Study On Inverse Model Of Negative Oxygen Ion Concentration Based On PM2.5

Air negative oxygen ion is known as life vitamin and is essential for people’s healthy life.Negative oxygen ion concentration value is also one of the important reference factors to evaluate the air quality level of indoor and forest scenic spots.Based on the important position of negative oxygen ions in various fields,it is imperative to strengthen the monitoring and research on the concentration of negative oxygen ions.The investigation found that it is common in China to purchase imported negative oxygen ion concentration sensor equipment to build a monitoring station to monitor negative oxygen ion concentration.Because the purchase cost of high-precision negative oxygen ion products is very high,it cannot meet the daily monitoring needs of the public and the long-term and multi-site monitoring needs of negative oxygen ion in the field.In order to reduce the monitoring cost and meet the needs of social monitoring,finding a new method for monitoring the concentration of negative oxygen ions will have high research value.Relevant literature shows that the negative oxygen ion concentration can maintain a good negative correlation with PM2.5 in the air.The correlation with other atmospheric factors is unstable,and it is easily affected by the regional environment.To study the mapping relationship between PM2.5 and negative oxygen ions,east lake campus of ZHEJIANG A&F UNIVERSITY,LINAN district,HANGZHOU city,was taken as the monitoring site.Used the self-built negative oxygen ion data collection platform which is combined with American EPEX negative oxygen ion detection equipment,to monitor the temperature,humidity,PM2.5 concentration and negative oxygen ion concentration.After preprocessing and correlation analysis of the collected data,the study confirmed that there is a long-term stable negative correlation between negative oxygen ion concentration and PM2.5,and the correlation coefficient is-0.7497Based on the negative correlation between PM2.5 and negative oxygen ion concentration,this study builds a negative oxygen ion concentration inversion model through nonlinear regression machine learning language.In order to select a more suitable algorithm model,the experiment uses support vector machine,BP neural network and GA-BP neural network algorithm for comparison and verification.The algorithm model takes the PM2.5 concentration value as input,and finally outputs the negative oxygen ion concentration value in the air.According to the prediction error of the final negative oxygen ion concentration of each algorithm model,this paper determines the GA-BP neural network algorithm model as the optimal model.The average error percentage of the negative oxygen ion concentration predicted by the GA-BP model is 7.46%,which can meet the needs of this study.On the basis of completing the construction of the model,this article carried out further portability verification of the model with the help of the Antigen negative oxygen ion detection instrument imported from Japan.Experiments show that the model has good stability and portability.In sunny weather,the output of the model is similar to the actual detection result of the Antianshi negative oxygen ion equipment,with an average error percentage of 5.89%.The construction and application of this research model will greatly reduce the monitoring cost of negative oxygen ion concentration.While promoting scientific research on negative oxygen ion concentration in colleges and universities,it can also meet people’s long-term monitoring needs of negative oxygen ion concentration and indirectly promote the development of negative oxygen ion concentration detection technology.