CN-115221784-B - Photovoltaic array reconstruction method and system based on double-objective optimization and computer equipment

CN115221784BCN 115221784 BCN115221784 BCN 115221784BCN-115221784-B

Abstract

The invention discloses a photovoltaic array reconstruction method, a system and computer equipment based on double-target optimization, wherein the method comprises the steps of establishing a photovoltaic array reconstruction model, step 110, and step 130, selecting a compromise solution from the pareto front by using a TOPSIS decision method, wherein the photovoltaic array reconstruction model comprises an optimization function taking output power maximization as a target, an optimization function taking switching times minimization as a target and a switching constraint condition, step 120, performing iterative optimization on the photovoltaic array reconstruction model by using a SPEA2 algorithm to obtain the pareto front, and further obtaining an optimal arrangement and combination scheme of the photovoltaic array. In the process of executing the photovoltaic array reconstruction by combining the SPEA2 algorithm and the TOPSIS decision method, the invention considers the problems of maximizing the output power and minimizing the switching times, can reduce the switching loss and simultaneously reduce the economic cost.

Inventors

LI WENJI
MENG DIE
LI CHUANGZHI
ZHANG GUIYUAN

Assignees

汕头大学

Dates

Publication Date: 20260508
Application Date: 20220711

Claims (6)

1. A photovoltaic array reconstruction method based on dual-objective optimization, the method comprising: Step 110, a photovoltaic array reconstruction model is established, wherein the photovoltaic array reconstruction model comprises an optimization function aiming at maximizing output power, an optimization function aiming at minimizing switching times and a switching constraint condition, and the optimization function aiming at maximizing the output power and the optimization function aiming at minimizing the switching times are recorded as a double-target optimization function; Step 120, performing iterative optimization on the photovoltaic array reconstruction model by using a SPEA2 algorithm to obtain a pareto front; 130, selecting a compromise solution from the pareto front by using a TOPSIS decision method, and further obtaining an optimal arrangement and combination scheme of the photovoltaic array; the step of performing iterative optimization on the photovoltaic array reconstruction model by using a SPEA2 algorithm to obtain the pareto front comprises the following steps: step 121, determining all photovoltaic modules currently shielded by shadows in the photovoltaic array, and simultaneously determining basic parameters required by a SPEA2 algorithm, including maximum iteration times, population sizes and external population sizes; Step 122, setting the current iteration number to k=0, and randomly initializing a plurality of initial arrangement combination schemes for generating the photovoltaic array in the photovoltaic array reconstruction model to obtain the current population Simultaneously creating an empty current external population ; 123, In the kth iteration, determining the current population by utilizing the fitness allocation strategy specified by the SPEA2 algorithm and combining the continuous variable discretization formula and the double-objective optimization function And the current external population The accurate fitness value of all individuals in the population is utilized to select and generate a new current external population by utilizing the environment specified by the SPEA2 algorithm Wherein the continuous variable discretization formula is applied to an optimization function aiming at minimizing the switching times, and the corresponding expression is as follows: in the formula, A first matrix formed by the electric switch states of all photovoltaic components on the photovoltaic array, wherein all elements in the first matrix are discrete values, The number of photovoltaic module columns arranged for the photovoltaic array, Is the first on the photovoltaic array A second matrix formed by the electric switch states of all the photovoltaic modules defined by the columns, wherein all elements in the second matrix are continuous values, and are obtained by iteration processing of a SPEA2 algorithm, Refers to a ranking function; 124, judging if k reaches the maximum iteration number, if so, then the new current external population is obtained The non-dominant individuals in (a) constitute the pareto front; if not, go to step 125; step 125, based on the new current external population Generation of a new current population using mating selection and crossover variation specified by the SPEA2 algorithm And assigning k+1 to k, and returning to step 123.
2. The photovoltaic array reconstruction method based on the dual objective optimization according to claim 1, wherein in the step 110, the expression of the dual objective optimization function is: Wherein: The switch constraint conditions are as follows: in the formula, For the output power of the photovoltaic array, For the corresponding number of switching times of the photovoltaic array, As the total output voltage of the photovoltaic array, As the total output current of the photovoltaic array, The number of rows of photovoltaic modules provided for the photovoltaic array, As a function of the sign of the symbol, Is the first on the photovoltaic array Line and th The current electrical switching state of the individual photovoltaic modules defined by the columns, Is the first on the photovoltaic array Line and th The initial electrical switching state of the individual photovoltaic modules defined by the columns, And Are all of a discrete value, and the two values are all of a discrete value, Is the first on the photovoltaic array Line and th The output current of the individual photovoltaic modules defined by the columns, Is the first on the photovoltaic array The total output voltage of all photovoltaic modules defined by the rows.
3. The photovoltaic array reconstruction method based on the dual-objective optimization of claim 1, wherein the implementation process of step 130 includes: Step 131, creating a decision matrix based on the pareto front comprising a plurality of pending permutation and combination schemes; step 132, determining a weighted decision matrix associated with the decision matrix based on the given weight vector; And 133, determining a positive ideal solution and a negative ideal solution based on the weighted decision matrix, calculating the weighted decision matrix by using a relative proximity maximization defined by a TOPSIS decision method as a screening criterion, and finally determining an optimal arrangement combination scheme from the decision matrix according to a calculation result.
4. The photovoltaic array reconstruction method based on the dual-objective optimization according to claim 3, wherein the optimal arrangement and combination scheme is an optimal electrical connection relationship between all photovoltaic modules defined by each column on the photovoltaic array.
5. A photovoltaic array reconstruction system based on dual objective optimization, the system comprising: The building module is used for building a photovoltaic array reconstruction model, wherein the photovoltaic array reconstruction model comprises an optimization function aiming at maximizing output power, an optimization function aiming at minimizing switching times and a switching constraint condition, and the optimization function aiming at maximizing the output power and the optimization function aiming at minimizing the switching times are recorded as double-target optimization functions; The optimization module is used for carrying out iterative optimization on the photovoltaic array reconstruction model by utilizing a SPEA2 algorithm to obtain a pareto front; The selecting module is used for selecting a compromise solution from the pareto front by utilizing a TOPSIS decision method so as to obtain an optimal arrangement and combination scheme of the photovoltaic array; the step of performing iterative optimization on the photovoltaic array reconstruction model by using a SPEA2 algorithm to obtain the pareto front comprises the following steps: step 121, determining all photovoltaic modules currently shielded by shadows in the photovoltaic array, and simultaneously determining basic parameters required by a SPEA2 algorithm, including maximum iteration times, population sizes and external population sizes; Step 122, setting the current iteration number to k=0, and randomly initializing a plurality of initial arrangement combination schemes for generating the photovoltaic array in the photovoltaic array reconstruction model to obtain the current population Simultaneously creating an empty current external population ; 123, In the kth iteration, determining the current population by utilizing the fitness allocation strategy specified by the SPEA2 algorithm and combining the continuous variable discretization formula and the double-objective optimization function And the current external population The accurate fitness value of all individuals in the population is utilized to select and generate a new current external population by utilizing the environment specified by the SPEA2 algorithm Wherein the continuous variable discretization formula is applied to an optimization function aiming at minimizing the switching times, and the corresponding expression is as follows: in the formula, A first matrix formed by the electric switch states of all photovoltaic components on the photovoltaic array, wherein all elements in the first matrix are discrete values, The number of photovoltaic module columns arranged for the photovoltaic array, Is the first on the photovoltaic array A second matrix formed by the electric switch states of all the photovoltaic modules defined by the columns, wherein all elements in the second matrix are continuous values, and are obtained by iteration processing of a SPEA2 algorithm, Refers to a ranking function; 124, judging if k reaches the maximum iteration number, if so, then the new current external population is obtained The non-dominant individuals in (a) constitute the pareto front; if not, go to step 125; step 125, based on the new current external population Generation of a new current population using mating selection and crossover variation specified by the SPEA2 algorithm And assigning k+1 to k, and returning to step 123.
6. A computer device, comprising: One or more processors; A memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications configured to perform the dual-objective optimization-based photovoltaic array reconstruction method of any one of claims 1 to 4.

Description

Photovoltaic array reconstruction method and system based on double-objective optimization and computer equipment Technical Field The invention relates to the technical field of photovoltaic array reconstruction, in particular to a photovoltaic array reconstruction method, a system and computer equipment based on double-objective optimization. Background The photovoltaic array is easy to be shielded by surrounding buildings, wires and other external factors to cause uneven illumination, so that the total output power of the photovoltaic array is reduced, and the hot spot effect can be generated even the overall electrical performance of the photovoltaic array is changed when the total output power is severe. From this point on, the technical staff proposed a photovoltaic array reconstruction way to solve the above problem, namely, changing the electrical connection state between a plurality of photovoltaic modules on the photovoltaic array by some means. At present, a scholars propose a photovoltaic array reconstruction method based on a TCT structure, namely, the photovoltaic array with local shading phenomenon is ordered and iterated according to a circuit principle according to a biological genetic law, the output power value after each iteration is evaluated until reaching a stable state, and then array adjustment is carried out according to the current arrangement sequence. However, the method has the disadvantages that the local compensation mode can effectively improve the total output power of the photovoltaic array, but the implementation complexity of introducing a hardware circuit is relatively high, and the array adjustment process needs to consume time and cost. In addition, scholars propose a photovoltaic array reconstruction method based on hierarchical inverter topology, namely, firstly classifying different inverters, then classifying photovoltaic modules in a shading state and photovoltaic modules not in the shading state, and respectively entering the inverters with different rated powers, thereby solving the influence of parallel mismatch and local shading. However, the method has certain disadvantages that the number of the inverters is large, the switching times of the switches are not considered, the problems of increased line loss, ageing of the switches and the like can occur, and the economic benefit is reduced. Disclosure of Invention The invention provides a photovoltaic array reconstruction method, a photovoltaic array reconstruction system and a photovoltaic array reconstruction computer device based on double-objective optimization, which are used for solving one or more technical problems in the prior art and at least providing a beneficial selection or creation condition. The embodiment of the invention provides a photovoltaic array reconstruction method based on double-objective optimization, which comprises the following steps: Step 110, building a photovoltaic array reconstruction model, wherein the photovoltaic array reconstruction model comprises an optimization function aiming at maximizing output power, an optimization function aiming at minimizing switching times and a switching constraint condition; Step 120, performing iterative optimization on the photovoltaic array reconstruction model by using a SPEA2 algorithm to obtain a pareto front; and 130, selecting a compromise solution from the pareto front by using a TOPSIS decision method, and further obtaining an optimal arrangement and combination scheme of the photovoltaic array. Further, in the step 110, the optimization function targeting the output power maximization and the optimization function targeting the switching times minimization are recorded as a dual-target optimization function, and the corresponding expression is: Wherein: The switch constraint conditions are as follows: Wherein f 1 is the output power of the photovoltaic array, f 2 is the number of switching times of a switch corresponding to the photovoltaic array, V our is the total output voltage of the photovoltaic array, I out is the total output current of the photovoltaic array, n row is the number of rows of the photovoltaic module arranged by the photovoltaic array, n col is the number of columns of the photovoltaic module arranged by the photovoltaic array, sign is a sign function, For the current electrical switching state of the individual photovoltaic modules defined by the p-th row and the q-th column on the photovoltaic array,For an initial electrical switching state of a single photovoltaic module defined by the p-th row and the q-th column on the photovoltaic array,AndAre discrete values, I pq is the output current of a single photovoltaic module defined by the p-th row and the q-th column on the photovoltaic array, and V mp is the total output voltage of all photovoltaic modules defined by the p-th row on the photovoltaic array. Further, the implementation process of step 120 includes: step 121, determining all photovoltaic modules