Power distribution network piecewise linearization joint opportunity constraint optimization scheduling method based on sample data, computer readable storage medium and program product

Abstract

A piecewise linearization joint opportunity constraint optimization scheduling method for a power distribution network based on sample data, a computer readable storage medium and a program product, comprising: establishing a joint opportunity constraint scheduling model, considering node voltage uncertainty and power injection imbalance risk, and taking a preset voltage out-of-limit probability as a boundary condition to obtain a joint opportunity constraint scheduling model; the operation risk of the system is controllable; using sample data and a Gaussian mixture model to precisely fit distribution characteristics of uncertainty variables, and converting chance constraints into linear constraints; in addition, a posterior iterative algorithm is adopted to dynamically adjust a joint default probability threshold, and conservative property and economical efficiency of the model are balanced. According to the method, the problem of uncertainty caused by access of renewable energy to the power distribution network can be solved, a generation power distribution scheme with both safety and economy is generated, the risks of node voltage out-of-limit and line blockage are effectively reduced, and safe and economical operation of the power distribution network is achieved.

Description

Technical Field

[0001] The invention relates to a renewable energy power generation distribution network dispatching technology, and in particular to a distribution network piecewise linearization joint opportunity constraint optimization dispatching method based on sample data. Background Art

[0002] At present, more and more distributed photovoltaic, distributed wind power and other renewable energy generation resources are connected to the distribution network. If the deterministic optimization scheduling method is adopted, the uncertainty of renewable energy will bring safety problems such as voltage over-limit and line blockage to the operation of the distribution network. In order to ensure the safe and economic operation of the distribution network under the background of increasing penetration of renewable energy, it is necessary to fully consider the impact of uncertainty in the optimization scheduling process.

[0003] Common optimization methods that consider uncertainty include scenario-based stochastic optimization, robust optimization, and chance-constrained programming. Stochastic optimization generates a large number of scenarios or selects typical scenarios to represent uncertainty, and optimizes and solves the problem by allocating the weight of the objective function according to the probability of different scenarios. This method has the problems of low computational efficiency and the assumed uncertainty distribution being incompatible with reality. Robust optimization does not require the distribution of the sample, but uses the sample data to construct an uncertainty set, requiring that the constraints must be met within the range of this uncertainty set. Although this method can ensure robustness and does not require an uncertainty distribution function, it is too conservative and lacks economy.

[0004] Chance constraint is an intuitive modeling method that considers uncertainty risk. It allows the constraint to be violated with a small probability so that the system operator can effectively balance robustness and optimality according to preferences. However, because the chance constraint is an implicit expression, it needs to be transformed to be solved. The main methods for dealing with chance constraints include scenario-based methods, sampling average approximation methods, and analytical formula-based methods. Among them, the scenario-based method requires a large number of data samples to be sampled, and requires that a certain number of sampled samples must meet the constraint conditions. This method has the problem of low computational efficiency and the economic efficiency of the optimization scheduling results is getting lower and lower as the number of samples increases. The sampling average approximation method approximates the probability of violation of the constraint of the sampled samples by introducing binary variables, which reduces the conservatism to a certain extent compared with the scenario-based method, but still requires a large number of sampling and introduces binary variables, causing the model to become a mixed integer programming problem. When there are more model conditions, its computational efficiency is too low. The method based on analytical formula assumes that the distribution function of the known uncertainty variable is a standard Gaussian distribution or other common distribution. The cumulative probability distribution function is solved to obtain the value of the corresponding quantile of the allowable risk value to solve the chance constraint. However, in practice, the distribution function of the uncertainty variable will not exactly conform to the specific distribution function, and the practical feasibility of this method is poor.

[0005] Therefore, as more and more distributed photovoltaic, distributed wind power and other renewable energy generation resources are connected to the distribution network, the traditional deterministic optimization scheduling method cannot handle the uncertainty of renewable energy, which will bring safety problems such as voltage over-limit and line blockage to the operation of the distribution network. In the context of the continuous increase in the penetration rate of renewable energy, it is necessary to fully consider the impact of uncertainty in the optimization and scheduling of the distribution network. At present, there is a lack of a method that can efficiently solve the problem of distribution network uncertainty optimization scheduling without assuming that the uncertainty satisfies a specific distribution function. In addition, stochastic optimization assumes a known distribution function of uncertainty, generates a large number of scenarios or selects typical scenarios to represent uncertainty, and optimizes and solves the problem according to the probability of different scenarios. This method has problems such as low computational efficiency and the assumed uncertainty distribution does not meet the actual situation. Furthermore, robust optimization does not require the distribution of samples, but uses the data of samples to construct an uncertainty set, requiring that the constraints must be met within the range of this uncertainty set. Although this method can ensure robustness and does not require a distribution function of uncertainty, it is too conservative and lacks economy. In addition, chance-constrained programming is a method that can effectively balance robustness and optimality, but it lacks an efficient solution method without assuming that the uncertainty variables meet a specific distribution function.

[0006] It should be noted that the information disclosed in the above background technology section is only used for understanding the background of the present application, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the invention

[0007] The main purpose of the present invention is to overcome the defects existing in the above-mentioned background technology and provide a distribution network piecewise linearization joint opportunity constrained optimization scheduling method based on sample data.

[0008] To achieve the above object, the present invention adopts the following technical solutions: A distribution network piecewise linearization joint opportunity constraint optimization scheduling method based on sample data includes the following steps: S1. Establish a joint opportunity constraint dispatch model for distribution network: Based on the uncertainty of node voltage in distribution network and the risk of unbalanced power injection, a joint opportunity constraint model is constructed, with the preset node voltage over-limit probability as the boundary condition to ensure that the system operation risk is controllable; S2. Piecewise linearization of joint chance constraints: The distribution characteristics of uncertain variables are fitted using sample data, the joint probability distribution of uncertain variables is accurately fitted through the Gaussian mixture model, and the piecewise linearization technique is used to transform the nonlinear joint chance constraints into linear constraints that can be solved analytically. S3. Joint default probability adjustment based on posterior iteration: Based on historical data and initial dispatch solutions, the joint default probability threshold is dynamically corrected through the posterior iteration algorithm to balance the conservatism and economy of the model. The distribution network optimization dispatch model is solved by combining the linearization constraints and the adjusted default probability to generate a power distribution plan that takes into account both safety and economy, and reduce the risks of node voltage exceeding the limit and line congestion.

[0009] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the distribution network piecewise linearization combined opportunity-constrained optimization scheduling method based on sample data.

[0010] A computer program product comprises a computer program, wherein when the computer program is executed by a processor, the method for distributing a distribution network piecewise linearization combined with opportunity-constrained optimization scheduling based on sample data is implemented.

[0011] The present invention has the following beneficial effects: Against the background of increasing penetration of renewable energy, the present invention provides an optimization scheduling method that can efficiently handle uncertainties in distribution networks. This method can accurately fit the distribution characteristics of uncertain variables through a Gaussian mixture model based on sample fitting, without assuming that these variables satisfy a specific distribution function. In addition, the method uses piecewise linearization technology to convert nonlinear chance constraints that are originally difficult to solve into easy-to-handle linear constraints, thereby improving computational efficiency. Through a joint default probability adjustment method based on posterior iteration, the present invention solves the conservative problem existing in existing methods, so that the model has better economy while ensuring system safety. The method of the present invention not only improves the safe and economical operation of the distribution network, but also reduces the risks of node voltage over-limit and line blocking, realizes the efficient solution of the joint chance constraint optimization scheduling model, and meets the requirements of the distribution network for safety and reliability under uncertainty conditions.

[0012] Other beneficial effects of the embodiments of the present invention will be further described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 Flowchart of the distribution network piecewise linearization combined with chance-constrained optimization scheduling method based on sample data implemented by the present invention. DETAILED DESCRIPTION

[0014] The following is a detailed description of the embodiments of the present invention. It should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention.

[0015] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0016] See also Figure 1 The embodiment of the present invention provides a distribution network piecewise linearization joint opportunity constrained optimization scheduling method based on sample data, comprising the following steps: Step S1, establishing a joint opportunity constraint dispatch model for the distribution network: Based on the uncertainty of node voltage in the distribution network and the risk of unbalanced power injection, a joint opportunity constraint model is constructed, with the preset node voltage over-limit probability as the boundary condition to ensure that the system operation risk is controllable.

[0017] In a preferred embodiment, step S1 specifically includes: constructing an optimization model with the goal of minimizing the operating cost of the distribution network, which includes the cost of purchasing electricity in the wholesale market and the cost of distributed generators (DG); modeling the power generation cost of DG as a quadratic function to reflect the increase in marginal power generation cost when the power generation power increases; determining the optimization variables, namely, the amount of energy purchased from the wholesale market, the most economical power generation power of the aggregator (AGG) and DG under the premise of system safety; introducing the node voltage amplitude and phase angle, as well as the nodes adjacent to the node, to represent the power flow constraints of the distribution network line; by defining a matrix, the linear relationship between the node power injection amount and the node voltage amplitude and phase angle is expressed in matrix form; considering the uncertainty of renewable energy output and load, the node absorbs unbalanced power by relaxing the node, and the voltage uncertainty is expressed in linear form; using joint opportunity constraints to constrain the node voltage to ensure that the overall system risk meets the preset voltage limit probability.

[0018] In a preferred embodiment, in step S1, the optimization of the dispatch model specifically includes: determining the wholesale market electricity price and the amount of electricity purchased from the wholesale market; determining the active power and reactive power output constraints of distributed generators; determining the charging and discharging power and dispatching costs of the aggregator (AGG); determining the active and reactive loads of the distribution network nodes; determining the day-ahead predicted distributed photovoltaic and wind turbine outputs; and by defining a matrix, expressing the linear relationship between the power injection amount of all nodes except the reference node in the linearized distribution network flow model and the node voltage amplitude and phase angle in matrix form.

[0019] Step S2, piecewise linearization of joint chance constraints: use sample data to fit the distribution characteristics of uncertain variables, accurately fit the joint probability distribution of uncertain variables through a Gaussian mixture model, and use piecewise linearization technology to transform the nonlinear joint chance constraints into a linear constraint form that can be analytically solved.

[0020] In a preferred embodiment, in step S2, the process of fitting uncertainty variables using a Gaussian mixture model specifically includes: defining a normalized representation of constraint conditions to uniformly represent the constraints of node power uncertainty; equivalently decomposing the constraint conditions into a single chance constraint with risk probability as a variable; calculating the overall joint violation probability of all possible constraint violation event combinations; using a Gaussian mixture model to fit the joint probability distribution function of historical data of node power uncertainty; determining the weight, mean and variance of each sub-distribution in the Gaussian mixture model; obtaining the Gaussian mixture model parameters under the normalized constraint conditions through affine transformation; and converting the chance constraint into the form of a cumulative density function to facilitate linear processing.

[0021] In a preferred embodiment, in step S2, the process of linearizing the chance constraint using piecewise linearization technology specifically includes: piecewise linearizing the cumulative density function of the standard Gaussian function; determining the segment parameters of each segment and the total number of segments; converting the cumulative density function into a form suitable for a specific problem through a translation transformation; introducing auxiliary variables to convert the linear sum form of the cumulative density function of the Gaussian mixture model into a linear form; through the piecewise linearization technology, converting the nonlinear chance constraint into a linear constraint form that can be analytically solved, so as to facilitate processing in the optimization model.

[0022] Step S3, joint default probability adjustment based on posterior iteration: Based on historical data and the initial dispatch solution, the joint default probability threshold is dynamically corrected through the posterior iteration algorithm to balance the conservatism and economy of the model, and the distribution network optimization dispatch model is solved in combination with the linearization constraints and the adjusted default probability to generate a power distribution plan that takes into account both safety and economy, and reduce the risks of node voltage exceeding the limit and line blocking.

[0023] In a preferred embodiment, step S3 specifically includes: initializing the iterative process, setting the number of iterations, the default probability allowed by the system, the historical data set, the initial value of the joint default probability and the convergence parameter; substituting the initial value of the joint default probability into the optimization scheduling model to solve the current scheduling decision; using the historical data set to evaluate the default probability corresponding to the current scheduling decision to determine whether there is a default; estimating the joint violation probability and calculating the overall joint violation probability of all possible constraint violation event combinations; dynamically adjusting the joint default probability threshold according to the evaluation results to balance the conservatism and economy of the model; checking the iteration termination condition, if the maximum value of the default probability is less than the allowed default probability or the maximum value of the joint default probability is less than the convergence parameter, then terminate the iteration; otherwise, update the joint default probability and continue to iterate.

[0024] The proposed distribution network piecewise linearization joint opportunity constraint optimization scheduling method based on sample data achieves accurate fitting of the distribution characteristics of uncertain variables by using the Gaussian mixture model technology based on sample fitting; the piecewise linearization technology is used to transform the opportunity constraints into linear constraints that are easy to solve; and the conservatism in the model is greatly improved by the posterior-based joint probability density adjustment algorithm. The efficient solution of the joint opportunity constraint optimization scheduling model is achieved, and the safe and economical operation of the distribution network is realized.

[0025] The specific embodiments of the present invention and algorithm examples thereof are further described below.

[0026] A piecewise linear joint opportunity constrained optimization scheduling method for distribution network based on sample data. The process is shown in Figure 1 Among them, the Gaussian mixture model based on sample fitting is used to achieve accurate fitting of the distribution characteristics of uncertain variables; the opportunity constraints based on the Gaussian mixture model are transformed into solvable linear constraints using the piecewise linearization method; and the problem of conservatism existing in the existing methods based on solving Boolean inequalities is solved through the joint default probability adjustment method based on posterior iteration.

[0027] The specific steps include: (1) Establishing a distribution network opportunity-constrained optimization dispatch model The optimization goal of the optimal dispatching model of the distribution network is to minimize the operating cost of the distribution network while meeting the constraints of the safe operation of the distribution network. The operating cost of the distribution network includes the cost of purchasing electricity from the wholesale market and the power generation cost of distributed generators (DG). The cost function of DG is a quadratic function. As the power generation increases, its marginal power generation cost will continue to increase. The optimization variable of the distribution network is to decide the energy purchased from the wholesale market and the most economical power generation of the aggregator (AGG) and DG under the premise of ensuring system safety: (1) In the formula, represents the electricity price in the wholesale market; Represents the amount of electricity purchased from the wholesale market; represents the active power of distributed generators; and Indicates the charge and discharge power of AGG; It represents the cost of dispatching AGG; , and represents the generation cost coefficient of distributed generators; A collection of nodes representing a distribution network.

[0028] The active and reactive output constraints of DG are expressed as follows: (2) (3) In the formula, and Indicates the upper and lower limits of active power output; and Indicates the upper and lower limits of reactive power output.

[0029] The constraints on the AGG active power are expressed as follows: (4) In the formula, and Represents the feasible domain parameters of AGG.

[0030] The power flow constraints of the distribution lines are as follows:

[0031] In the formula, , ; represents the node voltage amplitude; represents the phase angle of the node voltage; Representatives and Nodes Adjacent nodes; and Represent the active and reactive loads of the node respectively; Equations (7) and (8) represent the active power balance constraints of nodes connected to AGG; and They represent the distributed photovoltaic and wind turbine outputs predicted a day ago, respectively.

[0032] By defining the matrix and , the linear relationship between the power injection amount of all nodes except the reference node in the above linearized distribution network flow model and the node voltage amplitude and phase angle can be expressed in matrix form:

[0033] In the formula, , , and They represent the injected active power, injected reactive power, node phase angle and node voltage amplitude except the relaxed nodes respectively; and yes and The matrix after removing the first row and the first column; and yes and The vector consisting of the first element in the first column of excluding ; and Represents the phase angle and voltage magnitude of the relaxed node.

[0034] Considering the uncertainty of renewable energy output and load predicted in the day ahead, it will lead to unbalanced node power injection. These unbalanced powers will be absorbed and balanced by slack nodes (nodes connected to the main grid). is reversible. According to equations (13) and (15), by By inverting, the voltage uncertainty associated with unbalanced power can be expressed in the following linear form:

[0035] In the formula, represents the vector of unbalanced power components of nodes except the relaxed nodes; Represents the uncertainty of the node voltage; A linear mapping matrix representing the unbalanced power to the node voltage change; Represents the inverse matrix.

[0036] Considering the uncertainty of node voltage caused by the uncertainty of node injection power, in order to ensure that the overall risk of the system meets the requirements, the node voltage is constrained by joint opportunity constraints:

[0037] In the formula, and Respectively represent the upper and lower limits of the node voltage; Represents the joint risk probability of system node voltage exceeding the limit.

[0038] (2) Joint Chance Constrained Linearization First, by defining and To unify and normalize the constraints in the formula:

[0039] According to the inclusion-exclusion principle, (19) can be equivalently decomposed into a single opportunity constraint with risk probability as the variable:

[0040] In the formula, Representation Constraints The probability of risk; represents the overall joint violation probability after summing up all possible constraint violation event combinations.

[0041] The joint probability distribution function of the historical data of node power uncertainty can be fitted by a Gaussian mixture model:

[0042] In the formula, It is the Gaussian mixture model m The weight of the sub-distribution; is the probability density function of the Gaussian distribution; represents the mean; Represents variance.

[0043] The parameters of the Gaussian mixture model can be obtained by using formula (16) Perform an affine transformation to get:

[0044] Formula (20) can be equivalently transformed into the form of cumulative density function:

[0045] According to the Gaussian mixture model, formula (21) can be expressed as the linear sum of the cumulative density functions of multiple Gaussian functions:

[0046] In the formula, is the cumulative density function of the Gaussian function. The cumulative density function of the standard Gaussian function can be linearized and approximated by piecewise linearization technology:

[0047] In the formula, and Represents the line segment parameters of the nth segment; Indicates the total number of segments. Can be Perform translation transformation to obtain:

[0048] By introducing auxiliary variables , formula (24) can be transformed into the following linear form:

[0049] (3) Joint default probability adjustment based on posterior iteration The following joint default probability adjustment algorithm based on posterior iteration is used to determine the formula (21) Values: 1) Iteration initialization: Initialize the number of iterations c Equal to 0; set the default probability allowed by the system ; Historical dataset ; Joint Default Probability Initial value of ; Convergence parameters . 2) Model solution: Set as To solve the current optimization scheduling model to get the current scheduling decision . 3) Scheduling decision evaluation: Use historical data sets to evaluate the default probability corresponding to the current decision Conduct an assessment:

[0050] In the formula, Is If the vector is composed of If there is any value greater than 0 then ,on the contrary . 4) Estimation of joint violation probability: The joint violation probability is estimated through the following public statement:

[0051] 5) Update the joint violation probability: If Greater than mean is overrated, so Set as To further constrain the risk, otherwise to update. 6) Iteration termination condition check: If The maximum value is less than or The maximum value is less than , terminate the iteration, otherwise, return to step 2 to continue the iteration.

[0052] An embodiment of the present invention further provides a storage medium for storing a computer program, which at least performs the above method when executed.

[0053] An embodiment of the present invention further provides a control device, comprising a processor and a storage medium for storing a computer program; wherein the processor is configured to execute at least the method described above when executing the computer program.

[0054] An embodiment of the present invention further provides a processor, wherein the processor executes a computer program and at least executes the method described above.

[0055] The storage medium can be implemented by any type of non-volatile storage device, or a combination thereof. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a ferromagnetic random access memory (Flash Memory), a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The storage medium described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0056] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0057] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0058] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0059] A person skilled in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, etc. Various media that can store program codes.

[0060] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0061] The methods disclosed in the several method embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0062] The features disclosed in several product embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new product embodiments.

[0063] The features disclosed in several method or device embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0064] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art of the present invention, several equivalent substitutions or obvious variations can be made without departing from the concept of the present invention, and the performance or use is the same, which should be regarded as belonging to the protection scope of the present invention.

Claims

1. A distribution network piecewise linearization joint opportunity constrained optimization scheduling method based on sample data, characterized in that: The following steps are involved: S1. Establish a joint opportunity constraint dispatch model for distribution network: Based on the uncertainty of node voltage in distribution network and the risk of unbalanced power injection, a joint opportunity constraint model is constructed, with the preset node voltage over-limit probability as the boundary condition to ensure that the system operation risk is controllable; S2. Piecewise linearization of joint chance constraints: The distribution characteristics of uncertain variables are fitted using sample data, the joint probability distribution of uncertain variables is accurately fitted through the Gaussian mixture model, and the piecewise linearization technique is used to transform the nonlinear joint chance constraints into linear constraints that can be solved analytically. S3. Joint default probability adjustment based on posterior iteration: Based on historical data and initial dispatch solutions, the joint default probability threshold is dynamically corrected through the posterior iteration algorithm to balance the conservatism and economy of the model. The distribution network optimization dispatch model is solved by combining the linearization constraints and the adjusted default probability to generate a power distribution plan that takes into account both safety and economy, and reduce the risks of node voltage exceeding the limit and line congestion.

2. The distribution network piecewise linearization combined opportunity constrained optimization scheduling method based on sample data according to claim 1 is characterized in that: Step S1 specifically includes: Construct an optimization model to minimize the distribution network operation cost, which includes the wholesale market purchase cost and distributed generator (DG) generation cost; The generation cost of DG is modeled as a quadratic function to reflect the increase in marginal generation cost when the generation power increases; Determine the optimization variables, i.e., the amount of energy purchased from the wholesale market, the most economical generation power of the aggregator (AGG) and DG under the premise of system security; The node voltage amplitude and phase angle, as well as the nodes adjacent to the node, are introduced to represent the power flow constraints of the distribution network. By defining a matrix, the linear relationship between the node power injection amount and the node voltage amplitude and phase angle is expressed in matrix form; Considering the uncertainty of renewable energy output and load, the unbalanced power is absorbed by relaxing nodes and the voltage uncertainty is expressed in a linear form; Joint opportunity constraints are used to constrain node voltages to ensure that the overall system risk meets the preset voltage limit probability.

3. The distribution network piecewise linearization combined opportunity constrained optimization scheduling method based on sample data according to claim 2 is characterized in that: In step S1, the optimization of the scheduling model specifically includes: Determine wholesale market electricity prices and the amount of electricity purchased from the wholesale market; Determine the active and reactive power output constraints of distributed generators; Determine the charging and discharging power and dispatching fees of the aggregator (AGG); Determine the active and reactive loads at distribution network nodes; Determine the day-ahead forecast of distributed photovoltaic and wind turbine output; By defining a matrix, the linear relationship between the power injection amount of all nodes except the reference node in the linearized distribution network power flow model and the node voltage amplitude and phase angle is expressed in matrix form.

4. The distribution network piecewise linearization combined opportunity constrained optimization scheduling method based on sample data according to claim 1 is characterized in that: In step S2, the process of fitting uncertain variables using the Gaussian mixture model specifically includes: Define a normalized representation of constraints to uniformly represent the constraints of node power uncertainty; Decompose the constraints into single chance constraints with risk probability as variable; Calculate the overall joint violation probability for all possible constraint violation event combinations; Use Gaussian mixture model to fit the joint probability distribution function of historical data of node power uncertainty; Determine the weight, mean, and variance of each subdistribution in the Gaussian mixture model; Obtain the parameters of the Gaussian mixture model under normalized constraints through affine transformation; The chance constraint is transformed into the form of cumulative density function to facilitate linearization.

5. The distribution network piecewise linearization combined opportunity constrained optimization scheduling method based on sample data according to claim 1 is characterized in that: In step S2, the process of linearizing the chance constraint using the piecewise linearization technique specifically includes: The cumulative density function of the standard Gaussian function is piecewise linearized and approximated; Determine the line segment parameters of each segment and the total number of segments; The cumulative density function is converted into a form suitable for a specific problem through a translation transformation; Auxiliary variables are introduced to transform the linear sum form of the cumulative density function of the Gaussian mixture model into a linear form; Through piecewise linearization technology, nonlinear chance constraints are transformed into linear constraints that can be solved analytically, so as to facilitate processing in the optimization model.

6. The distribution network piecewise linearization combined opportunity constrained optimization scheduling method based on sample data according to claim 1 is characterized in that: Step S3 specifically includes: Initialize the iteration process, set the number of iterations, the default probability allowed by the system, the historical data set, the initial value of the joint default probability and the convergence parameters; By substituting the initial value of the joint default probability into the optimal scheduling model, the current scheduling decision is obtained; Use historical data sets to evaluate the default probability corresponding to the current scheduling decision to determine whether there is a default situation; Estimate the joint violation probability, and calculate the overall joint violation probability for all possible constraint violation event combinations; Dynamically adjust the joint default probability threshold based on the evaluation results to balance the conservatism and economy of the model; Check the iteration termination condition. If the maximum value of the default probability is less than the allowed default probability or the maximum value of the joint default probability is less than the convergence parameter, terminate the iteration; otherwise, update the joint default probability and continue the iterative solution.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the distribution network piecewise linearization joint opportunity constrained optimization scheduling method based on sample data as described in any one of claims 1 to 6 is implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the distribution network piecewise linearization joint opportunity constrained optimization scheduling method based on sample data as described in any one of claims 1 to 6 is implemented.

Images