Voltage optimization method and system

Abstract

The invention discloses a voltage optimization method and system. The method comprises the following steps: establishing a first target model, wherein the first target model is used for determining photovoltaic processing uncertainty; based on the first target model, a first scene set is generated in combination with a first algorithm, and the first scene set comprises a plurality of scenes; presetting a first prediction control strategy, and establishing a first optimization model based on the first scene set; and solving the first optimization model, and carrying out voltage optimization. The voltage optimization system can effectively cope with the uncertainty of photovoltaic output, and can more accurately predict and process the voltage fluctuation problem by establishing the first target model and the first scene set. Besides, by presetting the first prediction control strategy and establishing the first optimization model, the system can optimize the voltage in real time, so that the stability of the power grid and the absorption capability of new energy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy system and power distribution system planning, and in particular to a voltage optimization method and system. Background Art

[0002] With the continuous development of photovoltaics, an increasing proportion of photovoltaics are connected to the grid as distributed energy, and the reverse flow is significant, which brings problems such as voltage over-limit. However, the photovoltaic output is highly random and fluctuates greatly, which brings new challenges to the consumption of new energy and the stable operation of the power grid. Therefore, a voltage optimization method based on photovoltaic uncertainty is urgently needed.

[0003] The traditional distribution network has no distributed energy, only loads, and the loads are more predictable, so the control method of the power grid is relatively simple. However, with people's attention to renewable energy, more and more photovoltaic power stations and wind power plants have been put into use, especially in recent years, they have been directly integrated into the distribution network, making the distribution network problem increasingly complicated. Therefore, new methods need to be proposed for active distribution networks. However, most of the methods for optimizing the operation of new distribution networks at this stage still use new energy deterministic models, ignoring the impact of output uncertainty and failing to cover the prediction errors of new energy. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a voltage optimization method and system, which can solve the problems mentioned in the background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a voltage optimization method, comprising:

[0009] establishing a first target model, the first target model being used to determine photovoltaic processing uncertainty;

[0010] Based on the first target model and in combination with a first algorithm, a first scene set is generated, where the first scene set includes a plurality of scenes;

[0011] Presetting a first prediction control strategy and establishing a first optimization model based on the first scenario set;

[0012] The first optimization model is solved to perform voltage optimization.

[0013] As a preferred solution of the voltage optimization method of the present invention, wherein: the establishing of the first optimization model based on the first scenario set includes:

[0014] Performing a first predictive control strategy process on all the scenarios in the first scenario set;

[0015] Obtaining a first optimization model after all scenarios are processed by a first predictive control strategy;

[0016] The first optimization model includes a first objective function and a first constraint condition.

[0017] As a preferred solution of the voltage optimization method described in the present invention, the first optimization model is a comprehensive model for all scenarios in the first scenario set.

[0018] As a preferred solution of the voltage optimization method of the present invention, the generating of the first scenario set based on the first target model and in combination with the first algorithm comprises:

[0019] Establishing a first probability density function based on the first target model;

[0020] Based on the first probability density function, performing first sampling and second sampling in combination with a first algorithm;

[0021] Composing a scene according to the first sampling and the second sampling results;

[0022] The results of all scenarios are recorded as the first scenario set.

[0023] As a preferred solution of the voltage optimization method described in the present invention, the first algorithm includes:

[0024] Performing a first sampling on a first probability density function of a first target model according to probability;

[0025] A second sampling is performed on the output magnitude obtained by the first target model.

[0026] As a preferred solution of the voltage optimization method described in the present invention, the first objective function and the first constraint condition include:

[0027] The first objective function is to minimize the sum of objective functions of several different scenarios;

[0028] The first constraint condition is a constraint condition for all scenarios;

[0029] The first constraint condition at least includes that the active-reactive control instruction in a certain scenario is the same as the active-reactive control instruction in the first optimization model.

[0030] As a preferred solution of the voltage optimization method of the present invention, the first sampling and the second sampling include:

[0031] The first sampling includes dividing the first probability density function of the first target model into equal parts according to probability into a first interval, and performing random sampling in a plurality of intervals after the first interval is divided into equal parts;

[0032] The second sampling includes dividing the output obtained by the first target model into a second interval, and selecting one or two samples from a plurality of intervals after the second interval is divided into equal parts.

[0033] In a second aspect, the present invention provides a voltage optimization system, comprising:

[0034] A first model building module, used to build a first target model, wherein the first target model is used to determine photovoltaic processing uncertainty;

[0035] A scene acquisition module, configured to generate a first scene set based on the first target model and in combination with a first algorithm, wherein the first scene set includes a plurality of scenes;

[0036] A second model building module, used for presetting a first prediction control strategy and building a first optimization model based on the first scenario set;

[0037] A solution module is used to solve the first optimization model to perform voltage optimization.

[0038] In a third aspect, the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned method when executing the computer program.

[0039] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the method described above when executed by a processor.

[0040] Compared with the prior art, the invention has the following beneficial effects: the invention proposes a voltage optimization method and system, establishes a first target model, and the first target model is used to determine the uncertainty of photovoltaic processing; based on the first target model, a first scenario set is generated in combination with a first algorithm, and the first scenario set includes several scenarios; a first prediction control strategy is preset, and a first optimization model is established based on the first scenario set; the first optimization model is solved to optimize the voltage. The voltage optimization system of the present application can effectively deal with the uncertainty of photovoltaic output, and by establishing the first target model and the first scenario set, the voltage fluctuation problem can be predicted and handled more accurately. In addition, by presetting the first prediction control strategy and establishing the first optimization model, the system can optimize the voltage in real time, thereby improving the stability of the power grid and the absorption capacity of new energy. In practical applications, the voltage optimization system of the present application can reduce the voltage over-limit problem caused by the fluctuation of photovoltaic output, reduce the risk of power grid operation, and improve the quality of electric energy. At the same time, the system can also reduce the dependence on traditional power grid regulation equipment and reduce the operating cost of the power grid. Specifically, a universal Gaussian mixture model is used to model the uncertainty of photovoltaic output, which is suitable for the characteristics of strong randomness and large fluctuation of photovoltaic output. The improved Latin hypercube important sampling method is used to generate scenarios, which not only reflects the uniformity of sampling, but also takes into account the comprehensiveness of extreme cases. The multi-prediction model predictive control method is used to optimize the multi-dimensional prediction scenario, so that the solution takes into account the uncertainty of photovoltaic output and improves the reliability of the system. This method can be applied to new multi-level distribution networks and is in line with the actual operation of existing distribution networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing 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 labor. Among them:

[0042] Figure 1 A method flow chart of a voltage optimization method and system provided by an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a multi-dimensional prediction model of a voltage optimization method and system provided by an embodiment of the present invention;

[0044] Figure 3 An internal structural diagram of a computer device of a voltage optimization method and system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0046] Example 1

[0047] Reference Figure 1-Figure 3 , is the first embodiment of the present invention, which provides a voltage optimization method and system, including:

[0048] There are some problems in the existing related technologies. For example, when dealing with the uncertainty of photovoltaic output, a deterministic model is often used, which leads to the neglect of the prediction error of new energy and fails to fully reflect the volatility of new energy.

[0049] The present application provides a method that can effectively solve the above-mentioned problems. Next, how to implement the voltage optimization method will be described in detail in combination with multiple embodiments.

[0050] Figure 1 A method flow chart of a voltage optimization method and system is shown, including:

[0051] S101, establishing a first target model, where the first target model is used to determine photovoltaic processing uncertainty;

[0052] In an optional embodiment, the first target model is to represent the uncertainty of the target photovoltaic output, and a probability model can be constructed to simulate the randomness of the photovoltaic output. The model can predict the fluctuation range of photovoltaic output based on historical data and real-time data, and adjust the voltage control strategy accordingly to adapt to the volatility of new energy. In addition, the model can also optimize the charging and discharging plan of the energy storage device based on the prediction results, thereby improving the stability and efficiency of the entire power system.

[0053] In an optional embodiment, the uncertainty of photovoltaic output can be modeled by using a general Gaussian mixture model, which can capture the multimodal characteristics of photovoltaic output, that is, the fluctuation pattern of photovoltaic output under different weather conditions. In this way, the system can more accurately predict the photovoltaic output under various weather conditions, thereby providing a more accurate reference for voltage control.

[0054] In an optional embodiment, uncertainty modeling can also be performed by using a neural network, which can learn the complex nonlinear relationship of photovoltaic output data to improve the accuracy of the prediction. Through training, the neural network can identify the key factors affecting photovoltaic output and assign weights to these factors to optimize the prediction results.

[0055] In an optional embodiment, the neural network model can also be updated in real time to adapt to changes in photovoltaic system performance and ensure the long-term effectiveness of the prediction model. In practical applications, the voltage optimization system can dynamically adjust the voltage level of the power system in combination with the prediction results of the neural network model to cope with the uncertainty of photovoltaic output, thereby ensuring the stability and reliability of power supply.

[0056] In an optional embodiment, historical data can also be analyzed by using a machine learning algorithm to identify and predict the pattern of photovoltaic output. The algorithm is able to process large amounts of data and extract key variables that affect photovoltaic output, such as temperature, humidity, cloud cover, etc. In this way, the system can adjust the prediction model in real time to adapt to changes in environmental conditions, further improving the accuracy of the prediction.

[0057] In an optional embodiment, the embodiment may also include a feedback mechanism for automatically adjusting model parameters according to the difference between the actual photovoltaic output and the predicted value to ensure continuous optimization of the prediction results.

[0058] In the embodiment of the present application, a general Gaussian mixture model is used to model the uncertainty of photovoltaic output. This modeling method can effectively capture the randomness and volatility of photovoltaic output. The Gaussian mixture model combines multiple Gaussian distributions to form a complex probability distribution, which can better describe the multi-peak and nonlinear characteristics of photovoltaic output.

[0059] In an optional embodiment, the model can be trained using historical data to determine the parameters of each Gaussian distribution, including the mean, variance, and mixing coefficient. In this way, the system can quantify the uncertainty of photovoltaic output and consider this uncertainty in the voltage optimization process to achieve more accurate and reliable voltage control.

[0060] For example, a general Gaussian mixture model is used to model the uncertainty of photovoltaic output as follows:

[0061]

[0062] Among them, ε pv is the prediction error of photovoltaic; K is the number of Gaussian distributions; is the weight coefficient; σ iis the standard deviation of the i-th Gaussian distribution; is the variance of the i-th Gaussian distribution; μ i The mathematical expectation of the i-th Gaussian distribution.

[0063] It should be noted that the establishment of the first objective model, which is used to determine the uncertainty of photovoltaic processing, can provide a more flexible and adaptable framework for voltage optimization. By establishing the first objective model, the system can model the uncertainty of photovoltaic output, so as to take into account the volatility of photovoltaic output in the voltage optimization process. The establishment of this model is based on the analysis of historical and real-time data of photovoltaic output, which can predict the fluctuation range of output and adjust the voltage control strategy accordingly. In addition, the model can also optimize the charging and discharging plan of energy storage equipment and improve the stability and efficiency of the power system. In practical applications, the voltage optimization system can dynamically adjust the voltage level of the power system in combination with the prediction results of the model to cope with the uncertainty of photovoltaic output, thereby ensuring the stability and reliability of power supply.

[0064] S102, generating a first scene set based on the first target model and in combination with the first algorithm, where the first scene set includes a plurality of scenes;

[0065] In an embodiment of the present application, generating a first scene set based on the first target model and in combination with the first algorithm includes:

[0066] Establishing a first probability density function based on a first target model;

[0067] Based on the first probability density function, performing first sampling and second sampling in combination with the first algorithm;

[0068] Composing a scene according to the first sampling and the second sampling results;

[0069] The results of all scenarios are recorded as the first scenario set.

[0070] In an optional embodiment, the first algorithm is to implement sampling of the first probability density function, so the first algorithm can be any suitable probability sampling algorithm. For example, a Latin hypercube sampling method can be used, which can provide a more uniform sample distribution than traditional random sampling, thereby improving the quality and efficiency of scene generation. Through Latin hypercube sampling, it can be ensured that the sample points in each dimension in the high-dimensional space are evenly distributed, which helps to capture the uncertainty characteristics of photovoltaic output.

[0071] In another optional embodiment, the first algorithm may also be a sampling logic preset by the user or relevant technical personnel, which may be customized based on specific scenario requirements to adapt to different photovoltaic output fluctuation patterns. For example, an adaptive sampling algorithm based on historical data and real-time data may be designed, which can dynamically adjust the sampling strategy according to the fluctuation characteristics of photovoltaic output, thereby generating a more realistic scenario set.

[0072] In another optional embodiment, the first algorithm may also be an improved Latin hypercube importance sampling method, which combines the uniformity of Latin hypercube sampling and the pertinence of importance sampling, and can more comprehensively cover the uncertainty space of photovoltaic output, while focusing on those extreme situations that have a greater impact on voltage fluctuations. Through this method, a scenario set that is both representative and pertinent can be generated, providing a solid data foundation for subsequent voltage optimization.

[0073] In the embodiment of the present application, because a general Gaussian mixture model is used to model the uncertainty of photovoltaic output, the model is realized as the first probability density function at the beginning of the design. Therefore, in the present application, the step of establishing the first probability density function based on the first target model can directly use the first target model as the first probability density function;

[0074] It should be noted that if the first target model is established by other methods, if the output of the first target model is not a probability, it is necessary to re-obtain the first probability density function of the first target model;

[0075] In an optional embodiment, there are many operations to obtain the probability density function of the model, for example, through the Monte Carlo simulation method. Monte Carlo simulation is a calculation method based on random sampling, which approximates complex mathematical problems through statistical analysis of a large number of random samples. In this embodiment, Monte Carlo simulation can be used to estimate the probability distribution of photovoltaic output, thereby obtaining the probability density function of the model. In this way, the possible changes of photovoltaic output under different conditions can be simulated, providing more accurate input data for the voltage optimization system.

[0076] In an optional embodiment, there are many operations to obtain the probability density function of the model, for example, by using the Bayesian inference method. Bayesian inference is a statistical method that uses prior knowledge and observation data to update the probability estimate of model parameters.

[0077] It should be noted that this application does not impose any restrictions on this operation, and relevant technical personnel can select the corresponding operation according to actual needs.

[0078] In the embodiment of the present application, the first algorithm includes:

[0079] Performing a first sampling on a first probability density function of a first target model according to probability;

[0080] A second sampling is performed on the output magnitude obtained by the first target model.

[0081] In an optional embodiment, the first sampling and the second sampling can use the same sampling method or different sampling strategies to adapt to different scene requirements. For example, the first sampling can use the Latin hypercube sampling method, while the second sampling can use the Monte Carlo simulation method. Through this combined sampling strategy, the uncertainty characteristics of photovoltaic output can be more comprehensively captured while ensuring sampling efficiency and quality.

[0082] In another optional embodiment, for example, the first sampling and the second sampling can also be optimized according to the characteristics of the photovoltaic output, for example, a more intensive sampling strategy is adopted during the period when the photovoltaic output fluctuates greatly, and the sampling frequency can be appropriately reduced during the relatively stable period. Through this flexible sampling method, it is possible to ensure that sufficient data points are obtained during the critical period to improve the accuracy and practicality of the scene set.

[0083] In the embodiment of the present application, the first sampling and the second sampling include:

[0084] The first sampling includes equally dividing the first probability density function of the first target model into a first interval according to probability, and performing random sampling in a plurality of intervals after the first interval is equally divided;

[0085] The second sampling includes dividing the output obtained by the first target model into a second interval, and selecting one or two samples in a plurality of intervals after the second interval is divided into equal parts.

[0086] For example, the first sampling is performed by dividing the probability density function of photovoltaic output into N equal parts according to probability. 1 intervals, in N 1 Random sampling is performed in the interval.

[0087] The second sampling divides the photovoltaic output into N according to the upper and lower limits. 2 intervals, in N 2 Importance sampling is performed in intervals, and 1 to 2 samples are selected in each interval.

[0088] These samples are randomly combined to generate multiple prediction scenarios.

[0089] It should be noted that based on the first target model, the first scenario set is generated in combination with the first algorithm. The first scenario set includes several scenarios that can provide a comprehensive view of the volatility of photovoltaic output. By generating multiple prediction scenarios, the voltage optimization system can simulate the changes in output under different conditions, so as to take these fluctuations into account in the voltage control strategy. These scenario sets provide rich data support for voltage optimization, enabling the system to more accurately predict and respond to the uncertainty of photovoltaic output, thereby improving the stability and efficiency of the entire power system.

[0090] S103, presetting a first prediction control strategy and establishing a first optimization model based on a first scenario set;

[0091] In an embodiment of the present application, establishing a first optimization model based on the first scene set includes:

[0092] Performing a first predictive control strategy on all scenes in a first scene set;

[0093] Obtaining a first optimization model after all scenarios are processed by a first predictive control strategy;

[0094] The first optimization model includes a first objective function and a first constraint condition.

[0095] In the embodiment of the present application, the first optimization model is a comprehensive model for all scenes in the first scene set.

[0096] In an optional embodiment, the first optimization model can use a linear programming method that can optimize the voltage control strategy based on the data in the scenario set. The linear programming method is implemented by minimizing or maximizing the objective function, which can be cost, loss or other performance indicators. At the same time, the first optimization model also contains a series of constraints that ensure that parameters such as voltage level, current size, equipment capacity, etc. during the optimization process are within a safe and operational range. In this way, the first optimization model can provide an economical and reliable voltage control solution for the power system.

[0097] In an optional embodiment, the solution process of the first optimization model may involve complex mathematical operations, so advanced algorithms such as interior point method, simplex method, etc. can be used to improve the solution efficiency and accuracy. These algorithms can handle large-scale optimization problems and find the optimal solution while satisfying all constraints.

[0098] In an optional embodiment, the interior point method is particularly suitable for processing large-scale linear programming problems. It avoids the limitation of the traditional simplex method in searching on the boundary by searching for the optimal solution inside the feasible domain, thereby improving the solution speed and stability.

[0099] In an optional embodiment, the simplex rule is another commonly used linear programming solution method, which gradually approaches the optimal solution in an iterative manner and is applicable to linear programming problems of various scales.

[0100] It should be noted that a suitable solution algorithm can be selected according to the scale and characteristics of the problem to ensure the solution efficiency and accuracy of the optimization model.

[0101] In the embodiment of the present application, the first optimization model is established to achieve dynamic adjustment of the voltage level of the power system to adapt to the uncertainty of photovoltaic output. By constructing a first optimization model containing multiple prediction scenarios, the system can comprehensively consider various possible output conditions and formulate the optimal voltage control strategy accordingly. This strategy can not only improve the stability and efficiency of the power system, but also reduce the risks caused by photovoltaic output fluctuations.

[0102] In the embodiment of the present application, the preset of the first prediction control strategy is based on an in-depth understanding of the volatility of photovoltaic output. By presetting a reasonable prediction control strategy, the system can respond quickly when the photovoltaic output fluctuates and adjust the voltage level to ensure the continuity and stability of power supply. The formulation of the prediction control strategy needs to comprehensively consider the historical data, real-time data, weather forecast and other information of photovoltaic output to ensure the accuracy and adaptability of the strategy.

[0103] In the embodiment of the present application, the first objective function and the first constraint condition are set to ensure that the optimization model achieves the optimal voltage control effect while meeting the power system operation requirements. The first objective function can be a combination of multiple performance indicators, such as minimizing voltage deviation, minimizing the number of charge and discharge times of energy storage equipment, etc. The first constraint condition includes voltage level restrictions, current size restrictions, equipment capacity restrictions, etc. These constraints ensure that the voltage control strategy in the optimization process is both safe and efficient.

[0104] In the embodiment of the present application, the first objective function and the first constraint condition include:

[0105] The first objective function is to minimize the sum of the objective functions of several different scenarios;

[0106] The first constraint condition is the constraint condition of all scenarios;

[0107] The first constraint condition at least includes that the active-reactive control instruction in a certain scenario is the same as the active-reactive control instruction in the first optimization model.

[0108] Exemplarily, the first predictive control strategy is an incremental proximal model predictive control strategy, which can be expressed as:

[0109]

[0110] Among them, α k >0 is the step size; i k is the scene index, i k = k mod S; k is the number of iterations; is the active-reactive control instruction after the k-1th iteration.

[0111] Exemplarily, the first objective function is to use the incremental proximal model predictive control strategy to establish a multi-prediction model predictive control voltage optimization model objective function for the generated multiple scenarios:

[0112]

[0113] Where S is the number of prediction scenarios; F i is the objective function of the i-th prediction scenario.

[0114] Exemplarily, the first constraint condition is to use the incremental proximal model predictive control strategy to establish a multi-scenario constraint of the multi-prediction model predictive control voltage optimization model for the generated multiple scenarios:

[0115]

[0116] in, is the active-reactive control command at time t of the i-th prediction scenario; (p, q) t It is the active-reactive control instruction that takes into account multiple prediction scenarios at time t.

[0117] It should be noted that presetting the first prediction control strategy and establishing the first optimization model based on the first scenario set can achieve real-time adjustment of the voltage level to adapt to the dynamic changes in the power grid load. By considering the prediction data under different scenarios, the optimization model can identify potential voltage deviations in advance and take corresponding control measures. In addition, the system can also dynamically update the prediction scenario set by monitoring the power grid status in real time, thereby improving the accuracy and reliability of voltage control. In this way, the voltage optimization system can not only optimize the current voltage level, but also predict and alleviate possible voltage problems in the future to ensure the stable operation of the power grid.

[0118] S104, solving the first optimization model to perform voltage optimization.

[0119] In an optional embodiment, the multi-prediction model predictive control voltage optimization model (i.e., the first optimization model) can be solved by, for example, a gurobi solver to achieve fast and accurate calculation results. The solver can handle large-scale optimization problems and supports a variety of optimization algorithms, thereby ensuring the efficiency and effectiveness of the optimization process.

[0120] In an optional embodiment, the voltage optimization system can receive real-time grid operation data through an interface connection with real-time grid data and input these data into the optimization model. Then, the solver calculates the optimal voltage control strategy based on the model and data to guide the operation of the grid.

[0121] In an optional embodiment, the first optimization model may be solved by, for example, a Cplex solver. The Cplex solver is also good at handling large-scale linear and nonlinear optimization problems and has a wide range of applications in the field of power system optimization. By using the Cplex solver, the voltage optimization system can further improve the calculation speed and optimization accuracy, thereby achieving more sophisticated and efficient voltage control in power grid operation.

[0122] In an optional embodiment, the first optimization model can also be solved by, for example, an Xpress solver. The Xpress solver is a powerful optimization software that provides a variety of algorithms to solve linear programming, integer programming, and mixed integer linear programming problems. In the voltage optimization system, the Xpress solver can handle complex constraints and objective functions and provide accurate optimization results. In addition, the Xpress solver also supports parallel computing and distributed processing, which gives it a significant advantage in processing large-scale data sets, and can quickly respond to real-time changes in the power grid, ensuring timely updates and execution of voltage control strategies. Through the application of these advanced solvers, the voltage optimization system can achieve precise control of the voltage level of the power system, thereby improving the stability and efficiency of the entire power grid.

[0123] In the embodiment of the present application, the gurobi solver is used to solve the multi-prediction model predictive control voltage optimization model (ie, the first optimization model), such as Figure 2 It is a schematic diagram of a multi-dimensional prediction model. Each figure is a prediction for a scenario, and t in the figure represents the same moment in time.

[0124] In summary, the present invention proposes a voltage optimization method, establishes a first target model, and the first target model is used to determine the uncertainty of photovoltaic processing; based on the first target model, a first scenario set is generated in combination with a first algorithm, and the first scenario set includes several scenarios; a first prediction control strategy is preset, and a first optimization model is established based on the first scenario set; the first optimization model is solved to optimize the voltage. The voltage optimization system of the present application can effectively deal with the uncertainty of photovoltaic output. By establishing the first target model and the first scenario set, the voltage fluctuation problem can be predicted and processed more accurately. In addition, by presetting the first prediction control strategy and establishing the first optimization model, the system can optimize the voltage in real time, thereby improving the stability of the power grid and the absorption capacity of new energy. In practical applications, the voltage optimization system of the present application can reduce the voltage over-limit problem caused by the fluctuation of photovoltaic output, reduce the risk of power grid operation, and improve the quality of electric energy. At the same time, the system can also reduce the dependence on traditional power grid regulation equipment and reduce the operating cost of the power grid. Specifically, a universal Gaussian mixture model is used to model the uncertainty of photovoltaic output, which is suitable for the characteristics of strong randomness and large fluctuation of photovoltaic output. The improved Latin hypercube important sampling method is used to generate scenarios, which not only reflects the uniformity of sampling, but also takes into account the comprehensiveness of extreme cases. The multi-prediction model predictive control method is used to optimize the multi-dimensional prediction scenario, so that the solution takes into account the uncertainty of photovoltaic output and improves the reliability of the system. This method can be applied to new multi-level distribution networks and is in line with the actual operation of existing distribution networks.

[0125] Example 2

[0126] In a preferred embodiment, a general Gaussian mixture model is used to model the uncertainty of photovoltaic output; an improved Latin hypercube importance sampling method is used to generate scenarios for the photovoltaic uncertainty output model; an incremental proximal model predictive control strategy is used to establish a multi-prediction model predictive control voltage optimization model for the generated multiple scenarios; and a gurobi solver is used to solve the multi-prediction model predictive control voltage optimization model.

[0127] Furthermore, a general Gaussian mixture model is used to model the uncertainty of photovoltaic output as follows:

[0128]

[0129] Where: ε pv is the prediction error of photovoltaic; K is the number of Gaussian distributions; is the weight coefficient; σ i is the standard deviation of the i-th Gaussian distribution; is the variance of the i-th Gaussian distribution; μ i The mathematical expectation of the i-th Gaussian distribution.

[0130] Furthermore, the method of using the improved Latin hypercube important sampling method to generate scenarios for the photovoltaic uncertainty output model is as follows:

[0131] First, the probability density function of photovoltaic output is divided into N 1 intervals, in N 1 Random sampling is performed in the interval.

[0132] Secondly, the photovoltaic output is divided into N according to the upper and lower limits. 2 intervals, in N 2 Importance sampling is performed in intervals, and 1 to 2 samples are selected in each interval.

[0133] Finally, these samples are randomly combined to generate multiple prediction scenarios.

[0134] Furthermore, the incremental proximal model predictive control strategy is used for the generated multiple scenarios to establish the multi-scenario constraints of the multi-prediction model predictive control voltage optimization model as follows:

[0135]

[0136] Where: is the active-reactive control command at time t of the i-th prediction scenario; (p, q) t It is the active-reactive control instruction that takes into account multiple prediction scenarios at time t.

[0137] Furthermore, the incremental proximal model predictive control strategy is used for the generated multiple scenarios to establish a multi-prediction model predictive control voltage optimization model with the objective function as follows:

[0138]

[0139] Where: S is the number of prediction scenarios; F i is the objective function of the i-th prediction scenario.

[0140] Furthermore, the incremental proximal model predictive control strategy is:

[0141]

[0142] Where: α k >0 is the step size; i k is the scene index, i k = k mod S; k is the number of iterations; is the active-reactive control instruction after the k-1th iteration.

[0143] In summary, the multi-level problem characteristics and regulation resources of the distribution network are considered, and the power quality and economic benefits are taken into account. The universal Gaussian mixture model is used to model the uncertainty of photovoltaic output, which is suitable for the characteristics of strong randomness and large fluctuations of photovoltaic output. The improved Latin hypercube important sampling method is used to generate scenarios, which not only reflects the uniformity of sampling, but also takes into account the comprehensiveness of extreme cases. The multi-prediction model predictive control method is used to optimize the multi-dimensional prediction scenario, so that the solution results take into account the uncertainty of photovoltaic output and improve the reliability of the system. This method can be applied to new multi-level distribution networks and is in line with the actual operation of existing distribution networks.

[0144] Example 3

[0145] This embodiment also provides a voltage optimization system, including:

[0146] A first model building module, used to build a first target model, the first target model is used to determine photovoltaic processing uncertainty;

[0147] A scene acquisition module, used to generate a first scene set based on a first target model and in combination with a first algorithm, wherein the first scene set includes a plurality of scenes;

[0148] A second model building module, used for presetting a first prediction control strategy and building a first optimization model based on a first scenario set;

[0149] The solution module is used to solve the first optimization model and perform voltage optimization.

[0150] The above-mentioned unit modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above-mentioned modules.

[0151] This embodiment also provides a computer device, which may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 3As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a voltage optimization method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball or a touch pad set on the computer device housing, or an external keyboard, touch pad or mouse, etc.

[0152] This embodiment further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0153] Establishing a first target model, the first target model is used to determine photovoltaic processing uncertainty;

[0154] Based on the first target model and in combination with the first algorithm, a first scene set is generated, where the first scene set includes a plurality of scenes;

[0155] Preset a first prediction control strategy and establish a first optimization model based on a first scenario set;

[0156] The first optimization model is solved to perform voltage optimization.

[0157] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0158] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0159] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0160] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0161] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0162] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0163] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A voltage optimization method, characterized in that: include: establishing a first target model, the first target model being used to determine photovoltaic processing uncertainty; Based on the first target model and in combination with a first algorithm, a first scene set is generated, where the first scene set includes a plurality of scenes; Presetting a first prediction control strategy and establishing a first optimization model based on the first scenario set; The first optimization model is solved to perform voltage optimization.

2. The voltage optimization method according to claim 1, characterized in that: The establishing of the first optimization model based on the first scene set comprises: Performing a first predictive control strategy process on all the scenarios in the first scenario set; Obtaining a first optimization model after all scenarios are processed by a first predictive control strategy; The first optimization model includes a first objective function and a first constraint condition.

3. The voltage optimization method according to claim 2, characterized in that: The first optimization model is a comprehensive model for all the scenarios in the first scenario set.

4. The voltage optimization method according to claim 3, characterized in that: The generating a first scene set based on the first target model and in combination with a first algorithm comprises: Establishing a first probability density function based on the first target model; Based on the first probability density function, performing first sampling and second sampling in combination with a first algorithm; Composing a scene according to the first sampling and the second sampling results; The results of all scenarios are recorded as the first scenario set.

5. The voltage optimization method according to claim 4, characterized in that: The first algorithm comprises: Performing a first sampling on a first probability density function of a first target model according to probability; A second sampling is performed on the output magnitude obtained by the first target model.

6. The voltage optimization method according to claim 5, characterized in that: The first objective function and the first constraint condition include: The first objective function is to minimize the sum of objective functions of several different scenarios; The first constraint condition is a constraint condition for all scenarios; The first constraint condition at least includes that the active-reactive control instruction in a certain scenario is the same as the active-reactive control instruction in the first optimization model.

7. The voltage optimization method according to claim 6, characterized in that: The first sampling and the second sampling include: The first sampling includes dividing the first probability density function of the first target model into equal parts according to probability into a first interval, and performing random sampling in a plurality of intervals after the first interval is divided into equal parts; The second sampling includes dividing the output obtained by the first target model into a second interval, and selecting one or two samples from a plurality of intervals after the second interval is divided into equal parts.

8. A voltage optimization system, characterized in that: include: A first model building module, used to build a first target model, wherein the first target model is used to determine photovoltaic processing uncertainty; A scene acquisition module, configured to generate a first scene set based on the first target model and in combination with a first algorithm, wherein the first scene set includes a plurality of scenes; A second model building module, used for presetting a first prediction control strategy and building a first optimization model based on the first scenario set; A solution module is used to solve the first optimization model to perform voltage optimization.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images