Research On Water Quality Parameter Prediction Model Based On Improved Neural Network

With the continuous progress of science and technology,many emerging technologies have been proposed,such as Artificial Intelligence(AI),Internet of Things(IoT),Cloud Computing,and so on.As an important part of the new generation of information technology,the Internet of things technology has been deeply studied in various scenarios,such as smart city,smart home,and environmental monitoring.The emergence of the Internet of things has greatly improved the convenience of people's life.However,with the increasing types and number of sensor devices,the amount of data to be processed is also increasing.How to efficiently and accurately process a large number of data has always been the focus of people's research.The data of the Internet of things has the characteristics of magnanimity,heterogeneity,and timeliness,so the fusion of data is a difficult problem which is worth studying.As a method of Data Fusion(DF),Neural Network(NN)shows its strong advantages in many kinds of data fusion schemes.It abstracts the neurons in the human brain and builds different models from the perspective of information processing to form different networks through different connections,which can be used for analysis and prediction.There are many advantages of the neural network,but also some disadvantages,such as local optimization,poor generalization ability.Therefore,other algorithms are often used to overcome its shortcomings in practical applications.Among these algorithms,the application of the heuristic algorithm is the most studied.Many heuristic algorithms have been proposed and applied in different Internet of things scenarios.This paper mainly studies the theories of the Internet of things,data fusion,and neural networks are studied.This paper analyzes the shortcomings of the Back Propagation(BP)neural network algorithm,which is one of the common neural network algorithms,as well as the common ideas of using a heuristic algorithm to improve.An improved BP neural network model based on Seagull Optimization Algorithm(SOA)is proposed,and the performance of the proposed model is analyzed and verified for specific water quality parameter prediction scenarios.Focusing on the shortcomings of the seagull optimization algorithm itself,an improved idea is proposed,and the performance of the improved algorithm is verified by using the reference function.Using the new algorithm to improve the neural network,a new model is obtained,which is Improved Seagull Optimization Algorithm(ISOA)to improve the BP neural network model.Simulation results show that the proposed improved algorithm has a significant improvement in prediction accuracy and can be better applied in practice.