Engine protection control method and system based on combustion condition of air cylinder

Abstract

The invention discloses an engine protection control method and system based on the combustion condition of an air cylinder, and the method comprises the steps: periodically sampling operation parameters of an engine, and determining a pressure ratio coefficient according to the actual pressure of an outlet or an inlet of a mixing valve and the actual pressure of an outlet or an inlet of a throttle valve when it is judged that an EGR valve is changed from meeting to not meeting EGR closed-loop enabling conditions; according to the pressure ratio coefficient and the current real-time rotating speed of the engine, the duration initial value when the target opening degree of the EGR valve is kept unchanged is determined; updating the residual duration value when the target opening degree of the EGR valve needs to be kept unchanged at the current moment; performing self-learning updating on the residual duration value self-learning coefficient and the supercharger response time self-learning coefficient; carrying out secondary self-learning updating on the residual duration value self-learning coefficient; and according to the residual duration value self-learning coefficient after secondary self-learning updating, a final residual duration value with the target opening degree of the EGR valve kept unchanged is obtained. According to the method, engine knock protection can be improved, and supercharging control response precision can be optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of engine control, and in particular to an engine protection control method and system based on cylinder combustion conditions. Background Art

[0002] Exhaust Gas Recirculation (EGR) takes exhaust gas from the exhaust and enters the intake system. Studies have shown that the EGR system has certain advantages in improving emissions, reducing fuel consumption and improving anti-knock capabilities.

[0003] The EGR closed-loop enabling conditions can be specifically found in patent CN118128651A "EGR valve target opening control method, device, equipment, medium and product". When the EGR closed-loop enabling conditions are met, a closed-loop control algorithm is used to control the actions of the mixing valve and the EGR valve to achieve the EGR rate follow-up. When the EGR closed-loop enabling conditions are not met, how to set the EGR valve target opening becomes a problem that needs to be solved. Therefore, the present invention proposes an engine protection control method and system based on cylinder combustion conditions from the aspects of knock and boost control accuracy. Summary of the invention

[0004] The main purpose of the present invention is to provide an engine protection control method and system based on cylinder combustion conditions to protect the engine when the EGR closed-loop enabling conditions are not met.

[0005] The technical solution adopted by the present invention is: an engine protection control method based on cylinder combustion conditions, comprising: periodically sampling the operating parameters of the engine, and when it is determined that the EGR valve changes from satisfying to not satisfying the EGR closed-loop enabling condition, determining the pressure ratio coefficient according to the actual pressure at the outlet or inlet of the mixing valve and the actual pressure at the outlet or inlet of the throttle valve;

[0006] Determine an initial value of the time period during which the EGR valve target opening is maintained unchanged according to the pressure ratio coefficient and the current real-time engine speed;

[0007] According to the change in the outlet pressure of the mixing valve, the correction coefficient determined based on the number of knocks, the duration self-learning update coefficient and the fluctuation of the difference between the target boost pressure and the actual boost pressure, the remaining duration value is updated while maintaining the target opening of the EGR valve unchanged;

[0008] Based on the difference between the updated remaining time value and the initial remaining time value and the change in the outlet pressure of the mixing valve, the remaining time value self-learning coefficient and the supercharger response time self-learning coefficient are updated through self-learning;

[0009] Based on the difference between the updated remaining time value and the initial remaining time value and the number of knocks of different knock intensities, the remaining time value self-learning coefficient is updated for a second time by self-learning;

[0010] The final remaining time value at which the target opening of the EGR valve remains unchanged is obtained by using the self-learning coefficient of the remaining time value after the secondary self-learning update.

[0011] According to the above technical solution, the method for determining the pressure ratio coefficient specifically includes: the pressure ratio coefficient is the product of the ratio of the actual pressure at the mixing valve outlet to the actual pressure at the mixing valve inlet and the pressure ratio coefficient correction coefficient, and the pressure ratio coefficient correction coefficient is calibrated by the ratio of the actual pressure at the throttle outlet to the actual pressure at the throttle inlet under the conditions of satisfying the pressure ratio coefficient correction coefficient calibration;

[0012] The pressure ratio coefficient correction coefficient calibration conditions include that, from the first sampling period in which the EGR closed-loop enabling condition is not met to the sampling period corresponding to the end of the initial value of the duration, when the target boost pressure remains unchanged, the fluctuation of the difference between the target boost pressure and the actual boost pressure is within a preset range.

[0013] According to the above technical solution, the sampling period corresponding to the end of the initial value of the duration is determined by the updated supercharger response time, and the updated supercharger response time τ=τ Boost ×(1+r Boost ), where τ Boost is the initial supercharger response time, r Boost It is the booster response time self-learning coefficient, the default value is 0.

[0014] The initial supercharger response time τ Boost The method for obtaining the time delay of the supercharger includes setting different EGR rates at different engine speeds and different actual intake air densities of the cylinders, and sampling data multiple times to obtain the average value of the supercharger response time.

[0015] According to the above technical solution, the fluctuation of the difference between the target boost pressure and the actual boost pressure is within the preset range.

[0016] p BoostErrFilter (N) = K BoostErr ×[p BoostErr (N)-p BoostErrFilter (N-1)]+p BoostErrFilter (N-1), from the first sampling period when the EGR closed-loop enabling condition is not met to the sampling period corresponding to the end of the initial value of the duration, all satisfy |p BoostErrFliter (N)-p BoostErr (N) <min[p BoostErr (N), p BoostErrFliter (N)]×r BoostErrLim , where p BoostErrFilter (N) is the pressure difference after first-order low-pass filtering in the Nth sampling period, m is the number of engine cylinders, n is the engine speed, k is the engine speedBoostErr is the preset pressure difference filter coefficient, p BoostErr (N) is the pressure difference p of the Nth sampling period BoostErr Original value, p BoostErrFilter (N-1) is the pressure difference after the first-order low-pass filter of the N-1th sampling period, r BoostErrLim is the preset limit factor.

[0017] According to the above technical solution, the method for obtaining the change in the mixing valve outlet pressure specifically includes: calculating the quotient of the initial supercharger response time and the preset sampling period, and taking the difference between the maximum value and the minimum value of the mixing valve outlet pressure within the sampling period corresponding to the quotient.

[0018] According to the above technical solution, the method for updating the remaining time value includes: updating the remaining time value t1'=t1×(1+k Knock )×[1+f(Δp AfMixAct , Δp BoostErr )]×(1+r t1 );

[0019] Among them, k Knock is the correction factor calibrated based on the number of knocks, k Knock =f(r PreRatio , Cnt LowKnock )×f(r PreRatio , Cnt MediumKnock )×f(r PreRatio , Cnt HighKnock ), which is calibrated based on ensuring that the fluctuation of the difference between the target boost pressure and the actual boost pressure is within a preset range within a period of time after the updated remaining time value ends and the preset time is delayed, and only low-intensity knocks less than the preset number of times occur in the cylinder; Δp AfMixAct is the change in the outlet pressure of the mixing valve, Δp BoostErr is the fluctuation of the difference between the target boost pressure and the actual boost pressure, f(Δp AfMixAct , Δp BoostErr ) is based on the difference between the target boost pressure and the actual boost pressure p BoostErr The fluctuation of is within the preset range; t1 It is the self-learning coefficient of the remaining time value.

[0020] According to the above technical solution, the method for performing initial self-learning update on the self-learning update coefficient of the duration and the self-learning coefficient of the supercharger response time specifically includes:

[0021] Calculate the difference between the updated remaining duration value and the initial remaining duration value;

[0022] According to the difference between the updated remaining time value and the initial remaining time value, the state of the current working condition is judged; the state includes upward learning state 1, upward learning state 2, downward learning state 1, downward learning state 2 and other states;

[0023] If the current working condition is in the upward learning state, the self-learning coefficient of the remaining time value increases; if the current working condition is in the downward learning state, the self-learning coefficient of the remaining time value decreases; if the current working condition is in other states, the self-learning coefficient of the remaining time value is not updated.

[0024] According to the above technical solution, the specific method for determining the state of the current working condition includes:

[0025] If the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C1 for more than the preset number of times CNT1, it is judged that the current working condition is in the upward learning state 1; if the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C1, and the difference between the updated remaining time value and the initial remaining time value in the last self-learning process is not greater than the preset value C2, it is judged that the current working condition is in the upward learning state 2;

[0026] If the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C2 for more than the preset number of times CNT2, it is judged that the current working condition is in downward learning state 1; if the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C2, and the difference between the updated remaining time value and the initial remaining time value in the last self-learning process is not greater than the preset value C1, it is judged that the current working condition is in downward learning state 2;

[0027] The remaining cases are judged as other states.

[0028] According to the above technical solution, when the current operating condition belongs to the upward learning state one, if the continuous number of times the mixing valve outlet pressure change exceeds the preset value A1 exceeds the preset number CNT1, the supercharger response time self-learning coefficient is increased by the preset value X1, and it is judged that the current operating condition belongs to the secondary upward learning state one; if the current operating condition belongs to the downward learning state one, if the continuous number of times the mixing valve outlet pressure change exceeds the preset value A2 exceeds the preset number CNT2, the supercharger response time self-learning coefficient is subtracted from the preset value X2, and it is judged that the current operating condition belongs to the secondary downward learning state one.

[0029] According to the above technical solution, in each self-learning update process, if the supercharger response time self-learning coefficient is updated, CNT1 and CNT2 are cleared.

[0030] According to the above technical scheme, the priorities of the sub-upward learning state 1, upward learning state 1, sub-downward learning state 1, downward learning state 1, upward learning state 2, downward learning state 2, and other working conditions are reduced in sequence. If it is judged that the current working condition is in a high priority state, the low priority state will no longer be judged.

[0031] According to the above technical solution, the method for secondary self-learning update includes:

[0032] The state of the current working condition is determined according to the difference between the updated remaining time value and the initial remaining time value and the number of knocks of different knock intensities;

[0033] The states include upward high-burst learning state 1, upward medium-burst learning state 1, upward low-burst learning state 1, upward-burst learning state 1, downward high-burst learning state 1, downward medium-burst learning state 1, downward low-burst learning state 1, downward-burst learning state 1 and other states;

[0034] If the current working condition is in the upward learning state, the self-learning coefficient of the remaining time value increases; if the current working condition is in the downward learning state, the self-learning coefficient of the remaining time value decreases; if the current working condition is in other states, the self-learning coefficient of the remaining time value is not updated.

[0035] According to the above technical solution, the method for determining the state of the current working condition specifically includes:

[0036] Under the condition that the current operating condition satisfies the condition that the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C1, if the number of knocks with high knock intensity exceeds the preset value A3 by more than the preset number CNT3, the preset value X3 is added to the self-learning coefficient of the remaining time value, and it is judged that the current operating condition is in the upward high-knock learning state one; if the number of knocks with medium knock intensity exceeds the preset value A4 by more than the preset number CNT4, the preset value X4 is added to the self-learning coefficient of the remaining time value, and it is judged that the current operating condition is in the upward medium-knock learning state one; if the number of knocks with low knock intensity exceeds the preset value A5 by more than the preset number CNT5, the preset value X5 is added to the self-learning coefficient of the remaining time value, and it is judged that the current operating condition is in the upward low-knock learning state one; if the current operating condition satisfies the condition that the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C1 by more than the preset number CNT6, the preset value X6 is added to the self-learning coefficient of the remaining time value, and it is judged that the current operating condition is in the upward knock learning state one;

[0037] Under the condition that the current operating condition satisfies the difference between the updated remaining time value and the initial remaining time value is not greater than the preset value C2, if the number of knocks with high knock intensity exceeds the preset value A6 by more than the preset number CNT7, the self-learning coefficient of the remaining time value is reduced by the preset value X7, and it is judged that the current operating condition is in the downward high-knock learning state one; if the number of knocks with medium knock intensity exceeds the preset value A7 by more than the preset number CNT8, the self-learning coefficient of the remaining time value is reduced by the preset value X8, and it is judged that the current operating condition is in the downward medium-knock learning state one; if the number of knocks with low knock intensity exceeds the preset value A8 by more than the preset number CNT9, the self-learning coefficient of the remaining time value is reduced by the preset value X9, and it is judged that the current operating condition is in the downward low-knock learning state one; if the current operating condition satisfies the difference between the updated remaining time value and the initial remaining time value is not greater than the preset value C2 by more than the preset number CNT 10 , then the remaining time value self-learning coefficient is reduced by the preset value X 10 , it is determined that the current working condition is in the downward shock learning state 1;

[0038] In other cases, the self-learning coefficient of the remaining duration value is not updated.

[0039] According to the above technical solution, in each self-learning update process, if the supercharger response time self-learning coefficient is updated, CNT3, CNT4, CNT5, CNT6, CNT7, CNT8, CNT9, CNT 10 Cleared.

[0040] According to the above technical scheme, the priorities of the upward high violent shock learning state 1, upward medium violent shock learning state 1, upward low violent shock learning state 1, upward violent shock learning state 1, downward high violent shock learning state 1, downward medium violent shock learning state 1, downward low violent shock learning state 1, downward violent shock learning state 1 and other states decrease in sequence. If it is judged that the current working condition is in a high priority state, the low priority state will no longer be judged.

[0041] Another aspect of the present invention provides an engine protection control system based on cylinder combustion conditions, which executes the above-mentioned engine protection control method based on cylinder combustion conditions.

[0042] The beneficial effects of the present invention are as follows: the present invention provides an engine protection control method and system based on cylinder combustion conditions, when the EGR valve does not meet the EGR closed-loop enabling conditions, the initial value of the time during which the EGR valve target opening remains unchanged is determined, and based on the change in the outlet pressure of the mixing valve and the number of knocks of different knock intensities, the remaining value of the time is updated in the process of maintaining the EGR valve target opening unchanged. The present invention optimizes the control of the EGR valve target opening after the EGR closed-loop enabling conditions are exited, improves engine knock protection, and optimizes the response accuracy of boost control.

[0043] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 is a flow chart of an engine protection control method based on cylinder combustion conditions according to an embodiment of the present invention;

[0046] Figure 2 is a logic diagram of an engine protection control method based on cylinder combustion conditions according to an embodiment of the present invention;

[0047] Figure 3 It is a structural diagram of an engine protection control system based on cylinder combustion conditions according to an embodiment of the present invention.

[0048] Figure numerals: 1. air filter; 2. mixing valve; 3. compressor; 4. throttle; 5. engine; 6. turbine; 7. catalyst; 8. particulate matter collector; 9. EGR cooler; 10. EGR valve; 11. temperature sensor; 12. differential pressure sensor. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0050] It should be noted that the illustrations provided in the embodiments of the present invention are only used to illustrate the basic concept of the present invention in a schematic manner. Therefore, the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout type may also be more complicated.

[0051] In the present invention, it is also necessary to explain that, if the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, if the terms "first" and "second" appear, they are only used for description and distinction purposes, and cannot be understood as indicating or implying relative importance.

[0052] Example 1

[0053] This embodiment provides an engine protection control method based on cylinder combustion conditions, and its process is as follows: Figure 1 As shown, the steps include:

[0054] S1, periodically sample the engine's operating parameters, and determine when the EGR valve changes from satisfying to not satisfying the EGR closed-loop enabling conditions, based on the actual pressure p at the outlet of the mixing valve. AfMixAct , Actual pressure at the mixing valve inlet p BfMixAct , Actual throttle outlet pressure p AfThrAct and the actual throttle inlet pressure p BfMixAct Determine the pressure ratio coefficient.

[0055] S101. Determine whether the EGR valve meets the EGR closed-loop enabling conditions. The EGR closed-loop enabling conditions can be specifically found in patent CN118128651A "EGR valve target opening control method, device, equipment, medium and product".

[0056] Specifically, it is determined whether the current operating condition meets the EGR closed-loop enabling condition in the previous sampling period, and the current sampling period does not meet the EGR closed-loop enabling condition. The sampling period of this example is 10ms.

[0057] S102, according to the actual pressure p of the mixing valve outlet AfMixAct The actual pressure p of the mixing valve inlet BfMixAct Ratio Actual throttle outlet pressure p AfThrAct The actual throttle inlet pressure p BfMixAct Ratio Together determine the pressure ratio coefficient r PreRatio .

[0058] Specifically, is the pressure ratio correction factor, and The larger the value, the worse the throttle control ability and the greater the risk of supercharger pressure control accuracy. Therefore, in order to avoid the problem of supercharger pressure control accuracy, The calibration parameters are shown in Table 1.

[0059]

[0060] Table 1

[0061] S2, according to the pressure ratio coefficient r PreRatio The initial value of the time for which the EGR valve target opening is maintained constant is determined by the current engine real-time speed n. The purpose of maintaining the EGR valve target opening constant is to maintain the stability of the EGR valve and prevent the boost inlet pressure from affecting the boost control responsiveness. That is, the EGR valve target opening pct is determined EGRDsrd =pct EGRDsrd (z) duration, where pct EGRDsrd (z) is the EGR valve target opening in the previous sampling cycle, that is, the EGR valve target opening when the EGR closed-loop condition is met in the last sampling cycle. The initial value of the duration t starts from the first sampling cycle in which the EGR closed-loop condition is not met.

[0062] S201, judging whether the initial value calibration condition of the duration is met.

[0063] Specifically, the judgment basis is that, when the target boost pressure is fixed, p BoostErrFilter (N) = K BoostErr ×[p BoostErr (N)-p BoostErrFilter (N-1)]+p BoostErrFilter (N-1), N after the first sampling period when the EGR closed-loop enabling condition is not met to the end of the initial value t of the duration Boost All sampling periods satisfy |p BoostFilter (N)-p BoostErr (N)| <min[p BoostErr (N), p BoostErrFliter (N)]×r BoostErrLim .

[0064] The N Boost The value depends on Supercharger response time τ = τ Boost ×(1+r Boost ), where r Boost is the initial supercharger response time τ Boost The self-learning coefficient has a default value of 0 and can be saved after the vehicle is powered off. The initial supercharger response time τ Boost is the time it takes for the mixture to flow from the supercharger compressor to the throttle valve outlet, which can be adjusted at different engine speeds n eng and the different actual intake air density rho of the cylinder Act The result is obtained by setting different EGR rates for multiple sampling data and taking the average value.

[0065] Among them, p BoostErrFilter (N) is the pressure difference after first-order low-pass filtering in the Nth sampling period, m is the number of engine cylinders, n is the engine speed, k is the engine speed BoostErr is the preset pressure difference filter coefficient, p BoostErr (N) is the pressure difference p of the Nth sampling period BoostErr Original value, p BoostErrFilter (N-1) is the pressure difference after the first-order low-pass filter of the N-1th sampling period, r BoostErrLim is a preset limit coefficient, which is 0.1 in this embodiment.

[0066] S202. Calibrate the initial value t of the duration on the premise that the initial value calibration conditions of the duration are met.

[0067] Specifically, the initial value of the duration t=f(r PreRatio ,n), and the pressure ratio coefficient r PreRatio In direct proportion to the actual engine speed n, and inversely proportional to the actual engine speed n. To meet the difference between the target boost pressure and the actual boost pressure p BoostErr The fluctuation of the pressure ratio coefficient r is within the preset range. When the actual engine speed n is the same, if the pressure ratio coefficient r PreRatio The smaller the value, the smaller the initial value t of the duration; in the pressure ratio coefficient r PreRatio Under the same circumstances, if the speed n is smaller, the initial value t of the duration is larger.

[0068] S3, based on the change in the outlet pressure of the mixing valve Δp AfMixAct , the knock frequency and duration self-learning update coefficient r of different knock intensities t1 , while maintaining the EGR valve target opening unchanged, the remaining time value t1 is updated. Excessive mixing valve outlet pressure change Δp AfMixAct This will have a bad impact on the stability of the boost pressure control, and thus affect the stability of the subsequent boost pressure control. Therefore, the mixing valve outlet pressure change Δp needs to be considered in this step. AfMixAct Preferably, the remaining duration value t1 is limited to a maximum value t max and the minimum value t min In order to avoid excessive adjustment and poor control robustness of the intake system, in this embodiment, t max Take 0.2s, t min Take 0s.

[0069] S301, obtaining the change in the outlet pressure of the mixing valve Δp AfMixAct and the correction factor k determined based on the number of knocks Knock .

[0070] Specifically, the mixing valve outlet pressure change ΔpAfMixAct Take the nearest The mixing valve outlet pressure p in the sampling period AfMixAct Maximum and mixing valve outlet pressure p AfMixAct Minimum difference.

[0071] If the mixing valve outlet pressure p AfMixAct The maximum value occurs later than the mixing valve outlet pressure p AfMixAct When the minimum value appears, the mixing valve outlet pressure changes by Δp AfMixAct is a positive value; if the mixing valve outlet pressure p AfMixAct The maximum value appears earlier than the mixing valve outlet pressure p AfMixAct When the minimum value appears, the mixing valve outlet pressure changes by Δp AfMixAct is a negative value; if the mixing valve outlet pressure p AfMixAct Maximum value and mixing valve outlet pressure p AfMixAct The minimum value appears at the same time, then the mixing valve outlet pressure change Δp AfMixAct is 0.

[0072] k Knock is the knock number and pressure ratio coefficient r based on different knock intensities PreRatio Determined correction factor, k Knock =f(r PreRatio , Cnt LowKnock )×f(r PreRatio , Cnt MediumKnock )×f(r PreRatio , Cnt HighKnock ), while ensuring that the difference between the target boost pressure and the actual boost pressure is BoostErr The fluctuation of k is calibrated under the condition that it is within the preset range. Knock , satisfying the end at t1' and delaying the preset time τ EGR After a period of time, that is, the engine completes 20 working cycles, 1 working cycle means that each cylinder of the engine completes 4 working strokes, and at most only a preset number of low-intensity knocks occur, which is 5 in this embodiment. τ EGR is the time it takes for exhaust gas to flow from the EGR valve to the cylinder, which can be calculated at different engine speeds n eng and the actual intake air density rho of different cylinders Act The average value of multiple sampling data at different EGR rates is obtained.

[0073] In the knock frequency Cnt LowKnock , Cnt MediumKnock , Cnt HighKnock Under the same condition, the pressure ratio coefficient r PreRatio The smaller the value, the correction factor k determined based on the number of knocks of different knock intensities KnockThe smaller the pressure ratio coefficient r PreRatio Under the same conditions, the knock frequency Cnt LowKnock , Cnt MediumKnock , Cnt HighKnock The larger the value, the greater f(r PreRatio , Δ EGRRatioErr ) is larger, to suppress the occurrence of knock.

[0074] S302, on the premise of satisfying the calibration basis, complete f(Δp AfMixAct , Δp BoostErr ) calibration.

[0075] f(Δp AfMixAct , Δp BoostErr ) The calibration basis is to meet the difference between the target boost pressure and the actual boost pressure p BoostErr Under the premise that the fluctuation of boost pressure difference is within the preset range, BoostErr Under the same conditions, the mixing valve outlet pressure p AfMixAct Change Δp AfMixAct The smaller the f(Δp AfMixAct , Δp BoostErr ) is smaller; at the outlet pressure p of the mixing valve AfMixAct Change Δp AfMixAct Under the same conditions, if the boost pressure difference change Δp BoostErr The larger the value, the greater f(Δp AfMixAct , Δp BoostErr ) is larger.

[0076] S4, based on the difference between the updated remaining time value t1' and the initial remaining time value t1, the mixing valve outlet pressure change Δp AfMixAct , update coefficient r for self-learning of duration t1 and the self-learning coefficient r of the booster response time Boost Perform self-learning updates.

[0077] Determine whether the current working condition meets the following conditions in sequence, with priority from high to low. When a high priority state is determined, the low priority state will no longer be determined.

[0078] (1) Secondary upward learning state 1: If t1'-t1 is greater than the preset value C1, which is 0.1s in this example, it means that the adjustment is too large due to the excessive fluctuation of the boost pressure difference, and Δp AfMixAct exceeds the preset value A1, which is ±20 kPa in this example, and the number of consecutive occurrences exceeds the preset number CNT1, which is 5 in this embodiment, indicating that the influence of the boost pressure difference causes the time t1 to increase continuously. In order to reduce the influence of the boost pressure difference fluctuation, r t1 =r t1 (z)+0.05, r Boost=r Boost (z)+0.1, where r t1 (z) is the learning value stored at the last learning time t1, r Boost (z) is the r stored in the last learning Boost Learning value. CNT1 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT1 is updated at most once during each driving cycle.

[0079] (2) Upward learning state 1: If t1'-t1 is greater than the preset value C1, which is 0.1s in this example, it means that the adjustment is too large due to the excessive fluctuation of the boost pressure difference, and the number of consecutive occurrences exceeds the preset number CNT1, which is 5 in this embodiment, indicating that the influence of the boost pressure difference causes the duration t1 to increase continuously. In order to reduce the influence of the boost pressure difference fluctuation, r t1 =r t1 (z) + 0.03. CNT1 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT1 is updated at most once during each driving cycle.

[0080] (3) Secondary downward learning state 1: If t1'-t1 is not greater than the preset value C2, which is -0.1s in this example, it means that the adjustment is too large due to the excessive fluctuation of the boost pressure difference, and Δp AfMixAct The value exceeds the preset value A2. In this example, ±20 kPa is taken. The number of consecutive occurrences exceeds the preset number CNT2, indicating that the influence of the boost pressure difference causes the duration t1 to decrease. In order to reduce the influence of the pressure difference fluctuation, r t1 =r t1 (z)-0.02, r Boost =r Boost (z) -0.1. CNT2 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT2 is updated at most once during each driving cycle.

[0081] (4) Downward learning state 1: If t1'-t1 is not greater than the preset value C2, which is -0.1s in this example, it means that the adjustment is too large due to the excessive fluctuation of the boost pressure difference, and the number of consecutive occurrences exceeds the preset number CNT2, indicating that the influence of the boost pressure difference causes the duration t1 to decrease continuously. In order to reduce the influence of the boost pressure difference fluctuation, r t1 =r t1 (z) -0.04. CNT2 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT2 is updated at most once during each driving cycle.

[0082] (5) Upward learning state 2: If t1'-t1 is greater than the preset value C1, which is 0.1s in this example, and t1'-t1 is not greater than the preset value C2 during the previous self-learning process, the adjustment fluctuation is too large, which has a greater impact on the pressure fluctuation. t1 =r t1 (z) + 0.02. The newly learned and stored learning value is used the next time the same situation is judged.

[0083] (6) Downward learning state 2: If t1'-t1 is greater than the preset value C2, in this example, -0.1s, and t1'-t1 in the previous self-learning process is not greater than the preset value C1, the adjustment fluctuation is too large, and the pressure fluctuation has a greater impact at this time, then r t1 =r t1 (z) -0.02. The newly learned and stored learning value is used the next time the judgment is made in this situation.

[0084] (7) Other states that do not satisfy the above states, r t1 and r Boost Remain unchanged.

[0085] S5. After completing the initial self-learning update, based on the difference between the remaining duration value and the initial duration value, and the number of knocks of different knock intensities, a secondary self-learning update is performed on the duration self-learning update coefficient.

[0086] Determine whether the current working condition meets the following conditions in sequence, with priority from high to low. When a high priority state is determined, the low priority state will no longer be determined.

[0087] (1) Upward high knock learning state 1: If t1'-t1 is greater than the preset value C1, which is 0.1s in this example, and the knock number Cnt of high knock intensity is satisfied at the same time HighKnock exceeds the preset value A3, which is 5 in this example, and the number of consecutive occurrences exceeds the preset number CNT3, which is 5 in this example, indicating that the impact of knock protection causes the duration t1 to increase continuously. In order to suppress the impact of engine knock, r t1 =r t1 (z)+0.05, and CNT3 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT3 is updated at most once during each driving cycle.

[0088] (2) Towards medium knock learning state 1: If t1'-t1 is greater than the preset value C1, which is 0.1s in this example, and the knock number Cnt of medium knock intensity MediumKnock exceeds the preset value A4, which is 5 in this example, and the number of consecutive occurrences exceeds the preset number CNT4, which is 5 in this example, indicating that the impact of knock protection causes the duration t1 to increase continuously. In order to suppress the impact of engine knock, r t1=r t1 (z)+0.025, and CNT4 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT4 is updated at most once during each driving cycle.

[0089] (3) Upward low knock learning state 1: If t1'-t1 is greater than the preset value C1, which is 0.1s in this example, and the knock number Cnt of low knock intensity LowKnock exceeds the preset value A5, which is 5 in this example, and the number of consecutive occurrences exceeds the preset number CNT5, which is 5 in this example, indicating that the impact of knock protection causes the duration t1 to increase continuously. In order to suppress the impact of engine knock, r t1 =r t1 (z)+0.01, and CNT5 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT5 is updated at most once during each driving cycle.

[0090] (4) Upward knock learning state 1: If t1'-t1 is greater than the preset value C1, which is 0.1s in this example, and the number of consecutive occurrences exceeds the preset number CNT6, which is 5 in this example, r t1 =r t1 (z)+0.005, and CNT6 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT6 is updated at most once during each driving cycle.

[0091] (5) Downward high knock learning state 1: If t1'-t1 is greater than the preset value C2, in this example, it is -0.1s, and the knock number Cnt of high knock intensity HighKnock exceeds the preset value A6, which is 5 in this example, and the number of consecutive occurrences exceeds the preset number CNT7, which is 5 in this example, indicating that the impact of knock protection causes the duration t1 to increase continuously. In order to suppress the impact of engine knock, r t1 =r t1 (z) -0.02, and CNT7 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT7 is updated at most once during each driving cycle.

[0092] (6) Downward medium knock learning state 1: If t1'-t1 is greater than the preset value C2, in this example, it is -0.1s, and the knock number Cnt of medium knock intensity MediumKnock exceeds the preset value A7, which is 5 in this example, and the number of consecutive occurrences exceeds the preset number CNT8, which is 5 in this example, indicating that the impact of knock protection causes the duration t1 to increase continuously. In order to suppress the impact of engine knock, r t1 =r t1(z) -0.03, and CNT8 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT8 is updated at most once during each driving cycle.

[0093] (7) Downward low knock learning state 1: If t1'-t1 is greater than the preset value C2, in this example, it is -0.1s, and the knock number Cnt of low knock intensity LowKnock exceeds the preset value A8, which is 5 in this example, and the number of consecutive occurrences exceeds the preset number CNT9, which is 5 in this example, indicating that the impact of knock protection causes the duration t1 to increase continuously. In order to suppress the impact of engine knock, r t1 =r t1 (z) -0.04, and CNT9 is cleared at the same time. The newly learned and stored learning value is used the next time the situation is judged. CNT9 is updated at most once during each driving cycle.

[0094] (8) Downward knock learning state 1: If t1'-t1 is greater than the preset value C2, in this example -0.1s, and the number of consecutive occurrences exceeds the preset number CNT 10 , in this example, we take 5, r t1 =r t1 (z)-0.07, while CNT 10 Clear. The newly learned and stored learning value is used the next time the situation is judged. 10 Updated at most once during each driving cycle.

[0095] (9) Other states that do not satisfy the above states, r t1 Remain unchanged.

[0096] S6. Obtain the final remaining time value for maintaining the target opening of the EGR valve unchanged according to the time self-learning update coefficient after the second self-learning update.

[0097] Specifically, based on the calculation formula in S3

[0098] t1'=t1×(1+k Knock )×[1+f(Δp AfMixAct , Δp BoostErr )]×(1+r t1 ) and r determined by self-learning update t1 Calculate the final remaining duration value that remains unchanged.

[0099] In summary, the method logic of this embodiment is as follows: Figure 2As shown, based on the basic setting that the EGR valve opening needs to be maintained unchanged after the EGR closed-loop enabling condition is exited, the initial value of the time for maintaining the EGR valve opening unchanged is determined, and during the maintenance process, the remaining maintenance time is optimized based on the occurrence of different levels of knock and boost response conditions, so as to obtain the final EGR closed-loop enabling exit EGR valve opening maintenance time, thereby achieving improvements in engine knock protection and boost control response accuracy.

[0100] Example 2

[0101] This embodiment provides an engine protection control system based on cylinder combustion. This system executes the engine protection control method based on cylinder combustion described in Embodiment 1. The structure is as follows: Figure 3 As shown, it includes: an air filter 1; a mixing valve 2; a compressor 3; a throttle 4; an engine 5; a turbine 6; a catalyst 7; a particulate matter collector 8; an EGR cooler 9; an EGR valve 10; a temperature sensor 11; and a pressure difference sensor 12.

[0102] In summary, the present invention provides an engine protection control method and system based on cylinder combustion conditions, based on the basic setting that the EGR valve opening needs to be maintained unchanged after the EGR closed-loop enabling condition is exited, the present invention first determines the initial value of the time length for maintaining the EGR valve opening unchanged, and during the maintenance process, the remaining maintenance time length is optimized based on the number of occurrences of different levels of knock and the boost response conditions, so as to obtain the final EGR valve opening maintenance time, thereby achieving improvements in engine knock protection and boost control response accuracy.

[0103] It should be pointed out that, according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, and two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0104] The order of execution of each step in the above embodiment does not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0105] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. An engine protection control method based on cylinder combustion conditions, characterized in that: include: The operating parameters of the engine are sampled periodically, and when it is determined that the EGR valve changes from satisfying to not satisfying the EGR closed-loop enabling condition, the pressure ratio coefficient is determined according to the actual pressure at the outlet or inlet of the mixing valve and the actual pressure at the outlet or inlet of the throttle valve; Determine an initial value of the time period during which the EGR valve target opening is maintained unchanged according to the pressure ratio coefficient and the current real-time engine speed; The remaining time required to maintain the target opening of the EGR valve unchanged at the current moment is updated according to the change in the outlet pressure of the mixing valve, the correction coefficient determined based on the number of knocks, the time self-learning update coefficient, and the fluctuation of the difference between the target boost pressure and the actual boost pressure; Based on the difference between the updated remaining time value and the initial remaining time value and the change in the outlet pressure of the mixing valve, the remaining time value self-learning coefficient and the supercharger response time self-learning coefficient are updated through self-learning; Based on the difference between the updated remaining time value and the initial remaining time value and the number of knocks of different knock intensities, the remaining time value self-learning coefficient is updated by secondary self-learning; The final remaining time value at which the target opening of the EGR valve remains unchanged is obtained by using the self-learning coefficient of the remaining time value after the secondary self-learning update.

2. The engine protection control method based on cylinder combustion conditions according to claim 1, characterized in that: The method for determining the pressure ratio coefficient specifically includes: the pressure ratio coefficient is the product of the ratio of the actual pressure at the mixing valve outlet to the actual pressure at the mixing valve inlet and the pressure ratio coefficient correction coefficient, and the pressure ratio coefficient correction coefficient is calibrated by the ratio of the actual pressure at the throttle outlet to the actual pressure at the throttle inlet under the conditions of satisfying the pressure ratio coefficient correction coefficient calibration; The pressure ratio coefficient correction coefficient calibration conditions include that, from the first sampling period in which the EGR closed-loop enabling condition is not met to the sampling period corresponding to the end of the initial value of the duration, when the target boost pressure remains unchanged, the fluctuation of the difference between the target boost pressure and the actual boost pressure is within a preset range.

3. The engine protection control method based on cylinder combustion conditions according to claim 2, characterized in that: The sampling period corresponding to the end of the initial value of the duration is determined by the updated supercharger response time, and the updated supercharger response time τ=τ Boost ×(1+r Boost ), where τ Boost is the initial supercharger response time, r Boost It is the self-learning coefficient of the booster response time, and the default value is 0; The initial supercharger response time τ Boost The method for obtaining the time delay of the supercharger includes setting different EGR rates at different engine speeds and different actual intake air densities of the cylinders, and sampling data multiple times to obtain the average value of the supercharger response time.

4. The engine protection control method based on cylinder combustion conditions according to claim 3, characterized in that: The fluctuation of the difference between the target boost pressure and the actual boost pressure is within the preset range. BoostErrFilter (N) = K BoostErr ×[p BoostErr (N)-p BoostErrFilter (N-1)]+p BoostErrFilter (N-1), from the first sampling period when the EGR closed-loop enabling condition is not met to the sampling period corresponding to the end of the initial value of the duration, all satisfy |p BoostErrFliter (N)-p BoostErr (N)| <min[p BoostErr (N), p BoostErrFliter (N)]×r BoostErrLim , where p BoostErrFilter (N) is the pressure difference after first-order low-pass filtering in the Nth sampling period, m is the number of engine cylinders, n is the engine speed, k is BoostErr is the preset pressure difference filter coefficient, p BoostErr (N) is the pressure difference p of the Nth sampling period BoostErr Original value, p BoostErrFilter (N-1) is the pressure difference after the first-order low-pass filter of the N-1th sampling period, r BoostErrLim is the preset limit factor.

5. The engine protection control method based on cylinder combustion conditions according to claim 3, characterized in that: The method for obtaining the mixing valve outlet pressure variation specifically includes: calculating the quotient of the initial supercharger response time and a preset sampling period, and taking the difference between the maximum mixing valve outlet pressure and the minimum mixing valve outlet pressure in the sampling period corresponding to the quotient.

6. The engine protection control method based on cylinder combustion conditions according to claim 5, characterized in that: The method for updating the remaining duration value includes: updating the remaining duration value t1'=t1×(1+k Knock )×[1+f(Δp AfMixAct , Δp BoostErr )]×(1+r t1 ); Among them, k Knock is the correction factor calibrated based on the number of knocks, k Knock =f(r PreRatio , Cnt LowKnock )×f(r PreRatio , Cnt MediumKnock )×f(r PreRatio , Cnt HighKnock ), Cnt LowKnock is the number of knocks with low knock intensity, Cnt MediumKnock is the knock number of medium knock intensity, Cnt HighKnock is the number of detonations with high detonation intensity, r PreRatio is the pressure ratio coefficient, which is calibrated based on ensuring that the fluctuation of the difference between the target boost pressure and the actual boost pressure is within a preset range within a period of time after the updated remaining time value ends and the preset time is delayed, and only low-intensity knocks less than the preset number of times occur in the cylinder; Δp AfMixAct is the change in the outlet pressure of the mixing valve, Δp BoostErr is the fluctuation of the difference between the target boost pressure and the actual boost pressure, f(Δp AfMixAct , Δp BoostErr ) is based on the difference between the target boost pressure and the actual boost pressure p BoostErr The fluctuation of is within the preset range; t1 It is the self-learning coefficient of the remaining time value.

7. The engine protection control method based on cylinder combustion conditions according to claim 6, characterized in that: The method for performing initial self-learning update on the self-learning update coefficient of the duration and the self-learning coefficient of the supercharger response time specifically includes: Calculate the difference between the updated remaining duration value and the initial remaining duration value; According to the difference between the updated remaining time value and the initial remaining time value, the state of the current working condition is judged; the state includes upward learning state 1, upward learning state 2, downward learning state 1, downward learning state 2 and other states; If the current working condition is in the upward learning state, the self-learning coefficient of the remaining time value increases; if the current working condition is in the downward learning state, the self-learning coefficient of the remaining time value decreases; if the current working condition is in other states, the self-learning coefficient of the remaining time value is not updated.

8. The engine protection control method based on cylinder combustion conditions according to claim 7, characterized in that: The specific method for determining the state of the current working condition includes: If the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C1 for more than the preset number of times CNT1, it is judged that the current working condition is in the upward learning state 1; if the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C1, and the difference between the updated remaining time value and the initial remaining time value in the last self-learning process is not greater than the preset value C2, it is judged that the current working condition is in the upward learning state 2; If the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C2 for more than the preset number of times CNT2, it is judged that the current working condition is in downward learning state 1; if the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C2, and the difference between the updated remaining time value and the initial remaining time value in the last self-learning process is not greater than the preset value C1, it is judged that the current working condition is in downward learning state 2; The remaining cases are judged as other states.

9. The engine protection control method based on cylinder combustion conditions according to claim 8, characterized in that: When the current operating condition belongs to the upward learning state one, if the continuous number of times the mixing valve outlet pressure change exceeds the preset value A1 exceeds the preset number CNT1, the supercharger response time self-learning coefficient is increased by the preset value X1, and it is judged that the current operating condition belongs to the secondary upward learning state one; if the current operating condition belongs to the downward learning state one, if the continuous number of times the mixing valve outlet pressure change exceeds the preset value A2 exceeds the preset number CNT2, the supercharger response time self-learning coefficient is subtracted from the preset value X2, and it is judged that the current operating condition belongs to the secondary downward learning state one.

10. The engine protection control method based on cylinder combustion conditions according to claim 9, characterized in that: During each self-learning update process, if the supercharger response time self-learning coefficient is updated, CNT1 and CNT2 are cleared.

11. The engine protection control method based on cylinder combustion conditions according to claim 10, characterized in that: The priorities of the sub-upward learning state 1, upward learning state 1, sub-downward learning state 1, downward learning state 1, upward learning state 2, downward learning state 2, and other working conditions decrease in sequence. If the current working condition is judged to be in a high priority state, the low priority state will no longer be judged.

12. The engine protection control method based on cylinder combustion conditions according to claim 11, characterized in that: The method for secondary self-learning update includes: The state of the current working condition is determined according to the difference between the updated remaining time value and the initial remaining time value and the number of knocks of different knock intensities; The states include upward high-burst learning state 1, upward medium-burst learning state 1, upward low-burst learning state 1, upward-burst learning state 1, downward high-burst learning state 1, downward medium-burst learning state 1, downward low-burst learning state 1, downward-burst learning state 1 and other states; If the current working condition is in the upward learning state, the self-learning coefficient of the remaining time value increases; if the current working condition is in the downward learning state, the self-learning coefficient of the remaining time value decreases; if the current working condition is in other states, the self-learning coefficient of the remaining time value is not updated.

13. The engine protection control method based on cylinder combustion conditions according to claim 12, characterized in that: The method for determining the state of the current working condition specifically includes: Under the condition that the current operating condition satisfies the condition that the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C1, if the number of knocks with high knock intensity exceeds the preset value A3 by more than the preset number CNT3, the preset value X3 is added to the self-learning coefficient of the remaining time value, and it is judged that the current operating condition is in the upward high-knock learning state one; if the number of knocks with medium knock intensity exceeds the preset value A4 by more than the preset number CNT4, the preset value X4 is added to the self-learning coefficient of the remaining time value, and it is judged that the current operating condition is in the upward medium-knock learning state one; if the number of knocks with low knock intensity exceeds the preset value A5 by more than the preset number CNT5, the preset value X5 is added to the self-learning coefficient of the remaining time value, and it is judged that the current operating condition is in the upward low-knock learning state one; if the current operating condition satisfies the condition that the difference between the updated remaining time value and the initial remaining time value is greater than the preset value C1 by more than the preset number CNT6, the preset value X6 is added to the self-learning coefficient of the remaining time value, and it is judged that the current operating condition is in the upward knock learning state one; Under the condition that the current operating condition satisfies the difference between the updated remaining time value and the initial remaining time value is not greater than the preset value C2, if the number of knocks with high knock intensity exceeds the preset value A6 by more than the preset number CNT7, the self-learning coefficient of the remaining time value is reduced by the preset value X7, and it is judged that the current operating condition is in the downward high-knock learning state one; if the number of knocks with medium knock intensity exceeds the preset value A7 by more than the preset number CNT8, the self-learning coefficient of the remaining time value is reduced by the preset value X8, and it is judged that the current operating condition is in the downward medium-knock learning state one; if the number of knocks with low knock intensity exceeds the preset value A8 by more than the preset number CNT9, the self-learning coefficient of the remaining time value is reduced by the preset value X9, and it is judged that the current operating condition is in the downward low-knock learning state one; if the current operating condition satisfies the difference between the updated remaining time value and the initial remaining time value is not greater than the preset value C2 by more than the preset number CNT 10 , then the remaining time value self-learning coefficient is reduced by the preset value X 10 , it is determined that the current working condition is in the downward shock learning state 1; In other cases, the self-learning coefficient of the remaining duration value is not updated.

14. The engine protection control method based on cylinder combustion conditions according to claim 13, characterized in that: In each self-learning update process, if the remaining time value self-learning coefficient is updated, CNT3, CNT4, CNT5, CNT6, CNT7, CNT8, CNT9, CNT 10 Cleared.

15. The engine protection control method based on cylinder combustion conditions according to claim 14, characterized in that: The priorities of the upward high violent shock learning state 1, upward medium violent shock learning state 1, upward low violent shock learning state 1, upward violent shock learning state 1, downward high violent shock learning state 1, downward medium violent shock learning state 1, downward low violent shock learning state 1, downward violent shock learning state 1 and other states decrease in sequence. If it is judged that the current working condition is in a high priority state, the low priority state will no longer be judged.

16. An engine protection control system based on cylinder combustion conditions, characterized in that: The system executes the engine protection control method based on cylinder combustion conditions as described in any one of claims 1-15.

Images