computing time.

7. The method for calculating relay protection settings of a coal mine based on topology identification of claim 6, wherein step S45 comprises:

s451, constructing a communication quality evaluation matrix, a calculation unit load matrix and a task allocation fitness function;

s452, initializing ant colony, wherein each ant represents a task allocation scheme, and selecting the next computing unit by adopting a roulette selection strategy according to a communication quality evaluation matrix, a computing unit load matrix and a task allocation fitness function by the ants to construct the task allocation scheme;

S453, determining the concentration of the pheromone according to the fitness function value of the task allocation scheme, and determining the task allocation scheme of each calculation unit according to the task allocation scheme with the highest concentration of the pheromone.

8. The method for calculating relay protection settings for a coal mine based on topology identification of claim 7, wherein step S453 comprises:

S4531, calculating the task load balance degree of each calculation unit in the task allocation scheme represented by each ant, wherein the calculation formula is that the load balance degree=1- (the load standard deviation of each calculation unit/the load average value of each calculation unit);

S4532, calculating a comprehensive evaluation index of the task allocation scheme according to the fitness function value and the load balance of the task allocation scheme, wherein the calculation formula is that the comprehensive evaluation index=alpha is the fitness function value+beta (1-load balance), wherein alpha and beta are weight coefficients, and alpha+beta=1;

S4533, updating a calculation formula of the pheromone according to the comprehensive evaluation index, wherein the calculation formula of the pheromone comprises the concentration= (1-volatilization coefficient) of the pheromone, the current concentration of the pheromone and the concentration increment of the pheromone;

S4534, selecting a task allocation scheme with the highest pheromone concentration by adopting a roulette selection strategy according to the updated pheromone concentration so as to determine the task allocation scheme of each calculation unit.

9. The method for calculating relay protection setting of a coal mine based on topology identification according to claim 1, wherein step S5 comprises:

s51, controlling each computing unit to read original setting parameters from the corresponding relay protection device, and determining the setting parameter category to be updated corresponding to the type of the distributed computing task;

S52, calculating new setting parameters corresponding to the setting parameter categories to be updated in parallel according to the power grid topology model;

s53, packaging a calculation result comprising a calculation unit identifier, a device identifier, a setting parameter category and a new setting parameter value into a result data packet according to a preset format;

S54, sending the result data packet to a corresponding check node, checking the result data packet by the check node according to a preset rule, requesting retransmission if the check fails, and recording if the check fails and exceeds a retransmission threshold;

and S55, if all the check nodes finish checking and pass, issuing the new setting parameters to the corresponding relay protection devices.

10. A coal mine relay protection setting calculation system based on topology identification, which is applied to the steps of the method as claimed in any one of claims 1 to 9, and comprises:

the acquisition module acquires the state information of the switch equipment in each underground area in real time, and compares the current state with the last time state to detect whether the switch state changes or not;

The transmission module is used for generating a reduced change information packet containing a switch equipment identifier, a new state value and a change occurrence time stamp when the switch state changes, and transmitting the reduced change information packet to a ground center calculation unit or an adjacent underground area calculation unit;

The updating module is used for receiving the simplified change information packet and updating the power grid topology model according to the simplified change information packet;

The task distribution module is used for cooperatively distributing relay protection setting calculation tasks according to the area, the calculation task complexity, the current load and the communication state of each calculation unit which are related to the topology model update;

And the execution issuing module is used for controlling each computing unit to execute the distributed relay protection setting computing task, synchronizing and checking the computing result among the related computing units, and issuing the synchronized and checked relay protection setting parameters to the corresponding relay protection devices.