CN116759667A - An energy management method and system for lithium battery energy storage box - Google Patents
An energy management method and system for lithium battery energy storage box Download PDFInfo
- Publication number
- CN116759667A CN116759667A CN202310752303.8A CN202310752303A CN116759667A CN 116759667 A CN116759667 A CN 116759667A CN 202310752303 A CN202310752303 A CN 202310752303A CN 116759667 A CN116759667 A CN 116759667A
- Authority
- CN
- China
- Prior art keywords
- power
- load
- energy storage
- data set
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/4207—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells for several batteries or cells simultaneously or sequentially
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- General Chemical & Material Sciences (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application provides an energy management method and system for a lithium battery energy storage box, which relate to the technical field of intelligent management, and are used for acquiring an energy storage load data set based on a historical energy storage record data set of a first lithium battery energy storage box to cluster to obtain N load clustering results, analyzing the N load clustering results by combining the power data set to obtain N power intervals, building a load-power mapping model, carrying out interval optimization by utilizing an optimization algorithm to obtain N power optimal values, managing the first lithium battery energy storage box, solving the technical problems that in the prior art, the energy storage box is not accurately scheduled due to energy loss caused by overload influencing energy supply time limit or idle load because of the fact that the load and the power are not matched, and carrying out self-adaptive adjustment by analyzing the mapping relation between the load and the power, so that the load and the power in the energy scheduling process are improved, and the accurate and reasonable energy scheduling is realized.
Description
Technical Field
The application relates to the technical field of intelligent management, in particular to an energy management method and system of a lithium battery energy storage box.
Background
And carrying out energy management on the lithium battery energy storage box so as to ensure the efficient and stable operation of an energy supply system, wherein the advantages and disadvantages of the energy management influence the optimized and safe scheduling of energy, and the problems of unstable load, peak-valley difference and the like in the energy supply process are solved by reasonable planning so as to ensure the comprehensive utilization efficiency of energy.
At present, energy scheduling management is mainly performed by establishing a special energy scheduling management system, but most of consideration factors are energy supply and demand configuration problems, and part of detail defects existing in the management process are ignored so as to influence the energy scheduling effect, and further optimization and promotion are required.
In the prior art, in the energy management of a lithium battery energy storage box, due to the fact that load and power are matched and considered for a long time, overload influences energy supply time limit or no load to cause energy loss, and energy scheduling is not accurate enough.
Disclosure of Invention
The application provides an energy management method and system of a lithium battery energy storage box, which are used for solving the technical problems that in the energy management of the lithium battery energy storage box in the prior art, due to the fact that load and power are not matched, energy consumption is caused by overload influencing energy supply time limit or no load, and energy scheduling is not accurate enough.
In view of the above problems, the present application provides an energy management method and system for a lithium battery energy storage box.
In a first aspect, the present application provides a method for energy management of a lithium battery energy storage tank, the method comprising:
acquiring a historical energy storage record data set of the first lithium battery energy storage box according to the data acquisition device;
analyzing according to the historical energy storage record data set to obtain an energy storage load data set and a power data set;
clustering the energy storage load data sets to obtain N load clustering results;
analyzing from the power data set based on the N load clustering results to obtain N power intervals, wherein the N load clustering results correspond to the N power intervals;
obtaining a load-power mapping model according to the N load clustering results and the N power intervals;
based on the load-power mapping model, optimizing the N power intervals by using an optimization algorithm to obtain N power figure of merit, and managing the first lithium battery energy storage box by using the N power figure of merit.
In a second aspect, the present application provides an energy management system for a lithium battery energy storage tank, the system comprising:
the data acquisition module is used for acquiring a historical energy storage record data set of the first lithium battery energy storage box according to the data acquisition device;
the data analysis module is used for analyzing according to the historical energy storage record data set to obtain an energy storage load data set and a power data set;
the data clustering module is used for clustering the energy storage load data sets to obtain N load clustering results;
the clustering result analysis module is used for analyzing the power data set based on the N load clustering results to obtain N power intervals, wherein the N load clustering results correspond to the N power intervals;
the model acquisition module is used for acquiring a load-power mapping model according to the N load clustering results and the N power intervals;
and the optimizing management module is used for optimizing the N power intervals respectively by utilizing an optimizing algorithm based on the load-power mapping model to obtain N power optimal values, and managing the first lithium battery energy storage box by using the N power optimal values.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
according to the energy management method of the lithium battery energy storage box, provided by the embodiment of the application, a historical energy storage record data set of a first lithium battery energy storage box is obtained according to the data acquisition device, and an energy storage load data set and a power data set are obtained through analysis; clustering the energy storage load data set to obtain N load clustering results, and analyzing the N load clustering results by combining the power data set to obtain N power intervals, wherein the N load clustering results correspond to the N power intervals; according to the N load clustering results and the N power intervals, a load-power mapping model is obtained, the N power intervals are optimized by an optimization algorithm to obtain N power figure-of-merit, the N power figure-of-merit is used for managing the first lithium battery energy storage box, the technical problems that in the energy management of the lithium battery energy storage box in the prior art, due to the fact that the load and the power are not matched and are not round, energy supply time limit is affected or no load is caused, energy scheduling is not accurate enough are solved, optimal power optimizing is conducted through analysis of the mapping relation between the load and the power, self-adaptive adjustment is conducted through modeling, load and power matching degree in the energy scheduling process is improved, and accurate and reasonable scheduling of energy is achieved.
Drawings
FIG. 1 is a schematic flow diagram of an energy management method for a lithium battery energy storage box;
fig. 2 is a schematic diagram of a flow chart for obtaining N load clustering results in an energy management method of a lithium battery energy storage box;
FIG. 3 is a schematic diagram of N power figure of merit optimization flow in an energy management method of a lithium battery energy storage tank;
fig. 4 is a schematic structural diagram of an energy management system of a lithium battery energy storage box according to the present application.
Reference numerals illustrate: the system comprises a data acquisition module 11, a data analysis module 12, a data clustering module 13, a clustering result analysis module 14, a model acquisition module 15 and an optimizing management module 16.
Detailed Description
The application provides an energy management method and system for a lithium battery energy storage box, which are used for collecting historical energy storage record data, analyzing the mapping relation between loads and power distribution, determining optimal power distribution parameters under different loads by utilizing an optimization analysis algorithm, and finally training and constructing a load self-adaptive regulation model. In addition, a load test scheme is designed to detect a model so as to realize reasonable management scheduling based on energy supply and demand, and the method is used for solving the technical problem that in the energy management of a lithium battery energy storage box in the prior art, due to the fact that load and power are matched and considered to be bad, overload influences energy supply time limit or no load to cause energy loss, and energy scheduling is not accurate enough.
Example 1
As shown in fig. 1, the present application provides an energy management method of a lithium battery energy storage box, the method is applied to an energy management system of the lithium battery energy storage box, the system is in communication connection with a data acquisition device, and the method includes:
step S100: acquiring a historical energy storage record data set of the first lithium battery energy storage box according to the data acquisition device;
specifically, energy management is performed on the lithium battery energy storage box so as to ensure efficient and stable operation of an energy supply system, wherein the advantages and disadvantages of energy management affect energy optimization and safety scheduling, and the problems of unstable load, peak-valley difference and the like in the energy supply process are solved by reasonable planning so as to ensure comprehensive utilization efficiency of energy. The energy management method of the lithium battery energy storage box is applied to the energy management system, the energy management system is a master control system for energy supply scheduling management, the system is in communication connection with the data acquisition device, and the data acquisition device is auxiliary acquisition equipment for acquiring related demand information of the lithium battery energy storage box.
Specifically, the first lithium battery energy storage box is a target energy storage box for energy scheduling, distribution and management. Setting a preset time interval, namely a time limit range for data acquisition, acquiring historical energy storage data of the first lithium battery energy storage box based on the preset time interval, including load equipment, energy supply states and the like, identifying the acquired data based on a data mapping time node, and generating the historical energy storage record data set by carrying out data serialization arrangement based on a time sequence. And taking the historical energy storage record data set as a reference basis, and carrying out energy scheduling operation analysis on the first lithium battery so as to ensure the fit degree of a follow-up analysis result and the first lithium battery energy storage box.
Step S200: analyzing according to the historical energy storage record data set to obtain an energy storage load data set and a power data set;
step S300: clustering the energy storage load data sets to obtain N load clustering results;
specifically, the historical energy storage record data of the first lithium battery energy storage box are obtained, load association part identification extraction is carried out on each piece of historical energy storage record data, and a plurality of pieces of energy storage load data are determined through integration regularity on the extracted data and are used as an energy storage load data set, wherein the energy storage load data set is provided with identifiers of load equipment, load specifications and the like; and determining execution data in the load energy supply process, determining distribution power corresponding to an energy storage load, and integrating power data to be used as the power data set of the first lithium battery energy storage box, wherein the energy storage load data set corresponds to the power data set in a mapping mode. Further, the energy storage load data set and the power data set are subjected to mapping relation analysis, clustering analysis processing is performed on the energy storage load data set based on a K-means clustering method, N load clustering results are generated, targeted class hierarchy optimal power distribution analysis is performed based on the N load clustering results, and analysis efficiency is effectively improved on the basis of guaranteeing actual matching of analysis results.
Further, as shown in fig. 2, the energy storage load data set is clustered to obtain N load clustering results, and step S300 of the present application further includes:
step S310: carrying out positive serialization processing on the energy storage load data set to obtain a serialized load data set;
step S320: determining a K value clustering interval according to the serialized load data set, wherein K is a positive integer greater than or equal to 2;
step S330: optimizing based on the K value clustering interval to obtain a first K value;
step S340: and clustering the energy storage load data set according to the first K value to obtain N load clustering results.
Further, the step S330 of the present application further includes:
step S331: obtaining a first sample quantization index by performing sample size analysis on the serialized load data set;
step S332: taking the first sample quantization index as a constraint condition for optimizing the K value clustering interval, and determining K value granularity;
step S333: and optimizing in the K-value-based clustering interval according to the K value granularity, and determining the first K value.
Specifically, clustering is performed on the energy storage load data set, specifically, based on the energy storage load data set, data load value measurement and calibration are performed, and then positive serialization processing is performed, namely, the load values are ordered from large to small, the order of corresponding data in the energy storage load data set is synchronously adjusted according to the ordering result, and the obtained data positive sequence is used as the serialization load data set. And determining an initial sequence node and an end sequence node based on the serialized load data set, and determining an integral data magnitude interval as the K value clustering interval. And carrying out interval division on the K value clustering interval, and determining a plurality of clustering item numbers, namely the number of the clustering intervals, as a first K value, wherein the first K value is a positive integer greater than or equal to 2.
Specifically, the data magnitude measurement is performed on the serialized load data set, the sample magnitude, that is, the sample data magnitude, the analysis and conversion are performed on the sample magnitude, the first sample quantization index, that is, the magnitude base of the characterization sample magnitude, is determined, the multi-stage sample quantization index is set in an exemplary self-defining manner, each stage corresponds to different sample magnitude intervals respectively, interval mapping attribution is performed on the sample magnitude, and the grade corresponding to the attribution interval is used as the first sample quantization index. And carrying out optimizing constraint on the K value clustering interval based on the first sample quantization index, wherein the higher the first sample quantization index is, the larger the sample quantity is, the larger the clustering data quantity in the corresponding single clustering interval is, carrying out attribution data item number constraint on the K value clustering interval, and determining the single interval attribution data quantity matched with the K value clustering interval as the K value granularity. And optimizing in the K value clustering interval based on the K value granularity, and determining the number of the optimal clustering intervals to serve as the first K value. And carrying out clustering interval planning through the sample size so as to ensure the preference of the number of the determined clustering intervals, and matching with the current clustering condition.
Further, K data are randomly extracted from the energy storage load data set based on the first K value, the K data are used as pre-selected clustering centers, distances between each data in the energy storage load data set and each pre-selected clustering center are calculated respectively, the energy storage load data set is distributed to the pre-selected clustering centers based on a nearby principle, and primary clustering is completed. And further reselecting a cluster center based on the primary division result, carrying out data metering attribution again, repeating for a plurality of times until a termination cluster condition is met, for example, the cluster center is unchanged, and the like, and taking the current cluster attribution result as the N load cluster results. The data division is performed by K-means clustering, so that the attribution division of the energy storage load data set can be simply and efficiently realized.
Step S400: analyzing from the power data set based on the N load clustering results to obtain N power intervals, wherein the N load clustering results correspond to the N power intervals;
specifically, mapping and corresponding to the power data set based on the N load clustering results, extracting energy storage load data contained in each load clustering result, mapping and extracting corresponding power data in the power data set, sorting the extracted power data from small to large, extracting head-tail power data as critical limiting power, determining corresponding power intervals, respectively carrying out power interval analysis and determination on the N load clustering results, and obtaining N power intervals, wherein the N load clustering results are in one-to-one correspondence with the N power intervals. The optimal power distribution parameter analysis determination under different loads can be further performed based on the N load clustering results and the N power intervals.
Step S500: obtaining a load-power mapping model according to the N load clustering results and the N power intervals;
specifically, mapping and associating the N load clustering results with the N power intervals, and then respectively performing data association connection in the single clustering results to serve as training sample data. Preferably, the training sample data is screened, and data deviating from the main body mapping rule is screened out, so that the accuracy of the data is ensured, and the subsequent training effect is improved. The load-power mapping model is generated by conducting neural network training based on the training sample data. Preferably, the training sample data can be divided, a training set and a testing set are determined and used for training and testing the model, when the output accuracy of the model determined by the testing does not reach the standard, the sample dividing proportion or the sample content is adjusted, and the training and testing are performed again until the output accuracy of the model is qualified. Based on the load-power mapping model, optimal power attribution analysis of different loads can be rapidly performed.
Step S600: based on the load-power mapping model, optimizing the N power intervals by using an optimization algorithm to obtain N power figure of merit, and managing the first lithium battery energy storage box by using the N power figure of merit.
Further, after the N power figures are obtained, step S600 of the present application further includes:
step S610-1: acquiring power data sources corresponding to the power data sets, wherein each power data source corresponds to one lithium battery energy storage element;
step S620-1: grouping the power data sets based on the power data sources to obtain M power data sets;
step S630-1: and taking the M power data sets as M adaptation variables, taking the total power value of the N power figures of merit as an adaptation target, and outputting N groups of power adaptation results based on the M adaptation variables.
Further, the step S630-1 of the present application further includes:
step S631-1: building a load self-adaptive adjustment model according to the N groups of power adaptation results;
step S632-1: acquiring real-time power data of the first lithium battery energy storage box;
step S633-1: inputting the real-time power data into the load self-adaptive adjustment model, and outputting a first self-adaptive adjustment result according to the load self-adaptive adjustment model, wherein the first self-adaptive adjustment result is load data based on the real-time power data;
step S634-1: and adjusting the load of the first lithium battery energy storage box according to the first self-adaptive adjustment result.
Further, the step S631-1 of the present application further comprises:
step S6311-1: obtaining a test sample data set, wherein the test sample data set is a load-power sample data set;
step S6312-1: inputting the load-power sample data set into the load self-adaptive adjustment model for testing, and obtaining a model test result;
step S6313-1: analyzing the model test result to obtain an error checking coefficient and a stable checking coefficient;
step S6314-1: and obtaining a first optimization instruction based on the error checking coefficient and the stability checking coefficient, and optimizing the load self-adaptive adjustment model according to the first optimization instruction.
Specifically, further, an optimization algorithm is embedded into the load-power mapping model, and the N power intervals are optimized to determine optimal power distribution parameters under different loads. The optimization algorithm may be an adaptive algorithm of the current scene, and is not specifically limited, and exemplary, based on a simulated annealing algorithm, the calibration iteration of the power data is performed for multiple times until reaching an iteration termination condition, the current optimal power is determined to be the optimal power, and the corresponding N power figure-of-merit values of the N power intervals are respectively determined.
Further, tracing the power data set, specifically, tracing and analyzing corresponding data acquisition components, namely the lithium battery energy storage elements, according to the acquisition time, the acquisition mode, the acquisition place and the like of the data, mapping and integrating the data tracing result to generate the power data sources, namely the data generation components corresponding to the power data, wherein each power data source corresponds to one lithium battery energy storage element. The difference of the lithium battery energy storage elements causes corresponding power differences, for example, the adaptive power of different loads is different, the load data corresponding to different power conditions is different, and the condition is subjected to refinement analysis, so that the preference of final power distribution can be further improved. And taking the power data source as a division standard, determining a plurality of power data corresponding to the same power data source, namely the same lithium battery energy storage element, according to the power data set, carrying out power data division attribution, and generating M power data sets, wherein M is the same as the class of the lithium battery energy storage element. And taking the M power data sets as M adaptation variables, taking the total power value in the N power figure-of-merit values as an adaptation target, and determining N x M configuration results to determine the optimal power distribution parameters corresponding to each data under different loads.
Specifically, the N sets of adaptation results are obtained, where any set of results includes a power figure of merit and a plurality of adjustable power data corresponding to the mapping. And taking the plurality of adjustable power data as hierarchical identification data, taking the corresponding power figure of merit as hierarchical decision data, determining N adjustment groups, taking the N adjustment groups as sample data, performing neural network training, and generating the load self-adaptive adjustment model. Based on the load self-adaptive adjustment model, the optimal power adaptive adjustment configuration of different loads can be directly carried out, and the objectivity and accuracy of an adjustment result are ensured.
Specifically, the load self-adaptive adjustment model is detected based on a load test scheme. And determining a plurality of groups of energy supply sample data through data statistics, and adjusting a data representation mode to generate a plurality of load-power sequences serving as the test sample data set. And inputting the load-power sample data set into the load self-adaptive adjustment model, and analyzing output power adjustment data by the model, namely the model test result. Performing analysis precision assessment of the load self-adaptive adjustment model based on the model test result, performing sample data analysis, determining power adaptability load data, performing calibration with the model test result, determining data adjustment deviation based on the calibration result, generating the error checking coefficient, and exemplarily, uniformly dividing a plurality of data adjustment deviation intervals, respectively configuring a plurality of error checking coefficients of corresponding grades, performing interval attribution on the data adjustment deviation, taking a configuration coefficient corresponding to the attribution result as the error checking coefficient, wherein the data adjustment deviation is in direct proportion to the error checking coefficient, and the error checking coefficient is characterization data for measuring adjustment deviation; and further determining the frequency of deviation adjustment data based on the calibration result, and generating the stability test coefficient, wherein the generation mode of the stability test coefficient is the same as that of the error test coefficient, the stability test coefficient is in direct proportion to the frequency of the deviation adjustment data, and the stability test coefficient is the characterization data of the steady analysis state of the measurement model. And further configuring an error checking coefficient threshold and a stable checking coefficient threshold, judging whether the error checking coefficient and the stable checking coefficient meet the corresponding checking coefficient threshold, and when the error checking coefficient and the stable checking coefficient do not meet the corresponding checking coefficient threshold, indicating that a model operation mechanism is abnormal and has optimization necessity, and generating the first optimization instruction, namely, a starting instruction for model optimization. And along with the receiving of the first optimizing instruction, optimizing and adjusting the load self-adaptive adjusting model, and exemplarily, rescreening test sample data for secondary training. By analyzing and evaluating the model operation mechanism, the abnormal output result of the model caused by the function limitation of the model is avoided.
Further, real-time power data acquisition is performed on the first lithium battery, the real-time power data is input into the load self-adaptive adjustment model, and adaptive load data of the real-time power data is determined through data matching analysis and mapping and is output as the first self-adaptive adjustment result. And determining the load condition of the first lithium battery energy storage box, and carrying out load adjustment on the first lithium battery energy storage box based on the first self-adaptive adjustment result to realize optimal energy supply.
Further, as shown in fig. 3, step S600 of the present application further includes:
step S610-2: acquiring parameter configuration of the first lithium battery energy storage box, wherein the parameter configuration comprises an energy storage battery pack parameter, an energy storage voltage parameter and an energy storage power supply parameter;
step S620-2: performing power loss analysis according to the energy storage battery pack parameter, the energy storage voltage parameter and the energy storage power supply parameter to obtain a first rated power loss;
step S630-2: and optimizing the N power figures of merit according to the first rated power loss, and outputting N secondary power figures of merit.
Specifically, in the running process of the first lithium battery energy storage box, certain power loss inevitably exists based on multidimensional influence factors, and further optimization and adjustment are required for the power loss in order to ensure the final load analysis accuracy. And acquiring the energy storage battery pack parameter, the energy storage voltage parameter and the energy storage power supply parameter of the first lithium battery energy storage box, wherein power loss is caused due to the conditions of energy dispatching response speed, voltage and current fluctuation and the like. And performing power analysis based on the parameters, determining the first rated power loss, for example, performing power metering based on standard configuration parameter data, performing deviation analysis with current power data, and taking a power difference value as the first rated power loss. And determining an adjusting direction and an adjusting scale based on the first rated power loss, and respectively adjusting and optimizing the N power figures of merit to generate N secondary power figures of merit. By carrying out loss analysis adjustment, the idealization of the determined power figure of merit can be further improved, and the scene matching degree is ensured.
Example 2
Based on the same inventive concept as the energy management method of a lithium battery energy storage tank in the foregoing embodiments, as shown in fig. 4, the present application provides an energy management system of a lithium battery energy storage tank, the system comprising:
the data acquisition module 11 is used for acquiring a historical energy storage record data set of the first lithium battery energy storage box according to the data acquisition device;
the data analysis module 12 is used for analyzing according to the historical energy storage record data set to obtain an energy storage load data set and a power data set;
the data clustering module 13 is used for clustering the energy storage load data sets to obtain N load clustering results;
the clustering result analysis module 14 is configured to analyze the power data set based on the N load clustering results, to obtain N power intervals, where the N load clustering results correspond to the N power intervals;
the model acquisition module 15 is configured to obtain a load-power mapping model according to the N load clustering results and the N power intervals;
and the optimizing management module 16 is configured to optimize the N power intervals by using an optimization algorithm based on the load-power mapping model, to obtain N power figure of merit, and manage the first lithium battery energy storage box with the N power figure of merit.
Further, the system further comprises:
the data processing module is used for carrying out positive serialization processing on the energy storage load data set to obtain a serialized load data set;
the clustering interval determining module is used for determining a K value clustering interval according to the serialized load data set, wherein K is a positive integer greater than or equal to 2;
the interval optimizing module is used for optimizing based on the K value clustering interval to obtain a first K value;
and the clustering result acquisition module is used for clustering the energy storage load data set according to the first K value to obtain N load clustering results.
Further, the system further comprises:
the sample size analysis module is used for obtaining a first sample quantization index by carrying out sample size analysis on the serialized load data set;
the granularity determining module is used for taking the first sample quantization index as a constraint condition for optimizing the K value clustering interval to determine K value granularity;
and the clustering interval optimizing module is used for optimizing in the K-value-based clustering interval according to the K value granularity and determining the first K value.
Further, the system further comprises:
the data source acquisition module is used for acquiring power data sources corresponding to the power data sets, wherein each power data source corresponds to one lithium battery energy storage element;
the data grouping module is used for grouping the power data sets based on the power data sources to obtain M power data sets;
and the adaptation analysis module is used for taking the M power data sets as M adaptation variables, taking the total power value of the N power figures of merit as an adaptation target, and outputting N groups of power adaptation results based on the M adaptation variables.
Further, the system further comprises:
the parameter configuration acquisition module is used for acquiring the parameter configuration of the first lithium battery energy storage box, wherein the parameter configuration comprises an energy storage battery pack parameter, an energy storage voltage parameter and an energy storage power supply parameter;
the loss analysis module is used for carrying out power loss analysis according to the energy storage battery pack parameter, the energy storage voltage parameter and the energy storage power supply parameter to obtain first rated power loss;
and the loss optimization module is used for optimizing the N power figures of merit according to the first rated power loss and outputting N secondary power figures of merit.
Further, the system further comprises:
the model building module is used for building a load self-adaptive adjustment model according to the N groups of power adaptation results;
the real-time power data acquisition module is used for acquiring real-time power data of the first lithium battery energy storage box;
the adjusting result output module is used for inputting the real-time power data into the load self-adaptive adjusting model and outputting a first self-adaptive adjusting result according to the load self-adaptive adjusting model, wherein the first self-adaptive adjusting result is load data based on the real-time power data;
and the load adjusting module is used for adjusting the load of the first lithium battery energy storage box according to the first self-adaptive adjusting result.
Further, the system further comprises:
the system comprises a sample acquisition module, a load-power sample acquisition module and a load-power sample acquisition module, wherein the sample acquisition module is used for acquiring a test sample data set;
the model test module is used for inputting the load-power sample data set into the load self-adaptive adjustment model for testing, and obtaining a model test result;
the coefficient acquisition module is used for analyzing the model test result to obtain an error check coefficient and a stable check coefficient;
the model optimization module is used for obtaining a first optimization instruction based on the error checking coefficient and the stability checking coefficient, and optimizing the load self-adaptive adjustment model according to the first optimization instruction.
The foregoing detailed description of the method for managing energy of a lithium battery energy storage tank will be clear to those skilled in the art, and the device disclosed in this embodiment is relatively simple in description, and relevant places refer to the method section for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. An energy management method for a lithium battery energy storage box, wherein the method is applied to an energy management system of the lithium battery energy storage box, the system is in communication connection with a data acquisition device, and the method comprises:
acquiring a historical energy storage record data set of the first lithium battery energy storage box according to the data acquisition device;
analyzing according to the historical energy storage record data set to obtain an energy storage load data set and a power data set;
clustering the energy storage load data sets to obtain N load clustering results;
analyzing from the power data set based on the N load clustering results to obtain N power intervals, wherein the N load clustering results correspond to the N power intervals;
obtaining a load-power mapping model according to the N load clustering results and the N power intervals;
based on the load-power mapping model, optimizing the N power intervals by using an optimization algorithm to obtain N power figure of merit, and managing the first lithium battery energy storage box by using the N power figure of merit.
2. The method of claim 1, wherein clustering the stored energy load data sets results in N load cluster results, the method comprising:
carrying out positive serialization processing on the energy storage load data set to obtain a serialized load data set;
determining a K value clustering interval according to the serialized load data set, wherein K is a positive integer greater than or equal to 2;
optimizing based on the K value clustering interval to obtain a first K value;
and clustering the energy storage load data set according to the first K value to obtain N load clustering results.
3. The method of claim 2, wherein optimizing based on the K-value cluster interval to obtain a first K-value comprises:
obtaining a first sample quantization index by performing sample size analysis on the serialized load data set;
taking the first sample quantization index as a constraint condition for optimizing the K value clustering interval, and determining K value granularity;
and optimizing in the K-value-based clustering interval according to the K value granularity, and determining the first K value.
4. The method of claim 1, wherein after obtaining the N power figures of merit, the method further comprises:
acquiring power data sources corresponding to the power data sets, wherein each power data source corresponds to one lithium battery energy storage element;
grouping the power data sets based on the power data sources to obtain M power data sets;
and taking the M power data sets as M adaptation variables, taking the total power value of the N power figures of merit as an adaptation target, and outputting N groups of power adaptation results based on the M adaptation variables.
5. The method of claim 1, wherein the method further comprises:
acquiring parameter configuration of the first lithium battery energy storage box, wherein the parameter configuration comprises an energy storage battery pack parameter, an energy storage voltage parameter and an energy storage power supply parameter;
performing power loss analysis according to the energy storage battery pack parameter, the energy storage voltage parameter and the energy storage power supply parameter to obtain a first rated power loss;
and optimizing the N power figures of merit according to the first rated power loss, and outputting N secondary power figures of merit.
6. The method of claim 4, wherein the outputting is based on N sets of power adaptation results for the M adaptation variables, the method further comprising:
building a load self-adaptive adjustment model according to the N groups of power adaptation results;
acquiring real-time power data of the first lithium battery energy storage box;
inputting the real-time power data into the load self-adaptive adjustment model, and outputting a first self-adaptive adjustment result according to the load self-adaptive adjustment model, wherein the first self-adaptive adjustment result is load data based on the real-time power data;
and adjusting the load of the first lithium battery energy storage box according to the first self-adaptive adjustment result.
7. The method of claim 6, wherein the method further comprises:
obtaining a test sample data set, wherein the test sample data set is a load-power sample data set;
inputting the load-power sample data set into the load self-adaptive adjustment model for testing, and obtaining a model test result;
analyzing the model test result to obtain an error checking coefficient and a stable checking coefficient;
and obtaining a first optimization instruction based on the error checking coefficient and the stability checking coefficient, and optimizing the load self-adaptive adjustment model according to the first optimization instruction.
8. An energy management system for a lithium battery energy storage box, the system in communication with a data acquisition device, the system comprising:
the data acquisition module is used for acquiring a historical energy storage record data set of the first lithium battery energy storage box according to the data acquisition device;
the data analysis module is used for analyzing according to the historical energy storage record data set to obtain an energy storage load data set and a power data set;
the data clustering module is used for clustering the energy storage load data sets to obtain N load clustering results;
the clustering result analysis module is used for analyzing the power data set based on the N load clustering results to obtain N power intervals, wherein the N load clustering results correspond to the N power intervals;
the model acquisition module is used for acquiring a load-power mapping model according to the N load clustering results and the N power intervals;
and the optimizing management module is used for optimizing the N power intervals respectively by utilizing an optimizing algorithm based on the load-power mapping model to obtain N power optimal values, and managing the first lithium battery energy storage box by using the N power optimal values.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310752303.8A CN116759667B (en) | 2023-06-26 | 2023-06-26 | Energy management method and system for lithium battery energy storage box |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310752303.8A CN116759667B (en) | 2023-06-26 | 2023-06-26 | Energy management method and system for lithium battery energy storage box |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116759667A true CN116759667A (en) | 2023-09-15 |
| CN116759667B CN116759667B (en) | 2025-12-16 |
Family
ID=87960604
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310752303.8A Active CN116759667B (en) | 2023-06-26 | 2023-06-26 | Energy management method and system for lithium battery energy storage box |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116759667B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105449719A (en) * | 2014-08-26 | 2016-03-30 | 珠海格力电器股份有限公司 | Distributed energy power supply control method, device and system |
| KR101875329B1 (en) * | 2017-06-12 | 2018-07-05 | 한국과학기술원 | Forecasting apparatus and method for power consumption |
| CN109460419A (en) * | 2018-09-19 | 2019-03-12 | 广东卓维网络有限公司 | A kind of equipment state change events monitoring method |
| CN114912546A (en) * | 2022-06-14 | 2022-08-16 | 广东电网有限责任公司 | Energy data aggregation method and device |
| CN115149553A (en) * | 2021-03-29 | 2022-10-04 | 国网河北省电力有限公司雄安新区供电公司 | Method and device for constructing distributed energy storage system |
| CN115358346A (en) * | 2022-09-13 | 2022-11-18 | 西南石油大学 | Source-load intelligent matching method of regional comprehensive energy system |
| CN115686124A (en) * | 2022-12-30 | 2023-02-03 | 南京积芯力科技有限公司 | Energy storage battery output power self-adjusting system and method based on safety protection |
-
2023
- 2023-06-26 CN CN202310752303.8A patent/CN116759667B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105449719A (en) * | 2014-08-26 | 2016-03-30 | 珠海格力电器股份有限公司 | Distributed energy power supply control method, device and system |
| KR101875329B1 (en) * | 2017-06-12 | 2018-07-05 | 한국과학기술원 | Forecasting apparatus and method for power consumption |
| CN109460419A (en) * | 2018-09-19 | 2019-03-12 | 广东卓维网络有限公司 | A kind of equipment state change events monitoring method |
| CN115149553A (en) * | 2021-03-29 | 2022-10-04 | 国网河北省电力有限公司雄安新区供电公司 | Method and device for constructing distributed energy storage system |
| CN114912546A (en) * | 2022-06-14 | 2022-08-16 | 广东电网有限责任公司 | Energy data aggregation method and device |
| CN115358346A (en) * | 2022-09-13 | 2022-11-18 | 西南石油大学 | Source-load intelligent matching method of regional comprehensive energy system |
| CN115686124A (en) * | 2022-12-30 | 2023-02-03 | 南京积芯力科技有限公司 | Energy storage battery output power self-adjusting system and method based on safety protection |
Non-Patent Citations (1)
| Title |
|---|
| 罗朋;吴健威;于跃;杨燕霞;: "基于k均值聚类算法的无均衡自适应锂电储能研究", 电气技术, no. 05, 15 May 2020 (2020-05-15), pages 40 - 47 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116759667B (en) | 2025-12-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114298863B (en) | Data acquisition method and system of intelligent meter reading terminal | |
| CN110991786B (en) | 10kV static load model parameter identification method based on similar daily load curve | |
| CN108376982B (en) | Load phase sequence identification method and device | |
| CN105046591A (en) | Method for evaluating electricity utilization energy efficiency of power consumer | |
| CN119051084B (en) | Dynamic load balancing optimization method and system for micro-grid based on historical electricity consumption data | |
| CN116911471B (en) | A transmission performance optimization method and system for power supply system | |
| CN115373370B (en) | Method and system for monitoring running state of programmable controller | |
| Lin et al. | Approximate mixed-integer programming solution with machine learning technique and linear programming relaxation | |
| CN118586688B (en) | Multi-level data processing method and system in smart power grid | |
| CN119293515B (en) | Electric energy meter error prediction method and system based on big data analysis | |
| CN118868239A (en) | Virtual Synchronous Adaptive Grid-Connected Control Strategy | |
| CN119417189A (en) | Intelligent scheduling method of construction equipment based on AI algorithm | |
| CN118798548A (en) | Edge-to-edge integrated management platform for distributed energy | |
| CN116823215B (en) | Intelligent operation and maintenance management and control method and system for power station | |
| CN116759667B (en) | Energy management method and system for lithium battery energy storage box | |
| CN119362609B (en) | A method for coordinated control of multi-energy power systems at spatiotemporal scales | |
| CN120150305A (en) | One-stop integrated management system for photovoltaic, energy storage and charging | |
| CN119204455A (en) | Multi-dimensional adjustable potential evaluation method of virtual power plant based on integrated intelligent model | |
| CN115629353B (en) | A method and system for pre-checking abnormal electric meter | |
| CN116933201A (en) | Method and system for identifying illegal electricity utilization behavior of low-voltage charging pile | |
| CN115936680A (en) | Intelligent order dispatching method and system for equipment operation and maintenance | |
| CN115759354A (en) | New energy power prediction method and system based on deep learning | |
| CN115576936B (en) | A method and system for intelligent calibration of electricity consumption data | |
| Ahmed et al. | Leveraging Artificial Intelligence for Assessing Metering Faults in Electric Power Systems. | |
| CN120851714B (en) | Power grid data quality management method based on reinforcement learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |