A cross-cycle fuel injection control method and system suitable for strong transient working conditions

Abstract

This invention discloses a cross-cycle injection control method and system suitable for strong transient operating conditions. Addressing the technical problems of traditional quasi-steady-state control in high-pressure common rail diesel engines under strong transient conditions—ignoring cross-cycle memory effects, air-fuel mismatch, combustion phase drift, and difficulty in balancing power and emissions—this invention extracts combustion characteristics by collecting cycle-by-cycle operating data, and constructs a cross-cycle combustion state vector by fusing combustion, intake, injection, and heat / oil film state estimates from the current cycle and at least one previous cycle. It calculates the strong transient degree index and cross-cycle memory coefficient to predict the combustion phase and combustion deterioration risk in the next cycle, and then collaboratively corrects the main injection timing, injection pressure, injection quantity change rate, and pre-injection / post-injection parameters to form a cycle-by-cycle closed-loop control. This invention effectively suppresses CA50 shift and prolonged combustion duration, balancing transient torque response, fuel economy, and emission control. It is adaptable to different hardware configurations and can be superimposed on existing ECU steady-state injection MAPs.

Description

Technical Field

[0001] This invention relates to the field of diesel engine combustion control and fuel injection control technology, and in particular to a cross-cycle fuel injection control method, control system, electronic control unit and storage medium for diesel engines under strong transient conditions such as rapid loading, rapid acceleration, rapid fluctuation of intake air pressure, boost lag and EGR transient changes. Background Technology

[0002] High-pressure common rail diesel engines frequently encounter strong transient conditions during operation, such as rapid loading, rapid acceleration, sudden speed changes, boost lag, and rapid fluctuations in EGR rate. Under these conditions, the time scales of fuel supply, intake boost, EGR response, cylinder wall thermal state, and cylinder wall oil film dynamic changes are mismatched, resulting in a severe mismatch between the fuel injection quantity demand and the actual intake air volume, oxygen concentration, and temperature field in the cylinder.

[0003] Currently, diesel engine injection control generally adopts steady-state calibration (MAP) or quasi-steady-state correction strategies. These strategies determine the main injection timing, rail pressure, injection quantity, and multiple injection parameters based solely on instantaneous parameters such as engine speed, load, intake air pressure, and coolant temperature during the current single cycle. This approach is only suitable for steady-state and gradually changing operating conditions and has significant limitations under strong transient operating conditions. The existing control logic only uses the current operating condition data, which cannot quantify the cross-cycle memory effect caused by the combustion phase of the previous cycle, residual exhaust gas, cylinder wall thermal state, and oil film residue. It is difficult to predict the coupling effect of the historical cycle on the ignition and combustion phase of the next cycle. During rapid loading, the rate of increase in fuel injection volume is much higher than the rate of increase in intake pressure and intake volume, which can easily lead to large-area localized rich combustion, resulting in intensified diffusion combustion, a shift of the CA50 heat release center to the rear, an increase in the peak particulate matter emission, and deterioration of fuel economy. Traditional control only calibrates CA50, smoke opacity, and NOx under steady-state conditions, lacks a mechanism to quantify the degree of transient changes, cannot identify the risk of combustion deterioration in advance, and can only passively compensate after combustion deterioration occurs; Existing transient compensation methods are limited. While simply limiting the amount of fuel injection can reduce smoke, it will significantly weaken the transient torque response. A single advance injection strategy will cause excessive cylinder pressure rise rate, increased combustion noise and NOx emissions, making it impossible to achieve a balance between multiple objectives such as power, fuel consumption and emissions.

[0004] In summary, existing fuel injection control schemes lack a predictive collaborative correction mechanism that integrates multiple cycle states, making it difficult to adapt to the control requirements of diesel engines under strong transient operating conditions. Summary of the Invention

[0005] This invention provides a diesel engine cross-cycle injection control method suitable for strong transient operating conditions, applied to high-pressure common rail diesel engines, comprising the following steps: S101. Collect the operating data of the diesel engine during the nth working cycle. The operating data includes cylinder pressure, crankshaft angle, speed, torque demand, intake pressure, intake temperature, rail pressure, and injection timing. S102. Synchronously filter and cycle-divide the cylinder pressure signal and operating data to obtain the nth cycle pressure-crankshaft angle sequence and intake / exhaust and fuel injection boundary parameters; S103. Calculate the apparent heat release rate and cumulative heat release based on cylinder pressure data, and extract combustion characteristic parameters such as SOC, CA10, CA50, CA90, combustion duration, IMEP, and maximum pressure rise rate. S104. By integrating the combustion characteristics, intake state, injection state, and hot / oil film state estimates of the nth cycle and at least one preceding cycle, a cross-cycle combustion state vector is constructed. S105. Calculate the strong transient degree index and the cross-cycle memory coefficient based on the cross-cycle combustion state vector to identify the current transient level and the risk of combustion deterioration in the next cycle. S106. By using a combustion phase prediction model and combining it with cross-cycle combustion state vectors, the combustion phase, fuel-rich zone risk, and post-oxidation capacity of the (n+1)th cycle are predicted. S107. Combining the predicted combustion phase with the target CA50, generate the injection correction amount for the (n+1)th cycle. The injection correction amount includes the main injection timing correction amount, the injection pressure correction amount, the injection quantity change rate limit amount, and the pre-injection / post-injection parameter correction amount. S108. The injection command after superimposed correction is sent to the injection drive module; after the n+1th cycle ends, the actual combustion response is collected to update the cross-cycle state vector and model parameters, forming a cycle-by-cycle closed-loop control.

[0006] Preferably, the strong transient intensity index is obtained by weighted summation of the injection quantity change amplitude, intake pressure target deviation, air-fuel ratio critical difference, CA50 offset, and cross-cycle memory coefficient; the cross-cycle memory coefficient is calculated by weighted summation of CA50 cycle offset, combustion duration change, wall oil film risk, intake pressure hysteresis, and late-stage oxidation loss; based on the relationship between the strong transient intensity index and the three preset thresholds, the control system switches between different control modes: When the strong transient intensity index is greater than the first threshold and not greater than the second threshold, the transient pre-response mode is entered. When the index is greater than the second threshold, it enters a strong compensation mode; When the index falls from above the third threshold to below the third threshold, it enters the recovery correction mode; Wherein, the first threshold is less than the second threshold, and the third threshold is less than or equal to the first threshold.

[0007] Preferably, the main injection timing correction is calculated by weighting the difference between the predicted CA50 and the target optimal CA50, the intake pressure lag difference, and the cross-cycle memory coefficient. If the predicted CA50 is shifted backward relative to the target CA50 and the maximum pressure rise rate does not exceed the limit, the main injection timing is advanced or the pre-injection amount is increased. If the predicted pressure rise rate exceeds the limit, the main injection advance amount is reduced, the pre-injection interval is adjusted, or the rate of change of the injection amount per cycle is limited.

[0008] Preferably, the injection pressure correction amount is adjusted based on the risk level of the fuel-rich zone. When the fuel-rich risk is high and the rail pressure system has an adjustment margin, the rail pressure is increased. The post-injection parameters are adjusted according to the subsequent oxidation capacity index: when the carbon soot risk is high and the oxidation capacity meets the standard, a small proportion of post-injection is used to enhance in-cylinder oxidation; when the risk of oil film on the wall is high, the post-injection amount is canceled or reduced.

[0009] Preferably, the cross-cycle combustion state vector includes the following parameters: intake pressure, intake temperature, exhaust pressure, single-cycle injection quantity, main injection timing, common rail pressure, CA10, CA50, CA90, peak heat release rate, IMEP, excess air coefficient, cylinder wall thermal state estimate, wall oil film risk estimate, and previous cycle memory coefficient.

[0010] Preferably, when the diesel engine is not equipped with a cylinder pressure sensor, the combustion phase observer is used to input speed fluctuations, injection parameters, intake pressure, exhaust temperature to estimate CA50, and peak heat release rate combustion characteristics, and the observed output values ​​are used to construct a cross-cycle combustion state vector.

[0011] A diesel engine cross-cycle injection control system suitable for strong transient operating conditions, integrated into the diesel engine ECU or a separate external controller, includes: a data acquisition module for acquiring sensor signals of cylinder pressure, crankshaft angle, engine speed, intake pressure, rail pressure, and injection command; a combustion feature extraction module for calculating the heat release rate and extracting SOC, CA10, CA50, and IMEP combustion features based on cylinder pressure data; a cross-cycle state estimation module for fusing current and previous cycle parameters to construct a cross-cycle combustion state vector and estimating cylinder wall thermal state and oil film risk; a strong transient discrimination module for calculating the strong transient degree index and cross-cycle memory coefficient to identify the transient level of the operating condition; a combustion phase prediction module for predicting the combustion phase, fuel richness, and oxidation risk of the next cycle; an injection command generation module for outputting correction commands for main injection timing, rail pressure, pre-injection / post-injection, and injection quantity change rate; and a closed-loop update module for updating model parameters and memory coefficients based on the actual combustion results of the next cycle.

[0012] When the program is executed by the processor, it implements the diesel engine cross-cycle injection control method applicable to strong transient operating conditions as described above.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention constructs a cross-cycle state vector and memory coefficient that includes multi-cycle combustion, intake, fuel injection, cylinder wall thermal state and oil film risk, overcoming the shortcomings of traditional quasi-steady-state control that ignores inter-cycle coupling and memory effects.

[0014] This invention uses a weighted calculation of the strong transient intensity index and memory coefficient to divide the control mode into three levels for adaptive switching, avoiding insufficient or excessive correction. At the same time, through the coordinated correction of main injection timing, rail pressure, pre-injection / post-injection and injection quantity change rate, it effectively suppresses CA50 shift, extended combustion duration and insufficient oxidation in the later stage without simply limiting the injection quantity, ensuring transient torque response and pressure rise rate constraints.

[0015] This invention predicts the risk of the next cycle in advance based on the combustion phase prediction model and adopts feedforward pre-response control to significantly shorten the control lag.

[0016] This invention supports both collecting measured combustion characteristics based on cylinder pressure sensors and estimating parameters on models without cylinder pressure sensors through a combustion phase observer. It is compatible with both high- and low-cost diesel engines and can be superimposed on existing ECU steady-state injection MAPs, with low development costs.

[0017] This invention employs a closed-loop update mechanism that iteratively optimizes parameters and memory coefficients based on actual combustion response. It can adapt to changes in altitude, ambient temperature, and operating conditions caused by engine aging, ensuring stable control accuracy over long-term operation. Attached Figure Description

[0018] Figure 1 This is a flowchart of the cross-cycle fuel injection control method of the present invention.

[0019] Figure 2 This is a schematic diagram of the overall structure of the cross-cycle fuel injection control system applicable to strong transient conditions according to the present invention.

[0020] Figure 3 This is a schematic diagram illustrating the relationship between cross-cycle state memory and next cycle fuel injection control in this invention.

[0021] Figure 4 This is a schematic diagram of the multiple injection timing correction under strong transient conditions according to the present invention. Detailed Implementation

[0022] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the scope of protection of the invention. Example

[0023] like Figure 1 and Figure 2As shown, this embodiment is applied to a heavy-duty high-pressure common-rail vehicle diesel engine. The hardware is equipped with a cylinder pressure sensor, a crank angle sensor, an intake pressure / temperature sensor, and a rail pressure sensor. The entire control system is integrated into the original vehicle ECU, corresponding to the Figure 2 overall architecture of the modular control system shown.

[0024] Step S101: The ECU collects operating data of the n-th working cycle cycle by cycle, including in-cylinder pressure, crank angle, real-time rotational speed, driver torque demand, intake pressure, intake temperature, exhaust pressure, common rail pressure, and main injection timing.

[0025] Step S102: Perform synchronous filtering processing on the cylinder pressure signal and various sensor data, segment the data according to the engine working cycle, correct the piston top dead center (TDC) deviation, and output the complete pressure-crank angle sequence, intake and exhaust boundary parameters, and basic fuel injection parameters of the n-th cycle.

[0026] Step S103: Calculate the apparent heat release rate and cumulative heat release based on the cylinder pressure-crank angle sequence, and extract combustion characteristic parameters: SOC, CA10, CA50, CA90, combustion duration CD, peak heat release rate HRRmax, maximum in-cylinder pressure Pmax, maximum pressure rise rate dP / dCAmax, and IMEP.

[0027] As shown in Figure 3 , all combustion characteristics, intake air and fuel injection parameters of the current n-th cycle and the previous 1 historical cycle are fused, and the cylinder wall thermal state estimation Twall, wall oil film risk mfilm, and the previous cycle memory coefficient M(n-1) are superimposed to construct a cross-cycle combustion state vector Xn. This vector participates in the construction of subsequent cross-cycle memory prediction and the fuel injection control relationship of the next cycle.

[0028] Step S105: Calculate the strong transient degree index It and the cross-cycle memory coefficient Mn by weighting based on the vector Xn; three levels of thresholds are obtained through bench calibration: a first threshold T1, a second threshold T2, and a third threshold T3, which satisfy T3≤T1<T2; when T1<It≤T2, the control system enters the transient pre-response mode; when It>T2, the control system enters the strong compensation mode; when It drops from a state higher than T3 to It<T3, the control system switches to the recovery correction mode. Three types of combustion deterioration risks for the next cycle, namely CA50 retarding, over-limit pressure rise rate, and insufficient late oxidation, are identified synchronously.

[0029] The above weighting coefficients a1~a5 and b1~b5 can be determined by bench test calibration or system identification methods according to the model of the target diesel engine, emission regulation requirements and power performance targets, which are conventional technical means in the art; for example, the value range of each coefficient can be set between 0.05 and 2.5, and the weight distribution can be adjusted according to the focus of transient control.

[0030] Step S106: Using a linear regression combustion phase prediction model, input the cross-cycle state vector Xn to predict the SOC, CA50, combustion duration, fuel-rich zone risk Rrich, and late oxidation capacity index Ioxi for the (n+1)th cycle.

[0031] Step S107: As Figure 4 As shown, the main injection timing correction ΔSOI is calculated based on the difference between the predicted CA50 and the target optimal CA50, the intake pressure lag difference, and the weighted average of the cross-cycle memory coefficient. If the predicted CA50 lags and the maximum pressure rise rate does not exceed the limit, the main injection timing is advanced or the pre-injection quantity is increased. If the predicted pressure rise rate exceeds the limit, the injection advance angle is reduced, the pre-injection interval is adjusted, and the single change range of the injection quantity is limited. Simultaneously, the injection pressure correction ΔPrail is calculated based on the fuel richness risk level. When the fuel richness risk is high and the rail pressure system has adjustment margin, the rail pressure is increased. The post-injection parameters are adjusted based on the soot risk level and the subsequent oxidation capacity index. When the soot risk is high and Ioxi meets the standard, a small proportion of post-injection is increased; when the oil film risk mfilm is too high, post-injection is canceled. Finally, all corrections for main injection timing, rail pressure, injection quantity change rate, and pre-injection / post-injection are summarized to form the following... Figure 4 The multi-jet timing correction strategy is shown.

[0032] Step S108: The correction amount is superimposed on the original vehicle steady-state injection MAP basic command, and the injection drive module is sent to execute the injection; after the (n+1)th cycle of combustion is completed, the cylinder pressure and combustion characteristics are collected again, and the cross-cycle state vector, memory coefficient Mn and prediction model weighting coefficient are updated to form a permanent cycle-by-cycle closed loop iteration.

[0033] Typical operating condition: The diesel engine is rapidly loaded from low load to high load, the fuel injection demand rises rapidly, and the intake boost lags. It is calculated that It > T1, so it enters the transient pre-response mode, moderately advances the main injection, and limits the fuel injection rate. After 2-3 cycles, the air-fuel ratio decreases, CA50 continues to shift backward, and Mn increases. It > T2, so it switches to the strong compensation mode, simultaneously increases the rail pressure, and matches a small amount of post-injection. When the intake pressure recovers and CA50 returns to the target range, It falls back to below T3, so it enters the recovery correction mode, gradually reduces all fuel injection corrections, and returns to steady-state calibration control. Example

[0034] This embodiment is adapted to small off-road high-pressure common rail diesel engines. The engine does not have a cylinder pressure sensor; combustion characteristics are estimated using a combustion phase observer. The overall control flow of this solution is similar to... Figure 1 The flowcharts shown are completely identical.

[0035] The only difference lies in step S103: the cylinder pressure heat release rate calculation process is cancelled, and the engine speed fluctuation, crankshaft instantaneous angular velocity, real-time fuel injection parameters, intake pressure, intake temperature, and exhaust temperature are input into the combustion phase observer. Combined with the bench calibration model, the equivalent SOC, CA50, HRRmax, and IMEP are estimated. The virtual combustion characteristics output by the observer directly replace the measured cylinder pressure combustion characteristics, are input into step S104 to construct the cross-cycle combustion state vector Xn, and are then further substituted into... Figure 3 The cross-cycle state memory logic is shown; subsequent calculations of strong transient exponents, memory coefficients, combustion phase prediction, and multi-parameter collaborative correction of fuel injection (such as...) in S105-S108 are also included. Figure 4 As shown in the figure, the closed-loop update logic is exactly the same as in Example 1. This solution eliminates the hardware cost of the cylinder pressure sensor, sacrificing only a small amount of combustion characteristic estimation accuracy, while still achieving all functions such as strong transient condition pre-response, graded compensation, and multi-injection parameter collaborative correction. Example

[0036] In advance, a lookup mapping table is established using 3D CFD simulation and engine bench calibration to determine the risk levels of the fuel-rich zone (Rrich), the risk of the fuel film on the combustion wall (mfilm), and the late-stage oxidation capacity index (Ioxi). The mapping inputs include fuel injection quantity, fuel injection pressure, main injection timing, intake pressure, engine speed, EGR rate, and the CA50 offset of the previous cycle. During control operation, the ECU looks up Rrich, mfilm, and Ioxi in real time for Mn weighted calculation, rail pressure correction, and switching of the post-injection strategy. When Rrich is high, a combination of increasing rail pressure, moderately advancing the main injection, and limiting the slope of fuel injection quantity change is prioritized. When the mfilm value is large, the post-injection quantity is directly suppressed. When Ioxi is low, further delaying the combustion phase is prohibited, and intake lag correction is prioritized to improve the air-fuel mixture.

[0037] like Figure 2 As shown, the entire controller consists of seven integrated modules, all of which are programmed into the ECU: ① Data acquisition module: Acquires sensor signals for cylinder pressure, crankshaft angle, speed, intake air, rail pressure, and fuel injection commands; ② Combustion Feature Extraction Module: Calculates heat release rate and extracts SOC, CA10, CA50, and IMEP combustion features; ③ Cross-loop state estimation module: Concatenates current and historical loop parameters to generate a state vector, and estimates Twall and mfilm, such as... Figure 3 The logic shown establishes relationships between loops; ④ Strong transient discrimination module: Calculates It and Mn to determine the three-level control mode; ⑤ Combustion Phase Prediction Module: Outputs the combustion phase, fuel richness, and oxidation risk for the next cycle; ⑥ Injection command generation module: Outputs main injection timing, rail pressure, pre-injection / post-injection, and injection quantity change rate correction commands, and executes them as follows: Figure 4 The timing correction strategy shown; ⑦ Closed-loop update module: Iterates the model weighting coefficients and memory coefficients based on the actual combustion data of the next cycle.

[0038] A Flash storage medium is provided, in which control program code is embedded; when the program is called and executed by the ECU processor, all control steps S101-S108 described in Embodiment 1 or Embodiment 2 are fully executed, and fuel injection correction commands are output in real time; the storage medium can be independently installed inside the ECU, or it can be used as an external storage chip to communicate with the vehicle controller.

[0039] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A cross-cycle fuel injection control method suitable for strong transient operating conditions, applied to a high-pressure common rail diesel engine, characterized in that, Includes the following steps: S101. Collect the operating data of the diesel engine during the nth working cycle. The operating data includes cylinder pressure, crankshaft angle, speed, torque demand, intake pressure, intake temperature, rail pressure, and injection timing. S102. Synchronously filter and cycle-divide the cylinder pressure signal and operating data to obtain the nth cycle pressure-crankshaft angle sequence and intake / exhaust and fuel injection boundary parameters; S103. Calculate the apparent heat release rate and cumulative heat release based on cylinder pressure data, and extract combustion characteristic parameters such as SOC, CA10, CA50, CA90, combustion duration, IMEP, and maximum pressure rise rate. S104. By integrating the combustion characteristics, intake state, injection state, and hot / oil film state estimates of the nth cycle and at least one preceding cycle, a cross-cycle combustion state vector is constructed. S105. Calculate the strong transient degree index and the cross-cycle memory coefficient based on the cross-cycle combustion state vector to identify the current transient level and the risk of combustion deterioration in the next cycle. S106. By using a combustion phase prediction model and combining it with cross-cycle combustion state vectors, the combustion phase, fuel-rich zone risk, and post-oxidation capacity of the (n+1)th cycle are predicted. S107. Combining the predicted combustion phase with the target CA50, generate the injection correction amount for the (n+1)th cycle. The injection correction amount includes the main injection timing correction amount, the injection pressure correction amount, the injection quantity change rate limit amount, and the pre-injection / post-injection parameter correction amount. S108. Send the fuel injection command with the superimposed correction amount to the fuel injection drive module; After the (n+1)th cycle ends, the actual combustion response is collected to update the cross-cycle state vector and model parameters, forming a cycle-by-cycle closed-loop control.

2. The cross-cycle fuel injection control method for strong transient operating conditions according to claim 1, characterized in that, The strong transient intensity index is obtained by weighted summation of the injection quantity change amplitude, intake pressure target deviation, air-fuel ratio critical difference, CA50 offset, and cross-cycle memory coefficient; the cross-cycle memory coefficient is calculated by weighted summation of CA50 cycle offset, combustion duration change, wall oil film risk, intake pressure hysteresis, and late-stage oxidation loss; based on the relationship between the strong transient intensity index and the three preset thresholds, the control system switches between different control modes: When the strong transient intensity index is greater than the first threshold and not greater than the second threshold, the transient pre-response mode is entered. When the index is greater than the second threshold, it enters a strong compensation mode; When the index falls from above the third threshold to below the third threshold, it enters the recovery correction mode; Wherein, the first threshold is less than the second threshold, and the third threshold is less than or equal to the first threshold.

3. The cross-cycle fuel injection control method for strong transient operating conditions according to claim 2, characterized in that, The main injection timing correction is calculated by weighting the difference between the predicted CA50 and the target optimal CA50, the intake pressure lag difference, and the cross-cycle memory coefficient. If the predicted CA50 is shifted backward relative to the target CA50 and the maximum pressure rise rate does not exceed the limit, the main injection timing is advanced or the pre-injection amount is increased. If the predicted pressure rise rate exceeds the limit, the main injection advance amount is reduced, the pre-injection interval is adjusted, or the rate of change of the injection amount per cycle is limited.

4. The cross-cycle fuel injection control method for strong transient operating conditions according to claim 1, characterized in that, Adjust the injection pressure correction based on the risk level of the fuel-rich zone. When the fuel-rich risk is high and the rail pressure system has an adjustment margin, increase the rail pressure. Adjust the post-injection parameters based on the subsequent oxidation capacity index: when the carbon soot risk is high and the oxidation capacity meets the standard, use a small proportion of post-injection to enhance in-cylinder oxidation; when the wall oil film risk is high, cancel or reduce the post-injection amount.

5. The cross-cycle fuel injection control method for strong transient operating conditions according to claim 1, characterized in that, The cross-cycle combustion state vector includes the following parameters: intake pressure, intake temperature, exhaust pressure, single-cycle injection quantity, main injection timing, common rail pressure, CA10, CA50, CA90, peak heat release rate, IMEP, excess air coefficient, cylinder wall thermal state estimate, wall oil film risk estimate, and previous cycle memory coefficient.

6. The cross-cycle fuel injection control method for strong transient operating conditions according to claim 1, characterized in that, When the diesel engine is not equipped with a cylinder pressure sensor, the combustion phase observer is used to input speed fluctuations, injection parameters, intake pressure, and exhaust temperature to estimate CA50 and peak heat release rate combustion characteristics, and the observed output values ​​are used to construct a cross-cycle combustion state vector.

7. A diesel engine cross-cycle fuel injection control system suitable for strong transient operating conditions, integrated into the diesel engine ECU or a separate external controller, characterized in that, include: The data acquisition module is used to collect sensor signals for cylinder pressure, crankshaft angle, speed, intake pressure, rail pressure, and fuel injection command. The combustion feature extraction module is used to calculate the heat release rate and extract SOC, CA10, CA50, and IMEP combustion features based on cylinder pressure data. The cross-cycle state estimation module is used to fuse current and previous cycle parameters to construct a cross-cycle combustion state vector and estimate cylinder wall thermal state and oil film risk; the strong transient discrimination module is used to calculate the strong transient degree index and cross-cycle memory coefficient to identify the transient level of the operating condition; the combustion phase prediction module is used to predict the combustion phase, fuel richness and oxidation risk of the next cycle; the injection command generation module is used to output the main injection timing, rail pressure, pre-injection / post-injection, and injection quantity change rate correction commands; the closed-loop update module is used to update the model parameters and memory coefficient according to the actual combustion results of the next cycle.

8. A computer-readable storage medium having a computer-executable program stored thereon, characterized in that, When the program is executed by the processor, it implements the cross-cycle fuel injection control method for strong transient operating conditions as described in any one of claims 1 to 6.

Images