Prediction Of AQI Based On BP Neural Network And SVR Algorithm Based On Optimization

With the development of industry and technology,air quality problems have attracted more and more attention.Governing air pollution is a long-standing problem that is difficult to achieve in the short term [1].In order to better calculate the Air Quality Index(AQI)and improve the calculation efficiency of the Air Quality Index(AQI)[2],this paper introduces an artificial neural network algorithm.The traditional air quality evaluation and prediction methods have the defects of large workload,difficult statistical data and low prediction accuracy.The method has good nonlinear fitting ability and has been widely used in dealing with nonlinear problems,especially BP neural network algorithm and SVM algorithm,which effectively solve some existing problems.The data used in this paper are all from the real-time monitoring data of China Environmental Monitoring Station Taiyuan City.Artificial Bee Colony Algorithm(ABC)is a meta-heuristic cluster intelligence algorithm that mimics the behavior of bee colony division to find honey sources.The main feature is that it does not need to understand the special information of the problem.It only needs to compare the advantages and disadvantages of the problem.Through the local optimization behavior of each artificial bee individual,the superior value is finally highlighted in the group,and the convergence speed is faster.However,the ABC algorithm may fall into local optimum.In view of this shortcoming,this paper proposes BP-ABC and BP-IABC model algorithms,mainly using ABC algorithm to optimize the weight and threshold of BP algorithm.The BP model is used to predict the AQI index.The experimental results show that the BP-IABC model has higher fitting accuracy and smaller mean square error than BP model and BP-ABC model.The Grey Wolf Optimizer(GWO)simulates the population hierarchy of the greywolf population and the activities surrounding the prey.The gray wolf is scattered to find the target,and the quality of the initial population directly affects the global convergence speed of the algorithm and the quality of the solution.In order to improve the global optimization ability and optimize the parameters of Support Vector Machine Regression Algorithm(SVR)with improved algorithm,the MGWO-SVR prediction model is established.The MGWO-SVR model and SVR model are used to predict and fit the air quality index of Taiyuan City.The simulation results show that the improved model has good performance in prediction accuracy and performance.