Parameter determination method and device for distributed photovoltaic cluster participating in auxiliary service, equipment, medium and product

Abstract

The invention relates to a parameter determination method and device for auxiliary service participation of a distributed photovoltaic cluster, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a distribution parameter constraint, historical prediction output error data and prediction output data; obtaining a minimum error interval according to distribution of historical prediction output error data; obtaining a predicted output interval according to the minimum error interval and the predicted output data; according to the distribution parameter constraint and the prediction output interval, constructing an upper-layer auxiliary service distribution efficiency optimal model and a lower-layer target cluster return optimal model, and obtaining a double-layer optimization model containing an auxiliary capacity parameter and a capacity distribution parameter; solving the double-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity distribution parameter; and obtaining a target parameter according to the target auxiliary capacity parameter and the target capacity distribution parameter. By adopting the method, the enthusiasm of the distributed photovoltaic cluster to participate in the auxiliary service can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for determining parameters of a distributed photovoltaic cluster participating in auxiliary services. Background Art

[0002] As the proportion of renewable energy power generation in the power grid continues to increase, the randomness and volatility of renewable energy power generation output have brought challenges to the safe and stable operation of the power grid, which has significantly increased the demand for auxiliary services such as peak load regulation and frequency regulation. In regional power grids with high penetration of distributed photovoltaics, distributed photovoltaics outside conventional units can participate in the auxiliary services of the power grid and obtain corresponding returns. Among them, in order to fully explore the peak load regulation and frequency regulation capabilities of distributed photovoltaics and promote their active participation in auxiliary services, it is necessary to study the relevant parameters of their participation in auxiliary services to maximize their returns from participating in auxiliary services.

[0003] However, in the relevant research on the participation of distributed photovoltaic clusters in ancillary services, the errors in the output data prediction of distributed photovoltaic clusters are usually not taken into account, nor is the impact of other factors in ancillary services on market competition considered, making it difficult to maximize returns. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for determining parameters of a distributed photovoltaic cluster participating in auxiliary services in order to address the above technical issues.

[0005] In a first aspect, the present application provides a method for determining parameters of a distributed photovoltaic cluster participating in ancillary services, including:

[0006] Obtaining allocation parameter constraints of the auxiliary service, obtaining historical predicted output error data of the target cluster and predicted output data in a target auxiliary period;

[0007] According to the distribution of the historical predicted output error data, a minimum error interval for predicting the output data of the target cluster under a preset confidence level is obtained;

[0008] Obtaining a predicted output interval of the target cluster in the target auxiliary period according to the minimum error interval and the predicted output data;

[0009] According to the allocation parameter constraints and the predicted output range, an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model are constructed to obtain a two-layer optimization model including auxiliary capacity parameters and capacity allocation parameters of the target cluster;

[0010] Solving the two-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity allocation parameter of the target cluster;

[0011] A target parameter for the target cluster to participate in the auxiliary service in the target auxiliary period is obtained according to the target auxiliary capacity parameter and the target capacity allocation parameter.

[0012] In one of the embodiments, the minimum error interval for predicting the output data of the target cluster under a preset confidence is obtained based on the distribution of the historical predicted output error data, including: using a skewness and kurtosis test method to test whether the historical predicted output error data conforms to a normal distribution; if the historical predicted output error data does not conform to a normal distribution, using a kernel density estimation method to estimate the probability distribution of the historical predicted output error data to obtain a probability density function of the historical predicted output error data; based on the probability density function and the preset confidence, obtaining the minimum value interval of the historical predicted output error data under the preset confidence; based on the minimum value interval, obtaining the minimum error interval for predicting the output data of the target cluster.

[0013] In one of the embodiments, obtaining the historical predicted output error data of the target cluster includes: obtaining the historical predicted output data, historical measured output data and historical installed capacity data of the target cluster at multiple historical moments; obtaining the historical predicted deviation value at the historical moment based on the historical predicted output data and the historical measured output data at each historical moment; standardizing the corresponding historical predicted deviation value based on the historical installed capacity data at each historical moment to obtain the historical predicted deviation standard value at each historical moment; obtaining the historical predicted output error data based on each historical prediction deviation standard value.

[0014] In one of the embodiments, according to the allocation parameter constraint and the predicted output range, an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster reward optimal model are constructed to obtain a two-layer optimization model including the auxiliary capacity parameters and capacity allocation parameters of the target cluster, including: taking the capacity allocation efficiency of the auxiliary service as the optimal upper-layer optimization goal, constructing the upper-layer auxiliary service allocation efficiency optimal model including the auxiliary capacity parameters and the capacity allocation parameters, the collaborative auxiliary capacity parameters and collaborative capacity allocation parameters of the collaborative auxiliary clusters that collaboratively participate in the auxiliary service, and the demand capacity parameters and demand capacity allocation parameters of the auxiliary service; wherein the capacity allocation parameters satisfy the allocation parameter constraints, and the auxiliary The auxiliary capacity parameter, the collaborative auxiliary capacity parameter and the demand capacity parameter satisfy the capacity coupling constraint, and the collaborative auxiliary capacity parameter satisfies the collaborative capacity value constraint; according to the predicted output range, the auxiliary capacity constraint of the target cluster in the target auxiliary period is obtained; so that the return value of the target cluster in the target auxiliary period is maximized as the lower-level optimization target, and the optimal return model of the lower-level target cluster including the auxiliary capacity parameter and the allocation return parameter is constructed; wherein, the auxiliary capacity parameter satisfies the auxiliary capacity constraint, and the allocation return parameter is the dual variable of the capacity coupling constraint; according to the upper-level auxiliary service allocation efficiency optimal model and the lower-level target cluster return optimal model, the two-layer optimization model is obtained.

[0015] In one of the embodiments, solving the two-layer optimization model to obtain the target auxiliary capacity parameters and target capacity allocation parameters of the target cluster includes: according to the dual relationship between the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model, using the Carlo-Kuhn-Tucker condition to convert the two-layer optimization model into a single-layer planning optimization model with equilibrium constraints; solving the single-layer planning optimization model to obtain the target auxiliary capacity parameters and the target capacity allocation parameters of the target cluster.

[0016] In one of the embodiments, according to the dual relationship between the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model, the Carlo-Kuhn-Tucker condition is used to convert the two-layer optimization model into a single-layer planning optimization model with equilibrium constraints, including: constructing a first dual variable of the allocation parameter constraint and a second dual variable of the collaborative capacity value constraint of the upper-layer auxiliary service allocation efficiency optimal model; according to the first dual variable, the second dual variable and the allocation reward parameter, using the Carlo-Kuhn-Tucker condition to convert the upper-layer optimization objective of the upper-layer auxiliary service allocation efficiency optimal model and the allocation parameter constraint, the collaborative capacity value constraint, and the capacity coupling constraint into the equilibrium constraint of the lower-layer target cluster reward optimal model; according to the lower-layer optimization objective, the equilibrium constraint and the auxiliary capacity constraint of the lower-layer target cluster reward optimal model, the single-layer planning optimization model is obtained.

[0017] In a second aspect, the present application also provides a parameter determination device for a distributed photovoltaic cluster to participate in auxiliary services, including:

[0018] A data acquisition module, used to acquire the allocation parameter constraints of the auxiliary service, acquire the historical predicted output error data of the target cluster and the predicted output data in the target auxiliary period;

[0019] An error interval acquisition module, used to obtain a minimum error interval for predicting the output data of the target cluster under a preset confidence level according to the distribution of the historical predicted output error data;

[0020] An output interval acquisition module, configured to obtain a predicted output interval of the target cluster in the target auxiliary period according to the minimum error interval and the predicted output data;

[0021] A model building module, used to build an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model according to the allocation parameter constraints and the predicted output range, and obtain a two-layer optimization model including the auxiliary capacity parameters and capacity allocation parameters of the target cluster;

[0022] A model solving module, used to solve the two-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity allocation parameter of the target cluster;

[0023] The target parameter acquisition module is used to obtain the target parameter of the target cluster participating in the auxiliary service in the target auxiliary period according to the target auxiliary capacity parameter and the target capacity allocation parameter.

[0024] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0025] Obtaining allocation parameter constraints of the auxiliary service, obtaining historical predicted output error data of the target cluster and predicted output data in a target auxiliary period;

[0026] According to the distribution of the historical predicted output error data, a minimum error interval for predicting the output data of the target cluster under a preset confidence level is obtained;

[0027] Obtaining a predicted output interval of the target cluster in the target auxiliary period according to the minimum error interval and the predicted output data;

[0028] According to the allocation parameter constraints and the predicted output range, an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model are constructed to obtain a two-layer optimization model including auxiliary capacity parameters and capacity allocation parameters of the target cluster;

[0029] Solving the two-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity allocation parameter of the target cluster;

[0030] A target parameter for the target cluster to participate in the auxiliary service in the target auxiliary period is obtained according to the target auxiliary capacity parameter and the target capacity allocation parameter.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0032] Obtaining allocation parameter constraints of the auxiliary service, obtaining historical predicted output error data of the target cluster and predicted output data in a target auxiliary period;

[0033] According to the distribution of the historical predicted output error data, a minimum error interval for predicting the output data of the target cluster under a preset confidence level is obtained;

[0034] Obtaining a predicted output interval of the target cluster in the target auxiliary period according to the minimum error interval and the predicted output data;

[0035] According to the allocation parameter constraints and the predicted output range, an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model are constructed to obtain a two-layer optimization model including auxiliary capacity parameters and capacity allocation parameters of the target cluster;

[0036] Solving the two-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity allocation parameter of the target cluster;

[0037] A target parameter for the target cluster to participate in the auxiliary service in the target auxiliary period is obtained according to the target auxiliary capacity parameter and the target capacity allocation parameter.

[0038] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0039] Obtaining allocation parameter constraints of the auxiliary service, obtaining historical predicted output error data of the target cluster and predicted output data in a target auxiliary period;

[0040] According to the distribution of the historical predicted output error data, a minimum error interval for predicting the output data of the target cluster under a preset confidence level is obtained;

[0041] Obtaining a predicted output interval of the target cluster in the target auxiliary period according to the minimum error interval and the predicted output data;

[0042] According to the allocation parameter constraints and the predicted output range, an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model are constructed to obtain a two-layer optimization model including auxiliary capacity parameters and capacity allocation parameters of the target cluster;

[0043] Solving the two-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity allocation parameter of the target cluster;

[0044] A target parameter for the target cluster to participate in the auxiliary service in the target auxiliary period is obtained according to the target auxiliary capacity parameter and the target capacity allocation parameter.

[0045] The above-mentioned parameter determination method, device, computer equipment, computer-readable storage medium and computer program product for the participation of distributed photovoltaic clusters in auxiliary services first obtain the allocation parameter constraints of the auxiliary services, the historical predicted output error data of the target cluster and the predicted output data in the target auxiliary period, and then obtain the minimum error interval for predicting the output data of the target cluster under a preset confidence level based on the distribution of the historical predicted output error data, and obtain the predicted output interval of the target cluster in the target auxiliary period based on the minimum error interval and the predicted output data, and then construct an upper-level auxiliary service allocation efficiency optimal model and a lower-level target cluster return optimal model based on the allocation parameter constraints and the predicted output interval, and obtain a two-layer optimization model including the auxiliary capacity parameters and capacity allocation parameters of the target cluster, and obtain the target auxiliary capacity parameters and target capacity allocation parameters of the target cluster by solving the two-layer optimization model, and finally obtain the target parameters for the target cluster to participate in the auxiliary service in the target auxiliary period based on the target auxiliary capacity parameters and the target capacity allocation parameters. This scheme analyzes the distribution of the historical forecast output error data of the target cluster, and obtains the minimum error interval for predicting the output data of the cluster according to the preset confidence level. In this way, the output of the target cluster in the target auxiliary period can be more accurately estimated by combining the forecast output data of the target cluster in the target auxiliary period and the minimum error interval, and a forecast output interval with more reference value can be obtained. At the same time, by constructing a two-layer optimization model including the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster return optimal model, the scheme can take into account the overall returns of the auxiliary service supply and demand sides in the target auxiliary period and the returns obtained by the target cluster. By optimizing and solving the two-layer optimization model, the target auxiliary capacity parameters and target capacity allocation parameters that can achieve the best balance between the two objectives can be obtained. The target parameters obtained can help maximize the return of the target cluster participating in the auxiliary service, and effectively improve the enthusiasm of the distributed photovoltaic cluster to participate in the auxiliary services of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0047] FIG1 is a flow chart of a method for determining parameters of a distributed photovoltaic cluster participating in auxiliary services in one embodiment;

[0048] FIG2 is a schematic diagram of the structure of a two-layer optimization model in one embodiment;

[0049] FIG3 is a schematic diagram of a process for obtaining a minimum error interval in one embodiment;

[0050] FIG4 is a schematic diagram of a process for obtaining historical predicted output error data in one embodiment;

[0051] FIG5 is a schematic diagram of a process for obtaining a double-layer optimization model in one embodiment;

[0052] FIG6 is a schematic diagram of a process of converting a two-layer optimization model into a single-layer planning optimization model with equilibrium constraints in one embodiment;

[0053] FIG7 is a structural block diagram of a device for determining parameters of a distributed photovoltaic cluster participating in auxiliary services in one embodiment;

[0054] FIG. 8 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] In one embodiment, as shown in FIG1 , a method for determining parameters of a distributed photovoltaic cluster participating in auxiliary services is provided. This embodiment uses the method applied to a server as an example for illustration. It is understandable that the method can also be applied to a terminal, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0057] Step S101 : obtaining allocation parameter constraints of auxiliary services, obtaining historical predicted output error data of a target cluster and predicted output data in a target auxiliary period.

[0058] The target cluster may be a distributed photovoltaic cluster including multiple distributed photovoltaic devices, which may participate in the auxiliary services of the regional power grid in the region where it is located, and the target auxiliary period may be the period during which the target cluster needs to participate in the auxiliary services. Exemplarily, the auxiliary services participated in by the target cluster may include valley filling, peak shaving, etc.

[0059] Exemplarily, when a distributed photovoltaic cluster participates in ancillary services, auxiliary capacity parameters and capacity allocation parameters may be declared. Optionally, the auxiliary capacity parameters may include valley filling capacity and peak shaving capacity that the distributed photovoltaic cluster may provide when participating in ancillary services. Optionally, the capacity allocation parameters may be used to indicate the return that the distributed photovoltaic cluster expects to receive for participating in ancillary services. Exemplarily, when the ancillary services participated in by the target cluster include valley filling and peak shaving, the capacity allocation parameters may include a first expected return value corresponding to a unit valley filling capacity and a second expected return value corresponding to a unit peak shaving capacity.

[0060] The allocation parameter constraint of the auxiliary service can be used to constrain the value of the capacity allocation parameter set by each distributed photovoltaic cluster participating in the auxiliary service. Exemplarily, when the capacity allocation parameter may include a first expected return value corresponding to a unit valley filling capacity and a second expected return value corresponding to a unit peak shaving capacity, the allocation parameter constraint may include a first return upper limit and a second return upper limit that respectively limit the first expected return value and the second expected return value.

[0061] The predicted output data of the target cluster in the target auxiliary period may include the predicted values ​​of the upper and lower limits of the output of the target cluster at each moment in the target auxiliary period. The historical predicted output error data of the target cluster may include the deviation data between the predicted values ​​obtained by predicting the output of the target cluster at multiple historical moments and the corresponding measured data.

[0062] It is understandable that the allocation parameter constraints of the auxiliary service and the historical predicted output error data and predicted output data of the target cluster may be pre-collected or counted and stored in a database. In this step, these data may be directly obtained from the database.

[0063] In some exemplary embodiments, a nonlinear auto-regressive neural network with exogenous inputs (NARX) may be used before this step to predict the predicted output data of the target cluster in the target auxiliary period.

[0064] Specifically, the NARX neural network is a dynamic neural network with memory and dynamic feedback functions. It can memorize and store historical data, so that historical output data can be calculated together with future output data, so that the network has dynamic performance and is not easy to lose information.

[0065] For example, the NARX neural network used in this embodiment may include a delay layer, an input layer, a hidden layer, and an output layer. , output data Perform c-order delay processing and provide it to the input layer. The input layer can provide the input data to the hidden layer. The hidden layer can process the data provided by the input layer and provide the processing results to the output layer. The output layer can output the final predicted processing data based on the processing results provided by the hidden layer.

[0066] Exemplarily, the process by which the hidden layer processes the data provided by the input layer can be expressed as:

[0067]

[0068] in, is the data output by the jth neuron in the hidden layer, is the number of neurons in the hidden layer, is the number of neurons in the input layer, m is the number of neurons in the output layer, is the connection weight between the input layer and the hidden layer, is the connection weight between the delay layer and the hidden layer, The input layer provides data to the hidden layer, The delay layer performs delay processing on the output data and provides the data to the hidden layer. is the hidden layer threshold, is the activation function of the hidden layer.

[0069] Exemplarily, the process of the output layer processing the data provided by the hidden layer can be expressed as:

[0070]

[0071] in, The output layer The output data of each neuron is is the data output by the jth neuron in the hidden layer, is the number of neurons in the hidden layer, is the connection weight between the hidden layer and the output layer, is the output layer threshold, is the number of neurons in the output layer, is the transfer function g between the hidden layer and the output layer.

[0072] Using the NARX neural network model, the output data of the distributed photovoltaic cluster in the target period can be predicted by inputting the historical output data, temperature, humidity, electricity price and other information of the distributed photovoltaic cluster.

[0073] The historical data samples and the corresponding result samples can be used to train the NARX neural network. Input the NARX neural network to be trained, and then calculate the result sample corresponding to the historical data sample. , and the model output results To calculate the model error:

[0074]

[0075] in, is the overall error of the model, The output layer The output of each neuron With the corresponding result sample The error value between is the number of neurons in the output layer.

[0076] Then, the weight parameters in the NARX neural network can be adjusted according to the above error values. Exemplarily, the adjustment process can be expressed as:

[0077]

[0078]

[0079] in, is the connection weight between the input layer and the hidden layer, is the connection weight between the delay layer and the hidden layer, is the connection weight between the hidden layer and the output layer, The learning rate for model training.

[0080] Step S102 : obtaining a minimum error interval for predicting the output data of the target cluster under a preset confidence level according to the distribution of historical predicted output error data.

[0081] According to the historical predicted output error data of the target cluster, statistics can be performed to obtain the probability distribution of the multiple deviation data contained therein. Then, the minimum error interval that can meet the preset confidence can be calculated by combining the preset confidence and the distribution of the historical predicted output error data.

[0082] For example, assuming is the probability density function of the historical forecast output error data, and its cumulative distribution function can be expressed as:

[0083]

[0084] Based on this function, we can get the prediction error random variable X in the interval The probability of occurrence is:

[0085]

[0086] Based on this, the minimum probability interval model can be constructed:

[0087]

[0088] in, To preset the confidence level, Represents the interval length under the preset reliability. Using this model, we can get The minimum error interval.

[0089] Step S103, obtaining the predicted output range of the target cluster in the target auxiliary period according to the minimum error range and the predicted output data.

[0090] Among them, based on the minimum error interval calculated previously and the predicted output data of the target cluster in the target assistance period, the upper and lower limits of the predicted output of the target cluster in the target assistance period considering the prediction error can be obtained, so that its predicted output interval in the target assistance period can be obtained.

[0091] Exemplarily, the predicted output data of the target cluster in the target assistance period may be added to the upper and lower limits of the minimum error interval to obtain the predicted output interval of the target cluster in the target assistance period.

[0092] Step S104, constructing an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster reward optimal model according to allocation parameter constraints and predicted output range, and obtaining a two-layer optimization model including auxiliary capacity parameters and capacity allocation parameters of the target cluster.

[0093] Among them, in this step, according to the scenario in which the target cluster participates in auxiliary services, a two-layer optimization model including an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster reward optimal model can be constructed.

[0094] Exemplarily, the two-layer optimization model can be shown in Figure 2. Among them, the upper-layer auxiliary service allocation efficiency optimal model can be used to describe the auxiliary service market clearing process, and the lower-layer target cluster return optimal model can be used to describe the game process of the target cluster participating in the auxiliary service.

[0095] Optionally, the upper-layer auxiliary service allocation efficiency optimal model can take the realization of the optimal capacity allocation efficiency of auxiliary services within the target auxiliary period as the upper-layer optimization target, use the auxiliary capacity parameters and capacity allocation parameters of the target cluster, as well as the coordinated auxiliary capacity parameters and coordinated capacity allocation parameters of other distributed photovoltaic clusters that collaboratively participate in auxiliary services, as well as the capacity demand of the auxiliary service market and the allocation parameter constraints on the allocation parameters, to describe the auxiliary service market clearing process and generate clearing results. Among them, the auxiliary capacity parameters of the target cluster can be obtained based on the declaration information received from the optimal return model of the lower-layer target cluster.

[0096] Optionally, the optimal model for the return of the lower target cluster can take the maximum return value of the target cluster participating in the auxiliary service as the lower optimization target, and use the auxiliary capacity parameter and the allocation return parameter of the target cluster to describe the return obtained by the target cluster participating in the auxiliary service. Among them, the constraints that the auxiliary capacity parameter needs to meet can be determined by the predicted output range, and the allocation return parameter can be obtained based on the clearing results received from the upper-level auxiliary service allocation efficiency optimal model.

[0097] Step S105 , solving the two-layer optimization model to obtain the target auxiliary capacity parameters and target capacity allocation parameters of the target cluster.

[0098] Step S106: Obtain target parameters for the target cluster to participate in the auxiliary service in the target auxiliary period according to the target auxiliary capacity parameters and the target capacity allocation parameters.

[0099] Among them, in step S105, by optimizing and solving the two-layer optimization model, the auxiliary capacity parameters and capacity allocation parameters that can achieve a balance between the optimization objectives of the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model can be obtained, so that the target auxiliary capacity parameters and target capacity allocation parameters of the target cluster participating in the auxiliary service during the target auxiliary period can be obtained.

[0100] Then, in step S106, the target parameters of the target cluster may be obtained according to the target auxiliary capacity parameters and the target capacity allocation parameters. Optionally, the target parameters may include the target auxiliary capacity parameters and the target capacity allocation parameters.

[0101] It is understandable that the target cluster can use the target parameter as the declaration information for its participation in the auxiliary service during the target auxiliary period, so as to maximize the return value of participating in the auxiliary service while considering various influencing factors in the auxiliary service market.

[0102] In the above-mentioned parameter determination method for the participation of distributed photovoltaic clusters in auxiliary services, the distribution of the historical predicted output error data of the target cluster is analyzed, and the minimum error interval for predicting the output data of the cluster is obtained according to the preset confidence level, so that the output of the target cluster in the target auxiliary period can be more accurately estimated by combining the predicted output data of the target cluster in the target auxiliary period and the minimum error interval, and a more reference-worthy predicted output interval can be obtained. At the same time, by constructing a two-layer optimization model including an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model, the scheme can take into account the overall returns of both the supply and demand sides of the auxiliary services in the target auxiliary period and the returns obtained by the target cluster at the same time, so that by optimizing and solving the two-layer optimization model, the target auxiliary capacity parameters and target capacity allocation parameters that can achieve the best balance between the two objectives can be obtained, and the target parameters obtained can be conducive to maximizing the return of the target cluster participating in the auxiliary service, and effectively improve the enthusiasm of the distributed photovoltaic cluster to participate in the auxiliary services of the power grid.

[0103] In an exemplary embodiment, as shown in FIG3 , obtaining the minimum error interval for predicting the output data of the target cluster under a preset confidence according to the distribution of historical predicted output error data may include:

[0104] Step S301, using the skewness and kurtosis test method to test whether the historical predicted output error data conforms to the normal distribution.

[0105] Among them, according to the historical predicted output error data, the skewness-kurtosis test method can be used to conduct a normality test on it to obtain the corresponding sample skewness and sample kurtosis.

[0106] Among them, the sample skewness can reflect the symmetry information of the overall distribution of the historical forecast output error data, and the sample kurtosis can reflect the steepness of the overall distribution density curve of the historical forecast output error data near its peak value. If the measured sample skewness is not 0 and the kurtosis is not 3, it means that the historical forecast output error data does not conform to the normal distribution.

[0107] Step S302: If the historical predicted output error data does not conform to the normal distribution, the kernel density estimation method is used to estimate the probability distribution of the historical predicted output error data to obtain the probability density function of the historical predicted output error data.

[0108] Wherein, when it is found that the historical predicted output error data does not conform to the normal distribution, the kernel density estimation method can be used to estimate its probability distribution to obtain the corresponding probability density function. For example, the probability density function of the historical predicted output error data can be expressed as:

[0109]

[0110] in, It is the deviation data in the historical forecast output error data. is the probability density function of the historical forecast output error data, s is the sample size, h is the window width, is the kernel function.

[0111] Optionally, after obtaining the probability density function of the historical predicted output error data, the chi-square test method may be used to test the fitting effect of the function on the distribution characteristics of the historical predicted output error data.

[0112] Step S303, according to the probability density function and the preset confidence, obtain the minimum value interval of the historical predicted output error data under the preset confidence.

[0113] Among them, the probability density function based on historical predicted output error data is , we can get its cumulative distribution function:

[0114]

[0115] Based on this function, we can get the prediction error random variable X in the interval The probability of occurrence is: Based on this, the minimum probability interval model can be constructed as shown below:

[0116]

[0117] in, To preset the confidence level, Represents the interval length under the preset confidence level. Using this model, we can obtain the historical prediction output error data under the preset confidence level. The minimum value range of .

[0118] Step S304: obtaining a minimum error interval for predicting the output data of the target cluster according to the minimum value interval.

[0119] Among them, based on the minimum value interval of the historical predicted output error data obtained previously, it can be used as the minimum error interval for predicting the output data of the target cluster in this step.

[0120] In this embodiment, by testing whether the historical predicted output error data conforms to the normal distribution, and using the kernel density method to estimate its distribution when it does not conform to the normal test, a probability density function that more accurately reflects the probability distribution of the historical predicted output error data can be obtained. And according to the probability density function and the preset confidence, the minimum value interval of the historical predicted output error data under the preset confidence is obtained, so that the interval in which the predicted deviation data is most concentrated under a certain probability can be determined more accurately. Therefore, the minimum error interval for predicting the output data of the target cluster obtained by using the minimum value interval can better estimate the predicted output interval of the target cluster in the target auxiliary period, which is conducive to improving the accuracy of subsequent modeling.

[0121] In an exemplary embodiment, as shown in FIG4 , obtaining historical predicted output error data of a target cluster may include:

[0122] Step S401, obtaining historical predicted output data, historical measured output data and historical installed capacity data of the target cluster at multiple historical moments.

[0123] Step S402, obtaining the historical prediction deviation value at each historical moment according to the historical prediction output data and the historical measured output data at each historical moment.

[0124] Step S403, based on the historical installed capacity data at each historical moment, the corresponding historical forecast deviation value is standardized to obtain the historical forecast deviation standard value at each historical moment.

[0125] Step S404, obtaining historical prediction output error data according to each historical prediction deviation standard value.

[0126] Specifically, in this embodiment, historical predicted output data, historical measured output data and historical installed capacity data of the target cluster at multiple historical moments can be obtained in step S401, and then historical predicted output error data of the target cluster can be calculated based on these data in steps S402 to S404.

[0127] In step S401, for multiple historical moments at which the output of the target cluster has been predicted, the historical predicted output data, historical measured output data, and historical installed capacity data corresponding to each historical moment may be obtained respectively. The historical predicted output data corresponding to a certain historical moment may include the predicted values ​​of the upper and lower limits of the output of the target cluster at that moment, the historical measured output data may include the measured values ​​of the upper and lower limits of the output of the target cluster at that moment, and the historical installed capacity data may include the installed capacity of the target cluster at that moment.

[0128] Then, in step S402, the historical forecast deviation value corresponding to each historical moment can be calculated based on the historical forecast output data and the historical measured output data corresponding to each historical moment. Then, in step S403, the historical forecast deviation value at the same moment can be standardized based on the historical installed capacity data corresponding to each historical moment to obtain the historical forecast deviation standard value corresponding to each historical moment.

[0129] Exemplarily, the calculation process of the historical prediction deviation value and the historical prediction deviation standard value can be shown as follows:

[0130]

[0131] in, is the historical prediction output data corresponding to the target cluster at the i-th historical moment, is the historical installed capacity data of the target cluster (here it is assumed that the installed capacity of the target cluster remains unchanged). represents the predicted output standard value of the target cluster corresponding to the i-th historical moment, is the historical measured output data of the target cluster at the i-th historical moment, represents the historical forecast deviation value corresponding to the i-th historical moment, Represents the standard value of the historical forecast deviation corresponding to the i-th historical moment.

[0132] Then, in step S404, historical prediction output error data including these data may be obtained according to the historical prediction deviation standard values ​​corresponding to each historical moment.

[0133] In this embodiment, the corresponding historical prediction deviation value is calculated based on the historical predicted output data and the historical measured output data of the target cluster, and then the historical prediction deviation value is standardized in combination with the historical installed capacity data to obtain the historical prediction deviation standard value. The influence of the installed capacity of the target cluster on the error value can be excluded in the statistical analysis of the historical predicted output error data of the target cluster, and the distribution of the error data can be analyzed more accurately to obtain analysis results that are more meaningful for reference.

[0134] In an exemplary embodiment, as shown in FIG5 , based on the allocation parameter constraints and the predicted output range, an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster reward optimal model are constructed to obtain a two-layer optimization model including auxiliary capacity parameters and capacity allocation parameters of the target cluster, which may include:

[0135] Step S501 takes the optimization of the capacity allocation efficiency of the auxiliary service in the target auxiliary time period as the upper-layer optimization goal, and constructs an upper-layer auxiliary service allocation efficiency optimization model including auxiliary capacity parameters and capacity allocation parameters, collaborative auxiliary capacity parameters and collaborative capacity allocation parameters of collaborative auxiliary clusters collaboratively participating in the auxiliary service, and demand capacity parameters and demand capacity allocation parameters of the auxiliary service.

[0136] The upper optimization goal is to optimize the capacity allocation efficiency of auxiliary services in the target auxiliary period, which can refer to maximizing the social welfare of auxiliary services in the target auxiliary period. The maximum social welfare can correspond to the maximum sum of producer surplus and consumer surplus, that is, the total return of the target cluster and collaborative auxiliary cluster participating in the auxiliary service and the demand side of the auxiliary service in the auxiliary service market is maximized.

[0137] For example, for the upper-layer optimization goal of optimizing the capacity allocation efficiency of the auxiliary service in the target auxiliary period, its objective function can be reflected as minimizing negative social welfare. For example, the upper-layer objective function shown in the following formula can be constructed:

[0138]

[0139] in, , are the first expected return value per unit valley filling capacity and the second expected return value per unit peak shaving capacity in the capacity allocation parameters declared by the target cluster, corresponding to the t-th moment in the target auxiliary period, respectively. , are the valley filling capacity and peak shaving capacity at the tth moment included in the auxiliary capacity parameters declared by the target cluster, , are the first collaborative expected return value per unit valley filling capacity and the second collaborative expected return value per unit peak shaving capacity corresponding to the t-th moment in the collaborative capacity allocation parameters reported by the collaborative auxiliary cluster, , are the collaborative valley filling capacity and collaborative peak shaving capacity at the tth moment included in the collaborative capacity allocation parameters declared by the collaborative auxiliary cluster, , are the demand valley filling capacity and demand peak shaving capacity that the power system needs to purchase at the tth moment in the demand capacity parameters of the ancillary services, , are the first demand return value of the unit valley filling capacity and the second demand return value of the unit peak shaving capacity corresponding to the t-th moment in the demand capacity allocation parameter of the auxiliary service. , It is both the decision variable of the lower-level problem and the clearing result of the upper-level problem; , The first upper limit of return and the second upper limit of return may correspond to the first expected return value and the second expected return value in the allocation parameter constraint, respectively. The collaborative auxiliary capacity parameter and the collaborative capacity allocation parameter of the collaborative auxiliary cluster may be calculated based on the historical data of the collaborative auxiliary cluster, and the demand capacity parameter and the demand capacity allocation parameter of the auxiliary service may be data obtained in advance and released by the relevant departments of the power system or the power trading center.

[0140] Among them, each parameter in the upper-layer auxiliary service allocation efficiency optimal model needs to meet the corresponding constraints, including: the capacity allocation parameter needs to meet the allocation parameter constraint, the auxiliary capacity parameter, the coordinated auxiliary capacity parameter and the demand capacity parameter need to meet the capacity coupling constraint, and the coordinated auxiliary capacity parameter needs to meet the coordinated capacity value constraint. Exemplarily, the constraint can be shown as follows:

[0141]

[0142] Wherein, the allocation parameter constraint may include the first expected return value in the capacity allocation parameter: and the second expected return The corresponding value range constraints respectively; the capacity coupling constraint may include the valley filling capacity included in the auxiliary capacity parameter , the collaborative valley filling capacity included in the collaborative auxiliary capacity parameter , and the demand capacity parameters include the demand peak shaving capacity The first capacity coupling constraint for supply and demand balance, and the peak shaving capacity included in the auxiliary capacity parameter , the collaborative peak shaving capacity included in the collaborative auxiliary capacity parameter And the demand capacity parameters include the demand peak shaving capacity The second capacity coupling constraint of the supply and demand balance between the two; the collaborative capacity value constraint may include the collaborative valley filling capacity included in the collaborative auxiliary capacity parameter The corresponding first collaborative capacity value constraint and collaborative peak shaving capacity The corresponding second collaborative capacity value constraint. , are respectively the first allocation reward parameter corresponding to the unit valley filling capacity and the second allocation reward parameter corresponding to the unit peak shaving capacity in the allocation reward parameters, and are also the dual variables corresponding to the first capacity coupling constraint and the second capacity coupling constraint respectively; , are the dual variables corresponding to the first collaborative capacity value constraint and the second collaborative capacity value constraint respectively.

[0143] Step S502: obtaining the auxiliary capacity constraint of the target cluster in the target auxiliary period according to the predicted output range.

[0144] According to the predicted output range of the target cluster in the target auxiliary period, the auxiliary capacity constraint that the auxiliary capacity provided by the target cluster in the target auxiliary period needs to satisfy can be determined. For example, the auxiliary capacity constraint can be shown as follows:

[0145]

[0146] In the formula, is the predicted output range of the target cluster at the tth moment in the target auxiliary period, is the output value of the target cluster at the tth moment, , , , are the dual variables of the corresponding constraints respectively.

[0147] Step S503, taking the maximum reward value of the target cluster in the target auxiliary period as the lower-level optimization target, constructing a lower-level target cluster reward optimal model including auxiliary capacity parameters and allocation reward parameters.

[0148] The lower layer optimization goal is to maximize the reward value of the target cluster in the target auxiliary period, which may refer to maximizing the reward value obtained by the target cluster participating in the auxiliary service in the target auxiliary period. Exemplarily, based on the lower layer optimization goal, a lower layer objective function as shown in the following formula may be constructed:

[0149]

[0150] In the formula, , are the valley filling capacity and peak shaving capacity at the tth moment included in the auxiliary capacity parameters declared by the target cluster, , They are respectively the first allocation reward parameter corresponding to the unit valley filling capacity and the second allocation reward parameter corresponding to the unit peak shaving capacity in the allocation reward parameter. Among them, the auxiliary capacity parameter needs to satisfy the aforementioned auxiliary capacity constraint, and the allocation reward parameter is the dual variable of the aforementioned capacity coupling constraint.

[0151] Step S504, obtaining a two-layer optimization model according to the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model.

[0152] Based on the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model constructed in the aforementioned process, a two-layer optimization model for the target cluster to participate in auxiliary services can be obtained in this step.

[0153] In this embodiment, according to the upper-level optimization goal of optimizing the capacity allocation efficiency of auxiliary services, a corresponding upper-level auxiliary service allocation efficiency optimal model is constructed, which can comprehensively consider the common returns of the target cluster, the collaborative auxiliary cluster and the auxiliary service demander participating in the auxiliary service in the auxiliary service market, and accurately describe the clearing process of the auxiliary service market; and according to the lower-level optimization goal of maximizing the return value of the target cluster in the target auxiliary period, a corresponding lower-level target cluster return optimal model is constructed, and the model can be used to reflect the game process of the target cluster participating in the auxiliary service. Furthermore, the two-layer optimization model constructed according to the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster return optimal model can effectively reflect the different goals that the target cluster needs to achieve in participating in the auxiliary service and reflect the interaction between the two through the interaction between the upper and lower layers. Then, the subsequent optimization solution using the two-layer optimization model can obtain a decision-making plan that can better balance the two optimization goals, which is conducive to the target cluster to obtain greater returns in participating in the auxiliary service.

[0154] In an exemplary embodiment, solving a two-layer optimization model to obtain target auxiliary capacity parameters and target capacity allocation parameters of a target cluster may include: according to the dual relationship between an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model, using the Carlo-Kuhn-Tucker condition to transform the two-layer optimization model into a single-layer planning optimization model with equilibrium constraints; solving the single-layer planning optimization model to obtain target auxiliary capacity parameters and target capacity allocation parameters of the target cluster.

[0155] Specifically, according to the dual relationship between the upper-level auxiliary service allocation efficiency optimal model and the lower-level target cluster return optimal model in the two-layer optimization model, the Karush-Kuhn-Tucker Conditions (KKT conditions) can be used to convert the two-layer optimization model into a single-layer planning optimization model with equilibrium constraints. Then, by optimizing the single-layer planning optimization model, the two-layer optimization model can be optimized and solved to obtain the target auxiliary capacity parameters and target capacity allocation parameters that can balance the two optimization objectives.

[0156] Optionally, in an exemplary embodiment, as shown in FIG6 , according to the dual relationship between the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model, the Carlo-Kuhn-Tucker condition is used to transform the two-layer optimization model into a single-layer planning optimization model with equilibrium constraints, which may include:

[0157] Step S601, constructing a first dual variable of the collaborative capacity value constraint of the upper-layer auxiliary service allocation efficiency optimal model, and constructing a second dual variable of the auxiliary capacity constraint of the lower-layer target cluster reward optimal model.

[0158] Among them, the constraints that each parameter in the upper-layer auxiliary service allocation efficiency optimal model needs to meet can be shown as follows:

[0159]

[0160] The collaborative capacity value constraint may include the collaborative valley filling capacity included in the collaborative auxiliary capacity parameter. The corresponding first collaborative capacity value constraint and collaborative peak shaving capacity The first dual variable may include the dual variable corresponding to the first collaborative capacity value constraint. And the dual variable corresponding to the second collaborative capacity value constraint It can be understood that for the allocation parameter constraints in the upper-layer auxiliary service allocation efficiency optimal model, the corresponding dual variables may include , .

[0161] Among them, the auxiliary capacity constraint of the optimal model of the lower target cluster return can be expressed as follows:

[0162]

[0163] Based on this, the second dual variable corresponding to the auxiliary capacity constraint can include , , , .

[0164] Step S602, based on the first dual variable, the second dual variable and the allocation reward parameter, the Carlo-Kuhn-Tucker condition is used to convert the upper-level optimization objectives of the upper-level auxiliary service allocation efficiency optimal model as well as the allocation parameter constraints, the collaborative capacity value constraints, and the capacity coupling constraints into the equilibrium constraints of the lower-level target cluster reward optimal model.

[0165] Step S603, obtaining a single-layer planning optimization model according to the lower-layer optimization target, balance constraint and auxiliary capacity constraint of the lower-layer target cluster reward optimal model.

[0166] Among them, according to the first dual variable, the second dual variable and the allocation reward parameter, the dual relationship between the upper-level auxiliary service allocation efficiency optimal model and the lower-level target cluster reward optimal model can be determined. Therefore, in step S602, the KKT condition can be used to convert the upper-level optimization objective of the upper-level auxiliary service allocation efficiency optimal model and the corresponding allocation parameter constraints, collaborative capacity value constraints, and capacity coupling constraints into the equilibrium constraints of the lower-level target cluster reward optimal model. Then in step S603, the original two-level problem can be converted into an equivalent mathematical program with equilibrium constraints (MPEC) problem in combination with the lower-level optimization objective, equilibrium constraints and auxiliary capacity constraints of the lower-level target cluster reward optimal model, thereby obtaining a single-level planning optimization model equivalent to the two-level optimization model.

[0167] For example, the equivalent mathematical programming problem with equilibrium constraints after conversion can be shown as follows:

[0168]

[0169]

[0170]

[0171]

[0172] In the formula, Used to express complementary constraints, express .

[0173] Among them, after the conversion, an equivalent mathematical programming problem with equilibrium constraints is obtained, which can include the lower-level optimization objectives, equilibrium constraints and auxiliary capacity constraints of the optimal model of the lower-level target cluster return.

[0174] Optionally, for the converted mathematical programming problem containing equilibrium constraints, the complementary constraints therein may be subjected to equivalent linearization processing.

[0175] For example, the upper and lower limit constraints for valley filling capacity in the auxiliary capacity constraint of the target cluster can be linearized as follows: , , , , , .

[0176] For example, the upper and lower limit constraints for the peak shaving capacity in the auxiliary capacity constraints of the target cluster can be linearized as follows: , , , , , .

[0177] For example, the collaborative capacity value constraint can be linearized as follows: , , , , , .

[0178] After completing the above transformation, all complementary constraints of the single-level planning optimization model can be linearized. Then, the single-level planning optimization model can be solved using a related optimization solver to obtain the target auxiliary capacity parameters and target capacity allocation parameters that can balance the two optimization objectives.

[0179] In this embodiment, for the solution of the two-layer optimization model, according to the dual relationship between the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster return optimal model, the KKT condition is used to convert the upper-layer auxiliary service allocation efficiency optimal model into the equilibrium constraint of the lower-layer target cluster return optimal model. This can convert the solution of the two-layer optimization model into the solution of an equivalent single-layer planning optimization model, which can effectively simplify the model solution process and facilitate obtaining more accurate solution results.

[0180] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0181] Based on the same inventive concept, the embodiment of the present application also provides a parameter determination device for a distributed photovoltaic cluster participating in ancillary services for implementing the parameter determination method for a distributed photovoltaic cluster participating in ancillary services involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more distributed photovoltaic clusters participating in ancillary services provided below can refer to the limitations of the parameter determination method for a distributed photovoltaic cluster participating in ancillary services above, and will not be repeated here.

[0182] In an exemplary embodiment, as shown in FIG7 , a parameter determination device 700 for a distributed photovoltaic cluster to participate in auxiliary services is provided, comprising:

[0183] The data acquisition module 701 is used to acquire the allocation parameter constraints of the auxiliary service, acquire the historical predicted output error data of the target cluster and the predicted output data in the target auxiliary period;

[0184] The error interval acquisition module 702 is used to obtain the minimum error interval for predicting the output data of the target cluster under a preset confidence level according to the distribution of the historical predicted output error data;

[0185] An output range acquisition module 703 is used to obtain a predicted output range of the target cluster in the target auxiliary period according to the minimum error range and the predicted output data;

[0186] A model building module 704 is used to build an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster reward optimal model according to the allocation parameter constraints and the predicted output range, and obtain a two-layer optimization model including the auxiliary capacity parameters and capacity allocation parameters of the target cluster;

[0187] A model solving module 705 is used to solve the two-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity allocation parameter of the target cluster;

[0188] The target parameter acquisition module 706 is configured to obtain, according to the target auxiliary capacity parameter and the target capacity allocation parameter, a target parameter for the target cluster to participate in the auxiliary service during the target auxiliary period.

[0189] In an exemplary embodiment, the error interval acquisition module 702 is also used to: use the skewness and kurtosis test method to test whether the historical predicted output error data conforms to the normal distribution; if the historical predicted output error data does not conform to the normal distribution, use the kernel density estimation method to estimate the probability distribution of the historical predicted output error data to obtain the probability density function of the historical predicted output error data; based on the probability density function and the preset confidence, obtain the minimum value interval of the historical predicted output error data under the preset confidence; based on the minimum value interval, obtain the minimum error interval for predicting the output data of the target cluster.

[0190] In an exemplary embodiment, the data acquisition module 701 is also used to: obtain the historical predicted output data, historical measured output data and historical installed capacity data of the target cluster at multiple historical moments; obtain the historical prediction deviation value at the historical moment based on the historical predicted output data and the historical measured output data at each historical moment; standardize the corresponding historical prediction deviation value based on the historical installed capacity data at each historical moment to obtain the historical prediction deviation standard value at each historical moment; obtain the historical prediction output error data based on each historical prediction deviation standard value.

[0191] In an exemplary embodiment, the model building module 704 is further used to: take the capacity allocation efficiency of the auxiliary service as the upper optimization goal to be optimal, construct the upper auxiliary service allocation efficiency optimal model including the auxiliary capacity parameter and the capacity allocation parameter, the collaborative auxiliary capacity parameter and the collaborative capacity allocation parameter of the collaborative auxiliary cluster that collaboratively participates in the auxiliary service, and the demand capacity parameter and the demand capacity allocation parameter of the auxiliary service; wherein the capacity allocation parameter satisfies the allocation parameter constraint, the auxiliary capacity parameter, the collaborative auxiliary capacity parameter and the demand capacity parameter satisfy the capacity coupling constraint, and the collaborative auxiliary capacity parameter satisfies the collaborative capacity value constraint; obtain the auxiliary capacity constraint of the target cluster in the target auxiliary period according to the predicted output interval; take the return value of the target cluster in the target auxiliary period as the lower optimization goal to be maximized, construct the lower target cluster return optimal model including the auxiliary capacity parameter and the allocation return parameter; wherein the auxiliary capacity parameter satisfies the auxiliary capacity constraint, and the allocation return parameter is the dual variable of the capacity coupling constraint; obtain the double-layer optimization model according to the upper auxiliary service allocation efficiency optimal model and the lower target cluster return optimal model.

[0192] In an exemplary embodiment, the model solving module 705 is also used to: according to the dual relationship between the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model, use the Carlo-Kuhn-Tucker condition to convert the two-layer optimization model into a single-layer planning optimization model with equilibrium constraints; solve the single-layer planning optimization model to obtain the target auxiliary capacity parameters and the target capacity allocation parameters of the target cluster.

[0193] In an exemplary embodiment, the model solving module 705 is also used to: construct a first dual variable of the allocation parameter constraint and a second dual variable of the collaborative capacity value constraint of the upper-level auxiliary service allocation efficiency optimal model; according to the first dual variable, the second dual variable and the allocation return parameter, use the Carlo-Kuhn-Tucker condition to convert the upper-level optimization objective of the upper-level auxiliary service allocation efficiency optimal model and the allocation parameter constraint, the collaborative capacity value constraint, and the capacity coupling constraint into the equilibrium constraint of the lower-level target cluster return optimal model; according to the lower-level optimization objective, the equilibrium constraint and the auxiliary capacity constraint of the lower-level target cluster return optimal model, obtain the single-layer planning optimization model.

[0194] Each module in the above-mentioned distributed photovoltaic cluster participating in auxiliary service parameter determination device can be implemented in whole or in part by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0195] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG8. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as allocation parameter constraints, historical predicted output error data, and predicted output data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a parameter determination method for a distributed photovoltaic cluster participating in ancillary services is implemented.

[0196] Those skilled in the art will understand that the structure shown in FIG. 8 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0197] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0198] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0199] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0200] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0201] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0202] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0203] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for determining parameters of a distributed photovoltaic cluster participating in auxiliary services, characterized in that: The method comprises: Obtaining allocation parameter constraints of the auxiliary service, obtaining historical predicted output error data of the target cluster and predicted output data in a target auxiliary period; According to the distribution of the historical predicted output error data, a minimum error interval for predicting the output data of the target cluster under a preset confidence level is obtained; Obtaining a predicted output interval of the target cluster in the target auxiliary period according to the minimum error interval and the predicted output data; According to the allocation parameter constraints and the predicted output range, an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model are constructed to obtain a two-layer optimization model including auxiliary capacity parameters and capacity allocation parameters of the target cluster; Solving the two-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity allocation parameter of the target cluster; A target parameter for the target cluster to participate in the auxiliary service in the target auxiliary period is obtained according to the target auxiliary capacity parameter and the target capacity allocation parameter.

2. The method according to claim 1, characterized in that The obtaining, based on the distribution of the historical predicted output error data, a minimum error interval for predicting the output data of the target cluster under a preset confidence level includes: The skewness and kurtosis test method is used to test whether the historical predicted output error data conforms to the normal distribution; If the historical predicted output error data does not conform to the normal distribution, the probability distribution of the historical predicted output error data is estimated by using the kernel density estimation method to obtain the probability density function of the historical predicted output error data; According to the probability density function and the preset confidence, obtaining a minimum value interval of the historical predicted output error data under the preset confidence; The minimum error interval for predicting the output data of the target cluster is obtained according to the minimum value interval.

3. The method according to claim 1, characterized in that The obtaining of historical predicted output error data of the target cluster includes: Obtaining historical predicted output data, historical measured output data, and historical installed capacity data of the target cluster at multiple historical moments; According to the historical predicted output data and the historical measured output data at each of the historical moments, a historical predicted deviation value at the historical moment is obtained; According to the historical installed capacity data at each of the historical moments, the corresponding historical forecast deviation values ​​are standardized to obtain the historical forecast deviation standard values ​​at each of the historical moments; The historical prediction output error data is obtained according to each of the historical prediction deviation standard values.

4. The method according to any one of claims 1 to 3, characterized in that: The method of constructing an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster reward optimal model according to the allocation parameter constraints and the predicted output range, and obtaining a two-layer optimization model including auxiliary capacity parameters and capacity allocation parameters of the target cluster, includes: Taking the optimization of the capacity allocation efficiency of the auxiliary service as the upper-layer optimization goal, constructing the upper-layer auxiliary service allocation efficiency optimization model including the auxiliary capacity parameter and the capacity allocation parameter, the collaborative auxiliary capacity parameter and the collaborative capacity allocation parameter of the collaborative auxiliary cluster collaboratively participating in the auxiliary service, and the demand capacity parameter and the demand capacity allocation parameter of the auxiliary service; wherein the capacity allocation parameter satisfies the allocation parameter constraint, the auxiliary capacity parameter, the collaborative auxiliary capacity parameter and the demand capacity parameter satisfy the capacity coupling constraint, and the collaborative auxiliary capacity parameter satisfies the collaborative capacity value constraint; According to the predicted output range, obtaining the auxiliary capacity constraint of the target cluster in the target auxiliary period; Taking the maximum reward value of the target cluster in the target auxiliary period as the lower optimization target, constructing the optimal reward model of the lower target cluster including the auxiliary capacity parameter and the allocation reward parameter; wherein the auxiliary capacity parameter satisfies the auxiliary capacity constraint, and the allocation reward parameter is the dual variable of the capacity coupling constraint; The two-layer optimization model is obtained according to the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model.

5. The method according to claim 4, characterized in that Solving the two-layer optimization model to obtain the target auxiliary capacity parameter and the target capacity allocation parameter of the target cluster includes: According to the dual relationship between the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster return optimal model, the two-layer optimization model is transformed into a single-layer planning optimization model with equilibrium constraints using the Carlo-Kuhn-Tucker condition; The single-layer planning optimization model is solved to obtain the target auxiliary capacity parameter and the target capacity allocation parameter of the target cluster.

6. The method according to claim 5, characterized in that According to the dual relationship between the upper-layer auxiliary service allocation efficiency optimal model and the lower-layer target cluster reward optimal model, the Carlo-Kuhn-Tucker condition is used to transform the two-layer optimization model into a single-layer planning optimization model with equilibrium constraints, including: Constructing a first dual variable of the allocation parameter constraint and a second dual variable of the collaborative capacity value constraint of the upper layer auxiliary service allocation efficiency optimal model; According to the first dual variable, the second dual variable and the allocation reward parameter, the upper optimization objective of the upper auxiliary service allocation efficiency optimal model and the allocation parameter constraint, the collaborative capacity value constraint, and the capacity coupling constraint are converted into the equilibrium constraint of the lower target cluster reward optimal model using the Carlo-Kuhn-Tucker condition; The single-layer planning optimization model is obtained according to the lower-layer optimization target, the balance constraint and the auxiliary capacity constraint of the lower-layer target cluster reward optimal model.

7. A parameter determination device for a distributed photovoltaic cluster participating in auxiliary services, characterized in that: The device comprises: A data acquisition module, used to acquire the allocation parameter constraints of the auxiliary service, acquire the historical predicted output error data of the target cluster and the predicted output data in the target auxiliary period; An error interval acquisition module, used to obtain a minimum error interval for predicting the output data of the target cluster under a preset confidence level according to the distribution of the historical predicted output error data; An output interval acquisition module, configured to obtain a predicted output interval of the target cluster in the target auxiliary period according to the minimum error interval and the predicted output data; A model building module, used to build an upper-layer auxiliary service allocation efficiency optimal model and a lower-layer target cluster return optimal model according to the allocation parameter constraints and the predicted output range, and obtain a two-layer optimization model including the auxiliary capacity parameters and capacity allocation parameters of the target cluster; A model solving module, used to solve the two-layer optimization model to obtain a target auxiliary capacity parameter and a target capacity allocation parameter of the target cluster; The target parameter acquisition module is used to obtain the target parameter of the target cluster participating in the auxiliary service in the target auxiliary period according to the target auxiliary capacity parameter and the target capacity allocation parameter.

8. 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 6 are implemented.

9. 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 6 are implemented.

10. A computer program product, comprising a computer program, 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 6 are implemented.

Images