Research on Calibration Method of NOx emission MAP Based on Fuzzy Model Mining
Facing the problems of excessive dependence on manual experience and high time cost in the process of determining the original NOx emission MAP of diesel engine,a NOx emission calibration method based on fuzzy model mining is proposed in this paper.Firstly,this method uses all collected NOx emission data to calculate the support degree of fuzzy rules,and realizes the mining and extraction of fuzzy rules through the selection of the maximum support.Secondly,by calculating the support of all data,the interference of outliers and noise is weakened and the robustness of the model is increased.In order to further improve the accuracy of the model,a parameter optimization process based on gradient descent method and input variable fuzzy space in-crement algorithm is designed in this paper.The whole algorithm is realized by multiple nested loop programs,which reduces the difficulty of model training and shortens the development period.It is suitable for engineering application.Finally,the modified method is applied to the calibration of the original NOx emission MAP of a diesel engine.The experimental results show the accuracy,robustness and practicability of the method.
Data miningFuzzy systemRule extractionNOx emissionCalibration
万方数据
维普
