Evaluation Of Ash Accumulation Degree Of Photovoltaic Arrays And Determination Of Optimal Cleaning Time

In recent years,photovoltaic power generation has developed rapidly in the field of clean energy.However,from the analysis of the operation and maintenance of photovoltaic power generation in recent years,the surface ash accumulation of arrays in long-term outdoor operation is serious,which has become a potential economic problem affecting the photovoltaic power output.Timely cleaning can improve the overall income of photovoltaic power stations.Therefore,Thesis aims at the maximum power gain brought by cleaning,and studies the modeling of ash accumulation array,evaluation of ash accumulation degree and the optimal cleaning time.Details are as follows:(1)The existing simulation methods of ash accumulation of photovoltaic array cannot accurately reflect the influence of ash accumulation on the output characteristics of the array.Based on the analysis of the influence of ash accumulation on the array,Thesis improved the ideal photovoltaic array from two aspects of light intensity and temperature,and established the ash accumulation analysis model of photovoltaic array under arbitrary light intensity and temperature.(2)Based on the ash accumulation analysis model of photovoltaic array,the output characteristics of array under different ash accumulation degree and ash mixed with other faults are simulated and analyzed.Taking light intensity,temperature,array voltage and current as the evaluation characteristics of ash accumulation degree,an array ash accumulation degree evaluation model based on Back Propagation neural network is established.The empirical test shows that the model has a good evaluation effect on the ash accumulation degree of the array,but when the ash accumulation array is mixed with other faults,the evaluation accuracy of the model decreases by 14.63%.(3)In view of the fact that the existing evaluation methods of ash accumulation degree of photovoltaic arrays cannot accurately resist the interference of other faults,Thesis analyzes the output characteristics of ash accumulation of photovoltaic arrays and other faults,and reveals that the electrical parameter of short circuit current can reflect the ash accumulation of photovoltaic arrays and is not susceptible to the interference of other faults.A Support Vector Machine model for evaluating ash accumulation degree based on Grid Search hyperparameter optimization is proposed,which takes short circuit current,light intensity and temperature as input characteristic parameters.Through experimental testing and analysis,it is proved that the accuracy of ash accumulation degree evaluation of the proposed method is higher than that of Decision Tree,GS-DT and e Xtreme Gradient Boosting evaluation methods,and it is not affected by the interference of other faults.(4)In view of the different cleaning time caused by the influence of the future weather on the electricity gain brought by cleaning,this paper uses the SARIMA model to predict the power of the array in the next period of time,and then combines the ash accumulation time model based on the measured data to calculate the electricity gain that can be obtained by cleaning at different times in the next period of time.The optimal cleaning time is determined by the maximum power gain.