NO_x Emission Prediction And Eco-driving Speed Control Strategy For Heavy Trucks

The rapid growth of the number of motor vehicles brings convenience to human life,but also leads to increasingly severe traffic congestion and environmental pollution problems.Heavy goods vehicles(HGVS)are the main contributors to road nitrogen oxide(NO_x)emissions,and as a major source of pollution,they have become the focus of research in the transportation field.Improving driving behavior and promoting ecological driving are important ways to reduce motor vehicle energy consumption and pollutant emissions.Aiming at the problem of high emission of heavy goods vehicles,this thesis starts from two aspects:emission prediction and ecological driving speed guidance.The main research contents and conclusions are as follows:(1)Design the collection method of driving condition data and NO_x emission data of heavy duty trucks,select test vehicles and routes,and apply remote on-board Diagnostics(OBD)system to read real-time operation data and NO_x emission data of multiple heavy duty trucks;Analyze the problems in the obtained data,design the data processing scheme,and carry out data cleaning,interpolation and smoothing processing on the original data to obtain higher quality data.(2)The key influencing factors of NO_x emission of heavy-duty trucks were analyzed,and speed and specific power,which are highly correlated with NO_x emission,were selected as input variables;The K-means clustering algorithm is used to divide the operating conditions into three types: low speed,medium speed and high speed.Polynomial emission prediction models under different working conditions are constructed,and Simulated Annealing(SA)algorithm based on Bayesian Information Criterion(BIC)criterion is adopted to simplify the calculation process.A NO_x emission prediction model with high accuracy was obtained by least square fitting.The comparison between the P(v,VSP)model proposed in this paper and the VTMicro model drew a conclusion: the P(v,VSP)model has the advantages of simple structure,low complexity and high accuracy,and has good universality.(3)Analyze the influence of driving behavior on heavy goods vehicle emissions.The judgment thresholds of 5 kinds of basic driving behaviors were determined,and the recognition algorithm was designed to analyze the influence of different driving behaviors on NO_x emission.The characteristic parameters of driving behavior were constructed,and the Principal Component Analysis(PCA)method was applied to reduce the dimensionality of a large number of characteristic parameters,and principal component factors that could represent driving behavior were selected,and the principal component factors were taken as input parameters of the emission model.A Long Short Term Memory(LSTM)emission prediction model based on attention mechanism was constructed.The accuracy of the model was high,and the NO_x emission of different driving behaviors could be well predicted.(4)Proposed ecological driving speed optimization strategies for road section environment and signal intersection environment respectively.The curve generated by the trigonometric vehicle speed planning method is smooth,which can realize smooth acceleration and deceleration.The trigonometric vehicle speed guidance method is used to optimize the driving behavior of the road section environment,guide the vehicle to slow acceleration and deceleration,and run smoothly,and the results show that about 20% NO_x emission can be reduced.For intersections with vehicle-road collaborative environment,speed guidance scenarios are divided according to signal timing and intersection driving behavior,and speed guidance strategies are designed for different scenarios,taking into account the goals of reducing emissions and improving traffic efficiency.Non-dominated Sorting Genetic Algorithm II(NSGA-II)with an elite strategy was used to solve the multi-objective optimization problem to get the optimal recommended speed trajectory under different scenarios.Finally,the feasibility and effectiveness of the proposed method are verified by comparing the intersection traffic under the guidance strategy of unguided strategy and trigonometric function method.