Engine exhaust gas recirculation flow control method
By using multiple linear regression and neural network models in the EGR system, combined with data acquisition and fault monitoring, precise control of EGR flow and opening is achieved, solving the problem of inaccurate control in traditional methods and improving the performance and reliability of the engine.
Patent Information
- Application Number
- CN202510717804.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-30
- Publication Date
- 2025-07-22
AI Technical Summary
The traditional EGR flow control method is based on simple rules or empirical formulas, and lacks precision and intelligence, resulting in insufficient accuracy of EGR control and diagnostic strategies.
A multivariate linear regression model based on engine intake volume, temperature, intake manifold pressure and exhaust pressure is adopted, combined with the neural network model, the EGR flow and opening degree is controlled and monitored. Through data acquisition, model construction and verification, precise control of flow and opening degree is achieved, and alarm is issued in the event of a fault.
It improves the accuracy and intelligence of EGR flow control, optimizes the combustion efficiency of the engine, reduces pollutant emissions, reduces maintenance costs and downtime, and simplifies hardware costs.
Smart Images
Figure CN120351069A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of engine exhaust gas recirculation (EGR), and more particularly, to a method for controlling the flow rate of engine exhaust gas recirculation. Background Art
[0002] In engine technology, the EGR system plays a key role in reducing nitrogen oxide emissions and improving combustion efficiency. Traditional EGR flow control methods often rely on simple rules or empirical formulas, and EGR control and diagnostic strategies need to be further refined and intelligentized. Summary of the Invention
[0003] The main technical problem to be solved by this application is how to optimize the control of engine exhaust gas recirculation.
[0004] To solve the above technical problem, on the one hand, this application provides a method for controlling the flow rate of engine exhaust gas recirculation, which includes the following steps:
[0005] Run the engine exhaust gas recirculation flow rate and opening control model;
[0006] Run the correction algorithm for the engine exhaust gas recirculation flow rate and opening control model to correct the engine exhaust gas recirculation flow rate and opening;
[0007] Run the fault monitoring algorithm for the engine exhaust gas recirculation system to monitor the faults of the engine exhaust gas recirculation system.
[0008] According to the method for controlling the flow rate of engine exhaust gas recirculation proposed in one aspect of this application, in the engine exhaust gas recirculation flow rate and opening control model, the required engine exhaust gas recirculation flow rate is determined based on the engine intake air volume and temperature, and the engine exhaust gas recirculation valve opening is calculated based on the intake manifold pressure and the exhaust pressure.
[0009] According to the method for controlling the flow rate of engine exhaust gas recirculation proposed in one aspect of this application, the process of establishing the engine exhaust gas recirculation flow rate and opening control model includes the following steps: data acquisition, model construction, and model verification.
[0010] According to the method for controlling the flow rate of engine exhaust gas recirculation proposed in one aspect of this application,
[0011] In the data acquisition step, the engine intake air volume, temperature, intake manifold pressure, exhaust pressure, and the corresponding engine exhaust gas recirculation flow rate and opening data are tested and recorded under different working conditions;
[0012] In the model construction step, according to the previously recorded data, mathematical modeling is used to construct the engine exhaust gas recirculation flow rate and opening control model under different working conditions;
[0013] In the model verification step, the constructed engine exhaust gas recirculation flow rate and opening control model is verified using test data.
[0014] According to the engine exhaust gas recirculation flow rate control method proposed in one aspect of the present application, in the model construction step, a multiple linear regression model is established, with the intake air volume, temperature, intake manifold pressure, and exhaust pressure as independent variables, and the engine exhaust gas recirculation flow rate and opening as dependent variables, and the model parameters are determined by the least squares method.
[0015] According to the engine exhaust gas recirculation flow rate control method proposed in one aspect of the present application, the process of establishing the correction algorithm for the engine exhaust gas recirculation flow rate and opening control model includes the following steps:
[0016] Construct a neural network model, in which, determine the input variables and output variables of the neural network model;
[0017] Train the neural network model, in which, simulate engine exhaust gas recirculation system failures using engine exhaust gas recirculation pipes with different diameters, test under different working conditions to obtain training data, and train the neural network model based on the training data.
[0018] According to the engine exhaust gas recirculation flow rate control method proposed in one aspect of the present application, the input variables include engine knock signals, crankshaft angular acceleration signals, intake air volume, temperature, intake manifold pressure, and exhaust pressure, and the output variables include correction values of the engine exhaust gas recirculation flow rate and opening.
[0019] According to the engine exhaust gas recirculation flow rate control method proposed in one aspect of the present application, in the engine exhaust gas recirculation system fault monitoring algorithm, monitor the correction amplitude of correcting the engine exhaust gas recirculation flow rate and opening using the correction algorithm for the engine exhaust gas recirculation flow rate and opening control model.
[0020] According to the engine exhaust gas recirculation flow rate control method proposed in one aspect of the present application, when the correction amplitude exceeds the threshold, it is determined that there is a blockage in the engine exhaust gas recirculation pipe, and an alarm is issued.
[0021] According to the engine exhaust gas recirculation flow rate control method proposed in one aspect of the present application, calculate the average value of the correction amplitude over a period of time, and when the average value exceeds the threshold, an alarm is issued.
[0022] According to the technical solution of the present application, the designed engine exhaust gas recirculation flow rate control method makes the EGR control and diagnosis strategy more precise and intelligent, thereby optimizing the EGR control. Description of the Drawings
[0023] The disclosure of the present application will be described with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of the present application. Among them:
[0024] Figure 1 Schematically shows the schematic diagram of the engine exhaust gas recirculation flow control method proposed according to an embodiment of the present application. Specific embodiments
[0025] The embodiments of the present application will be described in detail below with reference to the accompanying drawings. Those skilled in the art will appreciate that these descriptions are only descriptive and exemplary and should not be construed as limiting the scope of protection of the present application.
[0026] The engine exhaust gas recirculation flow control method according to a specific embodiment of the present application mainly includes the following three aspects:
[0027] Basic EGR flow-opening model, in which, based on bench data, the EGR flow and opening control models under different working conditions are constructed. Its basic principle is summarized as: determining the required EGR flow based on parameters such as engine intake air volume and temperature, and calculating the EGR valve opening based on parameters such as intake manifold pressure and exhaust pressure.
[0028] Flow-opening model correction algorithm, in which, a neural network model is introduced, and based on the training data under different blockage conditions of the EGR pipeline, an EGR flow-opening correction algorithm based on engine knock and crankshaft angular acceleration signals is designed.
[0029] EGR system fault monitoring algorithm, in which, continuously monitor the correction of the EGR flow-opening model predicted by the neural network. When the correction amplitude of this model is excessive, the blockage condition existing in the EGR pipeline can be judged. At the same time, based on the actual combustion stability of the engine, if the blockage seriously affects the emission performance, the system selects to report an EGR system fault for repair and maintenance.
[0030] Refer to Figure 1 As shown, according to an embodiment of the present application, the engine exhaust gas recirculation flow control method includes: implementation of the basic EGR flow-opening model, implementation of the flow-opening model correction algorithm, and implementation of the EGR system fault monitoring algorithm.
[0031] The implementation of the basic EGR flow-opening model includes data collection, model construction, and model verification.
[0032] During the data collection process: install an EGR flow sensor on the engine bench, and conduct tests under different working conditions, record parameters such as engine intake air volume, temperature, intake manifold pressure, exhaust pressure, etc., as well as the corresponding EGR flow and opening data.
[0033] During the model construction process: Based on the collected data, mathematical modeling methods are used to construct EGR flow rate and opening control models under different working conditions. For example, a multiple linear regression model is established, with parameters such as intake air volume, temperature, intake manifold pressure, and exhaust pressure as independent variables, and EGR flow rate and opening as dependent variables. The model parameters are determined by the least squares method.
[0034] During the model verification process: New test data is used to verify the constructed model to ensure the accuracy and reliability of the model. The errors between the model prediction values and the actual measurement values can be calculated, such as the root mean square error and the mean absolute error, to evaluate the performance of the model.
[0035] The implementation of the flow rate - opening model correction algorithm includes the construction of a neural network model and the training of the neural network model.
[0036] During the construction process of the neural network model: A suitable neural network structure is selected, such as a feed - forward neural network. The input and output variables of the neural network are determined. The input variables include engine knock signal, crankshaft angular acceleration signal, intake air volume, temperature, intake manifold pressure, exhaust pressure, etc., and the output variables are the corrected values of EGR flow rate and opening.
[0037] During the training process of the neural network model: On the engine test bench, EGR system failures are simulated using EGR pipes with different diameters, and tests are carried out under multiple working conditions to obtain a large amount of training data. Based on these training data, the neural network is trained.
[0038] When engine knock occurs, it indicates that the engine combustion intensity is relatively large, which may mean that the EGR flow rate is too low. At this time, the neural network model can adjust the EGR flow rate and opening control model according to the knock signal, and appropriately increase the opening of the EGR valve to reduce the combustion intensity and improve the combustion stability.
[0039] When the engine angular acceleration is too low, it represents that the engine combustion deteriorates, which may be caused by too high an EGR flow rate. In this case, the neural network model can adjust the EGR flow rate and opening control model according to the angular acceleration signal, and appropriately reduce the EGR flow rate to improve the combustion condition.
[0040] Knock and crankshaft speed are important parameters reflecting the engine combustion state. Based on these two signals, a correction algorithm for the EGR flow rate - opening model can be designed and applied to the control of the EGR valve and the fault monitoring of the EGR system.
[0041] The implementation of the EGR system fault monitoring algorithm includes model correction monitoring, fault diagnosis, and maintenance.
[0042] During the model correction monitoring process: Continuously monitor the correction of the EGR flow - opening model predicted by the neural network. A correction amplitude threshold can be set. When the model correction amplitude exceeds this threshold, it is determined that there is a blockage in the EGR pipeline. For example, record the amplitude of each model correction and calculate the average correction amplitude over a period of time. If the average correction amplitude exceeds the threshold, a blockage alarm is issued.
[0043] During the fault diagnosis and repair process: When the system reports an EGR system fault, perform fault diagnosis. A fault diagnostic instrument can be used to read the fault codes in the engine control unit to determine the specific location and cause of the fault. Perform repairs and maintenance according to the fault diagnosis results, such as cleaning or replacing components such as the EGR valve and EGR pipeline.
[0044] The EGR flow control method according to this embodiment has the following beneficial effects: It improves the accuracy of EGR flow control, can better meet the requirements of the engine under different working conditions, improves combustion efficiency, enhances the performance and reliability of the engine, and reduces pollutant emissions; By introducing a neural network model to monitor and analyze the combustion situation in real time, it can timely adjust the EGR flow and opening control model, improving the performance and reliability of the engine; It can timely detect blockages in the EGR pipeline and system faults, ensuring the normal operation of the engine, reducing maintenance costs and downtime; There is no need to install an EGR flow sensor, saving hardware costs and reducing system complexity.
Claims
1. A method for controlling the exhaust gas recirculation flow rate of an engine, characterized in that, The engine exhaust gas recirculation flow control method includes the following steps: Run the engine exhaust gas recirculation flow and opening control model; Run the correction algorithm for the engine exhaust gas recirculation flow and opening control model to correct the engine exhaust gas recirculation flow and opening; Run the engine exhaust gas recirculation system fault monitoring algorithm to monitor the faults of the engine exhaust gas recirculation system.
2. The engine exhaust gas recirculation flow control method according to claim 1, wherein In the engine exhaust gas recirculation flow and opening control model, the required engine exhaust gas recirculation flow is determined based on the engine intake air volume and temperature, and the engine exhaust gas recirculation valve opening is calculated based on the intake manifold pressure and the exhaust pressure.
3. The method for controlling the exhaust gas recirculation flow rate of an engine according to claim 1, wherein The process of establishing the engine exhaust gas recirculation flow and opening control model includes the following steps: data acquisition, model construction, and model verification.
4. The engine exhaust gas recirculation flow control method according to claim 3, wherein In the data acquisition step, the engine intake air volume, temperature, intake manifold pressure, exhaust pressure, and the corresponding engine exhaust gas recirculation flow and opening data are tested and recorded under different working conditions; In the model construction step, according to the previously recorded data, mathematical modeling is used to construct the engine exhaust gas recirculation flow and opening control model under different working conditions; In the model verification step, the constructed engine exhaust gas recirculation flow and opening control model is verified using the test data.
5. The method for controlling the exhaust gas recirculation flow rate of an engine according to claim 4, characterized in that, In the model construction step, a multiple linear regression model is established, with the intake air volume, temperature, intake manifold pressure, and exhaust pressure as independent variables, and the engine exhaust gas recirculation flow and opening as dependent variables, and the model parameters are determined by the least squares method.
6. The method for controlling the exhaust gas recirculation flow rate of an engine according to claim 1, wherein, The process of establishing the correction algorithm for the engine exhaust gas recirculation flow and opening control model includes the following steps: Construct a neural network model, in which the input variables and output variables of the neural network model are determined; Train the neural network model, in which, the faults of the engine exhaust gas recirculation system are simulated using engine exhaust gas recirculation pipes with different diameters, and tests are carried out under different working conditions to obtain training data, and the neural network model is trained based on the training data.
7. The engine exhaust gas recirculation flow control method according to claim 6, characterized in that, The input variables include the engine knock signal, crankshaft angular acceleration signal, intake air volume, temperature, intake manifold pressure, and exhaust pressure, and the output variables include the correction values of the engine exhaust gas recirculation flow and opening.
8. The method for controlling the exhaust gas recirculation flow rate of an engine according to claim 1, characterized in that, In the engine exhaust gas recirculation system fault monitoring algorithm, monitor the correction amplitude of the correction of the engine exhaust gas recirculation flow and opening using the correction algorithm for the engine exhaust gas recirculation flow and opening control model.
9. The engine exhaust gas recirculation flow control method according to claim 8, wherein When the correction amplitude exceeds the threshold, it is determined that there is a blockage in the engine exhaust gas recirculation pipe, and an alarm is issued.
10. The engine exhaust gas recirculation flow control method according to claim 9, wherein Calculate the average value of the correction amplitude over a period of time, and when the average value exceeds the threshold, an alarm is issued.