Motor vehicle exhaust concentration prediction method and system based on machine learning

Abstract

The invention discloses a motor vehicle exhaust concentration prediction method and system based on machine learning, and the method comprises the following steps: S1, importing vehicle exhaust data from a corresponding database into an excel or csv file, and carrying out the preprocessing of the data, and obtaining the preprocessed exhaust data; s2, constructing a model framework, combining a neural network method and a support vector regression method in machine learning, adopting a multi-model fusion method, and fusing two models to perform tail gas concentration inversion; s3, utilizing the preprocessed tail gas data to train a tail gas concentration inversion fusion model, wherein a neural network part is used for memorizing deep-level tail gas data, and support vector regression is used for inverting the tail gas concentration; and S4, inputting vehicle tail gas data to predict the tail gas concentration by using the trained tail gas concentration inversion fusion model. The method is high in detection precision and easy to popularize; and multiple feature processing is performed, so that the accuracy is high, and the accuracy can be improved by performing feature processing on the original data twice through the neural network model and the support vector regression model.

Description

technical field [0001] The invention relates to the technical field of environmental measurement and detection, in particular to a method and system for predicting the concentration of motor vehicle exhaust gas based on machine learning. Background technique [0002] The development of transportation technology has brought many conveniences to human life, but it has also brought exhaust pollution that threatens human health. In particular, the characteristics of many types of motor vehicles, wide distribution and fast flow have aggravated the pollution of the environment, not only causing adverse effects on the environment but also seriously endangering human health. Therefore, the detection of vehicle exhaust concentration under different environments and driving conditions is a key issue in improving air quality. However, there are still many problems in the current method of tail gas concentration inversion, such as high cost, no universality and low detection accuracy, ...

AI Technical Summary

Problems solved by technology

[0006] A method and system for predicting the concentration of motor vehicle exhaust gas based on machine learning proposed by the present invention can effectively overcome the disadvantage of high cost, has universal applicability, and has high detection accuracy

Images