Self-adaptive control method and device for supercharging system of natural gas engine

Abstract

The invention provides a self-adaptive control method and device for a supercharging system of a natural gas engine, and belongs to the field of engines. The rotating speed, the air inlet flow, first information and the required torque gradient of the natural gas engine are obtained, whether the natural gas engine is in the low-load working condition or not is judged based on the rotating speed and the air inlet flow, and the first information comprises the accelerator opening degree change rate, the rotating speed change rate, the supercharging pressure and the load change slope; when the natural gas engine is in the low-load working condition, determining whether the current working condition is a steady-state working condition or a transient working condition according to the first information; and under the condition that the current working condition is the steady-state working condition, the opening degree of the deflation valve is controlled to be in a full-open state, under the condition that the current working condition is the transient working condition, the transient working condition is divided into multiple transient grades according to the accelerator opening degree change rate and the required torque gradient, and the opening degree of the deflation valve is controlled according to the transient grades. According to the scheme, the problem that the economical efficiency and the dynamic property of the natural gas engine under low load are difficult to balance is solved.

Description

Technical Field

[0001] The present application relates to the field of engines, and in particular to a method for adaptively controlling a boosting system of a natural gas engine, an adaptive control device for a boosting system of a natural gas engine, a computer-readable storage medium, and an electronic device. Background Art

[0002] In engine technology, the application of turbocharging has enhanced the performance of natural gas engines. However, due to the limited control strategy of the traditional bleed valve in the supercharging system, it is difficult to achieve a good balance between economy and transient power response. The timely opening and closing of the bleed valve directly affects the performance of the engine under different operating conditions. The control strategy of the turbocharger bleed valve under low-load conditions still faces technical difficulties.

[0003] Most existing bleed valve control strategies rely on fixed thresholds. This means the valve is fully open during steady-state low-load conditions to reduce pumping losses, while fully closed during transient acceleration to rapidly build boost pressure. In short, this control approach forces either full bleed or full boost during low-load operation, making it difficult to balance both fuel economy and power. Summary of the Invention

[0004] The main purpose of this application is to provide a method for adaptively controlling a boost system of a natural gas engine, an adaptive control device for a boost system of a natural gas engine, a computer-readable storage medium, and an electronic device, so as to at least solve the problem in the prior art that it is difficult to balance the economy and power performance of natural gas engines at low loads.

[0005] To achieve the above objectives, according to one aspect of the present application, a method for adaptively controlling a supercharging system of a natural gas engine is provided, comprising: obtaining a speed, an intake air flow rate, first information, and a required torque gradient of the natural gas engine; and based on the speed and intake air flow rate, determining whether the natural gas engine is in a low-load operating condition using a preset speed and intake air flow map, wherein the first information includes a throttle opening change rate, a speed change rate, a boost pressure, and a load change slope; and the required torque gradient is the ratio of the torque change required by the natural gas engine within a preset time window to the preset time window; when the natural gas engine is in the low-load operating condition, determining whether the current operating condition is a steady-state condition or a transient operating condition based on the first information; when the current operating condition is the steady-state condition, controlling the opening of a purge valve to be fully open; and when the current operating condition is the transient condition, using fuzzy logic control to classify the transient operating condition into multiple transient levels based on the throttle opening change rate and the required torque gradient, and controlling the opening of the purge valve based on the transient levels.

[0006] Optionally, the method further includes: obtaining historical hidden state information and historical observed state information within a first preset time period, wherein the historical hidden state information includes historical steady-state operating conditions and historical transient operating conditions, and the historical observed state information includes historical throttle opening change rate, historical speed change rate, historical boost pressure and historical load change slope; determining the historical hidden state information and the historical observed state information as a training data set, and inputting the training data set into a hidden Markov model for training until the hidden Markov model reaches a preset convergence condition, thereby obtaining a target hidden Markov model for future state prediction; inputting the current observed state into the target hidden Markov model for prediction, thereby obtaining a hidden state sequence within a second preset time period and a change probability of the hidden state sequence; comparing the change probability of the hidden state sequence with a preset steady-state probability threshold, and adjusting the opening of the air release valve when the change probability exceeds the preset steady-state probability threshold.

[0007] Optionally, when the natural gas engine is in the low-load operating condition, the current operating condition is determined to be a steady-state operating condition or a transient operating condition based on the first information, including: monitoring the throttle opening change rate, the speed change rate, the boost pressure and the load change slope; when the preset conditions are met within the preset transient window length, the current operating condition is determined to be the transient operating condition, wherein the preset conditions include the throttle opening change rate being greater than a first threshold, the speed change rate being greater than a second threshold, the boost pressure fluctuation range being greater than a third threshold, and the absolute value of the load change slope being greater than a fourth threshold.

[0008] Optionally, after determining whether the natural gas engine is in a low-load operating condition using a preset speed and an intake air flow rate MAP, the method further includes: when the natural gas engine is not in the low-load operating condition, performing PID control on the purge valve based on a preset boost pressure of the current operating condition of the natural gas engine, wherein the preset boost pressure is obtained by pre-calibration, and for each operating point under a non-low-load operating condition, there is a unique corresponding preset boost pressure.

[0009] Optionally, when the current operating condition is the transient operating condition, fuzzy logic control is adopted to divide the transient operating condition into multiple transient levels according to the throttle opening change rate and the required torque gradient, including: when the current operating condition is the transient operating condition, fuzzy logic control is adopted to fuzzify the throttle opening change rate into a first fuzzy set, a second fuzzy set and a third fuzzy set, and to fuzzify the required torque gradient into a first level, a second level and a third level; a rule base is set based on the first fuzzy set, the second fuzzy set, the third fuzzy set, the first level, the second level and the third level; and the rule base is used to divide the transient operating condition into multiple transient levels, and the transient levels include first-level transient, second-level transient and third-level transient.

[0010] Optionally, the opening of the air bleed valve is controlled according to the transient level, including: if the transient level is a level one transient, using a first preset air bleed valve closing slope to control the opening of the air bleed valve; if the transient level is a level two transient, using a second preset air bleed valve closing slope to control the opening of the air bleed valve; if the transient level is a level three transient, using a third preset air bleed valve closing slope to control the opening of the air bleed valve, wherein the first preset air bleed valve closing slope, the second preset air bleed valve closing slope and the third preset air bleed valve closing slope are determined during the test bench calibration process.

[0011] Optionally, the preset speed and intake air flow MAP include a hysteresis interval, and when the changes in the speed and intake air flow of the natural gas engine are within the hysteresis interval, it is determined that the judgment of the current operating condition does not change.

[0012] According to another aspect of the present application, an adaptive control device for a supercharging system of a natural gas engine is provided, comprising: a first acquisition unit, configured to acquire a speed, an intake air flow rate, first information, and a required torque gradient of the natural gas engine, and based on the speed and intake air flow rate, determine whether the natural gas engine is in a low-load operating condition using a preset speed and intake air flow map, wherein the first information includes a throttle opening change rate, a speed change rate, a boost pressure, and a load change slope, and the required torque gradient is a ratio of a torque change required by the natural gas engine within a preset time window to the preset time window; a determination unit, configured to determine, when the natural gas engine is in the low-load operating condition, whether a current operating condition is a steady-state condition or a transient operating condition based on the first information; and a first control unit, configured to control the opening of a purge valve to be fully open if the current operating condition is the steady-state condition, and to, when the current operating condition is the transient condition, employ fuzzy logic control to classify the transient operating condition into multiple transient levels based on the throttle opening change rate and the required torque gradient, and to control the opening of the purge valve based on the transient levels.

[0013] According to another aspect of the present application, a computer-readable storage medium is provided, which includes a stored program, wherein when the program is run, the device where the computer-readable storage medium is located is controlled to execute any one of the aforementioned methods for adaptively controlling a boost system of a natural gas engine.

[0014] According to another aspect of the present application, an electronic device is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing any one of the aforementioned methods for adaptive control of a boost system of a natural gas engine.

[0015] The technical solution of the present application is applied to obtain the speed, intake air flow, first information, and required torque gradient of a natural gas engine. Based on the speed and intake air flow, a preset speed and intake air flow MAP is used to determine whether the natural gas engine is in a low-load operating condition. The first information includes the throttle opening change rate, the speed change rate, the boost pressure, and the load change slope. The required torque gradient is the ratio of the torque change required by the natural gas engine within a preset time window to the preset time window. When the natural gas engine is in a low-load operating condition, the first information is used to determine whether the current operating condition is a steady-state condition or a transient operating condition. When the current operating condition is a steady-state condition, the opening of the purge valve is controlled to be fully open. When the current operating condition is a transient condition, fuzzy logic control is used to classify the transient operating condition into multiple transient levels based on the throttle opening change rate and the required torque gradient, and the opening of the purge valve is controlled based on the transient level. In this solution, by identifying whether the natural gas engine is in a low-load operating condition, if it is in a low-load condition, a steady-state or transient judgment is made. In the transient condition, fuzzy logic control is used to classify the transient condition, and the opening of the bleed valve is controlled according to the classification to achieve refined control, avoiding power hysteresis or energy waste caused by transient response in the existing technology, and allowing the bleed valve to be fully open at steady-state low load, which can reduce unnecessary pumping loss caused by exhaust gas driving the turbine, and has better economy, thereby solving the problem of difficult balance between economy and power performance of natural gas engines at low load. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings that constitute part of this application are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation on this application. In the drawings:

[0017] Figure 1 A hardware structure block diagram of a mobile terminal for executing an adaptive control method for a boosting system of a natural gas engine provided in an embodiment of the present application is shown;

[0018] Figure 2 A flow chart of a method for adaptively controlling a boosting system of a natural gas engine according to an embodiment of the present application is shown;

[0019] Figure 3 The figure shows the preset speed and intake air flow rate MAP of a natural gas engine boost system adaptive control method provided in accordance with an embodiment of the present application;

[0020] Figure 4 A flowchart of a specific adaptive control method for a boosting system of a natural gas engine provided in accordance with an embodiment of the present application is shown;

[0021] Figure 5A schematic diagram of the adjustment slope of the bleed valve opening under a graded transient working condition of a specific adaptive control method for a supercharging system of a natural gas engine provided in an embodiment of the present application is shown;

[0022] Figure 6 A structural block diagram of an adaptive control device for a boosting system of a natural gas engine provided according to an embodiment of the present application is shown.

[0023] The above drawings include the following reference numerals:

[0024] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. DETAILED DESCRIPTION

[0025] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0026] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] As introduced in the background technology, the existing air release valve control strategy adopts either full air release or full boost when running at low load, which makes it difficult to take into account both economy and power at the same time. In order to solve the problem of the difficulty in balancing the economy and power of natural gas engines at low load in the existing technology, the embodiments of the present application provide a natural gas engine boost system adaptive control method, a natural gas engine boost system adaptive control device, a computer-readable storage medium and an electronic device.

[0029] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0030] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for an adaptive control method of a natural gas engine boost system according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown) a processor 102 (the processor 102 may include but is not limited to a microprocessor MCU or a programmable logic device FPGA and other processing devices) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0031] The memory 104 can be used to store computer programs, such as software programs and modules of application software, such as the computer program corresponding to the device information display method in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implementing the above-mentioned method. The memory 104 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned networks include but are not limited to the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, abbreviated as NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] In this embodiment, a method for adaptively controlling a boost system of a natural gas engine running on a mobile terminal, a computer terminal, or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] Figure 2 FIG. 1 is a flow chart of an adaptive control method for a boost system of a natural gas engine according to an embodiment of the present application. Figure 2 As shown, the method includes the following steps:

[0034] Step S201: Obtaining the speed, intake air flow, first information, and required torque gradient of the natural gas engine, and based on the speed and intake air flow, determining whether the natural gas engine is in a low-load operating condition using a preset speed and intake air flow MAP, wherein the first information includes a throttle opening change rate, a speed change rate, a boost pressure, and a load change slope, and the required torque gradient is the ratio of the torque change required by the natural gas engine within a preset time window to the preset time window;

[0035] Specifically, first obtain the key data of the operation of the natural gas engine (hereinafter referred to as the engine), including speed, intake flow, first information (including throttle opening change rate, speed change rate, boost pressure and load change slope) and required torque gradient. These parameters are crucial for real-time monitoring of the engine status and making adaptive control decisions. The engine speed can be obtained through a crankshaft position sensor or a camshaft position sensor, and the intake flow can be obtained through an intake flow sensor. By comparing the real-time speed and intake flow of the engine, and matching them with the preset speed and intake flow MAP, it is determined whether the engine is working in a low-load condition. The preset speed and intake flow MAP are calibrated based on a large amount of experimental data, and can reflect the load condition of the engine under a specific combination of speed and intake flow. The preset speed and intake flow MAP are as follows: Figure 3 As shown, based on the preset speed and intake air flow MAP, it is determined whether the exhaust valve is in the area where it can be fully opened or fully closed (i.e., low load condition determination). Figure 3 , the area that can be fully opened and fully closed is 1, that is, when in area 1, the natural gas engine is in a low-load condition.

[0036] The first information includes the throttle position change rate, speed change rate, boost pressure, and load change slope. These are key indicators reflecting the engine's current operating status and operating condition trends. The throttle position change rate refers to the magnitude of the throttle position change over a short period of time. It reflects the driver's operating intention and is an important indicator for identifying transient operating conditions. The throttle position change rate can be obtained from the throttle position sensor. The speed change rate indicates the rate of change of engine speed over time and is also a key parameter for determining transient operating conditions. The speed change rate can be calculated from the speed obtained by the crankshaft position sensor or camshaft position sensor. For example, the speed difference between two sampling points is calculated and divided by the time difference to obtain the instantaneous speed change rate. Boost pressure directly affects the intake air volume, which in turn affects the engine's power output and combustion efficiency. Boost pressure can be obtained from the boost pressure sensor. The load change slope reflects the load change trend over time and is helpful for predicting future operating conditions and adjusting control strategies. Load can be estimated by monitoring intake air flow and engine speed. The load change slope is the amount of load change per unit time. The load change slope is calculated by continuously monitoring the load.

[0037] The demand torque gradient measures the engine's ability to increase or decrease torque per unit time. It is calculated by calculating the ratio of the change in demand torque within a preset time window to the length of that window. The demand torque gradient helps quickly respond to varying levels of torque demand, particularly in transient conditions, allowing timely adjustment of the blow-off valve opening to meet power response and energy conservation requirements. The demand torque gradient is calculated by combining information such as the driver's accelerator pedal position, vehicle acceleration, and driving conditions. The demand torque gradient is calculated by monitoring changes in demand torque, specifically the change in demand torque per unit time.

[0038] Step S201 significantly improves the operating efficiency and power response of natural gas engines under low-load conditions. Specifically, by real-time monitoring and analysis of key parameters such as speed, intake air flow, throttle opening rate of change, speed rate of change, boost pressure, load change slope, and required torque gradient, combined with preset speed and intake air flow map, it accurately determines whether the engine is in a low-load condition. This establishes a comprehensive operating condition identification system, laying a solid foundation for subsequent selection of the purge valve control strategy.

[0039] Step S202: when the natural gas engine is in the low-load operating condition, determining whether the current operating condition is a steady-state operating condition or a transient operating condition based on the first information;

[0040] Specifically, a steady-state operating condition refers to an engine operating condition in which its operating parameters (such as speed, intake air volume, and load) remain essentially stable over a prolonged period. In this case, the throttle opening rate of change, speed rate of change, boost pressure, and load slope all exhibit minimal fluctuation or are nearly zero, indicating that the engine is operating smoothly. A transient operating condition refers to a situation in which the engine's state rapidly changes, such as during acceleration, deceleration, or a sudden load increase. During a transient operating condition, the various parameters in the first information will experience significant fluctuations, such as a sharp increase in the throttle opening rate of change, a significant speed rate of change, and a corresponding increase in the boost pressure and load slopes. Significant fluctuations in these indicators indicate that the engine is experiencing a transient operating condition. Continuously monitoring the first information (throttle opening rate of change, speed rate of change, boost pressure, and load slope) reveals the dynamic relationship between driver input and engine response. Preset thresholds or criteria (such as the magnitude of the throttle opening rate of change or whether the speed rate of change exceeds a certain threshold) are used to determine whether the current operating condition is transient or still within the steady-state range.

[0041] During low-load operation of a natural gas engine, real-time monitoring and analysis of primary information, such as the rate of change of throttle opening, speed, boost pressure, and load slope, enables precise differentiation between steady-state and transient engine operating conditions. This differentiation ensures that the purge valve control strategy accurately responds to the engine's true operating mode, significantly improving engine flexibility and efficiency under low-load conditions.

[0042] Step S203, when the above-mentioned current operating condition is the above-mentioned steady-state operating condition, the opening of the air release valve is controlled to be in a fully open state. When the above-mentioned current operating condition is the above-mentioned transient operating condition, fuzzy logic control is adopted to divide the above-mentioned transient operating condition into multiple transient levels according to the above-mentioned throttle opening change rate and the above-mentioned required torque gradient, and the opening of the above-mentioned air release valve is controlled according to the above-mentioned transient levels.

[0043] Specifically, when the current operating condition is determined to be a steady-state condition, the opening of the bleed valve is maintained at a fully open state. Steady-state conditions mean that the engine operating parameters are relatively stable, and the torque and power requirements do not change drastically. The setting of a fully open air valve can minimize pumping losses and improve combustion efficiency, thereby significantly reducing fuel consumption under low-load conditions and enhancing the economic performance of the engine. This is because at steady-state low loads, the engine does not need to generate too much compressed air through the turbocharger. A fully open air valve allows excess exhaust gas to be discharged directly, avoiding unnecessary energy loss.

[0044] However, once a transient condition is detected, such as when the engine is facing acceleration or a sudden load change, a more complex fuzzy logic control algorithm is activated. Based on the throttle opening rate and the required torque gradient, the transient condition is subdivided into multiple levels, each corresponding to a different speed and amplitude of the bleed valve closing slope, thus achieving refined control of the transient response. The core of fuzzy logic control lies in its ability to handle non-binary, imprecise inputs such as the throttle opening rate and the required torque gradient. These two parameters may inherently exhibit uncertainty or gradual changes, rather than simply being "on" or "off." By mapping these inputs to predefined fuzzy sets, the severity of the transient condition can be assessed, and the optimal bleed valve control strategy is then selected based on this assessment. For example, for mild transient conditions (such as gentle acceleration), a gentler bleed valve closing slope can be selected to ensure smooth power delivery while avoiding excessive pumping losses. For intense transient conditions (such as emergency acceleration), a steeper bleed valve closing slope can be used to rapidly build higher boost pressure to meet sudden torque demands and ensure the engine's power response. This hierarchical control mechanism allows bleed valve adjustment to more closely match the engine's actual needs, providing sufficient power support in transient conditions while maintaining good economic performance in steady states. This achieves a dynamic balance between power and economy for natural gas engines under varying operating conditions.

[0045] In summary, by distinguishing between steady-state and transient operating conditions and combining fuzzy logic control, the problem of the engine having to consider both economy and power response under low-load conditions is effectively solved, and the adaptability and efficiency of the control system in dealing with complex operating environments are improved.

[0046] This embodiment integrates multiple dimensions of information, including speed, intake air flow, throttle opening rate, speed change rate, boost pressure, and load change slope, and combines them with preset speed and intake air flow MAP to accurately determine whether the natural gas engine is in a low-load phase. It also precisely diagnoses whether the operating state is steady or transient. Under steady-state conditions, the purge valve remains fully open, minimizing pumping losses and significantly improving fuel economy. Under transient conditions, this method uses fuzzy logic control to intelligently classify transient states based on the dynamic changes in throttle opening rate and demand torque gradient, thereby precisely regulating the purge valve opening and ensuring rapid response and smooth transitions in power output under varying transient demand scenarios. This method not only effectively balances the power and fuel economy of natural gas engines under low-load conditions, but also significantly improves engine operating flexibility and overall performance through intelligent operating condition recognition and adaptive control, thereby resolving the difficult balance between fuel economy and power in natural gas engines under low-load conditions.

[0047] During the specific implementation process, the above method also includes: obtaining historical hidden state information and historical observation state information within a first preset time period, wherein the above historical hidden state information includes historical steady-state operating conditions and historical transient operating conditions, and the above historical observation state information includes historical throttle opening change rate, historical speed change rate, historical boost pressure and historical load change slope; determining the above historical hidden state information and the above historical observation state information as a training data set, and inputting the above training data set into the hidden Markov model for training until the above hidden Markov model reaches the preset convergence condition, thereby obtaining a target hidden Markov model for future state prediction; inputting the current observation state into the above target hidden Markov model for prediction, thereby obtaining a hidden state sequence within a second preset time period and a change probability of the above hidden state sequence; comparing the above change probability of the change probability of the above hidden state sequence with a preset steady-state probability threshold, and adjusting the opening of the above air release valve when the above change probability exceeds the above preset steady-state probability threshold.

[0048] This embodiment introduces an operating condition prediction mechanism based on a hidden Markov model (HMM) to further improve the adaptive control accuracy of natural gas engine power and fuel economy under low-load conditions. This mechanism collects and analyzes historical operating condition data to predict future operating conditions, preemptively adjusting the bleed valve opening to ensure the engine can adapt more smoothly and efficiently to upcoming changes. Within the Hidden Markov Model framework, hidden states refer to internal system states that cannot be directly observed, while observed states are external manifestations that can be directly measured by sensors or other devices. In this embodiment, hidden states correspond to the engine operating condition category: steady-state or transient, while observed states include specific measurable parameters such as throttle opening rate of change, speed rate of change, boost pressure, and load slope.

[0049] Specifically, engine operating data from a past period (i.e., a first preset time period) is acquired, including historical throttle opening rate of change, speed rate of change, boost pressure, and load slope. This data reflects the engine's past operating conditions and external load variations. The first preset time period is primarily based on engineering experience and experimental research, as well as dynamic response and control theory. Engineers typically determine the appropriate length of the preset time period based on previous experimental data and engine operating characteristics. For example, this is determined by analyzing the time required for the engine to transition from one operating condition to another, as well as the rate of change and stability of various parameters (such as throttle opening, speed, and boost pressure) during this transition. The first preset time period should be long enough to capture typical operating condition cycles, while also avoiding data redundancy and computational overhead caused by excessive length. The dynamic response characteristics of the purge valve control and boost pressure systems are also important considerations in setting the first preset time period, which involves the concepts of delay time and response time in control system theory. For example, if the bleed valve takes several seconds to fully open or close, the first preset time period should at least include this time span so that the model can take into account the dynamic behavior of the bleed valve. At the same time, the first preset time period also needs to match the engine's torque response time to ensure that under transient conditions, the adjustment of the bleed valve can promptly affect the engine's output torque, thereby better adapting to the driver's operating intentions and road conditions. In addition, the setting of the first preset time period is also affected by noise filtering, computing resources, prediction accuracy and robustness. In short, the setting of the first preset time period is a comprehensive consideration that needs to balance factors such as engine dynamic characteristics, control response time, noise filtering requirements and computing resource constraints. The purpose is to build a model that can accurately reflect historical operating conditions and efficiently predict future states, so as to achieve precise control of the natural gas engine boost system.

[0050] Next, based on this historical data, historical operating conditions are identified and categorized into historical steady-state conditions and historical transient conditions. This process involves in-depth data analysis, identifying periods of stable engine operation with minimal parameter changes as historical steady-state conditions, and periods of large parameter fluctuations requiring rapid engine response as historical transient conditions.

[0051] Historical hidden state information and historical observed state information are combined into a training dataset for learning and training a hidden Markov model. The hidden Markov model infers a hidden state sequence based on the observed state information, i.e., a time series of steady-state or transient operating conditions. Model parameters are continuously updated to enable the model to accurately predict the relationship between the observed and hidden states until the model reaches a preset convergence condition. This means that the model's prediction accuracy meets a predetermined standard or the prediction error no longer improves significantly. The preset convergence condition can be set based on a maximum number of iterations or a parameter change threshold. Model training can be set to a maximum number of iterations. Once this limit is reached, training will terminate regardless of whether the model is truly stable. This helps prevent the model from falling into an endless training loop and also limits training time and computing resource consumption. During each iteration, model parameters (such as state transition probabilities and emission probabilities) are updated. If the parameter change within several consecutive iterations is less than a preset threshold, this indicates that the model parameters have stabilized and the learning curve has flattened. At this point, the model is considered to have converged and training is terminated. After model training is complete, a target hidden Markov model is obtained. This target hidden Markov model can be used to predict the hidden state sequence within a second preset time period based on current observed state information (such as the real-time throttle opening rate of change and engine speed rate of change). This is the series of possible steady-state and transient operating conditions that the engine may experience. The probability of a hidden state sequence change refers to the probability of transitioning from one hidden state to another at each moment in the second preset time period. For example, the probability of transitioning from a steady state to a transient state, or from a transient state back to a steady state, is predicted. This probability distribution not only predicts the hidden state at the next moment but also provides information about the uncertainty of future state transitions, which is crucial for understanding the continuity and randomness of system behavior. The second preset time period is primarily determined by the characteristics of the engine operating state transitions and the control system's response time requirements. The selection of the second preset time period needs to comprehensively consider the effectiveness of the operating condition prediction, the response time of the control system, engineering practice experience, and safety and performance considerations. Specifically, it needs to be long enough to capture potential operating condition changes, but not too long that the prediction becomes inaccurate or outdated. Generally speaking, the lower limit of this period is determined by the conversion time of the engine operating condition (such as from steady state to transient) and the complexity of the prediction model; the control system needs time to receive the prediction information and make corresponding adjustments, such as adjusting the opening of the bleed valve. The second preset time period should be greater than or equal to the time required for the control system to respond, so as to ensure that the prediction information can be effectively utilized before the actual operating condition changes and pre-adaptive adjustments can be made; based on past test and usage data, it can be understood on what time scale the engine tends to undergo operating condition conversion under specific operating conditions.For example, based on experience, the transition from a low-load steady state to a transient acceleration in a natural gas engine may occur within a few seconds to more than ten seconds. Therefore, the second preset time period should cover this range. To ensure that the engine can obtain the required power in a timely manner under all circumstances while avoiding unnecessary energy waste, the second preset time period should also take into account the needs of safety redundancy and performance optimization. For example, in the case of emergency acceleration, a shorter prediction time window may be more advantageous because it requires the system to react more quickly.

[0052] The predicted probability of the hidden state sequence change is compared with a preset steady-state probability threshold. When the probability of the hidden state sequence change exceeds the preset steady-state probability threshold, it indicates that the engine is about to enter or is currently in a transient operating condition. Based on this, the bleed valve opening is adjusted in advance to optimize engine power output and smoothness. This probabilistic prediction-based active control improves the ability to respond to future operating condition changes and avoids control delays caused by passive responses. Specifically, when the probability of the hidden state sequence change exceeds the preset steady-state probability threshold, it indicates that the engine is about to enter or is currently in a transient operating condition. Based on this prediction, the bleed valve opening is adjusted in advance to ensure timely engine response to transient conditions while maintaining smooth operation and efficiency. The specific steps for adjusting the bleed valve opening are: Selecting a corresponding bleed valve closing slope based on the transient level. Higher transient levels result in steeper bleed valve closing slopes, allowing for a rapid increase in boost pressure and intake air volume to meet sudden increases in power demand. The bleed valve opening is adjusted in advance based on the selected bleed valve closing slope.

[0053] By introducing the hidden Markov model, it is not only possible to respond to the current operating conditions, but also to anticipate the changing trends of future operating conditions, allowing for early responses and avoiding the power hysteresis or energy waste that may be caused by delayed responses in traditional control. By learning historical data through the hidden Markov model, it is possible to more accurately predict the conversion of engine operating conditions, especially the transition from transient to steady state or vice versa under low load, thereby achieving more precise bleed valve control, ensuring that the engine's power and economy can achieve the best balance under various operating conditions, and significantly improving the engine's overall performance and operating efficiency. In short, through the learning and prediction of the hidden Markov model, it is possible not only to judge the engine's operating condition type in real time based on the current observation state, but also to predict the changing trends of operating conditions in the short term in the future, thereby achieving more refined and forward-looking supercharging system control, significantly improving the flexibility, power and economy of natural gas engines under low-load conditions.

[0054] Furthermore, when the natural gas engine is in the low-load operating condition, the current operating condition is determined to be a steady-state operating condition or a transient operating condition based on the first information, including: monitoring the throttle opening change rate, the speed change rate, the boost pressure and the load change slope; when the preset conditions are met within the preset transient window length, the current operating condition is determined to be the transient operating condition, wherein the preset conditions include the throttle opening change rate being greater than a first threshold, the speed change rate being greater than a second threshold, the boost pressure fluctuation range being greater than a third threshold, and the absolute value of the load change slope being greater than a fourth threshold.

[0055] Specifically, the throttle opening change rate directly reflects the driver's operating intention, especially during acceleration or deceleration, the rapid change of throttle opening is an important sign of transient operating conditions; the speed change rate, observes the rate of change of engine speed over time, sudden increase or decrease in speed, especially large changes beyond the normal steady-state fluctuation range, indicates the occurrence of transient operating conditions; the boost pressure fluctuation range, evaluates the fluctuation of the boost pressure provided by the turbocharger over time. During steady-state operation, the boost pressure is relatively stable; under transient conditions, the boost pressure will fluctuate greatly; the load change slope, monitors the rate of change of engine load over time. The increase in the load change slope means that the engine responds quickly to changes in external conditions, which is also a feature of transient operating conditions.

[0056] The preset transient window length is a short period of time used to collect and analyze key parameters to determine whether a transient operating condition has occurred. The preset transient window length should be selected based on engine characteristics and control system response time to ensure timely capture of transient events without being too long, which may cause identification delays. The preset conditions in this embodiment include: A throttle opening change rate greater than a first threshold: If the throttle opening changes too rapidly within the preset transient window length, exceeding the pre-set first threshold, this indicates that the engine is accelerating and entering a transient operating condition. A speed change rate greater than a second threshold: Similarly, rapid changes in speed also indicate a transient operating condition. Therefore, if the speed change rate exceeds the second threshold, this change is considered a transient signal. A boost pressure fluctuation range greater than a third threshold: Significant fluctuations in boost pressure, if exceeding the third threshold, indicate a transition from steady state to transient state. A load change slope greater than a fourth threshold: Rapid changes in engine load (whether increasing or decreasing), if the absolute value of the load change slope exceeds the fourth threshold, are also considered triggers for a transient operating condition. The first, second, third, and fourth thresholds were determined based on a comprehensive analysis of bench test data, engine physical model analysis, typical driver operating patterns, the speed and range of operating condition changes, and rigorous considerations of safety and performance. By quantitatively analyzing the engine's transition characteristics between steady-state and transient conditions, combined with historical data statistics and physical limits, each threshold was designed to differentiate between different operating conditions, ensuring timely and accurate response to transient events while avoiding the negative impact of over-regulation.

[0057] By continuously monitoring these key parameters and checking for the presence of pre-defined conditions within a preset transient window (i.e., the throttle opening rate of change exceeding a first threshold, the speed rate of change exceeding a second threshold, the boost pressure fluctuation range exceeding a third threshold, and the absolute value of the load slope exceeding a fourth threshold), the system can quickly and effectively determine whether the engine is currently in a transient state. If all of these conditions are met, the current operating condition is marked as transient; otherwise, it is considered a steady-state condition. The preset transient window length is determined based on both the average time it takes for the engine to transition from steady-state to transient state across different driving modes and the minimum period required for the control system to sense and respond to the change. By analyzing extensive experimental data and historical operating records, combined with predictions from engine physics models, the system identifies a time range that captures early signals of transient conditions while avoiding the effects of delayed response due to an overly long window or the inability to filter noise due to a too short window.

[0058] By real-time monitoring of the throttle opening rate of change, the speed rate of change, boost pressure fluctuations, and the load change slope, combined with the threshold conditions in the preset transient window length, it is possible to quickly identify whether the engine is entering a transient state. This mechanism ensures accurate response at the initial stage of transient conditions. By adjusting the opening strategy of the bleed valve, it effectively balances power and economy, reduces unnecessary pumping losses, and enhances the engine's power response speed and smoothness. More importantly, it can adapt to diverse driving scenarios and driver operating habits, ensuring that the engine can maintain optimal performance under complex and changing operating conditions, thereby improving the stability and reliability of the entire system. At the same time, by monitoring multiple parameters rather than a single indicator, the robustness and accuracy of operating condition judgment are improved, the possibility of misjudgment is reduced, and stable operation is ensured under complex operating conditions.

[0059] In some embodiments of the present application, after determining whether the natural gas engine is in a low-load operating condition using a preset speed and intake flow rate MAP, the method further includes: when the natural gas engine is not in the low-load operating condition, performing PID control on the purge valve based on a preset boost pressure of the current operating condition of the natural gas engine, wherein the preset boost pressure is obtained by pre-calibration, and for each operating point under non-low-load conditions, there is a unique corresponding preset boost pressure.

[0060] Specifically, when the natural gas engine is operating under medium-to-high load conditions or other specific operating conditions, a PID control algorithm (proportional-integral-differential control) is used to fine-tune the purge valve. For each operating point outside low-load conditions, a pre-calibrated preset boost pressure is used. This value is precisely calculated based on the engine's characteristic curve and ideal operating conditions. This calibration process typically involves extensive testing and data analysis to determine the optimal boost pressure for each engine operating condition, optimizing both power and fuel economy.

[0061] When a natural gas engine is operating under non-low-load conditions, the preset boost pressure for the current operating conditions is determined by searching a preset MAP (map or characteristic curve) based on real-time monitored engine speed and intake air flow. Subsequently, a PID algorithm is used to compare the actual measured boost pressure with the preset boost pressure, calculate the deviation, and adjust the opening of the blow-off valve accordingly. The PID controller uses the proportional term (P) to quickly respond to deviations, the integral term (I) to eliminate static errors, and the differential term (D) to predict future changes. This allows for quick and accurate adjustment of the blow-off valve to achieve the preset boost pressure, ensuring optimal engine performance under all operating conditions.

[0062] This control strategy ensures more intelligent and personalized adjustment of the blow-off valve under non-low-load conditions. By dynamically adjusting boost pressure, it can meet the high demands for power performance under medium and high-load conditions while also taking into account fuel economy during non-steady-state operation. This sophisticated control logic improves overall engine performance. In short, by introducing dynamic PID control and pre-set boost pressure under non-low-load conditions, intelligent adjustment of the boost system is achieved across the entire engine operating range, thereby comprehensively improving the engine's overall performance.

[0063] This embodiment can introduce a fusion control model that can simultaneously consider multiple working conditions such as steady state, transient state and the initial stage of engine startup, so that the adjustment of the bleed valve is more accurate. The model can identify the characteristics of different working conditions and integrate them to provide a global optimal control strategy. When it is identified that the engine is in a low-load state at the initial stage of startup, it automatically switches to a special control mode to avoid unnecessary pumping air losses during the startup phase, while ensuring that the boost pressure is quickly established. As the engine runs stably, it gradually transitions to conventional transient and steady-state control modes. The multi-condition fusion intelligent control strategy can solve the problem of boost system control in the initial stage of engine startup and achieve a smooth transition from startup to operation. Through comprehensive consideration of multiple working conditions, the intelligence and flexibility of the control system are improved, and it can cope with more complex driving conditions, and reduce energy waste in the initial stage of startup, further improving the economy and environmental performance of the entire engine working condition.

[0064] In order to further increase the power and economy of the engine under transient conditions, when the above-mentioned current operating condition is the above-mentioned transient operating condition, fuzzy logic control is adopted, and the above-mentioned transient operating condition is divided into multiple transient levels according to the above-mentioned throttle opening change rate and the above-mentioned required torque gradient, including: when the above-mentioned current operating condition is the above-mentioned transient operating condition, the above-mentioned fuzzy logic control is adopted to fuzzify the above-mentioned throttle opening change rate into a first fuzzy set, a second fuzzy set and a third fuzzy set, and fuzzify the above-mentioned required torque gradient into a first level, a second level and a third level; a rule base is set based on the above-mentioned first fuzzy set, the second fuzzy set, the third fuzzy set, the first level, the second level and the third level; the above-mentioned rule base is used to divide the above-mentioned transient operating condition into multiple above-mentioned transient levels, and the above-mentioned transient levels include first-level transient, second-level transient and third-level transient.

[0065] Specifically, when a natural gas engine is determined to be experiencing a transient operating condition, fuzzy logic control is introduced to break down the degree of the transient into multiple levels, enabling more precise adjustment of the purge valve response. This process involves two key parameters: the throttle opening rate and the required torque gradient. The throttle opening rate is divided into three fuzzy sets: the first fuzzy set (indicating light acceleration, low throttle opening rate), the second fuzzy set (indicating moderate acceleration, medium throttle opening rate), and the third fuzzy set (indicating urgent acceleration, high throttle opening rate). By fuzzifying the throttle opening rate, the system better understands and reflects the driver's true intention, enabling appropriate control decisions even when the throttle operation is between two clear levels. Similarly, the required torque gradient is fuzzified into levels 1 (small), 2 (medium), and 3 (large). This reflects the magnitude of the change in the engine's required power output during a transient, further refining the description of the transient operating condition.

[0066] Next, based on the fuzzified throttle rate of change and the required torque gradient, a rule base is established to define the specific transient classification levels for different combinations. For example, a high throttle rate of change and a large required torque gradient trigger the highest level of transient response, while a low throttle rate of change and a small required torque gradient result in the lowest level of transient response. This rule base accurately assigns transient conditions to one of three levels: level 1, level 2, or level 3, based on the instantaneous throttle rate of change and the required torque gradient. Each level corresponds to a different bleed valve closing slope and control strategy, ensuring adequate power response while minimizing unnecessary energy loss and optimizing engine operating efficiency.

[0067] The use of fuzzy logic control enables a more intelligent response to driver input, particularly during transient conditions. This overcomes the control discontinuity and over- or under-response issues inherent in traditional control strategies, often caused by the use of fixed thresholds. By fine-tuning transient levels, the adjustment of the blow-off valve becomes smoother and more efficient, improving not only driving comfort and safety but also engine power and economy under transient conditions.

[0068] In other embodiments of the present application, the opening of the above-mentioned bleed valve is controlled according to the above-mentioned transient level, including: if the above-mentioned transient level is a first-level transient, the opening of the above-mentioned bleed valve is controlled by using a first preset bleed valve closing slope; if the above-mentioned transient level is a second-level transient, the opening of the above-mentioned bleed valve is controlled by using a second preset bleed valve closing slope; if the above-mentioned transient level is a third-level transient, the opening of the above-mentioned bleed valve is controlled by using a third preset bleed valve closing slope, wherein the above-mentioned first preset bleed valve closing slope, the above-mentioned second preset bleed valve closing slope and the above-mentioned third preset bleed valve closing slope are determined during the test bench calibration process.

[0069] Specifically, the system uses fuzzy logic to determine the engine's transient level and automatically selects the appropriate bleed valve closing slope for adjustment. This mechanism ensures an appropriate response based on the intensity of the transient condition, avoiding unnecessary energy loss due to overreaction or power loss due to underreaction. Specifically, level 1 transient corresponds to mild acceleration or slight load changes. In these situations, the bleed valve opening is controlled using the first preset bleed valve closing slope. This results in a gentler response, ensuring a smooth transition during the transient and minimizing the impact on driving comfort. When transient conditions intensify slightly, such as during normal acceleration, the system switches to the second preset bleed valve closing slope. This is faster than the first level, more effectively increasing boost pressure to meet the engine's demand for higher torque. For urgent acceleration or high load changes, the third preset bleed valve closing slope is used, providing the fastest response level and designed to quickly build the required boost pressure to meet sudden power demands.

[0070] The determination of the first, second, and third preset bleed valve closing slopes requires calibration on a professional test bench. This process typically includes: Simulating different transient levels: Reproducing various transient operating conditions on the test bench, including light acceleration and deceleration, moderate acceleration, and emergency acceleration, to collect engine operating data under these conditions; Parameter optimization: By continuously adjusting the bleed valve closing slope, find the optimal slope value that can quickly respond to power demand without excessively sacrificing economy at each transient level; Verification and calibration: Applying the selected closing slopes to actual operating conditions, repeatedly testing and verifying them to ensure they can produce the expected results in the actual driving environment; Standard setting: Finally, based on the results of the test bench calibration, corresponding bleed valve closing slope standards are set for different transient levels.

[0071] By controlling the bleed valve closing slope that matches the transient level, the boost system can provide appropriate power support under transient conditions of varying intensities while minimizing pumping losses and other forms of energy waste. This graded response strategy improves the engine's power performance and fuel economy, especially under frequently changing driving conditions, demonstrating greater flexibility and efficiency, thereby enhancing user satisfaction and driving experience. At the same time, the preset slope based on bench calibration ensures the reliability and consistency of the control strategy. In short, by clarifying the bleed valve closing slope corresponding to different transient levels, more accurate and efficient management of the engine's transient conditions is achieved, which is of great significance to improving the overall performance and operating economy of natural gas engines.

[0072] This embodiment can introduce an adaptive learning mechanism to continuously optimize the parameters of fuzzy logic control through real-time data feedback, such as the boundaries of fuzzy sets, the weights of rule bases, etc., that is, it will automatically adapt to changes in different engine and vehicle characteristics during operation without the need for frequent manual adjustments. Specifically, through a parameter self-optimization algorithm based on machine learning, the slope control of the bleed valve is optimally matched with the actual pumping loss, transient response time and power output characteristics of the engine. The algorithm automatically adjusts the slope control curve through online learning to minimize fuel consumption and maximize power response. The adaptive learning mechanism reduces dependence on manual calibration and adjustment, reduces maintenance costs, achieves adaptation to dynamic environments and engine characteristics, and improves the robustness and adaptability of the system. Through continuous optimization, performance can be continuously improved over time to better meet user needs.

[0073] In some further embodiments of the present application, the preset speed and intake flow MAP include a hysteresis interval. When the changes in the speed and intake flow of the natural gas engine are within the hysteresis interval, it is determined that the judgment of the current operating condition does not change.

[0074] Specifically, the preset speed and intake flow MAP is used to store the engine's set parameters at different speeds and intake flows, such as the opening of the exhaust valve or the target value of the boost pressure. However, since the speed and intake flow fluctuate due to various factors (such as load changes, driver operation, etc.) during engine operation, if appropriate measures are not taken, frequent switching between different control modes will lead to unstable control and reduced engine performance. To solve the above problems, a hysteresis range is introduced in the preset speed and intake flow MAP. The hysteresis range can be found in Figure 3 . Simply put, the hysteresis interval sets an allowable range for changes in speed and intake flow. Within this range, although these two parameters have changed, the current operating conditions will not be re-evaluated, and the control strategy of the bleed valve will not be changed. For example, suppose the hysteresis interval is set so that the speed change does not exceed the preset value ΔN, and the intake flow change does not exceed the preset value ΔQ. When the control system determines that the natural gas engine is in a low-load operating condition based on the current speed and intake flow, and adopts a corresponding control strategy, even if these two parameters fluctuate in the subsequent short period of time, as long as the change amplitude is less than ΔN and ΔQ, it will be determined that the judgment of the current operating condition is still valid, and no new control mode switch will be triggered. Only when the parameter change exceeds the upper limit of the hysteresis interval will the operating condition be re-evaluated and the corresponding control adjustment be made.

[0075] The hysteresis band significantly improves robustness against minor disturbances, reduces unnecessary control actions, and avoids engine instability caused by frequent switching of control modes. Furthermore, this mechanism simplifies control logic, reduces computational burden, and makes control more efficient. It also ensures smooth operation and safety in complex driving environments, enhancing the overall driving experience and engine performance.

[0076] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the adaptive control method for the boosting system of the natural gas engine of the present application will be described in detail below with reference to specific embodiments.

[0077] This embodiment relates to a specific adaptive control method for a supercharging system of a natural gas engine, such as Figure 4 As shown, it includes area judgment, initial operating condition identification, transient classification and air release valve slope control, and operating condition prediction control.

[0078] 1. Regional judgment:

[0079] First, based on the pre-calibrated MAP based on engine speed and intake air flow (preset speed and intake air flow MAP), determine whether the engine is in the area where the exhaust valve can be fully opened or fully closed (i.e., low load condition judgment). The preset speed and intake air flow MAP is a calibrated MAP, which can be found in Figure 3 , the area that can be fully opened and fully closed is 1. When it is not in low load condition, the air release valve is PID controlled according to the preset boost pressure (each operating point has its corresponding set boost pressure); when it is in low load condition, it enters Figure 3 The next step is determined within the dotted box. This pre-calibrated MAP has a certain hysteresis range (the hysteresis range is also a calibration value, the purpose of which is to prevent sudden changes in the opening of the exhaust valve). This prevents fluctuations in the exhaust valve opening caused by changes in operating conditions, which may affect engine stability.

[0080] 2. Initial working condition identification:

[0081] Key operating condition characteristics can be captured by driver data (throttle opening rate), environmental parameters (intake air temperature / pressure), system status (current boost pressure, speed), and historical data (load change trends) to accurately distinguish between steady-state and transient conditions. In this embodiment, a transient condition is identified if the throttle opening rate of change is greater than 10% / s, the speed rate of change is greater than 200 rpm / s, the boost pressure fluctuation is greater than 0.5 bar / s, and the absolute value of the load change slope is greater than 1% / s. The transient window length is 1 second. Time window filtering (with a steady-state window length of 5 seconds) is added to the relevant signal processing modules to prevent short-term disturbances from being mistaken for transients.

[0082] When the initial working condition is identified as steady state, the bleed valve is fully opened to ensure economic efficiency. On the contrary, when the initial working condition is identified as transient, the next step is transient classification judgment.

[0083] 3. Transient classification and air release valve slope control:

[0084] Fuzzy logic control is used to dynamically divide the transient level through the throttle opening change rate, required torque gradient, etc., to avoid the disconnection of step response caused by fixed thresholds. Fuzzify the input signal, for example, divide the throttle change rate into three fuzzy sets (0%~10% is low-mild acceleration, 5%~20% is medium-medium acceleration, 15%~30% is high-emergency acceleration, and the overlap of the intervals reflects "fuzziness"), and divide the required torque gradient into small, medium, and large. Formulate a rule base. For example, if the throttle change rate is "high" and the torque gradient is "large", the transient level is level 3. If the throttle change rate is "medium" and the torque gradient is "medium", the transient level is level 2. Clarify the output, and finally convert the fuzzy rule calculation results into specific transient levels (levels 1 / 2 / 3) and the bleed valve closing slope. The schematic diagram of the bleed valve opening adjustment slope under graded transient conditions is as follows. Figure 5 shown.

[0085] The specific transient level classification, the corresponding bleed valve transition slope (bleed valve closing slope) and application scenarios are shown in Table 1.

[0086] Table 1 Transient level and bleed valve transition slope correspondence

[0087]

[0088] 4. Working condition prediction control:

[0089] The system consists of two types of states: hidden states (not directly observable, such as steady state and transient states) and observed states (directly measurable, such as sensor signals). By collecting historical data, the relationship between these two states is learned. A hidden Markov model (HMM) is employed, which consists of two states: hidden and observed. For example, if the hidden state is a level 3 transient, the probability of observing a throttle change rate greater than 20% is 80%. The system then infers the most likely hidden state based on the current observed signal. This is the prediction phase of the HMM. Based on the current observed state, the most likely hidden state sequence is inferred. This helps predict the probability of the transient level changing from level 3 to level 2 to level 1 within the next 5 seconds. This predicts the probability of short-term changes in the future and compares it with a preset steady-state probability threshold calibrated on the test bench (the preset steady-state probability threshold can be calibrated based on big data, with different calibration values for different aircraft models). This module then determines the next operating condition's bleed valve action. This module predicts the probability distribution of the next operating condition, helping to proactively adjust the bleed valve state and ensure more effective action.

[0090] Figure 4 "0: Target Fully Closed" and "1: Target PID Control" represent different operating modes for the blow-off valve. These modes correspond to control strategies for different engine operating conditions, aiming to optimize engine performance, particularly the balance between power and economy. "0: Target Fully Closed" applies when the engine enters transient conditions, such as hard acceleration or a sudden load change, where the goal is to rapidly increase output torque. In these situations, the control system immediately closes the blow-off valve, forcing exhaust gas to flow entirely through the turbine, accelerating its speed and rapidly increasing boost pressure and intake air volume to meet power requirements. This immediate response mode is particularly suitable for scenarios requiring a rapid power boost. "1: Target PID Control" refers to non-transient or steady-state operation, where the control system uses a PID algorithm to fine-tune the blow-off valve opening to maintain optimal boost pressure. PID control dynamically adjusts proportional, integral, and derivative parameters based on the deviation of real-time engine conditions (such as speed and boost pressure) from preset targets, ensuring stable operation of the boost system, avoiding energy waste, and optimizing economy without sacrificing power response. "0: Target Fully Closed" is one of the control system's responses under extreme transient conditions (such as Level 3 transient - emergency acceleration), quickly closing the air bleed valve to a fully closed position to increase power output. "1: Target PID Control" maintains the air bleed valve opening at an optimal level through the PID control algorithm under other operating conditions (including steady-state and non-extreme transient conditions), responsive to transient demands while also considering economic efficiency.

[0091] This embodiment uses sensor signals and time window filtering to determine whether the engine is currently in steady state or transient state. Fuzzy logic control is used to categorize transient operating conditions into levels 1 (mild), 2 (moderate), and 3 (emergency), with corresponding bleed valve slopes. This achieves refined control, avoiding the power lag and energy waste caused by the "one-size-fits-all" transient response of traditional control. Predicting the probability distribution of the next operating condition helps proactively adjust the bleed valve state, resulting in more appropriate actions. Compared to traditional control methods, this embodiment offers the following core advantages: It balances response speed and smoothness: transient grading achieves a progressive response, while predictive control offsets mechanical delays through proactive action. Power is instantly available when overtaking, and there's no jerk when releasing the throttle. Economy is significantly improved: Traditional control methods fully close the bleed valve under low load conditions, resulting in high pumping losses and poor economy. This embodiment fully opens the bleed valve under steady-state low load conditions, achieving better economy. It adapts to complex road conditions: Accurately identifying operating condition predictions adjusts the bleed valve opening to the optimal value for the specific road conditions.

[0092] The embodiment of the present application also provides an adaptive control device for a boosting system of a natural gas engine. It should be noted that the adaptive control device for a boosting system of a natural gas engine in the embodiment of the present application can be used to execute the adaptive control method for a boosting system of a natural gas engine provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementations, and the details that have been explained will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and conceivable.

[0093] The following is an introduction to the adaptive control device for the boost system of a natural gas engine provided in an embodiment of the present application.

[0094] Figure 6 FIG is a block diagram of a natural gas engine boost system adaptive control device according to an embodiment of the present application. Figure 6 As shown, the device includes a first acquiring unit 10 , a determining unit 20 and a first control unit 30 . The first acquisition unit is configured to acquire a speed, intake air flow rate, first information, and a required torque gradient of the natural gas engine, and determine whether the natural gas engine is in a low-load operating condition based on the speed and intake air flow rate and a preset speed and intake air flow map. The first information includes a throttle opening change rate, a speed change rate, a boost pressure, and a load change slope. The required torque gradient is a ratio of a torque change required by the natural gas engine within a preset time window to the preset time window. The determination unit is configured to determine, when the natural gas engine is in the low-load operating condition, whether the current operating condition is a steady-state condition or a transient operating condition based on the first information. The first control unit is configured to control the opening of the purge valve to be fully open if the current operating condition is the steady-state condition, and to utilize fuzzy logic control to classify the transient operating condition into multiple transient levels based on the throttle opening change rate and the required torque gradient if the current operating condition is the transient condition, and to control the opening of the purge valve based on the transient levels.

[0095] This embodiment integrates multiple dimensions of information, including speed, intake air flow, throttle opening rate, speed change rate, boost pressure, and load change slope, and combines them with preset speed and intake air flow MAP to accurately determine whether the natural gas engine is in a low-load phase. It also precisely diagnoses whether the operating state is steady or transient. Under steady-state conditions, the purge valve remains fully open, minimizing pumping losses and significantly improving fuel economy. Under transient conditions, this method uses fuzzy logic control to intelligently classify transient states based on the dynamic changes in throttle opening rate and demand torque gradient, thereby precisely regulating the purge valve opening and ensuring rapid response and smooth transitions in power output under varying transient demand scenarios. This method not only effectively balances the power and fuel economy of natural gas engines under low-load conditions, but also significantly improves engine operating flexibility and overall performance through intelligent operating condition recognition and adaptive control, thereby resolving the difficult balance between fuel economy and power in natural gas engines under low-load conditions.

[0096] In a specific implementation process, the device further includes a second acquisition unit, a training unit, a prediction unit, and an adjustment unit. The second acquisition unit is configured to acquire historical hidden state information and historical observed state information within a first preset time period, wherein the historical hidden state information includes historical steady-state operating conditions and historical transient operating conditions, and the historical observed state information includes historical throttle opening change rate, historical speed change rate, historical boost pressure, and historical load change slope. The training unit is configured to determine the historical hidden state information and the historical observed state information as a training data set, and input the training data set into a hidden Markov model for training until the hidden Markov model reaches a preset convergence condition, thereby obtaining a target hidden Markov model for future state prediction. The prediction unit is configured to input the current observed state into the target hidden Markov model for prediction, thereby obtaining a hidden state sequence within a second preset time period and a probability of change of the hidden state sequence. The adjustment unit is configured to compare the change probability of the hidden state sequence with a preset steady-state probability threshold, and adjust the opening of the purge valve if the change probability exceeds the preset steady-state probability threshold.

[0097] By introducing the hidden Markov model, it is not only possible to respond to the current operating conditions, but also to anticipate the changing trends of future operating conditions, allowing for early responses and avoiding the power hysteresis or energy waste that may be caused by delayed responses in traditional control. By learning historical data through the hidden Markov model, it is possible to more accurately predict the conversion of engine operating conditions, especially the transition from transient to steady state or vice versa under low load, thereby achieving more precise bleed valve control, ensuring that the engine's power and economy can achieve the best balance under various operating conditions, and significantly improving the engine's overall performance and operating efficiency. In short, through the learning and prediction of the hidden Markov model, it is possible not only to judge the engine's operating condition type in real time based on the current observation state, but also to predict the changing trends of operating conditions in the short term in the future, thereby achieving more refined and forward-looking supercharging system control, significantly improving the flexibility, power and economy of natural gas engines under low-load conditions.

[0098] Furthermore, the determination unit includes a monitoring module and a determination module. The monitoring module is used to monitor the throttle opening change rate, the speed change rate, the boost pressure, and the load change slope. The determination module is used to determine that the current operating condition is the transient operating condition when preset conditions are met within a preset transient window length. The preset conditions include the throttle opening change rate being greater than a first threshold, the speed change rate being greater than a second threshold, the boost pressure fluctuation range being greater than a third threshold, and the absolute value of the load change slope being greater than a fourth threshold.

[0099] By real-time monitoring of the throttle opening rate of change, the speed rate of change, boost pressure fluctuations, and the load change slope, combined with the threshold conditions in the preset transient window length, it is possible to quickly identify whether the engine is entering a transient state. This mechanism ensures accurate response at the initial stage of transient conditions. By adjusting the opening strategy of the bleed valve, it effectively balances power and economy, reduces unnecessary pumping losses, and enhances the engine's power response speed and smoothness. More importantly, it can adapt to diverse driving scenarios and driver operating habits, ensuring that the engine can maintain optimal performance under complex and changing operating conditions, thereby improving the stability and reliability of the entire system. At the same time, by monitoring multiple parameters rather than a single indicator, the robustness and accuracy of operating condition judgment are improved, the possibility of misjudgment is reduced, and stable operation is ensured under complex operating conditions.

[0100] In some embodiments of the present application, the above-mentioned device also includes a second control unit for determining whether the above-mentioned natural gas engine is in a low-load operating condition using a preset speed and an intake flow rate MAP. If the above-mentioned natural gas engine is not in the above-mentioned low-load operating condition, the second control unit is used to perform PID control on the above-mentioned purge valve based on a preset boost pressure of the current operating condition of the above-mentioned natural gas engine, wherein the above-mentioned preset boost pressure is obtained by pre-calibration, and for each operating point under non-low-load conditions, there is a unique corresponding preset boost pressure.

[0101] This control strategy ensures more intelligent and personalized adjustment of the blow-off valve under non-low-load conditions. By dynamically adjusting boost pressure, it can meet the high demands for power performance under medium and high-load conditions while also taking into account fuel economy during non-steady-state operation. This sophisticated control logic improves overall engine performance. In short, by introducing dynamic PID control and pre-set boost pressure under non-low-load conditions, intelligent adjustment of the boost system is achieved across the entire engine operating range, thereby comprehensively improving the engine's overall performance.

[0102] In order to further improve the power and economy of the engine under transient operating conditions, the first control unit includes a first fuzzy module, a second fuzzy module, and a classification module. The first fuzzy module is configured to, when the current operating condition is the transient operating condition, employ the fuzzy logic control to fuzzify the throttle opening rate of change into a first fuzzy set, a second fuzzy set, and a third fuzzy set, and to fuzzify the required torque gradient into a first level, a second level, and a third level. The second fuzzy module is configured to establish a rule base based on the first, second, and third fuzzy sets, the first level, the second level, and the third level. The classification module is configured to utilize the rule base to classify the transient operating condition into a plurality of transient levels, including a first-level transient, a second-level transient, and a third-level transient.

[0103] The use of fuzzy logic control enables a more intelligent response to driver input, particularly during transient conditions. This overcomes the control discontinuity and over- or under-response issues inherent in traditional control strategies, often caused by the use of fixed thresholds. By fine-tuning transient levels, the adjustment of the blow-off valve becomes smoother and more efficient, improving not only driving comfort and safety but also engine power and economy under transient conditions.

[0104] In some other embodiments of the present application, the first control unit includes a first control module, a second control module, and a third control module. The first control module is configured to control the opening of the purge valve using a first preset purge valve closing slope if the transient level is a first-level transient; the second control module is configured to control the opening of the purge valve using a second preset purge valve closing slope if the transient level is a second-level transient; and the third control module is configured to control the opening of the purge valve using a third preset purge valve closing slope if the transient level is a third-level transient. The first, second, and third preset purge valve closing slopes are determined during a test bench calibration process.

[0105] By controlling the bleed valve closing slope that matches the transient level, the boost system can provide appropriate power support under transient conditions of varying intensities while minimizing pumping losses and other forms of energy waste. This graded response strategy improves the engine's power performance and fuel economy, especially under frequently changing driving conditions, demonstrating greater flexibility and efficiency, thereby enhancing user satisfaction and driving experience. At the same time, the preset slope based on bench calibration ensures the reliability and consistency of the control strategy. In short, by clarifying the bleed valve closing slope corresponding to different transient levels, more accurate and efficient management of the engine's transient conditions is achieved, which is of great significance to improving the overall performance and operating economy of natural gas engines.

[0106] In some further embodiments of the present application, the preset speed and intake flow MAP include a hysteresis interval. When the changes in the speed and intake flow of the natural gas engine are within the hysteresis interval, it is determined that the judgment of the current operating condition does not change.

[0107] The hysteresis band significantly improves robustness against minor disturbances, reduces unnecessary control actions, and avoids engine instability caused by frequent switching of control modes. Furthermore, this mechanism simplifies control logic, reduces computational burden, and makes control more efficient. It also ensures smooth operation and safety in complex driving environments, enhancing the overall driving experience and engine performance.

[0108] The adaptive control device for a natural gas engine's boosting system includes a processor and memory. The first acquisition unit, determination unit, first control unit, and the like are stored as program units in the memory, and the processor executes the program units stored in the memory to implement corresponding functions. The aforementioned modules are all located in the same processor; alternatively, the aforementioned modules may be located in different processors in any combination.

[0109] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0110] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program. When the program is executed, the device where the computer-readable storage medium is located is controlled to execute the adaptive control method for the boosting system of the natural gas engine.

[0111] An embodiment of the present invention provides an electronic device comprising a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the aforementioned method for adaptively controlling a natural gas engine's boosting system. The device herein may be a server, a PC, a PAD, a mobile phone, or the like.

[0112] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that initializes the steps of the above-mentioned adaptive control method for a boosting system of a natural gas engine.

[0113] Obviously, those skilled in the art will appreciate that the various modules or steps of the present invention described above can be implemented using a general-purpose computing device, can be centralized on a single computing device, or can be distributed across a network of multiple computing devices. They can be implemented using program code executable by the computing device, and thus, can be stored in a storage device and executed by the computing device. In some cases, the steps shown or described herein can be performed in a different order than that shown, or can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0114] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0115] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0116] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0117] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0118] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0119] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0120] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can implement information storage using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0121] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0122] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for adaptively controlling a boost system of a natural gas engine, characterized in that: include: Obtaining a speed, intake air flow, first information, and a required torque gradient of a natural gas engine, and determining whether the natural gas engine is in a low-load operating condition based on the speed and intake air flow using a preset speed and intake air flow MAP, wherein the first information includes a throttle opening change rate, a speed change rate, a boost pressure, and a load change slope, and the required torque gradient is a ratio of a torque change required by the natural gas engine within a preset time window to the preset time window; When the natural gas engine is in the low-load operating condition, determining, based on the first information, whether the current operating condition is a steady-state operating condition or a transient operating condition; When the current operating condition is the steady-state operating condition, the opening of the air release valve is controlled to be in a fully open state. When the current operating condition is the transient operating condition, fuzzy logic control is adopted to divide the transient operating condition into multiple transient levels according to the throttle opening change rate and the required torque gradient, and the opening of the air release valve is controlled according to the transient levels.

2. The method according to claim 1, characterized in that The method further comprises: Acquiring historical hidden state information and historical observed state information within a first preset time period, wherein the historical hidden state information includes historical steady-state operating conditions and historical transient operating conditions, and the historical observed state information includes historical throttle opening change rate, historical speed change rate, historical boost pressure, and historical load change slope; Determining the historical hidden state information and the historical observed state information as a training data set, and inputting the training data set into a hidden Markov model for training until the hidden Markov model reaches a preset convergence condition, thereby obtaining a target hidden Markov model for future state prediction; Inputting the current observation state into the target hidden Markov model for prediction to obtain a hidden state sequence within a second preset time period and a probability of change of the hidden state sequence; The change probability of the change probability of the hidden state sequence is compared with a preset steady-state probability threshold, and when the change probability exceeds the preset steady-state probability threshold, the opening of the purge valve is adjusted.

3. The method according to claim 1, characterized in that When the natural gas engine is in the low-load operating condition, determining whether the current operating condition is a steady-state operating condition or a transient operating condition according to the first information includes: monitoring the throttle opening change rate, the speed change rate, the boost pressure, and the load change slope; When the preset conditions are met within the preset transient window length, the current operating condition is determined to be the transient operating condition, wherein the preset conditions include the throttle opening change rate being greater than a first threshold, the speed change rate being greater than a second threshold, the boost pressure fluctuation range being greater than a third threshold, and the absolute value of the load change slope being greater than a fourth threshold.

4. The method according to claim 1, wherein After determining whether the natural gas engine is in a low-load operating condition by using the preset speed and intake air flow rate MAP, the method further includes: When the natural gas engine is not in the low-load operating condition, the PID control is performed on the purge valve based on the preset boost pressure of the current operating condition of the natural gas engine. The preset boost pressure is obtained by pre-calibration, and for each operating point under non-low-load conditions, there is a unique corresponding preset boost pressure.

5. The method according to claim 1, wherein When the current operating condition is the transient operating condition, fuzzy logic control is used to classify the transient operating condition into multiple transient levels according to the throttle opening change rate and the required torque gradient, including: When the current operating condition is the transient operating condition, the fuzzy logic control is used to fuzzify the throttle opening change rate into a first fuzzy set, a second fuzzy set, and a third fuzzy set, and the required torque gradient is fuzzified into a first level, a second level, and a third level; Setting a rule base based on the first fuzzy set, the second fuzzy set, the third fuzzy set, the first level, the second level, and the third level; The transient operating conditions are divided into a plurality of transient levels using the rule base, and the transient levels include primary transient, secondary transient and tertiary transient.

6. The method according to claim 1, characterized in that Controlling the opening of the air release valve according to the transient level includes: If the transient level is a level 1 transient, a first preset purge valve closing slope is used to control the opening of the purge valve; If the transient level is a level 2 transient, the opening of the purge valve is controlled by using a second preset purge valve closing slope; If the transient level is level 3, the opening of the purge valve is controlled by using the third preset purge valve closing slope. The first preset purge valve closing slope, the second preset purge valve closing slope, and the third preset purge valve closing slope are determined during a test bench calibration process.

7. The method according to claim 1, characterized in that The preset speed and intake air flow rate MAP include a hysteresis range. When the changes in the speed and intake air flow rate of the natural gas engine are within the hysteresis range, it is determined that the current operating condition is unchanged.

8. An adaptive control device for a natural gas engine supercharging system, characterized in that: include: a first acquisition unit, configured to acquire a speed, an intake air flow rate, first information, and a required torque gradient of the natural gas engine, and determine whether the natural gas engine is in a low-load operating condition based on the speed and the intake air flow rate and using a preset speed and intake air flow MAP, wherein the first information includes a throttle opening change rate, a speed change rate, a boost pressure, and a load change slope, and the required torque gradient is a ratio of a torque change required by the natural gas engine within a preset time window to the preset time window; a determining unit, configured to determine, when the natural gas engine is in the low-load operating condition, whether the current operating condition is a steady-state operating condition or a transient operating condition according to the first information; The first control unit is used to control the opening of the air release valve to be in a fully open state when the current operating condition is the steady-state operating condition, and to adopt fuzzy logic control when the current operating condition is the transient operating condition, to divide the transient operating condition into multiple transient levels according to the throttle opening change rate and the required torque gradient, and to control the opening of the air release valve according to the transient levels.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the adaptive control method for the boosting system of a natural gas engine according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the adaptive control method of the boosting system of the natural gas engine according to any one of claims 1 to 7.

Images