Soot model efficient calibration method for DPF efficiency monitoring

Abstract

The invention relates to a boot model efficient calibration method for DPF efficiency monitoring. The method comprises the following steps that (a) a real vehicle collects boot model input data; (b) carrying out the building of a virtual boot model in the Simulink; performing precision comparison on the virtual soot model and the soot model to ensure that the virtual soot model has enough precision, and obtaining a shaped soot model in Simulink; and (c) performing operation on the boot model input data acquired by the real vehicle in the set Simulink in the boot model, comparing the data with a laboratory measurement value, and ensuring that an error meets a precision requirement. The calibration of the soot model can be completed in only 20 hours, the precision is greatly improved, and the time cost and the labor cost are greatly saved.

Description

Technical Field

[0001] The present invention relates to the technical field of automobiles, and relates to an efficient calibration method for a soot model for DPF efficiency monitoring. Background Art

[0002] The requirements for PM in emission regulations lead to the need for a DPF (Diesel Particulate Filter) in the diesel engine aftertreatment system to reduce PM in the tailpipe emissions, and it is necessary to monitor the DPF efficiency so that when the DPF fails to a certain extent, the control system can detect the fault of low DPF efficiency.

[0003] The current technical route for monitoring DPF efficiency is as follows: when the PM carried in the exhaust gas upstream of the DPF reaches a certain mass (m1), the mass of PM leaking downstream of the DPF (m2) is detected, and the leakage ratio k = m2 / m1. When k is greater than a certain value, it can be considered that there is a fault of low DPF efficiency; among them, m2 can be equivalently measured by a PM sensor, while m1 is calculated and output by the soot model in the ECU. The calibration work of the soot model requires a lot of time and experimental resources.

[0004] The Chinese invention patent with the application number 202210417637.5 discloses a calibration method, device and electronic equipment for DPF filtration efficiency, which mainly relates to the technical field of automobiles. Among them, the method includes: running a preset script based on an integral Map and obtaining running data, and then judging whether the running data meets the preset conditions. If not, continue to run the preset script after adjusting the calibration data in the integral Map until the running data meets the preset conditions, where the calibration data at least includes the DPF filtration efficiency value. By using the running data obtained by running the preset script to calibrate the DPF filtration efficiency in the integral Map, the calibration time can be reduced and the calibration efficiency can be improved; however, the above method of running a preset script based on an integral Map is not applicable to the soot model.

[0005] In addition, since PM is closely related to the level of the lambda signal of the air-fuel mixture in the cylinder during combustion, and the lambda signal is largely controlled by the EGR rate, the existing soot model calibration method calibrates the soot model by collecting PM in the exhaust gas at different EGR rates. This method is very cumbersome and requires collecting all-round data of PM at multiple EGR rates. Moreover, although different EGR rates can be achieved in the laboratory environment, it is impossible to make the total intake air volume at each EGR rate consistent with the total intake air volume during the actual use of the vehicle, and the total intake air volume also has a certain influence on PM in the exhaust gas. This results in that the deviation of the soot model calibrated by the EGR rate can only be controlled within about 30%, and it takes about 120 hours to complete the calibration of the soot model by the EGR rate method, with a long working cycle and being cumbersome. Summary of the Invention

[0006] The object of the present invention is to overcome the defects of the existing calibration method, which has a long and cumbersome working cycle, and to provide an efficient calibration method for the soot model for DPF efficiency monitoring.

[0007] To achieve the above object, the technical solution adopted by the present invention is: an efficient calibration method for the soot model for DPF efficiency monitoring, comprising the following steps: (a) Collecting the input data of the soot model during real vehicle operation; (b) Building a virtual soot model in Simulink; comparing the accuracy of the virtual soot model with the soot model to ensure that the virtual soot model has sufficient accuracy, and obtaining the finalized soot model in Simulink; (c) Performing operations on the input data of the soot model collected during real vehicle operation in the finalized soot model in Simulink, and comparing it with the laboratory measurement values to ensure that the error meets the accuracy requirements.

[0008] Optimally, the input data of the soot model includes engine speed, engine fuel quantity, filtered value of the reciprocal of the lambda signal, reciprocal of the lambda signal, and engine operating mode.

[0009] Further, step (a) is specifically: (a1) Fixing the engine in the Normal mode and using laboratory equipment to collect data of the engine at different speeds and fuel quantities; (a2) Fixing the engine in the RHU mode and using laboratory equipment to collect data of the engine at different speeds and fuel quantities; (a3) Fixing the engine in the Normal mode and using laboratory equipment to collect data of the engine during the WLTC and NEDC cycles; (a4) Fixing the engine in the RHU mode and using laboratory equipment to collect data of the engine during the WLTC and NEDC cycles; (a5) Not fixing the engine mode and using laboratory equipment to collect data of the engine during the WLTC and NEDC cycles.

[0010] Optimally, step (b) is specifically: (b1) According to the description of the soot model in the ECU control software manual, using Matlab / Simulink to build a virtual soot model to reproduce the operation process of the soot model; (b2) After the virtual soot model in Simulink is built, import one set of the input data of the soot model collected from the real vehicle into the Matlab workspace. The virtual soot model in Simulink can calculate and output the soot value based on the imported input quantity. (b3) Compare the error between the soot value output by the virtual soot model in Simulink and the soot value output by the soot model in the ECU. When the error is too large, compare the differences in the operation logics between the two models and correct the soot value output by the virtual soot model in Simulink until the error meets the accuracy requirements, and obtain the finalized soot model in Simulink.

[0011] Further, in step (b3), when the absolute value of the error is greater than 5%, compare the differences in the operation logics between the two models and correct the soot value output by the virtual soot model in Simulink until the absolute value of the error ≤ 5%.

[0012] Optimally, step (c) is specifically as follows: (c1) Perform operations on the input data of the model collected from the real vehicle in the finalized soot model in Simulink. (c2) Compare with the laboratory measurement value and compare the error between the two. When the error is too large, adjust the calibration data of the finalized soot model in Simulink until the error meets the accuracy requirements.

[0013] Further, in step (c2), when the absolute value of the error is greater than 5%, adjust the calibration data of the finalized soot model in Simulink until the error ≤ 5%.

[0014] Due to the application of the above technical solution, the present invention has the following advantages compared with the prior art: The high-efficiency calibration method of the soot model for DPF efficiency monitoring of the present invention can complete the calibration of the soot model in only 20 hours and greatly improve the accuracy, saving a large amount of time cost and labor cost. Description of the Drawings

[0015] Figure 1 It is a flowchart of the high-efficiency calibration method of the soot model for DPF efficiency monitoring of the present invention. Figure 2 It is a schematic diagram of the input data of the model collected from the real vehicle of the present invention (Normal mode). Figure 3 It is a schematic diagram of the input data of the model collected from the real vehicle of the present invention (RHU mode). Figure 4 It is a schematic diagram of the quality output of the soot model of the present invention. Figure 5 Schematic diagram of the quality output of the soot model in Simulink finalized for the present invention; Figure 6 Comparison chart of the calculated values and laboratory measured values of the soot model in Simulink finalized for the present invention. Specific implementation manners

[0016] The present invention will be further described below in conjunction with the embodiments shown in the accompanying drawings.

[0017] As Figure 1 shown, an efficient calibration method for a soot model for DPF efficiency monitoring includes the following steps: (a) Collect input data of the soot model on a real vehicle; the input data of the soot model includes engine speed, engine fuel quantity, lambda signal inverse filtering value, lambda signal inverse, engine operating mode, etc.

[0018] Step (a) is specifically: (a1) Fix the engine in the Normal mode and use laboratory equipment (smoke meter) to collect data of the engine at different speeds and fuel quantities; (a2) Fix the engine in the RHU mode and use laboratory equipment to collect data of the engine at different speeds and fuel quantities; (a3) Fix the engine in the Normal mode and use laboratory equipment to collect data of the engine during the WLTC and NEDC cycles; (a4) Fix the engine in the RHU mode and use laboratory equipment to collect data of the engine during the WLTC and NEDC cycles; (a5) Do not fix the engine mode and use laboratory equipment to collect data of the engine during the WLTC and NEDC cycles.

[0019] In this embodiment, as Figure 2As shown, calibrate dmSotBasNrm_MAP based on the engine speed nEng and the engine fuel quantity qEmi, and output the steady-state soot model mass flow rate dmSotBasNrm_mp in Normal mode; calibrate dmSotLamCorNrm_MAP based on the difference between the filtered value rLamRecRef_mp of the reciprocal of the lambda signal and the reciprocal rLamRecEGS of the lambda signal, and the reciprocal rLamRecEGS of the lambda signal, and output the dynamically corrected soot model mass flow rate dmSotLamCorNrm_mp in Normal mode; add the steady-state soot model mass flow rate dmSotBasNrm_mp in Normal mode to the dynamically corrected soot model mass flow rate dmSotLamCorNrm_mp in Normal mode, output the soot model mass flow rate dmSotEngNrm1 in Normal mode, multiply by the constant 0.001 to convert the unit from mg / s to g / s, perform an integration operation, and output the soot model mass mSotEngNrm1_mp in Normal mode.

[0020] As Figure 3 As shown, calibrate dmSotBasNrm_MAP based on the engine speed nEng and the engine fuel quantity qEmi, and output the steady-state soot model mass flow rate dmSotBasNrm_mp in RHU mode; calibrate dmSotLamCorNrm_MAP based on the difference between the filtered value rLamRecRef_mp of the reciprocal of the lambda signal and the reciprocal rLamRecEGS of the lambda signal, and the reciprocal rLamRecEGS of the lambda signal, and output the dynamically corrected soot model mass flow rate dmSotLamCorNrm_mp in RHU mode; add the steady-state soot model mass flow rate dmSotBasNrm_mp in RHU mode to the dynamically corrected soot model mass flow rate dmSotLamCorNrm_mp in RHU mode, output the soot model mass flow rate dmSotEngNrm1 in RHU mode, multiply by the constant 0.001 to convert the unit from mg / s to g / s, perform an integration operation, and output the soot model mass mSotEngNrm1_mp in RHU mode.

[0021] (b) Build a virtual soot model in Simulink; compare the accuracy of the virtual soot model with the soot model to ensure that the virtual soot model has sufficient accuracy and obtain the finalized soot model in Simulink.

[0022] Step (b) is specifically as follows: (b1) According to the description of the soot model in the ECU control software manual, use Matlab / Simulink to build a virtual soot model to reproduce the operation process of the soot model; (b2) After the virtual soot model in Simulink is built, import a set of input data of the soot model collected from the actual vehicle into the Matlab workspace. The virtual soot model in Simulink can calculate the soot value based on the imported input quantity and output it; (b3) Compare the error between the soot value output by the virtual soot model in Simulink and the soot value output by the soot model in the ECU. When the error is too large, compare the differences in the operation logics between the two models and correct the soot value output by the virtual soot model in Simulink until the error meets the accuracy requirements, and obtain the finalized soot model in Simulink. When the absolute value of the error is greater than 5%, compare the differences in the operation logics between the two models and correct the soot value output by the virtual soot model in Simulink until the absolute value of the error ≤ 5%.

[0023] Specifically, as Figure 4 shown, the engine operating mode stOpModeAct is used as a switch to select the soot model mass flow rate dmSotEngNrm1 in the RHU mode or the soot model mass flow rate dmSotEngNrm1 in the Normal mode to output the soot model mass flow rate dmSotEGSum, multiply it by the constant 0.001 to convert the unit from mg / s to g / s, perform an integration operation, and output the soot model mass mSotEGSum (that is, output the virtual soot model mass mSotEGSum). After the soot model mass flow rate dmSotEGSum passes through the calibration tiPT1SotFlwSim_C filter, it outputs the filtered soot model (that is, the finalized soot model in Simulink) mass flow rate dmSotEGSumFlt.

[0024] (c) Perform operations on the input data of the soot model collected from the actual vehicle in the finalized soot model in Simulink, and compare it with the laboratory measurement values to ensure that the error meets the accuracy requirements.

[0025] Specifically: (c1) Perform operations on the model input data collected from the actual vehicle in the calibrated soot model in Simulink; (c2) Compare with the laboratory measurement values to compare the errors between the two. When the error is too large, adjust the calibration data of the calibrated soot model in Simulink until the error meets the accuracy requirements: when the absolute value of the error is greater than 5%, adjust the calibration data of the calibrated soot model in Simulink until the error ≤ 5%.

[0026] As Figure 5 shown, the filtered soot model mass flow dmSotEGSumFlt can be compared with the soot model mass flow dmSotEngMon_INCA in INCA; the filtered soot model mass flow dmSotEGSumFlt is multiplied by the constant 0.001 to convert the unit from mg / s to g / s, and an integration operation is performed to output the filtered soot model mass mSotEGSumFlt; the laboratory equipment soot mass flow dmSot_483 is multiplied by the constant 0.001 to convert the unit from mg / s to g / s, and an integration operation is performed to output the laboratory equipment soot mass mSot_483; the filtered soot model mass mSotEGSumFlt and the laboratory equipment soot mass mSot_483 can be compared for deviation and percentage deviation. As Figure 6 shown, the accuracy display between the calibrated soot model in Simulink and the laboratory measurement values: the integral value of the calibrated soot model in Simulink is generally 4.4% higher than the laboratory measurement integral value.

[0027] The above-mentioned efficient calibration method for the soot model used for DPF efficiency monitoring can complete the calibration of the soot model in only 20 hours and greatly improve the accuracy, saving a large amount of time cost and labor cost.

[0028] The above embodiments are only used to illustrate the technical concept and characteristics of the present invention, and their purpose is to enable those familiar with this technology to understand the content of the present invention and implement it accordingly, and cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit and essence of the present invention should be covered within the protection scope of the present invention.

Claims

1. An efficient calibration method for a soot model used in DPF efficiency monitoring, characterized in that It includes the following steps: (a)Collect the input data of the soot model from the actual vehicle; (b)Build a virtual soot model in Simulink; compare the accuracy of the virtual soot model with the soot model to ensure that the virtual soot model has sufficient accuracy, and obtain the finalized soot model in Simulink; (c)Operate the input data of the soot model collected from the actual vehicle in the finalized soot model in Simulink, and compare it with the laboratory measurement value to ensure that the error meets the accuracy requirements.

2. The method for efficiently calibrating the soot model for DPF efficiency monitoring according to claim 1, characterized in that: The input data of the soot model includes engine speed, engine fuel quantity, the filtered value of the reciprocal of the lambda signal, the reciprocal of the lambda signal, and the engine operating mode.

3. The method for efficiently calibrating the soot model for DPF efficiency monitoring according to claim 2, wherein Step (a) is specifically as follows: (a1)Fix the engine in the Normal mode and use laboratory equipment to collect data of the engine at different speeds and fuel quantities; (a2)Fix the engine in the RHU mode and use laboratory equipment to collect data of the engine at different speeds and fuel quantities; (a3)Fix the engine in the Normal mode and use laboratory equipment to collect data of the engine during the WLTC and NEDC cycles; (a4)Fix the engine in the RHU mode and use laboratory equipment to collect data of the engine during the WLTC and NEDC cycles; (a5)Do not fix the engine mode and use laboratory equipment to collect data of the engine during the WLTC and NEDC cycles.

4. The method for efficiently calibrating a soot model for DPF efficiency monitoring according to claim 1, characterized in that, Step (b) is specifically as follows: (b1)According to the description of the soot model in the ECU control software manual, use Matlab / Simulink to build a virtual soot model to reproduce the operation process of the soot model; (b2)After the virtual soot model in Simulink is built, import a set of input data of the soot model collected from the actual vehicle into the Matlab workspace. The virtual soot model in Simulink can calculate and output the soot value according to the imported input quantity; (b3)Compare the error between the soot value output by the virtual soot model in Simulink and the soot value output by the soot model in the ECU; when the error is too large, compare the differences in the operation logics between the two models and correct the soot value output by the virtual soot model in Simulink until the error meets the accuracy requirements, and obtain the finalized soot model in Simulink.

5. The method for efficiently calibrating the soot model for DPF efficiency monitoring according to claim 4, characterized in that: In step (b3), when the absolute value of the error is greater than 5%, compare the differences in the operation logics between the two models and correct the soot value output by the virtual soot model in Simulink until the absolute value of the error ≤ 5%.

6. The method for efficiently calibrating a soot model for DPF efficiency monitoring according to claim 1, wherein Step (c) is specifically as follows: (c1)Operate the input data of the model collected from the actual vehicle in the finalized soot model in Simulink; (c2) Compare it with the laboratory measurement value and compare the errors between the two; when the error is too large, adjust the calibration data of the finalized soot model in Simulink until the error meets the accuracy requirements.

7. The method for efficiently calibrating the soot model for DPF efficiency monitoring according to claim 6, wherein: In step (c2), when the absolute value of the error is greater than 5%, adjust the calibration data of the soot model in the calibrated Simulink until the error ≤ 5%.

Images