Multi-parameter coordinated calibration method and system for transient working condition of hybrid internal combustion engine

Abstract

The application provides a transient working condition multi-parameter collaborative calibration method and system of a hybrid internal combustion engine, and the method comprises the following steps: acquiring a data stream in an on-board diagnostic system, and screening feature information of seven key dimensions from the data stream; creating a gene map matching engine, and inputting the feature information of the seven key dimensions into the gene map matching engine; comparing the feature information of the seven key dimensions with pre-stored working conditions in the gene map matching engine, obtaining an optimal working condition, and outputting a corresponding basic parameter package; combining knock energy spectrum characteristics and oxygen sensor offset, and performing matrix operation to optimize the basic parameter package, and obtaining a corrected parameter package; and performing boundary control on an ignition angle based on operating parameters of the internal combustion engine, and using the boundary-controlled ignition angle to constrain the ignition angle in the operation process of the internal combustion engine. The application can dynamically correct each parameter of the internal combustion engine.

Description

Technical Field

[0001] This invention relates to the field of internal combustion engine control technology, and in particular to a method and system for transient multi-parameter collaborative calibration of a hybrid internal combustion engine. Background Technology

[0002] In the operation of hybrid vehicles, the power source switching process (pure electric → engine mode) is a crucial link in ensuring the continuity of vehicle power and fuel economy. During this process, the internal combustion engine needs to respond to torque requests rapidly within an extremely short period of 0.5 to 2 seconds to achieve a smooth power transition. However, traditional calibration parameters based on steady-state operating conditions are difficult to adapt to the drastic changes under transient operating conditions. This sudden change in operating conditions is not only reflected in the instantaneous increase in torque demand, but also involves the coordinated changes of multiple parameters such as engine speed, intake air volume, and fuel injection.

[0003] Currently, the industry has conducted extensive research on this technical challenge, mainly focusing on the following two directions:

[0004] One method is parameter switching control based on preset rule tables. This type of technology establishes a correspondence between operating conditions and control parameters to achieve rapid parameter switching. For example, the SAE 2019-01-5007 standard details the parameter adjustment rules for switching power sources in hybrid vehicles under different operating conditions. This standard divides different operating condition ranges based on parameters such as vehicle speed and load, and presets corresponding control parameters for each range. This allows for rapid adjustment of the internal combustion engine's operating state according to predetermined rules during power source switching.

[0005] Second, real-time optimization using model predictive control (MPC). By establishing a dynamic model of the engine and combining current operating condition information with predicted future operating conditions, the control parameters of the internal combustion engine are optimized in real time. In transient operating conditions where the power source switches, model predictive control can predict changes in the engine's operating state in advance and adjust the control parameters in advance based on the prediction results to achieve more accurate torque response and more efficient power output.

[0006] Disadvantages of existing technology

[0007] (1) Insufficient parameter synergy

[0008] Most existing technologies employ independent adjustment of multiple parameters. While this approach can control the engine's operating state to some extent, it fails to fully consider the interactions and synergies between these parameters. This leads to misalignment of the combustion phase during transient conditions such as power source switching. For example, when adjusting fuel injection parameters, neglecting the coordinated changes in ignition angle and intake air volume can result in incomplete fuel combustion, causing power output fluctuations, increased emissions, and other problems that severely impact engine performance and the vehicle's driving experience.

[0009] (2) Significant response delay

[0010] Current predictive algorithms generally suffer from long computation times, often exceeding the cycle of operating condition changes, typically reaching 150ms. During the power source switching process in hybrid vehicles, operating conditions change extremely rapidly, demanding very high real-time performance from control parameters. If the predictive algorithm cannot complete the calculation and output optimized control parameters within a short time, it cannot adapt to changes in operating conditions promptly, leading to a lag in engine torque response and impacting vehicle power performance and driving smoothness.

[0011] (3) Poor boundary adaptability

[0012] Current technologies typically use fixed knock thresholds to control engine performance. However, during long-term use, engine performance changes due to wear and aging of components. Fixed knock thresholds cannot be dynamically adjusted according to the actual condition of the engine and cannot effectively compensate for the effects of engine aging and deterioration. As the engine ages, the original knock thresholds may no longer be applicable, potentially leading to misjudgments or missed knock detections, thus affecting the normal operation and lifespan of the engine. Summary of the Invention

[0013] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method for multi-parameter collaborative calibration of a hybrid internal combustion engine under transient operating conditions, which aims to solve the technical problems mentioned in the background art.

[0014] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0015] A method for transient multi-parameter collaborative calibration of a hybrid internal combustion engine includes the following steps:

[0016] Acquire the data stream from the vehicle diagnostic system and filter out feature information from seven key dimensions from the data stream;

[0017] Create a gene map matching engine and input feature information from 7 key dimensions into the gene map matching engine;

[0018] The feature information of the seven key dimensions is compared with the pre-stored working conditions in the gene map matching engine to obtain the optimal working conditions and output the corresponding basic parameter package.

[0019] By combining the characteristics of the detonation energy spectrum and the oxygen sensor offset, the basic parameter package is optimized through matrix operations to obtain a modified parameter package;

[0020] Based on the operating parameters of the internal combustion engine, the ignition angle is controlled by boundary, and the ignition angle after boundary control is used to constrain the ignition angle during the operation of the internal combustion engine.

[0021] According to one aspect of the above technical solution, in the step of "acquiring the data stream from the vehicle diagnostic system and filtering out feature information of 7 key dimensions from the data stream", the data stream is the real-time operating condition information of the internal combustion engine.

[0022] According to one aspect of the above technical solution, the optimization of the basic parameter package by combining the detonation energy spectrum characteristics and oxygen sensor offset through matrix operations includes the following specific steps:

[0023] The detonation signal collected by the detonation sensor is analyzed, and the frequency band of the detonation signal from 3kHz to 15kHz is focused to calculate the detonation energy value.

[0024] A model for oxygen sensor offset compensation is constructed. The deviation between the measured value and the target value of the oxygen sensor is compared and multiplied by a coefficient to obtain the oxygen offset compensation value.

[0025] Based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, the basic weights of each parameter in the basic parameter package are optimized through matrix operations to obtain a modified parameter package.

[0026] According to one aspect of the above technical solution, the step of boundary control of the ignition angle based on the operating parameters of the internal combustion engine, and using the boundary-controlled ignition angle to constrain the ignition angle during the operation of the internal combustion engine, specifically includes:

[0027] The cumulative operating hours of the internal combustion engine, the knock frequency within a predetermined time and the fuel octane number are obtained, and the knock offset is calculated through the knock threshold offset model.

[0028] The detonation offset is compared with the detonation offset threshold.

[0029] If the knock offset does not meet the knock offset threshold, the temperature of the coolant is obtained, and the ignition angle is controlled by modifying the ignition angle boundary formula.

[0030] The ignition angle during the operation of an internal combustion engine is constrained by the ignition angle after boundary control.

[0031] According to one aspect of the above technical solution, the analysis of the detonation signal collected by the detonation sensor focuses on the frequency band of the detonation signal from 3kHz to 15kHz to calculate the detonation energy value. Specific steps include:

[0032] The detonation signal acquired by the detonation sensor is subjected to bandpass filtering to obtain a filtered detonation signal. The frequency of the bandpass filter is 3kHz to 15kHz.

[0033] The filtered detonation signal is subjected to a sliding convolution operation using a one-dimensional convolution kernel to obtain a convolution sequence;

[0034] The convolutional sequence is activated using the ReLU activation function to obtain the detonation feature signal;

[0035] The root mean square of the detonation characteristic signal is calculated within a calculation window to obtain the detonation energy value.

[0036] According to one aspect of the above technical solution, the step of optimizing the basic weights of each parameter in the basic parameter package through matrix operations based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, to obtain a corrected parameter package, specifically includes:

[0037] Calculate the ratio of the detonation energy value to the detonation threshold to obtain the detonation energy ratio;

[0038] The detonation energy ratio and the oxygen offset compensation value are used to form a state vector;

[0039] A weight matrix is ​​constructed based on the influence of the basic weight, the knock energy ratio, and the oxygen offset compensation value on the ignition angle, EGR rate, and fuel injection quantity, respectively.

[0040] The updated weight matrix is ​​obtained by multiplying the state vector and the weight matrix.

[0041] The updated weight matrix is ​​normalized to obtain the final optimized matrix;

[0042] By adjusting the weights in the final optimization matrix, the basic weights of ignition angle, EGR rate, and fuel injection quantity are dynamically adjusted to obtain a correction parameter package.

[0043] This invention also provides a multi-parameter collaborative calibration system for transient operating conditions of a hybrid internal combustion engine, comprising:

[0044] Acquisition module: used to acquire the data stream from the vehicle diagnostic system and filter out feature information in 7 key dimensions from the data stream;

[0045] Input module: Used to create a gene map matching engine, inputting feature information from 7 key dimensions into the gene map matching engine;

[0046] Output module: Used to compare the feature information of the 7 key dimensions with the pre-stored working conditions in the gene map matching engine to obtain the optimal working conditions and output the corresponding basic parameter package;

[0047] Optimization module: Used to combine the detonation energy spectrum characteristics and oxygen sensor offset, and optimize the basic parameter package through matrix operations to obtain a corrected parameter package;

[0048] Control module: Used to perform boundary control on the ignition angle based on the operating parameters of the internal combustion engine, and to constrain the ignition angle during the operation of the internal combustion engine using the boundary-controlled ignition angle.

[0049] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0050] By selecting feature information from seven key dimensions and inputting it into a gene mapping matching engine, a basic parameter package is obtained. Then, by combining knock energy spectrum features and oxygen sensor offset, the basic parameter package is optimized through matrix operations to obtain a modified parameter package. This enhances the synergistic performance between parameters, resulting in more precise combustion phases and reduced computational response delay during transient power source switching. By using the operating parameters of the internal combustion engine to perform boundary control of the ignition angle, the ignition angle constrained during engine operation can be transformed from passive to active control. This allows for proactive prediction and compensation for changes in knock risk caused by engine aging and fuel differences. The control is more precise, comprehensively considering multiple dimensions such as time, number of impacts, and fuel quality, significantly improving the adaptability and robustness of the control system. Balancing safety and efficiency, it provides greater optimization space for improving power and economy while ensuring the lifetime reliability of the engine. Attached Figure Description

[0051] Figure 1 This is a flowchart of the transient multi-parameter collaborative calibration method for a hybrid internal combustion engine in the first embodiment of the present invention;

[0052] Figure 2 This is a structural block diagram of the transient operating condition multi-parameter collaborative calibration system for a hybrid internal combustion engine in the second embodiment of the present invention;

[0053] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0054] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0055] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0057] Please see Figure 1 The figure shows a method for transient multi-parameter collaborative calibration of a hybrid internal combustion engine according to the first embodiment of the present invention, which includes the following steps:

[0058] S10, acquire the data stream from the vehicle diagnostic system, and filter out feature information of 7 key dimensions from the data stream;

[0059] S20, Create a gene map matching engine and input the feature information of 7 key dimensions into the gene map matching engine;

[0060] S30, compare the feature information of the 7 key dimensions with the pre-stored working conditions in the gene map matching engine to obtain the optimal working conditions and output the corresponding basic parameter package.

[0061] S40, combining the detonation energy spectrum characteristics and oxygen sensor offset, the basic parameter package is optimized through matrix operations to obtain a modified parameter package;

[0062] S50 uses the operating parameters of the internal combustion engine to perform boundary control on the ignition angle, and uses the boundary-controlled ignition angle to constrain the ignition angle during the operation of the internal combustion engine.

[0063] Understandably, this invention selects feature information from seven key dimensions, inputs it into a gene mapping matching engine to obtain a basic parameter package, and then optimizes the basic parameter package through matrix operations, combining knock energy spectrum features and oxygen sensor offset, to obtain a corrected parameter package. This enhances the synergistic performance between parameters, resulting in more precise combustion phases and reduced computational response delay during transient power source switching. By using the operating parameters of the internal combustion engine to perform boundary control of the ignition angle, and using the boundary-controlled ignition angle to constrain the ignition angle during engine operation, it can transform passive control into active control, proactively predicting and compensating for changes in knock risk caused by engine aging and fuel differences. The control is more precise, comprehensively considering multiple dimensions such as time, number of impacts, and fuel quality, significantly improving the adaptability and robustness of the control system. Balancing safety and efficiency, it provides greater optimization space for improving power and economy while ensuring the lifetime reliability of the engine.

[0064] Furthermore, in step S10, the data stream is the real-time operating condition information of the internal combustion engine. While the vehicle is in motion, the system collects the On-Board Diagnostic (OBD) data stream at a frequency of 100 times per second (100Hz), and this data contains the real-time operating condition information of the internal combustion engine. Through a lightweight feature extraction module, the system filters out feature information in seven key dimensions from the massive amount of data, generating a 7-dimensional feature vector [F1-F7], much like taking a "snapshot" of the internal combustion engine's operating status.

[0065] Specifically, the data stream in step S10 includes:

[0066] Engine torque T, engine speed N, coolant temperature T w Intake pressure P in Inlet air temperature T in Throttle opening θ and air-fuel ratio λ are sampled at 10 points within a 100ms window.

[0067] Feature information in 7 key dimensions was extracted from 7 data streams:

[0068] Torque change rate ;

[0069] Speed ​​gradient ;

[0070] Coolant temperature rise rate ;

[0071] Intake pressure change rate ;

[0072] Average intake air temperature ;

[0073] Throttle opening change ;

[0074] Standard deviation of air-fuel ratio ,in ;

[0075] Finally, a feature vector F=[F1,F2,F3,F4,F5,F6,F7] is generated every 100ms. T In the above formula, the subscript of the parameter indicates the serial number of the sampling point.

[0076] In steps S20 and S30, the generated 7-dimensional feature vector is input into the gene mapping matching engine. This engine acts like a "database," pre-stored with data on over 500 different working condition segments. The system compares the current feature vector with the pre-stored working conditions, finds the best matching condition, and outputs the basic parameter package P. base This is equivalent to finding a "basic operating scheme" for the current operating conditions of the internal combustion engine. The parameters in the basic parameter package are ignition angle, EGR rate, and fuel injection quantity.

[0077] Specifically, step S20 includes the following steps:

[0078] Several typical operating conditions of the vehicle during operation are obtained, and feature vectors Gm=[Gm1, Gm2, Gm3, Gm4, Gm5, Gm6, Gm7] for each typical operating condition are calculated based on feature information from seven key dimensions. T Gm and F are isomorphic. Gm1 has the same meaning as F1, except that one is under typical operating conditions and the other is under current operating conditions.

[0079] Under each of the typical operating conditions, obtain the optimal typical parameter package P. m= [θ ign-m EGR m Fuel m ] T ;θ ign-m EGR represents the ignition angle under the m-th operating condition. m Fuel represents the EGR rate under the m-th operating condition. m This represents the fuel injection quantity under the m-th operating condition;

[0080] Collect the eigenvectors Gm and typical parameter packages P under all typical operating conditions. m Constructs a gene mapping matching engine;

[0081] The feature vector F acquired in real time is input into the gene map matching engine.

[0082] Specifically, step S30 includes the following steps:

[0083] Calculate the Euclidean distance between the real-time acquired feature vector F and the feature vector Gm under the typical working condition;

[0084] The real-time acquired feature vector F is matched with the feature vector Gm under the typical working condition with the minimum Euclidean distance, and this typical working condition is taken as the optimal working condition, using the typical parameter package P of the optimal working condition. m This serves as the basic parameter package for the current operating conditions.

[0085] Furthermore, the specific steps of step S40 include:

[0086] The detonation signal collected by the detonation sensor is analyzed, and the frequency band of the detonation signal from 3Hz to 15Hz is focused to calculate the detonation energy value.

[0087] A model for oxygen sensor offset compensation is constructed. The deviation between the measured value and the target value of the oxygen sensor is compared and multiplied by a coefficient to obtain the oxygen offset compensation value.

[0088] Based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, the basic weights of each parameter in the basic parameter package are optimized through matrix operations to obtain a modified parameter package.

[0089] Specifically, the detonation signal collected by the detonation sensor is analyzed, focusing on the frequency band of the detonation signal from 3kHz to 15kHz to calculate the detonation energy value. The specific steps include:

[0090] The detonation signal acquired by the detonation sensor is subjected to bandpass filtering to obtain a filtered detonation signal. The frequency of the bandpass filter is 3kHz to 15kHz.

[0091] V bp =BPF [3kHz,15kHz] (V knock (t));

[0092] Where BPF represents bandpass filtering, t represents the time of the signal, and V knock Indicates the detonation signal, V bp This represents the filtered detonation signal;

[0093] The filtered detonation signal is subjected to a sliding convolution operation using a one-dimensional convolution kernel to obtain a convolution sequence;

[0094] ;

[0095] Where C represents the convolution sequence, j represents the internal index of the one-dimensional convolution kernel, j=1,2,3, K represents the one-dimensional convolution kernel, and t iThis represents the signal at time i;

[0096] The convolutional sequence is activated using the ReLU activation function to obtain the detonation feature signal;

[0097] ;

[0098] Where R(t) i () indicates the characteristic signal of detonation;

[0099] The root mean square of the detonation characteristic signal is calculated within a calculation window to obtain the detonation energy value;

[0100] ;

[0101] Where N represents the number of sampling points within the calculation window, E knock This indicates the detonation energy value.

[0102] Furthermore, the construction of the oxygen sensor offset compensation model involves comparing the deviation between the measured value and the target value of the oxygen sensor, multiplying by a coefficient, and obtaining the oxygen offset compensation value. Specific steps include:

[0103] Obtain the measured value and target value of the oxygen sensor;

[0104] An oxygen sensor offset compensation model is constructed. The measured value and the target value of the oxygen sensor are substituted into the oxygen sensor offset compensation model, multiplied by the compensation coefficient, and the oxygen offset compensation value is calculated.

[0105] Δλ= U (λ meas - λ target );

[0106] Where Δλ represents the oxygen offset compensation value, λ meas λ represents the measured value of the oxygen sensor. target The target value of the oxygen sensor is represented by U, and the compensation coefficient is U=f. map (Ne, L), f map This indicates a mapping operation, which is obtained by looking up the values ​​of engine speed Ne and engine load L in a two-dimensional mapping table.

[0107] Furthermore, the step of optimizing the basic weights of each parameter in the basic parameter package through matrix operations based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, to obtain a corrected parameter package, specifically includes:

[0108] Calculate the ratio of the detonation energy value to the detonation threshold to obtain the detonation energy ratio;

[0109] ;

[0110] Among them, E b E represents the knock threshold. c Indicates the detonation energy ratio;

[0111] The detonation energy ratio and the oxygen offset compensation value are used to form a state vector;

[0112] ;

[0113] Among them, S n This represents the state vector, with 1 indicating a bias term;

[0114] A weight matrix is ​​constructed based on the influence of the basic weight, the knock energy ratio, and the oxygen offset compensation value on the ignition angle, EGR rate, and fuel injection quantity, respectively.

[0115]

[0116] Among them, A new Let A represent the weight matrix. 11 A 21 A 31 These represent the base weights of ignition angle, EGR rate, and fuel injection quantity, respectively. 12 A 22 A 32 These represent the effects of knock energy ratio on the base weights of ignition angle, EGR rate, and injection quantity, respectively. 13 A 23 A 33 These represent the effects of oxygen offset compensation values ​​on the base weights of ignition angle, EGR rate, and fuel injection quantity, respectively.

[0117] The updated weight matrix is ​​obtained by multiplying the state vector and the weight matrix.

[0118] W new =A new ·S n ;

[0119] The updated weight matrix is ​​normalized to obtain the final optimized matrix;

[0120] ;

[0121] Among them, W final This represents the final optimization matrix. α Indicates a weighted index. β Indicates a circular index;

[0122] By adjusting the weights in the final optimization matrix, the basic weights of ignition angle, EGR rate, and fuel injection quantity are dynamically adjusted to obtain a correction parameter package.

[0123] Furthermore, the step of performing boundary control on the ignition angle based on the operating parameters of the internal combustion engine, and using the boundary-controlled ignition angle to constrain the ignition angle during the operation of the internal combustion engine, specifically includes:

[0124] The cumulative operating hours of the internal combustion engine, the knock frequency within a predetermined time and the fuel octane number are obtained, and the knock offset is calculated through the knock threshold offset model.

[0125] δ knock = K1 (1-e (-t / τ) ) + K2 N knock + K3 (RON);

[0126] Where, δ knock The value represents the knock offset, t represents the cumulative operating hours of the internal combustion engine, τ represents the time constant, which is a parameter that determines the degradation rate. The smaller τ is, the faster the degradation reaches saturation. K1 is the saturation coefficient. When time t is much greater than τ, this term approaches 1 infinitely. N knock K2 represents the number of detonations within a predetermined time (30 minutes), and K2 represents the single impact damage coefficient, reflecting the impact of each impact event on δ. knock The average contribution, where RON is the fuel octane number and K3 is the sensitivity coefficient of the impact of fuel octane number changes on knock risk;

[0127] The detonation offset is compared with the detonation offset threshold.

[0128] If the knock offset does not meet the knock offset threshold, the temperature of the coolant is obtained, and the ignition angle is controlled by modifying the ignition angle boundary formula.

[0129] θ ignlim' = θ ignlim - γ δ knock ;

[0130] Where, θ ignlim' θ represents the ignition angle after boundary control. ignlim The value represents the original ignition angle, and γ represents the compensation gain, which is automatically adjusted according to the coolant temperature: when the coolant temperature is below 60℃, γ is set to 0.3; when the temperature is between 60℃ and 90℃, γ is set to 0.5; when the temperature is above 90℃, γ is set to 0.7.

[0131] The ignition angle during the operation of an internal combustion engine is constrained by the ignition angle after boundary control.

[0132] In summary, the transient multi-parameter collaborative calibration method for hybrid internal combustion engines in the above embodiments of the present invention can dynamically adjust the various parameters of the internal combustion engine to improve the synergy of the parameters.

[0133] Please refer to Figure 2 The figure shows a transient multi-parameter collaborative calibration system for a hybrid internal combustion engine according to a second embodiment of the present invention, comprising:

[0134] Acquisition module 11: used to acquire the data stream from the on-board diagnostic system and filter out feature information of 7 key dimensions from the data stream; the data stream is the real-time operating condition information of the internal combustion engine;

[0135] Input module 12: Used to create a gene map matching engine, inputting feature information of 7 key dimensions into the gene map matching engine;

[0136] Output module 13: is used to compare the feature information of the seven key dimensions with the pre-stored working conditions in the gene map matching engine to obtain the optimal working conditions and output the corresponding basic parameter package;

[0137] Optimization module 14: Used to combine the detonation energy spectrum characteristics and oxygen sensor offset, and optimize the basic parameter package through matrix operations to obtain a corrected parameter package;

[0138] The optimization module 14 is specifically used to: analyze the detonation signal collected by the detonation sensor, focus the frequency band of the detonation signal from 3kHz to 15kHz, and calculate the detonation energy value;

[0139] A model for oxygen sensor offset compensation is constructed. The deviation between the measured value and the target value of the oxygen sensor is compared and multiplied by a coefficient to obtain the oxygen offset compensation value.

[0140] Based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, the basic weights of each parameter in the basic parameter package are optimized through matrix operations to obtain a modified parameter package.

[0141] The analysis of the detonation signal collected by the detonation sensor focuses on the frequency band of the detonation signal from 3kHz to 15kHz to calculate the detonation energy value. Specific steps include:

[0142] The detonation signal acquired by the detonation sensor is subjected to bandpass filtering to obtain a filtered detonation signal. The frequency of the bandpass filter is 3kHz to 15kHz.

[0143] The filtered detonation signal is subjected to a sliding convolution operation using a one-dimensional convolution kernel to obtain a convolution sequence;

[0144] The convolutional sequence is activated using the ReLU activation function to obtain the detonation feature signal;

[0145] The root mean square of the detonation characteristic signal is calculated within a calculation window to obtain the detonation energy value;

[0146] The step of optimizing the basic weights of each parameter in the basic parameter package using matrix operations based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, to obtain a corrected parameter package, specifically includes:

[0147] Calculate the ratio of the detonation energy value to the detonation threshold to obtain the detonation energy ratio;

[0148] The detonation energy ratio and the oxygen offset compensation value are used to form a state vector;

[0149] A weight matrix is ​​constructed based on the influence of the basic weight, the knock energy ratio, and the oxygen offset compensation value on the ignition angle, EGR rate, and fuel injection quantity, respectively.

[0150] The updated weight matrix is ​​obtained by multiplying the state vector and the weight matrix.

[0151] The updated weight matrix is ​​normalized to obtain the final optimized matrix;

[0152] By using the weights in the final optimization matrix, the basic weights of ignition angle, EGR rate and fuel injection quantity are dynamically corrected to obtain a correction parameter package.

[0153] Control module 15: Used to perform boundary control on the ignition angle based on the operating parameters of the internal combustion engine, and to constrain the ignition angle during the operation of the internal combustion engine using the boundary-controlled ignition angle.

[0154] The control module 15 is specifically used to: acquire the cumulative operating hours of the internal combustion engine, the knock frequency within a predetermined time and the fuel octane number, and calculate the knock offset through the knock threshold offset model;

[0155] The detonation offset is compared with the detonation offset threshold.

[0156] If the knock offset does not meet the knock offset threshold, the temperature of the coolant is obtained, and the ignition angle is controlled by modifying the ignition angle boundary formula.

[0157] The ignition angle during the operation of an internal combustion engine is constrained by the ignition angle after boundary control.

[0158] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0159] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for transient multi-parameter collaborative calibration of a hybrid internal combustion engine, characterized in that, Includes the following steps: Acquire the data stream from the vehicle diagnostic system and filter out feature information from seven key dimensions from the data stream; Create a gene map matching engine and input feature information from 7 key dimensions into the gene map matching engine; The feature information of the seven key dimensions is compared with the pre-stored working conditions in the gene map matching engine to obtain the optimal working conditions and output the corresponding basic parameter package. By combining the characteristics of the detonation energy spectrum and the oxygen sensor offset, the basic parameter package is optimized through matrix operations to obtain a modified parameter package; Based on the operating parameters of the internal combustion engine, the ignition angle is controlled by boundary, and the ignition angle after boundary control is used to constrain the ignition angle during the operation of the internal combustion engine. In the step of "acquiring the data stream from the on-board diagnostic system and filtering out feature information of 7 key dimensions from the data stream", the data stream is the real-time operating condition information of the internal combustion engine. The optimization of the basic parameter package by combining the detonation energy spectrum characteristics and oxygen sensor offset through matrix operations includes the following specific steps: The detonation signal collected by the detonation sensor is analyzed, and the frequency band of the detonation signal from 3kHz to 15kHz is focused to calculate the detonation energy value. A model for oxygen sensor offset compensation is constructed. The deviation between the measured value and the target value of the oxygen sensor is compared and multiplied by a coefficient to obtain the oxygen offset compensation value. Based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, the basic weights of each parameter in the basic parameter package are optimized through matrix operations to obtain a modified parameter package. The method of boundary control of ignition angle based on the operating parameters of the internal combustion engine, and using the boundary-controlled ignition angle to constrain the ignition angle during the operation of the internal combustion engine, includes the following specific steps: The cumulative operating hours of the internal combustion engine, the knock frequency within a predetermined time and the fuel octane number are obtained, and the knock offset is calculated through the knock threshold offset model. The detonation offset is compared with the detonation offset threshold. If the knock offset does not meet the knock offset threshold, the temperature of the coolant is obtained, and the ignition angle is controlled by modifying the ignition angle boundary formula. The ignition angle during the operation of an internal combustion engine is constrained by the ignition angle after boundary control.

2. The transient multi-parameter collaborative calibration method for a hybrid internal combustion engine according to claim 1, characterized in that, The analysis of the detonation signal collected by the detonation sensor focuses on the frequency band of the detonation signal from 3kHz to 15kHz to calculate the detonation energy value. Specific steps include: The detonation signal acquired by the detonation sensor is subjected to bandpass filtering to obtain a filtered detonation signal. The frequency of the bandpass filter is 3kHz to 15kHz. The filtered detonation signal is subjected to a sliding convolution operation using a one-dimensional convolution kernel to obtain a convolution sequence; The convolutional sequence is activated using the ReLU activation function to obtain the detonation feature signal; The root mean square of the detonation characteristic signal is calculated within a calculation window to obtain the detonation energy value.

3. The transient multi-parameter collaborative calibration method for a hybrid internal combustion engine according to claim 2, characterized in that, The step of optimizing the basic weights of each parameter in the basic parameter package using matrix operations based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, to obtain a corrected parameter package, specifically includes: Calculate the ratio of the detonation energy value to the detonation threshold to obtain the detonation energy ratio; The detonation energy ratio and the oxygen offset compensation value are used to form a state vector; A weight matrix is ​​constructed based on the influence of the basic weight, the knock energy ratio, and the oxygen offset compensation value on the ignition angle, EGR rate, and fuel injection quantity, respectively. The updated weight matrix is ​​obtained by multiplying the state vector and the weight matrix. The updated weight matrix is ​​normalized to obtain the final optimized matrix; By adjusting the weights in the final optimization matrix, the basic weights of ignition angle, EGR rate, and fuel injection quantity are dynamically adjusted to obtain a correction parameter package.

4. A multi-parameter collaborative calibration system for transient operating conditions of a hybrid internal combustion engine, characterized in that, include: Acquisition module: used to acquire the data stream from the on-board diagnostic system and filter out feature information of 7 key dimensions from the data stream; the data stream is the real-time operating condition information of the internal combustion engine; Input module: Used to create a gene map matching engine, inputting feature information from 7 key dimensions into the gene map matching engine; Output module: Used to compare the feature information of the 7 key dimensions with the pre-stored working conditions in the gene map matching engine to obtain the optimal working conditions and output the corresponding basic parameter package; Optimization module: Used to combine the detonation energy spectrum characteristics and oxygen sensor offset, and optimize the basic parameter package through matrix operations to obtain a corrected parameter package; The optimization module is specifically used to: analyze the detonation signal collected by the detonation sensor, focus the frequency band of the detonation signal from 3kHz to 15kHz, and calculate the detonation energy value; A model for oxygen sensor offset compensation is constructed. The deviation between the measured value and the target value of the oxygen sensor is compared and multiplied by a coefficient to obtain the oxygen offset compensation value. Based on the basic weights of each parameter in the basic parameter package, the ratio of the detonation energy value to the detonation threshold, and the oxygen offset compensation value, the basic weights of each parameter in the basic parameter package are optimized through matrix operations to obtain a modified parameter package. Control module: Used to perform boundary control on the ignition angle based on the operating parameters of the internal combustion engine, and to constrain the ignition angle during the operation of the internal combustion engine using the boundary-controlled ignition angle; The control module is specifically used to: acquire the cumulative operating hours of the internal combustion engine, the knock frequency within a predetermined time and the fuel octane number, and calculate the knock offset through the knock threshold offset model; The detonation offset is compared with the detonation offset threshold. If the knock offset does not meet the knock offset threshold, the temperature of the coolant is obtained, and the ignition angle is controlled by modifying the ignition angle boundary formula. The ignition angle during the operation of an internal combustion engine is constrained by the ignition angle after boundary control.

Images