Data Correction of Grid Air Quality Monitor Based on CIWOA-BP Neural Network
Air pollution is a serious threat to human health.In recent years,the miniature air quality monitor based on sensor technology(referred to as miniature monitoring station)has the advantages of small size and low cost,which conforms to the cur-rent grid and refined air quality management mode.However,the electrochemical sensors used in this device have complex gas cross-interference,which greatly affects the accuracy of its test.Aiming at the problem that the cross-interference is nonlinear and difficult to be described by explicit mathematical expressions,a chaotic mapping adaptive inertia weight whale optimization algo-rithm optimized back propagation(CIWOA-BP)neural network model was applied for the data correction of the miniature moni-toring station.The CIWOA-BP algorithm combined the advantages of BP neural network that is good at dealing with nonlinear black box problems with the ability of CIWOA for seeking global optimization.The results show that the miniature monitoring sta-tion corrected by CIWOA-BP can realize accurate quantitative analysis of NO2,CO,O3 and SO2 in gas mixture,and the goodness of fit(R2)between the calculated values and actual values of these four gases is more than 0.97.Moreover,the performance of this method is superior to that of sample,multiple linear regression and traditional BP neural network,which can effectively im-prove the monitoring accuracy of device for air pollutants.
grid air quality monitorminiature monitoring stationCIWOABP neural networkelectrochemical sensorsgas cross-interference
NETL
NSTL
万方数据
