| In recent years,with the development of water quality monitoring technology,BOD has gained more and more research due to the advantages in monitoring organic pollutants.Microbial sensor method is more suitable for monitoring BOD in water because of the advantages of good timeliness,high sensitivity,high accuracy and simple operation.It is used in the research of BOD monitor widely.To solve the problem of BOD monitoring in seawater,a microbial film sensor based on microbial film reactor was prepared.Then an BOD monitoring system was built based on microbial film sensor.The BOD standard solution and actual sea water samples were measured.Finally,the influence of main environmental parameters in the monitoring process was eliminated.The main contents of this paper include:Firstly,based on the principle of microbial film sensor,the BOD microbial sensor was prepared by using the inner wall of spiral glass tube as the carrier and the microorganism in polluted seawater as the microbial source.The performance of the sensor,including response time,accuracy,stability and service life,was tested.Finally,the working conditions of BOD microbial film sensor are selected and optimized,which lays a foundation for the construction of BOD monitoring system.Secondly,based on the analysis of the working conditions of microbial film sensors and the functional requirements of the system,an monitoring system for seawater BOD is designed and built,which realizes the functions of automatic liquid feeding,constant temperature control,data acquisition and processing.The performance of the monitoring system is analyzed,including the determination of standard curve,accuracy test and recovery rate test.Then the actual water samples of Qinhuangdao were measured and compared with the standard dilution method.The performance of BOD monitoring system for seawater in practical application is analyzed.Finally,based on the principle of information fusion,an improved extreme learning machine(ELM)algorithm based on particle swarm optimization(PSO)is proposed.PSO algorithm is used to optimize the output connection weights and hidden layer neuron thresholds in ELM algorithm to reduce the calculation error of ELM algorithm.Then,the water quality parameters pH,SS and ORP,which have great influence on the activity of microbial membrane,are used as the input of the model for network in-depth learning to eliminate the influence of parameters and output the final BOD value of seawater samples and improve the accuracy of the monitoring results of seawater BOD monitoring system. |
