Research On Gasoline Engine Exhaust Emission Prediction System Based On OBD Data Flow

In the 21st century,air pollution caused by exhaust emissions from automobiles has become one of the main sources of pollution in cities.With the rapid development of China's automobile industry and the gradual improvement of people's living standards,the utilization rate of automobiles will be further enhanced.In the future,a series of environmental problems dominated by motor vehicle pollution,including smog,will become increasingly prominent.Each country is gradually increasing the limits of exhaust emissions for vehicles in use.However,at present,a complete vehicle exhaust emission detection system is costly,complicated in operation,large in floor space,limited in use conditions,and cannot be promoted to the market.Moreover,the existing methods can only judge whether the vehicle exhaust emission is up to standard.But they cannot accurately and quickly obtain the cause of excessive exhaust emission,and provide guidance for the use of the vehicle.So they lack practicality.The OBD system can realize real-time monitoring of vehicle status.Also,it can record and store data stream information.Through data flow analysis,the working state of the automotive electronic control system or related components can be analyzed and judged.Based on the above research background,this topic uses the OBD data stream to conduct a certain exploration and research on the gasoline engine exhaust emission prediction system.By establishing a pollutant emission evaluation system,it used vehicle state parameters to predict the pollutant emission level of the vehicle.It could provide a reference for diagnosing the cause of vehicle failure and provide reasonable advice for the owner and service unit to maintain and use the vehicle.Also,it can be an effective supplement to the implementation of the I/M system.Firstly,it analyzed the formation mechanism and related influencing factors of CO,HC and NO_X in the exhaust gas.The BP neural network model prediction was selected by analyzing the characteristics of gasoline engine exhaust emission and comparing various prediction methods.Exhaust emission level.Next,the OBD diagnostic instrument was used to obtain the car data flow information.According to the data flow information,load,speed,water temperature,intake air temperature,intake air amount,oxygen content in exhaust gas,target air-fuel ratio,ignition advance angle retardation angle,fuel pressure,throttle opening and intake manifold temperature were selected as the main state parameter of the gasoline engine.And it analyzed the effects of various characteristic parameters on the discharge levels of the three main pollutants.Based on relevant regulations and expert experience,it determined the evaluation level of the forecasting system.Based on the BP neural network diagnosis idea,the determined characteristic parameters were taken as inputs,the vehicle emission evaluation level was output,and the prediction model was constructed.Then,in order to further improve the accuracy of the prediction model and achieve a more comprehensive and accurate prediction of gasoline engine exhaust emissions,it used Genetic Algorithm to optimize the BP neural network model.As the result,the CO emission prediction model and HC emission of gasoline engine based on GA-BP neural network were established respectively.Predictive model and NO_X emission prediction model.The comparison shows that the accuracy of the optimized prediction model was significantly improved.Moreover,through the three sets of vehicle state data,the established exhaust emission prediction model was verified and optimized,which proves the reliability and accuracy of the three prediction models.Finally,the established gasoline engine exhaust emission prediction system was applied to the actual vehicle for test verification,and provided effective recommendations for the maintenance and use of the vehicle.