Search

CN-116862068-B - Transformer substation planning distribution robust optimization method and system considering excitation type response uncertainty

CN116862068BCN 116862068 BCN116862068 BCN 116862068BCN-116862068-B

Abstract

The invention provides a robust optimization method and a robust optimization system for transformer substation planning distribution, which are used for considering excitation type response uncertainty. Firstly, comprehensively considering subscription cost, response cost and punishment income of excitation type demand response, and establishing a substation site selection and volume fixation mixed integer linear programming model. And secondly, constructing a response power uncertainty fuzzy set based on the 1-norm and the ++norm, and establishing a transformer substation planning distribution robust optimization model considering excitation type response uncertainty on the basis of improving the mixed integer linear programming model. And finally, decomposing the model into a main problem and a sub problem, and providing a distributed robust optimization model iterative solving method generated based on the columns and the constraints. The method provided by the invention uses the distributed robust optimization model, can fully consider the randomness of the user demand response, calculates and matches the load characteristics in the model, effectively reduces the peak value of the load curve of the transformer substation, and ensures the economy and the robustness of the planning scheme.

Inventors

  • XU ZHENGYANG
  • FAN QINGFEI
  • LI JUNJIE
  • GAO KUNYANG

Assignees

  • 天津大学
  • 天津天电清源科技有限公司

Dates

Publication Date
20260508
Application Date
20230714

Claims (7)

  1. 1. A robust optimization method for transformer substation planning distribution considering excitation type response uncertainty is characterized by comprising the following specific steps: step1, comprehensively considering signing cost, response cost and punishment income of demand response, and establishing a substation site selection and volume fixation mixed integer linear programming model considering excitation type demand response; Step2, constructing a response power uncertain fuzzy set based on 1-norm and +_norm, and on the basis of improving a mixed integer linear programming model, constructing a two-stage three-layer distribution robust optimization model based on a multi-discrete scene, wherein the construction Step of the robust optimization model comprises the following steps: 1) Building an uncertainty fuzzy set of demand responses Considering uncertainty of user will, the actual response result P t has deviation from the power grid demand response instruction, and P t fluctuates in a certain range, and modeling is carried out on a fuzzy set of uncertainty: Firstly, obtaining a plurality of actual scenes through historical data, screening N k finite discrete scenes and probability distribution p k,0 under each scene through a scene clustering means, and constructing a confidence set based on 1-norm and ++norm to limit fluctuation change of the probability distribution by considering that the scenes cannot represent the actual probability distribution: wherein, ψ 1 and ψ ∞ represent confidence intervals of 1-norm and + -norm limitation respectively, P is the vector form of scene probability P k , P 0 is the vector form of initial probability P k,0 of each scene; Is a vector consisting of N k positive real numbers corresponding to P; K is the number of sample scenes, alpha 1 and alpha ∞ are the confidence levels of the establishment of the psi 1 and the psi ∞ respectively, so that the probability distribution confidence level set is limited by the 1-norm and the + -norm simultaneously, the situation of extreme is avoided, and the psi=ψ 1 ∩Ψ ∞ is as follows: , 2) Two-stage distributed robust optimization model considering response power uncertainty The uncertainty of the considered demand response is that the user cannot fully meet the response requirement when receiving the response command, and the uncertainty exists in the response power, so that the default power is generated, and therefore, the calculation formula of the demand response cost C DR2 in the operation stage is rewritten as follows: meanwhile, the actual response power can be changed due to the uncertainty of the demand response, so that load curve fluctuation in the power supply range of the transformer substation is caused, and the N-1 safety constraint of the transformer substation is influenced, and the constraint formula is rewritten as follows: The uncertainty of the user demand response in the operation stage can influence the planning stage, so that the planning model can be decomposed into two stages after the uncertainty is taken into account, wherein the first stage is the planning stage, the decision variables comprise the relation between the position and the capacity of the transformer substation, the connection relation between the load and the transformer substation and the demand response subscription capacity, the second stage is the operation stage, the decision variables are the response power of the user, the uncertainty exists in the actual response power, and the first stage comprises x is 、y ij and x is 、y ij Represented by vector x, second stage decision variables Represented by vector d, the distributed robust model based on discrete scenes can be expressed as follows: Wherein: a linear coefficient matrix for the first stage objective function; n k represents the total number of discrete scenes of the probability distribution, k is the number of each scene, p k represents the probability of the scene at k, and the constraint condition form transformation is as follows: Wherein C, E, F, G, H, m, n, u, v represent the matrix or vector form corresponding to the variables above, and the first two formulas correspond to the equality constraint and the inequality constraint of the first stage variables; the third inequality constraint is a capacity coupling inequality of the first stage variable and the second stage variable, and the last inequality constraint corresponds to the demand response inequality constraint of the second stage; Step3, aiming at the distributed robust optimization model, a main problem and sub-problem iterative algorithm generated based on columns and constraints is provided.
  2. 2. The method for robust optimization of transformer substation planning distribution taking account of uncertainty of excitation type response of claim 1, wherein the building of the mixed integer linear programming model in Step1 comprises 1) a demand response cost model of a power grid company The power supply and use parties firstly sign contracts, define the maximum power which can be responded by users in a good signing period, sign signing capacity P 1 and generate signing cost, send response instructions to the users in advance when the power consumption is high in peak during the operation period of the power distribution network, the users respond to the power P t according to the instructions and accumulate to generate response cost, the part of the response power of the users which is less than the instruction requirement is illegal power P t,f and accumulate to generate punishment benefits, and the calculation formula of the demand response cost paid to the single user by the power grid company every year is as follows: the system comprises a power grid company, a user, a price management system and a price management system, wherein C DR is demand response cost paid to the user by the power grid company every year, sigma 1 、σ 2 、σ 3 is signing cost, response cost and unit price of punishment income respectively, P is the maximum load of the user, and lambda is the ratio of the maximum signable capacity of the user to the maximum load of the user and represents the capacity of the load to participate in demand response; 2) Transformer substation planning model considering demand response The transformer substation planning aims to reduce the investment construction cost of the transformer substation and the main line as much as possible on the premise of meeting the load electricity demand of a target year and various planning constraints, and meanwhile, various costs related to demand response are considered; a) Decision variables, wherein the decision variables comprise a transformer station position and capacity selection, a load and transformer station connection relation, a demand response signing capacity and response power of each period, x is is a Boolean variable and represents whether an ith transformer station position selects an ith transformer station type, each type of transformer station to be selected corresponds to different transformer station capacities, a capacity 0 option is added in the types of transformer stations to be selected, if a certain transformer station position selects a 0 capacity type, the position is not selected for constructing a transformer station, the position selection and the capacity selection are unified, and the nonlinear problem caused by multiplication of the two variables in the calculation of the construction cost of the transformer station is solved; The continuous variable represents the demand response subscription capacity of the jth load; as a continuous variable, representing the response power of the j-th load in the t period; b) Objective function: minC=C S +C L +C DR1 +C DR2 The method comprises the steps of taking C as the total cost, taking C S 、C L 、C DR1 and C DR2 as construction year cost, line construction year cost, demand response signing cost and response cost of a transformer respectively, taking r 0 as the discount rate, taking ms as depreciation year of the transformer, taking N P as the number of positions to be selected of the transformer substation, taking N S as the number of types of the transformer substation to be selected, taking C Ss as construction cost of the s-th type of the transformer substation to be selected, taking beta as a line unit cost coefficient, taking ml as depreciation year of the line, taking N L as the number of load points, taking d ij as distance from a transformer substation i to a load j, taking P j as the maximum load quantity of the j-th load point, and taking sigma j,1 and sigma j,2 as unit prices of demand response signing cost and response cost of the j-th load respectively; c) Constraint conditions: The capacity of the transformer substation is restricted to be unique, and one transformer substation construction position can only select one transformer substation type: the load points are attributed to unique constraint, and when the power supply range is divided, one upper-level transformer station corresponding to one load point is provided with only one upper-level transformer station: The maximum power supply radius constraint, r max , is the maximum value of the medium voltage line power supply radius specified in the power supply and distribution design specification: Based on the power grid safe operation principle, after any transformer in the transformer substation fails, the residual transformer needs to meet the requirement that all loads in the power supply range operate for 2 hours, and the maximum load rate e s of the transformer substation in normal operation has the following inequality constraint: Wherein J i is a load set in the power supply range of the ith transformer substation, P j,t is the power of the jth load in the t period, cos phi j is the power factor of the jth load, and S s is the capacity of the type of the ith transformer substation to be selected; and (3) demand response constraint, wherein the demand response subscription capacity of each load point is not more than the maximum response capacity of each load point, and the user response capacity is not more than the subscription capacity in running:
  3. 3. The robust optimization method of substation planning distribution taking account of excitation type response uncertainty as set forth in claim 1, wherein the iterative algorithm in Step3 includes: Based on a constraint generation algorithm, decomposing a model into a main problem MP and a sub problem SP, and then solving an optimal solution through iteration, wherein the purpose of MP solving is to obtain an optimal planning scheme meeting the constraint of a known probability distribution under the condition of a limited discrete scene, and the objective function and constraint condition of MP are described as follows: Wherein L is the lower-layer demand response running cost, the upper corner mark r represents the r-th iteration, the probability distribution of each iteration except the 1 st iteration is obtained by solving an SP, and the MP problem is solved to obtain a global optimal solution And corresponding planning decision variables ; SP solution is based on MP optimization results Under the condition that the capacity, the power supply range and the demand response subscription capacity of the transformer substation are known, the load time sequence characteristic and the demand response characteristic are matched, the worst probability distribution P k of the response power is found, and then the distribution is provided for MP so as to perform the next iterative calculation, and meanwhile, the load time sequence characteristic and the demand response characteristic are calculated according to the obtained load time sequence characteristic Updating the globally optimal solution, the objective function of the SP may be described as follows: As can be seen from the above equation, the min problem in each scene is independent, and the parallel method is used for simultaneous calculation, for example, the internal optimization result of the kth scene is that The objective function of the SP may be converted to: Solving the MP and the SP problems by using an MILP model and a linear programming model respectively, and transmitting an SP optimization result P k to the MP for iterative calculation until the global optimal solution of two adjacent iterations is obtained And stopping iteration when the difference value is smaller than a specified threshold value, and obtaining the optimal planning cost and the decision variable value.
  4. 4. A substation planning distribution robust optimization system accounting for excitation response uncertainty, for performing the method of claim 1, the system comprising: (1) The data input and processing module is used for carrying out matrixing processing on input load prediction data and various planning parameter information; (2) The main problem solving module obtains an optimal planning scheme and planning layer decision variables meeting the known probability distribution constraint under the condition of a limited discrete scene; (3) And the sub-problem solving module is used for fixing the decision variables of the planning layer obtained by the main problem solving module, solving the decision variables of the operation layer by taking the minimum operation cost as a target, and finding out the worst probability distribution of the response power.
  5. 5. A substation planning distribution robust optimization system taking account of excitation response uncertainty as in claim 4, further comprising the following modules: The initialization module is used for initializing and setting iteration solving parameters, and solving corresponding robustness scene probability through a known scene probability intervention algorithm so as to form iteration; The judging module is used for judging whether the planning result is converged or not and stopping iterative solution; And the output module is used for outputting the planning scheme and the decision variables.
  6. 6. A non-transitory computer readable storage medium storing computer instructions that cause the computer to perform the method of any one of claims 1 to 3.
  7. 7. A computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the method of any of claims 1 to 3.

Description

Transformer substation planning distribution robust optimization method and system considering excitation type response uncertainty Technical Field The invention belongs to the field of power distribution network planning, and relates to a robust optimization method and a robust optimization system for substation planning with uncertainty of excitation response. Background In recent years, energy transformation strategy in China is rapidly advanced, the electrification degree of terminal energy is continuously improved, peak load of a power distribution network is increased year by year, and huge pressure is brought to planning investment of a transformer substation. Demand response is an effective means to solve this problem. However, although the demand response can reduce the load peak to some extent, there is some uncertainty in the response due to the influence of human decision-making factors. How to finely consider the demand response and the uncertainty thereof in the transformer substation planning is an important problem to be solved currently. The transformer substation planning relates to site selection, volume fixing and power supply range division of a transformer substation, is a large-scale nonlinear optimization problem comprising multiple types of decision variables, and an early planning method mainly comprises two types of heuristic methods and a layered decoupling method. The heuristic method algorithm can obtain an optimal solution or an approximate optimal solution when solving a large-scale problem, is easy to sink into local optimum, and the power supply range division is mostly distributed nearby, so that the problems of unreasonable planning, too low or too high load rate and the like are caused, and the hierarchical decoupling method is essentially to decouple the large-scale nonlinear problem into an upper layer of sub-problem and a lower layer of sub-problem, and realizes site selection and power supply range division aiming at each to-be-selected capacity combination scheme generated by an upper layer, wherein the capacity combination scheme generated by the method has the possibility of being incapable of enumerating completely. Many studies have considered demand responses in substation planning, which can improve the comprehensive load characteristics of the substation and reduce the investment cost of substation planning, but all have not considered the uncertainty of the demand responses, so that the planning scheme may have insufficient planning investment of the target year. Some researches are based on a random optimization method to model uncertainty of the position and the size of target annual load prediction, uncertainty of photovoltaic output and the like in consideration of uncertainty factors in transformer substation planning, but the method needs to know uncertainty factor probability distribution, is difficult to acquire in practical application, and therefore causes insufficient target annual planning investment, and other researches adopt a robust optimization method to process the uncertainty, namely consider the uncertainty factor probability distribution in the worst scene, and the obtained planning scheme is more conservative. In recent years, the advantages of the distributed robust optimization method in terms of uncertainty treatment are gradually paid attention to by students, the characteristics of random optimization and robust optimization are combined, and the optimization result shows good performance in terms of economy and conservation. Disclosure of Invention The technical problem to be solved by the invention is how to establish a substation planning mathematical model considering the demand response and the uncertainty thereof, solve the problem that the traditional substation planning method is easy to fall into local optimum, and ensure the economy and the robustness of the obtained planning scheme. The present invention solves the above problems by the following means. A robust optimization method for transformer substation planning distribution considering excitation type response uncertainty is characterized by comprising the following specific steps: step1, comprehensively considering signing cost, response cost and punishment income of demand response, and establishing a substation site selection and volume fixation mixed integer linear programming model considering excitation type demand response; Step2, constructing a response power uncertain fuzzy set based on 1-norm and +_norm, and constructing a two-stage three-layer distribution robust optimization model based on a multi-discrete scene on the basis of improving a mixed integer linear programming model; Step3, aiming at the distributed robust optimization model, a main problem and sub-problem iterative algorithm generated based on columns and constraints is provided. Furthermore, the hybrid integer linear programming model set forth in Step1 is built by 1) a grid company demand