CN116033025A - Distribution network automation computing task scheduling method and system based on cloud-edge collaboration - Google Patents

Abstract

The invention relates to the technical field of power grid dispatching and discloses a cloud edge cooperation-based power distribution network automation computing task dispatching method and a cloud edge cooperation-based power distribution network automation computing task dispatching system.

Description

Distribution network automation computing task scheduling method and system based on cloud-edge collaboration

Technical Field

The present invention relates to the field of power grid dispatching technology, and in particular to a method and system for dispatching automated computing tasks in a distribution network based on cloud-edge collaboration.

Background Art

The automation business of distribution network is of great significance to the safe and economic operation of distribution network. The cloud-edge collaborative system can provide computing resources with high accessibility and large total amount, effectively supporting the automation business of new power system under the background of explosive growth of access objects, increasingly complex operation scenarios and continuous expansion of task scale. Various automation businesses are mapped to computing businesses in the cloud-edge collaborative system during execution. In order to ensure the real-time requirements of the business, sufficient computing resources need to be allocated to the computing business.

Traditional distribution network automation systems do not adopt cloud-edge architecture. Each subsystem is independent and vertically isolated, and lacks a computing resource sharing mechanism. Therefore, the corresponding computing business requests are executed within the corresponding system, and there is no need to pay attention to computing business scheduling.

The existing general cloud-edge collaborative scheduling methods are mostly targeted at Internet application scenarios, and pay more attention to call loss rate, load balancing rate, etc. On the one hand, they fail to adapt to the high real-time requirements of some distribution network automation services. On the other hand, they lack distributed scheduling methods that are compatible with the distribution network structure.

Summary of the invention

The present invention provides a distribution network automation computing task scheduling method and system based on cloud-edge collaboration, which solves the technical problems that the existing cloud-edge collaborative scheduling method fails to adapt to the high real-time requirements of some distribution network automation services and lacks a distributed scheduling method that is compatible with the distribution network structure.

In view of this, the first aspect of the present invention provides a method for scheduling distribution network automation computing tasks based on cloud-edge collaboration, applying a cloud-edge collaboration system, wherein the cloud-edge collaboration system includes a cloud node, a core communication network, an edge layer and a plurality of power terminal devices, wherein the edge layer includes a plurality of edge nodes corresponding one-to-one to a plurality of the power terminal devices, each edge node is communicatively connected, the edge layer is communicatively connected to the cloud node through the core communication network, and the power terminal device is communicatively connected to its corresponding edge node; the method comprises the following steps:

S1. Send a service instance request message to the corresponding edge node through the power terminal device, and initialize its corresponding edge node as the source node;

S2. Parsing the service instance request message through the source node to obtain its corresponding real-time requirement level;

S3. Based on a preset scheduling program, the node with the best scheduling is obtained as the execution node according to the real-time requirement level, and the service instance request message is executed through the execution node.

Preferably, the service instance request message includes a real-time requirement level, and the real-time requirement level is represented as 0, 1 and 2 from high to low.

Preferably, step S3 specifically includes:

S301, identifying the real-time requirement level through the source node, if the real-time requirement level is 0, using the source node as an execution node, and executing the service instance request message locally; if the real-time requirement level is 1 or 2, executing step S302;

S302, calculating the execution delay expectation of the source node according to a preset delay expectation calculation formula by the source node, and sending the service instance request message, the real-time requirement level and the execution delay expectation of the source node to the cloud node;

S303, calculating the expected execution delay of the cloud node according to the preset expected delay calculation formula by the cloud node;

S304, identifying the real-time requirement level through the cloud node, if the real-time requirement level is 1, executing step S305, if the real-time requirement level is 2, executing step S306;

S305, starting single-cloud unilateral collaborative scheduling, specifically including: comparing the expected execution delays of the service instance request message at the source node and the cloud node respectively through the following optimization problem function, obtaining a node with the smallest expected execution delay as an execution node, and executing the service instance request message through the execution node, wherein the formula is:

sti＝FN _gor i＝Cld

Where i is the node, E[l(g,i)] is the expected execution delay, g is the service instance request message, FNg is the source node, and Cld is the cloud node;

S306, starting global cloud-edge collaborative scheduling, specifically including: sending a calculation request to other edge nodes through the cloud node, calculating the execution delay expectation of the corresponding edge node according to the preset delay expectation calculation formula at the other edge node, and sending the execution delay expectation of the edge node to the cloud node;

S307, using the cloud node to compare the expected execution delays of the business instance request message at the source node, the cloud node and the other edge nodes according to the optimization problem function, obtaining a node with the smallest expected execution delay as an execution node, and executing the business instance request message through the execution node.

Preferably, the preset delay expectation calculation formula is:

E[l(g,i)]＝l ^sch (g,i)+E[l ^que (g,i)]+l ^tr (g,i)+l ^cpt (g,i)

Where l ^sch (g,i) is the scheduling delay, l ^tr (g,i) is the transmission delay, E[l ^que (g,i)] is the expected queuing delay, and l ^cpt (g,i) is the calculation delay.

Where Emer _G(g) is the real-time requirement level of the service instance request message, and ε is the delay of a single communication;

为源节点与节点i之间的传输速度；Where D _G(g) is the data volume of the service instance request message,

is the transmission speed between the source node and node i;

Where w _G(g) is the computing load of the service instance request message, and _ci is the computing capacity of node i;

为卸载系数，

为卸载系数

的期望，E[n_i,G]为单位时间τ内接收业务案例种类G请求数量的期望。In the formula, τ is the unit time,

is the unloading coefficient,

is the unloading factor

E[ni _,G ] is the expectation of the number of requests for business case type G received within a unit time τ.

In a second aspect, the present invention provides a distribution network automation computing task scheduling system based on cloud-edge collaboration, which applies a cloud-edge collaboration system, wherein the cloud-edge collaboration system includes a cloud node, a core communication network, an edge layer, and a plurality of power terminal devices, wherein the edge layer includes a plurality of edge nodes corresponding one-to-one to a plurality of the power terminal devices, each edge node is communicatively connected, the edge layer is communicatively connected to the cloud node through the core communication network, and the power terminal device is communicatively connected to its corresponding edge node; its scheduling system includes:

A request sending module, used to send a service instance request message to a corresponding edge node through a power terminal device, and initialize its corresponding edge node as a source node;

A parsing module, used for parsing the service instance request message through the source node to obtain its corresponding real-time requirement level;

The scheduling module is used to obtain the optimally scheduled node as the execution node based on the preset scheduling program and the real-time requirement level, and execute the service instance request message through the execution node.

Preferably, the service instance request message includes a real-time requirement level, and the real-time requirement level is represented as 0, 1 and 2 from high to low.

Preferably, the scheduling module specifically includes:

a first scheduling module, configured to identify the real-time requirement level through the source node, and if the real-time requirement level is 0, use the source node as an execution node and execute the service instance request message locally; if the real-time requirement level is 1 or 2, execute the first calculation module;

The first calculation module is used to calculate the execution delay expectation of the source node according to a preset delay expectation calculation formula through the source node, and send the service instance request message, the real-time requirement level and the execution delay expectation of the source node to the cloud node;

A second calculation module, used to calculate the execution delay expectation of the cloud node according to the preset delay expectation calculation formula through the cloud node;

A scheduling judgment module, used for identifying the real-time requirement level through the cloud node, and executing the second scheduling module if the real-time requirement level is 1, and executing the third scheduling module if the real-time requirement level is 2;

The second scheduling module is used to start single-cloud unilateral collaborative scheduling, specifically including: comparing the expected execution delays of the service instance request message at the source node and the cloud node respectively through the following optimization problem function, obtaining the node with the smallest expected execution delay as the execution node, and executing the service instance request message through the execution node, wherein the formula is:

sti＝FN _gor i＝Cld

Where i is the node, E[l(g,i)] is the expected execution delay, g is the service instance request message, FNg is the source node, and Cld is the cloud node;

The third scheduling module is used to start global cloud-edge collaborative scheduling, specifically including: sending a calculation request to other edge nodes through the cloud node, calculating the execution delay expectation of the corresponding edge node according to the preset delay expectation calculation formula at the other edge node, and sending the execution delay expectation of the edge node to the cloud node;

An execution module is used to compare the expected execution delays of the business instance request message at the source node, the cloud node and other edge nodes respectively through the cloud node according to the optimization problem function, obtain the node with the smallest expected execution delay as the execution node, and execute the business instance request message through the execution node.

Preferably, the preset delay expectation calculation formula is:

E[l(g,i)]＝l ^sch (g,i)+E[l ^que (g,i)]+l ^tr (g,i)+l ^cpt (g,i)

Where l ^sch (g,i) is the scheduling delay, l ^tr (g,i) is the transmission delay, E[l ^que (g,i)] is the expected queuing delay, and l ^cpt (g,i) is the calculation delay.

Where Emer _G(g) is the real-time requirement level of the service instance request message, and ε is the delay of a single communication;

为源节点与节点i之间的传输速度；Where D _G(g) is the data volume of the service instance request message,

is the transmission speed between the source node and node i;

Where w _G(g) is the computing load of the service instance request message, and _ci is the computing capacity of node i;

为卸载系数，

为卸载系数

的期望，E[n_i,G]为单位时间τ内接收业务案例种类G请求数量的期望。In the formula, τ is the unit time,

is the unloading coefficient,

is the unloading factor

E[ni _,G ] is the expectation of the number of requests for business case type G received within a unit time τ.

It can be seen from the above technical solutions that the present invention has the following advantages:

The present invention realizes the communication connection between cloud nodes, multiple edge nodes and multiple power terminal devices by applying the cloud-edge collaborative system, sends a business instance request message to the corresponding edge node through the power terminal device, parses the business instance request message, and obtains its corresponding real-time requirement level, taking into account the differences in real-time requirements of distribution network computing services, based on a preset scheduling program, obtains the best scheduling node as the execution node according to the real-time requirement level, and executes the business instance request message through the execution node, thereby providing a distributed structure compatible with the distribution network structure, which can adapt to the high real-time requirements of some distribution network automation services, can avoid high business timeout rate, and improve the resource utilization of cloud edge nodes.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of the structure of a cloud-edge collaboration system provided by an embodiment of the present invention;

FIG2 is a flow chart of a method for scheduling automatic computing tasks in a distribution network in a cloud-edge collaborative manner provided by an embodiment of the present invention;

Figure 3 is a structural diagram of a cloud-edge collaborative distribution network automation computing task scheduling system provided by an embodiment of the present invention.

DETAILED DESCRIPTION

In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

The present invention provides a distribution network automation computing task scheduling method based on cloud-edge collaboration, which applies a cloud-edge collaboration system. As shown in FIG1 , the cloud-edge collaboration system includes a cloud node, a core communication network, an edge layer, and a plurality of power terminal devices, wherein the edge layer includes a plurality of edge nodes corresponding one-to-one to a plurality of power terminal devices, each edge node is communicatively connected, the edge layer is communicatively connected to the cloud node through the core communication network, and the power terminal device is communicatively connected to its corresponding edge node;

As shown in Figure 2, the distribution network automation computing task scheduling method based on cloud-edge collaboration includes the following steps:

S1. Send a service instance request message to the corresponding edge node through the power terminal device, and initialize its corresponding edge node as the source node;

The source node refers to the edge node that receives the service instance request.

S2. Parse the service instance request message through the source node to obtain its corresponding real-time requirement level;

The service instance request message includes a real-time requirement level, and the real-time requirement level is represented as 0, 1, and 2 from high to low.

S3. Based on the preset scheduling program, the optimally scheduled node is obtained as the execution node according to the real-time requirement level, and the service instance request message is executed through the execution node.

After the service instance request message arrives at the edge node, the scheduling program is started. According to the level of real-time requirement, the scheduling program enters different branches. In the scheduling architecture of the present invention, high real-time requirement services will preempt low real-time requirement services, and services with equal real-time requirements are executed on a first-come, first-served basis. The preemption process only occurs in the queuing stage, i.e., "queuing". Among them, the execution node refers to the node that executes the service instance request message.

It should be noted that this embodiment provides a distribution network automation computing task scheduling method based on cloud-edge collaboration. By applying the cloud-edge collaboration system, the communication connection between cloud nodes, multiple edge nodes and multiple power terminal devices is realized. The business instance request message is sent to the corresponding edge node through the power terminal device, and the business instance request message is parsed to obtain its corresponding real-time requirement level. The real-time requirement differences of the distribution network computing service are taken into account. Based on the preset scheduling program, the optimal scheduling node is obtained as the execution node according to the real-time requirement level, and the business instance request message is executed by the execution node, thereby providing a distributed structure adapted to the distribution network structure, which can adapt to the high real-time requirements of some distribution network automation services, can avoid high service timeout rates, and improve the resource utilization of cloud edge nodes.

In a specific embodiment, step S3 specifically includes:

S301, identifying the real-time requirement level through the source node. If the real-time requirement level is 0, the source node is used as the execution node and the service instance request message is executed locally; if the real-time requirement level is 1 or 2, executing step S302;

S302, calculating the execution delay expectation of the source node according to a preset delay expectation calculation formula by the source node, and sending the service instance request message, the real-time requirement level and the execution delay expectation of the source node to the cloud node;

S303, calculating the expected execution delay of the cloud node according to a preset expected delay calculation formula through the cloud node;

S304, identifying the real-time requirement level through the cloud node, if the real-time requirement level is 1, executing step S305, if the real-time requirement level is 2, executing step S306;

S305, starting single-cloud unilateral collaborative scheduling, specifically including: comparing the expected execution delays of the service instance request message at the source node and the cloud node respectively through the following optimization problem function, obtaining the node with the smallest expected execution delay as the execution node, and executing the service instance request message through the execution node, wherein the formula is:

sti＝FN _gor i＝Cld

Where i is the node, E[l(g,i)] is the expected execution delay, g is the service instance request message, FNg is the source node, and Cld is the cloud node;

Among them, single-cloud unilateral collaborative scheduling means comparing the expected latency of the business instance request message when it is executed on the source node and the cloud node respectively, and selecting the smaller one for scheduling.

S306, starting global cloud-edge collaborative scheduling, specifically including: sending a calculation request to other edge nodes through the cloud node, calculating the execution delay expectation of the corresponding edge node according to a preset delay expectation calculation formula at the other edge node, and sending the execution delay expectation of the edge node to the cloud node;

Among them, global cloud-edge collaborative scheduling is to compare the expected delays of business instance request messages when they are executed at the source node, other edge nodes, and cloud nodes, and select the smaller one for scheduling.

S307. Compare the expected execution delays of the business instance request message at the source node, cloud node and other edge nodes respectively through the cloud node according to the optimization problem function, obtain the node with the smallest expected execution delay as the execution node, and execute the business instance request message through the execution node.

In a specific embodiment, the preset delay expectation calculation formula is:

E[l(g,i)]＝l ^sch (g,i)+E[l ^que (g,i)]+l ^tr (g,i)+l ^cpt (g,i)

Where l ^sch (g,i) is the scheduling delay, l ^tr (g,i) is the transmission delay, E[l ^que (g,i)] is the expected queuing delay, and l ^cpt (g,i) is the calculation delay.

Where Emer _G(g) is the real-time requirement level of the service instance request message, and ε is the delay of a single communication;

The above process requires multiple communications between nodes, which results in scheduling delay. Since the content of communication transmission in the scheduling stage is relatively small, the delay of a single communication is relatively small, which is recorded as ε.

为源节点与节点i之间的传输速度；Where D _G(g) is the data volume of the service instance request message,

is the transmission speed between the source node and node i;

The data volume refers to how much data needs to be transmitted to complete the request message of the service instance.

Where w _G(g) is the computing load of the service instance request message, and _ci is the computing capacity of node i;

The computing load refers to how many instructions are needed to complete the service instance request message.

为卸载系数，

为卸载系数

的期望，E[n_i,G]为单位时间τ内接收业务案例种类G请求数量的期望。In the formula, τ is the unit time,

is the unloading coefficient,

is the unloading factor

E[ni _,G ] is the expectation of the number of requests for business case type G received within a unit time τ.

The offloading coefficient refers to the proportion of business case request information with a real-time requirement of 1 at the edge node that is offloaded to the cloud node.

The above is a detailed description of an embodiment of a distribution network automation computing task scheduling method based on cloud-edge collaboration provided by the present invention. The following is a detailed description of an embodiment of a distribution network automation computing task scheduling system based on cloud-edge collaboration provided by the present invention.

The present invention provides a distribution network automation computing task scheduling system based on cloud-edge collaboration, which applies the cloud-edge collaboration system. As shown in Figure 1, the cloud-edge collaboration system includes a cloud node, a core communication network, an edge layer and multiple power terminal devices, wherein the edge layer includes multiple edge nodes corresponding one-to-one to multiple power terminal devices, and each edge node is communicatively connected to each other. The edge layer is communicatively connected to the cloud node through the core communication network, and the power terminal device is communicatively connected to its corresponding edge node.

For ease of understanding, please refer to Figure 3, the scheduling system includes:

The request sending module 100 is used to send a service instance request message to a corresponding edge node through a power terminal device, and initialize its corresponding edge node as a source node;

The parsing module 200 is used to parse the service instance request message through the source node to obtain its corresponding real-time requirement level;

The scheduling module 300 is used to obtain the optimally scheduled node as the execution node based on the preset scheduling program and the real-time requirement level, and execute the service instance request message through the execution node.

In a specific embodiment, the service instance request message includes a real-time requirement level, where the real-time requirement levels are represented as 0, 1, and 2 from high to low.

In a specific embodiment, the scheduling module specifically includes:

A first scheduling module is used to identify the real-time requirement level through a source node. If the real-time requirement level is 0, the source node is used as an execution node and the service instance request message is executed locally; if the real-time requirement level is 1 or 2, the first calculation module is executed;

A first calculation module, configured to calculate the execution delay expectation of the source node according to a preset delay expectation calculation formula through the source node, and send the service instance request message, the real-time requirement level and the execution delay expectation of the source node to the cloud node;

A second calculation module is used to calculate the execution delay expectation of the cloud node according to a preset delay expectation calculation formula through the cloud node;

A scheduling judgment module is used to identify the real-time requirement level through the cloud node. If the real-time requirement level is 1, the second scheduling module is executed; if the real-time requirement level is 2, the third scheduling module is executed;

The second scheduling module is used to start single-cloud unilateral collaborative scheduling, which specifically includes: comparing the expected execution delays of the service instance request message at the source node and the cloud node respectively through the optimization problem function of the following formula, obtaining the node with the smallest expected execution delay as the execution node, and executing the service instance request message through the execution node, and the formula is:

sti＝FN _gor i＝Cld

Where i is the node, E[l(g,i)] is the expected execution delay, g is the service instance request message, FNg is the source node, and Cld is the cloud node;

The third scheduling module is used to start global cloud-edge collaborative scheduling, specifically including: sending a calculation request to other edge nodes through the cloud node, calculating the execution delay expectation of the corresponding edge node according to a preset delay expectation calculation formula at the other edge node, and sending the execution delay expectation of the edge node to the cloud node;

The execution module is used to compare the expected execution delays of the business instance request message at the source node, cloud node and other edge nodes respectively through the cloud node according to the optimization problem function, obtain the node with the smallest expected execution delay as the execution node, and execute the business instance request message through the execution node.

In a specific embodiment, the preset delay expectation calculation formula is:

E[l(g,i)]＝l ^sch (g,i)+E[l ^que (g,i)]+l ^tr (g,i)+l ^cpt (g,i)

Where l ^sch (g,i) is the scheduling delay, l ^tr (g,i) is the transmission delay, E[l ^que (g,i)] is the expected queuing delay, and l ^cpt (g,i) is the calculation delay.

Where Emer _G(g) is the real-time requirement level of the service instance request message, and ε is the delay of a single communication;

为源节点与节点i之间的传输速度；Where D _G(g) is the data volume of the service instance request message,

is the transmission speed between the source node and node i;

Where w _G(g) is the computing load of the service instance request message, and _ci is the computing capacity of node i;

为卸载系数，

为卸载系数

的期望，E[n_i,G]为单位时间τ内接收业务案例种类G请求数量的期望。In the formula, τ is the unit time,

is the unloading coefficient,

is the unloading factor

E[ni _,G ] is the expectation of the number of requests for business case type G received within a unit time τ.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (8)

1. The cloud edge cooperative system is applied, and comprises cloud nodes, a core communication network, an edge layer and a plurality of power terminal devices, wherein the edge layer comprises a plurality of edge nodes which are in one-to-one correspondence with the power terminal devices, each edge node is in communication connection, the edge layer is in communication connection with the cloud nodes through the core communication network, and the power terminal devices are in communication connection with the corresponding edge nodes; the method is characterized by comprising the following steps:

s1, sending a service instance request message to a corresponding edge node through power terminal equipment, and initializing the corresponding edge node as a source node;

s2, analyzing the service instance request message through the source node to obtain a corresponding real-time requirement level;

and S3, based on a preset scheduling program, obtaining a node with optimal scheduling according to the real-time requirement level as an executing node, and executing the service instance request message through the executing node.

2. The cloud-edge collaboration-based power distribution network automatic computing task scheduling method according to claim 1, wherein the service instance request message comprises real-time requirement levels, and the real-time requirement levels are sequentially represented as 0, 1 and 2 from high to low.

3. The cloud edge collaboration-based power distribution network automatic calculation task scheduling method according to claim 2, wherein step S3 specifically includes:

s301, identifying the real-time requirement level through the source node, and if the real-time requirement level is 0, taking the source node as an executing node and executing the service instance request message on site; if the real-time requirement level is 1 or 2, executing step S302;

s302, calculating an execution delay expected of the source node according to a preset delay expected calculation formula through the source node, and sending the service instance request message, the real-time requirement level and the execution delay expected of the source node to the Yun Jiedian;

s303, calculating an execution delay expectation of the cloud node according to the preset delay expectation calculation formula through the cloud node;

s304, identifying the real-time requirement level through the cloud node, if the real-time requirement level is 1, executing step S305, and if the real-time requirement level is 2, executing step S306;

s305, starting Shan Yun unilateral cooperative scheduling, which specifically comprises the following steps: comparing the execution delay expectations of the service instance request message at the source node and the cloud node respectively through the following optimization problem function, obtaining a node with the minimum execution delay expectations as an execution node, and executing the service instance request message through the execution node, wherein the formula is as follows:

s.t.i=FN _g or i=cld

In the formula, i is a node, E [ l (g, i) ] is an execution delay expectation, g is a service instance request message, FNg is a source node, and Cld is a cloud node;

s306, starting global cloud edge cooperative scheduling, which specifically comprises the following steps: sending calculation requests to other edge nodes through the cloud node, calculating execution delay expectations of corresponding edge nodes at the other edge nodes according to the preset delay expectations calculation formula, and sending the execution delay expectations of the edge nodes to the Yun Jiedian;

s307, comparing the execution delay expectations of the service instance request message at the source node, the cloud node and other edge nodes according to the optimization problem function through the cloud node, obtaining the node with the minimum execution delay expectations as an execution node, and executing the service instance request message through the execution node.

4. The cloud edge collaboration-based power distribution network automation computing task scheduling method according to claim 3, wherein the preset delay expectation calculation formula is:

E[l(g,i)]＝l ^sch (g,i)+E[l ^que (g,i)]+l ^tr (g,i)+l ^cpt (g,i)

in the formula ,l^sch (g, i) is scheduling delay, l ^tr (g, i) is transmission delay, E [ l ] ^que (g,i)]For queuing delay expectations, l ^cpt (g, i) is a calculated time delay; wherein,

wherein Emer _G(g) The real-time requirement level of the service instance request message is that epsilon is the time delay of single communication;

in the formula ,D_G(g) The data volume of the message is requested for the service instance,

the transmission speed between the source node and the node i is;

in the formula ,w_G(g) Requesting messages for service instancesCalculating the load, c _i The power calculation capacity of the node i;

where τ is the unit time,

for unloading coefficients +.>

For unloading coefficients->

E [ n ] _i,G ]The expected number of requests for the business case category G is received for a unit time τ.

5. The cloud edge cooperative system comprises cloud nodes, a core communication network, an edge layer and a plurality of power terminal devices, wherein the edge layer comprises a plurality of edge nodes which are in one-to-one correspondence with the power terminal devices, each edge node is in communication connection, the edge layer is in communication connection with the cloud nodes through the core communication network, and the power terminal devices are in communication connection with the corresponding edge nodes; wherein the scheduling system comprises:

the request sending module is used for sending a service instance request message to the corresponding edge node through the power terminal equipment and initializing the corresponding edge node as a source node;

the analyzing module is used for analyzing the service instance request message through the source node to obtain the corresponding real-time requirement level;

the scheduling module is used for obtaining the node with optimal scheduling as an executing node according to the real-time requirement level based on a preset scheduling program, and executing the service instance request message through the executing node.

6. The cloud-edge collaboration-based power distribution network automated computing task scheduling system of claim 5, wherein the service instance request message comprises a real-time requirement level, and the real-time requirement level is represented as 0, 1 and 2 in sequence from high to low.

7. The cloud-edge collaboration-based power distribution network automation computing task scheduling system of claim 6, wherein the scheduling module specifically comprises:

the first scheduling module is used for identifying the real-time requirement level through the source node, taking the source node as an executing node if the real-time requirement level is 0, and executing the service instance request message on site; if the real-time requirement level is 1 or 2, executing a first calculation module;

the first calculation module is configured to calculate, by using the source node according to a preset delay expected calculation formula, an execution delay expected of the source node, and send the service instance request message, the real-time requirement level, and the execution delay expected of the source node to the Yun Jiedian;

the second calculation module is used for calculating the execution delay expectation of the cloud node according to the preset delay expectation calculation formula through the cloud node;

the scheduling judging module is used for identifying the real-time requirement level through the cloud node, executing the second scheduling module if the real-time requirement level is 1, and executing the third scheduling module if the real-time requirement level is 2;

the second scheduling module is configured to start Shan Yun unilateral cooperative scheduling, and specifically includes: comparing the execution delay expectations of the service instance request message at the source node and the cloud node respectively through the following optimization problem function, obtaining a node with the minimum execution delay expectations as an execution node, and executing the service instance request message through the execution node, wherein the formula is as follows:

s.t.i＝FN _g or i=cld

In the formula, i is a node, E [ l (g, i) ] is an execution delay expectation, g is a service instance request message, FNg is a source node, and Cld is a cloud node;

the third scheduling module is configured to start global cloud edge cooperative scheduling, and specifically includes: sending calculation requests to other edge nodes through the cloud node, calculating execution delay expectations of corresponding edge nodes at the other edge nodes according to the preset delay expectations calculation formula, and sending the execution delay expectations of the edge nodes to the Yun Jiedian;

and the execution module is used for comparing the execution delay expectations of the service instance request message at the source node, the cloud node and other edge nodes respectively according to the optimization problem function through the cloud node, obtaining the node with the minimum execution delay expectations as an execution node, and executing the service instance request message through the execution node.

8. The cloud edge collaboration-based power distribution network automation computing task scheduling system as claimed in claim 7, wherein the preset delay expectation calculation formula is:

E[l(g,i)]＝l ^sch (g,i)+E[l ^que (g,i)]+l ^tr (g,i)+l ^cpt (g,i)

in the formula ,l^sch (g, i) is scheduling delay, l ^tr (g, i) is transmission delay，E[l ^que (g,i)]For queuing delay expectations, l ^cpt (g, i) is a calculated time delay; wherein,

wherein Emer _G(g) The real-time requirement level of the service instance request message is that epsilon is the time delay of single communication;

in the formula ,D_G(g) The data volume of the message is requested for the service instance,

the transmission speed between the source node and the node i is;

in the formula ,w_G(g) Requesting a computational load of messages for a service instance, c _i The power calculation capacity of the node i;

where τ is the unit time,

for unloading coefficients +.>

For unloading coefficients->

E [ n ] _i,G ]The expected number of requests for the business case category G is received for a unit time τ. />

Images