CN-121998470-A - Grid-connected performance evaluation method and device for photovoltaic power generation system

CN121998470ACN 121998470 ACN121998470 ACN 121998470ACN-121998470-A

Abstract

The invention discloses a grid-connected performance evaluation method and device of a photovoltaic power generation system, and particularly relates to the technical field of grid-connected performance evaluation of the photovoltaic power generation system; then, based on the extracted features, a system state identification model is constructed, multidimensional feature weight distribution is dynamically carried out by combining correction factors, and further, each dimension evaluation comprehensive score and each system grid-connected performance comprehensive score are obtained and evaluated respectively, and an optimization mechanism is triggered to a management terminal according to an evaluation abnormal result; according to the invention, the characteristics are extracted through four dimensions of topological structure, equipment matching, electric grid connection and energy efficiency loss, and a system state recognition mode and a dynamic weight distribution mechanism are constructed, so that the weight of each dimension can be adaptively adjusted according to the actual running state of the system, and a clear directional guide is provided for the formulation of a subsequent optimization strategy.

Inventors

WANG FEI
ZHANG JIAHAO
WANG WENQI
ZHANG JIARUO

Assignees

大唐蒲城第二发电有限责任公司

Dates

Publication Date: 20260508
Application Date: 20251212

Claims (9)

1. A grid-connected performance evaluation method of a photovoltaic power generation system is characterized by comprising the following steps: S1, collecting topological structure data, electrical parameters, equipment running state data and system energy efficiency loss data of a grid connection point of a photovoltaic power generation system through a monitoring device to obtain a grid connection performance data set; S2, carrying out feature extraction on the grid-connected performance data set to obtain a grid-connected performance multi-dimensional feature set, wherein the grid-connected performance multi-dimensional feature set comprises topological structure dimension features, equipment matching dimension features, electrical grid-connected dimension features and energy efficiency loss dimension features; S3, constructing a system state identification model based on the grid-connected performance multi-dimensional feature set, and dynamically carrying out multi-dimensional feature weight distribution by combining the identified system state mode and the correction factors to obtain a grid-connected performance evaluation dynamic weight distribution result; s4, calculating the comprehensive evaluation score of each dimension based on the grid-connected performance multi-dimensional feature set and the grid-connected performance evaluation dynamic weight distribution result, and calculating the grid-connected performance comprehensive score of the system through the comprehensive evaluation score of each dimension; And S5, evaluating the comprehensive evaluation scores of the dimensions and the comprehensive scores of the grid-connected performance of the system, triggering an optimization mechanism according to the abnormal evaluation results, and transmitting the information of the optimization mechanism to a management terminal for human-computer interaction.
2. The grid-connected performance evaluation method of the photovoltaic power generation system according to claim 1, wherein the topological structure dimension feature in S2 is extracted based on topological structure data in a grid-connected performance dataset, and an electrical wiring rationality index I to and an N-1 redundancy index I re are calculated respectively; The electrical wiring rationality index I to is obtained by weighting a preset backup switch configuration redundancy score f sw , a power outage overhaul isolation score f is and a wiring complexity penalty factor f co of an estimated electrical wiring point, wherein I to =a1×f sw +a2×f is -a3×f co , a1, a2 and a3 are respectively corresponding weight coefficients, the condition that a1+a2=1 is satisfied, and a3 is independent penalty weight; The N-1 redundancy index I re is obtained by calculating the average value of N-1 verification passing rates R j of N1 preset evaluation devices in the photovoltaic power generation system, wherein R j is a piecewise function, if the actual maximum power generation total active power P af j of all the rest devices of the system is more than or equal to the product of the rated total power P ra of the system and the N-1 verification passing threshold value eta th after the critical device j fails and exits operation, R j =1, otherwise R j =P af j /(P ra ×η th .
3. The grid-connected performance evaluation method of the photovoltaic power generation system is characterized in that the equipment matching dimension characteristic in the S2 is obtained by respectively calculating an inverter-component capacity matching index I mis and a reactive compensation equipment response index I q based on equipment operation state data in a grid-connected performance data set, wherein the I mis is obtained by subtracting average mismatch deviation of all N inverters from 1, and the mismatch deviation is obtained by subtracting an absolute difference value of 1 from a ratio P ac J of the total power P dc J of the nominal direct current side and the rated alternating current side of all photovoltaic components of a J-th inverter; t is an evaluation time window, Q ref (T) and Q act (T) are respectively a reactive power reference value required by the system at the moment T and reactive power actually output by reactive compensation, lambda is a decay constant, and the unit is one-half of the unit of reactive power.
4. The grid-connected performance evaluation method of the photovoltaic power generation system is characterized in that the electrical grid-connected dimensional characteristics in the S2 are extracted based on electrical data in a grid-connected performance data set, a voltage safety operation index I vs and a harmonic-power grid fitness index I har are calculated respectively, the I vs is a piecewise function, if the voltage V pcc of a public connection point where the photovoltaic power generation system and an upper public power grid interact is larger than or equal to the rated voltage V nom ,I vs of the power grid within an evaluation time window T and is 1, if a voltage safety lower limit threshold V low ＜V pcc ＜V nom ,I vs =(V pcc -V low )/(V nom -V low is 0, and if V pcc ≤V low ,I vs is 0, the I har is calculated by combining an H-order harmonic current effective value I h and a corresponding H-order harmonic current allowable upper limit value I h,max of the public connection point under a preset highest harmonic frequency H.
5. The grid-connected performance evaluation method of the photovoltaic power generation system is characterized in that the energy efficiency loss dimension characteristic in S2 is obtained by extracting the energy efficiency loss dimension characteristic based on system energy efficiency loss data in a grid-connected performance dataset, calculating a system comprehensive efficiency index I eff and a preset evaluation line transmission efficiency index I loss respectively, wherein the I eff is obtained by integrating total alternating current power P aac (T) output by the system at the moment T, solar irradiance G (T) at the moment T, total area A of a photovoltaic array and conversion efficiency eta stc under the standard test condition of the photovoltaic module in a combination manner in an evaluation time window T, and the I loss is obtained by subtracting the ratio of average loss power P lo-ke of N2 preset evaluation lines to average output power P tot of the system in the evaluation time window T through 1, namely I loss =1-P lo-ke /P tot .
6. The grid-connected performance evaluation method of the photovoltaic power generation system according to claim 1, wherein the grid-connected performance evaluation dynamic weight distribution result in the step S3 is obtained by respectively constructing a system steady state model M1, a system fluctuation state model M2, a system fault state model M3 and a system grid weak model M4 based on an N-1 redundancy index I re , a voltage safety operation index I vs , a system comprehensive efficiency index I eff and an inverter health rate I nb in a multi-dimensional feature set of the grid-connected performance, wherein the inverter health rate I nb is obtained by the ratio of the number of inverters normally generating power in the photovoltaic power generation system to the total number of inverters; Based on the N-1 redundancy index I re , the inverter-component capacity matching index I mis , the voltage safety operation index I vs and the preset evaluation line transmission efficiency index I loss in the grid-connected performance multidimensional feature set, a correction factor C t of a topological structure dimension, a correction factor C e of a device matching dimension, a correction factor C g of an electric grid-connected dimension and a correction factor of an energy efficiency loss dimension are calculated respectively C L ;C t =2-I re ,C e =2-I mis ,C g =1+max(0,1-I vs ),C L =2-I loss ; And dynamically carrying out multidimensional feature weight distribution by combining the identified system state mode and the correction factor to obtain the weight omega m of each dimension, thus obtaining the dynamic weight distribution result of grid-connected performance evaluation.
7. The grid-connected performance evaluation method of the photovoltaic power generation system according to claim 1, wherein the grid-connected performance integrated score of the system in S4 is obtained by firstly carrying out weighted summation on indexes calculated in all dimensional characteristics of grid-connected performance and weights corresponding to the indexes to obtain all dimensional evaluation integrated scores SC m , and then carrying out weighted summation on all dimensional evaluation integrated scores S m and weights omega m of all the dimensions to obtain the grid-connected performance integrated score SC of the system.
8. The grid-connected performance evaluation method of the photovoltaic power generation system is characterized in that in S5, optimization mechanism information is determined to be local abnormality if the grid-connected performance integrated score SC is larger than or equal to a corresponding threshold SC th and each dimension evaluation integrated score SC m is larger than or equal to a corresponding exclusive threshold SC m,th , the grid-connected performance of the photovoltaic power generation system is good, if SC is larger than or equal to SC th , SC m ＜SC m,th exists and the number of abnormal dimensions is smaller than or equal to N3, the corresponding dimension optimization mechanism is triggered, the optimization mechanism information is generated and transmitted to a management terminal for human-computer interaction, the optimization mechanism information comprises a system grid-connected performance integrated score, an abnormal dimension evaluation integrated score, a current system state mode and a corresponding dimension abnormal optimization suggestion, otherwise, the flow of SC < SC th is shifted, if SC < SC th , the difference value delta SC m of SC m,th subtracted SC m is calculated, the difference value delta SC m which is larger than or equal to 0 is screened out, the optimization priority is ordered according to the size, the optimization mechanism information is generated and transmitted to the management terminal for human-computer interaction, and the optimization mechanism information comprises the system grid-connected performance integrated score, the abnormal dimension integrated score priority, the current system state mode and the corresponding dimension optimization suggestion are generated.
9. A grid-connected performance evaluation device of a photovoltaic power generation system, configured to use a grid-connected performance evaluation method of a photovoltaic power generation system according to any one of claims 1 to 8, comprising: The grid-connected performance data perception module is used for acquiring topological structure data, electrical data, equipment running state data and system energy efficiency loss data of a grid-connected point of the photovoltaic power generation system through the monitoring device to obtain a grid-connected performance data set; The grid-connected performance multi-dimensional feature extraction module is used for carrying out feature extraction on the grid-connected performance data set to obtain a grid-connected performance multi-dimensional feature set, wherein the grid-connected performance multi-dimensional feature set comprises topological structure dimension features, equipment matching dimension features, electrical grid-connected dimension features and energy efficiency loss dimension features; The grid-connected performance evaluation dynamic weight distribution module is used for constructing a system state identification model based on the grid-connected performance multi-dimensional feature set, dynamically carrying out multi-dimensional feature weight distribution by combining the identified system state mode and the correction factor to obtain a grid-connected performance evaluation dynamic weight distribution result; The grid-connected performance multi-dimensional coupling evaluation module is used for calculating the comprehensive evaluation score of each dimension based on the grid-connected performance multi-dimensional feature set and the grid-connected performance evaluation dynamic weight distribution result, and calculating the grid-connected performance comprehensive score of the system through the comprehensive evaluation score of each dimension; And the grid-connected performance evaluation and optimization module evaluates the comprehensive scores of the dimension evaluation and the comprehensive scores of the grid-connected performance of the system, triggers an optimization mechanism according to the evaluation abnormal result, and transmits the information of the optimization mechanism to the management terminal for human-computer interaction.

Description

Grid-connected performance evaluation method and device for photovoltaic power generation system Technical Field The invention relates to the technical field of grid-connected performance evaluation of photovoltaic power generation systems, in particular to a grid-connected performance evaluation method and device of a photovoltaic power generation system. Background The distributed photovoltaic power generation is a photovoltaic power generation facility built near a user site, the power generation, grid connection and use are completed nearby by adopting a self-power-utilization and residual power internet surfing mode, and the grid connection performance of the distributed photovoltaic power generation system directly influences the stability, the electric energy quality and the system operation efficiency of a power grid after the power grid is connected with the power grid along with the transformation of a global energy structure and the rapid development of renewable energy sources. However, grid-tie performance of a distributed photovoltaic power generation system is affected by a number of factors including, but not limited to, equipment selection, electrical wiring, reactive compensation, grid access conditions, and the like. The existing grid-connected performance evaluation method of the photovoltaic power generation system generally only carries out evaluation (such as power factor, harmonic content and the like) around the electrical grid-connected indexes, and carries out weighted summation on all evaluation indexes by adopting a fixed weight distribution system through an analytic hierarchy process, an entropy weight method and the like to obtain a comprehensive evaluation result of grid-connected performance, so that the evaluation of the grid-connected performance of the photovoltaic power generation system is realized. In the prior art, grid-connected performance evaluation can be realized, but the method has some limitations, such as multi-focusing on static monitoring of electrical parameters, lack of fusion analysis of multidimensional data such as topological structure, equipment matching, energy efficiency loss and the like, and lack of dynamic weight adjustment and decision optimization mechanisms based on real-time running states, so that evaluation results are one-sided and decision response lag, effective support cannot be provided for intelligent operation and maintenance of a photovoltaic power station, evaluation results lag, and immediate and accurate guidance cannot be provided for online optimization regulation and control of a system. Therefore, there is a need for a comprehensive performance assessment method that can be depth-aware of operating conditions, weight dynamic adjustment, and closed-loop linkage with optimization control. Disclosure of Invention In order to overcome the above-mentioned drawbacks of the prior art, embodiments of the present invention provide a method and an apparatus for evaluating grid-connected performance of a photovoltaic power generation system, so as to solve the problems set forth in the background art. In order to achieve the above purpose, the invention provides a grid-connected performance evaluation method of a photovoltaic power generation system, comprising the following steps: S1, collecting topological structure data, electrical parameters, equipment running state data and system energy efficiency loss data of a grid connection point of a photovoltaic power generation system through a monitoring device to obtain a grid connection performance data set; S2, carrying out feature extraction on the grid-connected performance data set to obtain a grid-connected performance multi-dimensional feature set, wherein the grid-connected performance multi-dimensional feature set comprises topological structure dimension features, equipment matching dimension features, electrical grid-connected dimension features and energy efficiency loss dimension features; S3, constructing a system state identification model based on the grid-connected performance multi-dimensional feature set, and dynamically carrying out multi-dimensional feature weight distribution by combining the identified system state mode and the correction factors to obtain a grid-connected performance evaluation dynamic weight distribution result; s4, calculating the comprehensive evaluation score of each dimension based on the grid-connected performance multi-dimensional feature set and the grid-connected performance evaluation dynamic weight distribution result, and calculating the grid-connected performance comprehensive score of the system through the comprehensive evaluation score of each dimension; And S5, evaluating the comprehensive evaluation scores of the dimensions and the comprehensive scores of the grid-connected performance of the system, triggering an optimization mechanism according to the abnormal evaluation results, and transmitting the information of the optimiza