Space-time prediction method for tail gas concentration of urban motor vehicles

Abstract

The invention discloses a space-time prediction method for the tail gas concentration of urban motor vehicles. Time-space data of the tail gas concentration of motor vehicles in a target city within ayear is collected and preprocessed, training, verification and test data sets based on Encoder-Decoder based lSTM neural network model are constructed, the training data set is sent the model for training to obtain a pre-training model parameter, the pre-training model parameter is tuned finely via the verification and test data sets, the model is further corrected to improve the model precision,the corrected model serves as a time-space prediction model for the tail gas concentration of urban motor vehicles, preprocessed concentration data of air pollutants within relatively long time of the target city service as input data of the model, the tail gas concentration of motor vehicles in certain area in the city at certain time can be output from the model, and thus, time-space predictionof the tail gas concentration is realized.

Description

technical field [0001] The invention belongs to the technical field of environmental monitoring, and relates to a motor vehicle exhaust concentration prediction method, in particular to a spatiotemporal prediction method of urban motor vehicle exhaust concentration. Background technique [0002] With social development and urban progress, in recent years, the number of motor vehicles in urban areas has continued to increase, and many social problems have emerged, such as serious urban traffic congestion, increased traffic accidents, motor vehicle exhaust pollution, drunk driving, etc. In big cities such as Beijing, Shanghai, and Guangzhou, motor vehicles have become the largest source of pollutants such as carbon monoxide, nitrogen oxides, and hydrocarbons. Since the emission of automobile exhaust gas is mainly between 0.3m and 2m, which happens to be within the breathing range of the human body, the health damage to the human body is very serious-stimulating the respiratory...

AI Technical Summary

Problems solved by technology

[0004] Due to the highly non-linearity of motor vehicle exhaust emissions, the concentration of pollutants is affected by multiple factors in the surrounding environment, including meteorological conditions, air environment, geographical environment, road conditions, traffic flow factors, etc., the existing technology can only be analyzed from time or space Prediction and prediction of pollutant concentration in a single dimension,

Images