A method of control of supercharging pressure

Abstract

This invention discloses a boost pressure control method, which includes four stages: a self-learning inactive stage, a self-learning stabilization stage, a self-learning activation stage, and a self-learning update stage. In the self-learning update stage, based on the average actual boost pressure before and after filtering, the average throttle inlet pressure before and after filtering, the average of the minimum actual boost pressure before filtering, or the average of the maximum actual boost pressure before filtering, a self-learning update coefficient for the minimum or maximum boost pressure is calculated and updated to the corresponding operating condition. The corresponding operating condition refers to engine speed, load, intake air temperature, coolant temperature, target intake pressure, atmospheric pressure, and target boost pressure fluctuations all not exceeding a preset range and their average values ​​being the same. Finally, the minimum or maximum boost pressure of the current operating condition is updated based on the self-learning update coefficient. This invention can perform self-learning of boost pressure, including minimum and maximum boost pressure.

Description

Technical Field

[0001] This invention relates to the field of engine control technology, and more specifically to a boost pressure control method. Background Technology

[0002] To respond to engine intake boost and torque increase requests, a boost system control is introduced, utilizing exhaust energy to achieve boost. The boost control employs a closed-loop control method. After the boost closed-loop control is activated, if the minimum boost pressure is too low, insufficient exhaust energy will cause fluctuations in boost pressure control; conversely, if the minimum boost pressure is too high, the operating range of the boost closed-loop control will narrow, failing to optimally utilize the boost advantage. Therefore, it is necessary to introduce minimum boost pressure control as a condition for activating the boost control closed-loop.

[0003] Patent CN201910988050.8, "Adaptive Closed-Loop System and Control Method for Boosting of Exhaust Gas Turbine Engines," states that "the opening condition of the exhaust bypass valve requires the sum of the exhaust back pressure and the boosted gas pressure to be greater than the spring preload. Therefore, the boosted gas pressure needs to exceed a certain calibrated value (called the minimum boost pressure) to overcome the spring preload. This calibrated value is determined by atmospheric pressure, engine speed, and air filter temperature, and is obtained through bench calibration; it is the minimum boost pressure." However, due to differences in the components of different engine boosting systems, or the aging of boosting system components, or the aging of other engine components leading to a shift in boost demand, it is necessary to iteratively update the minimum boost pressure to meet the differences in different engine bodies and the optimal operating range of boost control under different engine life cycles, thereby controlling the stability of the intake pressure.

[0004] The maximum boost pressure is used to estimate the maximum gas volume of the exhaust gas turbocharger system and protect the turbocharger. After the boost closed-loop control is activated, if the maximum boost pressure is too high, it will cause turbocharger surge; if the maximum boost pressure is too low, it will narrow the operating range of the boost closed-loop control, thus failing to make the best use of the boost advantage. Therefore, it is necessary to introduce maximum boost pressure control to reduce the opening of the boost actuator and reduce the boost pressure when the maximum boost pressure range is exceeded.

[0005] Patent CN202010599296.9, "A method and device for controlling the gas volume of an exhaust gas turbocharger system," estimates the maximum boost pressure p after the compressor. CompOutMax However, due to differences in turbocharger system components between different engines, or the aging of turbocharger system components, or the aging of other engine components, the turbocharger demand may shift. At the same time, when the operating environment changes, the maximum boost pressure needs to be iteratively updated to meet the differences between different engine bodies and the optimal operating range of boost control under different engine life cycles, thereby improving the vehicle's power requirements and protecting the turbocharger body. Summary of the Invention

[0006] The purpose of this invention is to provide a boost pressure control method that iteratively updates the minimum or maximum boost pressure to control the stability of the intake pressure and protect the turbocharger and engine.

[0007] The technical solution adopted in this invention is as follows:

[0008] A boost pressure control method includes four stages: a self-learning inactive stage, a self-learning stabilization stage, a self-learning activation stage, and a self-learning update stage.

[0009] In the self-learning inactive phase, determine whether the activation conditions for boost pressure self-learning are met; if yes, proceed to the self-learning stabilization phase; otherwise, remain in the self-learning inactive phase.

[0010] In the self-learning stabilization phase, it is determined whether the stabilization conditions for boost pressure self-learning are met. If both the activation and stabilization conditions are met, the self-learning activation phase begins. If the activation conditions are met but the stabilization conditions are not met, the self-learning stabilization phase is maintained. If the activation conditions are not met, the self-learning inactive phase is returned to.

[0011] During the self-learning activation phase, the engine speed, load, intake air temperature, coolant temperature, target intake pressure, atmospheric pressure, target boost pressure, actual boost pressure before filtering, actual boost pressure after filtering, throttle inlet pressure before filtering, throttle inlet pressure after filtering, and minimum or maximum actual boost pressure before filtering are accumulated over a certain period of time. Once these conditions are met, the system enters the self-learning update phase. If the activation conditions are not met during the self-learning activation phase, the system returns to the inactive self-learning phase.

[0012] During the self-learning update phase, the average value of each parameter accumulated over a certain period of time is calculated. Based on the average actual boost pressure before filtering, the average actual boost pressure after filtering, the average throttle inlet pressure before filtering, the average throttle inlet pressure after filtering, and the average minimum or maximum actual boost pressure before filtering, the self-learning update coefficient of the minimum boost pressure or the self-learning update coefficient of the maximum boost pressure is calculated, and this coefficient is updated to the corresponding operating condition. The corresponding operating condition refers to the engine speed, load, intake air temperature, water temperature, target intake air pressure, atmospheric pressure, and target boost pressure fluctuations all not exceeding the preset range and their average values ​​are the same.

[0013] Update the minimum or maximum boost pressure for the current operating condition based on the self-learning update coefficient of the minimum or maximum boost pressure.

[0014] Furthermore, the activation conditions for the minimum boost pressure self-learning include:

[0015] The engine is running, and the engine has been running for longer than the preset time.

[0016] The boost control is currently in a closed-loop control active state;

[0017] The difference between the current target boost pressure and the real-time minimum boost pressure does not exceed the preset pressure difference range; if the minimum boost pressure has not been updated, the real-time minimum boost pressure is the original minimum boost pressure; if the minimum boost pressure has been updated, the real-time minimum boost pressure is the updated minimum boost pressure.

[0018] The target intake pressure fluctuation is within the preset pressure fluctuation range;

[0019] The air-fuel ratio control is in a closed-loop control state;

[0020] The engine speed exceeds the preset speed value, but the engine speed fluctuation does not exceed the preset speed fluctuation range;

[0021] Load fluctuations shall not exceed the preset load fluctuation range;

[0022] The engine coolant temperature is within the preset range, and the engine coolant temperature fluctuation does not exceed the preset range.

[0023] The intake air temperature is within the preset intake air temperature range, and the intake air temperature fluctuation does not exceed the preset intake air temperature fluctuation range.

[0024] The atmospheric pressure fluctuation range shall not exceed the preset atmospheric pressure fluctuation range;

[0025] No faults were detected in the booster system;

[0026] When all of the above conditions are met, the activation condition for minimum boost pressure self-learning is satisfied.

[0027] Furthermore, the preset time is 5 minutes, the preset differential pressure range is ±1 kPa, the preset pressure fluctuation range is ±1%, the preset speed is 600 rpm, the preset speed fluctuation range is ±1%, the preset load fluctuation range is ±1%, the preset water temperature range is 0℃ to 100℃, the preset water temperature fluctuation range is ±1%, the preset intake air temperature range is 30℃ to 80℃, the preset intake air temperature fluctuation range is ±1%, and the preset atmospheric pressure fluctuation range is ±0.5 kPa.

[0028] Furthermore, the activation conditions for the self-learning of maximum boost pressure include:

[0029] The engine is running, and the engine has been running for longer than the preset time.

[0030] The boost control is currently in a closed-loop control active state;

[0031] The difference between the current target boost pressure and the real-time maximum boost pressure does not exceed the preset pressure difference range; if the maximum boost pressure has not been updated, the real-time maximum boost pressure is the original maximum boost pressure; if the maximum boost pressure has been updated, the real-time maximum boost pressure is the updated maximum boost pressure.

[0032] The average target boost pressure is greater than the preset target boost pressure value;

[0033] The target intake pressure fluctuation is within the preset pressure fluctuation range;

[0034] The throttle is fully open;

[0035] The air-fuel ratio control is in a closed-loop control state;

[0036] The ignition angle efficiency is greater than the preset value for ignition angle efficiency;

[0037] The engine speed exceeds the preset speed value, but the engine speed fluctuation does not exceed the preset speed fluctuation range;

[0038] Load fluctuations shall not exceed the preset load fluctuation range;

[0039] The engine coolant temperature is within the preset range, and the engine coolant temperature fluctuation does not exceed the preset range.

[0040] The intake air temperature is within the preset intake air temperature range, and the intake air temperature fluctuation does not exceed the preset intake air temperature fluctuation range.

[0041] The atmospheric pressure fluctuation range shall not exceed the preset atmospheric pressure fluctuation range;

[0042] No faults were detected in the booster system;

[0043] When all of the above conditions are met, the activation condition for maximum boost pressure self-learning is satisfied.

[0044] Furthermore, the preset time is 5 minutes, the preset differential pressure range is ±1 kPa, the preset pressure fluctuation range is ±1%, the preset ignition angle efficiency is 0.95, the preset speed is 600 rpm, the preset speed fluctuation range is ±1%, the preset load fluctuation range is ±1%, the preset water temperature range is 0℃ to 100℃, the preset water temperature fluctuation range is ±1%, the preset intake air temperature range is 30℃ to 80℃, the preset intake air temperature fluctuation range is ±1%, and the preset atmospheric pressure fluctuation range is ±0.5 kPa.

[0045] Furthermore, the target boost pressure preset value depends on atmospheric pressure and engine speed, and this preset value is not less than atmospheric pressure;

[0046] This preset value is equal to atmospheric pressure + f1(n,p) Amb ), where n is the engine speed, p Amb Atmospheric pressure.

[0047] Furthermore, the stabilization conditions for boost pressure self-learning include:

[0048] The self-learning stabilization phase has been entered beyond the preset first time;

[0049] The self-learning of the minimum boost pressure has not been updated beyond the preset second time;

[0050] When all of the above conditions are met, the stability condition for minimum boost pressure self-learning is satisfied.

[0051] Furthermore, the actual boost pressure and throttle inlet pressure filtering methods include:

[0052] x Filter (N)=K x ×[x Raw (N)-x Filter [(N-1)]+x Filter (N-1)

[0053] In the formula, x Raw The signal before filtering is the actual boost pressure or throttle inlet pressure, x Raw (N) represents the signal before filtering in the Nth sampling period, x Filter The filtered signal, x Filter (N) represents the filtered signal after the Nth sampling period, x Filter (N-1) is the filtered signal after the (N-1)th sampling period, where N = 1, 2, 3, ..., x Filter (0) equals the signal before filtering at the 0th sampling period, K x is a coefficient.

[0054] Furthermore, the self-learning update coefficient calculation method for the minimum boost pressure includes:

[0055]

[0056]

[0057] Where, p preThrAvg p represents the average throttle inlet pressure after filtering. preThrRawAvg p is the average throttle inlet pressure before filtering. BoostActAvg p represents the average actual boost pressure after filtering. BoostActRawAvgis the average value of the actual boost pressure before filtering;

[0058] If:

[0059] 1) C1 > 0.05 and C2 > 0.05, then:

[0060]

[0061] 2) C1 > 0.05 and C2 < 0.01, then:

[0062]

[0063] where the weighting coefficient k1 takes 0.9;

[0064] 3) C1 > 0.05 and 0.01 ≤ C2 ≤ 0.05, then:

[0065]

[0066] where the weighting coefficient k2 takes 0.8;

[0067] 4) 0.01 < C1 ≤ 0.05 and C2 > 0.05, then:

[0068]

[0069] where the weighting coefficient k3 takes 0.1;

[0070] 5) 0.01 < C1 ≤ 0.05 and C2 < 0.01, then:

[0071]

[0072] where the weighting coefficient k4 takes 0.6;

[0073] 6) 0.01 < C1 ≤ 0.05 and 0.01 ≤ C2 ≤ 0.05, then:

[0074]

[0075] where the weighting coefficient k5 takes 0.3;

[0076] 7) C1 ≤ 0.01 and C2 > 0.05, then:

[0077]

[0078] where the weighting coefficient k6 takes 0.02;

[0079] 8) C1 ≤ 0.01 and C2 < 0.01, then:

[0080]

[0081] The weighting coefficient k7 is set to 0.12;

[0082] 9) If C1≤0.01 and 0.01≤C2≤0.05, then:

[0083]

[0084] The weighting coefficient k8 is set to 0.08;

[0085] In the formula, r BoostMinAdapt r is a self-learning update coefficient that is updated at most once during a single driving cycle. BoostMinAdapt (z) represents the self-learning update coefficient from the previous self-learning update, p BoostActMinRawAvg r represents the average of the minimum actual boost pressure before filtering. Adpat It is an adjustment coefficient for the minimum boost pressure and can be continuously updated through self-learning.

[0086] Furthermore, the self-learning update method for the minimum boost pressure adjustment coefficient includes:

[0087] After the self-learning update coefficients are updated, the calculation is repeated:

[0088] and

[0089] If it still exists:

[0090] 1) If C1 > 0.05 and C2 > 0.05, then r Adpat =r Adpat (z)+0.05

[0091] 2) If C1 ≤ 0.01 and C2 < 0.01, then r Adpat =r Adpat (z)-0.02

[0092] 3) In other cases, r Adpat =r Adpat (z)

[0093] In the formula, r Adpat (z) is the adjustment coefficient of the last self-learning update.

[0094] Furthermore, the self-learning update coefficient calculation method for the maximum boost pressure includes:

[0095]

[0096]

[0097] Where, p preThrAvg p represents the average throttle inlet pressure after filtering. preThrRawAvgis the average value of the throttle inlet pressure before filtering, p BoostActAvg is the average value of the actual boost pressure after filtering, p BoostActRawAvg is the average value of the actual boost pressure before filtering;

[0098] If:

[0099] 1) C1 > 0.05 and C2 > 0.05, then:

[0100]

[0101] 2) C1 > 0.05 and C2 < 0.01, then:

[0102]

[0103] where the weighting coefficient k1 takes 0.9;

[0104] 3) C1 > 0.05 and 0.01 ≤ C2 ≤ 0.05, then:

[0105] where the weighting coefficient k2 takes 0.8;

[0106] 4) 0.01 < C1 ≤ 0.05 and C2 > 0.05, then:

[0107] where the weighting coefficient k3 takes 0.1;

[0108] 5) 0.01 < C1 ≤ 0.05 and C2 < 0.01, then:

[0109] where the weighting coefficient k4 takes 0.6;

[0110] 6) 0.01 < C1 ≤ 0.05 and 0.01 ≤ C2 ≤ 0.05, then:

[0111] where the weighting coefficient k5 takes 0.3;

[0112] 7) C1 ≤ 0.01 and C2 > 0.05, then:

[0113]

[0114] where the weighting coefficient k6 takes 0.02;

[0115] 8) C1 ≤ 0.01 and C2 < 0.01, then:

[0116]

[0117] where the weighting coefficient k7 takes 0.12;

[0118] 9) If C1≤0.01 and 0.01≤C2≤0.05, then:

[0119]

[0120] The weighting coefficient k8 is set to 0.08;

[0121] In the formula, r BoostMaxAdapt r is a self-learning update coefficient that is updated at most once during a single driving cycle. BoostMaxAdapt (z) represents the self-learning update coefficient from the previous self-learning update, p BoostActMaxRawAvg r represents the average of the actual maximum boost pressure before filtering. Adpat It is an adjustment coefficient for the maximum boost pressure and can be continuously updated through self-learning.

[0122] Furthermore, the self-learning update method for the maximum boost pressure adjustment coefficient includes:

[0123] After the self-learning update coefficients are updated, the calculation is repeated:

[0124] and

[0125] If it still exists:

[0126] 1) If C1 > 0.05 and C2 > 0.05, then r Adpat =r Adpat (z)-0.02

[0127] 2) If C1 ≤ 0.01 and C2 < 0.01, then r Adpat =r Adpat (z)+0.01

[0128] 3) In other cases, r Adpat =r Adpat (z)

[0129] In the formula, r Adpat (z) is the adjustment coefficient of the last self-learning update.

[0130] Furthermore, the minimum boost pressure for the current operating condition is updated based on the self-learning update coefficient of the minimum boost pressure as follows:

[0131] p BoostMinNew =p BoostMinDsrd ×(1+r BoostMinAdapt )

[0132] In the formula, p BoostMinNew For the updated minimum boost pressure, p BoostMinDsrd For the original minimum boost pressure, r BoostMinAdapt The self-learning update coefficient is the minimum boost pressure.

[0133] The maximum boost pressure under the current operating condition is updated based on the self-learning update coefficient of the maximum boost pressure as follows:

[0134] p BoostMaxNew =p BoostMaxDsrd ×(1+r BoostMaxAdapt )

[0135] In the formula, p BoostMaxNew For the updated maximum boost pressure, p BoostMaxDsrd r is the original maximum boost pressure. BoostMaxAdapt This is the self-learning update coefficient for the maximum boost pressure.

[0136] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0137] Regardless of differences in engine manufacturing, changes in operating conditions, or differences in engine lifecycle, this invention can self-learn the minimum boost pressure, improving the stability of the boost pressure control system; and self-learn the maximum boost pressure to meet the optimal operating range of boost control under different engine body differences and different engine lifecycles, thereby improving the vehicle's power requirements and protecting the turbocharger body. Attached Figure Description

[0138] Figure 1 This is a flowchart of the pressure control method for boosting pressure. Detailed Implementation

[0139] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0140] This invention first proposes a design method for a minimum boost pressure. When the target boost pressure is lower than its minimum boost pressure, the boost pressure control system is shut down, i.e., boost pressure is not allowed to be activated. This serves as the minimum limit when setting the target boost pressure, thus avoiding problems such as poor intake control stability and suboptimal power and economy caused by improper target setting during the boost pressure target setting process.

[0141] This invention performs iterative update learning of the minimum boost pressure when the engine operating conditions are stable, and learns the self-learning update coefficient r of the minimum boost pressure. BoostMinAdapt This ensures the accuracy of self-learning. Once the most basic prerequisites are met, the self-learning process can begin. For example... Figure 1As shown, the activation condition for the minimum boost pressure self-learning is as follows:

[0142] 1) The engine is running, and the engine running time exceeds the preset time, which is 5 minutes in this example;

[0143] 2) The boost control is currently in a closed-loop control active state;

[0144] 3) Current target boost pressure p BoostDsrd (Current target boost pressure p) BoostDsrd The method for obtaining this information can be found in patent CN202010109549.X, "Method for determining the target boost pressure of an exhaust gas turbocharged engine, storage medium," along with the real-time minimum boost pressure p. BoostMin (If not learned, then the original minimum boost pressure p) CompOutMin If it has been learned, then the difference between the learned and updated minimum boost pressure and the pressure is not greater than the preset range. In this example, it is ±1 kPa.

[0145] 4) The target intake pressure (i.e. the target value of the gas pressure entering the cylinder) fluctuates little; in this example, it is taken as ±1%.

[0146] 5) The air-fuel ratio control is in a closed-loop control state;

[0147] 6) The engine speed exceeds the preset value, which is 600 rpm in this example; and the engine speed fluctuation does not exceed the preset range, which is ±1% in this example;

[0148] 7) The load (intake density of fresh air entering the cylinder) fluctuation shall not exceed the preset range, which is ±1% in this example;

[0149] 8) The engine coolant temperature is within a certain range (0℃ to 100℃ in this example), and the engine coolant temperature fluctuation during self-learning is small (±1% in this example).

[0150] 9) The intake air temperature is within a certain range (30℃ to 80℃ in this example), and the intake air temperature fluctuation when entering self-learning is small, which is no more than ±1% in this example;

[0151] 10) Atmospheric pressure fluctuations shall not exceed ±0.5 kPa;

[0152] 11) No faults were detected in the booster system.

[0153] If any activation condition is not met at any stage of the self-learning process, the self-learning process terminates and enters the inactive phase. When the activation conditions are met, the self-learning process based on minimum boost pressure can be attempted, starting with the minimum boost pressure stabilization phase.

[0154] Upon entering the self-learning stabilization phase, the purpose of the stabilization phase is to ensure that the self-learning activation conditions are stable and reliable. During the self-learning stabilization phase, the minimum pressure boosting self-learning activation phase will begin when the following stabilization conditions are met:

[0155] 1) The self-learning stabilization phase exceeds the preset time t0, which is 5 seconds in this example;

[0156] 2) The self-learning of minimum boost pressure has not been updated for more than the preset time t1 (in this example, 500 hours). The self-learning count is updated once the self-learning of minimum boost pressure is completed. If the learning interval is too long, the difference in the learning value each time may be caused by the aging of engine parts, rather than the learning of accurate information.

[0157] If the above stability conditions are not met, but the activation conditions are met, the system remains in the self-learning stable phase. If the above conditions are met, but the activation conditions are not met, the system returns to the self-learning inactive phase. When both the above conditions and the activation conditions are met, the system enters the next phase, namely the minimum boost pressure self-learning activation phase.

[0158] During the self-learning activation phase for minimum boost pressure, the following parameters are accumulated over a certain time interval t2 (6 seconds in this example): total engine speed, total load, total intake air temperature, total coolant temperature, total target intake pressure, total atmospheric pressure, total target boost pressure, total actual boost pressure before filtering, total actual boost pressure after filtering, total throttle inlet pressure before filtering, total throttle inlet pressure after filtering, and the sum of the minimum actual boost pressure before filtering. Once time t2 is satisfied, the system proceeds to the next phase, the minimum boost pressure self-learning update phase.

[0159] The filtering algorithms for actual boost pressure and throttle inlet pressure are as follows:

[0160] x Filter (N)=K x ×[x Raw (N)-x Filter [(N-1)]+x Filter (N-1)

[0161] Where, x R aw represents the signal before filtering (actual boost pressure or throttle inlet pressure), x Raw (N) represents the signal before filtering in the Nth sampling period, x F ilter is the filtered signal, x Filter (N) represents the filtered signal after the Nth sampling period, x Filter (N-1) is the filtered signal after the (N-1)th sampling period, where N = 1, 2, 3, ..., x Filter(0) equals the signal before filtering at the 0th sampling period (the 0th sampling period refers to the moment when the self-learning activation phase has just begun); the sampling period interval Δt is 10ms in this example. K x The coefficients for the actual boost pressure and throttle inlet pressure in this example are both set to 0.1.

[0162] It should be noted that the minimum actual boost pressure before filtering is the minimum value within a nearby range. For example, there are multiple sampling periods within a certain time t2, and each sampling period has a minimum value. In addition, the sampling periods of each parameter may be inconsistent.

[0163] Minimum boost pressure self-learning update phase:

[0164] For each operating condition (defined as: engine speed, load, intake air temperature, coolant temperature, target intake air pressure, atmospheric pressure, and target boost pressure fluctuations all not exceeding ±1%, and their average values ​​being the same), the minimum boost pressure learning coefficient r for each operating condition is stored in the EEPROM (Electrically Erasable Programmable Read-Only Memory). BoostMinAdapt It has an initial default value of 0, which is updated in the EEPROM after the minimum boost pressure self-learning is completed. The minimum boost pressure self-learning storage phase mainly performs the following tasks:

[0165] 1) Calculate the average engine speed, average load, average intake air temperature, average coolant temperature, average target intake air pressure, average atmospheric pressure, average target boost pressure, average actual boost pressure before filtering, average actual boost pressure after filtering, average throttle inlet pressure before filtering, average throttle inlet pressure after filtering, and average minimum actual boost pressure before filtering for a cumulative time t2 during the self-learning phase of entering minimum boost pressure.

[0166] These average values ​​are mainly divided into two parts: one part is used to determine the operating conditions, and the other part is used to calculate the self-learning update coefficient for the minimum boost pressure. The operating conditions can be pre-entered into the memory, or the current operating conditions can be compared with existing operating conditions in the memory, and if it is a new operating condition, it can be stored in the memory.

[0167] 2) The learning coefficient r for the minimum boost pressure under the operating conditions of average engine speed, average load, average intake air temperature, average coolant temperature, average target intake air pressure, average atmospheric pressure, and average target boost pressure is calculated. BoostMinAdapt The learned values ​​are updated in the EEPROM under the corresponding operating conditions (engine speed, load, intake air temperature, water temperature, target intake pressure, atmospheric pressure, and target boost pressure).

[0168] Self-learning coefficient r of the minimum boost pressure BoostMinAdapt The learning method is as follows:

[0169]

[0170]

[0171] Where p preThrAvg is the average value of the throttle inlet pressure after filtering, p preThrRawAvg is the average value of the throttle inlet pressure before filtering, p BoostActAvg is the average value of the actual boost pressure after filtering, p BoostActRawAvg is the average value of the actual boost pressure before filtering. p BoostActMinRawAvg is the average value of the minimum of the actual boost pressure before filtering.

[0172] If:

[0173] 1) C1 > 0.05 and C2 > 0.05, then:

[0174]

[0175] 2) C1 > 0.05 and C2 < 0.01, then:

[0176] Where the weighting coefficient k1 takes 0.9.

[0177] 3) C1 > 0.05 and 0.01 ≤ C2 ≤ 0.05, then:

[0178] Where the weighting coefficient k2 takes 0.8.

[0179] 4) 0.01 < C1 ≤ 0.05 and C2 > 0.05, then:

[0180] Where the weighting coefficient k3 takes 0.1.

[0181] 5) 0.01 < C1 ≤ 0.05 and C2 < 0.01, then:

[0182] Where the weighting coefficient k4 takes 0.6.

[0183] 6) 0.01 < C1 ≤ 0.05 and 0.01 ≤ C2 ≤ 0.05, then:

[0184] Where the weighting coefficient k5 takes 0.3.

[0185] 7) C1 ≤ 0.01 and C2 > 0.05, then:

[0186] The weighting coefficient k6 is set to 0.02.

[0187] 8) If C1≤0.01 and C2<0.01, then:

[0188] The weighting coefficient k7 is set to 0.12.

[0189] 9) If C1≤0.01 and 0.01≤C2≤0.05, then:

[0190] The weighting coefficient k8 is set to 0.08.

[0191] In the formula, r BoostMinAdapt (z) represents the self-learning update coefficient from the previous self-learning update. The self-learning update coefficient is updated at most once during one driving cycle. Adpat The adjustment coefficient is set to 0 by default and can be continuously updated through self-learning, and is saved after the vehicle is powered off.

[0192] In r BoostMinAdapt Once confirmed, the minimum boost pressure will be updated immediately to:

[0193] p BoostMinNew =p BoostMinDsrd ×(1+r BoostMinAdapt )

[0194] In the formula, p BoostMinNew For the updated minimum boost pressure, p BoostMinDsrd For the original minimum boost pressure, r BoostMinAdapt This is the self-learning update coefficient for the minimum boost pressure.

[0195] 1+r coming soon BoostMinAdapt Multiply by the original minimum boost pressure p CompOutMin Once the real-time updated minimum boost pressure is obtained, the entire updated minimum boost pressure is completed and used as a closed-loop activation condition in the boost closed-loop control.

[0196] Among them, the adjustment coefficient r of the minimum boost pressure Adpat The method for determining it is as follows:

[0197] Read the updated version again and

[0198] If it still exists:

[0199] 1) If C1 > 0.05 and C2 > 0.05, then r Adpat =r Adpat (z)+0.05;

[0200] 2) If C1 ≤ 0.01 and C2 < 0.01, then r Adpat =r Adpat (z)-0.02;

[0201] 3) In other cases, r Adpat =r Adpat (z).

[0202] In the formula, r Adpat (z) is the adjustment coefficient of the last self-learning update. The self-learning update coefficient is updated at most once during one driving cycle.

[0203] The above completes the description of the minimum boost pressure control method.

[0204] This invention also proposes a design method for maximum boost pressure, similar to the design method for minimum boost pressure described above. The maximum boost pressure is used to estimate the maximum gas volume of the exhaust gas turbocharger system to protect the turbocharger.

[0205] This invention performs iterative update learning of the maximum boost pressure when the engine is operating under stable conditions, and learns the self-learning update coefficient r of the maximum boost pressure. BoostMaxAdapt This ensures the accuracy of self-learning. Once the most basic prerequisites are met, the self-learning process can begin. For example... Figure 1 As shown, the activation condition for the self-learning of maximum boost pressure is as follows:

[0206] 1) The engine is running, and the engine running time exceeds the preset time, which is 5 minutes in this example;

[0207] 2) The boost control is currently in a closed-loop control active state;

[0208] 3) Current target boost pressure p BoostDsrd (Current target boost pressure p) BoostDsrd The method for obtaining this information can be found in patent CN202010109549.X, "Method for determining the target boost pressure of an exhaust gas turbocharged engine, storage medium," and the real-time maximum boost pressure p. BoostMax (If not learned, then the original maximum boost pressure p) CompOutMax If it has been learned, the difference between the maximum boost pressure (the one that has been learned and updated) and the preset range shall not exceed ±1 kPa in this example.

[0209] 4) Target boost pressure average value p BoostDsrdAvg The value is greater than a preset value, which depends on atmospheric pressure and engine speed, and is not less than atmospheric pressure. The preset value is equal to atmospheric pressure + f1(n,p). Amb ), where n is the engine speed, p A mb is atmospheric pressure. f1(n,p)Amb The values ​​for this example are shown in Table 1.

[0210] Table 1f1(n,p) Amb Value table

[0211]

[0212] 5) The target intake pressure (i.e. the target value of the gas pressure entering the cylinder) fluctuates little; in this example, it is taken as ±1%.

[0213] 6) The throttle is fully open;

[0214] 7) The air-fuel ratio control is in a closed-loop control state;

[0215] 8) The ignition angle efficiency is greater than the preset value; in this example, it is set to 0.95.

[0216] 9) The engine speed exceeds the preset value, which is 600 rpm in this example; and the engine speed fluctuation does not exceed the preset range, which is ±1% in this example;

[0217] 10) The load (intake density of fresh air entering the cylinder) fluctuation shall not exceed the preset range, which is ±1% in this example;

[0218] 11) The engine coolant temperature is within a certain range (0℃ to 100℃ in this example), and the engine coolant temperature fluctuation during self-learning is small (±1% in this example).

[0219] 12) The intake air temperature is within a certain range (30℃ to 80℃ in this example), and the intake air temperature fluctuation when entering self-learning is small, which is no more than ±1% in this example;

[0220] 13) Atmospheric pressure fluctuations shall not exceed ±0.5 kPa;

[0221] 14) No faults were detected in the booster system.

[0222] If any activation condition is not met at any stage of the self-learning process, the self-learning process terminates and enters the inactive phase. When the activation conditions are met, the self-learning process based on maximum boost pressure can be attempted, starting with the stabilization phase based on maximum boost pressure.

[0223] Upon entering the self-learning stabilization phase, the purpose of this phase is to ensure that the self-learning activation conditions are stable and reliable. During the self-learning stabilization phase, the maximum pressure self-learning activation phase will begin when the following stabilization conditions are met:

[0224] 1) The self-learning stabilization phase exceeds the preset time t0, which is 5 seconds in this example;

[0225] 2) The self-learning of the maximum boost pressure has not been updated for more than the preset time t1 (in this example, it is 500 hours; the self-learning count is updated once the self-learning of the maximum boost pressure is completed. If the learning interval is too long, the difference in the learning value each time may be caused by the aging of engine parts, rather than the learning of accurate information).

[0226] If the above stability conditions are not met, but the activation conditions are met, the system remains in the self-learning stable phase. If the above conditions are met, but the activation conditions are not met, the system returns to the self-learning inactive phase. When both the above conditions and the activation conditions are met, the system enters the next phase, namely the maximum boost pressure self-learning activation phase.

[0227] During the self-learning activation phase for maximum boost pressure, the following parameters are accumulated over a certain time interval t2 (6 seconds in this example): total engine speed, total load, total intake air temperature, total coolant temperature, total target intake pressure, total atmospheric pressure, total target boost pressure, total actual boost pressure before filtering, total actual boost pressure after filtering, total throttle inlet pressure before filtering, total throttle inlet pressure after filtering, and the sum of the maximum actual boost pressure before filtering. Once time t2 is satisfied, the system proceeds to the next phase, the maximum boost pressure self-learning update phase.

[0228] The filtering algorithms for actual boost pressure and throttle inlet pressure are as follows:

[0229] x Filter (N)=K x ×[x Raw (N)-x Filter [(N-1)]+x Filter (N-1)

[0230] Where, x R aw represents the signal before filtering (actual boost pressure or throttle inlet pressure), x Raw (N) represents the signal before filtering in the Nth sampling period, x F ilter is the filtered signal, x Filter (N) represents the filtered signal after the Nth sampling period, x Filter (N-1) is the filtered signal after the (N-1)th sampling period, where N = 1, 2, 3, ..., x Filter (0) equals the signal before filtering at the 0th sampling period (the 0th sampling period refers to the moment when the self-learning activation phase has just begun); the sampling period interval Δt is 10ms in this example. K x The coefficients for the actual boost pressure and throttle inlet pressure in this example are both set to 0.1.

[0231] It should be noted that the maximum actual boost pressure before filtering is the largest value within a nearby range. For example, there are multiple sampling periods within a certain time t2, and each sampling period has a maximum value. Furthermore, the sampling periods for different parameters may be inconsistent.

[0232] Maximum boost pressure self-learning update phase:

[0233] For each operating condition (defined as: engine speed, load, intake air temperature, coolant temperature, target intake air pressure, atmospheric pressure, and target boost pressure fluctuations all not exceeding ±1%, and their average values ​​being the same), the maximum boost pressure learning coefficient r for each operating condition is stored in the EEPROM (Electrically Erasable Programmable Read-Only Memory). BoostMaxAdapt It has an initial default value of 0, which is updated in the EEPROM after the maximum boost pressure self-learning is completed. The maximum boost pressure self-learning storage phase mainly performs the following tasks:

[0234] 1) Calculate the average engine speed, average load, average intake air temperature, average coolant temperature, average target intake air pressure, average atmospheric pressure, average target boost pressure, average actual boost pressure before filtering, average actual boost pressure after filtering, average throttle inlet pressure before filtering, average throttle inlet pressure after filtering, and average maximum actual boost pressure before filtering for a certain period of time t2 during the self-learning phase of entering maximum boost pressure.

[0235] These average values ​​are mainly divided into two parts: one part is used to determine the operating conditions, and the other part is used to calculate the self-learning update coefficient for the maximum boost pressure. The operating conditions can be pre-entered into the memory, or the current operating conditions can be compared with existing operating conditions in the memory, and if it is a new operating condition, it can be stored in the memory.

[0236] 2) The learning coefficient r for the maximum boost pressure under the operating conditions of average engine speed, average load, average intake air temperature, average coolant temperature, average target intake pressure, average atmospheric pressure, and average target boost pressure is calculated. BoostMaxAdapt The learned values ​​are updated in the EEPROM under the corresponding operating conditions (engine speed, load, intake air temperature, water temperature, target intake pressure, atmospheric pressure, and target boost pressure).

[0237] Self-learning coefficient r of maximum boost pressure BoostMaxAdapt The learning methods are as follows:

[0238]

[0239]

[0240] where p preThrAvg is the average value of the throttle inlet pressure after filtering, p preThrRawAvg is the average value of the throttle inlet pressure before filtering, p BoostActAvg is the average value of the actual boost pressure after filtering, p BoostActRawAvg is the average value of the actual boost pressure before filtering. p BoostActMaxRawAvg is the average value of the maximum of the actual boost pressure before filtering.

[0241] If:

[0242] 1) C1 > 0.05 and C2 > 0.05, then:

[0243]

[0244] 2) C1 > 0.05 and C2 < 0.01, then:

[0245] where the weighting coefficient k1 is taken as 0.9.

[0246] 3) C1 > 0.05 and 0.01 ≤ C2 ≤ 0.05, then:

[0247] where the weighting coefficient k2 is taken as 0.8.

[0248] 4) 0.01 < C1 ≤ 0.05 and C2 > 0.05, then:

[0249] where the weighting coefficient k3 is taken as 0.1.

[0250] 5) 0.01 < C1 ≤ 0.05 and C2 < 0.01, then:

[0251] where the weighting coefficient k4 is taken as 0.6.

[0252] 6) 0.01 < C1 ≤ 0.05 and 0.01 ≤ C2 ≤ 0.05, then:

[0253] where the weighting coefficient k5 is taken as​​​​​​​​​​​​​​

[0258] 9) If C1≤0.01 and 0.01≤C2≤0.05, then:

[0259] The weighting coefficient k8 is set to 0.08.

[0260] In the formula, r BoostMaxAdapt (z) represents the self-learning update coefficient from the previous self-learning update. The self-learning update coefficient is updated at most once during one driving cycle. Adpat The adjustment coefficient is set to 0 by default and can be continuously updated through self-learning, and is saved after the vehicle is powered off.

[0261] In r BoostMaxAdapt Once confirmed, the maximum boost pressure will be updated immediately to:

[0262] p BoostMaxNew =p BoostMaxDsrd ×(1+r BoostMaxAdapt )

[0263] In the formula, p BoostMaxNew For the updated maximum boost pressure, p BoostMaxDsrd r is the original maximum boost pressure. BoostMaxAdapt This is the self-learning update coefficient for the maximum boost pressure.

[0264] 1+r coming soon BoostMaxAdapt Multiplied by the original maximum boost pressure p CompOutMax Once the real-time updated maximum boost pressure is obtained, the updated maximum boost pressure is completed and used to estimate the maximum gas volume of the exhaust gas turbocharger system.

[0265] Among them, the adjustment coefficient r of the maximum boost pressure Adpat The method for determining it is as follows:

[0266] Read the updated version again and

[0267] If it still exists:

[0268] 1) If C1 > 0.05 and C2 > 0.05, then r Adpat =r Adpat (z)-0.02;

[0269] 2) If C1 ≤ 0.01 and C2 < 0.01, then r Adpat =r Adpat (z)+0.01;

[0270] 3) In other cases, r Adpat =r Adpat (z).

[0271] In the formula, r Adpat (z) is the adjustment coefficient of the last self-learning update. The self-learning update coefficient is updated at most once during one driving cycle.

[0272] The above completes the description of the pressure control method for the exhaust gas turbocharging system.

[0273] In summary, regardless of differences in engine manufacturing or engine lifecycle, the ability to self-learn minimum and maximum boost pressure improves the stability of the boost pressure control system and protects the turbocharger and engine.

[0274] It should be noted that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0275] It should be noted that, depending on the implementation needs, the various steps / components described in this application can be broken down into more steps / components, or two or more steps / components or parts of the operation of steps / components can be combined into new steps / components to achieve the purpose of this invention.

[0276] Those skilled in the art will readily understand that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for controlling boosted pressure, characterized in that, It includes four stages: self-learning inactive stage, self-learning stabilization stage, self-learning activation stage, and self-learning update stage; In the self-learning inactive phase, determine whether the activation conditions for boost pressure self-learning are met; if yes, proceed to the self-learning stabilization phase; otherwise, remain in the self-learning inactive phase. In the self-learning stabilization phase, it is determined whether the stabilization conditions for the boost pressure self-learning are met. If both the activation and stabilization conditions are met, the self-learning activation phase is entered. If the activation conditions are met but the stabilization conditions are not met, the self-learning stabilization phase is maintained. If the activation conditions are not met, return to the self-learning inactive stage; During the self-learning activation phase, the engine speed, load, intake air temperature, coolant temperature, target intake pressure, atmospheric pressure, target boost pressure, actual boost pressure before filtering, actual boost pressure after filtering, throttle inlet pressure before filtering, and throttle inlet pressure after filtering are accumulated over a certain period of time. Simultaneously, the minimum or maximum actual boost pressure before filtering is accumulated. Once the aforementioned conditions are met within the specified time, the system enters the self-learning update phase. If the activation conditions are not met during the self-learning activation phase, the system returns to the inactive self-learning phase. During the self-learning update phase, the average values ​​of each parameter accumulated over a certain period of time are calculated. Based on the average actual boost pressure before filtering, the average actual boost pressure after filtering, the average throttle inlet pressure before filtering, the average throttle inlet pressure after filtering, and the average minimum or maximum actual boost pressure before filtering, the self-learning update coefficient for the minimum or maximum boost pressure is calculated, and this coefficient is updated to the corresponding operating condition. The corresponding operating condition refers to the engine speed, load, intake air temperature, water temperature, target intake air pressure, atmospheric pressure, and target boost pressure fluctuations all not exceeding the preset range and their average values ​​are the same. Update the minimum or maximum boost pressure for the current operating condition based on the self-learning update coefficient of the minimum or maximum boost pressure.

2. The boosting pressure control method according to claim 1, characterized in that, The activation conditions for minimum boost pressure self-learning include: The engine is running, and the engine has been running for longer than the preset time. The boost control is currently in a closed-loop control active state; The difference between the current target boost pressure and the real-time minimum boost pressure does not exceed the preset pressure difference range; if the minimum boost pressure has not been updated, the real-time minimum boost pressure is the original minimum boost pressure; if the minimum boost pressure has been updated, the real-time minimum boost pressure is the updated minimum boost pressure. The target intake pressure fluctuation is within the preset pressure fluctuation range; The air-fuel ratio control is in a closed-loop control state; The engine speed exceeds the preset speed value, but the engine speed fluctuation does not exceed the preset speed fluctuation range; Load fluctuations shall not exceed the preset load fluctuation range; The engine coolant temperature is within the preset range, and the engine coolant temperature fluctuation does not exceed the preset range. The intake air temperature is within the preset intake air temperature range, and the intake air temperature fluctuation does not exceed the preset intake air temperature fluctuation range. The atmospheric pressure fluctuation range shall not exceed the preset atmospheric pressure fluctuation range; No faults were detected in the booster system; When all of the above conditions are met, the activation condition for minimum boost pressure self-learning is satisfied.

3. The booster pressure control method according to claim 1, characterized in that, The activation conditions for self-learning maximum boost pressure include: The engine is running, and the engine has been running for longer than the preset time. The boost control is currently in a closed-loop control active state; The difference between the current target boost pressure and the real-time maximum boost pressure does not exceed the preset pressure difference range; if the maximum boost pressure has not been updated, the real-time maximum boost pressure is the original maximum boost pressure; if the maximum boost pressure has been updated, the real-time maximum boost pressure is the updated maximum boost pressure. The average target boost pressure is greater than the preset target boost pressure value; The target intake pressure fluctuation is within the preset pressure fluctuation range; The throttle is fully open; The air-fuel ratio control is in a closed-loop control state; The ignition angle efficiency is greater than the preset value for ignition angle efficiency; The engine speed exceeds the preset speed value, but the engine speed fluctuation does not exceed the preset speed fluctuation range; Load fluctuations shall not exceed the preset load fluctuation range; The engine coolant temperature is within the preset range, and the engine coolant temperature fluctuation does not exceed the preset range. The intake air temperature is within the preset intake air temperature range, and the intake air temperature fluctuation does not exceed the preset intake air temperature fluctuation range. The atmospheric pressure fluctuation range shall not exceed the preset atmospheric pressure fluctuation range; No faults were detected in the booster system; When all of the above conditions are met, the activation condition for maximum boost pressure self-learning is satisfied; The target boost pressure preset value depends on atmospheric pressure and engine speed, and this preset value is not less than atmospheric pressure; This preset value equals atmospheric pressure + ,in n Engine speed, Atmospheric pressure.

4. The booster pressure control method according to claim 1, characterized in that, The stable conditions for boost pressure self-learning include: The self-learning stabilization phase has been entered beyond the preset first time; The self-learning of the minimum boost pressure has not been updated beyond the preset second time; When all of the above conditions are met, the stability condition for minimum boost pressure self-learning is satisfied.

5. The booster pressure control method according to claim 1, characterized in that, Actual boost pressure and throttle inlet pressure filtering methods include: In the formula, This is the signal before filtering, which is the actual boost pressure or throttle inlet pressure. The signal before filtering in the Nth sampling period. This is the filtered signal. The filtered signal for the Nth sampling period. This is the filtered signal after the (N-1)th sampling period, where N = 1, 2, 3, ... It equals the unfiltered signal at the 0th sampling period. is a coefficient.

6. The booster pressure control method according to claim 1, characterized in that, The self-learning update coefficient calculation methods for minimum boost pressure include: in, This represents the average throttle inlet pressure after filtering. This represents the average throttle inlet pressure before filtering. This represents the average actual boost pressure after filtering. This represents the average actual boost pressure before filtering. if: 1) and ,but: 2) and ,but: Among them, the weighting coefficient Take 0.9; 3) and ,but: Among them, the weighting coefficient Take 0.8; 4) and ,but: Among them, the weighting coefficient Take 0.1; 5) and ,but: Among them, the weighting coefficient Take 0.6; 6) and ,but: Among them, the weighting coefficient Take 0.3; 7) and ,but: Among them, the weighting coefficient Take 0.02; 8) and ,but: Among them, the weighting coefficient Take 0.12; 9) and ,but: Among them, the weighting coefficient Take 0.08; In the formula, The coefficients are self-learning and updated at most once during a single driving cycle. This is the self-learning update coefficient from the last self-learning update. This represents the average of the minimum actual boost pressure before filtering. It is an adjustment coefficient for the minimum boost pressure and can be continuously updated through self-learning.

7. The booster pressure control method according to claim 6, characterized in that, The self-learning update method for the minimum boost pressure adjustment coefficient includes: After the self-learning update coefficients are updated, the calculation is repeated: and If it still exists: 1) and ,but 2) and ,but 3) In other cases, In the formula, This is the adjustment factor from the last self-learning update.

8. The booster pressure control method according to claim 1, characterized in that, The self-learning update coefficient calculation methods for maximum boost pressure include: in, This represents the average throttle inlet pressure after filtering. This represents the average throttle inlet pressure before filtering. This represents the average actual boost pressure after filtering. This represents the average actual boost pressure before filtering. if: 1) and ,but: 2) and ,but: Among them, the weighting coefficient Take 0.9; 3) and ,but: Among them, the weighting coefficient Take 0.8; 4) and ,but: Among them, the weighting coefficient Take 0.1; 5) and ,but: Among them, the weighting coefficient Take 0.6; 6) and ,but: Among them, the weighting coefficient Take 0.3; 7) and ,but: Among them, the weighting coefficient Take 0.02; 8) and ,but: Among them, the weighting coefficient Take 0.12; 9) and ,but: Among them, the weighting coefficient Take 0.08; In the formula, The coefficients are self-learning and updated at most once during a single driving cycle. This is the self-learning update coefficient from the last self-learning update. This represents the average of the actual maximum boost pressure before filtering. It is an adjustment coefficient for the maximum boost pressure and can be continuously updated through self-learning.

9. The booster pressure control method according to claim 8, characterized in that, The self-learning update method for the adjustment coefficient of maximum boost pressure includes: After the self-learning update coefficients are updated, the calculation is repeated: and If it still exists: 1) and ,but 2) and ,but 3) In other cases, In the formula, This is the adjustment factor from the last self-learning update.

10. The booster pressure control method according to claim 1, characterized in that, The minimum boost pressure for the current operating condition is updated based on the self-learning update coefficient of the minimum boost pressure as follows: In the formula, The updated minimum boost pressure. This is the original minimum boost pressure. The self-learning update coefficient is the minimum boost pressure. The maximum boost pressure under the current operating condition is updated based on the self-learning update coefficient of the maximum boost pressure as follows: In the formula, This is the updated maximum boost pressure. This is the original maximum boost pressure. This is the self-learning update coefficient for the maximum boost pressure.

Images