Research On Photovoltaic Array Fault Early Warning Based On Artificial Intelligence Technology Expert System

With the development of science and technology and the use of non-renewable energy in the energy industry,non-renewable energy is becoming less and less,and the pollution caused to the environment has become more and more serious.The trend of exploring and developing new energy industries has become unstoppable,and development In the new energy industry,photovoltaic power generation is highly appreciated because of its clean and pollution-free advantages.Photovoltaic power generation uses solar energy to convert it into electrical energy.To allow the photovoltaic array to have plenty of sunlight,most photovoltaic power plants are built in relatively remote,sunny,and strong areas.With the vigorous development of the photovoltaic industry,there are more and more photovoltaic power stations,the use efficiency is getting higher and higher,and the failure of photovoltaic arrays is increasing.In daily life,the photovoltaic array will be unstoppable due to the influence of external natural factors,or human factors,and the aging of its components and some failures will occur unstoppably.A photovoltaic array is one of the most important pieces of equipment for photovoltaic power generation,and the loss caused by failure to troubleshoot in time will not be estimated.Therefore,it is extremely urgent to improve the failure prediction capability of photovoltaic arrays.This paper focuses on the problem of photovoltaic array failure prediction,and the main contributions are as follows:First,a fault diagnosis and reasoning method based on production rules is proposed.A fault tree is established for various fault causes of the photovoltaic array power generation system,and the production rules of the photovoltaic array power generation system are obtained by analysis,and then the credibility of the production rules is calculated The degree factor is used to calculate the degree of similarity with the fact condition,to compare with the condition in the rule.If it is greater than that,the rule is triggered and the reliability of the conclusion is obtained.The second is to establish a case-based reasoning fault diagnosis method,and after analyzing the characteristic attributes of the photovoltaic array,choose to use the knowledge representation method of the matrix to express the fault information of the photovoltaic array,and then calculate the similarity between the case to be diagnosed and the single attribute in the source case After the degree of similarity and the overall similarity,the degree of similarity between the case to be diagnosed and each source case is obtained.Next,a knowledge diagnosis method based on neural network reasoning is established.Using the self-learning ability of neural networks and solving complex problems,it can effectively diagnose faults,use feature vectors to express the characteristics of the fault,and collect the characteristics of the fault.The vector input into the neural network can get the probabilities of various causes of the failure phenomenon.The deficiencies of the D-S evidence theory algorithm are proposed.Jousselme,the amount of evidence agreement and the strength of evidence conflict are used to improve the evidence conflict measurement standard,and the concepts of weight coefficient and information entropy are introduced to improve the evidence combination rule.It effectively solves the problems of high conflict evidence when the evidence is fused,which leads to the failure of fusion and a one-vote veto.Finally,the conclusions obtained based on the three reasoning methods of rules,cases,and neural networks are fused through the improved DS evidence theory proposed in this article,and the conclusions after the fusion of DS evidence theories are judged whether they are the most supported,and if so,then Take this conclusion as the final result and make corresponding early warning judgments.Compared with the traditional methods,the photovoltaic array fault diagnosis method proposed in this paper not only improves the efficiency of fault diagnosis and the utilization rate of fault information but also improves the accuracy of fault diagnosis and reduces the missed diagnosis compared with the single expert system diagnosis method.Rate,verify the effectiveness of the method in this paper.The functions of the expert system mentioned in this article are realized through MATLAB and ACCESS database so that ordinary workers can also perform fault early warning and monitoring of the photovoltaic array power generation system through this system.