Combustion optimization control system and method for dual-fuel micro-jet engine

Abstract

The invention discloses a combustion optimization control system and method for a dual-fuel micro-jet engine, and relates to the technical field of dual-fuel engines, the combustion optimization control system comprises a working condition monitoring module, a fuel injection compensation control module and a fuel execution module, the fuel injection compensation control module is used for dynamically correcting fuel injection parameters based on real-time working conditions, and the fuel execution module is used for dynamically correcting fuel injection parameters based on real-time working conditions. The fuel injection compensation control module comprises an injection parameter prediction unit, a compensation amount calculation unit and an injection correction unit. According to the invention, a fuel injection compensation control module is installed, a support vector machine model is adopted to predict reference parameters and is combined with dynamic compensation of a self-adaptive controller, millisecond-level fuel injection correction is realized, in the fuel-gas mode switching process, the combustion phase offset is controlled within + / -0.8 degree CA, the rotating speed fluctuation rate is reduced to be below 5%, and the fuel-gas mode switching is realized. And the torque sudden change and flameout risks are effectively eliminated.

Description

Technical Field

[0001] The present invention relates to the technical field of dual-fuel engines, and in particular to a combustion optimization control system and method for a dual-fuel micro-injection engine. Background Art

[0002] Dual-fuel micro-injection engines, a key development in clean power technology, significantly reduce nitrogen oxide and particulate matter emissions by utilizing a combustion mode that uses natural gas as the primary fuel and a trace amount of diesel as a pilot. Existing combustion optimization control systems typically control the pilot fuel injection process based on fixed parameter mapping tables, which can maintain combustion stability under steady-state conditions. However, when the engine switches from fuel to gas mode, traditional open-loop control strategies struggle to adapt to dynamic operating conditions due to factors such as sudden changes in intake flow velocity, changes in in-cylinder turbulence intensity, and fluctuations in residual exhaust gas. This leads to mismatches in pilot fuel injection quantity and timing. Experimental studies have shown that when these mismatches exceed ±0.1 mg / cyc or ±0.5°CA, combustion phase shifts exceeding 3°CA, causing speed fluctuations to exceed 15% or even combustion interruption, severely hindering the engineering application of dual-fuel technology.

[0003] Although some studies have attempted to introduce cylinder pressure feedback control in recent years, due to the strong nonlinear characteristics of the combustion process and the delay in sensor signal transmission, the response time of the conventional PID control algorithm in the transient process of mode switching is still greater than 20ms, and it is unable to effectively suppress millisecond-level speed mutations. Therefore, it is urgent to develop a real-time dynamic compensation mechanism with working condition adaptability to break through the technical bottleneck of mode switching instability.

[0004] Patent CN105971746B discloses a control system for a medium-speed micro-injection ignition dual-fuel engine for ships. The patent implements multiple injection control and stops the injection mode of a certain channel; the same control unit is used to control the fuel supply of the engine, facilitating the synchronous and precise control of the fuel mode switching process, achieving a smooth transition of the engine, and optimizing the operating parameters of the dual-fuel engine.

[0005] The aforementioned patent consists of an engine operation monitoring unit, a safety protection unit, a fuel injection and speed control unit, and a parameter modification unit, all installed in a control box. Each unit is connected via a redundant CAN bus to enable operational data exchange and logical parameter modification. The system also interacts with a human-machine interface via a Modbus bus for display parameter interaction. The system can implement multiple injection control and stop injection modes in specific channels. The same control unit is used to control the engine's fuel supply, facilitating synchronous and precise control of fuel mode switching, achieving smooth engine transitions, and optimizing dual-fuel engine operating parameters. However, the aforementioned patent still has shortcomings in terms of adaptive real-time dynamic compensation.

[0006] To this end, the present application proposes a combustion optimization control system and method for a dual-fuel micro-injection engine that is capable of adaptive real-time dynamic compensation. Summary of the Invention

[0007] The object of the present invention is to provide a combustion optimization control system and method for a dual-fuel micro-injection engine to solve the technical problems raised in the above background technology.

[0008] To achieve the above objectives, the present invention provides the following technical solutions: a combustion optimization control system for a dual-fuel micro-injection engine, comprising an operating condition monitoring module, a pilot oil compensation control module, and a fuel execution module, wherein the pilot oil compensation control module is used to dynamically modify the pilot oil injection parameters based on the real-time operating conditions;

[0009] The pilot oil compensation control module includes: an injection parameter prediction unit, a compensation amount calculation unit, and an injection correction unit;

[0010] The injection parameter prediction unit is configured with an offline trained support vector machine model and an online self-learning algorithm, and outputs a pilot fuel quantity reference value and an injection timing reference value;

[0011] The compensation calculation unit integrates an adaptive controller based on Lyapunov stability, and uses the angular acceleration deviation fed back by the crankshaft position sensor and the combustion phase offset fed back by the in-cylinder pressure sensor to calculate the pilot fuel amount compensation coefficient and the injection timing compensation;

[0012] The injection correction unit has a built-in microsecond time window correction algorithm, which performs real-time superposition correction on the reference value output by the injection parameter prediction unit according to the compensation coefficient and compensation amount output by the compensation amount calculation unit.

[0013] Preferably, the operating condition monitoring module is connected to the pilot oil compensation control module via a high-speed CAN bus for collecting engine operating status data;

[0014] The working condition monitoring module includes: crankshaft position sensing unit, cylinder pressure sensing unit and air-fuel ratio detection unit;

[0015] Integrated magnetoelectric pulse generator and digital signal processor to output crankshaft angle signal and speed fluctuation characteristic value in real time;

[0016] The in-cylinder pressure sensing unit integrates a piezoelectric force-sensitive element and a charge amplifier circuit, and outputs cylinder pressure change rate data in real time through the cylinder head mounting hole;

[0017] The air-fuel ratio detection unit uses a wide-range oxygen sensor, which is arranged in the front section of the exhaust pipe to detect the actual value of the excess air coefficient.

[0018] Preferably, the fuel execution module is connected to the pilot oil compensation control module via a PWM signal line of the electronic control unit ECU, for accurately executing the fuel injection instruction;

[0019] The fuel execution module includes: a pilot oil injection unit, a fuel gas injection unit and a drive circuit unit;

[0020] The pilot oil injection unit uses a piezoelectric crystal micro-injector with a response time of ≤0.1ms, and performs micro-fuel injection based on the corrected pilot oil reference value;

[0021] The gas injection unit is equipped with a high-speed electromagnetic natural gas injection valve, which controls the natural gas injection pulse width based on the electronic control unit ECU instructions;

[0022] The drive circuit unit integrates an H-bridge drive chip and an overcurrent protection circuit to convert the PWM signal into the injection valve drive current.

[0023] Preferably, the system is further designed with a combustion stability assessment module, which is connected to the pilot oil compensation control module via a high-speed CAN bus and is used to determine the combustion instability state in real time and trigger the compensation mechanism;

[0024] The combustion stability assessment module includes: an instability feature extraction unit, a mode switching judgment unit and a historical data storage unit;

[0025] The instability feature extraction unit integrates a fast Fourier transform hardware accelerator to calculate the pressure oscillation energy value in the 0.5-5kHz frequency range based on the output signal of the in-cylinder pressure sensing unit;

[0026] The mode switching decision unit integrates a dual-threshold comparator circuit and outputs an activation signal when the speed fluctuation characteristic value is greater than 250 rpm or the pressure oscillation energy value is greater than 200 J / deg;

[0027] The historical data storage unit uses FRAM non-volatile memory to record the compensation parameters and results of each mode switching process.

[0028] Preferably, the combustion stability assessment module further comprises an online calibration interface;

[0029] The online calibration interface includes: calibration parameter input port and self-learning feedback port;

[0030] The calibration parameter input port supports RS485 communication protocol and receives the speed fluctuation threshold and pressure oscillation energy threshold input by external calibration equipment;

[0031] The self-learning feedback port transmits the compensation effect data in the historical data storage unit back to the support vector machine model of the pilot oil compensation control module.

[0032] Preferably, the system is further designed with a fault tolerance module, which is connected to the pilot oil compensation control module via an SPI bus to ensure the safety of the injection process;

[0033] The fault tolerance module includes: a signal redundancy check unit, an injection safety latch unit and a fault code generation unit;

[0034] The signal redundancy check unit is equipped with a phase difference detection circuit and a window comparator to compare the phase difference between the crankshaft and camshaft position sensor pulse signals. When the deviation is greater than 0.5°CA, an abnormality flag is triggered.

[0035] The injection safety latch unit integrates a numerical comparator and a parameter switch, which switches to the preset fixed injection parameters when the pilot fuel quantity compensation coefficient is greater than 15%;

[0036] The fault code generation unit is configured with a diagnostic protocol stack and a message generator, and sends fault codes associated with historical data indexes to the vehicle controller.

[0037] Preferably, the adaptive controller of the compensation amount calculation unit includes:

[0038] The phase offset input circuit is connected to the charge amplifier output terminal of the in-cylinder pressure sensor and receives the combustion phase offset voltage signal;

[0039] The angular acceleration conversion circuit is connected to the digital output interface of the crankshaft position sensor unit to convert the crankshaft pulse signal into a digital angular acceleration;

[0040] The gain parameter generator uses FPGA programmable logic devices to generate time-varying gain parameters based on the excess air coefficient change rate, speed fluctuation characteristic value and angular acceleration;

[0041] The compensation amount operator integrates a multiplier and an adder circuit to output a digital signal of the pilot fuel amount compensation coefficient and the injection timing compensation amount.

[0042] Preferably, the gain parameter generator performs the following operations:

[0043] (1) Using a 12-bit ADC to collect the excess air coefficient change rate analog signal of the air-fuel ratio detection unit;

[0044] (2) Reading the digital value of the speed fluctuation characteristic value from the SPI interface of the crankshaft position sensor unit;

[0045] (3) receiving the second-order derivative value of the crankshaft angular acceleration output by the angular acceleration conversion circuit;

[0046] (4) Calculation of time-varying gain parameters is completed in FPGA programmable logic devices:

[0047]

[0048] Where: α, β and γ are calibration constants, determined by bench calibration test, dλ / dt is the rate of change of excess air coefficient, Δn is the speed fluctuation characteristic value, d 2 θ / dt 2 is the crankshaft angular acceleration, K p is the proportional gain parameter, K i is the integral gain parameter, K d is the differential gain parameter;

[0049] Preferably, the method comprises the following steps:

[0050] S1. System initialization configuration: The operator burns the support vector machine model parameters and gain coefficients into the FRAM memory through an external calibration device;

[0051] S2, sensor calibration start: use a dedicated calibration tool to perform zero drift correction on the in-cylinder pressure sensor;

[0052] S3, mode switching activation: the operator triggers the mode switching command from fuel oil to gas on the control panel;

[0053] S4. Compensation result recording: manually read the compensation effect report of the historical data storage unit and back it up to an external device.

[0054] Preferably, the method further comprises the following steps:

[0055] S11. The operator connects the USB port of the calibration device to the RS485 port of the online calibration interface, selects the pre-stored calibration constant file through the calibration software interface and performs the burning operation;

[0056] S21. Connect the charge amplifier output of the in-cylinder pressure sensor to the analog input channel of the calibration tool. The operator adjusts the zero point potentiometer to an error of less than ±0.5% based on the offset displayed by the calibration tool.

[0057] S31. Select the "Dual Fuel Switch" button on the engine control unit (ECU) operation interface. When the instrument panel displays a speed fluctuation > 250 rpm, manually confirm the activation of pilot oil compensation.

[0058] S41. Connect to the host computer through the data export interface of the FRAM memory, and the operator executes the data reading instruction and saves the compensation parameter log file.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] 1. This invention incorporates a pilot oil compensation control module, employing a support vector machine model to predict reference parameters and combining it with dynamic compensation via an adaptive controller to achieve millisecond-level pilot oil injection correction. During fuel-to-gas mode switching, the combustion phase offset is controlled within ±0.8°CA, and the speed fluctuation rate is reduced to below 5%, effectively eliminating the risk of torque mutation and flameout.

[0061] 2. This invention utilizes a working condition monitoring module that integrates a magnetoelectric encoder, a piezoelectric ceramic sensor, and a wide-band oxygen sensor for simultaneous multi-source data acquisition. It outputs speed fluctuation characteristic values, cylinder pressure change rate, and measured values ​​of excess air coefficient in real time, providing a high-precision input reference for dynamic compensation.

[0062] 3. This invention incorporates a combustion stability assessment module that uses a fast Fourier transform hardware accelerator to calculate pressure oscillation energy in real time. This, combined with a dual-threshold comparator circuit, enables instability prediction. When speed fluctuations exceed 250 rpm or pressure oscillation energy exceeds 200 J / deg, a compensation mechanism is triggered within 3ms, mitigating instability risks by more than 85% in advance.

[0063] 4. The present invention is equipped with a fault-tolerant module, uses a phase difference detection circuit to verify the consistency of crankshaft and camshaft signals, and cooperates with a numerical comparator to monitor the compensation coefficient in real time. When the ignition fuel quantity compensation coefficient is greater than 15% or the phase deviation is greater than 0.5°CA, it immediately switches to the preset safety parameters and generates a diagnostic code, reducing the misinjection failure rate from the industry average of 5% to below 0.1%. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 Schematic diagram of the workflow of the present invention. DETAILED DESCRIPTION

[0065] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0066] In the description of the present invention, it should be noted that the terms "upper," "lower," "inner," "outer," "front end," "rear end," "both ends," "one end," "the other end," and the like, indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0067] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "installed," "provided with," "connected," etc., should be understood in a broad sense. For example, "connected" may refer to a fixed connection, a detachable connection, or an integral connection; it may refer to a mechanical connection or an electrical connection; it may refer to a direct connection or an indirect connection through an intermediate medium; it may refer to internal communication between two components. Those skilled in the art will be able to understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0068] See also Figure 1 One embodiment of the present invention provides a combustion optimization control system for a dual-fuel micro-injection engine. When the engine switches from fuel to gas mode, the crankshaft position sensor of the operating condition monitoring module collects speed fluctuation characteristic values ​​in real time, the in-cylinder pressure sensor simultaneously outputs cylinder pressure change rate data, and the wide-band oxygen sensor detects the actual measured value of the excess air coefficient at the front end of the exhaust pipe. This data is transmitted to the pilot oil compensation control module via a high-speed CAN bus.

[0069] Within the pilot oil compensation control module, the injection parameter prediction unit invokes an offline-trained support vector machine model, taking the received speed fluctuation characteristic values, cylinder pressure change rate, and measured values ​​of the excess air coefficient as inputs, and outputs a pilot oil quantity reference value and an injection timing reference value. The compensation calculation unit simultaneously activates an adaptive controller based on Lyapunov stability, combining the angular acceleration deviation feedback from the crankshaft position sensor and the combustion phase offset feedback from the in-cylinder pressure sensor to calculate the pilot oil quantity compensation coefficient and injection timing compensation. The injection correction unit activates a microsecond time window correction algorithm, adding the compensation coefficient and compensation amount output by the compensation calculation unit to the reference value within a 50μs time window to generate the final corrected pilot oil injection parameters.

[0070] Furthermore, after the engine control unit (ECU) receives the mode switch command, the operating condition monitoring module's magnetoelectric encoder continuously monitors the crankshaft angle signal. When a speed fluctuation characteristic exceeds a preset threshold, the piezoelectric ceramic sensor immediately collects the current cylinder pressure change rate data. The wide-band oxygen sensor simultaneously locks onto the measured excess air coefficient. Based on historical training data, the support vector machine model maps a pilot fuel quantity baseline of 0.85 mg / cyc and an injection timing baseline of -12.5°CA. Simultaneously, the adaptive controller calculates a pilot fuel quantity compensation factor of +8% and an injection timing compensation of +0.6°CA based on the real-time angular acceleration deviation of 0.3 rad / s² and the combustion phase offset of 2.2°CA. The injection correction unit uses a hardware adder to perform the superposition calculation of the baseline and compensation values ​​within a microsecond time window, outputting the corrected pilot fuel injection parameters to the fuel execution module.

[0071] See also Figure 1 The present invention provides an embodiment of a combustion optimization control system for a dual-fuel micro-injection engine. The crankshaft position sensing unit of the operating condition monitoring module captures flywheel sprocket pulse signals via a magnetoelectric encoder. A digital signal processor converts these signals into digital crankshaft angle values ​​and speed fluctuation characteristic values. The piezoelectric ceramic sensor of the in-cylinder pressure sensing unit converts mechanical vibrations at the cylinder head bolt holes into charge signals, which are then output via a charge amplifier circuit as an analog cylinder pressure change rate signal. The wide-band oxygen sensor of the air-fuel ratio detection unit detects exhaust oxygen concentration and outputs the measured excess air coefficient value using a calibration curve. These data are transmitted to the pilot oil compensation control module via a high-speed CAN bus.

[0072] The fuel execution module receives the corrected pilot oil injection parameters: the pilot oil injection unit's piezoelectric crystal micro-injector opens its needle valve within 0.1ms, injecting a trace amount of diesel at a rate of 0.918mg / cyc. The fuel gas injection unit's high-speed electromagnetic natural gas injection valve opens with a 5.2ms pulse width based on ECU instructions, injecting the main fuel, natural gas. The drive circuit unit's H-bridge driver chip converts the PWM signal into a 12V / 2A drive current, and the overcurrent protection circuit monitors the risk of current exceeding the limit in real time.

[0073] Furthermore, when the engine enters the transient mode switching condition, the digital signal processor of the crankshaft position sensing unit collects flywheel sprocket pulses at a resolution of 0.1°CA. When the change rate of the interval between adjacent pulses exceeds 3%, it generates a speed fluctuation characteristic value of 252 rpm. The charge amplifier of the in-cylinder pressure sensing unit amplifies the 200pC charge signal output by the piezoelectric ceramic sensor into a ±5V analog voltage, which is then converted by the ADC into a digital cylinder pressure change rate of -0.8MPa / °CA. The wide-band oxygen sensor calculates the exhaust oxygen concentration using the Nernst equation and outputs a measured excess air coefficient value of 1.05. The driver circuit unit of the fuel actuator module analyzes the PWM duty cycle signal sent by the pilot oil compensation control module. The H-bridge driver chip outputs a 12V voltage to drive the piezoelectric crystal micro injector, which opens at a crankshaft angle position of -11.9°CA for 87μs to achieve an injection of 0.918mg / cyc. The high-speed electromagnetic natural gas injection valve is synchronously controlled to open for 5.2ms to complete the main gas injection.

[0074] See also Figure 1 The present invention provides an embodiment of a combustion optimization control system for a dual-fuel micro-injection engine. The instability feature extraction unit of the combustion stability assessment module receives cylinder pressure change rate data output by the in-cylinder pressure sensing unit, performs spectrum analysis on the 0.5-5kHz frequency band signal using an integrated fast Fourier transform hardware accelerator, and calculates the pressure oscillation energy value. The mode switching decision unit compares the real-time speed fluctuation feature value and the pressure oscillation energy value with a preset threshold value. When any parameter exceeds the limit, an activation signal is sent to the pilot oil compensation control module via the high-speed CAN bus. The historical data storage unit uses a FRAM memory to record the current compensation parameters and combustion phasing results.

[0075] The calibration parameter input port of the online calibration interface receives the speed fluctuation threshold correction value sent by the external calibration equipment via the RS485 communication protocol, such as adjusting it from 250 rpm to 230 rpm. The self-learning feedback port transmits the compensated post-combustion phase offset recorded in the historical data storage unit, such as -0.7°CA, back to the support vector machine model, triggering the online self-learning algorithm to update the model weights.

[0076] Furthermore, during the engine mode switching process, the fast Fourier transform hardware accelerator of the instability feature extraction unit captures cylinder pressure change rate data at a 100kHz sampling rate. After a 1024-point FFT operation, it extracts the energy peak of 185J / deg at the 4.2kHz frequency point. The dual-threshold comparator circuit of the mode switching decision unit simultaneously detects the speed fluctuation characteristic value of 268rpm, which exceeds the 250rpm threshold and immediately outputs a high-level activation signal to the pilot oil compensation control module. The historical data storage unit writes the current pilot oil quantity compensation coefficient (+8%), injection timing compensation (+0.6°CA), and post-compensation combustion phase offset (-0.7°CA) into the FRAM storage area. The operator adjusts the speed fluctuation threshold from 250rpm to 230rpm via an external calibration device connected to the RS485 port. The self-learning feedback port automatically uploads this compensation data packet to the support vector machine model, which updates the speed weight coefficient from 0.35 to 0.41 through an incremental learning algorithm.

[0077] See also Figure 1 The present invention provides an embodiment of a combustion optimization control system for a dual-fuel micro-injection engine. The signal redundancy check unit of the fault tolerance module compares the pulse rising edge time difference between the crankshaft position sensor and the camshaft position sensor through a phase difference detection circuit. When a phase deviation greater than 0.5°CA is detected, an abnormal flag is triggered. The numerical comparator of the injection safety latch unit monitors the pilot fuel quantity compensation coefficient in real time. When the compensation coefficient is greater than 15%, the system switches to a preset safety parameter of 0.75mg / cyc pilot fuel quantity and -10°CA injection timing. The fault code generation unit generates a CAN message containing a fault type code such as P0300 and a historical data index based on the diagnostic protocol stack and FRAM address 0x5A3 and sends it to the vehicle controller.

[0078] Furthermore, during the engine mode switching stage, the window comparator of the signal redundancy check unit continuously monitors the rising edge time difference of the crankshaft and camshaft sensor pulses. When the crankshaft sensor pulse is delayed by 0.8ms due to flywheel tooth clearance, corresponding to a phase deviation of 0.6°CA, the window comparator outputs an over-limit signal to trigger an abnormal flag. The injection safety latch unit simultaneously detects that the ignition fuel quantity compensation coefficient reaches +18%. The numerical comparator immediately outputs a switching instruction, and the parameter switcher forcibly switches the injection parameters from the dynamic compensation value, 1.02mg / cyc, -13.2°CA, to the preset safety parameters, 0.75mg / cyc, -10°CA; the fault code generation unit calls the UDS diagnostic protocol stack to generate a CAN message containing the fault code P0303, misfire in the third cylinder, and the associated historical data storage address 0x5A3, and sends it to the vehicle controller at a baud rate of 500kbps.

[0079] See also Figure 1The present invention provides an embodiment of a combustion optimization control system for a dual-fuel micro-injection engine. The phase offset input circuit of the compensation calculation unit receives the combustion phase offset voltage signal output by the charge amplifier of the in-cylinder pressure sensor, such as -1.2V corresponding to 3°CA. The angular acceleration conversion circuit analyzes the digital pulse signal of the crankshaft position sensor unit and calculates the crankshaft angular acceleration, such as -420 rad / s, through second-order differentiation. 2 ; The gain parameter generator uses FPGA programmable logic devices, and the excess air coefficient change rate is 0.15s -1 , the speed fluctuation characteristic value 240rpm and angular acceleration generate time-varying gain parameters, K p =0.32, K i =0.08, K d =0.15; the compensation operator calculates K through the multiplier p e(t), the adder combines the integral and differential terms, ultimately outputting a pilot fuel quantity compensation factor of +7.2% and an injection timing compensation of +0.5°CA;

[0080] Furthermore, the charge amplifier of the in-cylinder pressure sensor outputs a combustion phase offset analog voltage of -1.2V. After calibration, it corresponds to a 3°CA delay. The 12-bit ADC of the phase offset input circuit quantizes it into a digital value of 3072. The angular acceleration conversion circuit reads the 0.1°CA timestamp data from the SPI interface of the crankshaft position sensor unit and calculates the current angular acceleration of -420rad / s using the second-order central difference method. 2 ; The FPGA of the gain parameter generator performs the following operations: collecting the excess air coefficient change rate of the air-fuel ratio detection unit 0.15s -1 , read the speed fluctuation characteristic value 240rpm, receive the angular acceleration second derivative value -1050rad / s 3 , according to the formula K p =0.0021×0.15=0.32, K i =0.00033×240=0.08, K d =0.00014×(-1050)=-0.15; the multiplier of the compensation arithmetic unit calculates the proportional term 0.32×3°CA=0.96°CA, the integral term 0.08×∫3°CA dt=0.24°CA, and the differential term (-0.15)×(-420)=63, which are then combined into an injection timing compensation amount of +0.5°CA via the adder; the pilot fuel amount compensation coefficient is synchronously outputted as +7.2%;

[0081] The calibration constants α, β, and γ in the gain parameter generator were determined through bench calibration tests. The specific process involved installing the engine on a dynamic dynamometer and connecting the crankshaft position sensor, in-cylinder pressure sensor, and wide-band oxygen sensor of the operating condition monitoring module to the emissions analyzer. An initial parameter set was input to the online calibration interface via the calibration equipment. First, a fuel mode switch was performed under steady-state conditions of 1500 rpm and 25% load. The α value was adjusted until the correlation coefficient between the excess air ratio change rate and the combustion phase offset exceeded 0.95. Then, under transient conditions of a 10 rpm / s load step, the β value was adjusted to ensure that the root mean square error between the speed fluctuation characteristic value and the pilot fuel compensation coefficient was less than 3%. Finally, under -10°C cold start conditions, the γ value was optimized to ensure an angular acceleration closed-loop response delay of less than 2 ms. Finally, α = 0.0021, β = 0.00033, and γ = 0.00014 were determined and written to the non-volatile memory of the gain parameter generator.

[0082] Working Principle: First, the operating condition monitoring module uses the crankshaft position sensing unit's magnetoelectric encoder to collect real-time crankshaft angle signals from the engine's flywheel side and calculate the speed fluctuation characteristic value. The in-cylinder pressure sensing unit's piezoelectric ceramic sensor simultaneously detects pressure vibrations at the cylinder head bolt holes and outputs cylinder pressure change rate data. The air-fuel ratio detection unit's wide-band oxygen sensor measures the exhaust oxygen concentration in the front section of the exhaust pipe and converts it into a measured value for the excess air coefficient. These parameters are transmitted to the pilot oil compensation control module via the high-speed CAN bus.

[0083] Next, the injection parameter prediction unit of the pilot fuel compensation control module calls the offline-trained support vector machine model, inputs the received speed fluctuation characteristic values, cylinder pressure change rate, and measured values ​​of the excess air coefficient into the model, and outputs the pilot fuel quantity reference value and injection timing reference value. The compensation calculation unit simultaneously activates the adaptive controller based on Lyapunov stability, combining the angular acceleration deviation feedback from the crankshaft position sensor and the combustion phase offset feedback from the in-cylinder pressure sensor to calculate the pilot fuel quantity compensation coefficient and injection timing compensation. The injection correction unit activates the microsecond time window correction algorithm, superimposes the compensation value on the reference value within the preset time window, generates the corrected pilot fuel injection parameters, and sends them to the fuel execution module via the PWM signal line of the electronic control unit (ECU).

[0084] Finally, the driving circuit unit of the fuel execution module converts the PWM signal into a driving current. The piezoelectric crystal micro-injector of the pilot oil injection unit performs micro-fuel injection according to the corrected pilot oil quantity reference value. The high-speed electromagnetic natural gas injection valve of the gas injection unit controls the natural gas injection pulse width based on the ECU instruction. The combustion stability evaluation module analyzes the cylinder pressure oscillation energy and speed fluctuation characteristics in real time. When the risk of instability is detected, the compensation mechanism is triggered to update the support vector machine model. At the same time, the fault tolerance module verifies the signal consistency and monitors the over-limit behavior of the compensation coefficient to ensure a safe closed-loop injection process.

[0085] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A combustion optimization control system for a dual-fuel micro-injection engine, characterized by: It includes a working condition monitoring module, a pilot oil compensation control module and a fuel execution module. The pilot oil compensation control module is used to dynamically correct the pilot oil injection parameters based on the real-time working condition; The pilot oil compensation control module includes: an injection parameter prediction unit, a compensation amount calculation unit, and an injection correction unit; The injection parameter prediction unit is configured with an offline trained support vector machine model and an online self-learning algorithm, and outputs a pilot fuel quantity reference value and an injection timing reference value; The compensation calculation unit integrates an adaptive controller based on Lyapunov stability, and uses the angular acceleration deviation fed back by the crankshaft position sensor and the combustion phase offset fed back by the in-cylinder pressure sensor to calculate the pilot fuel amount compensation coefficient and the injection timing compensation; The injection correction unit has a built-in microsecond time window correction algorithm, which performs real-time superposition correction on the reference value output by the injection parameter prediction unit according to the compensation coefficient and compensation amount output by the compensation amount calculation unit.

2. A combustion optimization control system for a dual-fuel micro-injection engine according to claim 1, characterized in that: The operating condition monitoring module is connected to the pilot oil compensation control module via a high-speed CAN bus for collecting engine operating status data; The working condition monitoring module includes: crankshaft position sensing unit, cylinder pressure sensing unit and air-fuel ratio detection unit; Integrated magnetoelectric pulse generator and digital signal processor to output crankshaft angle signal and speed fluctuation characteristic value in real time; The in-cylinder pressure sensing unit integrates a piezoelectric force-sensitive element and a charge amplifier circuit, and outputs cylinder pressure change rate data in real time through the cylinder head mounting hole; The air-fuel ratio detection unit uses a wide-range oxygen sensor, which is arranged in the front section of the exhaust pipe to detect the actual value of the excess air coefficient.

3. The combustion optimization control system for a dual-fuel micro-injection engine according to claim 1, characterized in that: The fuel execution module is connected to the pilot oil compensation control module via a PWM signal line of the electronic control unit ECU for accurately executing the fuel injection command; The fuel execution module includes: a pilot oil injection unit, a fuel gas injection unit and a drive circuit unit; The pilot oil injection unit uses a piezoelectric crystal micro-injector with a response time of ≤0.1ms, and performs micro-fuel injection based on the corrected pilot oil reference value; The gas injection unit is equipped with a high-speed electromagnetic natural gas injection valve, which controls the natural gas injection pulse width based on the electronic control unit ECU instructions; The drive circuit unit integrates an H-bridge drive chip and an overcurrent protection circuit to convert the PWM signal into the injection valve drive current.

4. The combustion optimization control system for a dual-fuel micro-injection engine according to claim 1, characterized in that: The system is also designed with a combustion stability assessment module, which is connected to the pilot oil compensation control module via a high-speed CAN bus and is used to determine the combustion instability state in real time and trigger the compensation mechanism; The combustion stability assessment module includes: an instability feature extraction unit, a mode switching judgment unit and a historical data storage unit; The instability feature extraction unit integrates a fast Fourier transform hardware accelerator to calculate the pressure oscillation energy value in the 0.5-5kHz frequency range based on the output signal of the in-cylinder pressure sensing unit; The mode switching decision unit integrates a dual-threshold comparator circuit and outputs an activation signal when the speed fluctuation characteristic value is greater than 250 rpm or the pressure oscillation energy value is greater than 200 J / deg; The historical data storage unit uses FRAM non-volatile memory to record the compensation parameters and results of each mode switching process.

5. The combustion optimization control system for a dual-fuel micro-injection engine according to claim 4, characterized in that: The combustion stability assessment module further comprises an online calibration interface; The online calibration interface includes: calibration parameter input port and self-learning feedback port; The calibration parameter input port supports RS485 communication protocol and receives the speed fluctuation threshold and pressure oscillation energy threshold input by external calibration equipment; The self-learning feedback port transmits the compensation effect data in the historical data storage unit back to the support vector machine model of the pilot oil compensation control module.

6. The combustion optimization control system for a dual-fuel micro-injection engine according to claim 1, characterized in that: The system is also designed with a fault tolerance module, which is connected to the pilot oil compensation control module via the SPI bus to ensure the safety of the injection process; The fault tolerance module includes: a signal redundancy check unit, an injection safety latch unit and a fault code generation unit; The signal redundancy check unit is equipped with a phase difference detection circuit and a window comparator to compare the phase difference between the crankshaft and camshaft position sensor pulse signals. When the deviation is greater than 0.5°CA, an abnormality flag is triggered. The injection safety latch unit integrates a numerical comparator and a parameter switch, which switches to the preset fixed injection parameters when the pilot fuel quantity compensation coefficient is greater than 15%; The fault code generation unit is configured with a diagnostic protocol stack and a message generator, and sends fault codes associated with historical data indexes to the vehicle controller.

7. The combustion optimization control system for a dual-fuel micro-injection engine according to claim 1, characterized in that: The adaptive controller of the compensation amount calculation unit includes: The phase offset input circuit is connected to the charge amplifier output terminal of the in-cylinder pressure sensor and receives the combustion phase offset voltage signal; The angular acceleration conversion circuit is connected to the digital output interface of the crankshaft position sensor unit to convert the crankshaft pulse signal into a digital angular acceleration; The gain parameter generator uses FPGA programmable logic devices to generate time-varying gain parameters based on the excess air coefficient change rate, speed fluctuation characteristic value and angular acceleration; The compensation amount operator integrates a multiplier and an adder circuit to output a digital signal of the pilot fuel amount compensation coefficient and the injection timing compensation amount.

8. The combustion optimization control system for a dual-fuel micro-injection engine according to claim 7, characterized in that: The gain parameter generator performs the following operations: (1) Using a 12-bit ADC to collect the excess air coefficient change rate analog signal of the air-fuel ratio detection unit; (2) Reading the digital value of the speed fluctuation characteristic value from the SPI interface of the crankshaft position sensor unit; (3) receiving the second-order derivative value of the crankshaft angular acceleration output by the angular acceleration conversion circuit; (4) Calculation of time-varying gain parameters is completed in FPGA programmable logic devices: ； Where: α, β and γ are calibration constants, determined by bench calibration test, dλ / dt is the rate of change of excess air coefficient, Δn is the speed fluctuation characteristic value, d 2 θ / dt 2 is the crankshaft angular acceleration, K p is the proportional gain parameter, K i is the integral gain parameter, K d is the differential gain parameter.

9. A dual-fuel micro-injection engine combustion optimization control method, applicable to the system according to any one of claims 1 to 8, characterized in that: The method comprises the following steps: S1. System initialization configuration: The operator burns the support vector machine model parameters and gain coefficients into the FRAM memory through an external calibration device; S2, sensor calibration start: use a dedicated calibration tool to perform zero drift correction on the in-cylinder pressure sensor; S3, mode switching activation: the operator triggers the mode switching command from fuel oil to gas on the control panel; S4. Compensation result recording: manually read the compensation effect report of the historical data storage unit and back it up to an external device.

10. The dual-fuel micro-injection engine combustion optimization control method according to claim 9, characterized in that: The method further comprises the following steps: S11. The operator connects the USB port of the calibration device to the RS485 port of the online calibration interface, selects the pre-stored calibration constant file through the calibration software interface and performs the burning operation; S21. Connect the charge amplifier output of the in-cylinder pressure sensor to the analog input channel of the calibration tool. The operator adjusts the zero point potentiometer to an error of less than ±0.5% based on the offset displayed by the calibration tool. S31. Select the "Dual Fuel Switch" button on the engine control unit (ECU) operation interface. When the instrument panel displays a speed fluctuation > 250 rpm, manually confirm the activation of pilot oil compensation. S41. Connect to the host computer through the data export interface of the FRAM memory, and the operator executes the data reading instruction and saves the compensation parameter log file.

Images