A capacitor compensation selection competitive optimization method and system

Abstract

The present application relates to the technical field of capacitor optimization, and provides a method and system for competitive optimization of capacitor compensation selection. The method includes: collecting power from the load of the target circuit to generate power characteristic information; analyzing the compensation capacity of the power characteristic information to generate the target compensation capacity; configuring three-phase co-compensation based on the target compensation capacity to generate a compensation memory library; taking phase compensation as the goal and searching with the target compensation capacity as the index to generate a compensation database; performing mutation cross iteration based on the compensation memory library and the compensation database to obtain the target compensation scheme; and compensating the target circuit based on the capacitor compensation selection in the target compensation scheme. The present application solves the technical problem that the traditional capacitor compensation method cannot effectively deal with the unbalanced three-phase load condition, resulting in unsatisfactory compensation effect, and realizes the optimization of capacitor compensation selection, improves the power factor of the power system, and reduces the effect of reactive power loss.

Description

Technical Field

[0001] The present application relates to the field of power electronics technology, specifically to the field of capacitor optimization technology, and in particular to a method and system for competitive optimization of capacitor compensation selection. Background Art

[0002] With the continuous advancement of power electronics technology and the increasing complexity of power system structures, the compensation role of capacitors in power systems has become increasingly critical. However, traditional capacitor compensation methods often find it difficult to provide effective compensation solutions when faced with unbalanced three-phase loads, resulting in unsatisfactory compensation effects, which in turn affects the power factor of the power system and increases reactive power losses. Three-phase load imbalance is a common problem in power systems, and its causes are diverse, such as asymmetric operation of equipment and uneven load distribution. When the three-phase load is unbalanced, traditional capacitor compensation methods can usually only provide a unified compensation solution, and cannot accurately compensate for the specific conditions of each phase, thus affecting the compensation effect. Summary of the invention

[0003] The present application provides a capacitor compensation selection competitive optimization method and system, aiming to solve the technical problem that the traditional capacitor compensation method cannot effectively cope with the three-phase load imbalance condition, resulting in unsatisfactory compensation effect.

[0004] In view of the above problems, the present application provides a method and system for competitive optimization of capacitor compensation selection.

[0005] The first aspect disclosed in the present application provides a competitive optimization method for capacitor compensation selection, the method comprising: using an intelligent measurement and control unit to collect power from a load of a target circuit to generate multiple power characteristic information of multiple loads; performing compensation capacity analysis on the multiple power characteristic information to generate a target compensation capacity; performing three-phase co-compensation configuration based on the target compensation capacity to generate an initial compensation memory library, wherein the initial compensation memory library includes multiple three-phase co-compensation schemes; taking phase compensation as the goal, searching with the target compensation capacity as an index to generate a historical phase compensation database, wherein the historical phase compensation database includes multiple historical phase compensation schemes; performing variation crossover iteration based on the initial compensation memory library and the historical phase compensation database until a preset number of cycles is met to obtain a target compensation scheme; and compensating the target circuit based on the capacitor compensation selection in the target compensation scheme.

[0006] Another aspect disclosed in the present application provides a capacitor compensation selection competition optimization system, the system comprising: a power acquisition module, the power acquisition module is used to use an intelligent measurement and control unit to collect power from a load of a target circuit, and generate multiple power characteristic information of multiple loads; a compensation capacity analysis module, the compensation capacity analysis module is used to perform compensation capacity analysis on the multiple power characteristic information, and generate a target compensation capacity; a three-phase co-compensation configuration module, the three-phase co-compensation configuration module is used to perform three-phase co-compensation configuration based on the target compensation capacity, and generate an initial compensation memory library, wherein the initial compensation memory library includes multiple three-phase co-compensation schemes; a retrieval module, the retrieval module is used to take phase compensation as a target, and use the target compensation capacity as an index to perform retrieval, and generate a historical phase compensation database, wherein the historical phase compensation database includes multiple historical phase compensation schemes; a variation cross iteration module, the variation cross iteration module is used to perform variation cross iteration based on the initial compensation memory library and the historical phase compensation database, until a preset number of cycles is met to obtain a target compensation scheme; a compensation module, the compensation module is used to compensate the target circuit based on the capacitor compensation selection in the target compensation scheme.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The above-mentioned capacitor compensation selection competition optimization method collects and records the power characteristic information of multiple loads in the circuit, which describes the power usage of each load in detail. Subsequently, the required target compensation capacity is calculated and determined by analyzing the power characteristic information. After the target compensation capacity is determined, the three-phase co-compensation configuration is performed according to this capacity, and an initial compensation memory library containing multiple possible three-phase co-compensation configurations is generated. After that, with phase compensation as the goal and the target compensation capacity as the index, the relevant phase compensation schemes are retrieved from the historical database, which have been successfully applied under similar conditions in the past, and a historical phase compensation database is formed. After obtaining the initial compensation memory library and the historical phase compensation database, a series of optimization iterations are performed. This process combines strategies such as mutation and crossover to find the optimal compensation scheme. This process will be repeated until the preset number of cycles is met and the target compensation scheme is obtained. Finally, according to the capacitor compensation selection in the obtained target compensation scheme, the target circuit is compensated to achieve the purpose of improving the power factor, reducing reactive power loss, and improving the power quality.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A schematic diagram of a flow chart of a capacitor compensation selection competition optimization method in one embodiment;

[0012] Figure 2 A diagram of a system architecture for competitive optimization of capacitor compensation selection in one embodiment.

[0013] Explanation of the accompanying drawings: power collection module 1, compensation capacity analysis module 2, three-phase co-compensation configuration module 3, retrieval module 4, variation cross iteration module 5, compensation module 6. DETAILED DESCRIPTION

[0014] The embodiments of the present application provide a capacitor compensation selection competitive optimization method and system to solve the technical problem that the traditional capacitor compensation method cannot effectively deal with the three-phase load imbalance condition, resulting in unsatisfactory compensation effect.

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0017] Embodiment 1

[0018] like Figure 1 As shown, the present application provides a capacitor compensation selection competition optimization method, the method comprising:

[0019] The intelligent measurement and control unit is used to collect power from the load of the target circuit and generate multiple power characteristic information of multiple loads.

[0020] In power systems, three-phase load imbalance is a common and important problem. Traditional capacitor compensation methods are often not ideal in dealing with this problem because they can usually only provide a unified compensation solution and cannot accurately compensate for the specific conditions of each phase. Using the capacitor compensation selection competition optimization method, the imbalance of the three-phase load can be monitored in real time, and an accurate compensation solution can be provided for each phase through an intelligent algorithm, thereby optimizing the compensation effect.

[0021] In an embodiment of the present application, in the optimization and management of the power system, a key step is to understand and evaluate the power usage of different loads in the circuit. In order to achieve this goal, the system terminal uses an intelligent measurement and control unit to accurately collect the power of the load of the target circuit. The intelligent measurement and control unit can monitor and record the power data of each load in the circuit in real time, including key parameters such as voltage and power factor. Through this process, multiple power characteristic information of multiple loads is obtained. This information reflects in detail the power consumption of each load at different times and under different working conditions, and provides data support for subsequent capacitor compensation.

[0022] Furthermore, the present application provides power characteristic information, including:

[0023] The plurality of power characteristic information includes a plurality of measured power factors, a plurality of load powers, and a plurality of load voltages.

[0024] Preferably, the intelligent measurement and control unit can monitor the working status of multiple loads in the target circuit in real time and collect multiple power characteristic information about these loads. This information includes multiple measured power factors, multiple load powers, and multiple load voltages. Among them, the measured power factor reflects the ratio between active power and apparent power when the load consumes electric energy, which helps the system terminal understand the power utilization efficiency of the load. Multiple load powers reflect the power consumption of each load within a specific time, which is crucial for evaluating the energy consumption level of the target circuit. Multiple load voltage data reveal the voltage distribution in the target circuit, which is critical for ensuring the normal operation of the equipment and avoiding damage caused by excessive or low voltage. In summary, the intelligent measurement and control unit provides the system terminal with multiple key power characteristic information of each load in the target circuit, which is the basis for evaluating the circuit operation status, optimizing the capacitor compensation selection, and other work.

[0025] Compensation capacity analysis is performed on the plurality of power characteristic information to generate a target compensation capacity.

[0026] In one embodiment, after obtaining multiple power characteristic information of multiple loads, the system terminal starts to perform compensation capacity analysis. Compensation capacity analysis is a comprehensive evaluation of these power characteristic information to determine the capacitor compensation capacity required to achieve ideal power conditions, such as improving power factor, reducing reactive power loss, etc. This analysis process is based on the trained compensation capacity identification network layer. Through compensation capacity analysis, the compensation capacity identification network layer generates a target compensation capacity, which is the basis for subsequent capacitor compensation selection. The determination of the target compensation capacity is intended to ensure that the capacitor can accurately compensate for the reactive power in the target circuit and improve the overall performance of the power system.

[0027] Furthermore, the present application provides a method for obtaining the target compensation capacity, the method further comprising:

[0028] Collect multiple sample measured power factors, multiple sample load powers, multiple sample load voltages and multiple sample compensation capacities as construction data; use the construction data for supervised training to generate a compensation capacity identification network layer; use the compensation capacity identification network layer to identify the multiple measured power factors, multiple load powers and the multiple load voltages to obtain the target compensation capacity.

[0029] Optionally, the system terminal collects a series of construction data through the intelligent measurement and control unit, including multiple sample measured power factors, multiple sample load powers, multiple sample load voltages, and multiple sample compensation capacities. These data are the basis for constructing the compensation capacity identification network layer. Once sufficient construction data is collected, the system terminal initializes a neural network and sets the number of network layers, the number of neurons in each layer, the activation function, the loss function, etc. according to the complexity of the problem and the characteristics of the data. Subsequently, the obtained construction data is divided into a training set, a validation set, and a test set. The training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the generalization ability of the model. Then, multiple measured power factors, multiple load powers, and multiple load voltages in the training set are passed through the neural network to calculate the output value predicted compensation capacity. After that, the predicted output value is compared with the sample compensation capacity, and the loss function value is calculated by the mean square error. Then, according to the loss function value, the error is back-propagated and the network weights and biases are updated through the gradient descent algorithm. Then, the process of forward propagation, loss calculation, and back-propagation is repeated until the preset number of iterations is reached. Finally, the validation set is used to evaluate the network obtained during the training process to check the performance of the network on unseen data. According to the validation results, the network parameters or network architecture are adjusted, such as increasing or decreasing the number of network layers, changing the number of neurons, adjusting the learning rate, etc. The final trained network is then tested with the test set to evaluate the generalization ability and accuracy of the network. If the test results meet the requirements, the trained compensation capacity identification network layer is output. If not, it is necessary to readjust the network parameters or select a more suitable network architecture for training. Further, the system terminal inputs the actual multiple measured power factor, load power, and load voltage data into the compensation capacity identification network layer. The network layer uses the previously learned knowledge to identify these data and outputs the corresponding compensation capacity prediction value, that is, the target compensation capacity. In summary, this process is to build a compensation capacity identification network layer by collecting key data and conducting supervised training, and use this network layer to predict the target compensation capacity. This method can improve the accuracy and efficiency of subsequent capacitor compensation selection and help optimize the performance of the power system.

[0030] A three-phase co-compensation configuration is performed based on the target compensation capacity to generate an initial compensation memory bank, wherein the initial compensation memory bank includes a plurality of three-phase co-compensation schemes.

[0031] In one embodiment, after the target compensation capacity is determined, the system terminal performs three-phase co-compensation configuration. This process is to allocate appropriate compensation amount to each phase in the three-phase circuit according to the target compensation capacity to ensure the stable operation of the target circuit. When performing three-phase co-compensation configuration, the system terminal first determines the basic strategy of three-phase co-compensation according to the specific structure and operation status of the target circuit. This includes deciding whether to adopt static compensation or dynamic compensation, whether to consider the volatility of the load, etc. Among them, static compensation is suitable for occasions with small load changes and low requirements for compensation speed; dynamic compensation is suitable for occasions with large load changes and high requirements for compensation speed. Subsequently, based on the target compensation capacity and the three-phase co-compensation strategy, multiple candidate three-phase co-compensation schemes are generated. These schemes cover different capacitor configuration methods, quantities, connection positions, etc. Afterwards, the generated candidate schemes are sorted to form an initial compensation memory library. Each scheme in this memory library specifies in detail the compensation amount that should be configured for each phase, as well as the type, quantity and connection method of capacitors that may be required, so as to select and adjust in the subsequent optimization process.

[0032] Taking phase compensation as a target and taking the target compensation capacity as an index for searching, a historical phase compensation database is generated, wherein the historical phase compensation database includes a plurality of historical phase compensation schemes.

[0033] In one embodiment, the system terminal takes phase compensation as a target. When it is necessary to determine a phase compensation scheme for a specific target compensation capacity, the system terminal uses this target compensation capacity as an index to search the historical database. The historical database includes phase compensation schemes that have been implemented in the past and their corresponding compensation capacities, implementation effects and other information. Subsequently, the system terminal constructs a historical phase compensation database based on the search results, which stores multiple historical phase compensation schemes and corresponding index values. Afterwards, the system terminal can quickly find matching or similar historical compensation schemes from the historical phase compensation database through the target compensation capacity, helping the system terminal to determine a suitable target compensation scheme more quickly.

[0034] Based on the initial compensation memory library and the historical phase compensation database, mutation crossover iteration is performed until a preset number of cycles is met to obtain a target compensation solution.

[0035] In one embodiment, the system terminal uses a mutation cross iteration method to optimize and search for the target compensation scheme based on the initial compensation memory and the historical phase compensation database. Specifically, the initial compensation memory provides a variety of possible three-phase co-compensation schemes, while the historical phase compensation database contains a collection of historical phase compensation schemes. The system terminal first numbers the schemes in the two databases to obtain multiple three-phase co-compensation scheme numbers and multiple historical phase compensation scheme numbers. Subsequently, multiple three-phase co-compensation scheme numbers are used as starting points, and a new scheme is generated by combining multiple historical phase compensation scheme numbers for mutation and cross operations. When performing mutation cross iteration, the system terminal updates the three-phase co-compensation scheme number according to a preset update rule, and stores the update result in the updated compensation memory. Afterwards, the system terminal randomly extracts and analyzes the updated compensation memory to obtain the stage updated three-phase co-compensation scheme number, and this extraction and analysis process ends at a preset number of cycles. Then, the one with the largest fitness is extracted from multiple updated three-phase co-compensation scheme numbers as the target compensation scheme, which is the optimal solution to meet the compensation requirements of the target circuit.

[0036] Further, the present application provides performing mutation crossover iteration based on the initial compensation memory library and the historical phase compensation database, and the method further includes:

[0037] Respectively number the multiple three-phase co-compensation schemes in the initial compensation memory bank and the multiple historical phase-splitting compensation schemes in the historical phase-splitting compensation database to obtain multiple three-phase co-compensation scheme numbers and multiple historical phase-splitting compensation scheme numbers; traverse the multiple three-phase co-compensation scheme numbers and the multiple historical phase-splitting compensation scheme numbers to perform compensation fitness evaluation, and generate multiple three-phase co-compensation scheme fitnesses and multiple historical phase-splitting compensation scheme fitnesses.

[0038] Preferably, in order to find the most suitable compensation scheme for the current target circuit from the initial compensation memory and the historical phase compensation database, the system terminal numbers and evaluates the fitness of the compensation schemes in the two. Specifically, the system terminal assigns a unique number to each three-phase co-compensation scheme in the initial compensation memory. This helps to track, manage and evaluate these schemes in the future. Similarly, a unique number is also assigned to each historical phase compensation scheme in the historical phase compensation database. These numbers will be used to quickly retrieve and reference past compensation schemes. Subsequently, the system terminal specifies the standards and indicators for compensation fitness evaluation. These indicators include the degree of improvement of the target circuit performance such as power factor improvement, load power reduction, and load voltage stability improvement. Then, according to the specific needs and goals of the target circuit, a weight is assigned to each indicator to ensure the accuracy of the evaluation results. After that, all three-phase co-compensation scheme numbers and historical phase compensation scheme numbers are traversed. For each number, the corresponding compensation scheme information is retrieved from the initial compensation memory or the historical phase compensation database. Then, the power simulation component is used to apply each compensation scheme to the simulation environment of the target circuit. The target circuit is run in the simulation environment and data on relevant performance indicators are collected. Then, all the collected performance indicator data are analyzed to obtain the minimum and maximum values ​​of each performance indicator. The difference between the current value and the minimum value is then ratioed to the difference between the maximum value and the minimum value to obtain the standardized indicator value. For each scheme, the system terminal performs a weighted summation of all the performance indicators of this scheme according to the assigned weights to obtain the fitness of this scheme. By repeating the above process, the system terminal finally obtains the fitness of multiple three-phase co-compensation schemes and the fitness of multiple historical phase-splitting compensation schemes. These fitness can help the system terminal fully understand the performance of each compensation scheme in the target circuit and provide support for the subsequent acquisition of the target compensation scheme.

[0039] A random number generator is used to perform mutation cross iteration on the multiple three-phase co-compensation scheme numbers according to the preset update rules and the multiple historical phase-separated compensation scheme numbers until the preset number of cycles is met to obtain the target compensation scheme.

[0040] Preferably, the system terminal uses a random number generator and preset update rules to mutate and cross-iterate multiple three-phase co-compensation scheme numbers. Specifically, the system terminal generates two random numbers through a random number generator, and by comparing the size relationship between the two random numbers, determines whether to directly obtain the target compensation scheme or to randomly sort multiple historical phase compensation scheme numbers. If it is determined to obtain the target compensation scheme, the system terminal will perform mutation crossover according to the preset update rules to generate an updated compensation memory library. Subsequently, according to the preset number of cycles, the updated compensation memory library is iteratively extracted, and the fitness values ​​of the two randomly extracted each time are compared, and the corresponding stage update three-phase co-compensation scheme number generation operation is performed according to the judgment result. Afterwards, the corresponding three-phase co-compensation scheme with the largest fitness among all the generated stage update three-phase co-compensation scheme numbers is extracted as the target compensation scheme.

[0041] Furthermore, the present application provides the preset update rule, and the method further includes:

[0042] The preset update rule is that when the fitness of the first historical phase-splitting compensation scheme number extracted from the multiple historical phase-splitting compensation scheme numbers is greater than or equal to the fitness of the first three-phase co-compensation scheme corresponding to the first three-phase co-compensation scheme number at the corresponding number position of the multiple three-phase co-compensation scheme numbers, the first three-phase co-compensation scheme number is replaced with the first historical phase-splitting compensation scheme number, and multiple updated three-phase co-compensation scheme numbers are obtained, and the updated multiple three-phase co-compensation scheme numbers are used as the updated compensation memory library.

[0043] Optionally, the update rule preset by the system terminal is an optimization process for improving the performance of the three-phase co-compensation scheme. Specifically, in each iteration, the system terminal randomly selects one from the historical phase-splitting compensation scheme numbers as the first historical phase-splitting compensation scheme number, and extracts the fitness value corresponding to the first historical phase-splitting compensation scheme number. Subsequently, the fitness of this historical scheme is compared with the fitness of the first three-phase co-compensation scheme corresponding to the corresponding position in the multiple three-phase co-compensation scheme numbers, that is, the first three-phase co-compensation scheme number with the same number. If the fitness of the historical phase-splitting compensation scheme is greater than or equal to the fitness of the first three-phase co-compensation scheme, the system terminal replaces the first three-phase co-compensation scheme number with the first historical phase-splitting compensation scheme number. In this way, an updated multiple three-phase co-compensation scheme numbers are obtained. As the number of iterations increases, the system terminal constructs an updated compensation memory based on the updated multiple three-phase co-compensation scheme numbers. The updated compensation memory stores schemes with higher fitness.

[0044] Furthermore, the present application provides a scheme for obtaining target compensation, and the method further comprises:

[0045] A first update three-phase co-compensation scheme number is randomly extracted from the update compensation memory bank to obtain a first update fitness; and a second update three-phase co-compensation scheme number is randomly extracted from the update compensation memory bank to obtain a second update fitness.

[0046] Optionally, after obtaining the updated compensation memory library, the system terminal randomly selects two schemes from the updated compensation memory library for evaluation. Specifically, the system terminal first randomly extracts a three-phase co-compensation scheme number from the updated compensation memory library as the first updated three-phase co-compensation scheme number, and then extracts the fitness value corresponding to the number as the first updated fitness. Subsequently, the number of another three-phase co-compensation scheme is randomly extracted from the updated compensation memory library as the second updated three-phase co-compensation scheme number, and the corresponding fitness value is also extracted as the second updated fitness.

[0047] Determine whether the second update fitness is greater than the first update fitness, and if so, use the second update three-phase co-compensation scheme number as the stage update three-phase co-compensation scheme number;

[0048] If not, then accept the second updated three-phase co-compensation scheme number as the stage updated three-phase co-compensation scheme number according to a certain probability;

[0049] After reaching the preset number of iterations, the updated three-phase co-compensation scheme corresponding to the maximum updated fitness value in the iteration process is used as the target compensation scheme.

[0050] Optionally, in each iteration, the size relationship between the first updated fitness and the second updated fitness is compared. If the second updated fitness is greater than the first updated fitness, it means that the second solution is better than the first solution in performance, so the system terminal directly selects the second updated three-phase co-compensation solution number as the updated three-phase co-compensation solution number of the current stage. However, if the second updated fitness is not greater than the first updated fitness, the system terminal does not directly eliminate the second solution, but introduces a random acceptance mechanism. This mechanism allows the system terminal to accept the second updated three-phase co-compensation solution number as the stage updated three-phase co-compensation solution number with a certain probability, even if its fitness value is low. The purpose of this is to increase the diversity of the search and prevent the algorithm from falling into a local optimal solution. In the search space, there may be multiple local optimal solutions, and randomly accepting a poorer solution helps the algorithm jump out of the current local optimal area and continue to explore other possible better solutions. The entire iterative process will be repeated until the preset number of iterations is reached. In each iteration, the system terminal updates the current optimal fitness value and the corresponding updated three-phase co-compensation solution number. Finally, after the iteration, the system terminal selects the updated three-phase co-compensation scheme corresponding to the maximum updated fitness value obtained during the entire iteration as the target compensation scheme. This scheme performs best under the preset performance indicators and is the final solution obtained by combining random selection and comparison.

[0051] Furthermore, the present application provides for randomly sorting the numbers of the multiple historical phase compensation schemes, and the method further includes:

[0052] The random number generator is constructed based on the rand function, and the first random number and the second random number are generated by using the random number generator; when the first random number is greater than the second random number, the multiple three-phase co-compensation scheme numbers are mutated and cross-iterated according to the multiple historical phase-splitting compensation scheme numbers in accordance with the preset update rule until the preset number of cycles is met to obtain the target compensation scheme; when the first random number is less than or equal to the second random number, the multiple historical phase-splitting compensation scheme numbers are randomly sorted, and the multiple historical phase-splitting compensation scheme numbers are updated according to the sorting result.

[0053] Optionally, the system terminal uses the rand function to construct a random number generator, which is used to generate random numbers, and these random numbers determine whether to perform random sorting. Subsequently, the system terminal generates two random numbers based on the random number generator, namely the first random number and the second random number. These two random numbers are used to determine the next optimization direction or operation. If the first random number is greater than the second random number, the system terminal uses the same steps as above to mutate and cross-iterate the current multiple three-phase co-compensation scheme numbers according to the preset update rules to obtain the target compensation scheme. However, if the first random number is less than or equal to the second random number, the system terminal randomly sorts the multiple historical phase compensation scheme numbers and updates these historical scheme numbers according to the sorting results. The purpose of this random sorting and updating is to increase the randomness of the crossover. In the optimization process, the introduction of randomness can help the algorithm jump out of the local optimal solution and explore a wider search space, so as to find a better global solution. In summary, this process combines deterministic rules, i.e., when the first random number is greater than the second random number, and randomness, i.e., when the first random number is less than or equal to the second random number, to maintain the diversity and effectiveness of the search during the optimization process.

[0054] The target circuit is compensated based on the capacitor compensation selection in the target compensation scheme.

[0055] In one embodiment, after obtaining the target compensation scheme, the system terminal actually compensates the target circuit according to the capacitor compensation selection in the scheme. The calculated and optimized optimal capacitor configuration scheme is applied to the target circuit to improve the performance parameters of the target circuit, such as power factor, load power, load voltage, etc. The capacitor compensation selection involves the capacity, quantity, connection method, etc. of the capacitor. The selection of these parameters will directly affect the compensation effect. Therefore, after determining the target compensation scheme, the system terminal installs the corresponding capacitor in the target circuit according to the capacitor compensation selection specified in the scheme, and performs necessary configuration and debugging to ensure that the compensation effect reaches the expected target.

[0056] In summary, the embodiments of the present application have at least the following technical effects:

[0057] The embodiment of the present application collects the load power characteristic information of the target circuit through an intelligent measurement and control unit, including measuring the power factor, load power and load voltage, and then analyzes and obtains the target compensation capacity. Based on this capacity, a three-phase co-compensation configuration is performed, and an initial compensation memory library and a historical phase compensation database are constructed. Subsequently, the initial scheme is continuously optimized through numbering, fitness evaluation and mutation cross iteration, and the competition and selection of compensation schemes are realized by combining the randomness of the random number generator and the preset update rules. During the iteration process, the scheme is updated according to the fitness of the scheme, and after meeting the preset number of iterations, the scheme with the highest fitness is selected as the target compensation scheme. In addition, a compensation capacity identification network layer is introduced to improve the prediction accuracy of the compensation capacity through supervised training. These technical effects jointly solve the technical problem that the traditional capacitor compensation method cannot effectively cope with the three-phase load imbalance, resulting in unsatisfactory compensation effect, and realize the optimization of capacitor compensation selection, improve the power factor of the power system, and reduce the effect of reactive power loss.

[0058] Embodiment 2

[0059] Based on the same inventive concept as the compensation selection competition optimization method of a capacitor in the aforementioned embodiment, Figure 2 As shown, the present application provides a capacitor compensation selection competition optimization system, the system comprising:

[0060] Power acquisition module 1: The power acquisition module 1 is used to collect power from the load of the target circuit using an intelligent measurement and control unit to generate multiple power characteristic information of multiple loads;

[0061] Compensation capacity analysis module 2: the compensation capacity analysis module 2 is used to perform compensation capacity analysis on the plurality of power characteristic information to generate a target compensation capacity;

[0062] Three-phase co-compensation configuration module 3: the three-phase co-compensation configuration module 3 is used to perform three-phase co-compensation configuration based on the target compensation capacity and generate an initial compensation memory library, wherein the initial compensation memory library includes multiple three-phase co-compensation schemes;

[0063] Retrieval module 4: the retrieval module 4 is used to take phase compensation as a target and the target compensation capacity as an index to perform retrieval and generate a historical phase compensation database, wherein the historical phase compensation database includes a plurality of historical phase compensation schemes;

[0064] Mutation crossover iteration module 5: the mutation crossover iteration module 5 is used to perform mutation crossover iteration based on the initial compensation memory library and the historical phase compensation database until a preset number of cycles is met to obtain a target compensation solution;

[0065] Compensation module 6: The compensation module 6 is used to compensate the target circuit based on the capacitor compensation selection in the target compensation scheme.

[0066] Furthermore, the power collection module 1 is also used to perform the following method:

[0067] The plurality of power characteristic information includes a plurality of measured power factors, a plurality of load powers, and a plurality of load voltages.

[0068] Furthermore, the power collection module 1 is also used to perform the following method:

[0069] collecting a plurality of sample measurement power factors, a plurality of sample load powers, a plurality of sample load voltages, and a plurality of sample compensation capacities as construction data;

[0070] Performing supervised training using the constructed data to generate a compensation capacity recognition network layer;

[0071] The compensation capacity identification network layer is used to identify the multiple measured power factors, the multiple load powers and the multiple load voltages to obtain the target compensation capacity.

[0072] Furthermore, the mutation crossover iteration module 5 is also used to perform the following method:

[0073] Respectively numbering the multiple three-phase co-compensation schemes in the initial compensation memory bank and the multiple historical split-phase compensation schemes in the historical split-phase compensation database to obtain multiple three-phase co-compensation scheme numbers and multiple historical split-phase compensation scheme numbers;

[0074] Traversing the multiple three-phase co-compensation scheme numbers and the multiple historical split-phase compensation scheme numbers to evaluate compensation fitness, and generating multiple three-phase co-compensation scheme fitnesses and multiple historical split-phase compensation scheme fitnesses;

[0075] A random number generator is used to perform mutation cross iteration on the multiple three-phase co-compensation scheme numbers according to the preset update rules and the multiple historical phase-separated compensation scheme numbers until the preset number of cycles is met to obtain the target compensation scheme.

[0076] Furthermore, the mutation crossover iteration module 5 is also used to perform the following method:

[0077] The preset update rule is that when the fitness of the first historical phase-splitting compensation scheme number extracted from the multiple historical phase-splitting compensation scheme numbers is greater than or equal to the fitness of the first three-phase co-compensation scheme corresponding to the first three-phase co-compensation scheme number at the corresponding number position of the multiple three-phase co-compensation scheme numbers, the first three-phase co-compensation scheme number is replaced with the first historical phase-splitting compensation scheme number, and multiple updated three-phase co-compensation scheme numbers are obtained, and the updated multiple three-phase co-compensation scheme numbers are used as the updated compensation memory library.

[0078] Furthermore, the mutation crossover iteration module 5 is also used to perform the following method:

[0079] Randomly extracting a first updated three-phase co-compensation scheme number from the updated compensation memory bank to obtain a first updated fitness;

[0080] Randomly extracting the second updated three-phase co-compensation scheme number from the updated compensation memory bank again to obtain the second updated fitness;

[0081] Determine whether the second update fitness is greater than the first update fitness, and if so, use the second update three-phase co-compensation scheme number as the stage update three-phase co-compensation scheme number;

[0082] If not, then accept the second updated three-phase co-compensation scheme number as the stage updated three-phase co-compensation scheme number according to a certain probability;

[0083] After reaching the preset number of iterations, the updated three-phase co-compensation scheme corresponding to the maximum updated fitness value in the iteration process is used as the target compensation scheme.

[0084] Furthermore, the mutation crossover iteration module 5 is also used to perform the following method:

[0085] Constructing the random number generator based on the rand function, and using the random number generator to generate a first random number and a second random number;

[0086] When the first random number is greater than the second random number, according to a preset update rule, the multiple three-phase co-compensation scheme numbers are mutated and cross-iterated according to the multiple historical phase-separated compensation scheme numbers until a preset number of cycles is met to obtain the target compensation scheme;

[0087] When the first random number is less than or equal to the second random number, the multiple historical phase-splitting compensation scheme numbers are randomly sorted, and the multiple historical phase-splitting compensation scheme numbers are updated according to the sorting result.

[0088] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0090] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A capacitor compensation selection competitive optimization method, characterized in that: The method comprises: The intelligent measurement and control unit is used to collect the power of the load of the target circuit and generate multiple power characteristic information of multiple loads; Performing compensation capacity analysis on the plurality of power characteristic information to generate a target compensation capacity; Performing three-phase co-compensation configuration based on the target compensation capacity to generate an initial compensation memory bank, wherein the initial compensation memory bank includes a plurality of three-phase co-compensation schemes; Taking phase compensation as a target, searching with the target compensation capacity as an index, and generating a historical phase compensation database, wherein the historical phase compensation database includes a plurality of historical phase compensation schemes; Performing mutation crossover iteration based on the initial compensation memory library and the historical phase compensation database until a preset number of cycles is met to obtain a target compensation solution; The target circuit is compensated based on the capacitor compensation selection in the target compensation scheme.

2. The method according to claim 1, characterized in that The plurality of power characteristic information includes a plurality of measured power factors, a plurality of load powers, and a plurality of load voltages.

3. The method according to claim 1, characterized in that Based on the initial compensation memory library and the historical phase compensation database, mutation crossover iteration is performed, and the method includes: Respectively numbering the multiple three-phase co-compensation schemes in the initial compensation memory bank and the multiple historical split-phase compensation schemes in the historical split-phase compensation database to obtain multiple three-phase co-compensation scheme numbers and multiple historical split-phase compensation scheme numbers; Traversing the multiple three-phase co-compensation scheme numbers and the multiple historical split-phase compensation scheme numbers to evaluate compensation fitness, and generating multiple three-phase co-compensation scheme fitnesses and multiple historical split-phase compensation scheme fitnesses; A random number generator is used to perform mutation cross iteration on the multiple three-phase co-compensation scheme numbers according to the preset update rules and the multiple historical phase-separated compensation scheme numbers until the preset number of cycles is met to obtain the target compensation scheme.

4. The method according to claim 3, characterized in that The preset update rule is that when the fitness of the first historical phase-splitting compensation scheme number extracted from the multiple historical phase-splitting compensation scheme numbers is greater than or equal to the fitness of the first three-phase co-compensation scheme corresponding to the first three-phase co-compensation scheme number at the corresponding number position of the multiple three-phase co-compensation scheme numbers, the first three-phase co-compensation scheme number is replaced with the first historical phase-splitting compensation scheme number, and multiple updated three-phase co-compensation scheme numbers are obtained, and the updated multiple three-phase co-compensation scheme numbers are used as the updated compensation memory library.

5. The method according to claim 4, characterized in that The method comprises: Randomly extracting a first updated three-phase co-compensation scheme number from the updated compensation memory bank to obtain a first updated fitness; Randomly extracting the second updated three-phase co-compensation scheme number from the updated compensation memory bank again to obtain the second updated fitness; Determine whether the second update fitness is greater than the first update fitness, and if so, use the second update three-phase co-compensation scheme number as the stage update three-phase co-compensation scheme number; If not, then accept the second updated three-phase co-compensation scheme number as the stage updated three-phase co-compensation scheme number according to a certain probability; After reaching the preset number of iterations, the updated three-phase co-compensation scheme corresponding to the maximum updated fitness value in the iteration process is used as the target compensation scheme.

6. The method according to claim 4, characterized in that The method comprises: Constructing the random number generator based on the rand function, and using the random number generator to generate a first random number and a second random number; When the first random number is greater than the second random number, according to a preset update rule, the multiple three-phase co-compensation scheme numbers are mutated and cross-iterated according to the multiple historical phase-separated compensation scheme numbers until a preset number of cycles is met to obtain the target compensation scheme; When the first random number is less than or equal to the second random number, the multiple historical phase-splitting compensation scheme numbers are randomly sorted, and the multiple historical phase-splitting compensation scheme numbers are updated according to the sorting result.

7. The method according to claim 2, characterized in that The method comprises: collecting a plurality of sample measurement power factors, a plurality of sample load powers, a plurality of sample load voltages, and a plurality of sample compensation capacities as construction data; Perform supervised training using the constructed data to generate a compensation capacity recognition network layer; The compensation capacity identification network layer is used to identify the multiple measured power factors, the multiple load powers and the multiple load voltages to obtain the target compensation capacity.

8. A capacitor compensation selection competition optimization system, characterized in that: The steps for implementing the method for competitive optimization of capacitor compensation selection as claimed in any one of claims 1 to 7 include: Power acquisition module: Use the intelligent measurement and control unit to collect the power of the load of the target circuit and generate multiple power characteristic information of multiple loads; Compensation capacity analysis module: performs compensation capacity analysis on the plurality of power characteristic information to generate a target compensation capacity; A three-phase co-compensation configuration module: performs three-phase co-compensation configuration based on the target compensation capacity, and generates an initial compensation memory library, wherein the initial compensation memory library includes a plurality of three-phase co-compensation schemes; Retrieval module: taking phase compensation as a target and the target compensation capacity as an index for retrieval, and generating a historical phase compensation database, wherein the historical phase compensation database includes a plurality of historical phase compensation schemes; Mutation crossover iteration module: performs mutation crossover iteration based on the initial compensation memory library and the historical phase compensation database until a preset number of cycles is met to obtain a target compensation solution; Compensation module: Compensating the target circuit based on the capacitor compensation selection in the target compensation scheme.

Images