A plasma combustion-assisted engine system based on active volatile gas pressurization and stratification

Abstract

This application discloses an engine plasma combustion-assisted system based on active volatile gas pressurization and stratification. The system includes a volatile gas state monitoring module, an active pressurization module, a stratified injection control module, a plasma combustion-assisted module, and a controller. The volatile gas state monitoring module acquires the concentration and pressure of volatile gas from the fuel tank in real time; the active pressurization module actively pressurizes the volatile gas from the fuel tank to the target supply pressure accordingly; the stratified injection control module determines the target injection quantity of volatile gas based on the concentration and the overall target excess air coefficient in the cylinder, and injects the pressurized volatile gas through an independent injection channel from the intake end to the compression section to the area around the ignition point, forming a stratified mixture that is rich in the ignition point but lean overall; the plasma combustion-assisted module modulates the discharge energy according to the overall leanness, activating an active ignition nucleus to ignite the lean mixture. This application frees the volatile gas supply from intake manifold vacuum dependence and improves the ignition reliability of lean mixtures.

Description

Technical Field

[0001] This application relates to the field of internal combustion engine fuel supply and combustion control technology, and in particular to an engine plasma combustion-assisted system based on active volatile gas pressurization and stratification. Background Technology

[0002] Engine fuel tanks continuously generate volatile gases during storage and operation. Current treatment methods rely on the vacuum in the intake manifold to passively draw out the volatile gases and mix them evenly with the air entering the cylinder before they participate in combustion.

[0003] This approach has the following drawbacks:

[0004] Firstly, the suction power of volatile gas comes from the vacuum in the intake manifold, and this vacuum fluctuates with the engine operating conditions, resulting in insufficient suction capacity of volatile gas under certain operating conditions, and the supply capacity of volatile gas is uncontrollable as the operating conditions change.

[0005] Secondly, the volatile gas is diluted after being evenly mixed with all the intake air, and the highly active components are dispersed throughout the cylinder, making it difficult to create local conditions conducive to ignition at the ignition point.

[0006] Third, fluctuations in volatile gas concentration are considered interference that needs to be suppressed in existing methods and are not utilized. As a result, the degree of volatile gas recovery and utilization is limited, and there is insufficient ignition reliability when the overall cylinder is lean. Summary of the Invention

[0007] To address the problems in existing technologies where volatile gas supply relies on passive intake manifold vacuum, resulting in uncontrollable fluctuations in supply capacity depending on operating conditions, dilution of highly reactive components after uniform mixing with all intake air, and insufficient ignition reliability when the overall cylinder mixture is lean, this application provides an engine plasma combustion-assisted system based on active volatile gas pressurization and stratification. This system acquires the concentration and pressure of volatile gas from the fuel tank in real time and uses this as control input. After active pressurization, the volatile gas is targeted and injected into the area around the ignition point from the intake end to the compression section, forming a localized enrichment and stratification. Then, modulated energy plasma discharge excites the active ignition core to ignite the overall lean mixture in the cylinder. This system enables the volatile gas supply to break free from dependence on intake manifold vacuum, forms a localized enrichment at the ignition point conducive to ignition, and improves the ignition reliability of the overall lean mixture in the cylinder.

[0008] Specifically, this application provides the following technical solutions:

[0009] In a first aspect, this application provides an engine plasma combustion-assisted system based on active volatile gas pressurization and stratification, comprising:

[0010] The volatile gas state monitoring module is used to acquire the concentration and pressure of volatile gases in the engine fuel tank in real time.

[0011] An active pressurization module is used to actively pressurize the volatile gas from the fuel tank to the target supply pressure based on the concentration and the pressure.

[0012] The stratified injection control module is used to determine the target injection amount of volatile gas based on the concentration and the overall target excess air coefficient in the cylinder, and to reverse the injection pulse width and injection phase. The pressurized volatile gas is then injected into the area around the ignition point through an independent injection channel from the intake end to the compression section, forming a stratification that is rich around the ignition point and thin overall in the cylinder.

[0013] The plasma combustion-assisted module is used to modulate the plasma discharge energy according to the overall leanness of the cylinder, and to excite active ignition nuclei in the enriched area around the ignition point to ignite the overall lean mixture in the cylinder.

[0014] The controller is used to coordinate the operation of the volatile gas state monitoring module, the active pressurization module, the stratified injection control module, and the plasma combustion-supporting module.

[0015] Optionally, the active pressurization module includes a high-frequency micro air pump, which pressurizes the volatile gas in stages in series in a two-stage manner when the target supply pressure is higher than the single-stage pressurization capacity.

[0016] Optionally, the stratified injection control module is further configured to advance the injection phase relative to the ignition phase based on the crankshaft angular velocity and the mixing transport time, so that the enriched area around the ignition point is formed at the ignition moment.

[0017] Optionally, the stratified injection control module is further configured to determine the target injection quantity of volatile gas based on the volume of the enrichment zone in the vicinity of the ignition site, the target local equivalence ratio, and the overall target excess air coefficient in the cylinder, and output the volatile gas through a low-flow injector.

[0018] Optionally, the volatile gas state monitoring module is further configured to filter and estimate the concentration and obtain its rate of change and combustible quality index. The controller determines the availability of volatile gas based on the combustible quality index and provides the combustible quality index to the plasma combustion-supporting module for modulating the plasma discharge energy.

[0019] Secondly, this application provides an engine plasma combustion control method based on active volatile gas pressurization and stratification, including:

[0020] Real-time acquisition of the concentration and pressure of volatile gases in the engine fuel tank;

[0021] Based on the concentration and the pressure, the volatile gas from the fuel tank is actively pressurized to the target supply pressure;

[0022] Based on the concentration and the overall target excess air coefficient in the cylinder, the target injection amount of volatile gas is determined and the injection pulse width and injection phase are solved. The pressurized volatile gas is then injected into the area around the ignition point through an independent injection channel from the intake end to the compression section, forming a layer that is rich around the ignition point and thin overall in the cylinder.

[0023] The plasma discharge energy is modulated according to the overall leanness of the cylinder to stimulate active ignition nuclei in the enriched area around the ignition point, thereby igniting the overall lean mixture in the cylinder.

[0024] Optionally, the pressurized volatile gas is injected into the area around the ignition point. Based on the crankshaft angular velocity and the mixing and transport time, the injection phase is advanced relative to the ignition phase, so that the enriched area around the ignition point is formed at the ignition moment.

[0025] Optionally, it also includes collecting combustion feedback, extracting combustion phase and cycle fluctuation indicators, closing-loop correcting the target injection amount of volatile gas, the injection phase and the plasma discharge energy, and calculating the cumulative re-burning utilization of volatile gas.

[0026] Optionally, it also includes arranging the triggering sequence of the acquisition, the boost, the injection and the excitation based on the crankshaft phase, and gradually reducing the volatile gas injection quantity to zero and switching to conventional fuel supply when volatile gas is detected to be unavailable or combustion abnormal.

[0027] The beneficial effects of this application are as follows: By using the concentration and pressure of the fuel tank volatiles as control inputs and supplying them through active pressurization, the supply of volatiles no longer depends on the intake manifold vacuum, thus allowing for on-demand supply of volatiles under various engine operating conditions; by injecting the pressurized volatiles through an independent injection channel from the intake end to the compression section to the vicinity of the ignition point, a layered structure is formed around the ignition point that is enriched while the overall cylinder is lean, ensuring that highly reactive volatile components are concentrated at the ignition point rather than being diluted by the entire intake air; by modulating the plasma discharge energy according to the overall leanness of the cylinder and activating active ignition nuclei in the enriched area around the ignition point, the air-fuel mixture can still be reliably ignited even under overall lean cylinder conditions. The above-mentioned technical means work together to allow volatiles to be recovered and re-supplyed for combustion under various engine operating conditions, while simultaneously improving the ignition reliability of the overall lean cylinder mixture. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the overall process of the engine plasma combustion control method provided in the embodiments of this application.

[0029] Figure 2 This is a schematic diagram of the module architecture of the engine plasma combustion-assisted system provided in the embodiments of this application.

[0030] Figure 3This is a schematic diagram of the pressure step response of the active pressurization chamber provided in an embodiment of this application.

[0031] Figure 4 This is a schematic diagram illustrating the relationship between plasma discharge energy and the overall excess air coefficient within the cylinder, as provided in the embodiments of this application.

[0032] Figure 5 This is a schematic diagram illustrating the closed-loop convergence comparison of combustion cycle fluctuations provided in an embodiment of this application.

[0033] Figure 6 This is a schematic diagram of a single-cycle multi-channel collaborative timing waveform provided in an embodiment of this application.

[0034] Figure 7 A radar diagram illustrating the multi-dimensional performance comparison provided in this application embodiment. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0036] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0037] This embodiment provides an engine plasma combustion-assisted system and its control method based on active volatile gas pressurization and stratification. In a specific implementation, the system acquires the concentration and pressure of volatile gas in the engine fuel tank in real time through a volatile gas state monitoring module. An active pressurization module actively pressurizes the volatile gas from the fuel tank to the target supply pressure. A stratified injection control module injects the pressurized volatile gas through an independent injection channel from the intake end to the compression section, surrounding the ignition point to form a locally enriched but generally lean stratification within the cylinder. The plasma combustion-assisted module modulates the plasma discharge energy according to the overall leanness of the cylinder, activating active ignition nuclei in the enriched area around the ignition point. This allows the supply of volatile gas to break free from dependence on intake manifold vacuum, forms a locally enriched area at the ignition point conducive to ignition, and reliably ignites the air-fuel mixture under generally lean cylinder conditions. This enables the recovery and return of volatile gas from the fuel tank and stable combustion under various engine operating conditions.

[0038] like Figure 2 As shown, the system for implementing the above method consists of a volatile gas state monitoring module, an active boosting module, a stratified injection control module, a plasma combustion-supporting module, and a controller connected to each of the above modules. In a specific implementation, the controller is an engine electronic control unit.

[0039] Before executing the above method, calibration is required. Specifically:

[0040] S000 performs baseline calibration of zero point, characteristic curves and thresholds for sensing and actuator components involved in volatile gas state monitoring, active pressurization, stratified injection and plasma combustion.

[0041] In this embodiment, the calibration service module reads the original electrical signals from the volatile gas concentration sensor and pressure sensor, the ambient temperature, and the self-learning correction values ​​stored in non-volatile memory before the last power outage. It then sequentially completes the calibration of the four types of components and a system self-test, ultimately outputting a set of calibration coefficients and a system readiness flag. If any calibration or self-test item fails, the calibration service module sets the system readiness flag to 0 and starts the system in normal fuel mode. Specifically:

[0042] S010 performs zero-point and sensitivity calibration on the concentration sensor and pressure sensor and adds temperature compensation.

[0043] In this embodiment, the calibration service module maps the raw voltage or digital values ​​output by the concentration sensor and pressure sensor into physical quantities of concentration and pressure. The output of the concentration sensor is approximately linear with the measured concentration within its operating range, but it exhibits zero-point drift and sensitivity deviation with temperature variations. Therefore, a two-point calibration is used to determine the linearity coefficient, and a temperature-dependent compensation term is superimposed. The calibration relationship for the concentration here is as follows: The pressure calibration relationship is as follows: .in and This is the raw output of the concentration and pressure sensor. and Output at zero point. This is the concentration sensitivity coefficient. This is the temperature compensation coefficient. To calibrate the reference temperature, This is the pressure sensitivity coefficient. Where the zero point... The median of 256 consecutive samples taken in a clean environment after power-on was used to suppress the influence of abnormal pulses on the zero point, rather than taking a simple average value.

[0044] For example, in a specific implementation, the concentration sensitivity coefficient Set as (Volume fraction per ADC count), its value is determined by two-point calibration, that is, after sampling two standard gases of known concentration to obtain two output points with high and low concentrations, the difference between the high and low concentrations is correlated with the difference between the corresponding outputs. Temperature compensation coefficient. Set as to The values ​​(per degree Celsius) were obtained by linear regression after retesting the standard gas at different temperatures; the calibration reference temperature. Set the temperature to 25 degrees Celsius.

[0045] Assume the median of 256 points continuously sampled by the concentration sensor in a clean environment during power-on is... (ADC counting), raw output of concentration sensor at a certain moment (ADC count), oil tank temperature Celsius, at this point the temperature compensation term If the value is 1, then the calibrated concentration is That is, the concentration of volatile gases is 5% by volume. When the oil tank temperature exceeds... When the calibration range is from 80 degrees Celsius to 100 degrees Celsius, the extrapolation deviation of the temperature compensation term increases. At this point, the confidence level of the flammability quality index is correspondingly lowered in the subsequent S100 to avoid making overconfident judgments about the concentration outside the calibration range. The pressure calibration is similar to the concentration calibration, using the zero-point output and sensitivity coefficient to map the original output of the pressure sensor to a pressure value in kilopascals.

[0046] S020 is a characteristic curve relating the control quantity, differential pressure, and output flow rate of the booster device.

[0047] In this embodiment, the calibration service module establishes the relationship between the output flow rate of the micro air pump and its driving duty cycle and inlet / outlet pressure difference, so as to provide a feedforward basis for the subsequent active pressurization stage.

[0048] Among them, the output flow rate of the micro air pump decreases as the outlet back pressure increases under a given duty cycle, which is approximately a family of characteristics that decreases linearly with the pressure difference. Therefore, a linear fitting relationship between the flow rate and the duty cycle and the pressure difference is adopted. ,in The pressure difference across the pump. For the pump output volumetric flow rate, For pump drive duty cycle, The gain of flow rate due to duty cycle, The back pressure sensitivity coefficient, This is the intercept.

[0049] For example, after measuring the flow rate of the micro air pump at multiple duty cycles and multiple back pressure points on a test bench, the coefficients of the above relationship are fitted in a least squares manner.

[0050] In a specific implementation, regarding duty cycle Pressure difference The measured output flow rate at kPa was 3.0 liters per minute, at a duty cycle of Pressure difference The measured speed at 1 kPa was 2.0 liters per minute, at a duty cycle of Pressure difference The measured speed at 1000 kPa was 1.4 L / min. Therefore, the least squares solution yielded... (liters per minute) (liters per minute per kilopascal) (Liters per minute). The above fitting results were verified, and... , Substituting kPa into the fitting relationship, we obtain The increase per minute is consistent with the measured value, indicating that the fitting coefficient is usable.

[0051] Optionally, for air pumps exhibiting strong nonlinearity, a lookup table plus bilinear interpolation method is used instead of the above linear fitting. That is, a grid table from duty cycle and pressure difference to output flow rate is directly stored, and bilinear interpolation is performed according to the current duty cycle and pressure difference during operation. In this embodiment, linear fitting is preferred to save storage overhead of the electronic control unit.

[0052] S030, a linearized relationship between the injection pulse width and injection quantity is fitted to the volatile gas injector.

[0053] In this embodiment, the calibration service module establishes the relationship between the injection quantity of the volatile gas injector and its driving pulse width, so that the subsequent stratified injection stage can inversely solve the injection pulse width from the target injection quantity. The relationship between the injection quantity and the pulse width is as follows: ,in The mass of volatile gas emitted in a single injection. For flow coefficient, For the jet pulse width, This refers to the ineffective opening time of the injector. To alleviate supply pressure, To calibrate the reference supply pressure, introduce One method is used to illustrate the effect of supply pressure on injection volume, where the orifice flow rate is approximately proportional to the square root of the pressure difference.

[0054] For example, the volatile gas injector is a small-flow-rate dedicated injector with a flow coefficient of... At the level of 0.4 mg / ms, to match the sub-milligram to several milligram volatile gas replenishment requirements of a single-cylinder, single-cycle system. In a specific implementation, the reference supply pressure is calibrated. kPa, ineffective start-up time of the injector Milliseconds, in supply pressure Actual measured injection pulse width at kPa Injection volume per millisecond milligrams, then the flow coefficient Milligrams per millisecond. This invalid on-time typically ranges from 0.4 milliseconds to 1.0 millisecond across different injectors, determined by weighing calibration of the injector near the zero-volume critical pulse width. Following this calibration, the corresponding injection pulse width can be subsequently derived from the target injection volume and the current supply pressure.

[0055] S040 is used to calibrate the breakdown threshold and the upper limit of the safe discharge current for plasma discharge devices.

[0056] In this embodiment, the calibration service module applies incremental drive to the high-energy plasma ignition coil and records its volt-ampere characteristics to determine the breakdown voltage threshold and the upper limit of the safe discharge current, which serve as the safety boundary for subsequent plasma discharge energy modulation. The breakdown voltage threshold is taken as the 95th percentile of multiple breakdown voltages to ensure reliable breakdown and the formation of a discharge channel at this threshold.

[0057] After completing the above four types of calibration, the calibration service module performs a system self-check, writes the calibration coefficients to the backup area of ​​non-volatile storage, and atomically switches to the effective area after the verification is passed, so as to avoid the coefficients being damaged due to power failure during the calibration process.

[0058] For example, when the self-test detects that the measured value of the breakdown voltage threshold exceeds the safety limit of the coil, such as when the breakdown voltage rises abnormally due to insulation deterioration, the calibration service module sets the system ready flag to 0, prohibits entering the volatile gas combustion mode, and starts the system in the normal fuel mode while recording the fault code, thereby prioritizing the basic operational safety of the engine when the component characteristics are abnormal.

[0059] The above method specifically includes the following steps:

[0060] S100 acquires the concentration and pressure of volatile gases in the engine fuel tank in real time.

[0061] like Figure 1 and Figure 2 As shown, this stage is executed periodically in phase with the crankshaft during engine operation, estimating the concentration and pressure of volatile gases in the fuel tank area in real time and assessing their combustibility, providing input for subsequent active turbocharging and stratified injection. In S100, the volatile gas state monitoring module reads the concentration sensor signal, pressure sensor signal, fuel tank temperature, and calibration coefficients obtained from S000. It first filters and estimates the concentration and pressure and calculates the concentration change rate, then calculates the combustibility index, and finally compares it with the lower combustible limit threshold to determine the usability indicator of volatile gases. This step quantifies concentration fluctuations, which are usually considered disturbances, into usable control signals.

[0062] S110 performs filtered estimation on the concentration and pressure observations and obtains the rate of change of concentration.

[0063] In S110, the volatile gas state monitoring module performs filtered estimation on the concentration observations calibrated by S000. The concentration sensor response contains noise and exhibits a certain lag. To provide a smooth and accurate concentration estimate and rate of change within a delay of no more than 20 milliseconds, the volatile gas concentration is modeled as a second-order state containing both concentration and its rate of change, and recursively estimated using Kalman filtering. Its state vector is... The prediction process is The state transition matrix The update process is as follows: The observation matrix , These are the calibrated concentration observation values. For Kalman gain, The sampling period is defined as follows: the observation noise variance is the variance of 1000 samples taken continuously by the sensor at a constant concentration, and the process noise covariance is used to adjust how quickly the estimate follows the observation.

[0064] For example, in a specific implementation, the sampling period Seconds, following the previously calibrated concentration observation That is, a volume fraction of 5%. Let the concentration estimate at the previous time step be... Concentration change rate estimation The prediction process provides the result per second. Let the current Kalman gain be... Then the new information is The updated concentration estimate is That is, a volume fraction of 5% and a concentration change rate of 1%. Every second. It is evident that when observations and predictions are consistent, the estimated value remains stable and the rate of change is preserved. This rate of concentration change is then used as a feedforward quantity to participate in the advance correction of stratified injection. A sampling period of 5 milliseconds allows the concentration estimate to follow a single concentration step within a delay of no more than 20 milliseconds, meeting the real-time requirement of synchronization with the crankshaft phase.

[0065] S120, calculates the flammability quality index based on concentration estimates and determines the usable mark for volatile gases accordingly.

[0066] In S120, the volatile gas state monitoring module quantifies the flammability of the current volatile gas as an ignition medium, which is used for the subsequent scheduling of plasma discharge energy. The flammability quality index is... ,in This refers to the concentration corresponding to the lower limit of flammability of the volatile gas. Stoichiometric concentration, This is a temperature correction factor. This indicates that the result within the parentheses is limited to between 0 and 1. When the concentration is below the lower flammability limit, the flammability quality index is 0, indicating that the volatile gas cannot be used as an ignition medium, and the volatile gas usability flag is set to 0 accordingly.

[0067] For example, in one specific implementation, the lower flammability limit of gasoline vapor corresponds to the concentration Take a volume fraction of 1.4%, stoichiometric concentration Take a volume fraction of 2%, temperature correction factor The concentration is set to 1.0 when evaporation and atomization conditions are favorable, and then adjusted to 0.8 when evaporation is slow and atomization deteriorates at low temperatures. This is based on the concentration estimated in S110. The combustible quality index is then... This means that the current concentration is much higher than the stoichiometric point, and the quality of the ignition layer is sufficient, so the availability flag for volatile gas is set to 1. At this time, the concentration, pressure, concentration change rate, combustible quality index, and availability flag are output together. Among them, the concentration and quality index are directly transmitted to subsequent stages for setting the pressurization target and modulating the discharge energy.

[0068] As a comparison, if the concentration is estimated at a certain moment... If the volume fraction is 1%, which is lower than the lower flammability limit of 1.4%, then the flammability quality index is 0 and the usability mark is 0. Based on this, the controller switches the cylinder back to the conventional fuel mode in this cycle and does not perform volatile gas injection and plasma combustion enhancement, thereby avoiding ineffective injection and waste of ignition energy when the volatile gas concentration is insufficient.

[0069] In a specific implementation, the volatile gas state monitoring module uses a circular buffer to store the concentration values ​​and corresponding timestamps of the most recent sampling points to detect abnormal concentration changes. When the concentration change rate between two adjacent sampling points exceeds a set abnormal threshold, it is determined to be a sensor malfunction or an abnormal event such as fuel tank filling. This sampling point is then discarded as an outlier to prevent the outlier from propagating through the Kalman filter and causing abrupt changes in the control quantities of subsequent stages. The abnormal threshold here is three times the maximum concentration change rate under normal driving conditions, serving as the boundary for determining sensor malfunction. The temperature correction factor in the combustible quality index reflects the influence of fuel tank temperature on volatile gas components: under low-temperature conditions, such as when the ambient temperature is below zero degrees Celsius, the vapor pressure of light components in gasoline decreases, the mass fraction of combustible components in the volatile gas decreases, and the quality of the volatile gas as an ignition medium decreases at the same concentration. Therefore, the temperature correction factor is adjusted to 0.8 at low temperatures to reflect this effect.

[0070] It should be noted that the combustible quality index can also be extended to a dynamic quality index that considers the concentration change rate. When the concentration change rate is positive and the amplitude is large, it indicates that the volatile gas is in a rapid generation stage, and so on.

[0071] S200, based on the concentration and the pressure, actively pressurize the volatile gas from the fuel tank to the target supply pressure.

[0072] This stage is executed after S100 determines that the volatile gas is available. It uses a high-frequency micro air pump to actively pressurize the volatile gas from the fuel tank, thereby replacing the passive desorption method that relies on intake manifold vacuum in traditional evaporative emission control. This makes the supply of volatile gas controllable under various operating conditions, including full load and boost.

[0073] In this embodiment, the active boosting module reads the concentration, pressure, and combustible quality index obtained in S100, as well as the target injection demand returned by the subsequent stratified injection stage and the air pump characteristics obtained in S000. First, it determines the target supply pressure according to the boosting target setting. Then, it controls the air pump's duty cycle using a feedforward superposition of the air pump characteristics and proportional-integral feedback to stabilize the actual pressure in the boosting chamber at the target supply pressure. Finally, after the actual pressure stabilizes within the target neighborhood, it outputs a boosting ready flag. Specifically, it includes the following sub-steps:

[0074] S210 determines the target supply pressure based on the target injection volume, injection demand, and back pressure at the injection moment.

[0075] In this embodiment, the active pressurization module determines a target supply pressure that ensures the injector delivers the required mass of volatile gas within the target pulse width while also allowing for a margin in back pressure at the injection moment. Since the injector's injection volume is approximately proportional to the square root of the supply pressure, and the injection must overcome the back pressure at the nozzle at the injection moment, the target supply pressure is determined by the larger of the metering requirement and the back pressure margin.

[0076]

[0077] in, To provide pressure for the target, The target volatile gas injection mass for this cycle is... For the corresponding target jet pulse width, The back pressure at the nozzle at the moment of injection. For back pressure margin, , and This is derived from the S000 calibration. The back pressure margin is set to 1.2 times the minimum pressure difference required to ensure that the spray completes penetration within the target phase window, in order to balance spray penetration and metering stability.

[0078] For example, in a specific implementation, based on the aforementioned calibration results, the volatile gas concentration is 5% by volume, the combustibility index is 1.0, and the flow coefficient of the low-flow volatile gas injector is... mg per millisecond, ineffective on-time Milliseconds, calibration reference supply pressure kPa.

[0079] The subsequent stratified injection stage will provide the target volatile gas injection mass for this cycle according to the stratified requirements. Milligrams, corresponding to the target jet pulse width Milliseconds, back pressure at the intake nozzle kPa, back pressure margin kPa. The measurement requirement item is: The kPa is because the target injection pulse width is obtained by inversely solving the equation that the supply pressure equals the reference pressure of 300 kPa, so the metering demand term is consistent with 300 kPa.

[0080] Back pressure margin term is kPa. The larger of the two values ​​represents the target supply pressure. kPa. When the target supply pressure cannot be reached due to the constraint of the maximum achievable pressure of the air pump or the setting value of the pressure relief valve of the booster chamber, the mass of the target volatile gas injected in the subsequent stratified injection stage shall be reduced accordingly to ensure that the supply pressure is consistent with the metering.

[0081] There is an engineering trade-off between selecting the volume of the booster chamber and setting the back pressure margin. If the volume of the booster chamber is too large, the air pump pressurization time will increase, and the system's response to changes in transient injection demand will be slower; if the volume of the booster chamber is too small, the proportion of pressure fluctuations in the chamber caused by each injection will increase, resulting in a decrease in metering accuracy.

[0082] For example, the booster chamber volume is between 50 and 200 ml, ensuring that the volume consumption corresponding to a single injection of 0.485 mg of volatile gas accounts for no more than 0.1% of the total booster chamber volume, thereby guaranteeing that the pressure inside the chamber remains stable between consecutive injections. The back pressure margin is set based on the physical process of spray penetration: after the volatile gas is ejected from the nozzle, it must penetrate into the spark plug gap within the target phase window. The spray penetration distance is approximately proportional to the square root of the pressure difference across the nozzle. When the pressure difference is insufficient, the spray penetration distance is shortened, causing the enrichment zone to deviate from the spark plug gap and reducing ignition reliability. Therefore, the back pressure margin is set to 1.2 times the minimum pressure difference required to ensure that the spray completes penetration within the target phase window, that is, a safety factor of 1.2 is used to cover the cycle-by-cycle fluctuations in the cylinder pressure.

[0083] S220 uses a combination of feedforward and proportional-integral feedback to control the pump duty cycle, thereby stabilizing the pressure in the booster chamber to the target supply pressure.

[0084] In this embodiment, the active booster module stabilizes the actual pressure in the booster chamber to the target supply pressure under back pressure disturbances and fluctuations in volatile gas consumption. The control quantity is synthesized from a feedforward term and a feedback term. The feedforward term is obtained by inverse solving the pump characteristics calibrated by S000, that is, finding the duty cycle reference to achieve the required supplementary flow rate under the known target supply pressure and pressure difference, expressed as:

[0085]

[0086] The feedback term uses incremental proportional-integral control to eliminate steady-state error, expressed as:

[0087]

[0088] Among them, error ;

[0089] The final duty cycle is the sum of the feedforward and feedback cycles, limited to between 0 and 1, i.e.:

[0090]

[0091] in, The required supplemental flow rate to maintain the target pressure is estimated based on leakage from the booster chamber and injection consumption. and The proportional and integral gains are determined by the booster chamber volume and the pump bandwidth. , This represents the pressure difference across the pump.

[0092] For example, in a specific implementation, the volatile gas consumption flow rate corresponding to the sub-milligram injection volume is extremely small. The supplementary flow rate required to maintain the pressure in the pressurization chamber is mainly used to compensate for chamber leakage. Therefore, the supplementary flow rate... The pressure ratio that a single stage of a miniature diaphragm pump can provide is typically no more than 2 to 3 times the pressure ratio due to the constraints of diaphragm stroke and valve sealing. The pressure ratio required to boost the pressure from the internal tank pressure of about 110 kPa to the target supply pressure of 300 kPa is about 2.7, which is close to the limit of a single stage and may not be achievable when the diaphragm is worn or the ambient temperature is high.

[0093] Therefore, a two-stage series pressurization method is adopted to increase the effective pressure difference of each stage. kPa. Substitute the above values ​​into the feedforward relationship and into the pump characteristic coefficient obtained from the S000 calibration. , , The feedforward duty cycle is obtained. The feedforward duty cycle is 0.80, which is less than 1, indicating that the duty cycle is not saturated and the boosting capacity meets the requirements. The selection of sub-milligram injection volume and two-stage series boosting here allows the duty cycle to have a margin and not become saturated.

[0094] At a certain control step, let the actual pressure of the pressurization chamber be... kPa, then the error kPa, previous beat error kPa, proportional gain Integral gain per kilopascal For every kilopascal, the feedback increment is The feedback term makes minor adjustments to the duty cycle to ensure smooth convergence, and the final duty cycle is the result of limiting the sum of the feedforward and feedback. Optionally, when the booster chamber volume is large and the hysteresis is significant, model predictive control can be used instead of the above proportional-integral feedforward control. In this embodiment, proportional-integral feedforward control is preferred to meet the real-time requirements.

[0095] S230 outputs a boosting ready flag after the pressure in the boosting chamber enters the vicinity of the target supply pressure and the set number of beats is continuously set.

[0096] like Figure 3 As shown, the active boosting module performs a ready hysteresis judgment on the boosting chamber pressure. Only after the actual pressure of the boosting chamber enters the neighborhood of 5% above and below the target supply pressure and the set number of beats is continuously set, will the boosting ready flag be set to 1, in order to avoid the pressure fluctuating near the boundary of the target neighborhood, which would cause the ready flag to flip frequently.

[0097] For example, in a specific implementation, when the actual pressure in the booster chamber enters the range of 300 kPa plus or minus 15 kPa and continues for 3 control cycles, the booster ready flag is set to 1. This flag is then used by the stratified injection stage as an injection trigger condition and by the timing arrangement stage.

[0098] In contrast, if the pressure in the boost chamber remains below 80% of the target supply pressure for a set time due to leakage, boost failure is determined, the boost ready flag is set to 0, and the controller switches back to the normal fuel mode, thereby ensuring the basic operation of the engine when boost capacity is insufficient.

[0099] The pressurization chamber is also equipped with a pressure relief valve. When the internal pressure of the pressurization chamber exceeds the set safety upper limit, the pressure relief valve automatically opens, releasing excess volatile gas back to the oil tank or activated carbon canister to protect the structural integrity of the pressurization chamber and pipeline. The opening pressure of the pressure relief valve is set to 1.3 times the target supply pressure, ensuring sufficient margin between the normal operating pressure and the pressure relief threshold. A pressure sensor is installed inside the pressurization chamber to provide real-time feedback on the chamber pressure. The sampling results from this pressure sensor are filtered using a moving average to suppress pulsations caused by the reciprocating motion of the air pump, ensuring a stable pressure value fed back to the control loop.

[0100] S300, based on the concentration and the overall target excess air coefficient in the cylinder, determines the target injection amount of volatile gas and reverses the injection pulse width and injection phase, and injects the pressurized volatile gas into the area around the ignition point through an independent injection channel from the intake end to the compression section, forming a layer that is rich around the ignition point and thin overall in the cylinder.

[0101] This stage, executed from the intake end to the compression section after the boost-ready flag is set to 1, involves targeting the boosted volatiles through independent volatile injectors installed near the spark plug and directed towards the spark plug gap. This creates a localized enrichment around the spark plug and an overall lean stratification within the cylinder. At this point, the conventional fuel already creates an overall lean background based on the target excess air coefficient within the cylinder. The additional targeted injection of volatiles into the enriched area near the spark plug increases the equivalence ratio of that localized area to the target local equivalence ratio, thereby spatially separating the locally ignitable area from the overall lean, low-emission environment.

[0102] In the S300, the stratified injection control module reads the concentration, boost chamber pressure, overall target excess air coefficient in the cylinder, engine speed, load, and injector calibration, as well as the correction amount fed back from the subsequent combustion feedback loop. It first determines the target volatile gas mass corresponding to the local target equivalence ratio around the spark plug, then reverse-engineers the injection pulse width, and finally determines the injection phase. The target volatile gas mass and target injection pulse width are then fed back to the active boost loop to set the target supply pressure. Specifically:

[0103] S310 determines the target volatile gas mass based on the volume of the enrichment zone near the ignition point, the target local equivalence ratio, and the overall target excess air coefficient in the cylinder.

[0104] In this embodiment, the stratified injection control module divides the combustion chamber into two spatial domains: a spark plug-adjacent enrichment zone and the overall cylinder equivalence zone. Given that a thin background has already formed in the overall cylinder equivalence zone, the module determines the additional volatile gas mass required to inject into the enrichment zone to achieve the target local equivalence ratio. The total fuel required for the enrichment zone to achieve the target local equivalence ratio is:

[0105]

[0106] Among them, the air quality in the enriched area was The fuel already provided by the overall rarefied background in this enriched region is:

[0107]

[0108] Therefore, the amount of volatile gas that needs to be supplemented is the difference between the total fuel and the background fuel, that is:

[0109]

[0110] in, The effective volume of the spark plug's adjacent enrichment zone is determined by the spray cone angle, penetration distance, and the size of the spark core at ignition time. The density of the gas at the intake end of the cylinder. For stoichiometric air-fuel ratio, For the target equivalent ratio in the enriched area, The overall excess air coefficient in the cylinder.

[0111] For example, the effective volume of the spark plug's neighborhood enrichment region cubic centimeters, taking the larger proportion of the clearance volume to correspond to the need for a larger ignition core in lean combustion; in-cylinder intake gas density g / cm³; stoichiometric air-fuel ratio of 14.7; target equivalence ratio in enrichment zone Overall excess air coefficient in the cylinder The air quality in the enriched area is... Milligrams.

[0112] Substituting the above values, we get... The amount of volatile gas that needs to be replenished in a single cylinder and single cycle is 0.485 mg, which is in the sub-milligram range and is matched with a special injector for low-flow volatile gas.

[0113] Because a larger effective volume in the enrichment zone requires more volatile gas and the enrichment layer is easily diluted, while a smaller effective volume results in insufficient ignition core energy; when the target local equivalence ratio is higher than 1.1, the enrichment zone becomes too rich, producing soot, while when it is lower than 0.9, ignition reliability decreases. Therefore, the target local equivalence ratio is set between 0.9 and 1.1, and the overall in-cylinder excess air coefficient is set between 1.3 and 1.6. This target volatile gas mass is then fed back to the active pressurization stage to set the target supply pressure.

[0114] It should be noted that the effective volume of the enrichment zone is not a geometrically fixed quantity, but rather an equivalent volume determined by the spray cone angle, spray penetration distance, tumble intensity of the in-cylinder airflow field, and the mixing and transport process from injection to ignition. When the spray cone angle is large, the volatile gas diffuses more radially, increasing the cross-sectional area of ​​the enrichment zone but decreasing the concentration gradient; when the spray penetration distance is short, the enrichment zone is concentrated near the nozzle and is difficult to cover the spark plug gap.

[0115] Optionally, by selecting a nozzle with a cone angle between 15 and 25 degrees and installing it 10 to 20 millimeters from the spark plug gap, the spray penetration distance can be made to precisely cover the spark plug gap, resulting in an effective volume of the enriched zone between 10 and 20 cubic centimeters. The in-cylinder tumble intensity increases with engine speed; the higher the speed, the faster the spray is transported and mixed by turbulence. The concentration gradient of the enriched layer at ignition time tends to be gentler. Therefore, under high-speed conditions, the mixing and transport time should be shorter, and the injection phase should be closer to the ignition phase.

[0116] S320 deduces the injection pulse width based on the target volatile gas mass, supply pressure, and injector flow characteristics.

[0117] In this embodiment, the stratified injection control module uses the target volatile gas mass obtained from S310, combined with the current supply pressure and the injector flow characteristics calibrated in S000, to inversely calculate the injection pulse width. The injection pulse width is:

[0118]

[0119] And on this basis, the pulse width correction amount fed back from the subsequent combustion feedback stage is superimposed, that is:

[0120]

[0121] in The injector flow coefficient, To alleviate supply pressure, To calibrate the reference supply pressure, This refers to the ineffective opening time of the injector. This is the pulse width correction amount.

[0122] For example, in one specific implementation, 0.485 milligrams of the aforementioned target volatile gas are received under pressure. The reference supply pressure is equal to 300 kPa, therefore the square root correction term of the supply pressure is 1, and the ejector flow coefficient is... mg per millisecond, ineffective on-time In milliseconds, the jet pulse width is Milliseconds. The injection pulse width, along with the target volatile gas mass, is fed back to the active pressurization stage as input to the target supply pressure measurement demand, thereby ensuring that the supply pressure and measurement remain consistent under the same set of parameters.

[0123] When there is a pulse width correction value provided by the subsequent combustion feedback loop, it is superimposed on the above inverse solution result to compensate for the drift of injector characteristics and the combustion deviation of each cycle.

[0124] S330 advances the injection phase relative to the ignition phase based on the crankshaft angular velocity and the mixing transport time, so that the enrichment zone around the ignition point is formed at the ignition moment.

[0125] In this embodiment, the stratified injection control module determines the injection phase so that the volatile gas spray forms a stable enrichment layer around the spark plug precisely at the moment of ignition. The injection phase is advanced by a crankshaft angle corresponding to the mixing and transport time from the ignition phase, i.e.:

[0126]

[0127] in For ignition phase, For the crankshaft angular velocity, satisfying , The mixing and transport time required for the spray to reach and form the target enrichment layer. If the mixing and transport time is too long, the enrichment layer will be homogenized by turbulence, so its value is usually between 1 millisecond and 3 milliseconds.

[0128] For example, engine speed Rotational speed per minute corresponds to a crankshaft angular velocity of 12 degrees per millisecond; mixed transport time Milliseconds; Ignition Phase The crankshaft rotation angle is 15 degrees before top dead center. Substituting the above value into the injection phase relationship, we get... The crankshaft angle, i.e., the injection phase, is 39 degrees before top dead center, falling within the stratified injection window from the end of the intake to the mid-compression section. This aligns with the engineering practice of stratified direct injection in the cylinder, ensuring stable stratification during compression and preventing premature homogenization of the spray by turbulence. This mixing and transport time is calibrated using the spray penetration distance and the in-cylinder vortex intensity.

[0129] Optionally, a mapping table for mixing and transport time is pre-calibrated using engine speed and load as an index, and bilinear interpolation is performed during operation based on the current speed and load. As a contrast, when the overall excess air coefficient in the cylinder is already below the lean-burn lower limit, for example, below 1.2, adding volatile gases would make the overall mixture too rich. In this case, the stratified injection control module rejects incremental injection, maintaining only the background fuel supply and lowering the target local equivalence ratio, thereby avoiding further enrichment when the mixture is already too rich, which would lead to soot.

[0130] In a specific implementation, the stratified injection control module maintains an effective volume mapping table for the enriched region and a mixed transport time mapping table indexed by engine speed and load. Each mapping table is stored in an 8-row by 8-column grid. During runtime, bilinear interpolation is performed in the grid based on the current speed and load to obtain the corresponding effective volume of the enriched region and the mixed transport time.

[0131] In addition, the effective volume mapping table of the enrichment zone was calibrated by bench testing combined with in-cylinder spray visualization, and the mixing transport time mapping table was calibrated by in-cylinder particle image velocimetry experiments. These two mapping tables provide dynamic stratification parameters as engine operating conditions change, enabling stratified injection control to form a stable enrichment zone around the spark plug across a wide operating range from idle to full load.

[0132] S400 modulates the plasma discharge energy according to the overall leanness of the cylinder, and activates the active ignition nucleus in the enriched area around the ignition point to ignite the overall lean mixture in the cylinder.

[0133] This stage, executed during the ignition phase after the formation of the enrichment layer around the spark plug in S300, uses electro-plasma to enhance the combustion of the air-fuel mixture. In S400, the plasma combustion enhancement module reads the target local equivalence ratio, overall excess air coefficient in the cylinder, combustible quality index, concentration, ignition base phase, and coil volt-ampere characteristics. It first calculates the target discharge energy and duration based on the overall leanness of the cylinder and the quality of the enrichment layer, then drives the high-energy plasma ignition coil to generate a continuous arc discharge in the spark plug gap. This excites a large active ignition nucleus in the enrichment region, which ignites the enrichment layer and subsequently ignites the overall lean mixture in the cylinder. Specifically:

[0134] S410 modulates the target discharge energy based on the overall excess air coefficient in the cylinder and the quality of the enriched layer.

[0135] In this embodiment, the plasma combustion-enhancing module determines the target discharge energy required for reliable ignition based on the overall lean-out level and the quality of the enriched layer within the cylinder. The target discharge energy, based on the baseline ignition energy, increases with increasing overall excess air coefficient, decreasing combustible quality index, and decreasing target local equivalence ratio, as expressed below:

[0136]

[0137] in, The baseline ignition energy under stoichiometric conditions and with good quality. The energy gain coefficient is the coefficient for the degree of rarefaction. This is the quality compensation coefficient. To prevent the lower limit of the denominator from being too small, The overall excess air coefficient in the cylinder. It refers to the flammability quality index. The target local equivalent ratio.

[0138] Therefore, the leaner the air-fuel mixture, the higher the ignition energy required; the lower the quality of the enrichment layer, the more compensation energy is required. The increased demand for ignition energy due to the overall leanness of the cylinder stems from the competition between the heat dissipation of the flame core and the chemical reaction rate: the leaner the air-fuel mixture, the larger the intermolecular distance of combustible molecules, the constant rate of heat transfer from the surface of the flame core to the surrounding unburned mixture, while the rate of chemical exothermic reaction decreases with decreasing concentration, thus increasing the minimum energy required for the flame core to maintain self-propagation.

[0139] It should be noted that the quality index's energy compensation reflects the difference in the activity of volatile gas components. A lower combustible quality index means that the proportion of heavy components in the volatile gas is higher or the temperature is too low, resulting in insufficient evaporation. The ignition nucleus requires more initial energy to overcome the higher ignition temperature. The relationship between the target discharge energy and the discharge duration is the initial energy stored in the energy storage capacitor minus the heat dissipation and radiation losses during the discharge process. The longer the discharge duration, the lower the energy utilization rate due to increased heat dissipation from the electrodes. Therefore, while meeting the ignition delay requirements, the discharge duration should be controlled within a relatively short time.

[0140] For example, the reference ignition energy millijoules, rarefaction level, energy gain coefficient Quality compensation coefficient Lower limit of denominator The aforementioned target local equivalent ratio Overall excess air coefficient in cylinder Combustible quality index Substituting the above values ​​into the relationship for target discharge energy, we obtain... Millijoules. This means that the ignition energy required under this rarefied condition is about 31 millijoules, which is in the tens of millijoules range, higher than the 1 to 3 millijoules required for conventional spark ignition.

[0141] The upper limit of the target discharge energy is constrained by the coil's safe volt-ampere and electrode ablation life. When the target discharge energy exceeds this upper limit, for example, exceeding 60 millijoules, it is limited to the upper limit and an alarm is triggered. If necessary, the ignition phase is reversed to reduce the energy demand caused by the overall lean-out of the cylinder. The actual discharge energy is the integral of the product of the discharge voltage and discharge current over the discharge duration, expressed as... ,in This is the discharge voltage. This is the discharge current. During the discharge duration, the discharge monitoring sub-component samples the discharge voltage and discharge current and integrates them in real time to obtain the actual discharge energy, which is then compared with the target discharge energy to achieve closed-loop energy storage regulation.

[0142] S420 drives the plasma discharge device to excite an active ignition core at the ignition point and verifies that the ignition core temperature meets the reliable ignition conditions.

[0143] In this embodiment, the plasma combustion-assisted module drives the high-energy plasma ignition coil to generate a continuous arc discharge in the spark plug gap, exciting the active ignition core. The ignition core temperature is determined by the temperature rise caused by the discharge energy on the mass of the ignition core, superimposed on the initial temperature of the enriched layer, and is expressed as:

[0144]

[0145] The corresponding ignition delay decreases exponentially with increasing ignition core temperature, as expressed in:

[0146]

[0147] The condition for reliable ignition is that the ignition delay is less than the discharge duration, meaning the discharge duration must cover the ignition delay. The ignition core temperature, For the initial temperature of the enrichment layer, As an energy utilization factor, For the ignition core mass, For isobaric specific heat, As the apparent activation energy, The gas constant is It is a pre-exponential factor.

[0148] like Figure 4 As shown, the target discharge energy increases with the increase of the overall excess air coefficient in the cylinder in order to maintain reliable ignition under lean conditions.

[0149] For example, the plasma ignition nucleus acts on the initial volume of the spark plug gap, which is on the order of cubic millimeters, taking the gap volume as an example. cubic millimeter, that is cubic centimeters, gas density is If the mass of the ignition nucleus is grams per cubic centimeter, then the mass of the ignition nucleus is . gram. Let the energy utilization factor be... Actual discharge energy Millijoules, or 0.0313 joules, are the specific heats at constant pressure. Joules per gram per Kelvin, initial temperature of enrichment layer Kelvin, substituting the above values ​​into the relationship for ignition core temperature, yields... Kelvin. The ignition core temperature is approximately 4612 Kelvin, which is within the reasonable range for plasma ignition, i.e., several thousand Kelvin. The corresponding ignition delay is extremely short, less than the discharge duration on the order of several hundred microseconds, thus enabling reliable ignition.

[0150] It should be noted that the ignition core mass in the ignition core temperature rise calculation is the initial spark core mass in the spark plug gap, which is in the microgram range, not the mass of the entire enrichment zone, which is in the milligram range; otherwise, the temperature rise estimate would be severely distorted. As a contrast, when the target discharge energy exceeds the coil's safe upper limit after adjustment, it is limited to the upper limit and an alarm is triggered. Simultaneously, the ignition phase is reversed to reduce the energy demand due to the overall lean-out in the cylinder, thus ensuring both electrode life and ignition reliability. In a specific implementation, plasma discharge energy modulation drive uses a controllable energy storage capacitor in conjunction with a switch. During the charging phase, the controller sets the charging termination voltage of the energy storage capacitor according to the target discharge energy. After the energy storage capacitor is charged to this voltage in the charging circuit, the switch disconnects the charging circuit to maintain energy storage. During the discharging phase, the switch connects the energy storage capacitor to the plasma ignition coil. The energy storage capacitor releases energy into the spark plug gap through the ignition coil, forming a plasma arc at the gap.

[0151] The plasma discharge used in this embodiment belongs to the category of thermal plasma, where the temperature of the discharge arc can reach thousands to tens of thousands of Kelvin. The active ignition nucleus formed at the spark plug gap contains a large number of free radicals and active particles. These active particles trigger a chain reaction during diffusion into the enrichment layer, igniting the gas mixture. Compared with the point-like ignition nucleus of conventional spark ignition, the ignition nucleus formed by the plasma arc is larger in volume and has a higher concentration of active particles, making it easier to ignite the gas mixture under lean conditions. The discharge voltage and current sampling circuit samples the voltage and current waveforms during the discharge process at high speed. The sampling results are numerically integrated over the discharge duration to obtain the actual discharge energy. The deviation between the actual discharge energy and the target discharge energy is used to correct the charging termination voltage of the energy storage capacitor in the next cycle, forming a closed-loop energy regulation.

[0152] In addition, the size of the spark plug gap also affects the discharge characteristics. When the gap is larger, the breakdown voltage increases and the arc extension space increases, resulting in a larger active ignition nucleus, which is beneficial for the ignition of lean mixtures. However, an excessively large gap will cause the breakdown voltage to exceed the power supply capacity of the coil. Therefore, in this embodiment, the spark plug gap is taken as 1.2 to 1.5 times that of the conventional spark plug gap to achieve a balance between breakdown reliability and ignition nucleus volume.

[0153] The S500 collects combustion feedback data, performs closed-loop correction of injection quantity, injection phase and discharge energy, and calculates the cumulative re-burning utilization of volatile gas.

[0154] This step is performed after each combustion cycle, feeding back the combustion results of this cycle to the aforementioned injection and discharge stages to reduce combustion deviation cycle by cycle and to calculate the cumulative re-burning utilization of volatile gases. In this embodiment, the feedback module collects ion current or cylinder pressure signals, the measured excess air coefficient corresponding to exhaust oxygen, misfire and knock indicators, and the target combustion phase. It first extracts combustion characteristics, then generates a correction amount using incremental proportional-integral conversion and feeds it back to the injection and discharge stages, while simultaneously accumulating and calculating the re-burning utilization of volatile gases. Specifically:

[0155] S510 extracts combustion phase and cycle fluctuation indicators from cylinder pressure or ion current signals.

[0156] In this embodiment, the feedback module extracts phase and stability indicators characterizing combustion quality from cylinder pressure or ion current curves. The combustion phase is characterized by the crankshaft angle corresponding to a cumulative heat release of 50%, calculated from the cylinder pressure using the first law of thermodynamics to determine the heat release rate. The indicated average effective pressure is obtained by dividing the cylinder pressure over the cylinder volume by the displacement, expressed as:

[0157]

[0158] The cyclic fluctuation index is characterized by the ratio of the standard deviation to the mean of the mean effective pressure across multiple cycles, expressed as: .in For displacement, and These are the standard deviation and mean of the average effective pressure of the multi-cycle indicator, respectively.

[0159] For example, the feedback module maintains a sliding window of the multi-cycle indicated mean effective pressure to calculate the cycle fluctuation index and maintains an annular buffer for the concentration history. When the cycle fluctuation index exceeds 5%, combustion instability is determined, the feedback module increases the correction amount of the discharge energy, and triggers the subsequent timing orchestration stage to evaluate whether to exit the stratified lean-burn mode and switch to conventional fuel, thereby prioritizing smooth engine operation when combustion stability decreases. This combustion phase and cycle fluctuation index are then used as inputs for closed-loop correction.

[0160] like Figure 5 As shown, under the closed-loop correction, the cyclic fluctuation index gradually converges from about 7.5% in the open-loop state to about 2.2% in the steady state, indicating that the closed-loop correction significantly improves combustion stability.

[0161] S520 generates corrections for injection quantity, injection phase, and discharge energy based on combustion phase deviation and stability deviation.

[0162] In this embodiment, the feedback module takes the deviation between the combustion characteristics and the target as input and generates a correction amount using incremental proportional-integral conversion. Taking combustion phase deviation-driven injection phase correction as an example, the injection phase correction amount is:

[0163]

[0164] Combustion phase deviation , and For proportional and integral gains. Similarly, the discharge energy correction is driven by the misfire rate and cycle fluctuation indicators, and the pulse width correction corresponding to the injection quantity is driven by the deviation of the measured excess air coefficient. Each correction has an upper limit on its amplitude to avoid closed-loop oscillation.

[0165] For example, if the target combustion phase is 8 degrees crankshaft rotation after top dead center, and the measured combustion phase is 11 degrees crankshaft rotation after top dead center, meaning combustion is delayed, then the combustion phase deviation is... crankshaft rotation angle, previous deviation crankshaft angle, proportional gain Integral gain Then the injection phase correction amount is The crankshaft angle correction advances injection and ignition by 0.6 degrees, placing it within a limit of no more than 3 degrees, thus bringing the combustion phase of the next cycle closer to the target. The upper limit of the injection phase correction is 3 degrees of crankshaft angle, and the upper limit of the discharge energy correction is 5 millijoules. This correction is then fed back to the aforementioned injection and discharge stages, superimposed on the calculated results of the injection pulse width, injection phase, and discharge energy.

[0166] S530, the cumulative amount of volatile gas reused.

[0167] In this embodiment, the feedback module weights and accumulates the target volatile gas mass for each cycle according to the degree of combustion completeness, and calculates the cumulative re-burning utilization of the volatile gas, expressed as: ,in The combustion efficiency factor is based on the degree of combustion completeness and is determined by the combustion phase and misfire determination. It is between 0 and 1.

[0168] For example, in one specific implementation, the target volatile gas mass for a single cycle is 0.485 mg, the combustion efficiency factor is 0.95, and the engine performs approximately 66.7 injections per second across four cylinders at 2000 rpm. Therefore, the volatile gas reburning utilization per unit time is... Milligrams per second, approximately 0.11 kilograms per hour, means the system re-burns and utilizes about 110 grams of volatile gas per hour. This cumulative re-burn utilization, along with the self-learning update, is stored in non-volatile memory for verifying the actual re-burning effect of fuel vapor recovery and refueling.

[0169] The S600 uses crankshaft phase as a reference to schedule the triggering sequence of each component, and when it detects that volatile gas is unavailable or combustion is abnormal, it gradually reduces the volatile gas injection quantity to zero and switches to conventional fuel supply.

[0170] This phase spans every engine cycle, uniformly scheduling the trigger times of each of the aforementioned stages based on the crankshaft phase, and performing a seamless mode switch when unavailable volatile gases or abnormal combustion is detected. In this embodiment, the collaborative scheduling module reads the crankshaft phase and speed, the status flags of each stage, and fault diagnosis information, establishes an event sequence table from monitoring, boost readiness, injection, ignition to feedback, manages data transmission and priority arbitration between stages, runs a fault detection degradation state machine, and outputs the trigger timing instructions for each stage, the system operating mode, and fault codes.

[0171] S610, within a single-cylinder cycle window, arranges the trigger times of each stage according to the crankshaft phase.

[0172] like Figure 6 As shown, the collaborative scheduling module arranges the triggering times of each stage within the time window of a single-cylinder cycle to ensure the entire link is closed. The triggering angle of each event is obtained by superimposing the reference phase with its respective phase offset, and the time margin between adjacent events must be greater than the execution time of that stage.

[0173] For example, at 2000 rpm, a single-cylinder cycle corresponds to a 720-degree crankshaft rotation, or two revolutions in a four-stroke cycle, which corresponds to approximately 60 milliseconds. Monitoring is triggered early in the intake phase, boost readiness is verified during the intake phase, injection is executed 39 degrees before top dead center, ignition is executed 15 degrees before top dead center, and feedback is collected during the expansion and exhaust phases and completed at the end of the cycle. The interval between the injection phase and the ignition phase is 24 degrees of crankshaft rotation, corresponding to 2 milliseconds at this speed, which is precisely the mixing and transport time, meeting the window requirement for the enrichment layer to form at the ignition moment, and is consistent with the stratified injection process. When there is competition for resources in the electronic control unit, priority scheduling is carried out according to the priorities of safety protection, ignition, injection, boost, monitoring, and feedback.

[0174] The S620 operates a fault detection and degradation state machine that seamlessly switches to normal fuel supply in case of an anomaly.

[0175] In the S620, the collaborative scheduling module operates a fault detection and degradation state machine encompassing three states: normal, degraded, and protection. In the normal state, volatile gas is used for combustion; in the degraded state, conventional fuel is used; and in the protection state, volatile gas injection and pressurization are stopped. When the volatile gas availability flag is 0, the pressurization readiness flag is 0, or the cyclic fluctuation index exceeds the limit, the system transitions from the normal state to the degraded state. When these conditions recover for several consecutive cycles, the system returns from the degraded state to the normal state. When safety faults such as overpressure, knocking, or coil overcurrent are detected, the system transitions from any state to the protection state. Mode switching must be seamless, meaning the volatile gas injection quantity gradually returns to zero and is synchronously compensated by conventional fuel; the engine must not stall.

[0176] For example, when the fuel tank's volatile gases are depleted during driving, and the concentration drops to 1% by volume, causing the volatile gas availability flag to be 0, the state machine transitions from a normal state to a degraded state. The volatile gas injection quantity gradually returns to zero over three cycles, with conventional fuel providing synchronous compensation. The engine transitions smoothly without stalling, and the system operating mode switches to conventional fuel. As another example, when knock intensity exceeds the limit, the state machine transitions from an arbitrary state to a protection state, immediately stopping volatile gas injection and boost, reversing the ignition phase, and recording a knock fault code. Yet another example is when the crankshaft signal is lost, causing the timing reference to fail, directly entering a protection state and switching to a limited operating mode with conventional fuel. Fault codes are written to the diagnostic area to support on-board diagnostic reading.

[0177] like Figure 2 As shown, the engine plasma combustion-assisted system based on active volatile gas pressurization and stratification provided in this application includes a volatile gas state monitoring module, an active pressurization module, a stratified injection control module, a plasma combustion-assisted module, and a controller connected to each of the above modules, wherein:

[0178] The volatile gas state monitoring module includes a concentration sensor, a pressure sensor, and a temperature sensor located in the fuel tank area, as well as a signal conditioning and acquisition circuit connected to the aforementioned sensors. The volatile gas state monitoring module acquires the concentration and pressure of the volatile gas in the fuel tank in real time, filters and estimates the concentration observations, obtains its rate of change and combustible quality index, and outputs the concentration, pressure, concentration change rate, combustible quality index, and volatile gas availability flag to the controller.

[0179] The active boosting module includes a high-frequency micro air pump, a boosting chamber with a pressure sensor and a pressure relief valve, and an air pump drive circuit. The air inlet of the high-frequency micro air pump is connected to the fuel tank, and the air outlet is connected to the boosting chamber. The active boosting module actively boosts the volatile gas from the fuel tank to the target supply pressure according to the concentration and pressure. When the target supply pressure is higher than the single-stage boosting capacity, the volatile gas is boosted in two stages in series. The pressure relief valve opens to protect the boosting chamber when the pressure exceeds the limit.

[0180] The stratified injection control module includes an independent volatile gas injector installed near the spark plug and facing the spark plug gap, an injection drive stage, and an injection strategy subtask running on the controller. Based on the concentration and the overall target excess air coefficient in the cylinder, the stratified injection control module determines the target volatile gas injection quantity, resolves the injection pulse width and injection phase, and injects the pressurized volatile gas from the intake end to the compression section through the independent volatile gas injector to the area around the spark plug, forming a stratified mixture that is rich around the spark plug and generally thin inside the cylinder.

[0181] The plasma combustion-assisted module includes a high-energy plasma ignition coil, a discharge energy modulation drive with a controllable energy storage capacitor and a switch, and a discharge voltage and current sampling circuit. The plasma combustion-assisted module modulates the plasma discharge energy according to the overall lean-out level in the cylinder, driving the high-energy plasma ignition coil to generate a continuous arc discharge in the spark plug gap. This arc discharges an active ignition nucleus in the enriched area around the spark plug, igniting the lean mixture in the cylinder. The discharge voltage and current sampling circuit samples the discharge process to obtain the actual discharge energy.

[0182] The controller is an engine electronic control unit, connected to the aforementioned modules, coordinating the operation of the volatile gas state monitoring module, the active boost module, the stratified injection control module, and the plasma combustion-assisted module. The controller also collects combustion feedback, extracts combustion phase and cycle fluctuation indicators, performs closed-loop correction of the target volatile gas injection quantity, the injection phase, and the plasma discharge energy, calculates the cumulative reburning utilization of volatile gas, and programs the triggering sequence of each module based on the crankshaft phase. When volatile gas unavailability or combustion abnormality is detected, the volatile gas injection quantity is gradually reduced to zero and the system switches to conventional fuel supply. The controller is connected to non-volatile memory, which stores calibration coefficients, self-learning update values, and cumulative reburning utilization.

[0183] Figure 7 As shown, compared to passive desorption relying on intake manifold vacuum, this system actively pressurizes the volatile gas to supply it on demand under various operating conditions, including full load, enabling the recovery and resupply of volatile gas under different engine operating conditions. By targeting the pressurized volatile gas to the spark plug area, a locally enriched but generally lean stratification is formed, concentrating highly reactive volatile gas components at the ignition point rather than diluting them with the entire intake air. By modulating the plasma discharge energy according to the overall leanness of the cylinder and activating active ignition nuclei in the enriched area around the spark plug, the air-fuel mixture can still be reliably ignited even under generally lean cylinder conditions. The above technologies work together to ensure that volatile gas is recovered and resupplyed for combustion under various engine operating conditions, while improving the ignition reliability of the generally lean cylinder mixture.

[0184] Those skilled in the art should understand that, in practice, the above modules can be implemented by corresponding hardware components in conjunction with the controller to execute instructions stored in the memory. The above descriptions are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

[0185] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit described above can be implemented in hardware.

[0186] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An engine plasma combustion-assisted system based on active volatile gas pressurization and stratification, characterized in that, include: The volatile gas state monitoring module is used to acquire the concentration and pressure of volatile gases in the engine fuel tank in real time. An active pressurization module is used to actively pressurize the volatile gas from the fuel tank to the target supply pressure based on the concentration and the pressure. The stratified injection control module is used to determine the target injection amount of volatile gas based on the concentration and the overall target excess air coefficient in the cylinder, and to reverse the injection pulse width and injection phase. The pressurized volatile gas is then injected into the area around the ignition point through an independent injection channel from the intake end to the compression section, forming a stratification that is rich around the ignition point and thin overall in the cylinder. The plasma combustion-assisted module is used to modulate the plasma discharge energy according to the overall leanness of the cylinder, and to excite active ignition nuclei in the enriched area around the ignition point to ignite the overall lean mixture in the cylinder. The controller is used to coordinate the operation of the volatile gas state monitoring module, the active pressurization module, the stratified injection control module, and the plasma combustion-supporting module.

2. The system according to claim 1, characterized in that, The active pressurization module includes a high-frequency micro air pump, which pressurizes the volatile gas in two stages in series when the target supply pressure is higher than the single-stage pressurization capacity.

3. The system according to claim 1, characterized in that, The layered injection control module is also used to advance the injection phase relative to the ignition phase according to the crankshaft angular velocity and the mixing transport time, so that the enriched area around the ignition point is formed at the ignition moment.

4. The system according to claim 3, characterized in that, The stratified injection control module is also used to determine the target injection quantity of volatile gas based on the volume of the enrichment zone in the vicinity of the ignition site, the target local equivalence ratio and the overall target excess air coefficient in the cylinder, and output the volatile gas through a small flow injector.

5. The system according to claim 1, characterized in that, The volatile gas state monitoring module is also used to filter and estimate the concentration and obtain its rate of change and combustible quality index. The controller determines the availability of volatile gas based on the combustible quality index and provides the combustible quality index to the plasma combustion-supporting module for modulating the plasma discharge energy.

6. A method for controlling engine plasma combustion based on active pressurization and stratification of volatile gases, characterized in that, include: Real-time acquisition of the concentration and pressure of volatile gases in the engine fuel tank; Based on the concentration and the pressure, the volatile gas from the fuel tank is actively pressurized to the target supply pressure; Based on the concentration and the overall target excess air coefficient in the cylinder, the target injection amount of volatile gas is determined and the injection pulse width and injection phase are solved. The pressurized volatile gas is then injected into the area around the ignition point through an independent injection channel from the intake end to the compression section, forming a layer that is rich around the ignition point and thin overall in the cylinder. The plasma discharge energy is modulated according to the overall leanness of the cylinder to stimulate active ignition nuclei in the enriched area around the ignition point, thereby igniting the overall lean mixture in the cylinder.

7. The method according to claim 6, characterized in that, The pressurized volatile gas is injected into the area around the ignition point. Based on the crankshaft angular velocity and mixing transport time, the injection phase is advanced relative to the ignition phase, so that the enriched area around the ignition point is formed at the moment of ignition.

8. The method according to claim 6, characterized in that, It also includes collecting combustion feedback, extracting combustion phase and cycle fluctuation indicators, closing-loop correcting the target injection amount of volatile gas, the injection phase and the plasma discharge energy, and calculating the cumulative re-burning utilization of volatile gas.

9. The method according to claim 6, characterized in that, It also includes arranging the triggering sequence of the acquisition, the boost, the injection and the excitation based on the crankshaft phase, and gradually reducing the volatile gas injection quantity to zero and switching to conventional fuel supply when volatile gas is detected to be unavailable or combustion is abnormal.

Images