A distribution network reliability optimization method and system
By monitoring the operating conditions of the distribution network and using genetic algorithms, optimization solutions are generated, and the problem of operating risks in the existing technology under high humidity and high load conditions is solved, and dynamic optimization and cost savings of the distribution network are achieved.
Patent Information
- Application Number
- CN202410080909.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-19
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-01-19
AI Technical Summary
When the existing distribution network optimization methods operate under high humidity and high load conditions, it is difficult to match targeted optimization solutions, resulting in increased operating risks, and a fixed optimization cycle is not easy to deal with dynamically changing environments and load conditions.
By monitoring the operating conditions of the distribution network, building operating condition coefficients and reliability coefficients, using the trained operating status model and genetic algorithm, an optimization solution is generated to reduce operating risks and achieve targeted optimization.
Quickly provide optimization plans to reduce operating risks, improve the reliability of the distribution network, reduce optimization costs, and adapt to dynamic environment and load changes.
Smart Images

Figure CN117910639B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power distribution network optimization, and in particular to a power distribution network reliability optimization method and system. Background Art
[0002] The distribution network refers to a power grid that receives electric energy from the transmission network or regional power plants and distributes it locally or step by step according to voltage to various users through distribution facilities. It is mainly composed of overhead lines, cables, towers, distribution transformers, disconnectors, reactive power compensators and some ancillary facilities, and plays an important role in distributing electric energy in the power grid.
[0003] Distribution networks are divided into high-voltage, medium-voltage and low-voltage distribution networks according to voltage levels. High-voltage distribution networks are usually 35-110kV, and 220kV can be used when the load density is large. Medium-voltage distribution networks are usually 6-10kV, and my country's medium-voltage distribution networks are mainly 10kV. Distribution networks play an important role in distributing electric energy in power grids. They have the characteristics of multiple voltage levels, complex network structures, diverse equipment types, multiple and wide operating points, and relatively poor safety environments. Therefore, there are relatively more safety risk factors in distribution networks.
[0004] In a Chinese invention patent with application publication number CN115378041A, a distribution network optimization method, system, distribution network, equipment and medium are disclosed. The method includes performing an injection power equivalent calculation on distributed energy and energy storage system in the distribution network to obtain a first operating parameter corresponding to the distributed energy and a second operating parameter corresponding to the energy storage system; setting multiple optimization targets corresponding to the distribution network according to the first operating parameter and the second operating parameter, as well as the topological structure of the distribution network; setting objective functions and / or constraints corresponding to the multiple optimization targets, solving the objective function according to the distribution network optimization algorithm to determine the distribution network optimization strategy, and adjusting the optimal configuration of the energy storage system according to the distribution network optimization strategy, wherein the distribution network optimization algorithm is used to solve the optimal solution of the objective function.
[0005] The optimization method recorded in the above application can comprehensively and systematically analyze the various factors affecting the optimization of the distribution network, and find the optimal solution from multiple solutions to ensure that the distribution network optimization strategy is the most consistent with the current distribution network optimization strategy. However, in addition to this, the existing optimization methods are usually based on fixed time intervals when optimizing the distribution network. However, if the air humidity in the area where the distribution network is located is high and its own operating load is also large, faults will occur during its operation, such as the number of interruptions and the duration of the interruptions. Under this condition, if the optimization is still to be performed at a fixed cycle, the operating risk of the distribution network will be greater, and it will not be easy to match a targeted optimization plan during the optimization.
[0006] To this end, the present invention provides a distribution network reliability optimization method and system. Summary of the invention
[0007] 1. Technical issues to be solved
[0008] In view of the deficiencies of the prior art, the present invention provides a distribution network reliability optimization method and system, which monitors the operating conditions of the distribution network and generates an operating condition data set from the monitoring data, thereby constructing the operating condition coefficient of the distribution network. If the operating condition coefficient exceeds the operating threshold, the reliability coefficient is generated from the working state data of the distribution network. If the reliability coefficient is lower than the state threshold, the distribution network is predicted using the trained distribution network operating state model, and a prediction data column is obtained. A correction coefficient is generated from the prediction data column. If the correction coefficient exceeds the correction threshold, an optimization model is obtained by genetic algorithm training, and the operation of the distribution network is optimized using the trained optimization model, and the corresponding optimization scheme is obtained, and the best scheme is screened out for output. One or more optimization schemes can be quickly given, and the operation of the distribution network can be improved in a targeted manner, and the operation risk can be reduced, thereby solving the problems in the background technology.
[0009] (II) Technical solution
[0010] To achieve the above objectives, the present invention is implemented through the following technical solutions:
[0011] A distribution network reliability optimization method, comprising:
[0012] Monitor the operating conditions of the distribution network and generate an operating condition data set from the monitoring data, and then construct the operating condition coefficient Fpx of the distribution network. If the operating condition coefficient Fpx exceeds the operating threshold, issue an evaluation instruction;
[0013] After receiving the evaluation command, the reliability coefficient Kes is generated by the working status data of the distribution network. If the reliability coefficient Kes is lower than the status threshold, a prediction command is issued externally;
[0014] Use the trained distribution network operation status model to make predictions on the distribution network, obtain the prediction data column, generate the correction coefficient Xes from the prediction data column, and issue an optimization instruction to the outside if the correction coefficient Xes exceeds the correction threshold;
[0015] After receiving the optimization instruction, the optimization model is obtained by genetic algorithm training, and the operation of the distribution network is optimized using the trained optimization model; the corresponding optimization scheme is obtained, and the best scheme is output according to the optimization coefficient Bo. The optimization coefficient Bo is obtained as follows: after linear normalization of the reliability coefficient Kes and the optimization cost Co, the corresponding data is mapped to the interval [0,1] according to the following formula:
[0016]
[0017] Weight coefficient: 0≤ρ≤1, 0≤θ≤1.
[0018] Furthermore, when the distribution network is in operation, the air humidity Rp in the environment where the distribution network is located and the operating load Rx of its operating equipment are monitored at the end of each monitoring cycle, and several groups of continuously acquired monitoring data are summarized to construct a distribution network operating condition data set.
[0019] Furthermore, the operating condition coefficient Fpx of the distribution network is generated from the operating condition data set. The specific method is as follows: after linear normalization of the air humidity Rp and the operating load Rx, the corresponding data are mapped to the interval [0,1], and then according to the following formula:
[0020]
[0021] Where i = 1, 2, ..., m, m is the number of monitoring cycles, is the qualified standard value of air humidity, It is the qualified standard value of the operating load; weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1.
[0022] Furthermore, after receiving the evaluation instruction, the number of interruptions Zc during the operation of the distribution network and the interruption duration Sc in each operation cycle are collected during the operation cycle, and several groups of data are aggregated to generate a working status data set in the distribution network; the reliability coefficient Kes of the distribution network is generated from the working status data set.
[0023] Furthermore, after receiving the prediction instruction, sample data of the distribution network is collected to train and obtain the distribution network operation status model, and several groups of prediction data continuously output by the distribution network operation status model are summarized to obtain the reliability coefficient Kes on each prediction node, and several reliability coefficients Kes are sorted in order according to the prediction nodes generated to generate a prediction data column.
[0024] Furthermore, the correction coefficient Xes is obtained as follows: obtain the difference between two adjacent correction coefficients in the prediction data column, and take the positive difference as the positive difference Tu, and the negative difference as the negative difference Cu; and then follow the following method:
[0025]
[0026] Weight coefficient: 0≤ζ≤1, 0≤χ≤1, i is the serial number of the negative difference Cu, i=1, 2, ..., k; is the mean of several negative differences, j is the serial number of the positive difference Tu, j = 1, 2, ..., p; is the mean of several positive differences.
[0027] Furthermore, the collected distribution network data is used to set algorithm parameters for the genetic algorithm based on the distribution network data, so as to construct an initial model by completing the parameter setting of the genetic algorithm, and the initial model is trained with the sample data to obtain the optimized model.
[0028] Furthermore, the constraints are defined according to the specification data of the distribution network; after setting the stopping criteria, the operation of the distribution network is optimized and iterated using the trained optimization model under the preset constraints until the stopping criteria are met, thereby obtaining an optimization solution that makes the reliability coefficient Kes exceed the state threshold.
[0029] Furthermore, the trained distribution network operation status model is used to simulate and analyze several output optimization schemes, and the corresponding analysis data is obtained. The corresponding reliability coefficient Kes and the optimization cost Co of each optimization scheme are generated again from the analysis data, thereby generating the corresponding optimization coefficient Bo.
[0030] A distribution network reliability optimization system, comprising:
[0031] A monitoring unit monitors the operating conditions of the distribution network and generates an operating condition data set from the monitoring data, thereby constructing an operating condition coefficient of the distribution network. If the operating condition coefficient exceeds an operating threshold, an evaluation instruction is issued;
[0032] The evaluation unit, after receiving the evaluation instruction, generates a reliability coefficient from the working status data of the distribution network, and if the reliability coefficient is lower than the status threshold, an external prediction instruction is issued;
[0033] The prediction unit uses the trained distribution network operation status model to make predictions on the distribution network, obtains the prediction data column, generates a correction coefficient from the prediction data column, and issues an optimization instruction to the outside if the correction coefficient exceeds the correction threshold;
[0034] After receiving the optimization instruction, the optimization unit obtains the optimization model through genetic algorithm training, optimizes the operation of the distribution network using the trained optimization model, obtains the corresponding optimization plan, and selects the best plan for output.
[0035] (III) Beneficial effects
[0036] The present invention provides a distribution network reliability optimization method and system, which have the following beneficial effects:
[0037] 1. Evaluate the operational reliability of the distribution network. If the operational reliability coefficient Kes of the distribution network is insufficient, it will have a significant negative impact on subsequent production and operation activities. If the current operating status of the distribution network fails to improve, it is necessary to determine the optimization time node and take optimization measures in a timely manner.
[0038] 2. Use the trained distribution network operation status model to predict the operation status of the distribution network. After obtaining several reliability coefficients Kes and corresponding prediction data columns, the next operation status of the distribution network can be evaluated and judged by obtaining the prediction data columns. If the operation status of the distribution network gradually improves, it is only necessary to maintain the current operation of the distribution network without additional optimization, which can reduce the operation cost.
[0039] 3. Based on the obtained prediction data column, a correction coefficient Xes is constructed. The operating status and operating trend of the distribution network can be evaluated based on the correction coefficient Xes. Therefore, a more accurate evaluation of the operating status of the distribution network can be made based on the obtained correction coefficient Xes, and it can be judged whether the distribution network needs to be optimized.
[0040] 4. Construct a trained optimization model based on the genetic algorithm, and after setting the optimization conditions and optimization targets, optimize the current working state of the distribution network to obtain several optimization schemes. Therefore, when the current operating state of the distribution network is difficult to meet expectations, and the next operating state is also difficult to meet expectations, one or more optimization schemes can be quickly given, which can improve the operation of the distribution network in a targeted manner and reduce operating risks.
[0041] 5. Based on the reliability of the scheme and the optimization cost, the optimization coefficient Bo is constructed. The target scheme with the largest optimization coefficient Bo is selected, and the second one is used as a backup scheme. This can maintain a high cost-effectiveness when optimizing the distribution network and save the optimization cost to the greatest extent. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 It is a schematic diagram of the flow chart of the method for optimizing the reliability of the distribution network of the present invention;
[0043] Figure 2 This is a schematic diagram of the structure of the distribution network reliability optimization system of the present invention. DETAILED DESCRIPTION
[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0045] See also Figure 1 The present invention provides a distribution network reliability optimization method, comprising:
[0046] Step 1: Monitor the operating conditions of the distribution network and generate an operating condition data set from the monitoring data, and then construct an operating condition coefficient Fpx of the distribution network. If the operating condition coefficient Fpx exceeds the operating threshold, issue an evaluation instruction;
[0047] Step 101: When the distribution network is in operation, determine the distribution area of each device in the distribution network, construct an electronic map covering the distribution area, and mark each device in the distribution network on the electronic map;
[0048] After setting the monitoring cycle, for example, one hour is used as a monitoring cycle, at the end of each monitoring cycle, the air humidity Rp in the environment where the distribution network is located and the operating load Rx of its operating equipment are monitored, and several groups of continuously acquired monitoring data are summarized to construct a distribution network operating condition data set;
[0049] Step 102: Generate the operating condition coefficient Fpx of the distribution network from the operating condition data set, and the specific method is as follows:
[0050] After linear normalization of the air humidity Rp and the operating load Rx, the corresponding data are mapped to the interval [0,1], and then according to the following formula:
[0051]
[0052] Where i = 1, 2, ..., m, m is the number of monitoring cycles, is the qualified standard value of air humidity, is the qualified standard value of the operating load; weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1; the weight coefficient is determined by referring to the hierarchical analysis method;
[0053] Combined with historical data and expected operation and management of the distribution network, the operation threshold is pre-built. If the operation condition coefficient Fpx exceeds the operation threshold, it means that the current operation status of the distribution network is poor and needs to be adjusted in time. Otherwise, the operation status of the distribution network may gradually deteriorate. At this time, an evaluation instruction is issued to the outside.
[0054] When using, combine the contents in steps 101 and 102:
[0055] When it is necessary to determine whether to optimize the distribution network, the operating status and operating environment of the distribution network are first monitored, and the operating condition coefficient Fpx is constructed based on the acquired monitoring data. If it exceeds expectations, it means that the operating status of the distribution network is poor. After a long period of continuous operation, the distribution network has accumulated more and more faults and is increasingly affected by the environment. Therefore, under this condition, it is necessary to make timely adjustments to the current operating status of the distribution network.
[0056] In the existing optimization methods, the optimization of the distribution network is usually carried out based on fixed time intervals. However, if the air humidity in the area where the distribution network is located is high and its own operating load is also large, faults will occur during its operation. For example, the number of interruptions and the duration of the interruptions will also increase. Under this condition, if the optimization is still carried out at a fixed period, the operating risk of the distribution network will be greater, and it will not be easy to match a targeted optimization plan during the optimization.
[0057] Step 2: After receiving the evaluation instruction, the reliability coefficient Kes is generated from the working status data of the distribution network. If the reliability coefficient Kes is lower than the status threshold, a prediction instruction is issued externally;
[0058] The step 2 includes the following contents:
[0059] Step 201: after receiving the evaluation instruction, set a number of operation cycles of equal length, collect the number of interruptions Zc during the operation of the distribution network and the interruption duration Sc in each operation cycle, that is, the sum of the duration of each interruption, aggregate several groups of data, and generate a working status data set in the distribution network;
[0060] Step 202: Generate the reliability coefficient Kes of the distribution network from the working status data set, and the specific method is as follows:
[0061] Perform linear normalization on the interruption times Zc and the interruption duration Sc, map the corresponding data to the interval [0,1], and then follow the following formula:
[0062]
[0063] Among them, k2 and k1 are weight coefficients, 0≤k1≤1, 0≤k2≤1, and k1+k2=1, is the tolerable value of the number of interruptions in each operating cycle, is the tolerable value of the interruption duration in each operation cycle, P i is the middle value of the evaluation in the ith operation cycle, is its mean; it should be noted that Zc i is the number of interruptions in the i-th operating cycle, Sc i is the interruption duration in the i-th operation cycle;
[0064] Based on historical data and management expectations for the distribution network, the status threshold is set in advance. If the reliability coefficient Kes is lower than the status threshold, it means that the current reliability of the distribution network is poor. If it cannot be handled in time, the probability of interruption risk or shutdown risk may be relatively high. At this time, a prediction instruction is issued to the outside to predict the operating status of the distribution network.
[0065] When using, combine the contents in steps 201 and 202:
[0066] After receiving the evaluation instruction, the operation reliability of the distribution network is evaluated based on the number and duration of distribution network interruptions when a fault occurs. If the operation reliability coefficient Kes of the distribution network is insufficient, it means that the risk of failure during the operation of the distribution network is high, which may have a greater negative impact on subsequent production and operation activities; if the current operating status of the distribution network fails to improve, it is necessary to determine the optimization time node and take optimization measures in a timely manner.
[0067] Step 3: Use the trained distribution network operation status model to make predictions for the distribution network, obtain the prediction data column, generate the correction coefficient Xes from the prediction data column, and if the correction coefficient Xes exceeds the correction threshold, issue an optimization instruction to the outside.
[0068] The step three includes the following contents:
[0069] Step 301: after receiving the prediction instruction, collect the topological structure data of the distribution network, such as lines, transformers, circuit breakers, fuses, etc.; collect the statistical data of the failure rate and repair time of the equipment; collect the load data of the equipment, including the load size, load mode, etc.; environmental data, such as temperature, humidity, etc.; summarize the collected data, perform data recognition on the collected data, obtain the corresponding data features, and construct a feature data set after summarizing;
[0070] The initial model is constructed using the BP neural network, and some data are extracted from the feature data set as training sets and test sets to train and test the initial model, and the trained initial model is used as the distribution network operation status model;
[0071] Step 302: Use the trained distribution network operation status model to make predictions for the distribution network. After setting equally spaced prediction nodes, continuously output several groups of prediction data to summarize and construct a prediction data set. Based on the prediction data in the prediction data set and in combination with historical data, obtain the reliability coefficient Kes on each prediction node, and sort the several reliability coefficients Kes in order according to the prediction nodes generated by them to generate a prediction data column.
[0072] When in use, after collecting data such as the operation and specifications of the distribution network, the distribution network operation status model is trained and obtained, and the operation status of the distribution network is predicted using the trained distribution network operation status model. After obtaining several reliability coefficients Kes and corresponding prediction data columns, the next operation status of the distribution network can be evaluated and judged by obtaining the prediction data columns. If the operation status of the distribution network gradually improves, it is only necessary to maintain the current operation of the distribution network without additional optimization, which can reduce operating costs.
[0073] Step 303, generating a correction coefficient Xes from the prediction data column, the specific method is as follows: obtain the difference between two adjacent correction coefficients in the prediction data column, and take the positive difference as the positive difference Tu, and the negative difference as the negative difference Cu; then follow the following method:
[0074]
[0075] Weight coefficient: 0≤ζ≤1, 0≤χ≤1, which can be obtained by the hierarchical analysis method, where i is the serial number of the negative difference Cu, i=1, 2, ..., k; is the mean of several negative differences, j is the serial number of the positive difference Tu, j = 1, 2, ..., p; is the mean of several positive differences;
[0076] A correction threshold is set in advance. If the correction coefficient Xes exceeds the correction threshold, it means that after the prediction, the operating status of the distribution network will still gradually deteriorate. If it cannot be optimized in time, the operating efficiency of the distribution network will gradually decrease until the frequency of interruptions becomes larger and the duration of interruptions becomes longer. At this time, an optimization instruction is issued to the outside.
[0077] When using, combine the contents in steps 301 to 303:
[0078] Based on the obtained prediction data column, a correction coefficient Xes is constructed. According to the correction coefficient Xes, the operating status and operating trend of the distribution network can be evaluated. Therefore, a more accurate evaluation of the operating status of the distribution network can be made based on the obtained correction coefficient Xes, and it can be judged whether the distribution network needs to be optimized.
[0079] Step 4: After receiving the optimization instruction, the optimization model is obtained by genetic algorithm training, the operation of the distribution network is optimized using the trained optimization model, the corresponding optimization plan is obtained, and the best plan is output;
[0080] The step 4 includes the following contents:
[0081] Step 401: using a genetic algorithm, according to the collected distribution network data, setting algorithm parameters for the genetic algorithm according to the distribution network data, such as population size, number of iterations, crossover rate and mutation rate, so as to complete the construction of an initial model by the genetic algorithm with parameter setting; training and testing the initial model by sample data, and using the trained initial model as an optimization model; defining constraint conditions according to the specification data of the distribution network, such as budget constraints, equipment capacity constraints, network topology constraints, etc.;
[0082] Step 402: after receiving the optimization instruction and setting the stop criteria, such as the maximum number of iterations or fitness convergence, the operation of the distribution network is optimized and iterated using the trained optimization model under preset constraints until the stop criteria are met, and an optimization solution is obtained that makes the reliability coefficient Kes exceed the state threshold;
[0083] When in use, after receiving the optimization instruction, a trained optimization model is constructed based on the genetic algorithm, and after setting the optimization conditions and optimization targets, the current working state of the distribution network is optimized, thereby obtaining several optimization schemes. Therefore, when the current operating state of the distribution network is difficult to meet expectations, and the next operating state is also difficult to meet expectations, one or more optimization schemes can be quickly given, which can improve the operation of the distribution network in a targeted manner and reduce operating risks.
[0084] Step 403: Use the trained distribution network operation status model to simulate and analyze the outputted optimization schemes, and obtain corresponding analysis data. Generate the corresponding reliability coefficient Kes and the optimization cost Co of each optimization scheme from the analysis data again, thereby generating the corresponding optimization coefficient Bo:
[0085] The specific method is as follows: After linear normalization of the reliability coefficient Kes and the optimization cost Co, the corresponding data is mapped to the interval [0,1], and then according to the following formula:
[0086]
[0087] Weight coefficient: 0≤ρ≤1, 0≤θ≤1, which can be obtained by referring to the hierarchical analysis method;
[0088] The optimization solution with the largest optimization coefficient Bo is taken as the target solution, and the target solution is output.
[0089] When using, combine the contents in steps 401 to 403:
[0090] On the basis of obtaining several optimization schemes, the optimization coefficient Bo is constructed based on the scheme reliability and optimization cost. The target scheme with the largest optimization coefficient Bo is taken as the target scheme, and the second one is taken as the backup scheme. This can maintain a high cost-effectiveness when optimizing the distribution network and save the optimization cost to the greatest extent.
[0091] See also Figure 2 The present invention provides a distribution network reliability optimization method and system, comprising:
[0092] A monitoring unit monitors the operating conditions of the distribution network and generates an operating condition data set from the monitoring data, thereby constructing an operating condition coefficient of the distribution network. If the operating condition coefficient exceeds an operating threshold, an evaluation instruction is issued;
[0093] The evaluation unit, after receiving the evaluation instruction, generates a reliability coefficient from the working status data of the distribution network, and if the reliability coefficient is lower than the status threshold, an external prediction instruction is issued;
[0094] The prediction unit uses the trained distribution network operation status model to make predictions on the distribution network, obtains the prediction data column, generates a correction coefficient from the prediction data column, and issues an optimization instruction to the outside if the correction coefficient exceeds the correction threshold;
[0095] After receiving the optimization instruction, the optimization unit obtains the optimization model through genetic algorithm training, optimizes the operation of the distribution network using the trained optimization model, obtains the corresponding optimization plan, and selects the best plan for output.
[0096] It should be noted that the hierarchical analysis method is a combination of qualitative and quantitative analysis methods. It can decompose complex problems into multiple levels. By comparing the importance of factors at each level, it can help decision makers make decisions on complex issues and determine the final decision plan. In this process, the hierarchical analysis method can be used to determine the weight coefficients of these indicators.
[0097] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0098] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0099] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0100] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0101] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0102] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0103] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.
[0104] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present 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. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
[0105] 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 distribution network reliability optimization method, characterized in that: include, The operating conditions of the distribution network are monitored and an operating condition data set is generated from the monitoring data, thereby constructing the operating condition coefficient Fpx of the distribution network. If the operating condition coefficient Fpx exceeds the operating threshold, the reliability coefficient Kes is generated from the working status data of the distribution network. If the reliability coefficient Kes is lower than the status threshold, the distribution network is predicted using the trained distribution network operating status model to obtain a prediction data column, and a correction coefficient Xes is generated from the prediction data column. If the correction coefficient Xes exceeds the correction threshold, an optimization model is obtained by genetic algorithm training, and the operation of the distribution network is optimized using the trained optimization model. Obtain the corresponding optimization plan, and select the best plan output according to the optimization coefficient Bo; The operating condition coefficient Fpx of the distribution network is generated from the operating condition data set. The specific method is as follows: After linear normalization of the air humidity Rp and the operating load Rx, the corresponding data are mapped to the interval [0,1], and then calculated according to the following formula: Where i = 1, 2, ..., m, m is the number of monitoring cycles, is the qualified standard value of air humidity, is the qualified standard value of the operating load; weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1; The correction coefficient Xes is obtained as follows: obtain the difference between two adjacent correction coefficients in the prediction data column, and take the positive difference as the positive difference Tu, and the negative difference as the negative difference Cu; then calculate it according to the following method: Weight coefficient: 0≤ζ≤1, 0≤χ≤1, i is the serial number of the negative difference Cu, i=1, 2, ..., k; is the mean of several negative differences, j is the serial number of the positive difference Tu, j = 1, 2, ..., p; is the mean of several positive differences; The optimization coefficient Bo is obtained as follows: After linear normalization of the reliability coefficient Kes and the optimization cost Co, the corresponding data is mapped to the interval [0,1] and calculated according to the following formula: Weight coefficient: 0≤ρ≤1, 0≤θ≤1; The reliability coefficient Kes of the distribution network is generated from the working status data set in the following way: Perform linear normalization on the interruption times Zc and the interruption duration Sc, map the corresponding data to the interval [0,1], and then calculate according to the following formula: Among them, k2 and k1 are weight coefficients, 0≤k1≤1, 0≤k2≤1, and k1+k2=1, is the tolerable value of the number of interruptions in each operating cycle, is the tolerable value of the interruption duration in each operation cycle, P i is the middle value of the evaluation in the ith operation cycle, is its mean; it should be noted that Zc i is the number of interruptions in the i-th operating cycle, Sc i is the interruption duration in the i-th operation cycle.
2. A distribution network reliability optimization method according to claim 1, characterized in that: When the distribution network is in operation, the air humidity Rp in the environment where the distribution network is located and the operating load Rx of its operating equipment are monitored at the end of each monitoring cycle, and several groups of continuously acquired monitoring data are summarized to construct a distribution network operating condition data set.
3. A distribution network reliability optimization method according to claim 1, characterized in that: After receiving the evaluation instruction, the number of interruptions Zc during the operation of the distribution network and the interruption duration Sc in each operation cycle are collected during the operation cycle, and several groups of data are aggregated to generate a working status data set in the distribution network; the reliability coefficient Kes of the distribution network is generated from the working status data set.
4. A distribution network reliability optimization method according to claim 1, characterized in that: After receiving the prediction instruction, sample data of the distribution network is collected to train and obtain the distribution network operation status model. Several groups of prediction data continuously output by the distribution network operation status model are summarized to obtain the reliability coefficient Kes on each prediction node, and several reliability coefficients Kes are sorted in order according to the prediction nodes generated to generate a prediction data column.
5. A distribution network reliability optimization method according to claim 1, characterized in that: The collected distribution network data is used to set algorithm parameters for the genetic algorithm based on the distribution network data, so as to build an initial model with the genetic algorithm having completed the parameter setting, and the initial model is trained with the sample data to obtain the optimized model.
6. A distribution network reliability optimization method according to claim 5, characterized in that: Define constraints based on the specification data of the distribution network; After setting the stopping criterion, the trained optimization model is used to optimize the operation of the distribution network under the preset constraints until the stopping criterion is met, and the optimization scheme that makes the reliability coefficient Kes exceed the state threshold is obtained.
7. A distribution network reliability optimization method according to claim 1, characterized in that: The trained distribution network operation status model is used to simulate and analyze several output optimization schemes, and the corresponding analysis data is obtained. The corresponding reliability coefficient Kes and the optimization cost Co of each optimization scheme are generated again from the analysis data, thereby generating the corresponding optimization coefficient Bo.
8. An optimization system for implementing the distribution network reliability optimization method according to any one of claims 1 to 7, characterized in that: include: A monitoring unit monitors the operating conditions of the distribution network and generates an operating condition data set from the monitoring data, thereby constructing an operating condition coefficient of the distribution network. If the operating condition coefficient exceeds an operating threshold, an evaluation instruction is issued; The evaluation unit, after receiving the evaluation instruction, generates a reliability coefficient from the working status data of the distribution network, and if the reliability coefficient is lower than the status threshold, an external prediction instruction is issued; The prediction unit uses the trained distribution network operation status model to make predictions on the distribution network, obtains the prediction data column, generates a correction coefficient from the prediction data column, and issues an optimization instruction to the outside if the correction coefficient exceeds the correction threshold; After receiving the optimization instruction, the optimization unit obtains the optimization model through genetic algorithm training, optimizes the operation of the distribution network using the trained optimization model, obtains the corresponding optimization plan, and selects the best plan for output.
