Power multi-network system resource fusion method, device, equipment and medium

Abstract

The invention provides an electric power multi-network system resource fusion method and device, equipment and a medium. Efficient linkage and real-time response of resources in a multi-network system environment are realized by determining an optimal path. Multi-network system sensing data fusion is realized by determining a comprehensive sensing result, and a hierarchical data fusion strategy is used for optimizing real-time processing and transmission efficiency of multi-source data, so that the data fusion capability of a power system is remarkably improved. Dynamic allocation and optimal scheduling of computing resources are achieved through resource allocation control, rapid recovery and continuity keeping can be achieved when nodes or networks are abnormal, the adaptability and flexibility of a system in a complex power environment are enhanced, and it is ensured that the system is kept stable in peak loads and emergencies. And a cross-hierarchy collaborative scheduling strategy is realized through cross-hierarchy task data transfer, so that the stability and reliability of the system when a single node or a local network is abnormal are ensured, and stable operation and real-time response of the power system in emergency and fault states are ensured.

Description

Technical Field

[0001] The present application relates to the field of power network technology, and in particular to a method, device, equipment and medium for integrating power multi-grid resources. Background Art

[0002] With the rapid development of power systems and smart grids, the number and types of devices in power networks are increasing, and communication, perception, and computing resources are widely distributed in various devices and nodes. Resource management methods in related technologies usually adopt decentralized or independent management modes, which makes it difficult to efficiently integrate these distributed resources, resulting in low resource utilization, slow response speed, and inability to adapt to the complex and changeable real-time business needs in power systems. Summary of the invention

[0003] In view of this, the purpose of this application is to propose a method, device, equipment and medium for the integration of multi-grid resources in power grids, which can realize the intelligent linkage and coordination of cross-network communication, perception and computing resources in a multi-grid wireless environment, so as to improve the real-time performance, resource utilization and adaptability of the power grid system.

[0004] Based on the above objectives, the present application provides a method for integrating power multi-grid resources, including:

[0005] Acquire the real-time communication characteristics of each network system, and determine the optimal path for the intelligent gateway to communicate with each network system according to the real-time communication characteristics;

[0006] Determine a network weight coefficient of each network, obtain fused feature data of each network through the optimal path, and determine a comprehensive perception result according to the network weight coefficient and the fused feature data;

[0007] The task priority is determined according to the comprehensive perception result and the task attribute of each network node task, and resource allocation control is performed according to the task priority and the current load of each network node.

[0008] Optionally, after determining the comprehensive perception result, the method further includes:

[0009] Determining whether there is an abnormal event based on the comprehensive sensing result;

[0010] In response to the absence of abnormal events, dynamically adjust the communication resources, sensing resources and computing resources according to the urgency of each network node task and preset constraints;

[0011] In response to the existence of the abnormal event, the abnormal network system and abnormal node corresponding to the abnormal event are determined, and the target optimal path corresponding to the abnormal network system corresponding to the abnormal event is determined, and the task data in the abnormal node is collected through the target optimal path, the target backup node of the abnormal node is determined, and the task data in the abnormal node is migrated to the target backup node.

[0012] Optionally, determining an optimal path for communicating with each network system according to the real-time communication characteristics includes:

[0013] Determine a state vector of each network system according to the real-time communication characteristics;

[0014] The objective function of each communication path is constructed according to the preset adjustment weight coefficient, and the objective function is solved according to the state vector to obtain the objective function value; and the optimal path is determined among multiple communication paths according to the objective function value, and the optimal path is dynamically updated according to the real-time communication characteristics.

[0015] Optionally, determining a comprehensive perception result according to the network weight coefficient and the fusion feature data includes:

[0016] Determine the number of intermediate nodes;

[0017] Determine the product of the network weight coefficient and the corresponding fusion feature data as a single perception result;

[0018] The individual perception results are globally fused according to the number of intermediate nodes to obtain the comprehensive perception result.

[0019] Optionally, determining the task priority according to the comprehensive perception result and the task attribute of each network node task includes:

[0020] Determining the real-time requirement parameters and importance requirement parameters of each network node task according to the comprehensive perception results;

[0021] Determine a real-time weight coefficient and an importance weight coefficient according to the task attributes, and determine the product of the real-time weight coefficient and the real-time requirement parameter as the real-time score, and determine the product of the importance requirement parameter and the importance weight coefficient as the importance score;

[0022] Determine the sum of the real-time score and the importance score as the total task score of the network node task;

[0023] The network node tasks are arranged in descending order according to the total score of the task to obtain the task priority.

[0024] Optionally, the performing resource allocation control according to the task priority and the current load of each network node includes:

[0025] Determine target tasks in order from front to back according to the task priorities, and determine target demand load according to the task attributes of the target tasks;

[0026] Determine a target network node corresponding to the target task, and determine a target idle load according to a maximum load of the target network node and a current load of the target network node;

[0027] In response to the target idle load being greater than or equal to the target demand load, allocating the target task to the target network node and updating the current load of the target network node;

[0028] In response to the target idle load being less than the target required load, dynamic load balancing control is performed among multiple networks according to the current task set of the target network node and the target required load.

[0029] Optionally, the performing dynamic load balancing control among multiple networks according to the current task set of the target network node and the target demand load includes:

[0030] Determine a migratable task in the current task set according to the target demand load;

[0031] Determine a node to be migrated with the largest idle load among the remaining nodes except the target node;

[0032] The migratable task is migrated to the node to be migrated, the target task is allocated to the target node, and the current load of the target node and the current load of the node to be migrated are updated.

[0033] Optionally, determining the migratable tasks in the current task set according to the target demand load includes:

[0034] Determine the task with the lowest priority in the current task set as the task to be migrated, and determine the sum of the required load of the task to be migrated and the target idle load as the available load;

[0035] In response to the available load being greater than or equal to the target required load, determining the task to be migrated as the migratable task;

[0036] In response to the available load being greater than or equal to the target required load, a new task to be migrated is determined according to the priority of each task in the task set until the available load is greater than or equal to the target required load, and all tasks to be migrated are determined as the migratable tasks.

[0037] Based on the same inventive concept, the present application also provides a power multi-grid resource fusion device, comprising:

[0038] The communication module is configured to: obtain the real-time communication characteristics of each network system, and determine the optimal path for the intelligent gateway to communicate with each network system according to the real-time communication characteristics;

[0039] The perception module is configured to: determine a network weight coefficient of each network, obtain fused feature data of each network through the optimal path, and determine a comprehensive perception result according to the network weight coefficient and the fused feature data;

[0040] The computing module is configured to: determine the task priority according to the comprehensive perception result and the task attribute of each network node task, and perform resource allocation control according to the task priority and the current load of each network node.

[0041] Optionally, the power multi-grid resource fusion device further includes:

[0042] An abnormality determination module is configured to: determine whether there is an abnormal event according to the comprehensive sensing result;

[0043] The resource dynamic adjustment module is configured to: in response to the absence of abnormal events, dynamically adjust the communication resources, the sensing resources and the computing resources according to the urgency of the tasks of each network node and the preset constraints;

[0044] The resource linkage module is configured to: in response to the existence of the abnormal event, determine the abnormal network system and abnormal node corresponding to the abnormal event, determine the target optimal path corresponding to the abnormal network system corresponding to the abnormal event, collect the task data in the abnormal node through the target optimal path, determine the target backup node of the abnormal node, and migrate the task data in the abnormal node to the target backup node.

[0045] Optionally, the communication module includes:

[0046] The state monitoring submodule is configured to: determine the state vector of each network system according to the real-time communication characteristics;

[0047] The path selection submodule is configured to: construct an objective function for each communication path according to a preset adjustment weight coefficient, and solve the objective function according to the state vector to obtain an objective function value; and determine the optimal path among multiple communication paths according to the objective function value, and dynamically update the optimal path according to the real-time communication characteristics.

[0048] Optionally, the perception module includes:

[0049] The quantity determination submodule is configured to: determine the quantity of intermediate nodes;

[0050] The single perception submodule is configured to: determine the product of the network weight coefficient and the corresponding fusion feature data as a single perception result;

[0051] The comprehensive perception submodule is configured to: globally fuse the individual perception results according to the number of intermediate nodes to obtain the comprehensive perception result.

[0052] Optionally, the calculation module includes:

[0053] The parameter determination submodule is configured to: determine the real-time requirement parameters and importance requirement parameters of each network node task according to the comprehensive perception result;

[0054] The score determination submodule is configured to: determine a real-time weight coefficient and an importance weight coefficient according to the task attribute, and determine the product of the real-time weight coefficient and the real-time requirement parameter as the real-time score, and determine the product of the importance requirement parameter and the importance weight coefficient as the importance score;

[0055] The total score determination submodule is configured to: determine the sum of the real-time score and the importance score as the total task score of the network node task;

[0056] The task sorting submodule is configured to: sort the network node tasks in descending order according to the total task scores to obtain the task priorities.

[0057] Optionally, the calculation module includes:

[0058] The demand load determination submodule is configured to: determine the target tasks in order from front to back according to the task priorities, and determine the target demand load according to the task attributes of the target tasks;

[0059] The idle load determination submodule is configured to: determine a target network node corresponding to the target task, and determine a target idle load according to a maximum load of the target network node and a current load of the target network node;

[0060] A first load distribution submodule is configured to: in response to the target idle load being greater than or equal to the target demand load, distribute the target task to the target network node and update the current load of the target network node;

[0061] The second load distribution submodule is configured to: in response to the target idle load being less than the target required load, perform dynamic load balancing control among multiple networks according to the current task set of the target network node and the target required load.

[0062] Optionally, the second load distribution submodule includes:

[0063] A migration task screening unit is configured to: determine a migratable task in the current task set according to the target demand load;

[0064] The migration node screening unit is configured to: determine the node to be migrated with the largest idle load among the remaining nodes except the target node;

[0065] The load balancing unit is configured to: migrate the migratable task to the node to be migrated, allocate the target task to the target node, and update the current load of the target node and the current load of the node to be migrated.

[0066] Optionally, the migration task screening unit includes:

[0067] The load calculation subunit is configured to: determine the task with the lowest priority in the current task set as the task to be migrated, and determine the sum of the required load of the task to be migrated and the target idle load as the available load;

[0068] A first task determination subunit is configured to: in response to the available load being greater than or equal to the target required load, determine the task to be migrated as the migratable task;

[0069] The second task determination subunit is configured to: in response to the available load being greater than or equal to the target demand load, determine a new task to be migrated according to the priority of each task in the task set, until the available load is greater than or equal to the target demand load, and determine all the tasks to be migrated as the migratable tasks.

[0070] Based on the same inventive concept, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method as described above when executing the computer program.

[0071] Based on the same inventive concept, the present application also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute the method as described above.

[0072] From the above, it can be seen that the power multi-grid resource fusion method, device, equipment and medium provided in the present application can obtain the real-time communication characteristics of each grid, and determine the optimal path for communicating with each grid based on the real-time communication characteristics; determine the grid weight coefficient of each grid, and obtain the fusion feature data of each grid through the optimal path, and determine the comprehensive perception result based on the grid weight coefficient and the fusion feature data; determine the task priority based on the comprehensive perception result and the task attributes of each grid node task, and perform resource allocation control based on the task priority and the current load of each grid node.

[0073] By determining the optimal path, efficient resource linkage and real-time response in a multi-grid environment can be achieved, significantly improving the system response speed and real-time performance, and realizing efficient linkage of synaesthesia computing resources. By determining the comprehensive perception results, multi-grid perception data fusion can be achieved. The hierarchical data fusion strategy is used to optimize the real-time processing and transmission efficiency of multi-source data, significantly improving the data fusion capability of the power system and providing a basis for the subsequent dynamic allocation and optimal scheduling of computing resources. Dynamic allocation and optimal scheduling of computing resources can be achieved through resource allocation control, which can quickly recover and maintain continuity when abnormalities occur in nodes or networks, enhancing the adaptability and flexibility of the system in complex power environments, and ensuring stability during peak loads and emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0075] Figure 1 A schematic diagram of the structure of the power system of the network system of the embodiment of the present application;

[0076] Figure 2 This is a flow chart of the method for integrating power multi-grid resources according to an embodiment of the present application;

[0077] Figure 3 A flowchart for determining an optimal path for communicating with each network system for an embodiment of the present application;

[0078] Figure 4 A flowchart for determining the comprehensive perception result for an embodiment of the present application;

[0079] Figure 5 A flowchart for determining task priorities for an embodiment of the present application;

[0080] Figure 6 A flowchart of resource allocation control for an embodiment of the present application;

[0081] Figure 7 A flow chart of dynamic load balancing control between multiple networks according to an embodiment of the present application;

[0082] Figure 8 This is a schematic diagram of the structure of a power multi-grid resource fusion device according to an embodiment of the present application;

[0083] Fig. 9 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0084] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0085] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0086] It should be understood herein that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction rather than having any limiting meaning.

[0087] Based on the description of the above background technology, the related technology also has the following situations:

[0088] With the rapid development of power systems and smart grids, the number and types of equipment in power networks are increasing, and communication, perception and computing resources are widely distributed in various types of equipment and nodes. The resource management methods in related technologies usually adopt decentralized or independent management modes, which makes it difficult to efficiently integrate these distributed resources, resulting in low resource utilization, slow response speed, and inability to adapt to the complex and changeable real-time business needs of power systems. For example, in applications such as power fault detection and load scheduling, collaborative work between multiple devices and multiple nodes is usually required, but the resource management methods in related technologies are difficult to achieve efficient linkage of communication, perception and computing resources, which often leads to data delays, slow processing and untimely system response. Especially in the wireless communication environment of multiple networks, due to the significant differences in communication standards, bandwidth limitations and transmission stability of different networks, the dispersion, heterogeneity and dynamic nature of resources further increase the difficulty of management. The traditional single network resource scheduling and static configuration methods can no longer meet the actual needs of smart grids.

[0089] Related technologies extract the initial grid structure of the distribution network, basic power parameter information of chargers and charging piles, typical daily load characteristic curves, typical daily traffic flow and other data information, and information on the coupling nodes of the distribution network and the transportation network; obtain candidate solutions for the optimal configuration of charging stations taking into account traffic flow information; build a multi-objective collaborative planning model for the distribution network taking into account the optimal configuration of charging stations; design a multi-stage solution strategy to obtain the final distribution network planning solution taking into account charging stations. Through the technology of the present invention, the collaborative planning of the transportation network and the distribution network taking into account the geographical strong correlation coupling of traffic flow and power load flow can be realized, which can better provide reference for the distribution network development planning department. This method has advantages in optimizing the planning of distribution networks and charging stations through geographical coupling, but there are significant shortcomings. It has a high degree of dependence on traffic flow and electric vehicle behavior data, and the data is difficult to obtain and the real-time performance is poor, which affects the accuracy of the solution. The queuing model used for charging station configuration is difficult to reflect the actual complex needs, which can easily lead to unreasonable resource allocation. In addition, the multi-stage solution strategy has high computational cost, and the efficiency may be limited when applied on a large scale.

[0090] Related technologies can also treat transmission network planning and distribution network planning as two planning sub-problems, select the transmission power at the connection point of the transmission network and the distribution network as a coupling variable to associate the two sub-problems, take into account the optimal power flow of the network, and establish a transmission network and distribution network coordination planning model based on the optimal power flow; use the coupling variable as a shared variable to realize communication between sub-problems, use the ADMM algorithm to solve the established model, and obtain a preliminary network planning scheme in which the power supply networks of each substation are interconnected; propose a partitioning sub-algorithm to independently divide the power supply networks of each substation in the preliminary planning scheme on the distribution network side, and optimize the preliminary planning scheme on the transmission network side. It can effectively link the transmission and distribution networks, and transform the large-scale transmission and distribution network planning problem into two sub-problems of transmission network planning and distribution network planning, reducing the scale of the model and the complexity of solution. This method proposes a method based on the alternating direction multiplier method and the optimal power flow in the optimization of the transmission and distribution network linkage, which can reduce the scale of the model and the complexity of solution, and achieve a certain degree of transmission and distribution network coordination. However, its shortcomings are more obvious. The method is highly dependent on coupled variables, especially on the shared variables of power transmission between the transmission network and the distribution network. If the information is incomplete or the transmission is interrupted, the solution may fail. In addition, the iteration process of the ADMM algorithm in large-scale networks is relatively complex, and there are problems with convergence and computational efficiency, especially when the load and grid topology change dynamically, it is difficult to quickly adapt and adjust.

[0091] Therefore, there are still problems such as difficulty in integrating multi-network resources, insufficient real-time response, low data fusion efficiency, and poor system stability. Communication, perception, and computing resources are independent in a multi-network wireless environment, resulting in low resource utilization, slow response, and inability to allocate resources in a timely manner in the face of sudden demands and environmental changes, affecting the adaptability and real-time performance of the system. In addition, the fusion processing efficiency of multi-source data is low, the processing bottleneck is obvious, and it is difficult to meet the real-time needs of data analysis. The stability and reliability of multi-network wireless networks are also poor, and system instability is easily caused by high load or transmission anomalies.

[0092] The power multi-grid resource fusion method, device, equipment and medium provided in the embodiments of the present application can obtain the real-time communication characteristics of each grid, and determine the optimal path for communicating with each grid based on the real-time communication characteristics; determine the grid weight coefficient of each grid, and obtain the fusion feature data of each grid through the optimal path, and determine the comprehensive perception result based on the grid weight coefficient and the fusion feature data; determine the task priority based on the comprehensive perception result and the task attributes of each grid node task, and perform resource allocation control based on the task priority and the current load of each grid node.

[0093] By determining the optimal path, efficient resource linkage and real-time response in a multi-grid environment can be achieved, significantly improving the system response speed and real-time performance, and realizing efficient linkage of synaesthesia computing resources. By determining the comprehensive perception results, multi-grid perception data fusion can be achieved. The hierarchical data fusion strategy is used to optimize the real-time processing and transmission efficiency of multi-source data, significantly improving the data fusion capability of the power system and providing a basis for the subsequent dynamic allocation and optimal scheduling of computing resources. Dynamic allocation and optimal scheduling of computing resources can be achieved through resource allocation control, which can quickly recover and maintain continuity when abnormalities occur in nodes or networks, enhancing the adaptability and flexibility of the system in complex power environments, and ensuring stability during peak loads and emergencies.

[0094] The following describes in detail the method for integrating power multi-grid resources provided by the embodiments of the present application in conjunction with the accompanying drawings.

[0095] In some embodiments, the power multi-grid resource fusion method is applied to the intelligent gateway of the power system, wherein the structural diagram of the power system including the multi-grid system is as follows: Figure 1 As shown, the intelligent gateway is connected to a pair of networks, each of which includes at least one intermediate node communicating with the intelligent gateway and multiple local nodes connected to the intermediate nodes (ie, the structure of network 2, network 3 and network n is the same as that of network 1).

[0096] The local node first pre-processes the data locally, including denoising, filtering and preliminary analysis, to reduce unnecessary data volume and transmission burden. Suppose the perception data is the matrix D i Represents the original data collected by the local node, where i is the index of the local node. The original data is denoised and smoothed using a low-pass filter to remove high-frequency noise to obtain the denoised data matrix D i '. Extract the main features of the data after removing noise, including the average value and change rate of voltage, current and temperature, and express the extracted features as vector F i .

[0097] At each intermediate node of the network, the local feature data from each local node is aggregated and initially fused. Assume that each intermediate node receives local feature data F of n local nodes. i , then the goal of the intermediate layer fusion is to aggregate all data into comprehensive features F mid , represents the overall perception result of the network. The received local feature data is fused by weighted average to reflect the data weight of each node. The fusion calculation formula is:

[0098]

[0099] Among them, ω irepresents the weight of the i-th local node, which is based on the reliability and data quality of the corresponding local node, and the F generated by weighted average mid Represents the fused feature data of the intermediate layer.

[0100] After determining the corresponding fusion feature data, it is transmitted to the intelligent gateway through the corresponding communication path. The intelligent gateway integrates the collected fusion feature data to achieve overall perception of the entire multi-grid power system, and provide a basis for the subsequent dynamic allocation and optimization of computing resources, as well as the adaptive linkage and cross-level collaborative scheduling mechanism of synergistic computing resources.

[0101] In some embodiments, Figure 2 As shown, the power multi-grid resource fusion method is applied to the intelligent gateway, including:

[0102] Step 201: Acquire the real-time communication characteristics of each network system, and determine the optimal path for the intelligent gateway to communicate with each network system according to the real-time communication characteristics.

[0103] In specific implementation, in a multi-network wireless environment (network is a system composed of networks and physical devices, such as 5G, NB-IoT, LoRa, etc.), the communication characteristics of each network (bandwidth, latency, signal strength, etc.) will change at any time. In order to realize the linkage of telecomputing (communication, perception and computing) resources, an adaptive path selection algorithm is deployed in the intelligent gateway. By real-time monitoring of the status parameters of each network, the current best performance communication path is dynamically selected as the optimal path, and seamless switching and intelligent load diversion are realized. While ensuring communication stability and real-time performance, efficient linkage of telecomputing resources is realized.

[0104] In the intelligent gateway device, the state parameters of each network system are collected in real time to form the real-time communication characteristics of multiple network systems, which are expressed as the state vector X i =[B i ,L i ,S i ], where B i represents the bandwidth of the ith network, L i represents the delay of the ith network, S i Indicates the signal strength of the ith network. The state monitoring submodule in the intelligent gateway updates the state vector of the network at regular intervals and transmits the data to the path selection submodule of the gateway. Among them, the state monitoring submodule and the path selection submodule are both submodules of the communication module.

[0105] The path selection submodule uses an adaptive path selection algorithm to determine the optimal path. The core of path selection is to consider the bandwidth, delay and signal strength of each network and dynamically select the optimal communication path. To this end, the objective function f(X i )=ω1·Bi -ω2·L2+ω3·s3, where ω1, ω2, and ω3 are adjustment weights to balance the relative importance of bandwidth, latency, and signal strength in different application scenarios. When the load is high, increase the weight of ω2 to reduce latency. When the signal is unstable, increase ω3 to give priority to paths with strong signals. Through this objective function, the system can select the optimal path based on real-time communication characteristics. opt :

[0106]

[0107] After determining the optimal path at the current moment, the performance of the optimal path will change with the change of network status, resulting in the optimal path not being the communication path with the best performance. At this time, it is necessary to dynamically switch the path of the optimal path, intelligently load balance and self-optimize the performance according to the real-time communication characteristics.

[0108] Dynamic path switching, load balancing, and self-optimization are combined to ensure efficient communication of the power system in a multi-grid environment. First, when the monitored real-time communication characteristics change (such as bandwidth reduction or delay increase), the power system will recalculate the objective function value based on the new state vector and dynamically switch to the new optimal path with the highest score to achieve seamless switching of the optimal path and avoid data transmission interruption. Under high load conditions, the adaptive algorithm further diverts data and assigns non-critical data streams to suboptimal paths, reducing the pressure on the main path and giving priority to the transmission of critical data. After each path selection and switching, the actual performance feedback of the optimal path (such as delay, packet loss rate, and stability) is recorded, and the weight parameters of the path selection are adjusted based on the feedback data to gradually optimize the algorithm and enhance the adaptability and stability of the system.

[0109] Step 202: Determine the network weight coefficient of each network, and obtain the fused feature data of each network through the optimal path, and determine the comprehensive perception result based on the network weight coefficient and the fused feature data.

[0110] In specific implementation, in a multi-grid power system, the perception data in each grid comes from a wide range of sources, and transmitting data between different nodes and levels (different local nodes constitute at least one level, and intermediate nodes constitute at least one level) will bring network burden. To ensure data accuracy and real-time performance, the embodiment of the present application deploys a hierarchical data fusion strategy in the intelligent gateway, which gradually processes and aggregates data at nodes of different levels, completes data integration layer by layer from local nodes to intermediate nodes of the intermediate layer, and then to the global gateway node of the intelligent gateway, and finally generates a comprehensive perception result at the system level.

[0111] The local node first preprocesses the data locally, including denoising, filtering and preliminary analysis, to reduce unnecessary data volume and transmission burden, obtain local feature numbers, and represent the extracted local feature numbers as vector F i At each intermediate node of the network, local feature data from each local node is aggregated and initially fused to form the fused feature data F mid .

[0112] The final global fusion is performed in the intelligent gateway, which integrates the data of all intermediate nodes and generates a comprehensive perception result at the system level. Assume that the global gateway node receives data from m intermediate nodes. (where j = 1, 2, 3, ..., m), the goal of global fusion is to form a system-level summed perception vector F G , perform weighted summation on the intermediate layer data from each network system, and integrate the feature data of each network system into the final system-level perception result:

[0113]

[0114] Among them, ω j Indicates the weight of each network system, reflecting the influence of the network system on the overall system. G That is the comprehensive perception result at the global system level.

[0115] By determining the comprehensive perception results, multi-grid perception data fusion is achieved. The hierarchical data fusion strategy is used to optimize the real-time processing and transmission efficiency of multi-source data, which significantly improves the data fusion capability of the power system and provides a basis for the subsequent dynamic allocation and optimization scheduling of computing resources.

[0116] Step 203: Determine the task priority according to the comprehensive perception result and the task attribute of each network node task, and perform resource allocation control according to the task priority and the current load of each network node.

[0117] In specific implementation, the goal of resource allocation control is to achieve dynamic allocation and optimal scheduling of computing resources. The dynamic allocation and optimal scheduling method of computing resources is used to achieve intelligent scheduling of computing resources and cross-network collaborative computing in a multi-network environment based on the task volume and current load of different nodes. This method effectively allocates and adjusts computing resources through task priority sorting and dynamic load balancing strategies, ensures efficient execution of key tasks, and balances the computing load of each node.

[0118] First, we need to prioritize tasks. In the process of allocating computing resources, we prioritize all tasks according to their attributes (such as urgency and importance) to ensure that computing resources for critical tasks are allocated first. i With task volume Ci and priority P i Two characteristics, define the priority sorting rules as follows: Task priority P i Calculation based on the real-time and importance of the task:

[0119] P i =αR i +βI i

[0120] Among them, R i Indicates the real-time requirements of the task, I i represents the importance requirement of the task, α and β are weight coefficients, α represents the real-time requirement parameter, and β represents the importance requirement parameter, which is used to balance real-time and importance. Task priority can be represented in the form of a task priority list. i The value of all tasks is sorted in descending order from high to low to form a task priority list [T1, T2, …, T n ], the task priority list can ensure that high-priority tasks get computing resources first.

[0121] Then, to achieve the optimal allocation of computing resources, the intelligent gateway allocates computing resources in turn according to the task priority. Assume that the computing nodes in the power system are {N1, N2, …, N k}, node N j The computing power is C j , the current load is L j To ensure that task T i The computational requirements of C i To be satisfied, it needs to be allocated to the node N whose computing load is lower than its computing capacity j On, the task T i Assign to node N j Requirements:

[0122] L j +C i ≤C j .

[0123] If the above conditions are met, it means that node N j Have enough computing power to perform the task. Then the task T i Assign to nodes N j , and node N j The current load is updated to:

[0124] L j =L j +C i .

[0125] Finally, in order to ensure that the node is not overloaded, dynamic load balancing is also required. When the computing load of a node exceeds the threshold, the system distributes the task to the computing nodes of other networks through the load balancing mechanism to ensure the computing efficiency of the system. Assume that the threshold load is L T , when node N j The current load L j When the threshold is exceeded, the dynamic load balancing strategy is activated to distribute tasks to other nodes: low-priority tasks are transferred from the overloaded node N to the j Migrate to node N with lower load m , to balance the load. Task migration must meet the following conditions:

[0126] L j -C i ≤L T And L m +C i ≤C m .

[0127] Among them, C m Represents node N m The threshold load, L m Represents node N m The current load.

[0128] After the migration is completed, update node N respectively j and node N m Current load:

[0129] L j =L j -C i ≤L T , L m =L m +C i .

[0130] Regularly check the load status of each node, dynamically adjust task allocation to ensure load balance of each node, and avoid excessive load at a single point affecting the overall performance of the system. After completing each task allocation and load balancing, the feedback data of the allocation results will be recorded, including information such as task completion time and computing consumption of each node, and based on this information, the task priority weight parameters α and β will be optimized and adjusted, so that the scheduling strategy is gradually optimized to improve the overall computing efficiency and response speed of the system.

[0131] In some embodiments, after determining the comprehensive perception result, the method further includes:

[0132] Step 221: Determine whether there is an abnormal event based on the comprehensive perception results.

[0133] In specific implementation, during the operation of the system, the perception module will monitor data changes in real time and use trend analysis to detect abnormal events (such as voltage fluctuations, current mutations, etc.) to determine whether there are abnormal events.

[0134] Step 2022: In response to the absence of abnormal events, dynamically adjust communication resources, perception resources and computing resources according to the urgency of each network node task and preset constraints.

[0135] In specific implementation, if there are no abnormal events, it is only necessary to dynamically balance communication, perception and computing. The dynamic balancing process adopts resource adaptive scheduling and multi-resource linkage coordination strategy. In the resource management process, the allocation ratio of communication resources, perception resources and computing resources is dynamically adjusted according to the real-time and urgency of the task to ensure the priority execution of key tasks. The system realizes the linkage scheduling of multiple resources through intelligent collaborative algorithms, so that various resources can achieve the optimal combination in a multi-network environment. Among them, using λ comm represents the adaptive ratio of communication resources in resource allocation, λ sense represents the adaptive ratio of perceptual resources in resource allocation, λ comp represents the adaptive proportion of computing resources in resource allocation, and represents the weights of communication resources, sensing resources, and computing resources respectively, where λ comm , sense , comp The preset constraints must be met:

[0136] λ comm +λ sense +λ comp =1

[0137] According to the urgency of the tasks of each network node, resources are dynamically adjusted under the above constraints to ensure that key resources are used for high-priority tasks first.

[0138] Step 2023: In response to the existence of an abnormal event, determine the abnormal network system and abnormal node corresponding to the abnormal event, and determine the target optimal path corresponding to the abnormal network system corresponding to the abnormal event, and collect the task data in the abnormal node through the target optimal path, determine the target backup node of the abnormal node, and migrate the task data in the abnormal node to the target backup node.

[0139] In specific implementation, if there is an abnormal event, high-frequency data collection is automatically triggered and rapid resource linkage is performed. Assume that the perception data is a time series D(t). When its rate of change exceeds the set threshold δ, the system triggers an event response. When an abnormal event is triggered, the abnormal network system and abnormal node corresponding to the abnormal event are determined, and the target optimal path corresponding to the abnormal network system corresponding to the abnormal event is determined. Multiple nodes respond collaboratively, and the task data in the abnormal node is collected through the target optimal path, and the task data is sent to the upper computing node through rapid data transmission between multiple nodes, ensuring that the intelligent gateway can quickly realize data collection to ensure rapid processing of abnormal situations.

[0140] After the task data collection in the abnormal node is completed, the cross-level resource allocation and fault tolerance mechanism is triggered. Establish resource pools between different levels and different network systems, and implement hierarchical resource allocation according to task requirements. In emergencies, resources at all levels can be flexibly called to ensure the continuity of critical tasks. In addition, the system has a fault tolerance function. When a node fails or the network is abnormal, the fault tolerance mechanism is automatically started to transfer the task data to the backup node or replace the path, thereby ensuring the stability and continuity of task execution. When a node failure is detected, the system will dynamically migrate the task to the backup node to achieve seamless switching of tasks.

[0141] When an abnormal event is triggered, the cross-level collaborative scheduling strategy is implemented by transferring task data across levels, ensuring the stability and reliability of the system in the event of single-node or local network anomalies, and ensuring the stable operation and real-time response of the power system in emergencies and fault conditions.

[0142] In summary, the power multi-grid resource fusion method provided by the embodiment of the present application. By determining the optimal path, efficient linkage and real-time response of resources in a multi-grid environment are achieved, the system response speed and real-time performance are significantly improved, and efficient linkage of synaesthesia computing resources is achieved. Multi-grid perception data fusion is achieved by determining the comprehensive perception results, and the hierarchical data fusion strategy is used to optimize the real-time processing and transmission efficiency of multi-source data, which significantly improves the data fusion capability of the power system and provides a basis for the subsequent dynamic allocation and optimization scheduling process of computing resources. The dynamic allocation and optimization scheduling of computing resources are realized through resource allocation control, which can quickly recover and maintain continuity when an abnormality occurs in a node or network, enhance the adaptability and flexibility of the system in a complex power environment, and ensure stability in peak loads and emergencies. And when an abnormal event is triggered, the cross-level collaborative scheduling strategy is realized by transferring task data across levels, which ensures the stability and reliability of the system when a single node or local network is abnormal, and ensures the stable operation and real-time response of the power system in emergencies and fault conditions.

[0143] In some embodiments, Figure 3As shown, the optimal path for communicating with each network system is determined based on the real-time communication characteristics, including:

[0144] Step 301: Determine the state vector of each network system according to the real-time communication characteristics.

[0145] In the specific implementation, the state parameters of each network system are collected in real time to form the real-time communication characteristics of multiple network systems, which are expressed as the state vector X i =[B i ,L i ,S i ], where B i represents the bandwidth of the ith network, L i represents the delay of the ith network, S i Represents the signal strength of the i-th network.

[0146] Step 302: construct an objective function for each communication path according to a preset adjustment weight coefficient, and solve the objective function according to the state vector to obtain an objective function value; determine an optimal path among multiple communication paths according to the objective function value, and dynamically update the optimal path according to real-time communication characteristics.

[0147] In specific implementation, the state monitoring submodule in the intelligent gateway updates the state vector of the network at regular intervals and transmits the data to the path selection submodule of the gateway. The path selection submodule uses an adaptive path selection algorithm to determine the optimal path. The core of path selection is to comprehensively consider the bandwidth, delay and signal strength of each network and dynamically select the optimal communication path. To this end, the objective function f(X i )=ω1·B i -ω2·L2+ω3·s3, where ω1, ω2, and ω3 are adjustment weight coefficients to balance the relative importance of bandwidth, delay, and signal strength in different application scenarios. When the load is high, increase the weight of ω2 to reduce delay. When the signal is unstable, increase ω3 to give priority to the path with strong signal. Through this objective function, the system can select the optimal path according to the real-time communication characteristics. opt :

[0148]

[0149] After determining the optimal path at the current moment, the performance of the optimal path will change with the change of network status, resulting in the optimal path not being the communication path with the best performance. At this time, it is necessary to dynamically switch the path of the optimal path, intelligently load balance and self-optimize the performance according to the real-time communication characteristics.

[0150] In some embodiments, Figure 4 As shown in the figure, the comprehensive perception result is determined based on the network weight coefficient and the fusion feature data, including:

[0151] Step 401: Determine the number of intermediate nodes.

[0152] In specific implementation, the final global fusion is carried out in the intelligent gateway, which integrates the data of all intermediate nodes and generates system-level comprehensive perception results, so the number of intermediate nodes m needs to be determined.

[0153] Step 402: Determine the product of the network weight coefficient and the corresponding fusion feature data as a single perception result.

[0154] In specific implementation, use ω j Represents the weight coefficient of the jth intermediate node, represents the fused feature data of the jth intermediate node, then the single perception result is

[0155] Step 403: globally integrate the individual perception results according to the number of intermediate nodes to obtain a comprehensive perception result.

[0156] In specific implementation, the global gateway node receives data from m intermediate nodes. (where j = 1, 2, 3, ..., m), the goal of global fusion is to form a system-level summed perception vector F G , perform weighted summation on the intermediate layer data from each network system, and integrate the feature data of each network system into the final system-level perception result:

[0157]

[0158] Among them, ω j Indicates the weight of each network system, reflecting the influence of the network system on the overall system. G That is the comprehensive perception result at the global system level.

[0159] By determining the comprehensive perception results, multi-grid perception data fusion is achieved. The hierarchical data fusion strategy is used to optimize the real-time processing and transmission efficiency of multi-source data, which significantly improves the data fusion capability of the power system and provides a basis for the subsequent dynamic allocation and optimization scheduling of computing resources.

[0160] In some embodiments, Figure 5 As shown in the figure, the task priority is determined based on the comprehensive perception results and the task attributes of each network node task, including:

[0161] Step 501: Determine the real-time requirement parameters and importance requirement parameters of each network node task based on the comprehensive perception result.

[0162] In the specific implementation, task priority sorting is required. In the process of allocating computing resources, all tasks are prioritized according to their urgency and importance to ensure that computing resources for key tasks are allocated first. Therefore, it is necessary to first determine the real-time demand parameter R. i , indicating the real-time requirement of the task and the importance requirement parameter I i , indicating the importance of the task.

[0163] Step 502: Determine the real-time weight coefficient and the importance weight coefficient according to the task attributes, and determine the product of the real-time weight coefficient and the real-time requirement parameter as the real-time score, and determine the product of the importance requirement parameter and the importance weight coefficient as the importance score.

[0164] In specific implementation, when determining the priority, the real-time requirement parameter α and the importance requirement parameter β must be determined first to balance real-time and importance. The product of the real-time weight coefficient and the real-time requirement parameter is determined as the real-time score αR i , and the product of the importance requirement parameter and the importance weight coefficient is determined as the importance score βI i .

[0165] Step 503: Determine the sum of the real-time score and the importance score as the total task score of the network node task.

[0166] When implementing, define the task priority P i The calculation process of the total score of the task is as follows:

[0167] P i =αR i +βI i .

[0168] Step 504: Arrange the network node tasks in descending order according to the total task scores to obtain the task priorities.

[0169] In specific implementation, task priority can be expressed in the form of task priority list. i The value of all tasks is sorted in descending order from high to low to form a task priority list [T1, T2, …, T n ], the task priority list can ensure that high-priority tasks get computing resources first.

[0170] In some embodiments, Figure 6 As shown, resource allocation control is performed based on task priority and the current load of each network node, including:

[0171] Step 601: Determine target tasks in order from front to back according to task priorities, and determine target demand load according to task attributes of the target tasks.

[0172] In the specific implementation, in order to achieve the optimal allocation of computing resources, the intelligent gateway allocates computing resources in sequence according to the task priority. If the task ranked first in the task priority is T i , then T i As the first target task, T i Take the example to illustrate that the target task T i The computational requirements are expressed in C i express.

[0173] Step 602: Determine a target network node corresponding to the target task, and determine a target idle load according to a maximum load of the target network node and a current load of the target network node.

[0174] In the specific implementation, it is assumed that the grid nodes {N1, N2, ..., N k}Node N j To execute the target task T i , then the node N j As the target network node, if the target network node N j The maximum load (also called computing power) is C j , target network node N j The current load is L j To ensure the target task T i The computational requirements of C i To be satisfied, it needs to be allocated to the node N whose computing load is lower than its computing capacity j Therefore, it is necessary to first determine the target idle load C of the target network node. j -L j , which indicates the maximum load that can be allocated to the target network node. The target network node can be allocated a load less than or equal to C j -L j task.

[0175] Step 603: In response to the target idle load being greater than or equal to the target demand load, the target task is allocated to the target network node, and the current load of the target network node is updated.

[0176] In specific implementation, if the target idle load is greater than or equal to the target demand load, it means that the target task T i Assign to target network node N j After that, meet:

[0177] L j +C i ≤C j .

[0178] Description Target network node N jHave enough computing power to perform the target task without overloading. Then the target task T i Assign to target network node N j , and the target network node N j The current load is updated to:

[0179] L j =L j +C i .

[0180] Step 604: In response to the target idle load being less than the target demand load, dynamic load balancing control among multiple networks is performed according to the current task set of the target network node and the target demand load.

[0181] In the specific implementation, in order to ensure that the node is not overloaded, dynamic load balancing is also required. Therefore, when the target idle load is less than the target demand load, dynamic load balancing control between multiple networks is performed based on the current task set of the target network node and the target demand load.

[0182] In some embodiments, Figure 7 As shown, dynamic load balancing control among multiple networks is performed according to the current task set and target demand load of the target network node, including:

[0183] Step 701: Determine migratable tasks in the current task set according to the target demand load.

[0184] In specific implementation, when the computing load of a node exceeds the threshold, the system distributes the task to computing nodes of other networks through the load balancing mechanism to ensure the computing efficiency of the system. Assume that the threshold load is L T , when node N j The current load L j When the threshold is exceeded, the dynamic load balancing strategy is activated to move low-priority tasks from the overloaded node N. j Migrate to node N with lower load m , to balance the load. However, migrating a low-priority task may still cause overload, so it is necessary to determine the migratable tasks in the current task set based on the target demand load.

[0185] Then, the migratable tasks are determined in the current task set according to the target demand load, including:

[0186] Step 7011: Determine the task with the lowest priority in the current task set as the task to be migrated, and determine the sum of the required load of the task to be migrated and the target idle load as the available load.

[0187] In specific implementation, the current task set is the target network node N jThe set of tasks being executed in the task set. If T1 is the task with the lowest priority in the current task set, then T1 is the task to be migrated. If the required load of T1 is C1, then the available load = C1 + C j -L j , indicating the target network node N after the task to be migrated is migrated j The size of the load that can be distributed.

[0188] Step 7012: In response to the available load being greater than or equal to the target required load, the task to be migrated is determined as a migratable task.

[0189] In specific implementation, if the available load is greater than or equal to the target required load, it means that migrating the task to be migrated T1 can achieve the allocation of the target task to the target network node N. j .

[0190] Step 7013: In response to the available load being greater than or equal to the target required load, new tasks to be migrated are determined according to the priority of each task in the task set until the available load is greater than or equal to the target required load, and all tasks to be migrated are determined as migratable tasks.

[0191] In specific implementation, if the available load is greater than or equal to the target required load, it means that migrating the task to be migrated T1 still cannot allocate the target task to the target network node N. j , the new tasks to be migrated are still determined according to the priority of each task in the task set, until the available load is greater than or equal to the target demand load, and all the tasks to be migrated are determined as migratable tasks. For example, assuming that the current task set is T1, T2, T3, T4 and T5 in order of priority from low to high, if the available load is greater than or equal to the target demand load after migrating T1, then continue to migrate T2. If the available load is greater than or equal to the target demand load after migrating T2, T1 and T2 are determined as migratable tasks. Otherwise, continue to migrate T3 until the available load is greater than or equal to the target demand load.

[0192] Step 702: Determine the node to be migrated with the largest idle load among the remaining nodes except the target node.

[0193] In the specific implementation, the premise of task migration is that there are nodes that can accommodate migratable tasks, so it is necessary to determine the node to be migrated with the largest idle load among the remaining nodes except the target node to ensure that the computing efficiency of the entire system can be increased after migrating the migratable tasks to the node to be migrated.

[0194] Step 703: Migrate the migratable tasks to the node to be migrated, assign the target task to the target node, and update the current load of the target node and the current load of the node to be migrated.

[0195] In specific implementation, when the computing load of a node exceeds the threshold, the system distributes the task to computing nodes of other networks through the load balancing mechanism to ensure the computing efficiency of the system. Assume that the threshold load is L T , when node N j The current load L j When the threshold is exceeded, the dynamic load balancing strategy is activated to distribute tasks to other nodes: low-priority tasks are transferred from the overloaded node N to the j Migrate to the node N with the lowest load to be migrated m , to balance the load. Task migration must meet the following conditions:

[0196] L j -C i ≤L T And L m +C i ≤C m .

[0197] Among them, C m Represents node N m The threshold load, L m Represents node N m The current load of the node to be migrated is guaranteed to be not overloaded after migrating the migratable tasks to the node to be migrated.

[0198] After the migration is completed, update node N respectively j and node N m Current load:

[0199] L j =L j -C i ≤L T , L m =L m +C i .

[0200] Regularly check the load status of each node, dynamically adjust task allocation to ensure load balance of each node, and avoid excessive load at a single point affecting the overall performance of the system. After completing each task allocation and load balancing, the feedback data of the allocation results will be recorded, including information such as task completion time and computing consumption of each node, and based on this information, the task priority weight parameters α and β will be optimized and adjusted, so that the scheduling strategy is gradually optimized to improve the overall computing efficiency and response speed of the system.

[0201] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.

[0202] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0203] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a power multi-grid resource fusion device.

[0204] refer to Figure 8 , power multi-grid resource fusion device, including:

[0205] The communication module 10 is configured to: obtain the real-time communication characteristics of each network system, and determine the optimal path for communicating with each network system according to the real-time communication characteristics;

[0206] The perception module 20 is configured to: determine a network weight coefficient of each network, obtain fused feature data of each network through an optimal path, and determine a comprehensive perception result according to the network weight coefficient and the fused feature data;

[0207] The calculation module 30 is configured to: determine the task priority according to the comprehensive perception result and the task attribute of each network node task, and perform resource allocation control according to the task priority and the current load of each network node.

[0208] In some embodiments, Figure 8 As shown, the power multi-grid resource fusion device also includes:

[0209] The abnormality determination module 40 is configured to: determine whether there is an abnormal event according to the comprehensive sensing result;

[0210] The resource dynamic adjustment module 50 is configured to: in response to the absence of abnormal events, dynamically adjust the communication resources, the sensing resources and the computing resources according to the urgency of each network node task and the preset constraints;

[0211] The resource linkage module 60 is configured to: in response to the existence of an abnormal event, determine the abnormal network system and abnormal node corresponding to the abnormal event, determine the target optimal path corresponding to the abnormal network system corresponding to the abnormal event, collect the task data in the abnormal node through the target optimal path, determine the target backup node of the abnormal node, and migrate the task data in the abnormal node to the target backup node.

[0212] In some embodiments, the communication module 10 includes:

[0213] The state monitoring submodule is configured to: determine the state vector of each network system according to the real-time communication characteristics;

[0214] The path selection submodule is configured to: construct an objective function for each communication path according to a preset adjustment weight coefficient, and solve the objective function according to the state vector to obtain an objective function value; and determine an optimal path among multiple communication paths according to the objective function value, and dynamically update the optimal path according to real-time communication characteristics.

[0215] In some embodiments, the perception module 20 includes:

[0216] The quantity determination submodule is configured to: determine the quantity of intermediate nodes;

[0217] The single perception submodule is configured to: determine the product of the network weight coefficient and the corresponding fusion feature data as the single perception result;

[0218] The comprehensive perception submodule is configured to globally fuse the individual perception results according to the number of intermediate nodes to obtain a comprehensive perception result.

[0219] In some embodiments, the computing module 30 includes:

[0220] The parameter determination submodule is configured to: determine the real-time requirement parameters and importance requirement parameters of each network node task according to the comprehensive perception result;

[0221] The score determination submodule is configured to: determine a real-time weight coefficient and an importance weight coefficient according to the task attribute, and determine the product of the real-time weight coefficient and the real-time requirement parameter as the real-time score, and determine the product of the importance requirement parameter and the importance weight coefficient as the importance score;

[0222] The total score determination submodule is configured to: determine the sum of the real-time score and the importance score as the total task score of the network node task;

[0223] The task sorting submodule is configured to sort the network node tasks in descending order according to the total task scores to obtain the task priority.

[0224] In some embodiments, the computing module further includes:

[0225] The demand load determination submodule is configured to: determine the target tasks in order from front to back according to the task priorities, and determine the target demand load according to the task attributes of the target tasks;

[0226] The idle load determination submodule is configured to: determine a target network node corresponding to a target task, and determine a target idle load according to a maximum load of the target network node and a current load of the target network node;

[0227] The first load distribution submodule is configured to: in response to the target idle load being greater than or equal to the target demand load, distribute the target task to the target network node and update the current load of the target network node;

[0228] The second load distribution submodule is configured to: in response to the target idle load being less than the target demand load, perform dynamic load balancing control among multiple networks according to the current task set of the target network node and the target demand load.

[0229] In some embodiments, the second load distribution submodule includes:

[0230] The migration task screening unit is configured to: determine a migratable task in the current task set according to a target demand load;

[0231] The migration node screening unit is configured to: determine the node to be migrated with the largest idle load among the remaining nodes except the target node;

[0232] The load balancing unit is configured to: migrate the migratable tasks to the nodes to be migrated, allocate the target tasks to the target nodes, and update the current load of the target nodes and the current load of the nodes to be migrated.

[0233] In some embodiments, the migration task screening unit includes:

[0234] The load calculation subunit is configured to: determine the task with the lowest priority in the current task set as the task to be migrated, and determine the sum of the required load and the target idle load of the task to be migrated as the available load;

[0235] The first task determination subunit is configured to: in response to the available load being greater than or equal to the target required load, determine the task to be migrated as a migratable task;

[0236] The second task determination subunit is configured to: in response to the available load being greater than or equal to the target required load, determine a new task to be migrated according to the priority of each task in the task set, until the available load is greater than or equal to the target required load, and determine all the tasks to be migrated as migratable tasks.

[0237] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0238] The device of the above embodiment is used to implement the corresponding power multi-grid resource integration method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0239] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for integrating multi-grid resources of electric power described in any of the above embodiments is implemented.

[0240] Fig. 9 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0241] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0242] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0243] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0244] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0245] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0246] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0247] The electronic device of the above-mentioned embodiment is used to implement the corresponding power multi-grid resource integration method in any of the above-mentioned embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0248] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the power multi-grid resource integration method as described in any of the above embodiments.

[0249] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0250] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the power multi-grid resource integration method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0251] It is understandable that before using the technical solutions of each embodiment of the present application, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0252] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly remind the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can independently choose whether to provide personal information to the electronic device, application, server, storage medium or other software or hardware that performs the operation of the technical solution of the present application according to the prompt message.

[0253] As an optional but non-limiting implementation, in response to receiving the user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0254] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation method of this application. Other methods that meet relevant laws and regulations may also be applied to the implementation method of this application.

[0255] A person skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0256] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0257] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0258] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the present application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the protection scope of the present application.

Claims

1. A method for integrating power multi-grid resources, characterized in that: include: Acquire the real-time communication characteristics of each network system, and determine the optimal path for the intelligent gateway to communicate with each network system according to the real-time communication characteristics; Determine a network weight coefficient of each network, obtain fused feature data of each network through the optimal path, and determine a comprehensive perception result according to the network weight coefficient and the fused feature data; The task priority is determined according to the comprehensive perception result and the task attribute of each network node task, and resource allocation control is performed according to the task priority and the current load of each network node.

2. The method for integrating power multi-grid resources according to claim 1, characterized in that: After determining the comprehensive perception result, the method further includes: Determining whether there is an abnormal event based on the comprehensive sensing result; In response to the absence of abnormal events, dynamically adjust the communication resources, sensing resources and computing resources according to the urgency of each network node task and preset constraints; In response to the existence of the abnormal event, the abnormal network system and abnormal node corresponding to the abnormal event are determined, and the target optimal path corresponding to the abnormal network system corresponding to the abnormal event is determined, and the task data in the abnormal node is collected through the target optimal path, the target backup node of the abnormal node is determined, and the task data in the abnormal node is migrated to the target backup node.

3. The method for integrating power multi-grid resources according to claim 1, characterized in that: The determining the optimal path for communicating with each network system according to the real-time communication characteristics includes: Determine a state vector of each network system according to the real-time communication characteristics; The objective function of each communication path is constructed according to the preset adjustment weight coefficient, and the objective function is solved according to the state vector to obtain the objective function value; and the optimal path is determined among multiple communication paths according to the objective function value, and the optimal path is dynamically updated according to the real-time communication characteristics.

4. The method for integrating power multi-grid resources according to claim 1, characterized in that: The determining of the comprehensive perception result according to the network weight coefficient and the fusion feature data includes: Determine the number of intermediate nodes; Determine the product of the network weight coefficient and the corresponding fusion feature data as a single perception result; The individual perception results are globally fused according to the number of intermediate nodes to obtain the comprehensive perception result.

5. The method for integrating power multi-grid resources according to claim 1, characterized in that: The determining of the task priority according to the comprehensive perception result and the task attribute of each network node task includes: Determining the real-time requirement parameters and importance requirement parameters of each network node task according to the comprehensive perception results; Determine a real-time weight coefficient and an importance weight coefficient according to the task attributes, and determine the product of the real-time weight coefficient and the real-time requirement parameter as the real-time score, and determine the product of the importance requirement parameter and the importance weight coefficient as the importance score; Determine the sum of the real-time score and the importance score as the total task score of the network node task; The network node tasks are arranged in descending order according to the total task scores to obtain the task priorities.

6. The method for integrating power multi-grid resources according to claim 1, characterized in that: The resource allocation control according to the task priority and the current load of each network node includes: Determine target tasks in order from front to back according to the task priorities, and determine target demand load according to the task attributes of the target tasks; Determine a target network node corresponding to the target task, and determine a target idle load according to a maximum load of the target network node and a current load of the target network node; In response to the target idle load being greater than or equal to the target demand load, allocating the target task to the target network node and updating the current load of the target network node; In response to the target idle load being less than the target required load, dynamic load balancing control is performed among multiple networks according to the current task set of the target network node and the target required load.

7. The method for integrating power multi-grid resources according to claim 6, characterized in that: The dynamic load balancing control among multiple networks according to the current task set of the target network node and the target demand load includes: Determine a migratable task in the current task set according to the target demand load; Determine a node to be migrated with the largest idle load among the remaining nodes except the target node; The migratable task is migrated to the node to be migrated, the target task is allocated to the target node, and the current load of the target node and the current load of the node to be migrated are updated.

8. The method for integrating power multi-grid resources according to claim 7, characterized in that: The determining of the migratable tasks in the current task set according to the target demand load includes: Determine the task with the lowest priority in the current task set as the task to be migrated, and determine the sum of the required load of the task to be migrated and the target idle load as the available load; In response to the available load being greater than or equal to the target required load, determining the task to be migrated as the migratable task; In response to the available load being greater than or equal to the target required load, a new task to be migrated is determined according to the priority of each task in the task set until the available load is greater than or equal to the target required load, and all tasks to be migrated are determined as the migratable tasks.

9. A device for integrating multi-grid resources of electric power, characterized in that: include: The communication module is configured to: obtain the real-time communication characteristics of each network system, and determine the optimal path for the intelligent gateway to communicate with each network system according to the real-time communication characteristics; The perception module is configured to: determine a network weight coefficient of each network, obtain fused feature data of each network through the optimal path, and determine a comprehensive perception result according to the network weight coefficient and the fused feature data; The computing module is configured to: determine the task priority according to the comprehensive perception result and the task attribute of each network node task, and perform resource allocation control according to the task priority and the current load of each network node.

10. The power multi-grid resource fusion device according to claim 9, characterized in that: Also includes: An abnormality determination module is configured to: determine whether there is an abnormal event according to the comprehensive sensing result; The resource dynamic adjustment module is configured to: in response to the absence of abnormal events, dynamically adjust the communication resources, the sensing resources and the computing resources according to the urgency of the tasks of each network node and the preset constraints; The resource linkage module is configured to: in response to the existence of the abnormal event, determine the abnormal network system and abnormal node corresponding to the abnormal event, determine the target optimal path corresponding to the abnormal network system corresponding to the abnormal event, collect the task data in the abnormal node through the target optimal path, determine the target backup node of the abnormal node, and migrate the task data in the abnormal node to the target backup node.

11. The power multi-grid resource fusion device according to claim 9, characterized in that: The communication module comprises: The state monitoring submodule is configured to: determine the state vector of each network system according to the real-time communication characteristics; The path selection submodule is configured to: construct an objective function for each communication path according to a preset adjustment weight coefficient, and solve the objective function according to the state vector to obtain an objective function value; and determine the optimal path among multiple communication paths according to the objective function value, and dynamically update the optimal path according to the real-time communication characteristics.

12. The power multi-grid resource fusion device according to claim 9, characterized in that: The perception module comprises: The quantity determination submodule is configured to: determine the quantity of intermediate nodes; The single perception submodule is configured to: determine the product of the network weight coefficient and the corresponding fusion feature data as a single perception result; The comprehensive perception submodule is configured to: globally fuse the individual perception results according to the number of intermediate nodes to obtain the comprehensive perception result.

13. The power multi-grid resource fusion device according to claim 9, characterized in that: The computing module comprises: The parameter determination submodule is configured to: determine the real-time requirement parameters and importance requirement parameters of each network node task according to the comprehensive perception result; The score determination submodule is configured to: determine a real-time weight coefficient and an importance weight coefficient according to the task attribute, and determine the product of the real-time weight coefficient and the real-time requirement parameter as the real-time score, and determine the product of the importance requirement parameter and the importance weight coefficient as the importance score; The total score determination submodule is configured to: determine the sum of the real-time score and the importance score as the total task score of the network node task; The task sorting submodule is configured to: sort the network node tasks in descending order according to the total task scores to obtain the task priorities.

14. The power multi-grid resource fusion device according to claim 9, characterized in that: The computing module comprises: The demand load determination submodule is configured to: determine the target tasks in order from front to back according to the task priorities, and determine the target demand load according to the task attributes of the target tasks; The idle load determination submodule is configured to: determine a target network node corresponding to the target task, and determine a target idle load according to a maximum load of the target network node and a current load of the target network node; A first load distribution submodule is configured to: in response to the target idle load being greater than or equal to the target demand load, distribute the target task to the target network node and update the current load of the target network node; The second load distribution submodule is configured to: in response to the target idle load being less than the target required load, perform dynamic load balancing control among multiple networks according to the current task set of the target network node and the target required load.

15. The power multi-grid resource fusion device according to claim 14, characterized in that: The second load distribution submodule includes: A migration task screening unit is configured to: determine a migratable task in the current task set according to the target demand load; The migration node screening unit is configured to: determine the node to be migrated with the largest idle load among the remaining nodes except the target node; The load balancing unit is configured to: migrate the migratable task to the node to be migrated, allocate the target task to the target node, and update the current load of the target node and the current load of the node to be migrated.

16. The power multi-grid resource fusion device according to claim 15, characterized in that: The migration task screening unit includes: The load calculation subunit is configured to: determine the task with the lowest priority in the current task set as the task to be migrated, and determine the sum of the required load of the task to be migrated and the target idle load as the available load; A first task determination subunit is configured to: in response to the available load being greater than or equal to the target required load, determine the task to be migrated as the migratable task; The second task determination subunit is configured to: in response to the available load being greater than or equal to the target demand load, determine a new task to be migrated according to the priority of each task in the task set, until the available load is greater than or equal to the target demand load, and determine all the tasks to be migrated as the migratable tasks.

17. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

18. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 8.

Images