Application Of Air Quality Assessment With Support Vector Machine Based On Improved Wolf Colony Algorithm

In recent years,there have been more and more examples of swarm intelligence optimization algorithms in scientific research and engineering applications.Wolvescens algorithm GWO is a new intelligent optimization algorithm.Due to simple parameter setting and large space for optimization,it is becoming more and more Many scholars used research and improvement and applied it to engineering practice in combination with other algorithms.In the current important period of machine learning and data mining,support vector machines as a new generation of learning systems based on the latest developments in statistical theory,has become the focus of scholars to study,it is a two-class classification model.With the development of society and economy and the improvement of people's living standards,the issue of urban air quality has received more and more attention and attention.According to “Ambient Air Quality Standard”(GB 3095-2012),air quality index is used to evaluate air quality..This paper proposes an improved wolves algorithm.At the same time,this method is used to solve the parameter optimization problem of the support vector machine model.The optimized support vector machine has achieved better classification results in the evaluation of air quality data..The main research contents of this article mainly include the following aspects:The first is to improve the problems existing in the wolves algorithm GWO.The optimization of the selection operator is used to increase the initial population quality of the wolves and increase the initial quality of wolves;The step size of the wolves increases,and the lifting model jumps out of the level of the local optimal solution.Based on the optimization of the position weights,the dynamic response performance of the model is improved.When the wolves relocate,the greater weight is to move closer to the wolf.Proximity to the wolf and wolf increases the speed at which the wolves approach the prey and increase the probability of catching the prey at a limited number of iterations.The benchmark function is used to verify the optimization effect of the wolves algorithm.The experimental results show that the optimized wolves algorithm is superior to the compared algorithm in terms of convergence speed and convergence accuracy.The second is to combine the improved wolves algorithm and support vector machine,optimize the penalty factor and kernel function parameters of the support vector machine model,and establish a support vector machine model based on the improved wolves swarm algorithm to classify air quality in Chengdu.Then collect Chengdu city air quality data,according to Chengdu air quality evaluation criteria,determine air quality assessment factors,preprocess data,support vector machine model simulation experiment,initialize wolf population parameters,use support vector machine model based on evaluation factors.Classify air quality.The final experimental results show that the optimized SVM model outperforms other comparison algorithms for air quality data set classification and evaluation performance.The improved support vector based model is used to classify air quality,and statistical analysis is performed on the classification results as a guide.The reference for people's production and life has made people more aware that air quality assessment has important practical significance in controlling air pollution,providing a scientific basis for the management and decision-making of environmental protection departments,and pave the way for the reduction or prevention of air pollution.Local governments lay the theoretical foundation for environmental governance and have very important practical significance for improving,building,and continuing to develop smart cities.