A nonlinear correction method for coal mine underground sensors based on genetic improved neural network
The working environment underground in coal mines is extremely harsh,including high humidity,high temperature,continuous vibration,and a large amount of dust.These factors may cause nonlinear interference to sensors,resulting in significant deviations between their output results and actual values.If this deviation is not corrected,it will seriously affect the production safety and efficiency of coal mines.Therefore,this study proposes a neural network method based on genetic algorithm optimiza-tion,aiming to correct the nonlinear error of underground sensors in coal mines.Determine the input and output of the calibration model based on the obtained sensor nonlinear error data.Construct a calibration model with neural networks as the core,and use genetic algorithms to optimize the weight and threshold parameters of the calibration model to improve its calibration performance.After the model training is completed,it is applied to the nonlinear correction of sensors in actual coal mine underground.The ex-perimental results show that after calibration by this research method,the gas concentration value is closer to the standard gas concentration value,and the error is significantly reduced.This fully demonstrates the effectiveness and reliability of this research method in practical applications.
万方数据
维普
