Research On Self-calibration Of Water Quality Monitoring Sensor Based On Improved Support Vector Machine

With the increasingly serious environmental pollution,the increase of industrial waste water,pesticide fertilizer used in agricultural production and domestic garbage and sewage,the water quality characteristics are more complex.Environmental pollution has attracted great attention of the society,and the monitoring of water quality has become a key concern of the state.At present,the traditional water quality monitoring methods have shortcomings in dealing with non-linear errors,which makes the evaluation results difficult to meet the standard requirements.Correct selection of correction algorithm model is the core content of water quality monitoring research.Based on the analysis of the research status of water quality monitoring at home and abroad,and taking turbidity parameters of water quality as the research object,this paper analyzed the principle of detection methods and related intelligent algorithms commonly used in water quality monitoring,designed turbidity sensor,and combined with the experimental platform and the analysis of experimental results,studied the related theories and technologies of support vector machine and grid search,particle swarm optimization,genetic algorithm and so on.Taking the turbidity of water quality measured by sensors as sample data,a self-tuning method of water quality parameters based on improved support vector machine is studied.The details are as follows:(1)Sensor design,study the principle of sensor detection,design turbidity sensor structure,light source,hardware circuit and control,acquisition,communication software,and elaborate the relationship between light intensity and detection voltage.(2)Regression algorithm of support vector machine is used as correction method for water quality monitoring.Taking turbidity parameters as an example,support vector machine is improved by grid search,genetic algorithm and particle swarm optimization.That is,penalty coefficients C and core parameters ? of two key parameters of support vector machine are searched comprehensively,and the best combination of parameters is selected to improve the efficiency and accuracy of water quality monitoring system.(3)Design of experimental platform and data analysis,including the design of data acquisition platform of lower computer,design of monitoring platform,improvement of support vector machine experiment and analysis of data results.In the paper,the application of improved support vector machine in water quality monitoring system is studied.The experimental results show that the improved support vector machine can accurately measure the parameters,meet the standard requirements and promote the development of water quality monitoring.It has certain theoretical and practical significance.