CN-122026454-A - Wind-solar storage and RPC capacity optimization configuration method of traction power supply system

CN122026454ACN 122026454 ACN122026454 ACN 122026454ACN-122026454-A

Abstract

The invention relates to the technical field of traction power supply system optimization, in particular to a wind-solar storage and RPC capacity optimization configuration method of a traction power supply system, which comprises the following steps of establishing a multi-source power balance model comprising a hybrid energy storage system, a power interaction system with a power grid, a railway power Regulator (RPC) and traction load; the method comprises the steps of constructing a multi-objective optimization model under constraint conditions by taking full life cycle cost minimization and new energy utilization maximization as optimization targets, carrying out collaborative optimization on capacity configuration variables and operation scheduling variables by adopting a mixed optimization algorithm based on combination of a black-wing iris optimization algorithm and NSGA-II, and outputting an optimal equipment capacity configuration scheme and a typical daily operation scheduling strategy. The invention provides a multi-objective optimization method taking new energy consumption and system economy into consideration aiming at the problems of wind-solar storage and RPC capacity optimization configuration in a traction power supply system. By introducing a BKA-NSGA-II hybrid optimization algorithm, the collaborative optimization of system capacity and operation scheduling is realized.

Inventors

JIANG KE
XU YIYUE
QIAN KANG
GENG LU
ZHANG ZHAO
WANG ZHANGFAN
PENG YUFEI
ZHOU JIANGSHAN
Feng Zhefei

Assignees

中国能源建设集团江苏省电力设计院有限公司

Dates

Publication Date: 20260512
Application Date: 20251215

Claims (11)

1. The wind-solar energy storage and RPC capacity optimization configuration method of the traction power supply system is characterized by comprising the following steps: Step one, establishing a multisource power balance model comprising a hybrid energy storage system, a power interaction system with a power grid, a railway power Regulator (RPC) and traction loads; Step two, constructing a multi-objective optimization model under constraint conditions by taking the minimization of the total life cycle cost and the maximization of the new energy utilization rate as optimization targets; step three, adopting a mixed optimization algorithm based on the combination of a black-wing iris optimization algorithm and a non-dominant ordering genetic algorithm II to carry out collaborative optimization on capacity configuration variables and operation scheduling variables; and step four, outputting an optimal equipment capacity configuration scheme and a typical daily operation scheduling strategy.
2. The method for optimizing configuration of wind-solar energy storage and RPC capacity of a traction power supply system according to claim 1, wherein in step one, in said multi-source power balance model, in a scheduling period with 5 minutes as a scheduling time interval The system should satisfy the following power balance constraints: ; (1) Wherein, the 、 Respectively the moments of time The output power of the photovoltaic power and the wind power, 、 The discharging power and the charging power of the storage battery are respectively, 、 The discharging power and the charging power of the super capacitor are respectively, For the power of the RPC, For the exchange of power between the system and the grid, For traction load power, the power balance equation ensures that the total energy supply of the system is equal to the total load at each moment, and dynamic power matching is realized.
3. The method for optimizing configuration of wind-solar energy storage and RPC capacity of a traction power supply system according to claim 2, wherein in step one, said hybrid energy storage system comprises a battery energy storage system and a super capacitor system; In a battery energy storage system, the state of a battery is determined by a state parameter SOC thereof, and an SOC updating formula is as follows: ; (2) Wherein: 、 In order to achieve the charge and discharge efficiencies respectively, For the capacity of the battery, The scheduling period is 5 minutes; The battery SOC should satisfy the following boundary conditions: ; (3) The battery charge and discharge power is limited by rated power, loss and service life reduction are prevented, and the mutual exclusion conditions of charge and discharge are set: ; (4) ; (5) in the super capacitor system, the super capacitor has the characteristics of quick response and high power density, and the modeling mode is similar to that of a storage battery: ; (6) and satisfies the following constraints: ; (7) ; (8) (9)。
4. The method for optimizing wind-solar energy storage and RPC capacity of traction power supply system as defined in claim 3, wherein in step one, said power interaction system with power grid can provide active support or absorb electric energy for railway power regulator in a certain range, the control strategy of RPC can be determined by scheduling system for power compensation, power quality improvement or energy storage assistance, and output power thereof The method meets the following conditions: ; (10) The power grid is used as a final power supply and regulation end, and the power interaction quantity is defined as that in general practice, traction loads are mainly purchased, and surplus electricity generated by braking feedback is fed into the power grid to be fed into a positive feedback meter: (11)。
5. The method for optimizing configuration of wind-solar storage and RPC capacity of a traction power supply system according to claim 4, wherein in the second step, the total life cycle cost includes annual cost such as equipment investment, system running cost, energy abandoning penalty cost, RPC use cost and energy storage system depreciation cost; the annual cost of equipment investment and the like is as follows: ; (12) ; (13) Wherein, the In order to be a capital recovery factor, In order to achieve the discount rate, For the life-time of the device, Investment cost per unit capacity; The system operation cost comprises power grid electricity purchasing cost, energy discarding punishment cost, RPC use cost and energy storage system depreciation cost, and the formula is as follows: ;(14) the electricity purchasing cost of the power grid is as follows: ;(15) Wherein, the For the moment of time Electricity purchasing unit of (2) is yuan/kWh; the energy discarding punishment cost is as follows: ; (16) Wherein, the The cost is punished for unit energy rejection, The energy is new energy power which is actually used for load or energy storage; The RPC is used at the following cost: (17) Wherein, the Unit cost (yuan/kWh) for RPC; The depreciation cost of the energy storage system is as follows: ;(18) Wherein, the , The invention converts annual cost into depreciation and spreads the depreciation to a scheduling period for the depreciation cost of an energy storage unit; The full Life Cycle Cost (LCC) is calculated by summing the annual capital investment and operating costs to get the minimum system first objective function: (19)。
6. the optimal configuration method for wind-solar storage and RPC capacity of a traction power supply system according to claim 5, wherein in the second step, the new energy utilization rate is defined as an energy ratio effectively used in wind-solar power generation, namely: ;(20) Wherein, the , And The photovoltaic and wind power are actually used, so that the maximum energy utilization efficiency objective function is as follows: ;(21)。
7. the method for optimizing configuration of wind-solar energy storage and RPC capacity of a traction power supply system according to claim 6, wherein in the second step, the constraint conditions include a power flow constraint, a hybrid energy storage state of charge constraint, an upper limit constraint, a lower limit constraint and an RPC constraint; the tide constraint is mainly used for guaranteeing energy flow balance of the system, and is shown in a formula (1); The hybrid energy storage state of charge constraint is: ;(22) The upper limit and the lower limit constraint are respectively represented by wind power generation, photovoltaic power generation, storage batteries, super capacitors and RPC upper limit and lower limit: ;(23) ;(24) ;(25) ;(26) ;(27) ;(28) ;(29) The RPC constraint is: (30)。
8. The wind-solar energy storage and RPC capacity optimization configuration method of a traction power supply system according to claim 7, wherein in the third step, the optimization algorithm of the iris nigricans is an emerging natural element heuristic optimization algorithm based on predation and migration behavior heuristics of the iris nigricans, and specifically comprises the following steps: Step 1, initializing population, wherein initial positions are given to iris nigromaculata individuals in the initial stage to form search space distribution, and the problem is assumed to be Dimension optimization problem, number of potential solutions is Initial population matrix The definition is as follows: ;(31) The individual positions are randomly generated as: ;(32) Wherein, the Is 1 and An integer of the number of the two, And Respectively the first Dimension of the first The upper limit and the lower limit of the iris nigricans, Is at In the initialization process, the BKA selects the individual with the optimal fitness value as the leader in the initial population and selects the global optimal initial leader by the following strategy : ; (33) In the middle of Representing the initial population of the plant, In order to find the function, In order to be an optimal fitness value, In order to adapt the function of the degree of adaptation, Is the first in the population An individual.
9. Step 2, an attack behavior, wherein the attack behavior of the iris nigricans is mainly represented as a hover observation-fast dive strategy in an attack stage, has stronger global searching and searching capability, and is a mathematical model of the attack behavior of the iris nigricans as follows: ; (34) ;(35) In the formula, And Respectively represent the first Step and No In step iteration Iris nigromaculata at the first A position of dimension, Is a random number between 0 and 1, Is a constant with a value of 0.9, Is the total number of iterations and, Is the number of iterations completed so far, Is a balance parameter; Step 3, migration behaviors, namely simulating dynamic update of a leader according to population fitness under natural conditions, wherein the leader has stronger local development capability, the strategy is characterized in that the dynamic evolution of winners and aptamers is realized, if the fitness value of the current population is smaller than that of random population, the leader gives up the leader and adds the lead into the migration population, and conversely, the leader leads the population to reach a destination, and the following is a mathematical model of the migration behaviors of the iris nigricans: ;(36) ;(37) In the formula, Is up to In the iteration, the iris nigromaculata is in A leader in dimension; And Respectively the first In a second time, the first time, In the second iteration Iris japonica A position on the dimension; Is an arbitrary Iris nigromaculata at the first place The current obtained in the second iteration is at The position of the dimension; Is an arbitrary Iris nigromaculata at the first place The first obtained in the iteration Fitness value of dimensional random position: Is a cauchy mutation; (38) Wherein, the above formula The result is output for the formula for scalar function values. Is the first Weight vector of sub-problem, expressed in given weight Lower solution Smaller values indicate that the solution is closer to the optimal equilibrium point under the weight definition, Is the first The number of functions of the object is the number of functions, Representing the first for the ideal point vector The optimum value that can be achieved in theory for the individual objective function.
10. The wind-solar storage and RPC capacity optimization configuration method of a traction power supply system according to claim 8, wherein in the third step, the non-dominant ranking genetic algorithm II is a classical and efficient multi-objective optimization algorithm and is improved on the basis of NSGA, the problems of high computational complexity, poor convergence and insufficient population diversity of the traditional genetic algorithm in multi-objective optimization are effectively solved by introducing non-dominant ranking, crowding distance calculation and elite retention strategies, the core idea of NSGA-II is to divide the population into different non-dominant grades, and preferably select individuals in the better grades for genetic operation, and meanwhile, the population diversity is ensured by measuring the density of the individuals through crowding distances, and the method specifically comprises the following steps: step 1, non-dominant ranking, for multi-objective optimization problem, one solution Another solution is governed If and only if all objective functions, solution Not inferior to solution And having at least one objective function, solution Strictly better than solution The method comprises the following steps: ;(39) NSGA-II divides populations into different non-dominant classes by fast non-dominant ranking For an optimal pareto front, including all non-dominant individuals; Step 2, calculating the crowding distance, namely, in order to maintain the diversity of the population, NSGA-II introduces the concept of the crowding distance, wherein for individuals in the same non-dominant class, the larger the crowding distance is, the more sparse other individuals around the individual are, the higher the reserved value is, and for individuals in one non-dominant class First, the individuals in the rank are sorted for each objective function value, and then, for each objective function Individual(s) The crowding distance of the boundary individual is generally set to infinity and the crowding distance is set to be the difference between the front and rear adjacent individuals on the objective function The calculation method of (2) is as follows: ;(40) Wherein, the And Is according to the object Ordered individuals Is used to determine the number of individuals in the group, And Is the object of Maximum and minimum values on pareto front; Step3, elite retention strategy, namely NSGA-II adopts elite retention strategy to make parent population And offspring populations Is combined to form Then select according to the non-dominant grade and crowding distance Individual best individual enters next generation The strategy ensures the convergence of the algorithm and the capability of searching the pareto optimal front edge, and prevents the loss of excellent individuals; The algorithm flow of NSGA-II is as follows: 1) Initializing, randomly generating an initial population The population size is ; 2) Non-dominant ranking is calculated from the crowding distance Non-dominant ranking is performed, and the crowding distance of each individual is calculated; 3) Genetic manipulation, by selection, crossover and mutation operations, to generate a population of offspring The specific operation is as follows: ① Selecting by analog binary crossing of selected parent individuals , Performing crossover operation to generate two sub-individuals , For the following Dimensional decision variables: ; ; (41) Wherein, the Is a probability of crossing Distribution index A determined random parameter; ② Variation operations, common polynomial variation, on individuals Performing mutation generation : ;(42) Wherein, the , Is the first The upper and lower bounds of the dimensional decision variables, Is a probability of variation Distribution index A determined disturbance term; 4) Merging and selecting parent population And offspring populations Combining to form a larger intermediate population ; 5) New population generation Non-dominant ranking is performed, and the crowding distance is calculated from the best non-dominant ranking The selection of individuals is started until a new population The size reaches If a certain non-dominant level If the individuals of (a) cannot be fully placed into the new population, then from the crowded distance Is selected for individuals with greater crowding distances until Reach to A subject; 6) And (5) iterating, namely repeating the steps 2-5 until the maximum iteration times are met.
11. The wind-solar storage and RPC capacity optimization configuration method of a traction power supply system according to claim 9, wherein in the third step, in order to fully exert the strong global searching capability of BKA and the advantages of NSGA-II in processing multi-objective problems, a BKA-NSGA-II mixing algorithm is provided, the mixing algorithm aims to accelerate searching of the optimal front edge of pareto through the exploration mechanism of BKA, maintain population diversity and obtain a high-quality pareto solution set by utilizing non-dominant sorting and crowding distance of NSGA-II, and combine the global exploration capability of BKA with the multi-objective processing framework of NSGA-II, so that the respective defects are hopefully made up, and the pareto optimal solution set which is widely and uniformly distributed is obtained while the optimizing efficiency is ensured; the BKA-NSGA-II mixing algorithm has the following flow: 1) Initializing, namely randomly generating initial iris nigromaculata population Population size of Each individual body Representing a wind-solar energy storage and RPC capacity configuration scheme, namely: ;(43) At the same time, for each capacity allocation scheme There is also a need to optimize the corresponding run schedule vector : ;(44) Combining capacity configuration and operation scheduling to form a complete individual: Setting algorithm parameters; 2) Fitness evaluation and non-dominant ranking for initial population Is a single unit of (a) or (b) Calculating multi-objective fitness value, namely economical goal and new energy consumption goal, according to capacity configuration and operation schedule Non-dominant ordering of NSGA-II, dividing into different pareto fronts And calculates the crowding distance of each individual ; 3) The algorithm of the iteration reaching the maximum iteration sub-times is terminated; 4) Output final output pareto optimal solution set, i.e. last population is at first non-dominant rank And the individuals represent wind-solar storage and RPC capacity configuration schemes of the traction power supply system under the double aims of new energy consumption and economy.

Description

Wind-solar storage and RPC capacity optimization configuration method of traction power supply system Technical Field The invention relates to the technical field of traction power supply system optimization, in particular to a wind-solar storage and RPC capacity optimization configuration method of a traction power supply system. Background The existing research is concentrated on the scheduling control of a single energy storage type or the capacity configuration of a new energy system, a capacity planning method for comprehensively considering the cooperation of hybrid energy storage and RPC in a traction power supply scene is lacking, the optimization target is mostly limited to economic or technical single-dimensional indexes, and the balance between the initial investment and long-term running cost of the system and the utilization efficiency of the new energy is difficult to be considered. Disclosure of Invention The invention aims to provide a wind-solar storage and RPC capacity optimization configuration method of a traction power supply system, which aims to solve the problems of redundancy in system configuration, low new energy consumption rate and high total life cycle cost in the prior art. In order to achieve the above object, the technical scheme provided is as follows: The wind-solar energy storage and RPC capacity optimization configuration method of the traction power supply system is characterized by comprising the following steps: Step one, establishing a multisource power balance model comprising a hybrid energy storage system, a power interaction system with a power grid, a railway power Regulator (RPC) and traction loads; Step two, constructing a multi-objective optimization model under constraint conditions by taking the minimization of the total life cycle cost and the maximization of the new energy utilization rate as optimization targets; Step three, adopting a mixed optimization algorithm based on the combination of a black wing iris optimization algorithm and NSGA-II to carry out collaborative optimization on capacity configuration variables and operation scheduling variables; and step four, outputting an optimal equipment capacity configuration scheme and a typical daily operation scheduling strategy. Preferably, in the step one, in the multi-source power balance model, in a scheduling period with 5 minutes as a scheduling time intervalThe system should satisfy the following power balance constraints: ; (1) Wherein, the 、Respectively the moments of timeThe output power of the photovoltaic power and the wind power,、The discharging and charging co-rates of the storage battery are respectively,、Respectively discharge and charge co-rate of the super capacitor,For the power of the RPC,For the exchange of power between the system and the grid,For traction load power, the power balance equation ensures that the total energy supply of the system is equal to the total load at each moment, and dynamic power matching is realized. Preferably, in step one, the hybrid energy storage system includes a battery energy storage system and a supercapacitor system; In a battery energy storage system, the state of the battery is determined by its state parameter SOC (State of Charge), and the SOC update formula is as follows: ; (2) Wherein: 、 In order to achieve the charge and discharge efficiencies respectively, For the capacity of the battery,The scheduling period is 5 minutes; The battery SOC should satisfy the following boundary conditions: ; (3) The battery charge and discharge power is limited by rated power, loss and service life reduction are prevented, and the mutual exclusion conditions of charge and discharge are set: ; (4) ; (5) in the super capacitor system, the super capacitor has the characteristics of quick response and high power density, and the modeling mode is similar to that of a storage battery: ; (6) and satisfies the following constraints: ; (7) ; (8) (9) 。 Preferably, in the first step, the power interaction system with the power grid provides active support or absorbs electric energy for a railway power Regulator (RPC) in a certain range, and the control strategy of the RPC can be determined by a dispatching system for power compensation, power quality improvement or energy storage assistance, and the output power of the RPC The method meets the following conditions: ; (10) The power grid is used as a final power supply and regulation end, and the power interaction quantity is defined as that in general practice, traction loads are mainly purchased, and surplus electricity generated by braking feedback is fed into the power grid to be fed into a positive feedback meter: (11)。 preferably, in the second step, the full life cycle cost includes annual cost such as equipment investment, system running cost, energy discarding penalty cost, RPC use cost and energy storage system depreciation cost; the annual cost of equipment investment and the like is as follows: ; (12) ; (13)