Research On Air Quality Prediction Technology Based On LSTM-CNN-DRL Integrated Model

With the continuous development of science and technology,rapid urbanization,air pollution is becoming increasingly serious,not only will seriously endanger the health of plants and animals and human beings,but also significantly affect the weather and climate,thus limiting the development of many developing countries and even developed countries.And bad air quality can seriously affect human health and even lead to death.Effective prediction of air quality is conducive to people’s concern and prevention of air pollution caused by health problems and take appropriate measures,but also facilitate the government to take timely decisions to control air pollution.Therefore,the study of air quality index prediction is particularly important.The purpose of this paper is to study the depth learning and depth reinforcement learning model for the prediction of air quality index,and to improve the accuracy and accuracy of the statistical prediction method of AQI against the background of the prediction of AQI in major cities in China.Aiming at the disadvantage of low precision and high threshold of domain knowledge of traditional air quality prediction methods,an integrated model of long short-term memory network and convolution neural network for air quality index prediction is proposed in this paper.The model combines LSTM network with CNN network through series mechanism to maximize LSTM pairing.The advantage of time series prediction and CNN’s advantage in feature recognition.After a lot of experiments and analysis,the validity of the model is verified.Aiming at the disadvantage of the traditional depth learning model in the field of air quality prediction,this paper proposes a depth reinforcement learning model for the prediction of air quality index grades.The model of reinforcement learning process is built by predicting the grade of air quality index,and a classification model of air quality index grade is constructed by using Double DQN algorithm in depth reinforcement learning.The model pioneered the application of the deep reinforcement learning technology in the field of air quality index prediction,and achieved some obvious results.Finally,in order to improve the accuracy of the air quality index prediction model proposed in this paper,the long short-term memory network for air quality index prediction,the convolution neural network series model and the depth reinforcement learning model for air quality index grade prediction are integrated to learn,and the DNN neural network is combined.The learning strategy of the network forms a new integrated model,which achieves better prediction results.