The purpose of this study is to solve the problem of oxidation-reduction reaction of ethylene/acetylene mixture gas in the electrochemical carbon monoxide element, which generates current and affects the output of the measurement circuit, resulting in severe distortion of the carbon monoxide gas concentration detection value in the sensor output and to ensure the correct carbon monoxide gas concentration monitoring value output by the multi-parameter sensor.
This paper proposes a least squares polynomial regression model to compensate for the deviation of carbon monoxide concentration detection values under the interference of various combustible and harmful gases on carbon monoxide electrochemical components.
The experiment showed that the model has fast training speed, high execution efficiency and simple algorithm. Compared with the radial basis function neural network model, it has significant advantages. Through experimental analysis and demonstration, the model can basically control the offset between (−2.2) and has great practical value.
This research topic is novel and highly original.
