Research On Performance Experiment And Prediction Of Solar PV/T System

Compared with the traditional photovoltaic power generation technology,the solar photovoltaic photothermal(PV/T)utilization technology has many advantages.The system can obtain additional heat.The cooling of the photovoltaic cell improves its photoelectric efficiency and greatly improves the comprehensive utilization rate of solar energy.PV/T collector is a key component for the comprehensive utilization rate of the PV/T system,and how to improve the photothermal performance of the system to improve the photoelectric performance is a current research hotspot.The research group has developed an ultra-thin superconducting solar heat-absorbing panel with a core thickness of less than5 mm,which is seamlessly bonded to the photovoltaic panel through lamination technology,eliminating contact thermal resistance,reducing energy loss and strengthening effective energy.This paper uses a combination of theoretical analysis and experimental testing to study its performance;considering that there are many influencing factors of the system performance and the relationship is more complicated,and the neural network can handle nonlinear problems and has strong learning ability,this paper establishes a PV/T system The BP neural network prediction model of photoelectric thermal performance seeks the model with the best prediction accuracy by comparing different optimization algorithms,influencing factors,and data volume.The research results provide data reference for the design and research of similar PV/T systems,and provide new research methods for the promotion of integrated solar energy application technology,which is of great significance to China's energy conservation and emission reduction business.This paper firstly conducts a theoretical analysis of the photovoltaic and energy transfer situation of the PV/T system,uses a new type of ultra-thin superconducting solar heat absorption core to design and build a set of flat solar PV/T system experimental test platform,to provide follow-up experimental research Theoretical support and test platform.Then this paper conducts an experimental study on the photovoltaic thermal performance of the constructed solar PV/T system in Beijing in summer and autumn.Based on long-term test data,this paper analyzes the daily average values of outdoor temperature,solar radiation intensity,power,panel temperature,inlet water temperature,photoelectric efficiency,and photothermal efficiency,and then comparatively analyzes the photoelectric photothermal characteristics of PV/T system under time-varying weather.It is concluded that the daily average photoelectric efficiency of the PV/T system ranges from3.11 to 13.71%,the average daily photothermal efficiency ranges from 7.35 to 56.65%,and the daily average energy saving rate ranges from 24.85 to 86.39%.From the point of view of primary energy saving rate,August is the best.Finally,based on the obtained large amount of experimental data and data characteristics,this paper establishes a PV/T system photoelectric thermal efficiency prediction model based on BP neural network.The structure and parameter design of the neural network are analyzed.The influence of different numbers of factors and data sample size on the neural network prediction model is also analyzed.It can be seen that the accuracy of the model will be improved with the increase of the influencing factors,the amount of data and the appropriate design parameters.Finally,the predicted value of the model is compared with the actual calculated value,and it is found that the predicted result of the photoelectric efficiency is consistent with the actual value,and the fitting of the numerical accuracy is highly consistent;while the predicted result of the photothermal efficiency is similar to the actual value.This paper uses neural networks to predict the photoelectric thermal performance of the system during the operation of the system,analyzes the complex relationship between the factors,and realizes the prediction of the photoelectric thermal performance of the PV/T system through multiple influencing factors,providing similar system design and research Data reference.