Engine cylinder pressure reconstruction method based on multi-physical field dynamic coupling mechanism

Abstract

The application discloses an engine in-cylinder pressure reconstruction method based on a multi-physical field dynamic coupling mechanism, and comprises the following steps: S1, a signal acquisition module acquires one or more of the following parameters of an engine: intake pressure, intake temperature, cooling water temperature and an engine electronic control unit (ECU) signal; S2, a complete injection rate-time curve is calculated through a constructed polynomial fitting model; S3, transient heat loss between engine in-cylinder working medium and a combustion chamber wall is calculated; S4, cumulative heat release including pre-injection combustion heat release rate and main-injection combustion heat release rate is calculated; and S5, an in-cylinder pressure time sequence of the engine in a complete working cycle is iteratively reconstructed. The application balances among precision, cost and reliability, and provides a feasible solution for intelligent control of the engine.

Description

Technical Field

[0001] This invention relates to the field of engine cylinder pressure reconstruction, and in particular to an engine cylinder pressure reconstruction method based on a multi-physics dynamic coupling mechanism. Background Technology

[0002] In the field of intelligent engine control technology, in-cylinder pressure is the most critical state parameter characterizing the combustion process, directly determining the engine's power, economy, noise level, and emissions performance. Achieving high-precision, low-cost, real-time online monitoring of in-cylinder pressure is of paramount and irreplaceable significance for implementing advanced combustion closed-loop control, realizing engine fault diagnosis and health management, and optimizing its overall operating efficiency.

[0003] However, directly mounting cylinder pressure sensors on existing mass-produced engines presents several challenges. Cylinder pressure sensors are inherently expensive, hindering large-scale commercial applications. Furthermore, the sensors require long-term operation in the extremely harsh high-temperature, high-pressure, and high-frequency mechanical shock environments within the engine cylinder, posing a significant challenge to their long-term reliability and durability. In addition, integrating and mounting the sensors on mass-produced engines, reliably packaging the signal leads, and ensuring measurement stability throughout their entire lifecycle all present substantial engineering challenges.

[0004] Therefore, indirectly "reconstructing" or predicting cylinder pressure through other easily measurable and acquireable signals from the engine has become a hot topic and mainstream alternative technology route that academia and industry have long been committed to solving. Currently, mainstream cylinder pressure reconstruction technologies are mainly divided into three categories based on modeling methodologies: "white-box models" built on rigorous first principles of physicochemistry, "black-box models" that rely entirely on data-driven approaches, and "gray-box models" that attempt to integrate physical mechanisms and data characteristics. However, these existing technologies all have significant technical bottlenecks when applied to complex and dynamically changing engine systems. Although white-box models have clear physical meaning, they are usually complex, computationally burdensome, and heavily reliant on a large number of boundary conditions and physical property parameters that are difficult to obtain accurately, making it difficult to meet the needs of real-time vehicle control. Black-box models may perform well under conditions where training data covers the operating conditions, but their internal reasoning process lacks physical interpretability and is essentially a "black box" operation. They have poor generalization ability under sparse or unseen operating conditions, low reliability of prediction results, and high risks in engineering applications. Existing gray box models, such as those that introduce additional knock sensors to measure vibration noise signals to aid reconstruction, or those that employ sensor fusion and filtering algorithms, while improving accuracy, inevitably introduce additional sensor costs and bring new challenges to the reliability of these sensors in harsh environments.

[0005] In summary, existing in-cylinder pressure reconstruction technologies, when applied to actual engine systems, struggle to achieve a good balance across multiple key performance dimensions, including reconstruction accuracy, model generalization and robustness, computational real-time performance, engineering implementation cost, and model physical interpretability. Consequently, they fail to meet the urgent needs of next-generation advanced engine intelligent control systems that require high precision, high reliability, and engineering applicability.

[0006] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this invention is to provide a method for reconstructing engine cylinder pressure based on the dynamic coupling mechanism of multiphysics fields, thereby solving the technical problems existing in the background art.

[0008] To achieve the above objectives, this invention provides a method for reconstructing engine cylinder pressure based on a multi-physics dynamic coupling mechanism, comprising: S1: a signal acquisition module acquires one or more of the following engine parameters: intake pressure, intake temperature, coolant temperature, and engine electronic control unit (ECU) signals; S2: an injection rate prediction module, based on the injection pressure and injection pulse width data contained in the ECU signals and combined with the known structural parameters of the injector, calculates the complete injection rate variation curve over time using a constructed polynomial fitting model. The injection rate prediction module divides the needle valve action stage data into multiple sub-models, including but not limited to: needle valve fast opening stage, needle valve slow opening stage, needle valve full opening stage, needle valve fluctuation stage, needle valve slow closing stage, and needle valve fast closing stage. The rate function for each stage includes... Specific parameters calibrated from experimental data; S3: The heat transfer prediction module uses the Hohenberg model to calculate the instantaneous heat transfer loss between the working fluid in the engine cylinder and the combustion chamber wall; S4: The cumulative heat release prediction module receives the injection rate curve output by the injection rate prediction module and calculates the cumulative heat release, including the pre-injection combustion heat release rate and the main injection combustion heat release rate, based on the concept of cumulative fuel mass in the cylinder; S5: The core reconstruction module takes the cumulative heat release output by the cumulative heat release prediction module, the heat transfer output by the heat transfer prediction module, the real-time cylinder volume and its rate of change calculated by the crankshaft angle sensor, and the initial pressure and temperature conditions as input, executes the simplified formula calculation of the first law of thermodynamics, and iteratively reconstructs the in-cylinder pressure time series of the engine in one complete working cycle.

[0009] In one embodiment of the present invention, step S2 includes: S21: establishing piecewise polynomial functions for the six stages based on the action characteristics of the needle valve, wherein the injection rate and injection pressure during the slow-opening stage of the needle valve are correlated through a fitting polynomial; S22: multiple calibration parameters involved in the sub-model of the injection rate module are determined by solving them together with the physical correlation formula between the parameters and the injection pressure or with experimental data, ensuring the model's adaptability to injector differences and different operating conditions; wherein the input time unit of the fitting polynomial is milliseconds, the output injection rate unit is milligrams per millisecond, and each stage function is a continuously differentiable function with respect to time, thereby ensuring the continuity, smoothness, and physical rationality of the calculated injection rate curve in time, providing high-fidelity input boundary conditions for subsequent combustion heat release calculations.

[0010] In one embodiment of the present invention, the fitting formula includes:

[0011] ;

[0012] ;

[0013] ;

[0014] ;

[0015] ;

[0016] ;

[0017] ;

[0018] In the above fitting formula, the time unit is ms, and the injection rate unit is mg / ms; IRt(1), IR t (2) IR t (3) IR t (4) IR t (5) and IR t (6) Represents the injection rates corresponding to the needle valve fast opening stage t1, needle valve slow opening stage t2, needle valve full opening stage t3, needle valve fluctuation stage t4, needle valve slow closing stage t5 and needle valve fast closing stage t6 respectively; Parameters i, j and k are calibrated parameters that can be calibrated based on test data. They are obtained by fitting the needle valve slow opening stage with different injection pressures. The values ​​of i2 and j2 are -0.53 and 1.64 respectively.

[0019] In one embodiment of the present invention, a derivation formula is also included for simultaneously solving the data of the fitting formula:

[0020] The derived formula includes:

[0021] (1)

[0022] (2)

[0023] (3)

[0024] (4)

[0025] Combining the aforementioned fitting formula, where the value of k1 is related to the change in injection pressure, the derived formula (1) is solved simultaneously to obtain the value of k1; through IR... t (5) and IR t (6) Perform a joint solution to obtain the values ​​of i6 and k6; IR t The parameter in (1) needs to be obtained from the value of k1 according to equation (4), and then the IR needs to be solved. t1 (1) = IR t1 (2) and IR t1 (1)'=IR t1 (2) Obtain the values ​​of i1 and j1 from the two equations. ； The value of Z is determined based on the peak injection rate.

[0026] In one embodiment of the present invention, the core calculation model of the cumulative heat release prediction module is based on physical derivation, and step S4 includes:

[0027] S41: Calculate the pre-injection heat release rate using a formula that includes combustion efficiency calibration parameters and pre-injection delay period;

[0028] S42: The heat release rate of the main injection is calculated by the main injection calibration parameters and combustion efficiency parameters, and is superimposed with the heat release rate of the residual fuel quantity.

[0029] S43: The combustion efficiency calibration parameters are related to the intake air temperature, intake air pressure and intake air composition, realizing the physical response of the combustion process to the real-time state inside the cylinder;

[0030] S44: The model's output is a cumulative heat release curve that changes over time, serving as the core energy input for calculating changes in cylinder pressure.

[0031] In one embodiment of the present invention, before performing the reconstruction calculation, the contribution analysis of the complete first law of thermodynamics is first performed, and the heat release rate curves output by the model are compared graphically with all terms included, the blow-by loss term ignored, the latent heat of fuel vaporization term ignored, and both terms ignored.

[0032] Secondly, the present invention provides an engine cylinder pressure reconstruction system based on a multiphysics dynamic coupling mechanism, characterized in that it includes a processor and a memory, wherein the memory stores computer-executable instructions, and when the instructions are executed by the processor, the engine cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism is realized.

[0033] Thirdly, a computer-readable storage medium storing a computer program thereon, characterized in that, when the program is executed by a computing device, it implements the engine cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism.

[0034] Compared with the prior art, the engine cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism of the present invention achieves the following effects:

[0035] 1. High-precision reconstruction: By modeling physical mechanisms and correlating dynamic parameters, it overcomes the generalization defects of data-driven models and achieves accurate prediction of cylinder pressure in a wide range of operating conditions (such as high altitude, high temperature, and high EGR), with significantly lower error than traditional methods.

[0036] 2. Real-time performance and low cost: Reuse existing sensors to avoid the installation of expensive cylinder pressure sensors; simplified models and staged calculations significantly reduce the computational load on the ECU, meeting the real-time control requirements of the vehicle.

[0037] 3. Engineering Deployability: The modular design facilitates integration into mass-produced ECUs, reduces dependence on calibration parameters, improves model robustness, and provides a reliable foundation for advanced functions such as combustion closed-loop control and fault diagnosis. Attached Figure Description

[0038] Appendix Figure 1 The fuel injection rate model verification results of the engine in-cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism provided in this application;

[0039] Appendix Figure 2 The CA50 model (AFM Model) verification results of the engine in-cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism provided in this application;

[0040] Appendix Figure 3 is This application provides a comparison of different terms of an engine in-cylinder pressure reconstruction method based on a multiphysics dynamic coupling mechanism;

[0041] Appendix Figure 4 The cylinder pressure reconstruction model verification results of the engine cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism provided in this application;

[0042] Appendix Figure 5A schematic diagram of the model for an engine cylinder pressure reconstruction method based on the dynamic coupling mechanism of multiphysics fields, provided in this application. Detailed Implementation

[0043] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments.

[0044] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0045] As shown in the figure, this invention provides a method for reconstructing engine in-cylinder pressure based on a multi-physics dynamic coupling mechanism, including:

[0046] S1: The signal acquisition module acquires one or more of the following engine parameters: intake pressure, intake temperature, coolant temperature, and engine electronic control unit (ECU) signals;

[0047] S2: The injection rate prediction module, based on the injection pressure and injection pulse width data contained in the ECU signal and combined with the known structural parameters of the injector, calculates the complete injection rate change curve over time through a constructed polynomial fitting model. The injection rate prediction module divides the needle valve action stage data into multiple sub-models, including but not limited to: needle valve fast opening stage, needle valve slow opening stage, needle valve full opening stage, needle valve fluctuation stage, needle valve slow closing stage, and needle valve fast closing stage. The rate function of each stage contains specific parameters calibrated from experimental data.

[0048] S3: The heat transfer prediction module uses the Hohenberg model to calculate the instantaneous heat transfer loss between the working fluid in the engine cylinder and the combustion chamber wall.

[0049] S4: The cumulative heat release prediction module receives the injection rate curve output by the injection rate prediction module and calculates the cumulative heat release, including the pre-injection combustion heat release rate and the main injection combustion heat release rate, based on the concept of in-cylinder cumulative fuel mass.

[0050] S5: The core reconstruction module takes the cumulative heat output by the cumulative heat release prediction module, the heat transfer output by the heat transfer prediction module, the real-time cylinder volume and its rate of change calculated by the crankshaft angle sensor, and the initial pressure and temperature conditions as input, and performs simplified formula calculations based on the first law of thermodynamics to iteratively reconstruct the cylinder pressure time series of the engine in a complete working cycle.

[0051] Specifically, this implementation defines the basic framework and input-output relationships of the entire reconstruction model, aiming to construct a closed-loop computation process. Its specific implementation path is as follows:

[0052] 1. Signal Input: The system reads the values ​​from the intake manifold absolute pressure sensor and intake air temperature sensor in real time via the standard interface of the engine electronic control unit. Simultaneously, it collects data from the engine coolant temperature sensor. All of these sensors are standard equipment in modern electronically controlled engines.

[0053] 2. ECU Signal Analysis: The injection pressure signal and injection pulse width signal for each injection event are read from within the ECU, which can also be directly provided by the ECU software. To calculate crankshaft angle or time, signals from the crankshaft position sensor also need to be acquired simultaneously.

[0054] 3. Injection Rate Calculation: The injector structural parameters are pre-existing as known constants in the model. The injection pressure and injection pulse width under the current operating conditions are input into the piecewise polynomial model to calculate the injection rate waveform covering the entire life cycle of the injection event.

[0055] The model constructed in this implementation method realizes a "soft sensing" technology, and its effect logic chain is clear:

[0056] Cost and Reliability Improvements: The core inputs of the model utilize sensors already standard on mass-produced engines and signals inherent within the ECU, completely eliminating the need for any form of in-cylinder pressure sensor or dedicated knock sensor. This not only avoids high direct costs but also thoroughly solves the reliability and durability challenges of sensors caused by extreme in-cylinder environments.

[0057] High precision and generalization driven by physical mechanisms: The model's computational path strictly follows the physical essence of in-cylinder energy conservation. Compared with traditional data-driven black-box or shallow gray-box models, this first-principles-based modeling strategy ensures that when the model is applied to new operating conditions not covered by the training data, its prediction results still have a solid physical basis and will not exhibit non-physical biases due to the failure of the data-driven mode. This achieves a qualitative leap from "data fitting" to "mechanism prediction," significantly improving the extrapolation accuracy and robustness of the model under sparse and unseen operating conditions.

[0058] Providing core observations for closed-loop control: The pressure curve output by the model in real time includes key combustion characteristic parameters such as peak combustion pressure (Pmax), average indicated pressure (IMEP), and pressure rise rate (dp / dθ). These parameters are direct feedback quantities for implementing advanced combustion closed-loop control. The controller can directly and accurately track the setpoints of these combustion targets by adjusting injection timing (SOI), injection quantity, and exhaust gas recirculation (EGR) rate, thereby achieving optimized combustion efficiency and pollutant control, and constructing a fully closed-loop link from combustion state perception to intelligent actuator adjustment.

[0059] Specifically, this implementation ensures the fidelity of the input source. The injection rate is the "fuel mass source term" of the combustion heat release rate and is one of the most critical inputs to the energy conservation equation. Traditional simplified models introduce significant errors. This implementation, through a finely piecewise fitting that closely matches the actual movement of the needle valve, restores the actual fuel supply dynamics to the greatest extent possible, providing high-fidelity boundary conditions for subsequent combustion calculations. Incorrect injection rate waveforms will cause phase and shape distortion of the heat release rate curve, ultimately directly leading to cylinder pressure reconstruction errors.

[0060] As a preferred embodiment of the present invention, step S2 includes: S21: establishing piecewise polynomial functions for the six stages based on the action characteristics of the needle valve, wherein the injection rate and injection pressure during the slow-opening stage of the needle valve are correlated through a fitting polynomial; S22: multiple calibration parameters involved in the sub-model of the injection rate module are determined by jointly solving them with the physical correlation formula with the injection pressure or with experimental data, ensuring the model's adaptability to injector differences and different operating conditions; wherein the input time unit of the fitting polynomial is milliseconds, the output injection rate unit is milligrams per millisecond, and each stage function is a continuously differentiable function with respect to time, thereby ensuring the continuity, smoothness, and physical rationality of the calculated injection rate curve in time, providing high-fidelity input boundary conditions for subsequent combustion heat release calculations.

[0061] As a preferred embodiment, the fitting formula includes:

[0062] ;

[0063] ;

[0064] ;

[0065] ;

[0066] ;

[0067] ;

[0068] ;

[0069] In the above fitting formula, the time unit is ms, and the injection rate unit is mg / ms; IRt(1), IR t (2) IR t (3) IR t (4) IR t (5) and IR t (6) Represents the injection rates corresponding to the needle valve fast opening stage t1, needle valve slow opening stage t2, needle valve full opening stage t3, needle valve fluctuation stage t4, needle valve slow closing stage t5 and needle valve fast closing stage t6 respectively; Parameters i, j and k are calibrated parameters that can be calibrated based on test data. They are obtained by fitting the needle valve slow opening stage with different injection pressures. The values ​​of i2 and j2 are -0.53 and 1.64 respectively.

[0070] As a preferred option, it also includes a derivation formula for simultaneously solving the data of the fitting formula:

[0071] The derived formula includes:

[0072] (1)

[0073] (2)

[0074] (3)

[0075] (4)

[0076] Combining the aforementioned fitting formula, where the value of k1 is related to the change in injection pressure, the derived formula (1) is solved simultaneously to obtain the value of k1; through IR... t (5) and IR t (6) Perform a joint solution to obtain the values ​​of i6 and k6; IR t The parameter in (1) needs to be obtained from the value of k1 according to equation (4), and then the IR needs to be solved. t1 (1) = IR t1 (2) and IR t1 (1)'=IR t1 (2) Obtain the values ​​of i1 and j1 from the two equations. ； The value of Z is determined based on the peak injection rate.

[0077] As a preferred embodiment, the core calculation model of the cumulative heat release prediction module is based on physical derivation, and step S4 includes:

[0078] S41: Calculate the pre-injection heat release rate using a formula that includes combustion efficiency calibration parameters and pre-injection delay period;

[0079] S42: The heat release rate of the main injection is calculated by the main injection calibration parameters and combustion efficiency parameters, and is superimposed with the heat release rate of the residual fuel quantity.

[0080] S43: The combustion efficiency calibration parameters are related to the intake air temperature, intake air pressure and intake air composition, realizing the physical response of the combustion process to the real-time state inside the cylinder;

[0081] S44: The model's output is a cumulative heat release curve that changes over time, serving as the core energy input for calculating changes in cylinder pressure.

[0082] Specifically, this implementation achieves adaptive and precise calculation of energy input: combustion heat release is the core driving force for in-cylinder pressure generation. This implementation method avoids directly solving for the chemical reaction rate of complex turbulent combustion, and instead introduces a parameterized model that is strongly correlated with the real-time operating state of the engine to achieve a highly information-compressed equivalent of the physical process.

[0083] For example, higher intake air temperatures accelerate the chemical preparation process of fuel, resulting in a shorter ignition delay. By correlating efficiency parameters with intake air temperature, this model can output a combustion heat release curve that is "earlier and potentially faster." This characteristic enables the model to adaptively predict various environments such as engine load, intake air heating, and high-altitude conditions, a physical response that is difficult for pure black-box models to achieve.

[0084] Separating Combustion Chemistry from Gas Physics: This model clearly defines the scope of "combustion science." It encapsulates the complex chemical energy release process of fuel into "a physics-driven empirical correlation formula." This means that the master equation (the first law of thermodynamics) for subsequent cylinder pressure reconstruction can focus on handling gas physics processes without embedding complex chemical reaction kinetics code. This significantly reduces the computational complexity of the overall model and the difficulty of coupled solution, enabling the model to run in real time on microprocessors.

[0085] As a preferred approach, before performing the reconstruction calculation, the contribution analysis of the complete first law of thermodynamics was first conducted. The heat release rate curves output by the model were compared graphically with all terms included, the blow-by loss term ignored, the latent heat of fuel vaporization term ignored, and both terms ignored.

[0086] As a preferred embodiment, the following formula is used, and calculations are performed using the Hohenberg model:

[0087] Cumulative heat release prediction model: Calculated using a method based on the concept of cumulative fuel mass in the cylinder. The input to this model is the injection rate curve, and the formula is:

[0088] (5)

[0089] (6)

[0090] (7)

[0091] in, These are the cumulative fuel mass in the cylinder (mg), the amount of fuel injected into the cylinder (mg), the in-cylinder fuel consumption rate (related to the air-fuel ratio and intake thermodynamic conditions, with calibration parameters ranging from 0.0009 to 0.0015), heat release, combustion efficiency, crankshaft angle (deg), ignition delay (deg), and the lower heating value of diesel. The principle is based on formula (5) to calculate the amount of fuel that can be completely burned at the current crankshaft angle. The first term is the amount of fuel injected into the cylinder that can be completely burned at the current crankshaft angle, and the second term is the amount of fuel remaining in the cylinder at the previous crankshaft angle. The second term is 0 before the injection begins. The sum of the two is the amount of fuel that can be completely burned in the cylinder at the current crankshaft angle. Then, according to equation (6), the heat release at the current crankshaft angle is calculated and added to the heat release at the previous crankshaft angle to obtain the cumulative heat release at the current crankshaft angle. Finally, the crankshaft angle CA50 was determined when the cumulative heat release reached 50% of the total cumulative heat release. The model's prediction results can be found in the [reference needed]. Figure 2 As shown.

[0092] The first law of thermodynamics is used to calculate the cylinder pressure model, and the formula is:

[0093] (8)

[0094] The heat transfer loss Qw is calculated as shown in Equation (9). In the equation, h represents the heat transfer coefficient (W / m²·K). As represents the area where heat loss occurs. T represents the cylinder temperature (determined according to the ideal gas equation). Tref = 363.15 K. The engine's heat transfer coefficient is calculated using Hohenberg's heat transfer coefficient relation (Equation (9)). V, P, and T represent the cylinder volume (m³), pressure (Pa), and temperature (K), respectively. vm represents the piston speed (m / s).

[0095] (9)

[0096] (10)

[0097] There are two scenarios for calculating the in-cylinder temperature T. The first is under the condition of known in-cylinder pressure, where the in-cylinder temperature is calculated using the ideal gas law, assuming the cylinder is a closed system. This scenario is mainly used in the analysis of the influence of each term in the first law of thermodynamics on the calculation of the heat release rate. The second scenario is under the condition of unknown in-cylinder pressure, i.e., cylinder pressure reconstruction. In this paper, it is assumed that the compression stroke before fuel injection is a reversible adiabatic process, and Equation 11 is used to calculate the in-cylinder temperature. The pressure is calculated using Equation 12. During the expansion stroke, since the fuel combustion releases heat and thus heats the working fluid in the cylinder, the calculation of the in-cylinder temperature needs to be combined with Equation 13 based on Equation 11. Since the change in volume between adjacent crankshaft angles is very small and can be ignored, the constant volume specific heat capacity cv is used for calculation. The calculation of ΔQ in Equation 13 will be introduced below. K1 is an empirical parameter, equal to 1.2.

[0098] (11)

[0099] (12)

[0100] (13)

[0101] To further reduce the model's calibration and accelerate its computation, the impact of each term in Equation 8 on the final heat release rate was analyzed, and comparisons were made between the results without cross-flow loss ( ), and the latent heat of vaporization of fuel oil ( The heat release rate calculated by the first law of thermodynamics model with and without considering two terms is compared to the heat release rate calculated by the model with and without considering two terms. It was found that when these two terms are ignored... or The calculated heat release rate is almost identical to that obtained using the complete formula; in other words, the calculation process is virtually the same. and The impact on the calculated heat release rate is negligible. Therefore, these two terms can be ignored when calculating heat release using the first law of thermodynamics. Thus, the principle formula for the cylinder pressure reconstruction model is shown in Equation 14. Model verification results are as follows... Figure 4 As shown.

[0102] (14)

[0103] Example 2:

[0104] This invention provides an engine cylinder pressure reconstruction system based on a multiphysics dynamic coupling mechanism. The system comprises a processor and a memory, wherein the memory stores computer-executable instructions. When the processor executes the instructions, it implements the engine cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism.

[0105] Example 3:

[0106] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the program is executed by a computing device, it implements the engine cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism.

[0107] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. A method for reconstructing engine in-cylinder pressure based on multiphysics dynamic coupling mechanism, characterized in that, include: S1: The signal acquisition module acquires one or more of the following engine parameters: intake pressure, intake temperature, coolant temperature, and engine electronic control unit (ECU) signals; S2: The injection rate prediction module, based on the injection pressure and injection pulse width data contained in the ECU signal and combined with the known structural parameters of the injector, calculates the complete injection rate change curve over time through a constructed polynomial fitting model. The injection rate prediction module divides the needle valve action stage data into multiple sub-models, including but not limited to: needle valve fast opening stage, needle valve slow opening stage, needle valve full opening stage, needle valve fluctuation stage, needle valve slow closing stage, and needle valve fast closing stage. The rate function of each stage contains specific parameters calibrated from experimental data. S3: The heat transfer prediction module uses the Hohenberg model to calculate the instantaneous heat transfer loss between the working fluid in the engine cylinder and the combustion chamber wall. S4: The cumulative heat release prediction module receives the injection rate curve output by the injection rate prediction module and calculates the cumulative heat release, including the pre-injection combustion heat release rate and the main injection combustion heat release rate, based on the concept of in-cylinder cumulative fuel mass. S5: The core reconstruction module takes the cumulative heat output by the cumulative heat release prediction module, the heat transfer output by the heat transfer prediction module, the real-time cylinder volume and its rate of change calculated by the crankshaft angle sensor, and the initial pressure and temperature conditions as input, and performs simplified formula calculations based on the first law of thermodynamics to iteratively reconstruct the cylinder pressure time series of the engine in a complete working cycle.

2. The engine cylinder pressure reconstruction method based on multiphysics dynamic coupling mechanism as described in claim 1, characterized in that, Step S2 includes: S21: Based on the action characteristics of the needle valve, establish piecewise polynomial functions for the six stages, wherein the injection rate and injection pressure during the slow-opening stage of the needle valve are correlated by fitting a polynomial. S22: The multiple calibration parameters involved in the sub-model of the injection rate module are determined by the physical correlation formula between the injection pressure and the injection pressure or by joint solution with the test data, to ensure the adaptability of the model to the differences of injectors and different operating conditions. The input time unit of the fitting polynomial is milliseconds, and the output injection rate unit is milligrams per millisecond. The functions of each stage are all continuously differentiable functions with respect to time, thereby ensuring the continuity, smoothness and physical rationality of the calculated injection rate curve in time, and providing high-fidelity input boundary conditions for subsequent combustion heat release calculations.

3. The engine cylinder pressure reconstruction method based on multiphysics dynamic coupling mechanism as described in claim 2, characterized in that, The fitting formula includes: ； ； ； ； ； ； ； In the above fitting formula, the time unit is ms, and the injection rate unit is mg / ms; IR t (1) IR t (2) IR t (3) IR t (4) IR t (5) and IR t (6) Represents the injection rates corresponding to the needle valve fast opening stage t1, needle valve slow opening stage t2, needle valve full opening stage t3, needle valve fluctuation stage t4, needle valve slow closing stage t5 and needle valve fast closing stage t6 respectively; Parameters i, j and k are calibrated parameters that can be calibrated based on test data. They are obtained by fitting the needle valve slow opening stage with different injection pressures. The values ​​of i2 and j2 are -0.53 and 1.64 respectively.

4. The engine cylinder pressure reconstruction method based on multiphysics dynamic coupling mechanism as described in claim 3, characterized in that, It also includes derivation formulas for simultaneously solving the data in claim 3: The derived formula includes: (1) (2) (3) (4) Combining the fitting formula of claim 3, where the value of k1 is related to the change in injection pressure, the derived formula (1) is solved simultaneously to obtain the value of k1; through IR... t (5) and IR t (6) Perform a joint solution to obtain the values ​​of i6 and k6; IR t The parameter in (1) needs to be obtained from the value of k1 according to equation (4), and then the IR needs to be solved. t1 (1) = IR t1 (2) and IR t1 (1)'=IR t1 (2) Obtain the values ​​of i1 and j1 from the two equations. ； The value of Z is determined based on the peak injection rate.

5. The engine cylinder pressure reconstruction method based on multiphysics dynamic coupling mechanism as described in claim 4, characterized in that, The core calculation model of the cumulative heat release prediction module is based on physical derivation, and step S4 includes: S41: Calculate the pre-injection heat release rate using a formula that includes combustion efficiency calibration parameters and pre-injection delay period; S42: The heat release rate of the main injection is calculated by the main injection calibration parameters and combustion efficiency parameters, and is superimposed with the heat release rate of the residual fuel quantity. S43: The combustion efficiency calibration parameters are related to the intake air temperature, intake air pressure and intake air composition, realizing the physical response of the combustion process to the real-time state inside the cylinder; S44: The model's output is a cumulative heat release curve that changes over time, serving as the core energy input for calculating changes in cylinder pressure.

6. The engine cylinder pressure reconstruction method based on multiphysics dynamic coupling mechanism according to claim 5, characterized in that, Before performing the reconstruction calculation, we first analyzed the contribution of each term in the complete first law of thermodynamics formula. We then compared the heat release rate curves output by the model with all terms included, with the blow-by loss term ignored, with the latent heat of fuel vaporization ignored, and with both terms ignored simultaneously.

7. An engine in-cylinder pressure reconstruction system based on multiphysics dynamic coupling mechanism, characterized in that, It includes a processor and a memory, the memory storing computer-executable instructions, which, when executed by the processor, implement the engine cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism as described in any one of claims 1-6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a computing device, it implements the engine cylinder pressure reconstruction method based on the multiphysics dynamic coupling mechanism as described in any one of claims 1-6.

Images