A deep learning approach to real-time CO concentration prediction at signalized intersection

Özet

Vehicle exhaust emissions at signalized intersections are the essential source of traffic-related pollution to pedestrians.Therefore, it is critical to predicting traffic emissions, especially the hazardous CO gas, with practicaland accurate methods. However, the CO emission and concentration at crosswalks can be influenced by thecomplex traffic conditions in a complicated way, making the prediction of CO concentration a challenging taskfor traditional statistical models. To this end, a hybrid machine learning framework is proposed in this study toinvestigate the concentration of CO emissions at pedestrian crosswalks. The proposed method firstly ranks keyinfluencing factors with a random forest approach. Then a prediction model with Multi-Variate Long Short-TermMemory (LSTM) neural networks based on the selected factors is developed. Data is collected at the field intersectionfor model training and validation. The autoregressive integrated moving average (ARIMA), supportvector machines (SVM), radial basis functions network (RBFN), nonlinear vector autoregressive (VAR) and gatedrecurrent unit (GRU) neural network are selected as the benchmark models to verify the performance of theproposed model. The Root Mean Square Errors (RMSE), Mean Absolute Error (MAE) and R square are calculatedto evaluate the performance of models comprehensively. The results indicated that the proposed model overwhelmsthe benchmark models in terms of prediction accuracy.