Communication network scheduling system and method for power system
By monitoring and classifying power system data in real time and optimizing communication resource allocation using annealing algorithm, the problems of inflexible resource allocation and insufficient scheduling strategies in traditional scheduling methods are solved, and the efficient and safe operation of the power system is achieved.
Patent Information
- Application Number
- CN202410872076.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-01
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-07-01
AI Technical Summary
The traditional power system communication network scheduling method lacks a comprehensive analysis of real-time power data and communication performance data, resulting in insufficient flexibility and efficiency in the allocation of communication resources, difficulty in adapting to complex and changing working environments, and lack real-time adjustment capabilities, making it difficult to deal with emergencies.
The information acquisition module is used to monitor the power system data in real time, and the data is divided into three states: normal, abnormal and fault through the information classification module. The strategy calculation module generates a scheduling strategy, uses annealing algorithm to optimize the task execution time, and adjusts the bandwidth and buffer according to the data state. Area size.
It realizes comprehensive analysis and accurate scheduling of power system data, quickly identify system conditions, dynamically adjust resource allocation, optimize communication resource utilization, avoid resource waste, and improve the operating efficiency and safety of power system.
Smart Images

Figure CN119011618A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of communication networks, and in particular to a communication network dispatching system and method for a power system. Background Art
[0002] In modern power systems, the dispatching of communication networks is crucial to ensuring the stability and efficiency of power supply. Traditional power system communication network dispatching methods often lack comprehensive analysis of real-time power data and communication performance data, resulting in inflexible and inefficient allocation of communication resources when power systems fail, thus affecting the overall operating efficiency and safety of the power systems. In addition, existing dispatching strategy optimization methods usually rely on fixed rules or simple algorithms, which are difficult to adapt to the complex and changeable working environment of power systems, resulting in insufficient real-time and accuracy of dispatching strategies.
[0003] In the prior art, publication number CN113890780A discloses a power communication scheduling method, which builds a communication network within a region based on a communication bus protocol, determines whether the communication service involves the transmission of key priority data, and sets a pre-reading threshold based on the computing power level of the communication system to determine which communication services need priority processing. When the progress of a communication service is not less than its pre-reading threshold, the service is added to the priority communication service group, and the expected transmission quality is determined by the transmission delay and packet loss rate of the service. All communication services in the priority communication service group are sorted according to the expected transmission quality, and the communication service with the lowest expected transmission quality is selected to start pre-reading.
[0004] The main problems with the above scheme are: setting the pre-reading threshold based on the computing power level requires an accurate assessment of the computing resources of the system in the entire area, otherwise it may cause the pre-reading operation to be overloaded and affect the overall performance; and the calculation of the expected transmission quality depends on the real-time delay and packet loss rate. The selection of these parameters should be based on the actual network conditions and business needs, and they change dynamically in actual situations. The above scheme lacks the ability to adjust in real time to the dynamic changes of the network and is difficult to respond to emergencies.
[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention
[0006] The object of the present invention is to provide a communication network dispatching system and method for an electric power system to solve the problems raised in the above background technology.
[0007] To achieve the above object, the present invention provides the following technical solutions:
[0008] A communication network dispatching system for a power system, specifically comprising:
[0009] An information collection module, used for real-time monitoring of power data and communication performance data of each terminal in the power system within an equal time period, normalizing the collected power data and communication performance data respectively to generate a first data set, wherein the power data includes current, voltage and power, and the communication performance data includes delay time, packet loss rate and throughput data volume;
[0010] An information classification module is used to collect power data and communication performance data of the power system, obtain a normal threshold and a fault threshold of the power system data, compare the first data set with the normal threshold and the fault threshold, classify all data into three states: normal, abnormal and fault, and generate a second data set, a third data set and a fourth data set respectively;
[0011] a strategy calculation module, configured to receive the second data set, the third data set and the fourth data set transmitted by the information classification module, allocate communication resources to the second data set according to the amount of resources just required to complete the task, reduce the communication resource allocation of the corresponding task according to the fluctuation of the data for the third data set, stop allocating communication resources for the fourth data set, generate a dispatch strategy after all communication data in the power system are allocated according to one strategy, and integrate all dispatch strategies to generate a strategy set;
[0012] The strategy optimization module is used to receive a strategy set, take minimizing the task execution time as the optimization goal, generate an objective function, randomly select a scheduling strategy from the strategy set as the initial solution, set an initial temperature of the annealing algorithm, and generate a cooling function. At the initial annealing temperature, the initial solution is searched in the neighborhood, and a new scheduling strategy is randomly selected from all adjacent scheduling strategies as the new solution. According to the Metropolis criterion, it is determined whether to accept the new solution. At the same time, the annealing temperature is gradually reduced according to the cooling function until the annealing termination condition is reached, and the scheduling strategy at this time is output;
[0013] The scheduling feedback module is used to receive the scheduling policy output by the policy optimization module and adjust the allocated bandwidth and buffer size according to the data type and the scheduling policy.
[0014] Furthermore, the formula for normalizing the power data is as follows:
[0015]
[0016] Among them, E0 represents the result of normalizing the power data, E represents the value of a type of power data currently collected, and E max Indicates the historical maximum value of the type of power data currently collected, E minIndicates the historical minimum value of the type of power data currently being collected. The types of power data include: current, voltage, and power;
[0017] The formula used to normalize the communication performance data is:
[0018]
[0019] Among them, L0 represents the result of normalizing the communication performance data, L represents a type of communication performance data currently collected, and L max Indicates the historical maximum value of the type of communication performance data currently collected, L min Indicates the historical minimum value of the communication performance data type currently collected. The communication performance data types include: delay, packet loss rate, and throughput.
[0020] Furthermore, the principle for classifying data into three states: normal, abnormal, and fault is as follows:
[0021] Analyze the collected power data and communication performance data to determine the normal fluctuation range of the power data and communication performance data when the power system operates normally;
[0022] The boundary point of the normal fluctuation range is set as the normal threshold;
[0023] Analyze the collected power data and communication performance data to determine the fluctuation range of the power data and communication performance data of the corresponding line when there is an interrupted line in the power system;
[0024] The data fluctuation range boundary point of the interrupted line is set as the fault threshold;
[0025] For real-time collected data, the principle for judging its current state is:
[0026] When S0≤S z When , the real-time data status is normal;
[0027] When S z ≤S0<S g When , the real-time data status is abnormal;
[0028] When S0 ≥ S g When , the real-time data status is faulty;
[0029] Among them, S0 represents the real-time collected power system data, S z Indicates the normal threshold of the corresponding data, S g represents the fault threshold of the corresponding data;
[0030] All the collected real-time power data and real-time and communication performance data are judged, all normal state data are extracted to generate a second data set, all abnormal state data are extracted to generate a third data set, and all fault state data are extracted to generate a fourth data set.
[0031] Furthermore, the principle for generating a scheduling strategy is:
[0032] For tasks corresponding to normal status data, bandwidth and buffer size that are just enough to complete the task execution are allocated;
[0033] For the tasks corresponding to the fault status data, no communication resources are allocated;
[0034] For the tasks corresponding to the abnormal state data, based on the task allocation strategy of the normal state data, a weight factor is generated for each data in the third data set, and the formula is as follows:
[0035]
[0036] D i =S ij -S z
[0037] D max =S g -S z
[0038]
[0039] Among them, V i represents the volatility of the i-th data, represents the historical average value of the i-th data, S ij represents the jth collection result of the i-th data, j represents the number of data collection, V max represents the maximum volatility, S z Indicates the normal threshold of the data, D i Indicates the degree of deviation of the i-th data from the normal state, D max Indicates the maximum deviation, S g represents the fault threshold of the data, α represents the influence ratio of volatility on the weight factor, β represents the influence ratio of deviation degree on the weight factor, and α+β=1, W i represents the weight factor of the i-th data in the third data set. By adjusting the values of α and β, different weight factors can be generated, thereby forming different scheduling schemes;
[0040] The formula for allocating communication resources according to the weight factor is:
[0041] P i=P base ×(1-W i )
[0042] Among them, P i Indicates the amount of resources allocated to the task corresponding to the i-th data, P base Indicates the amount of resources allocated in normal state.
[0043] Furthermore, the formula for generating the objective function is:
[0044]
[0045] Among them, T represents the execution time of a task, N represents the total number of tasks that need to be dispatched in a power system dispatch, and T exec (i) represents the computation time of the i-th task, T wait (i) represents the waiting time of the i-th task, T comm (i) represents the communication delay of the i-th task.
[0046] Furthermore, the formula for generating the cooling function is:
[0047] t i+1 =γt i
[0048] where t i represents the annealing temperature of the i-th iteration, t i+1 represents the annealing temperature of the (i+1)th iteration, and γ represents the temperature decay rate.
[0049] Furthermore, the formula for determining whether to accept a new solution according to the Metropolis criterion is:
[0050]
[0051] Among them, P represents the probability of accepting a new solution, F(i) represents the task execution time corresponding to the current scheduling strategy, F(i+1) represents the scheduling data corresponding to the next scheduling strategy, that is, the new solution, and t represents the annealing temperature. During the annealing process, the annealing temperature gradually decreases with the cooling function. At this time, the probability of taking a larger new solution is also gradually decreasing. When the annealing temperature drops to the lowest, the probability of taking a larger new solution is 0, and the annealing termination condition is met.
[0052] Furthermore, the principle for adjusting the allocated bandwidth and buffer size according to the data type and the scheduling strategy is as follows:
[0053] For data in normal state, computing resources are sufficient, and task computing time is mainly affected by task complexity, so there is no need to adjust bandwidth and buffer size. To reduce waiting time, increase bandwidth and buffer size. To reduce communication delay time, increase bandwidth size. In general, bandwidth and buffer size in normal state remain stable or increase slightly, and reserve resources for possible abnormal situations.
[0054] For data in abnormal states, in order to optimize task computing time, reduce bandwidth allocation for non-critical tasks and ensure computing resources for critical tasks; in order to reduce waiting time, reduce bandwidth for non-critical tasks and increase bandwidth and buffer size for critical tasks; in order to reduce communication delay time, reduce bandwidth for non-critical tasks and increase bandwidth for critical tasks; in general, in abnormal states, reduce bandwidth resource allocation for non-critical tasks and increase bandwidth and buffer resources for critical tasks, giving priority to ensuring execution time and stability of critical tasks;
[0055] The critical tasks are essential to the operation of the system and have execution time requirements. If they fail or are delayed, they will have a significant impact on the main functions or services of the system. The specific tasks involved are: system monitoring, security control, real-time data transmission, and tasks that need to be scheduled first in the specific scheduling process;
[0056] The non-critical tasks are not essential to the operation of the system. Even if they fail or are delayed, they will not have a serious impact on the main functional services of the system. The specific tasks involved are: data analysis, report generation, system maintenance, and tasks that do not need to be scheduled in priority during the specific scheduling process;
[0057] Stop allocating bandwidth and buffer for data in the fault state.
[0058] The present invention also provides a communication network dispatching method for a power system, the method is implemented by the above-mentioned communication network dispatching system for a power system, and the specific steps include:
[0059] Step 1: real-time monitoring of power data and communication performance data of each terminal in the power system within an equal time period, normalizing the collected power data and communication performance data respectively to generate a first data set, wherein the power data includes current, voltage and power, and the communication performance data includes delay time, packet loss rate and throughput data volume;
[0060] Step 2: Collect power data and communication performance data of the power system, obtain the normal threshold and fault threshold of the power system data, compare the first data set with the normal threshold and the fault threshold, divide all data into three states: normal, abnormal and fault, and generate the second data set, the third data set and the fourth data set respectively;
[0061] Step 3: receiving the second data set, the third data set and the fourth data set, allocating communication resources to the second data set according to the amount of resources just needed to complete the task, reducing the communication resource allocation of the corresponding task according to the fluctuation of the data for the third data set, and stopping allocating communication resources for the fourth data set, generating a dispatching strategy after allocating all communication data in the power system according to one strategy, and integrating all dispatching strategies to generate a strategy set;
[0062] Step 4: Receive the strategy set, take minimizing the task execution time as the optimization goal, generate the objective function, randomly select a scheduling strategy from the strategy set as the initial solution, set an initial temperature of the annealing algorithm, and generate a cooling function. At the initial annealing temperature, perform a neighborhood search on the initial solution, randomly select a new scheduling strategy from all adjacent scheduling strategies as the new solution, determine whether to accept the new solution according to the Metropolis criterion, and gradually reduce the annealing temperature according to the cooling function until the annealing termination condition is reached, and output the scheduling strategy at this time;
[0063] Step 5: Receive the scheduling policy output by the policy optimization module, and adjust the allocated bandwidth and buffer size according to the data type and the scheduling policy.
[0064] Compared with the prior art, the present invention has the following beneficial effects:
[0065] The present invention adopts a real-time information collection and classification mechanism, which can effectively monitor and analyze the power data and communication performance data of the power system, ensure the comprehensiveness and accuracy of the data basis for scheduling decisions, and can quickly identify the system situation by dividing the data into three states: normal, abnormal and fault. Different strategies are adopted to respond to data in different states, reducing the complexity while ensuring the coverage of the strategy set, and finding the scheduling strategy with the shortest task execution time in the strategy set through the annealing algorithm, and adjusting the bandwidth and buffer size of each task according to the scheduling strategy and data status; the present invention can effectively deal with abnormal and fault conditions that may occur at any time in the dynamic power system through real-time monitoring and dynamic adjustment, and reduce the resource allocation in abnormal state, avoid resource waste, and optimize the utilization rate of communication resources. BRIEF DESCRIPTION OF THE DRAWINGS
[0066] Figure 1 This is a schematic diagram of the system modules of an embodiment of the present invention;
[0067] Figure 2 The figure is a schematic diagram of the method flow of an embodiment of the present invention. DETAILED DESCRIPTION
[0068] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.
[0069] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include 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 may 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.
[0070] Example:
[0071] See also Figure 1 , the present invention provides a technical solution:
[0072] A communication network dispatching system for a power system, specifically comprising:
[0073] An information collection module, used for real-time monitoring of power data and communication performance data of each terminal in the power system within an equal time period, normalizing the collected power data and communication performance data respectively to generate a first data set, wherein the power data includes current, voltage and power, and the communication performance data includes delay time, packet loss rate and throughput data volume;
[0074] In this embodiment, the formula for normalizing the power data is as follows:
[0075]
[0076] Among them, E0 represents the result of normalizing the power data, E represents the value of a type of power data currently collected, and E max Indicates the historical maximum value of the type of power data currently collected, E min Indicates the historical minimum value of the type of power data currently being collected. The types of power data include: current, voltage, and power;
[0077] The formula used to normalize the communication performance data is:
[0078]
[0079] Among them, L0 represents the result of normalizing the communication performance data, L represents a type of communication performance data currently collected, and L maxIndicates the historical maximum value of the type of communication performance data currently collected, L min Indicates the historical minimum value of the communication performance data type currently collected.
[0080] An information classification module is used to collect power data and communication performance data of the power system, obtain a normal threshold and a fault threshold of the power system data, compare the first data set with the normal threshold and the fault threshold, classify all data into three states: normal, abnormal and fault, and generate a second data set, a third data set and a fourth data set respectively;
[0081] In this embodiment, the principle for classifying data into three states of normal, abnormal and fault is as follows:
[0082] Analyze the collected power data and communication performance data to determine the normal fluctuation range of the power data and communication performance data when the power system operates normally;
[0083] The boundary point of the normal fluctuation range is set as the normal threshold;
[0084] Analyze the collected power data and communication performance data to determine the fluctuation range of the power data and communication performance data of the corresponding line when there is an interrupted line in the power system;
[0085] The data fluctuation range boundary point of the interrupted line is set as the fault threshold;
[0086] For real-time collected data, the principle for judging its current state is:
[0087] When S0≤S z When , the real-time data status is normal;
[0088] When S z ≤S0<S g When , the real-time data status is abnormal;
[0089] When S0 ≥ S g When , the real-time data status is faulty;
[0090] Among them, S0 represents the real-time collected power system data, S z Indicates the normal threshold of the corresponding data, S g represents the fault threshold of the corresponding data;
[0091] All the collected real-time power data and real-time and communication performance data are judged, all normal state data are extracted to generate a second data set, all abnormal state data are extracted to generate a third data set, and all fault state data are extracted to generate a fourth data set.
[0092] a strategy calculation module, configured to receive the second data set, the third data set and the fourth data set transmitted by the information classification module, allocate communication resources to the second data set according to the amount of resources just required to complete the task, reduce the communication resource allocation of the corresponding task according to the fluctuation of the data for the third data set, stop allocating communication resources for the fourth data set, generate a dispatch strategy after all communication data in the power system are allocated according to one strategy, and integrate all dispatch strategies to generate a strategy set;
[0093] In this embodiment, the principle for generating a scheduling strategy is as follows:
[0094] For tasks corresponding to normal status data, bandwidth and buffer size that are just enough to complete the task execution are allocated;
[0095] For the tasks corresponding to the fault status data, no communication resources are allocated;
[0096] For the tasks corresponding to the abnormal state data, based on the task allocation strategy of the normal state data, a weight factor is generated for each data in the third data set, and the formula is as follows:
[0097]
[0098] D i =S ij -S z
[0099] D max =S g -S z
[0100]
[0101] Among them, V i represents the volatility of the i-th data, represents the historical average value of the i-th data, S ij represents the jth collection result of the i-th data, j represents the number of data collection, V max represents the maximum volatility, S z Indicates the normal threshold of the data, D i Indicates the degree of deviation of the i-th data from the normal state, D max Indicates the maximum deviation, S g represents the fault threshold of the data, α represents the influence ratio of volatility on the weight factor, β represents the influence ratio of deviation degree on the weight factor, and α+β=1, W i represents the weight factor of the i-th data in the third data set. By adjusting the values of α and β, different weight factors can be generated, thereby forming different scheduling schemes;
[0102] Weight factor W i It indicates that the i-th data in the third data set, relative to the data in the normal state in the first data set, the proportion of the communication resources allocated is reduced, reflecting the deviation of the data of the second data set from the normal state data, and the deviation is measured by the volatility and the degree of deviation, the standard deviation of the second data set reflects the fluctuation of the data, and the deviation between the real-time data in the second data set and the data threshold reflects the deviation, and the weight factor is proportional to the volatility and the degree of deviation of the second data set; the values of α and β are adjusted, and several weight factors are generated for one data, which in turn affects the allocation strategy, and the generated allocation strategy;
[0103] The formula for allocating communication resources according to the weight factor is:
[0104] P i =P base ×(1-W i )
[0105] Among them, P i Indicates the amount of resources allocated to the task corresponding to the i-th data, P base Indicates the amount of resources allocated under normal conditions;
[0106] P i It reflects the actual amount of resources that should be allocated for the data in the i-th second data set under the influence of the weight factor. It is generated by subtracting the reduced part affected by the weight factor compared with the amount of resources allocated under the normal state. The actual amount of resources allocated is proportional to the amount of resources allocated under the normal state and inversely proportional to the weight factor.
[0107] The strategy optimization module is used to receive a strategy set, take minimizing the task execution time as the optimization goal, generate an objective function, randomly select a scheduling strategy from the strategy set as the initial solution, set an initial temperature of the annealing algorithm, and generate a cooling function. At the initial annealing temperature, the initial solution is searched in the neighborhood, and a new scheduling strategy is randomly selected from all adjacent scheduling strategies as the new solution. According to the Metropolis criterion, it is determined whether to accept the new solution. At the same time, the annealing temperature is gradually reduced according to the cooling function until the annealing termination condition is reached, and the scheduling strategy at this time is output;
[0108] In this embodiment, the formula for generating the objective function is:
[0109]
[0110] Among them, T represents the execution time of a task, N represents the total number of tasks that need to be scheduled, and T exec (i) represents the computation time of the i-th task, T wait(i) represents the waiting time of the i-th task, T comm (i) represents the communication delay of the i-th task;
[0111] The task execution time T reflects a task that requires resource allocation. It is the sum of all the time for allocating resources, scheduling resources, data transmission, task waiting, and executing calculations. As these times increase, the task execution time will also increase.
[0112] Among them, T exec (i) represents the execution computing time, which is the time required for the task to be actually executed on the computing resources, including all the computing steps of data processing and algorithm execution in the task; T wait (i) represents the time that a task spends waiting for resource allocation and other tasks to complete, including the time for allocating resources, scheduling resources, and task waiting time. It can be obtained by calculating the difference between the task arrival time and the time when the task starts to execute; T comm (i) represents the communication delay time, which reflects the time of data transmission and is determined by the size of the data, network bandwidth and network congestion. Different scheduling strategies will generate different task execution times. The annealing algorithm is used to find the schedule with the shortest task execution time among all scheduling strategies.
[0113] In this embodiment, the formula for generating the cooling function is:
[0114] t i+1 =γt i
[0115] where t i represents the annealing temperature of the i-th iteration, t i+1 represents the annealing temperature of the (i+1)th iteration, γ represents the temperature decay rate, which is a constant less than 1 determined according to actual conditions and can be taken as 0.98. The cooling function reflects the rate of change of the annealing temperature as the number of iterations increases. The lower the temperature decay rate, the faster the rate of change of the annealing temperature.
[0116] In this embodiment, the formula for determining whether to accept a new solution according to the Metropolis criterion is:
[0117]
[0118] Among them, P represents the probability of accepting a new solution, T(i) represents the task execution time corresponding to the current scheduling strategy, T(i+1) represents the scheduling data corresponding to the next scheduling strategy, that is, the new solution, and t represents the annealing temperature. During the annealing process, the annealing temperature gradually decreases with the cooling function. At this time, the probability of taking a larger new solution is also gradually decreasing. When the annealing temperature drops to the lowest, the probability of taking a larger new solution is 0, and the annealing termination condition is met.
[0119] The probability of accepting a new solution is calculated using the Metropolis criterion in the simulated annealing algorithm, which is used to decide whether to accept a new solution, i.e., a new scheduling strategy. The simulated annealing algorithm is an optimization algorithm that searches the solution space by simulating the solid annealing process and gradually lowering the "temperature" according to the cooling function to find the global optimal solution or the approximate optimal solution.
[0120] If the task execution time T(i+1) of the new solution is less than the scheduling data T(i) of the current solution, that is, the new solution is better, then the probability of accepting the new solution is 1, which means that if the new solution is better than the current solution, the new solution is always accepted; if the scheduling data T(i+1) of the new solution is greater than or equal to the task execution time T(i) of the current solution, that is, the new solution is not better than the current solution, then the probability of accepting the new solution is
[0121] Here, t represents the annealing temperature, which is a control parameter that determines the "randomness" of the algorithm during the search process. When t is large, even if the new solution is not better than the current solution, there is a greater probability of accepting the new solution to prevent the algorithm from falling into the local optimal solution. As t decreases, the probability of accepting a worse solution will also decrease, and the algorithm will gradually tend to accept a better solution. As the annealing process proceeds, the temperature t gradually decreases according to the predetermined cooling function, resulting in a smaller and smaller probability of accepting a worse solution until the temperature drops to the minimum, and the algorithm no longer accepts any solution worse than the current solution. At this time, the termination condition of the annealing algorithm is met, and an approximate optimal solution is obtained.
[0122] A scheduling feedback module is used to receive the scheduling policy output by the policy optimization module and adjust the allocated bandwidth and buffer size according to the data type and the scheduling policy;
[0123] In this embodiment, the principle for adjusting the allocated bandwidth and buffer size according to the data type and scheduling strategy is as follows:
[0124] For data in normal state, computing resources are sufficient, and task computing time is mainly affected by task complexity, so there is no need to adjust bandwidth and buffer size. To reduce waiting time, increase bandwidth and buffer size. To reduce communication delay time, increase bandwidth size. In general, bandwidth and buffer size in normal state remain stable or increase slightly, and reserve resources for possible abnormal situations.
[0125] For data in abnormal states, in order to optimize task computing time, reduce bandwidth allocation for non-critical tasks and ensure computing resources for critical tasks; in order to reduce waiting time, reduce bandwidth for non-critical tasks and increase bandwidth and buffer size for critical tasks; in order to reduce communication delay time, reduce bandwidth for non-critical tasks and increase bandwidth for critical tasks; in general, in abnormal states, reduce bandwidth resource allocation for non-critical tasks and increase bandwidth and buffer resources for critical tasks, giving priority to ensuring execution time and stability of critical tasks;
[0126] The critical tasks are essential to the operation of the system and have execution time requirements. If they fail or are delayed, they will have a significant impact on the main functions or services of the system. Critical tasks require more computing resources, bandwidth, and buffers. The specific tasks involved are: system monitoring, security control, real-time data transmission, and tasks that need to be scheduled first in the specific scheduling process;
[0127] The non-critical tasks are not essential to the operation of the system. Even if they fail or are delayed, they will not have a serious impact on the main functional services of the system. The demand for resources is relatively small, and their resource allocation can be reduced when resources are tight. The specific tasks involved are: data analysis, report generation, system maintenance, and tasks that do not need to be scheduled in priority during the specific scheduling process;
[0128] Stop allocating bandwidth and buffer for data in the fault state.
[0129] See also Figure 2 The present invention also provides a communication network scheduling method for a power system, the method is executed by the above-mentioned communication network scheduling method for a power system, and the specific steps include:
[0130] Step 1: real-time monitoring of power data and communication performance data of each terminal in the power system within an equal time period, normalizing the collected power data and communication performance data respectively to generate a first data set, wherein the power data includes current, voltage and power, and the communication performance data includes delay time, packet loss rate and throughput data volume;
[0131] Step 2: Collect power data and communication performance data of the power system, obtain the normal threshold and fault threshold of the power system data, compare the first data set with the normal threshold and the fault threshold, divide all data into three states: normal, abnormal and fault, and generate the second data set, the third data set and the fourth data set respectively;
[0132] Step 3: receiving the second data set, the third data set and the fourth data set, allocating communication resources to the second data set according to the amount of resources just needed to complete the task, reducing the communication resource allocation of the corresponding task according to the fluctuation of the data for the third data set, and stopping allocating communication resources for the fourth data set, generating a dispatching strategy after allocating all communication data in the power system according to one strategy, and integrating all dispatching strategies to generate a strategy set;
[0133] Step 4: Receive the strategy set, take minimizing the task execution time as the optimization goal, generate the objective function, randomly select a scheduling strategy from the strategy set as the initial solution, set an initial temperature of the annealing algorithm, and generate a cooling function. At the initial annealing temperature, perform a neighborhood search on the initial solution, randomly select a new scheduling strategy from all adjacent scheduling strategies as the new solution, determine whether to accept the new solution according to the Metropolis criterion, and gradually reduce the annealing temperature according to the cooling function until the annealing termination condition is reached, and output the scheduling strategy at this time;
[0134] Step 5: Receive the scheduling policy output by the policy optimization module, and adjust the allocated bandwidth and buffer size according to the data type and the scheduling policy.
[0135] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0136] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.
[0137] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0138] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.
Claims
1. A communication network dispatching system for a power system, characterized in that: Specifically include: An information collection module, used for real-time monitoring of power data and communication performance data of each terminal in the power system within an equal time period, normalizing the collected power data and communication performance data respectively to generate a first data set, wherein the power data includes current, voltage and power, and the communication performance data includes delay time, packet loss rate and throughput data volume; An information classification module is used to collect power data and communication performance data of the power system, obtain a normal threshold and a fault threshold of the power system data, compare the first data set with the normal threshold and the fault threshold, classify all data into three states: normal, abnormal and fault, and generate a second data set, a third data set and a fourth data set respectively; a strategy calculation module, configured to receive the second data set, the third data set and the fourth data set transmitted by the information classification module, allocate communication resources to the second data set according to the amount of resources just required to complete the task, reduce the communication resource allocation of the corresponding task according to the fluctuation of the data for the third data set, stop allocating communication resources for the fourth data set, generate a dispatch strategy after all communication data in the power system are allocated according to one strategy, and integrate all dispatch strategies to generate a strategy set; The strategy optimization module is used to receive a strategy set, take minimizing the task execution time as the optimization goal, generate an objective function, randomly select a scheduling strategy from the strategy set as the initial solution, set an initial temperature of the annealing algorithm, and generate a cooling function. At the initial annealing temperature, the initial solution is searched in the neighborhood, and a new scheduling strategy is randomly selected from all adjacent scheduling strategies as the new solution. According to the Metropolis criterion, it is determined whether to accept the new solution. At the same time, the annealing temperature is gradually reduced according to the cooling function until the annealing termination condition is reached, and the scheduling strategy at this time is output; The scheduling feedback module is used to receive the scheduling policy output by the policy optimization module and adjust the allocated bandwidth and buffer size according to the data type and the scheduling policy.
2. A communication network dispatching system for a power system according to claim 1, characterized in that: In the information collection module, the formula for normalizing the power data is as follows: Among them, E0 represents the result of normalizing the power data, E represents the value of a type of power data currently collected, and E max Indicates the historical maximum value of the type of power data currently collected, E min Indicates the historical minimum value of the type of power data currently being collected. The types of power data include: current, voltage, and power; The formula used to normalize the communication performance data is: Among them, L0 represents the result of normalizing the communication performance data, L represents a type of communication performance data currently collected, and L max Indicates the historical maximum value of the type of communication performance data currently collected, L min Indicates the historical minimum value of the communication performance data type currently collected. The communication performance data types include: delay, packet loss rate, and throughput.
3. A communication network dispatching system for a power system according to claim 1, characterized in that: The principle for classifying data into three states of normal, abnormal and fault in the information classification module is as follows: Analyze the collected power data and communication performance data to determine the normal fluctuation range of power data and communication performance data when the power system is operating normally: The boundary point of the normal fluctuation range is set as the normal threshold; Analyze the collected power data and communication performance data to determine the fluctuation range of the power data and communication performance data of the corresponding line when there is an interrupted line in the power system; The data fluctuation range boundary point of the interrupted line is set as the fault threshold; For real-time collected data, the principle for judging its current state is: When S0≤S z When , the real-time data status is normal; When S z ≤S0<S g When , the real-time data status is abnormal; When S0 ≥ S g When , the real-time data status is faulty; Among them, S0 represents the real-time collected power system data, S z Indicates the normal threshold of the corresponding data, S g represents the fault threshold of the corresponding data; All the collected real-time power data and real-time and communication performance data are judged, all normal state data are extracted to generate a second data set, all abnormal state data are extracted to generate a third data set, and all fault state data are extracted to generate a fourth data set.
4. The communication network dispatching system for a power system according to claim 1, characterized in that: The principle on which a scheduling strategy is generated in the strategy calculation module is: For tasks corresponding to normal status data, bandwidth and buffer size that are just enough to complete the task execution are allocated; For the tasks corresponding to the fault status data, no communication resources are allocated; For the tasks corresponding to the abnormal state data, based on the task allocation strategy of the normal state data, a weight factor is generated for each data in the third data set, and the formula is as follows: D i =S ij -S z D max =S g -S z Among them, V i represents the volatility of the i-th data, represents the historical average value of the i-th data, S ij represents the jth collection result of the i-th data, j represents the number of data collection, V max represents the maximum volatility, S z Indicates the normal threshold of the data, D i Indicates the degree of deviation of the i-th data from the normal state, D max Indicates the maximum deviation, S g represents the fault threshold of the data, α represents the influence ratio of volatility on the weight factor, β represents the influence ratio of deviation degree on the weight factor, and α+β=1, W i represents the weight factor of the i-th data in the third data set. By adjusting the values of α and β, different weight factors can be generated, thereby forming different scheduling schemes; The formula for allocating communication resources according to the weight factor is: P i =P base ×(1-W i ) Among them, P i Indicates the amount of resources allocated to the task corresponding to the i-th data, P base Indicates the amount of resources allocated in normal state.
5. The communication network dispatching system for a power system according to claim 1, characterized in that: The formula for generating the objective function in the strategy optimization module is: Among them, T represents the execution time of a task, N represents the total number of tasks that need to be dispatched in a power system dispatch, and T exec (i) represents the computation time of the i-th task, T wait (i) represents the waiting time of the i-th task, T comm (i) represents the communication delay of the i-th task.
6. A communication network dispatching system for a power system according to claim 1, characterized in that: The formula for generating the cooling function in the strategy optimization module is: t i+1 =γt i where t i represents the annealing temperature of the i-th iteration, t i+1 represents the annealing temperature of the (i+1)th iteration, and γ represents the temperature decay rate.
7. The communication network dispatching system for a power system according to claim 1, characterized in that: In the strategy optimization module The formula for determining whether to accept a new solution according to the Metropolis criterion is: Among them, P represents the probability of accepting a new solution, F(i) represents the task execution time corresponding to the current scheduling strategy, F(i+1) represents the scheduling data corresponding to the next scheduling strategy, that is, the new solution, and t represents the annealing temperature. During the annealing process, the annealing temperature gradually decreases with the cooling function. At this time, the probability of taking a larger new solution is also gradually decreasing. When the annealing temperature drops to the lowest, the probability of taking a larger new solution is 0, and the annealing termination condition is met.
8. The communication network dispatching system for a power system according to claim 1, characterized in that: The principle for adjusting the allocated bandwidth and buffer size according to the data type and scheduling strategy in the scheduling feedback module is: For data in normal state, computing resources are sufficient, and task computing time is mainly affected by task complexity, so there is no need to adjust bandwidth and buffer size. To reduce waiting time, increase bandwidth and buffer size. To reduce communication delay time, increase bandwidth size. In general, bandwidth and buffer size in normal state remain stable or increase slightly, and reserve resources for possible abnormal situations. For data in abnormal states, in order to optimize task computing time, reduce bandwidth allocation for non-critical tasks and ensure computing resources for critical tasks; in order to reduce waiting time, reduce bandwidth for non-critical tasks and increase bandwidth and buffer size for critical tasks; in order to reduce communication delay time, reduce bandwidth for non-critical tasks and increase bandwidth for critical tasks; in general, in abnormal states, reduce bandwidth resource allocation for non-critical tasks and increase bandwidth and buffer resources for critical tasks, giving priority to ensuring execution time and stability of critical tasks; The critical tasks are essential to the operation of the system and have execution time requirements. If they fail or are delayed, they will have a significant impact on the main functions or services of the system. The specific tasks involved are: system monitoring, security control, real-time data transmission, and tasks that need to be scheduled first in the specific scheduling process; The non-critical tasks are not essential to the operation of the system. Even if they fail or are delayed, they will not have a serious impact on the main functional services of the system. The specific tasks involved are: data analysis, report generation, system maintenance, and tasks that do not need to be scheduled in priority during the specific scheduling process; Stop allocating bandwidth and buffer for data in the fault state.
9. A communication network dispatching method for a power system, characterized in that: The method is performed by the communication network dispatching system for a power system according to any one of claims 1 to 8, and includes: Step 1: real-time monitoring of power data and communication performance data of each terminal in the power system within an equal time period, normalizing the collected power data and communication performance data respectively to generate a first data set, wherein the power data includes current, voltage and power, and the communication performance data includes delay time, packet loss rate and throughput data volume; Step 2: Collect power data and communication performance data of the power system, obtain the normal threshold and fault threshold of the power system data, compare the first data set with the normal threshold and the fault threshold, divide all data into three states: normal, abnormal and fault, and generate the second data set, the third data set and the fourth data set respectively; Step 3: receiving the second data set, the third data set and the fourth data set, allocating communication resources to the second data set according to the amount of resources just needed to complete the task, reducing the communication resource allocation of the corresponding task according to the fluctuation of the data for the third data set, and stopping allocating communication resources for the fourth data set, generating a dispatching strategy after allocating all communication data in the power system according to one strategy, and integrating all dispatching strategies to generate a strategy set; Step 4: Receive the strategy set, take minimizing the task execution time as the optimization goal, generate the objective function, randomly select a scheduling strategy from the strategy set as the initial solution, set an initial temperature of the annealing algorithm, and generate a cooling function. At the initial annealing temperature, perform a neighborhood search on the initial solution, randomly select a new scheduling strategy from all adjacent scheduling strategies as the new solution, determine whether to accept the new solution according to the Metropolis criterion, and gradually reduce the annealing temperature according to the cooling function until the annealing termination condition is reached, and output the scheduling strategy at this time; Step 5: Receive the scheduling policy output by the policy optimization module, and adjust the allocated bandwidth and buffer size according to the data type and the scheduling policy.
Citation Information
Patent Citations
Electric power communication network scheduling method
CN113890780A
Power distribution network scheduling task optimization method and system
CN113822485A
Data center energy-saving regulation and control method and system based on AI optimization
CN116414214A
Automatic policy change management scheme for DiffServ-enabled MPLS networks
US20080247314A1
Method and system for allocating power resources of data center to microservices
WO2021008225A1
