Research On Soot Load Estimation Of Diesel DPF Based On Model

The rapid development of China’s automobile industry brings the rapid increase in car ownership,but the automobile exhaust inevitably causes air pollution.In which,particulate matter（PM）is one of the main pollutants in diesel engine exhaust,and it is easy to cause great harm to the air and human health.Diesel Particulate Filter（DPF）is the most effective mean for reducing exhaust particulates.When the PM deposited in the DPF reaches a certain amount,it needs to be regenerated.The key technologies of the active regeneration method are the soot load estimation of the DPF and the regeneration temperature control.With the data of the DPF bench test,the relationship between the pressure drop across the DPF,the exhaust volume flow,and the soot load under steady-state operations is analyzed,and the soot load characteristics are obtained.In the WHTC operation,the pressure drop across the DPF exhibits a fast fluctuating state,and the higher the carbon load,the greater the pressure drop of the DPF.Based on the similarities of gas flow and current,as well as the theoretical analysis of the fluid dynamics of the exhaust gas,a DPF equivalent circuit model is established,which is also established in MATLAB/Simulink software.Based on the test data of WHTC operation,the parameter identification is performed with the least squares method.The relationship between the parameter and the soot load is obtained,and the results of soot load which vary with time are obtained under the WHTC operation.Finally,the simulation value of DPF pressure drop is compared with the experimental value,which verifies the accuracy of the model.Double extended Kalman filter is used for soot load estimation under WHTC operation.With the pressure drop across the DPF,the exhaust volume flow,and the exhaust temperature measured by the sensors,the state and parameter of the system can be estimated simultaneously,and then the soot load value can be obtained based on the relationship between the parameter and the soot load.When the simulation initial value is accurate,the error is 0.2g/L,or 5%,relative to the test value when the regeneration timing of 4 g/L is reached.When the initial value of the simulation is inaccurate,the convergence time required for low and medium soot load is about 500s,and the convergence time for high soot load is about 900s.Dual H infinity filter is also used for soot load estimation under WHTC operation.The biggest difference between this method and the dual extended Kalman filter is that it does not require high accuracy of the model,does not require the statistical characteristics of the noise known,and can minimize the maximum error.These characteristics make the dual H infinity filter more advantageous.When the simulation initial value is accurate,the error is 0.16 g/L,or 4%,relative to the test value when the regeneration timing of 4 g/L is reached.In the case of inaccurate simulation initial value,the convergence speed is faster than dual extended Kalman filter at low soot load,and the convergence time required for medium and high soot load is comparable to double Kalman filter.Overall it shows a higher degree of robustness.When the soot load in the DPF reaches a certain value,it needs to be regenerated,and the active regeneration requires the exhaust gas temperature to reach the particulate fire point.In this paper,the DPF active regenerative light-off temperature is controlled for the DOC system.The feedback linearization control is used to control the fuel injection concentration at the DOC inlet so that the DOC outlet exhaust temperature can reach the fire point 650°C.The spray injection concentration at DOC inlet should be controlled at 1950 to 2000ppm at the operation in which the DOC inlet temperature is 250°C and the airspeed is 60,000 h^-1,causing no secondary pollution.