Reactive compensation optimization method, device and equipment for power system and readable storage medium

Abstract

The invention discloses a reactive compensation optimization method, device and equipment for a power system and a readable storage medium, and the method comprises the steps: obtaining current meteorological data, current power operation data and historical power operation data of a target region, and predicting the load data of the target region based on the current meteorological data, the current power operation data and the historical power operation data in combination with a deep learning algorithm; taking load data and current power operation data as input data, taking network loss minimization and voltage deviation minimization as targets, taking node power balance, node voltage upper and lower limits and reactive compensation upper and lower limits as constraints, and constructing a multi-target optimization model for generating a reactive compensation optimization scheme; setting a variation rate based on the load data and the current power operation data; and combining the mutation rate with a genetic algorithm, and solving the multi-objective optimization model to obtain a reactive compensation optimization scheme. Therefore, the reactive power compensation optimization scheme can be dynamically generated through the multi-target optimization model, the operation state change of the power system is responded in real time, and the optimal reactive power distribution of the power system is ensured.

Description

Technical Field

[0001] The present application relates to the field of power grid technology, and more specifically, to a method, device, equipment and readable storage medium for optimizing reactive power compensation in a power system. Background Art

[0002] In the power system, reactive power is an important factor in maintaining voltage stability, reducing line losses and maintaining power system safety. Reasonable allocation of reactive power can not only improve the power quality of the power grid, but also reduce the energy loss of the power system, so that the power grid can maintain stable operation under different load levels. However, with the large-scale access of new energy, the optimization of reactive power compensation faces new challenges. The access of new energy is usually volatile and uncertain, so the power grid needs a more flexible reactive power compensation solution to cope with frequent power fluctuations and voltage instability.

[0003] Traditional reactive power compensation optimization methods mainly optimize reactive power compensation for static conditions, and cannot effectively cope with dynamic changes in power system operation, especially the frequent fluctuations caused by the access of new energy. Therefore, it is urgent to provide a reactive power compensation optimization method that can adapt to the dynamic changes of power systems. Summary of the invention

[0004] In view of this, the present application provides a method, device, equipment and readable storage medium for optimizing reactive power compensation in a power system, which are used to address the shortcomings of the reactive power compensation optimization methods in the prior art that they cannot effectively cope with dynamic changes in the operation of the power system.

[0005] In order to achieve the above objectives, the proposed solution is as follows:

[0006] A method for optimizing reactive power compensation in a power system, comprising:

[0007] Obtain current meteorological data, current power operation data and historical power operation data of the target area;

[0008] Based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, the load data of the target area is predicted;

[0009] Taking the load data and the current power operation data as input data, minimizing network loss and voltage deviation as goals, and taking node power balance, node voltage upper and lower limits, and reactive power compensation upper and lower limits as constraints, a multi-objective optimization model for generating a reactive power compensation optimization solution is constructed;

[0010] Setting a mutation rate based on the load data and the current power operation data;

[0011] The mutation rate is combined with a genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization scheme, which is used to optimize reactive power compensation in a power system.

[0012] Optionally, also include:

[0013] Based on the current power operation data, monitor whether the voltage of each node in the current power system exceeds a safe range;

[0014] When the voltage at any node exceeds a safety range, a reactive power compensation value is calculated, and reactive compensation is performed based on the reactive power compensation value.

[0015] Optionally, the predicting load data of a target area based on the meteorological data, the current power operation data and the historical power operation data in combination with a deep learning algorithm includes:

[0016] Obtain the load forecasting function built based on the deep learning algorithm;

[0017] Based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, predict the new energy power generation data of the target area;

[0018] The new energy power generation data, the meteorological data, the current power operation data and the historical power operation data are combined with the load prediction function to predict the load data of the target area.

[0019] Optionally, the current power operation data includes current load information;

[0020] The setting of the variation rate based on the load data and the current power operation data includes:

[0021] Comparing the load data with the current load information to evaluate the degree of load fluctuation;

[0022] The variation rate is set based on the load fluctuation degree.

[0023] Optionally, combining the mutation rate with a genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization solution includes:

[0024] Randomly generating multiple feasible solutions of the multi-objective optimization model as first-generation feasible solutions;

[0025] Calculate the network loss value and voltage error corresponding to each feasible solution of the latest generation;

[0026] Based on the network loss value and voltage error of each feasible solution of the latest generation, the fitness value of each feasible solution of the latest generation is calculated;

[0027] According to the size of the fitness value, multiple elite feasible solutions are selected from the latest generation of feasible solutions, and based on the mutation rate, crossover and mutation operations are performed on each elite feasible solution to generate multiple feasible solutions of the new generation;

[0028] If the fitness value of the latest generation of feasible solutions is greater than the fitness value of the previous generation of feasible solutions, the latest generation of feasible solutions is retained, and the process returns to the step of selecting multiple elite feasible solutions from the latest generation of feasible solutions according to the size of the fitness value to enter the next generation population until the network loss and voltage deviation are minimized;

[0029] If the fitness value of the latest generation of feasible solutions is less than the fitness value of the previous generation of feasible solutions, the feasible solutions of the previous generation are retained as the latest generation of feasible solutions, and the process returns to execute the step of selecting multiple elite feasible solutions from the latest generation of feasible solutions according to the size of the fitness value to enter the next generation population until the network loss is minimized and the voltage deviation is minimized.

[0030] Optionally, obtain the current power operation data of the target area, including:

[0031] Integrate the current node voltages, current load information, current renewable energy outputs, current reactive powers and current active powers collected by various sensors in the target area of ​​the power system to form current power operation data.

[0032] A reactive power compensation optimization device for a power system, comprising:

[0033] An acquisition module is used to acquire current meteorological data, current power operation data and historical power operation data of the target area;

[0034] A prediction module, used to predict the load data of a target area based on the meteorological data, the current power operation data and the historical power operation data in combination with a deep learning algorithm;

[0035] A construction module is used to construct a multi-objective optimization model for generating a reactive compensation optimization scheme by taking the load data and the current power operation data as input data, minimizing network loss and voltage deviation as goals, and taking node power balance, node voltage upper and lower limits, and reactive compensation upper and lower limits as constraints;

[0036] A setting module, used for setting a mutation rate based on the load data and the current power operation data;

[0037] A solution module is used to combine the mutation rate with a genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization solution, and the reactive power compensation optimization solution is used to optimize reactive power compensation in a power system.

[0038] A reactive power compensation optimization device for a power system, comprising a memory and a processor;

[0039] The memory is used to store programs;

[0040] The processor is used to execute the program to implement the various steps of the above-mentioned power system reactive power compensation optimization method.

[0041] A readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, each step of the above-mentioned power system reactive power compensation optimization method is implemented.

[0042] It can be seen from the above technical solutions that the method for optimizing reactive power compensation in a power system provided by the present application can obtain the current meteorological data, current power operation data and historical power operation data of the target area; based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, predict the load data of the target area; based on this, the present application can use a deep learning algorithm to perform load forecasting, thereby improving the accuracy of the forecast, thereby improving the accuracy of the reactive power compensation optimization scheme; subsequently, using the load data and the current power operation data as input data, with the goal of minimizing network loss and minimizing voltage deviation, and with node power balance, node voltage upper and lower limits, and reactive compensation upper and lower limits as constraints, a multi-objective optimization model for generating a reactive power compensation optimization scheme is constructed; based on this, the present application can be implemented through Construct a multi-objective optimization model, a multi-level dynamic optimization reactive power optimization strategy, and currently respond to changes in the operating state of the power system in order to dynamically adjust the reactive power compensation equipment to cope with dynamic changes in the operation of the power system; at the same time, the multi-objective optimization model integrates node power balance, voltage upper and lower limits, and compensation upper and lower limits to further ensure the practicality and reliability of this application; then, the mutation rate can be set based on the load data and the current power operation data; the mutation rate is combined with the genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization scheme, which is used to optimize the reactive power compensation of the power system; based on this, this application combines the genetic algorithm to iteratively solve the reactive power compensation optimization scheme. In this process, the mutation rate is adjusted with reference to the load forecast and current load information to enhance the optimization capability of the scheme. It can be seen that this application can combine deep learning and genetic algorithms to perform global reactive power optimization on the entire power system in the target area, avoiding the problem of uneven reactive power distribution during local optimization. At the same time, this application can dynamically generate reactive power compensation optimization schemes through multi-objective optimization models, respond to changes in the dynamic operating state of the power system in real time, and ensure the optimal reactive power distribution of the power system under different load conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0044] Figure 1 A flow chart of a method for optimizing reactive power compensation in a power system disclosed in an embodiment of the present application;

[0045] Figure 2 A structural block diagram of a reactive power compensation optimization device for a power system disclosed in an embodiment of the present application;

[0046] Figure 3 This is a hardware structure block diagram of a power system reactive power compensation optimization device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0047] 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, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0048] The embodiment of the present application provides a method for optimizing reactive power compensation in a power system. The method can be applied to various power systems or reactive power supplement systems, and can also be applied to various computer terminals or smart terminals. The execution subject can be a processor or server of the computer terminal or smart terminal. The method flow chart of the method is as follows: Figure 1 As shown, specifically including:

[0049] Next, combine Figure 1 The reactive power compensation optimization method of the power system of the present application is introduced in detail, including the following steps:

[0050] Step S1, obtaining current meteorological data, current power operation data and historical power operation data of the target area.

[0051] Specifically, current meteorological data related to the output of renewable energy in the target area may be obtained, for example, wind intensity and light intensity.

[0052] The current node voltage, current load information, current reactive power and current active power in the target area can be obtained as current power operation data;

[0053] The historical node voltage, historical new energy output, historical load information, historical reactive power and historical active power of the previous n moments can be obtained as historical power operation data.

[0054] Step S2: Based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, predict the load data of the target area.

[0055] Specifically, a model built based on a deep learning algorithm can be obtained, and meteorological data, current power operation data, and historical power operation data can be input into the prediction model to obtain the load data predicted by the prediction model.

[0056] Based on load data, the investment in compensation equipment can be adjusted in advance to reduce unnecessary voltage fluctuations and losses.

[0057] Step S3, using the load data and the current power operation data as input data, minimizing network losses and voltage deviations as goals, and taking node power balance, node voltage upper and lower limits, and reactive compensation upper and lower limits as constraints, to construct a multi-objective optimization model for generating a reactive compensation optimization solution.

[0058] Specifically, the network loss minimization function and the voltage deviation minimization function can be used as objective functions, and the node power balance, node voltage upper and lower limits, and reactive compensation upper and lower limits can be used as constraints to build a model, and the load data and current power operation data can be input into the model to obtain a multi-objective optimization function.

[0059] Among them, the network loss minimization function can be shown as follows:

[0060]

[0061] The voltage deviation minimization function can be shown as follows:

[0062]

[0063] Node power balancing can be done as follows:

[0064]

[0065]

[0066] The node voltage upper and lower limits can be as follows:

[0067]

[0068] The upper and lower limits of reactive power compensation can be shown as follows:

[0069]

[0070] In the formula, For Node and The conductance between For Node The voltage, For Node The voltage, is the node voltage phase angle difference, is the active power loss between nodes; is the reference voltage value, usually 1.0 pu; For Node Active power of For Node Reactive power; For Node Active power of load; For Node The reactive power of the load; For Node and The susceptance between is the lower limit of node voltage, generally set to 0.95 pu; is the upper limit of node voltage, which is generally set to 1.05 pu; is the output reactive power of the reactive compensation device, is the minimum output of the reactive power compensation device; It is the maximum output of the reactive power compensation device.

[0071] Step S4: setting a variation rate based on the load data and the current power operation data.

[0072] Specifically, the variation rate may be set with reference to the degree of difference between the load data and the current load information in the current power operation data.

[0073] Step S5: combining the mutation rate with a genetic algorithm to solve the multi-objective optimization model and obtain a reactive power compensation optimization scheme, wherein the reactive power compensation optimization scheme is used to optimize reactive power compensation for the power system.

[0074] Specifically, the genetic algorithm is combined to iteratively calculate the feasible solution of the multi-objective optimization model. During the iterative calculation process, the population is cross-mutated based on the mutation rate to generate the optimal feasible solution.

[0075] According to the reactive compensation optimization plan, the reactive compensation equipment (such as capacitor banks, SVC, STATCOM) can be put into operation and removed in real time to ensure the stable operation of the power system.

[0076] At the same time, the present application can also combine the mutation rate with particle swarm optimization or ant colony algorithm to solve the multi-objective optimization model and obtain a reactive power compensation optimization solution.

[0077] It can be seen from the above technical solutions that the method for optimizing reactive power compensation in a power system provided by the present application can obtain the current meteorological data, current power operation data and historical power operation data of the target area; based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, the load data of the target area is predicted; based on this, the present application can use a deep learning algorithm to perform load prediction, improve the accuracy of the prediction, and thus improve the accuracy of the reactive power compensation optimization scheme; subsequently, using the load data and the current power operation data as input data, with the goal of minimizing network loss and voltage deviation, and with node power balance, node voltage upper and lower limits, and reactive compensation upper and lower limits as constraints, a multi-objective optimization model for generating a reactive power compensation optimization scheme is constructed; based on this , this application can construct a multi-objective optimization model, a multi-level dynamic optimization reactive power optimization strategy, and currently respond to changes in the operating state of the power system in order to dynamically adjust the reactive compensation equipment; at the same time, the multi-objective optimization model integrates node power balance, voltage upper and lower limits, and compensation upper and lower limits, further ensuring the practicality and reliability of this application; then, the mutation rate can be set based on the load data and the current power operation data; the mutation rate is combined with the genetic algorithm to solve the multi-objective optimization model to obtain a reactive compensation optimization scheme, which is used to optimize the reactive compensation of the power system; based on this, this application combines the genetic algorithm to iteratively solve the reactive compensation optimization scheme. In this process, the mutation rate is adjusted with reference to the load forecast and current load information to enhance the optimization capability of the scheme. It can be seen that this application can combine deep learning and genetic algorithms to perform global reactive power optimization on the entire power system in the target area, avoiding the problem of uneven reactive distribution during local optimization. At the same time, this application can dynamically generate reactive compensation optimization schemes through multi-objective optimization models, respond to changes in the operating state of the power system in real time, and ensure the optimal reactive power distribution of the power system under different load conditions.

[0078] In some embodiments of the present application, although the solution time of the multi-objective optimization model is short, in order to ensure accuracy, it is impossible to generate a reactive compensation optimization plan in real time and continuously. There is a certain time interval between the generation of two reactive compensation optimization plans. In the interval between the generation of the reactive compensation optimization plan, there may be accidents that cause the node voltage to be too high or too low. At this time, reactive compensation needs to be performed in time to avoid power system operation failures. Therefore, the present application adds a compensation process. Next, the compensation process will be described in detail. The steps are as follows:

[0079] S001. Based on the current power operation data, monitor whether the voltage of each node in the current power system exceeds a safe range.

[0080] Specifically, the node voltage collected by the sensor can be obtained in real time, and the node voltage can be filtered using a filtering algorithm;

[0081] The filtered node voltage can be compared with a preset safety range to assess whether it exceeds the safety range.

[0082] The safety range can be calculated based on the reference voltage of the corresponding node.

[0083] S002. When the voltage at any node exceeds a safety range, a reactive power compensation value is calculated, and reactive compensation is performed based on the reactive power compensation value.

[0084] Specifically, the reactive power compensation value may be calculated based on a preset control gain, a node voltage exceeding a safety range and a corresponding reference voltage, and reactive compensation may be performed based on the reactive power compensation value.

[0085] Among them, a reactive power compensation calculation function can be used to calculate the reactive power compensation value based on a preset control gain, a node voltage exceeding a safety range and a corresponding reference voltage.

[0086] The reactive power compensation calculation function can be shown as follows:

[0087]

[0088] is the preset control gain; is the number of node voltages exceeding the safety range; is the nth node voltage that exceeds the safety range; is the reference voltage value corresponding to the nth node voltage that exceeds the safety range.

[0089] It can be seen from the above technical solution that compared with the previous embodiment, this embodiment newly provides a compensation process. Through the above compensation process, the voltage of each node can be detected in real time, and after the node voltage exceeds the safety range, each node is compensated by reactive power to ensure the safe and stable operation of the power system.

[0090] In some embodiments of the present application, the process of obtaining the current power operation data of the target area in step S1 is described in detail, and the steps are as follows:

[0091] S10, integrating each current node voltage, each current load information, each current new energy output, each current reactive power and each current active power collected by various sensors in the target area of ​​the power system to form current power operation data.

[0092] Specifically, the current node voltage, current load information, current reactive power, current new energy output, and current active power collected by various sensors in the target area of ​​the power system can be obtained;

[0093] A filtering algorithm, such as a Kalman filter, may be used to clean and integrate the collected current node voltages, current load information, current reactive powers, and current active powers.

[0094] Among them, the filtering algorithm can be as follows:

[0095]

[0096] is the filtering result; The value collected by the sensor; is the measurement matrix; is the Kalman gain.

[0097] It can be seen from the above technical solution that this embodiment provides an optional method for obtaining current power operation data. Through the above method, data can be collected through sensors, thereby improving the accuracy and real-time performance of the current power operation data and facilitating timely acquisition of current power operation data.

[0098] In some embodiments of the present application, step S2, based on the meteorological data, the current power operation data and the historical power operation data, combined with the deep learning algorithm, predicts the load data of the target area, and the steps are as follows:

[0099] S20. Obtain a load prediction function based on a deep learning algorithm.

[0100] Specifically, the load forecasting function can be shown as follows:

[0101]

[0102] For time The forecast load at the time, For the front Load history data at each moment, For meteorological data, Data on new energy generation.

[0103] S21. Based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, predict the new energy power generation data of the target area.

[0104] Specifically, the new energy power generation data of the target area can be predicted based on meteorological data, current output of new energy in current power operation data and current meteorological data, historical new energy output and historical meteorological data in historical power operation data, combined with deep learning algorithms.

[0105] S23, combining the new energy power generation data, the meteorological data, the current power operation data and the historical power operation data with the load prediction function to predict the load data of the target area.

[0106] Specifically, the new energy power generation data, the meteorological data, the current power operation data and the historical power operation data are substituted into the load prediction function to predict the load data of the target area.

[0107] It can be seen from the above technical solution that this embodiment provides an optional method for predicting the load data of the target area based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm. The above method can be further combined with a deep learning algorithm to predict the output of new energy, and on the basis of the output of new energy, the load is predicted, thereby further improving the load prediction accuracy of the present application.

[0108] In some embodiments of the present application, step S4, the process of setting the mutation rate based on the load data and the current power operation data is described in detail, and the steps are as follows:

[0109] S40: Compare the load data with the current load information to evaluate the degree of load fluctuation.

[0110] Specifically, the difference between the load data and the current load information may be calculated, the difference indicating the degree of load fluctuation.

[0111] S41. Setting the variation rate based on the load fluctuation degree.

[0112] Specifically, the mutation rate can be set according to the size of the difference.

[0113] The difference is directly proportional to the mutation rate.

[0114] It can be seen from the above technical solution that this embodiment provides an optional method for setting the mutation rate based on the load data and the current power operation data. Through the above method, the mutation rate can be set according to the degree of load fluctuation, which can better maintain the balance between exploration and optimization and further ensure the reliability of the generated reactive compensation optimization scheme.

[0115] In some embodiments of the present application, the process of setting the variation rate based on the load fluctuation degree in step S41 is described in detail, and the steps are as follows:

[0116] S410: taking the product of the load fluctuation degree and the preset proportional weight as the variation rate.

[0117] Specifically, the proportional weight can be calculated in advance based on the results of multiple tests;

[0118] The product of the difference and the proportional weight can be used as the mutation rate.

[0119] It can be seen from the above technical solution that this embodiment provides an optional way to set the variation rate according to the load fluctuation degree, and the variation rate can be better determined by the above method.

[0120] In some embodiments of the present application, the process of combining the mutation rate with a genetic algorithm to solve the multi-objective optimization model and obtain a reactive power compensation optimization solution in step S5 is described in detail, and the steps are as follows:

[0121] S50. Randomly generate multiple feasible solutions of the multi-objective optimization model as first-generation feasible solutions.

[0122] Specifically, multiple feasible solutions of the multi-objective optimization model can be randomly generated as first-generation feasible solutions.

[0123] S51. Calculate the network loss value and voltage error corresponding to each feasible solution of the latest generation.

[0124] Specifically, the network loss value and voltage error of each feasible solution generated most recently may be calculated.

[0125] S52. Calculate the fitness value of each feasible solution of the latest generation based on the network loss value and voltage error of each feasible solution of the latest generation.

[0126] Specifically, the network loss value and voltage error of each feasible solution may be substituted into a preset fitness calculation function to calculate the fitness value of each feasible solution.

[0127] The fitness calculation function can be shown as follows:

[0128]

[0129] and are network loss and voltage error respectively, and is the weight coefficient.

[0130] S53. According to the size of the fitness value, multiple elite feasible solutions are selected from the latest generation of feasible solutions, and based on the mutation rate, each elite feasible solution is subjected to crossover and mutation operations to generate multiple feasible solutions of a new generation; whether the fitness value of the latest generation of feasible solutions is greater than the fitness value of the previous generation of feasible solutions; if so, execute step S53; if not, execute step S54.

[0131] Specifically, N feasible solutions with the highest fitness values ​​can be selected from the latest generation of feasible solutions as elite feasible solutions.

[0132] Based on the mutation rate, each elite feasible solution can be crossover and mutation operated to obtain a new generation of feasible solutions.

[0133] In order to retain elite genes, the fitness values ​​of the upper and lower generations can be compared. If the next generation is not as good as the previous generation, the next generation will be deleted, and the previous generation will be used as the parent to generate the offspring; if the previous generation is not as good as the next generation, the next generation will be retained, and the next generation will be used as the parent to generate the offspring.

[0134] Therefore, it can be evaluated whether the fitness value of the latest generation of feasible solutions is greater than the fitness value of the previous generation of feasible solutions; if so, step S53 is executed; if not, step S54 is executed.

[0135] S53, retaining the latest generation of feasible solutions, and returning to step S53 to enter the next generation population until the network loss is minimized and the voltage deviation is minimized.

[0136] Specifically, if the fitness value of the latest generation of feasible solutions is greater than the fitness value of the previous generation of feasible solutions, the latest generation of feasible solutions is retained, and the latest generation is used as the parent generation to generate the child generation.

[0137] S54, retaining each feasible solution of the previous generation as the latest generation of feasible solutions, and returning to execute step S53 to enter the next generation population until the network loss is minimized and the voltage deviation is minimized.

[0138] Specifically, if the fitness value of the latest generation of feasible solutions is less than the fitness value of the previous generation of feasible solutions, then the feasible solutions of the previous generation are retained as the latest generation of feasible solutions, that is, the previous generation is used as the parent generation to generate the child generation.

[0139] It can be seen from the above technical solution that this embodiment provides an optional method of combining the mutation rate with the genetic algorithm to solve the multi-objective optimization model and obtain a reactive compensation optimization solution. The reactive compensation optimization solution can be iteratively solved through the above method to further ensure the accuracy of the reactive compensation optimization solution.

[0140] Next, we will combine Figure 2 The reactive power compensation optimization device for a power system provided in the present application is introduced in detail. The reactive power compensation optimization device for a power system provided below can be compared with the reactive power compensation optimization method for a power system provided above.

[0141] See also Figure 2 It can be found that the power system reactive power compensation optimization device can include:

[0142] An acquisition module 10 is used to acquire current meteorological data, current power operation data and historical power operation data of a target area;

[0143] A prediction module 20, configured to predict the load data of a target area based on the meteorological data, the current power operation data and the historical power operation data in combination with a deep learning algorithm;

[0144] A construction module 30 is used to construct a multi-objective optimization model for generating a reactive compensation optimization scheme, taking the load data and the current power operation data as input data, minimizing network loss and voltage deviation as goals, and taking node power balance, node voltage upper and lower limits, and reactive compensation upper and lower limits as constraints;

[0145] A setting module 40, configured to set a variation rate based on the load data and the current power operation data;

[0146] The solution module 50 is used to combine the mutation rate with the genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization solution, and the reactive power compensation optimization solution is used to optimize reactive power compensation for the power system.

[0147] The power system reactive power compensation optimization device may also include:

[0148] A node voltage monitoring unit, used to monitor whether the voltage of each node in the current power system exceeds a safe range based on the current power operation data;

[0149] The reactive power compensation value calculation unit is used to calculate the reactive power compensation value when the voltage of any node exceeds the safety range, and perform reactive compensation based on the reactive power compensation value.

[0150] Prediction modules may include:

[0151] A load forecasting function acquisition unit, used to acquire a load forecasting function constructed based on a deep learning algorithm;

[0152] A new energy power generation data prediction unit, used to predict the new energy power generation data of the target area based on the meteorological data, the current power operation data and the historical power operation data in combination with a deep learning algorithm;

[0153] The load data prediction unit is used to combine the new energy power generation data, the meteorological data, the current power operation data and the historical power operation data with the load prediction function to predict the load data of the target area.

[0154] Setting up the module can include:

[0155] A load fluctuation degree evaluation unit, used for comparing the load data with the current load information to evaluate the load fluctuation degree;

[0156] The load fluctuation degree utilizing unit is used to set the variation rate based on the load fluctuation degree.

[0157] Load fluctuation degree utilization units may include:

[0158] The product calculation component is used to take the product between the load fluctuation degree and the preset proportional weight as the variation rate.

[0159] Solving modules can include:

[0160] A first solving unit, used for randomly generating a plurality of feasible solutions of the multi-objective optimization model as first generation feasible solutions;

[0161] The second solving unit is used to calculate the network loss value and voltage error corresponding to each feasible solution of the latest generation;

[0162] A third solution unit is used to calculate the fitness value of each feasible solution of the latest generation based on the network loss value and voltage error of each feasible solution of the latest generation;

[0163] The fourth solution unit is used to select multiple elite feasible solutions from the latest generation of feasible solutions according to the size of the fitness value, and perform crossover and mutation operations on each elite feasible solution based on the mutation rate to generate multiple feasible solutions of the new generation;

[0164] The fifth solving unit is used for retaining the latest generation of feasible solutions if the fitness value of the latest generation of feasible solutions is greater than the fitness value of the previous generation of feasible solutions, and returning to execute the step of selecting multiple elite feasible solutions from the latest generation of feasible solutions according to the size of the fitness value to enter the next generation population until the network loss is minimized and the voltage deviation is minimized;

[0165] The sixth solving unit is used to retain the feasible solutions of the previous generation as the latest generation of feasible solutions if the fitness value of the latest generation of feasible solutions is less than the fitness value of the previous generation of feasible solutions, and return to execute the step of selecting multiple elite feasible solutions from the latest generation of feasible solutions according to the size of the fitness value to enter the next generation population until the network loss is minimized and the voltage deviation is minimized.

[0166] Acquisition modules may include:

[0167] The current power operation data acquisition unit is used to integrate the current node voltages, current load information, current new energy outputs, current reactive powers and current active powers collected by various sensors in the target area of ​​the power system to form current power operation data.

[0168] The power system reactive power compensation optimization device provided in the embodiment of the present application can be applied to power system reactive power compensation optimization equipment, such as PC terminals, cloud platforms, servers and server clusters. Figure 3 The hardware structure diagram of the power system reactive power compensation optimization device is shown in FIG. Figure 3 , the hardware structure of the power system reactive power compensation optimization device may include: at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4;

[0169] In the embodiment of the present application, the number of the processor 1, the communication interface 2, the memory 3, and the communication bus 4 is at least one, and the processor 1, the communication interface 2, and the memory 3 communicate with each other through the communication bus 4;

[0170] The processor 1 may be a central processing unit CPU, or an application-specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention, etc.;

[0171] The memory 3 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), etc., such as at least one disk memory;

[0172] The memory stores a program, and the processor can call the program stored in the memory, wherein the program is used to:

[0173] Obtain current meteorological data, current power operation data and historical power operation data of the target area;

[0174] Based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, the load data of the target area is predicted;

[0175] Taking the load data and the current power operation data as input data, minimizing network loss and voltage deviation as goals, and taking node power balance, node voltage upper and lower limits, and reactive power compensation upper and lower limits as constraints, a multi-objective optimization model for generating a reactive power compensation optimization solution is constructed;

[0176] Setting a mutation rate based on the load data and the current power operation data;

[0177] The mutation rate is combined with a genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization scheme, which is used to optimize reactive power compensation in a power system.

[0178] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0179] The embodiment of the present application further provides a readable storage medium, which may store a program suitable for execution by a processor, wherein the program is used to:

[0180] Obtain current meteorological data, current power operation data and historical power operation data of the target area;

[0181] Based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, the load data of the target area is predicted;

[0182] Taking the load data and the current power operation data as input data, minimizing network loss and voltage deviation as goals, and taking node power balance, node voltage upper and lower limits, and reactive power compensation upper and lower limits as constraints, a multi-objective optimization model for generating a reactive power compensation optimization solution is constructed;

[0183] Setting a mutation rate based on the load data and the current power operation data;

[0184] The mutation rate is combined with a genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization scheme, which is used to optimize reactive power compensation in a power system.

[0185] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0186] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0187] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0188] The above description of the disclosed embodiments enables professionals and technicians in the field to implement or use the present application. Various modifications to these embodiments will be apparent to professionals and technicians in the field, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. The various embodiments of the present application can be combined with each other. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for optimizing reactive power compensation in a power system, characterized in that: include: Obtain current meteorological data, current power operation data and historical power operation data of the target area; Based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, the load data of the target area is predicted; Taking the load data and the current power operation data as input data, minimizing network loss and voltage deviation as goals, and taking node power balance, node voltage upper and lower limits, and reactive power compensation upper and lower limits as constraints, a multi-objective optimization model for generating a reactive power compensation optimization solution is constructed; Setting a mutation rate based on the load data and the current power operation data; The mutation rate is combined with a genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization scheme, which is used to optimize reactive power compensation in a power system.

2. The method for optimizing reactive power compensation in a power system according to claim 1, characterized in that: Also includes: Based on the current power operation data, monitor whether the voltage of each node in the current power system exceeds a safe range; When the voltage at any node exceeds a safety range, a reactive power compensation value is calculated, and reactive compensation is performed based on the reactive power compensation value.

3. The method for optimizing reactive power compensation in a power system according to claim 1, characterized in that: The method of predicting the load data of the target area based on the meteorological data, the current power operation data and the historical power operation data in combination with a deep learning algorithm includes: Obtain the load forecasting function built based on the deep learning algorithm; Based on the meteorological data, the current power operation data and the historical power operation data, combined with a deep learning algorithm, predict the new energy power generation data of the target area; The new energy power generation data, the meteorological data, the current power operation data and the historical power operation data are combined with the load prediction function to predict the load data of the target area.

4. The method for optimizing reactive power compensation in a power system according to claim 1, characterized in that: The current power operation data includes current load information; The setting of the variation rate based on the load data and the current power operation data includes: Comparing the load data with the current load information to evaluate the degree of load fluctuation; The variation rate is set based on the load fluctuation degree.

5. The method for optimizing reactive power compensation in a power system according to claim 4, characterized in that: The step of setting the variation rate based on the load fluctuation degree includes: The product of the load fluctuation degree and the preset proportional weight is used as the variation rate.

6. The method for optimizing reactive power compensation in a power system according to claim 1, characterized in that: The step of combining the mutation rate with a genetic algorithm to solve the multi-objective optimization model and obtain a reactive power compensation optimization scheme includes: Randomly generating multiple feasible solutions of the multi-objective optimization model as first-generation feasible solutions; Calculate the network loss value and voltage error corresponding to each feasible solution of the latest generation; Based on the network loss value and voltage error of each feasible solution of the latest generation, the fitness value of each feasible solution of the latest generation is calculated; According to the size of the fitness value, multiple elite feasible solutions are selected from the latest generation of feasible solutions, and based on the mutation rate, crossover and mutation operations are performed on each elite feasible solution to generate multiple feasible solutions of the new generation; If the fitness value of the latest generation of feasible solutions is greater than the fitness value of the previous generation of feasible solutions, the latest generation of feasible solutions is retained, and the process returns to the step of selecting multiple elite feasible solutions from the latest generation of feasible solutions according to the size of the fitness value to enter the next generation population until the network loss and voltage deviation are minimized; If the fitness value of the latest generation of feasible solutions is less than the fitness value of the previous generation of feasible solutions, the feasible solutions of the previous generation are retained as the latest generation of feasible solutions, and the process returns to execute the step of selecting multiple elite feasible solutions from the latest generation of feasible solutions according to the size of the fitness value to enter the next generation population until the network loss is minimized and the voltage deviation is minimized.

7. The method for optimizing reactive power compensation in a power system according to any one of claims 1 to 6, characterized in that: Obtain current power operation data of the target area, including: Integrate the current node voltages, current load information, current renewable energy outputs, current reactive powers and current active powers collected by various sensors in the target area of ​​the power system to form current power operation data.

8. A reactive power compensation optimization device for a power system, characterized in that: include: An acquisition module is used to acquire current meteorological data, current power operation data and historical power operation data of the target area; A prediction module, used to predict the load data of a target area based on the meteorological data, the current power operation data and the historical power operation data in combination with a deep learning algorithm; A construction module is used to construct a multi-objective optimization model for generating a reactive compensation optimization scheme by taking the load data and the current power operation data as input data, minimizing network loss and voltage deviation as goals, and taking node power balance, node voltage upper and lower limits, and reactive compensation upper and lower limits as constraints; A setting module, used for setting a mutation rate based on the load data and the current power operation data; A solution module is used to combine the mutation rate with a genetic algorithm to solve the multi-objective optimization model to obtain a reactive power compensation optimization solution, and the reactive power compensation optimization solution is used to optimize reactive power compensation in a power system.

9. A reactive power compensation optimization device for a power system, characterized in that: including memory and processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the power system reactive power compensation optimization method as described in any one of claims 1-7.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the method for optimizing reactive power compensation in a power system as described in any one of claims 1 to 7 is implemented.

Images