Engine combustion noise control method and device, electronic equipment and target engine

Abstract

The invention relates to the technical field of engines, in particular to an engine combustion noise control method and device, electronic equipment and a target engine. Attribute information, current working condition information and current working environment information corresponding to the target engine are obtained; based on the attribute information, the current working condition information and the current working environment information, a target oil injection strategy corresponding to the target engine is determined; the target oil injection strategy comprises a target pre-injection strategy and a target main injection strategy or only comprises the target main injection strategy; the target pilot injection strategy comprises a pilot injection oil quantity and a pilot injection advance angle; the target main injection strategy comprises a main injection fuel quantity, a main injection advance angle and main injection duration; and based on the target oil injection strategy, the oil injection mode of the target engine is controlled, so that combustion noise corresponding to the target engine is reduced. The accuracy of the determined target oil injection strategy corresponding to the target engine is ensured, and the combustion noise corresponding to the target engine is reduced. And a combustion noise control technology which is more efficient, reliable and environment-friendly is provided.

Description

Technical Field

[0001] The present invention relates to the field of engine technology, and in particular to an engine combustion noise control method, device, electronic equipment and a target engine. Background Art

[0002] In the current booming modern automobile industry, engine performance optimization has always been a core research topic, and the effective control of combustion noise is a key link. Combustion noise not only reduces the comfort of driving experience, but also may indirectly reflect the abnormal working conditions of the engine, posing a potential threat to the reliability and durability of the engine.

[0003] For a long time, in order to overcome the problem of combustion noise, existing technologies have mainly focused on the operating conditions of the engine, focusing on the precise control of the throttle opening or the ignition advance angle. In terms of throttle opening control, engineers use the filter coefficient to fine-tune the control signal, hoping to accurately change the degree of throttle opening, thereby optimizing the engine's intake volume to achieve effective intervention in the combustion process. In terms of ignition advance angle control, the ignition timing is usually corrected by increasing the ignition advance angle offset, striving to achieve the best combustion process under different working conditions.

[0004] However, this traditional technical path has exposed many unavoidable defects in real applications. The working environment of the engine is extremely complex, and various electronic signals are intertwined, making the filter signal extremely susceptible to interference from other signals. Once this interference occurs, it will seriously distort the control instructions for the throttle opening, causing deviations in throttle control. When the throttle cannot accurately adjust the intake volume as expected, the combustion process of the engine will fall into disorder, causing abnormal combustion. This will not only cause the fuel to fail to burn fully, causing a significant increase in fuel consumption, greatly reducing the fuel economy of the engine and increasing the user's cost of use; at the same time, the incompletely burned fuel will also produce a large amount of harmful emissions, such as carbon monoxide, hydrocarbons and nitrogen oxides, which seriously exceed environmental protection standards and cause more serious pollution to the atmospheric environment.

[0005] With increasingly stringent environmental regulations and increasing user requirements for engine performance, the traditional technical means of improving combustion noise based on throttle opening and ignition advance angle control can no longer meet the current development needs. Therefore, exploring a more efficient, reliable and environmentally friendly combustion noise control technology has become an important issue that needs to be solved in the automotive industry. Summary of the invention

[0006] In view of this, the present invention provides an engine combustion noise control method, device, electronic equipment and target engine to solve the problem of providing a more efficient, reliable and environmentally friendly combustion noise control technology.

[0007] In a first aspect, the present invention provides a method for controlling engine combustion noise, the method comprising:

[0008] Obtaining the property information, current working condition information and current working environment information corresponding to the target engine;

[0009] Based on the attribute information, the current working condition information and the current working environment information, a target injection strategy corresponding to the target engine is determined; the target injection strategy includes a target pre-injection strategy and a target main injection strategy or only includes a target main injection strategy; the target pre-injection strategy includes a pre-injection amount and a pre-injection advance angle; the target main injection strategy includes a main injection amount, a main injection advance angle and a main injection duration;

[0010] Based on the target injection strategy, the injection mode of the target engine is controlled to reduce the combustion noise corresponding to the target engine.

[0011] The engine combustion noise control method provided by the embodiment of the present application obtains the property information, current working condition information and current working environment information corresponding to the target engine. By obtaining the property information, current working condition information and current working environment information of the target engine to determine the target injection strategy, the individual differences and real-time operating status of the engine can be fully considered. Based on the property information, current working condition information and current working environment information, the target injection strategy corresponding to the target engine is determined, and the accuracy of the target injection strategy corresponding to the determined target engine is ensured. Based on the target injection strategy, the injection mode of the target engine is controlled to reduce the combustion noise corresponding to the target engine. The target pre-injection strategy included in the target injection strategy has a significant effect on reducing combustion noise through precise control of the pre-injection amount and pre-injection advance angle. The pre-injection injection strategy is released in the low-load area, and the pre-injected fuel can be mixed with air in advance to form a more uniform mixture, thereby improving the combustion conditions. A reasonable pre-injection advance angle can make the combustion closer to the ideal state, reduce the pressure rise rate at the initial stage of combustion, and thus reduce the combustion noise. At the same time, the precise implementation of the pre-injection strategy can achieve noise reduction without worsening fuel consumption, thereby improving the overall performance of the engine. The target main injection strategy included in the target injection strategy accurately controls the main injection amount and main injection advance angle, and works in conjunction with the pre-injection strategy (if pre-injection exists), or plays a key role independently when only the main injection strategy is used. The precise main injection amount ensures the provision of the appropriate amount of fuel under different operating conditions, ensuring full and stable combustion, and avoiding abnormal combustion noise caused by too much or too little fuel. A reasonable main injection advance angle can enable the main injection fuel to participate in combustion at the best time, and cooperate with the engine's piston movement, valve opening and closing, etc., to further optimize the combustion process and reduce combustion noise. For example, when the engine is running at high speed and high load, the appropriate main injection strategy can ensure power output while effectively controlling the combustion noise at a low level. It provides a more efficient, reliable and environmentally friendly combustion noise control technology.

[0012] In an alternative embodiment, based on the attribute information, the current operating condition information, and the current working environment information, determining a target injection strategy for a target engine, includes:

[0013] Inputting the attribute information, the current operating condition information, and the current working environment information into a preset injection strategy determination model;

[0014] The preset injection strategy determination model extracts features from the attribute information, the current operating condition information, and the current working environment information, and based on the extracted features, outputs a target injection strategy for the target engine.

[0015] In the engine combustion noise control method provided by the embodiments of the present application, the attribute information, the current operating condition information, and the current working environment information are input into a preset injection strategy determination model. The preset injection strategy determination model extracts features from the attribute information, the current operating condition information, and the current working environment information, and based on the extracted features, outputs a target injection strategy for the target engine. The preset injection strategy determination model can quickly process the input target engine attribute information, current operating condition information, and current working environment information. Compared with manual analysis or traditional simple rule judgment, through algorithms and data training, the model can accurately extract key features in a short time and generate a suitable target injection strategy based on these features, ensuring the accuracy of the generated target injection strategy. The preset injection strategy determination model with sufficient training has strong adaptability and generalization ability. It can process the attribute information of various different types of engines, as well as the operating conditions and working environment information within a wide range. Whether it is engines of different brands and models, or in various extreme working environments (such as high-temperature and high-cold regions) and complex operating conditions (such as frequent start-stop, rapid acceleration and deceleration), the model can generate a suitable injection strategy according to the input information. This makes the solution have a wider application scenario and can adapt to diverse user needs and actual usage situations. In addition, the above method reduces the influence of human factors on the determination of the injection strategy and reduces the error caused by misjudgment of human factors.

[0016] In an alternative embodiment, the training process of the preset injection strategy determination model includes:

[0017] Obtaining training attribute information, training operating condition information, and training working environment information corresponding to a plurality of training engines;

[0018] Based on the training attribute information, the training operating condition information, and the training working environment information, determining a training injection strategy corresponding to the training engine;

[0019] Generate a training dataset based on the training attribute information, training condition information, training working environment information, and training fuel injection strategy corresponding to each training engine;

[0020] Train an initial fuel injection strategy determination network based on the training dataset to obtain a preset fuel injection strategy determination model.

[0021] The engine combustion noise control method provided by the embodiments of the present application obtains the training attribute information, training condition information, and training working environment information corresponding to multiple training engines; determines the training fuel injection strategy corresponding to the training engine based on the training attribute information, training condition information, and training working environment information; and generates a training dataset based on the training attribute information, training condition information, training working environment information, and training fuel injection strategy corresponding to each training engine. Thus, the training of the preset fuel injection strategy determination model is based on a large amount of real and rich data. These data cover the situations of different engines under various working conditions and environments. The preset fuel injection strategy determination model can learn the mapping relationship between different conditions and the optimal fuel injection strategy. Therefore, in practical applications, for the specific situation of the target engine, a more accurate fuel injection strategy can be output, effectively reducing combustion noise and improving engine performance. Train an initial fuel injection strategy determination network based on the training dataset to obtain a preset fuel injection strategy determination model. During the training process, the model can adjust its own parameters such as weights and biases according to the feedback of the training data to minimize the error between the predicted fuel injection strategy and the actual training fuel injection strategy. Through multiple iterative trainings, the performance of the model is gradually improved, and the processing ability for different input information and the output accuracy of the fuel injection strategy will be improved, thus more effectively reducing engine combustion noise and improving the comprehensive performance of the engine.

[0022] In an optional implementation manner, determining the training fuel injection strategy corresponding to the training engine based on the training attribute information, training condition information, and training working environment information includes:

[0023] Obtain an initial pre-injection strategy and an initial main injection strategy;

[0024] For each training engine, obtain the first fuel consumption trend line and the first noise trend line generated after fuel injection based on the initial pre-injection strategy and the initial main injection strategy under the training condition information and the training working environment information;

[0025] Obtain the second fuel consumption trend line and the second noise trend line generated after fuel injection based on the initial main injection strategy under the training condition information and the training working environment information;

[0026] Compare the first fuel consumption trend line corresponding to the training engine with the second fuel consumption trend line, and compare the first noise trend line with the second noise trend line;

[0027] Based on the comparison results, determine the first training engine suitable for initiating the initial pre-injection strategy and the second training engine not suitable for initiating the initial pre-injection strategy;

[0028] For the first training engine, determine the first training fuel injection strategy corresponding to the training condition information and the training working environment information of the first training engine; the first training fuel injection strategy includes a training pre-injection strategy and a training main injection strategy;

[0029] For the second training engine, determine the second training fuel injection strategy corresponding to the training condition information and the training working environment information of the second training engine; the second training fuel injection strategy only includes a training main injection strategy.

[0030] The engine combustion noise control method provided by the embodiments of the present application obtains an initial pre-injection strategy and an initial main injection strategy. For each training engine, obtain the first fuel consumption trend line and the first noise trend line generated after fuel injection based on the initial pre-injection strategy and the initial main injection strategy under the training condition information and the training working environment information of the training engine; obtain the second fuel consumption trend line and the second noise trend line generated after fuel injection based on the initial main injection strategy under the training condition information and the training working environment information of the training engine; compare the first fuel consumption trend line corresponding to the training engine with the second fuel consumption trend line, and compare the first noise trend line with the second noise trend line; according to the comparison results, determine the first training engine suitable for starting the initial pre-injection strategy and the second training engine not suitable for starting the initial pre-injection strategy. Thus, it can accurately judge which training engines are suitable for starting the pre-injection strategy and which are not. This comparative analysis based on actual operation data can fully consider the performance differences of different training engines under specific working conditions and environments, avoiding the adverse effects that may be brought about by blindly adopting the pre-injection strategy, such as the problem of deteriorated fuel consumption on some engines, and improving the pertinence and accuracy of fuel injection strategy formulation. For the first training engine, determine the first training fuel injection strategy corresponding to the first training engine under the training condition information and the training working environment information; the first training fuel injection strategy includes a training pre-injection strategy and a training main injection strategy. For the first training engine suitable for starting the pre-injection strategy, determining the first training fuel injection strategy including the training pre-injection strategy and the training main injection strategy can effectively reduce combustion noise without deteriorating fuel consumption and improve the comprehensive performance of the engine. The pre-injection strategy can improve the combustion process, make the combustion more stable, reduce the pressure fluctuation in the initial stage of combustion, and thus reduce noise. For the second training engine, determine the second training fuel injection strategy corresponding to the second training engine under the training condition information and the training working environment information; the second training fuel injection strategy only includes a training main injection strategy. For the second training engine not suitable for starting the pre-injection strategy, only the training main injection strategy is adopted, avoiding the negative impact brought about by inappropriate pre-injection, ensuring that the engine operates in the best state under this working condition and environment, and maintaining good performance.

[0031] In an alternative embodiment, determining the first training fuel injection strategy corresponding to the first training engine under the training condition information and the training working environment information includes:

[0032] Based on the training attribute information corresponding to the first training engine, generate a variety of test pre-injection strategies and test main injection strategies;

[0033] Based on each test pre-injection strategy and each test main injection strategy, generate multiple groups of test fuel injection strategies; each group of test fuel injection strategies includes a test pre-injection strategy and a test main injection strategy;

[0034] Obtain the engine combustion noise data and engine-related data generated after fuel injection based on each group of test injection strategies under the training condition information and training working environment information of the first training engine. The engine-related data includes at least one of engine power, engine torque, engine fuel consumption rate, and engine emissions.

[0035] Based on each group of test injection strategies and their corresponding engine combustion noise data and engine-related data, determine the first training injection strategy of the first training engine under the training condition information and training working environment information.

[0036] The engine combustion noise control method provided by the embodiments of the present application generates a variety of test pre-injection strategies and test main injection strategies based on the training attribute information corresponding to the first training engine, so as to comprehensively explore the possible injection strategy combination space. Then, based on each test pre-injection strategy and each test main injection strategy, multiple groups of test injection strategies are generated; obtain the engine combustion noise data and engine-related data generated after fuel injection based on each group of test injection strategies under the training condition information and training working environment information of the first training engine. This multi-dimensional data collection and evaluation method can more comprehensively understand the comprehensive impact of different injection strategies on engine performance. For example, not only can we focus on reducing combustion noise, but also consider fuel economy (fuel consumption rate), power output (engine power and torque), and environmental protection performance (emissions) at the same time, which helps to determine an injection strategy that can achieve a better balance in all aspects and improve the overall performance of the engine. Based on each group of test injection strategies and their corresponding engine combustion noise data and engine-related data, determine the first training injection strategy of the first training engine under the training condition information and training working environment information. By analyzing and comparing these data, we can accurately find the injection strategy parameter combination that can make the engine achieve the best performance under the given training condition information and training working environment information. Compared with relying solely on empirical values or simple parameter adjustments, this method based on a large amount of test data can more accurately determine the optimal injection strategy, thereby effectively reducing combustion noise and meeting the requirements of other performance indicators at the same time.

[0037] In an alternative embodiment, based on each group of test injection strategies and their corresponding engine combustion noise data and engine-related data, determining the first training injection strategy of the first training engine under the training condition information and training working environment information includes:

[0038] Construct an objective function based on the engine combustion noise data and engine-related data corresponding to each group of test injection strategies; the objective function is generated based on at least one of the engine combustion noise, fuel consumption rate, nitrogen oxide emissions, and particulate matter emissions corresponding to the first training engine;

[0039] Construct a target constraint condition based on the training attribute information corresponding to the first training engine; the target constraint condition includes an injection strategy constraint and an engine operating condition constraint;

[0040] Construct a target optimization model based on the objective function and the target constraint condition;

[0041] Solve the target optimization model to determine the first training injection strategy corresponding to the first training engine under the training condition information and the training working environment information.

[0042] The engine combustion noise control method provided by the embodiments of the present application constructs an objective function based on the engine combustion noise data and the engine-related data corresponding to each group of test injection strategies. Thus, the goal of engine performance optimization can be scientifically quantified. Instead of vaguely pursuing the improvement of engine performance, the objective function is clearly constructed with specific indicators such as combustion noise, fuel consumption rate, and emissions, making the optimization direction clearer and more specific. For example, if the weight of combustion noise in the objective function is relatively large, then during the optimization process, more emphasis will be placed on reducing combustion noise while taking into account other indicators, thereby achieving targeted engine performance optimization. Construct a target constraint condition based on the training attribute information corresponding to the first training engine; the target constraint condition includes an injection strategy constraint and an engine operating condition constraint. All kinds of limiting factors in the actual operation of the engine are comprehensively considered. The injection strategy constraint ensures that the injection parameters are within the range allowed by the engine hardware and technology, avoiding unreasonable injection settings; the engine operating condition constraint ensures that the engine operates under safe and stable working conditions. This comprehensive setting of constraint conditions makes the optimization result more practically feasible and avoids the situation that is theoretically feasible but actually cannot be implemented. Construct a target optimization model based on the objective function and the target constraint condition. The target optimization model can accurately describe the relationship between the injection strategy and the engine performance, as well as the limitations of various constraint conditions on the injection strategy. Solve the target optimization model to determine the first training injection strategy corresponding to the first training engine under the training condition information and the training working environment information. This first training injection strategy is determined after comprehensively considering various performance indicators and actual operation limitations of the engine. Therefore, this first training injection strategy can effectively improve the comprehensive performance of the engine, not only reducing combustion noise, but also achieving a better balance in terms of fuel economy, power output, emissions, etc. For example, while reducing combustion noise, ensuring that the fuel consumption rate of the engine does not increase significantly, the emissions meet the environmental protection standards, and improving the overall competitiveness of the engine.

[0043] In an alternative embodiment, solving the target optimization model to determine the first training injection strategy corresponding to the first training engine under the training condition information and the training working environment information includes:

[0044] Detect whether the injection strategies of each group of tests meet the target constraint conditions;

[0045] If the test injection strategy meets the target constraint conditions, calculate the objective function value corresponding to the test injection strategy based on the objective function;

[0046] Update each test injection strategy according to the objective function value corresponding to each test injection strategy to generate a new injection strategy;

[0047] Detect whether each new injection strategy meets the target constraint conditions, and calculate the objective function value corresponding to the new injection strategy based on the objective function;

[0048] Until the new injection strategy meets the target constraint conditions and the objective function value corresponding to the new injection strategy is optimal, determine the new injection strategy as the first training injection strategy corresponding to the first training engine under the training condition information and the training working environment information.

[0049] The engine combustion noise control method provided by the embodiments of the present application detects whether each group of test injection strategies meets the target constraint conditions, and can timely exclude those injection strategies that do not meet the actual operation limits of the engine during the optimization process. If the test injection strategy meets the target constraint conditions, then based on the objective function, calculate the objective function value corresponding to the test injection strategy, so as to accurately quantify the comprehensive impact of each test injection strategy on the engine performance. According to the objective function values corresponding to each test injection strategy, update each test injection strategy to generate a new injection strategy; detect whether each new injection strategy meets the target constraint conditions, and based on the objective function, calculate the objective function value corresponding to the new injection strategy. By continuously adjusting the injection strategy parameters and re-evaluating whether they meet the constraint conditions and calculating the objective function values, the optimal solution can be gradually approached. This iterative optimization method can fully explore better injection strategies and continuously improve the performance of the engine, such as further reducing combustion noise and improving fuel economy. Until the new injection strategy meets the target constraint conditions and the objective function value corresponding to the new injection strategy is optimal, then determine the new injection strategy as the first training injection strategy of the first training engine corresponding to the training condition information and the training working environment information. The finally determined first training injection strategy is obtained under the condition of meeting the target constraint conditions and having the optimal objective function value. This means that this injection strategy can not only theoretically make the comprehensive performance of the engine reach the best, but also ensure the stable and reliable operation of the engine in actual operation. A suitable injection strategy can make the combustion process of the engine more stable, reduce the occurrence of abnormal combustion phenomena, thereby extending the service life of the engine and reducing the maintenance cost. In addition, this optimization process takes into account various factors and constraint conditions, and can generate injection strategies adapted to different training condition information and training working environment information. Whether the engine operates at different speeds and loads, or under different environmental temperatures, pressures and other conditions, the injection strategies obtained through this optimization process can better adapt to these complex situations and ensure that the engine can maintain good performance in various working conditions and environments.

[0050] In a second aspect, the present invention provides an engine combustion noise control device, which includes:

[0051] An acquisition module, configured to acquire the attribute information, the current condition information, and the current working environment information corresponding to the target engine;

[0052] A determination module, configured to determine the target injection strategy corresponding to the target engine based on the attribute information, the current condition information, and the current working environment information; the target injection strategy includes a target pre-injection strategy and a target main injection strategy or only includes a target main injection strategy; the target pre-injection strategy includes a pre-injection fuel quantity and a pre-injection advance angle; the target main injection strategy includes a main injection fuel quantity, a main injection advance angle, and a main injection duration;

[0053] The control module is used to control the fuel injection mode of the target engine based on the target fuel injection strategy to reduce the combustion noise corresponding to the target engine.

[0054] The engine combustion noise control device provided in the embodiment of the present application obtains the attribute information, current working condition information and current working environment information corresponding to the target engine. By obtaining the attribute information, current working condition information and current working environment information of the target engine to determine the target injection strategy, the individual differences and real-time operating status of the engine can be fully considered. Based on the attribute information, current working condition information and current working environment information, the target injection strategy corresponding to the target engine is determined, ensuring the accuracy of the target injection strategy corresponding to the determined target engine. Based on the target injection strategy, the injection mode of the target engine is controlled to reduce the combustion noise corresponding to the target engine. The target pre-injection strategy included in the target injection strategy has a significant effect on reducing combustion noise through precise control of the pre-injection amount and pre-injection advance angle. The pre-injection injection strategy is released in the low-load area, and the pre-injected fuel can be mixed with air in advance to form a more uniform mixture, thereby improving the combustion conditions. A reasonable pre-injection advance angle can make the combustion closer to the ideal state, reduce the pressure rise rate at the initial stage of combustion, and thus reduce the combustion noise. At the same time, the precise implementation of the pre-injection strategy can achieve noise reduction without worsening fuel consumption, thereby improving the overall performance of the engine. The target main injection strategy included in the target injection strategy accurately determines the main injection amount and main injection advance angle, and works in conjunction with the pre-injection strategy (if pre-injection exists), or plays a key role independently when only the main injection strategy is used. The precise main injection amount ensures the provision of the appropriate amount of fuel under different operating conditions, ensuring full and stable combustion, and avoiding abnormal combustion noise caused by too much or too little fuel. A reasonable main injection advance angle can enable the main injection fuel to participate in combustion at the best time, and cooperate with the engine's piston movement, valve opening and closing, etc., to further optimize the combustion process and reduce combustion noise. For example, when the engine is running at high speed and high load, the appropriate main injection strategy can ensure power output while effectively controlling the combustion noise at a low level. It provides a more efficient, reliable and environmentally friendly combustion noise control technology.

[0055] In a third aspect, the present invention provides an electronic device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the engine combustion noise control method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0056] In a fourth aspect, the present invention provides a target engine, an electronic device of the target engine and an engine body, wherein the electronic device is used to execute the engine combustion noise control method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0058] Figure 1 is a schematic flowchart of an engine combustion noise control method according to an embodiment of the present invention;

[0059] Figure 2 is a schematic flowchart of another engine combustion noise control method according to an embodiment of the present invention;

[0060] Figure 3 is a structural block diagram of an engine combustion noise control device according to an embodiment of the present invention;

[0061] Figure 4 is a schematic hardware structure diagram of an electronic device according to an embodiment of the present invention. Specific Embodiments

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0063] In the current booming development of the modern automotive industry, the performance optimization of engines has always been a core research topic, and the effective control of combustion noise is a key link among them. Combustion noise not only reduces the comfort of the driving and riding experience but may also indirectly reflect abnormal conditions inside the engine, posing a potential threat to the reliability and durability of the engine.

[0064] For a long time, to overcome the problem of combustion noise, the prior art has mainly focused on the operating conditions of the engine and emphasized precise regulation of the throttle opening or ignition advance angle. In terms of throttle opening control, engineers fine-tune the control signal with a filtering coefficient, hoping to precisely change the opening degree of the throttle, thereby optimizing the intake air volume of the engine to effectively intervene in the combustion process. In the control of the ignition advance angle, usually, the ignition timing is corrected by increasing the ignition advance angle offset, striving to achieve the best combustion state under different operating conditions.

[0065] However, this traditional technical path has exposed many unavoidable defects in real applications. The working environment of the engine is extremely complex, and various electronic signals are intertwined, making the filter signal extremely susceptible to interference from other signals. Once this interference occurs, it will seriously distort the control instructions for the throttle opening, causing deviations in throttle control. When the throttle cannot accurately adjust the intake volume as expected, the combustion process of the engine will fall into disorder, causing abnormal combustion. This will not only cause the fuel to fail to burn fully, causing a significant increase in fuel consumption, greatly reducing the fuel economy of the engine and increasing the user's cost of use; at the same time, the incompletely burned fuel will also produce a large amount of harmful emissions, such as carbon monoxide, hydrocarbons and nitrogen oxides, which seriously exceed environmental protection standards and cause more serious pollution to the atmospheric environment.

[0066] With increasingly stringent environmental regulations and increasing user requirements for engine performance, the traditional technical means of improving combustion noise based on throttle opening and ignition advance angle control can no longer meet the current development needs. Therefore, exploring a more efficient, reliable and environmentally friendly combustion noise control technology has become an important issue that needs to be solved in the automotive industry.

[0067] Based on this, an embodiment of the present application provides a method for controlling engine combustion noise. It should be noted that the execution subject of the method for controlling engine combustion noise provided in the embodiment of the present application can be a device for controlling engine combustion noise, and the device for controlling engine combustion noise can be implemented as part or all of an electronic device through software, hardware, or a combination of software and hardware, wherein the electronic device can be a control device in a target vehicle or a control device in an engine. The embodiment of the present application does not specifically limit the electronic device. In the following method embodiments, the execution subject is an electronic device for example.

[0068] According to an embodiment of the present invention, an embodiment of an engine combustion noise control method 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.

[0069] In this embodiment, a method for controlling engine combustion noise is provided, which can be used in the above-mentioned electronic equipment. Figure 1 : is a flow chart of an engine combustion noise control method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0070] Step S101, obtaining attribute information, current operating condition information and current working environment information corresponding to the target engine.

[0071] Among them, the attribute information of the target engine may include basic engine structure parameters, key component parameters, fuel injection system parameters, and other attribute information. For a spark-ignition engine, ignition system parameters may also be included.

[0072] Among them, the basic structure parameters may include the type of the engine (such as gasoline engine, diesel engine, hybrid engine, etc.), the number of cylinders, the cylinder arrangement form (in-line, V-type, horizontally opposed, etc.), and the number of strokes (four-stroke, two-stroke). The key component parameters may include the piston diameter, piston stroke, and compression ratio. The fuel injection system parameters may include the type of fuel injector (electromagnetic fuel injector, piezoelectric fuel injector, etc.), the number of fuel injectors, and the injection characteristics of the fuel injector (injection angle, injection rate, etc.). The other attribute information may include the manufacturer, model year, and design purpose of the engine (such as passenger car engine, commercial vehicle engine, construction machinery engine, etc.). The ignition system parameters may include the type of spark plug and the ignition advance angle range.

[0073] Among them, the current operating condition information may include the current engine speed and the current engine load. The current working environment information includes the current ambient temperature, the current ambient pressure, and the current ambient humidity.

[0074] The electronic device can receive the attribute information, current operating condition information, and current working environment information corresponding to the target engine input by the user, and can also receive the attribute information, current operating condition information, and current working environment information corresponding to the target engine sent by other devices. The electronic device can also search for the attribute information corresponding to the target engine in the storage space, and obtain the current engine speed and the current engine load of the target engine based on the sensor. Then, the electronic device detects the current ambient temperature based on the temperature sensor, detects the current ambient pressure based on the pressure sensor, and detects the current ambient humidity based on the humidity sensor.

[0075] The embodiments of the present application do not specifically limit the manner in which the electronic device obtains the attribute information, current operating condition information, and current working environment information corresponding to the target engine.

[0076] Step S102, based on the attribute information, current operating condition information, and current working environment information, determine the target fuel injection strategy corresponding to the target engine.

[0077] Among them, the target fuel injection strategy includes a target pre-injection strategy and a target main injection strategy or only includes a target main injection strategy; the target pre-injection strategy includes the pre-injection fuel quantity and the pre-injection advance angle; the target main injection strategy includes the main injection fuel quantity, the main injection advance angle, and the main injection duration.

[0078] Specifically, the electronic device can determine the target injection strategy corresponding to the target engine by looking up a table based on the corresponding relationships between the attribute information, the current operating condition information, and the current working environment information and the injection strategies respectively.

[0079] This step will be introduced in detail below.

[0080] Step S103: Control the injection mode of the target engine based on the target injection strategy to reduce the combustion noise corresponding to the target engine.

[0081] Specifically, the electronic device identifies the target injection strategy to determine whether the target injection strategy includes a target pre-injection strategy. If the target injection strategy includes the target pre-injection strategy, the electronic device identifies the target pre-injection strategy to determine the pre-injection fuel quantity and the pre-injection advance angle in the target pre-injection strategy. Then, fuel is injected into the cylinder of the target engine according to the pre-injection advance angle and the pre-injection fuel quantity. The pre-injected fuel mixes with air in the cylinder in advance and starts to burn, making the rising rates of the in-cylinder temperature and pressure during the combustion of the main-injected fuel relatively gentle, thereby reducing the combustion noise. Then, the electronic device identifies the target main-injection strategy to determine the main-injection fuel quantity and the main-injection advance angle. Then, fuel is injected into the cylinder of the target engine according to the main-injection fuel quantity and the main-injection advance angle.

[0082] If the target injection strategy does not include the target pre-injection strategy, the electronic device identifies the target main-injection strategy to determine the main-injection fuel quantity and the main-injection advance angle. Then, fuel is injected into the cylinder of the target engine according to the main-injection fuel quantity and the main-injection advance angle.

[0083] The engine combustion noise control method provided by the embodiment of the present application obtains the property information, current working condition information and current working environment information corresponding to the target engine. By obtaining the property information, current working condition information and current working environment information of the target engine to determine the target injection strategy, the individual differences and real-time operating status of the engine can be fully considered. Based on the property information, current working condition information and current working environment information, the target injection strategy corresponding to the target engine is determined, and the accuracy of the target injection strategy corresponding to the determined target engine is ensured. Based on the target injection strategy, the injection mode of the target engine is controlled to reduce the combustion noise corresponding to the target engine. The target pre-injection strategy included in the target injection strategy has a significant effect on reducing combustion noise through precise control of the pre-injection amount and pre-injection advance angle. The pre-injection injection strategy is released in the low-load area, and the pre-injected fuel can be mixed with air in advance to form a more uniform mixture, thereby improving the combustion conditions. A reasonable pre-injection advance angle can make the combustion closer to the ideal state, reduce the pressure rise rate at the initial stage of combustion, and thus reduce the combustion noise. At the same time, the precise implementation of the pre-injection strategy can achieve noise reduction without worsening fuel consumption, thereby improving the overall performance of the engine. The target main injection strategy included in the target injection strategy accurately determines the main injection amount and main injection advance angle, and works in conjunction with the pre-injection strategy (if pre-injection exists), or plays a key role independently when only the main injection strategy is used. The precise main injection amount ensures the provision of the appropriate amount of fuel under different operating conditions, ensuring full and stable combustion, and avoiding abnormal combustion noise caused by too much or too little fuel. A reasonable main injection advance angle can enable the main injection fuel to participate in combustion at the best time, and cooperate with the engine's piston movement, valve opening and closing, etc., to further optimize the combustion process and reduce combustion noise. For example, when the engine is running at high speed and high load, the appropriate main injection strategy can ensure power output while effectively controlling the combustion noise at a low level. It provides a more efficient, reliable and environmentally friendly combustion noise control technology.

[0084] In this embodiment, a method for controlling engine combustion noise is provided, which can be used in the above-mentioned electronic equipment. Figure 2 : is a flow chart of an engine combustion noise control method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0085] Step S201, obtaining attribute information, current operating condition information and current working environment information corresponding to the target engine.

[0086] For details about this step, please refer to the above description of step S101, which will not be elaborated here.

[0087] Step S202, determining a target fuel injection strategy corresponding to the target engine based on the attribute information, the current operating condition information and the current working environment information.

[0088] Among them, the target fuel injection strategy includes a target pre-injection strategy and a target main injection strategy or only includes the target main injection strategy; the target pre-injection strategy includes the pre-injection fuel quantity and the pre-injection advance angle; the target main injection strategy includes the main injection fuel quantity, the main injection advance angle, and the main injection duration.

[0089] Specifically, the above step S202 may include the following steps:

[0090] Step S2021: Input the attribute information, the current operating condition information, and the current working environment information into a preset fuel injection strategy determination model.

[0091] Specifically, the electronic device inputs the attribute information, the current operating condition information, and the current working environment information into a preset fuel injection strategy determination model.

[0092] Step S2022: The preset fuel injection strategy determination model extracts features from the attribute information, the current operating condition information, and the current working environment information, and based on the extracted features, outputs the target fuel injection strategy corresponding to the target engine.

[0093] Specifically, the preset fuel injection strategy determination model can convert the attribute information of the target engine into a one-hot encoding or a specific numerical vector form. For example, for the cylinder arrangement form, in-line, V-type, and horizontally opposed respectively correspond to different encoding values, so that the preset fuel injection strategy determination model can effectively identify and process them. The preset fuel injection strategy determination model can use the sliding window technique to process the current speed and current load data of the target engine within a continuous time period. For example, calculate statistical features such as the mean, variance, and change rate of the current speed and current load within the window to capture the dynamic change trend of the target engine operating condition. In addition, the preset fuel injection strategy determination model can also adopt a long short-term memory network (LSTM) to process the operating condition data in the form of time series, which can effectively remember the changes in the operating condition over a period of time in the past, so as to better predict the fuel injection strategy required at the current moment.

[0094] Then, the preset fuel injection strategy determination model can use a convolutional neural network (CNN) to extract features from the current working environment information. The convolutional layer of the CNN can learn the local correlation features between the current working environment information, such as the influence features of the co-variation of the environmental temperature and pressure on the intake air density.

[0095] After feature extraction, the preset injection strategy determination model inputs these fused features into the decision-making module. The decision-making module can be a multi-layer perceptron (MLP) or a model based on the Transformer architecture. The MLP performs non-linear transformation and combination on the input features through multiple hidden layers, and finally obtains the parameter values of the target injection strategy at the output layer. The model based on the Transformer architecture, on the other hand, can better capture the global dependencies between different features and improve the accuracy of decision-making.

[0096] During the decision-making process of the model, the preset injection strategy determination model will also refer to the injection strategies and the corresponding engine performance feedback data under similar working conditions and environments in history. Through analogical reasoning, the current decision is corrected and optimized. For example, if the current working condition and environmental characteristics are similar to those of a successful case in history, the model can preferentially refer to the injection strategy of this case and make fine-tuning according to the current minor differences.

[0097] Finally, the preset injection strategy determination model determines whether the target engine is suitable for starting the pilot injection strategy under the current working condition information and the current working environment information. If the target engine is suitable for starting the pilot injection strategy, the target injection strategy output by the preset injection strategy determination model includes the target pilot injection strategy and the target main injection strategy. If the target engine is not suitable for starting the pilot injection strategy, the target injection strategy output by the preset injection strategy determination model only includes the target main injection strategy.

[0098] Among them, for the target pilot injection strategy, the preset injection strategy determination model outputs the pilot injection quantity and the pilot injection advance angle. The determination of the pilot injection quantity will comprehensively consider the working conditions of the engine, the environment, and optimization objectives such as combustion noise and emissions. For example, at low load and low ambient temperature, the model may appropriately increase the pilot injection quantity to improve the evaporation and mixing effect of the fuel and reduce combustion noise. The setting of the pilot injection advance angle is optimized according to the engine speed, load, and dynamic characteristics of the combustion process. Innovatively, the model can output a dynamically changing pilot injection advance angle curve instead of a fixed advance angle value to better adapt to the real-time changes of the combustion process of the engine under different working conditions.

[0099] For the target main injection strategy, the preset injection strategy determination model outputs the main injection quantity and the main injection advance angle. The main injection quantity is mainly calculated according to the load demand and power performance requirements of the engine, while taking into account fuel economy and emission indicators. The model will use optimization algorithms to find the main injection quantity that optimizes the comprehensive performance under various constraint conditions. The determination of the main injection advance angle will combine factors such as combustion efficiency and knock risk. Innovatively, the model can dynamically adjust the main injection advance angle according to the real-time operating state of the engine to achieve adaptive optimization of the combustion process. For example, when it is detected that the combustion is insufficient, the main injection advance angle is automatically advanced to improve the combustion efficiency.

[0100] In an alternative embodiment, the training process of the preset fuel injection strategy determination model may include the following steps:

[0101] Step a1, obtain the training attribute information, training condition information, and training working environment information corresponding to multiple training engines.

[0102] Specifically, the electronic device may receive the training attribute information, training condition information, and training working environment information corresponding to multiple training engines input by the user, or may receive the training attribute information, training condition information, and training working environment information corresponding to multiple training engines sent by other devices. The electronic device may also search for the training attribute information corresponding to multiple training engines in the storage space, and obtain the training condition information of multiple training engines based on sensors. Then, the electronic device detects the training working environment information corresponding to multiple training engines based on temperature sensors.

[0103] The embodiments of the present application do not specifically limit the manner in which the electronic device obtains the training attribute information, training condition information, and training working environment information corresponding to multiple training engines.

[0104] Step a2, determine the training fuel injection strategy corresponding to the training engine based on the training attribute information, training condition information, and training working environment information.

[0105] Specifically, the above step a2 may include the following steps:

[0106] Step a21, obtain the initial pre-injection strategy and the initial main injection strategy.

[0107] Specifically, the electronic device may receive the initial pre-injection strategy and the initial main injection strategy input by the user, or may receive the initial pre-injection strategy and the initial main injection strategy sent by other devices. Among them, the initial pre-injection strategy and the initial main injection strategy may be determined based on expert experience. The electronic device may also generate the initial pre-injection strategy and the initial main injection strategy based on expert experience. The embodiments of the present application do not specifically limit the manner in which the electronic device obtains the initial pre-injection strategy and the initial main injection strategy.

[0108] Step a22, for each training engine, obtain the first fuel consumption trend line and the first noise trend line generated after fuel injection based on the initial pre-injection strategy and the initial main injection strategy under the training condition information and the training working environment information.

[0109] Specifically, the electronic device determines the initial pre-injection strategy and the initial main injection strategy. Among them, the initial pre-injection strategy includes parameter settings such as pre-injection fuel quantity and pre-injection advance angle. For example, the pre-injection fuel quantity is set to 0.3 milliliters, and the pre-injection advance angle is set to 10 degrees of crankshaft rotation angle. The initial main injection strategy covers the main injection fuel quantity and the main injection advance angle. Assuming that the main injection fuel quantity is estimated to be 5 milliliters according to the engine load, the main injection advance angle is set to 20 degrees of crankshaft rotation angle before top dead center. These initial values can be obtained based on experience or preliminary theoretical calculations.

[0110] Then, the electronic device determines the training working condition information of the training engine. For example, the engine speed is set at multiple discrete values between 1500 revolutions per minute and 3000 revolutions per minute, and the engine load is set in stages from a low load of 20% to a high load of 80%. At the same time, the training working environment information is determined. For example, the ambient temperature is set in the range of 20°C to 40°C, and the atmospheric pressure is a fluctuating value near the standard atmospheric pressure, etc.

[0111] Then, during the engine operation, based on the high-precision fuel consumption sensor installed in the fuel supply system, the fuel flow data is collected in real time. The fuel consumption sensor converts the collected analog signal into a digital signal and transmits it to the data acquisition system. The data acquisition system records the fuel consumption data at a certain time interval (such as 1 second), so as to obtain a fuel consumption data sequence that changes with time under this working condition and injection strategy. The electronic device processes the collected fuel consumption data, with time as the abscissa and fuel consumption as the ordinate. First, the electronic device removes the outliers in the data, such as the data points that deviate significantly from the normal range due to sensor failures or short-term interferences. Then, the electronic device uses data fitting methods, such as polynomial fitting or exponential fitting, and selects a suitable fitting function according to the change trend of the fuel consumption data. For example, if the fuel consumption data shows an approximately linear growth trend, a first-order polynomial fitting can be used to obtain the functional relationship between fuel consumption and time, and then the first fuel consumption trend line is drawn.

[0112] In addition, during the engine operation, based on the acoustic sensor, the engine combustion noise data is synchronously collected. The microphone converts the received sound signal into an electrical signal, and after signal amplification and filtering processing, it is transmitted to the data acquisition system. The data acquisition system also records the noise data at a time interval of 1 second, including information such as the sound pressure level and frequency components of the noise. For the noise data, the electronic device also uses time as the abscissa and the noise sound pressure level as the ordinate. The electronic device performs spectral analysis on the noise data to understand the frequency distribution of the noise. Remove the background noise component in the noise data to highlight the engine combustion noise. Adopt a data processing method similar to the fuel consumption data, select a suitable fitting function to fit the noise data, obtain the trend of the noise changing with time, and draw the first noise trend line. At the same time, the main frequency components of the noise can be marked on the trend line for subsequent analysis of the relationship between the cause of the noise and the injection strategy.

[0113] Step a23: Obtain the second fuel consumption trend line and the second noise trend line generated after fuel injection based on the initial main injection strategy under the training engine's training condition information and training working environment information.

[0114] Specifically, in the same way as obtaining the above first fuel consumption trend line and the second fuel consumption trend line, the electronic device obtains the second fuel consumption trend line and the second noise trend line generated after fuel injection based on the initial main injection strategy under the training engine's training condition information and training working environment information.

[0115] Step a24: Compare the first fuel consumption trend line corresponding to the training engine with the second fuel consumption trend line, and compare the first noise trend line with the second noise trend line.

[0116] Specifically, the electronic device compares the first fuel consumption trend line corresponding to the training engine with the second fuel consumption trend line, and compares the first noise trend line with the second noise trend line.

[0117] Step a25: According to the comparison results, determine the first training engine suitable for turning on the initial pre-injection strategy and the second training engine not suitable for turning on the initial pre-injection strategy.

[0118] Specifically, if the first fuel consumption trend line does not deteriorate compared to the second fuel consumption trend line, and the first noise trend line is significantly optimized compared to the second noise trend line, then determine that this training engine is the first training engine suitable for turning on the initial pre-injection strategy. If the first fuel consumption trend line deteriorates compared to the second fuel consumption trend line, and / or the first noise trend line is not optimized compared to the second noise trend line, then determine that this training engine is the first training engine not suitable for turning on the initial pre-injection strategy.

[0119] Optionally, the electronic device can determine the first maximum fuel consumption and the first average fuel consumption corresponding to the first fuel consumption trend line, and determine the second maximum fuel consumption and the second average fuel consumption corresponding to the second fuel consumption trend line, and then compare the first maximum fuel consumption and the first average fuel consumption with the second maximum fuel consumption and the second average fuel consumption respectively. If the first maximum fuel consumption is less than or equal to the second maximum fuel consumption, and the first average fuel consumption is less than or equal to the second average fuel consumption, then determine that the first fuel consumption trend line does not deteriorate compared to the second fuel consumption trend line.

[0120] Similarly, the electronic device can determine the first maximum noise and the first average noise corresponding to the first noise trend line, and determine the second maximum noise and the second average noise corresponding to the second noise trend line. Then, it compares the first maximum noise and the first average noise with the second maximum noise and the second average noise respectively. If the first maximum noise is less than the second maximum noise and the first average noise is less than the second average noise, it is determined that the first noise trend line is significantly optimized compared to the second noise trend line.

[0121] Step a26: For the first training engine, determine the first training fuel injection strategy corresponding to the training condition information and the training working environment information. The first training fuel injection strategy includes a training pre-injection strategy and a training main injection strategy.

[0122] Specifically, the above step a26 may include the following steps:

[0123] Step a261: Based on the training attribute information corresponding to the first training engine, generate multiple trial pre-injection strategies and trial main injection strategies.

[0124] Specifically, the electronic device can initially determine the pre-injection fuel quantity range according to the engine displacement and compression ratio included in the training attribute information corresponding to the first training engine. Then, determine the pre-injection advance angle interval according to the combustion characteristics and speed range of the engine. Then, according to the pre-injection fuel quantity range and the pre-injection advance angle interval, use different parameter combination methods. Uniform sampling can be used. For example, divide the pre-injection fuel quantity range into several equal parts, such as at intervals of 0.5 mg / cycle, and select values such as 0.5, 1, 1.5, 2 mg / cycle; make a similar division for the pre-injection advance angle, such as at intervals of 5° crankshaft angle, and select values such as 10°, 15°, 20°, 25°, 30° crankshaft angle. Then, fully combine these pre-injection fuel quantity and pre-injection advance angle values to form multiple trial pre-injection strategies.

[0125] The electronic device determines the main injection fuel quantity range according to the load demand of the first training engine, and determines the main injection advance angle interval according to the combustion characteristics and speed range of the engine. Then, according to the main injection fuel quantity and injection pressure corresponding to the first training engine, determine the main injection duration range.

[0126] The electronic device uses methods such as uniform sampling to perform parameter combination within the determined main injection fuel quantity range, main injection advance angle interval, and main injection duration range. Select different main injection fuel quantity, main injection advance angle, and main injection duration values for combination to form multiple trial main injection strategies.

[0127] Step a262: Based on each trial pre-injection strategy and each trial main injection strategy, generate multiple groups of trial fuel injection strategies.

[0128] Among them, each group of test injection strategies includes a test pre-injection strategy and a test main injection strategy.

[0129] Specifically, the electronic device can randomly combine each test pre-injection strategy and each test main injection strategy based on each test pre-injection strategy and each test main injection strategy to generate multiple groups of test injection strategies.

[0130] Step a263, obtain the engine combustion noise data and engine-related data generated after injection based on each group of test injection strategies under the training condition information and training working environment information of the first training engine.

[0131] Among them, the engine-related data includes at least one of engine power, engine torque, engine fuel consumption rate, and engine emissions.

[0132] Specifically, the electronic device can obtain the engine combustion noise data generated after injection based on each group of test injection strategies under the training condition information and training working environment information of the first training engine through an acoustic sensor arranged at a suitable position around the first training engine.

[0133] Then, the electronic device measures the engine torque and engine speed of the output shaft based on a torque sensor installed on the engine output shaft, and uses the power calculation formula (engine power = engine torque × engine speed ÷ 9550) to obtain the engine power.

[0134] The electronic device measures the fuel volume change based on a high-precision fuel consumption sensor installed in the fuel supply pipeline to calculate the fuel consumption. The sensor is installed at a position close to the injector to ensure that the measured fuel quantity is the actual fuel consumption of the engine.

[0135] Finally, the electronic device obtains the engine emissions based on an emissions sensor installed in the engine exhaust system. For example, a nitrogen oxide (NOx) sensor is used to measure the nitrogen oxide content in the exhaust gas, and a particulate matter sensor is used to measure the particulate matter number or mass concentration. The sensor should be installed at a position where the exhaust temperature is stable and the gas flow is uniform to ensure the accuracy of the measurement data.

[0136] Step a264, determine the first training injection strategy corresponding to the first training engine under the training condition information and training working environment information based on each group of test injection strategies and their corresponding engine combustion noise data and engine-related data.

[0137] Specifically, the above step a264 may include the following steps:

[0138] Step a2641, construct an objective function based on the engine combustion noise data and engine-related data corresponding to each group of test injection strategies.

[0139] Among them, the objective function is generated based on at least one of the engine combustion noise, fuel consumption rate, nitrogen oxide emissions, and particulate matter emissions corresponding to the first training engine.

[0140] Specifically, the electronic device can determine the weight information corresponding to the engine combustion noise, fuel consumption rate, nitrogen oxide emissions, and particulate matter emissions respectively according to the importance corresponding to the engine combustion noise, fuel consumption rate, nitrogen oxide emissions, and particulate matter emissions of the first training engine.

[0141] Then, based on the engine combustion noise, fuel consumption rate, nitrogen oxide emissions, and particulate matter emissions and the corresponding weight information respectively, an objective function is constructed.

[0142] Suppose the electronic device selects four indicators: combustion noise (N), fuel consumption rate (FC), engine power (P), and nitrogen oxide emissions (NOx), and their corresponding weights are w1, w2, w3, and w4 respectively. Then the objective function (F) can be expressed as: F = w1×N + w2×FC + w3×P + w4×NOx. In this expression, each indicator is multiplied by its corresponding weight and then added together to obtain the objective function value. The objective function value comprehensively reflects the comprehensive performance of the engine under different fuel injection strategies.

[0143] Step a2642, construct an objective constraint condition based on the training attribute information corresponding to the first training engine.

[0144] Among them, the objective constraint conditions include fuel injection strategy constraints and engine operating condition constraints.

[0145] Specifically, the electronic device can determine the value range of each parameter in the fuel injection strategy according to the design parameters and actual working ability of the first training engine. For example, the pre-injection fuel quantity cannot be negative, and due to the hardware limitations of the engine fuel injection system, there is a maximum allowable pre-injection fuel quantity, which can be expressed as an inequality constraint: 0 ≤ pre-injection fuel quantity ≤ pre-injection fuel quantity max. In addition, the pre-injection advance angle also has a reasonable angle range. Beyond this range, it may cause abnormal engine combustion or unstable operation, so there is pre-injection advance angle min ≤ pre-injection advance angle ≤ pre-injection advance angle max. For example, to ensure the normal operation of the first training engine, determine the lower and upper limits of the main injection fuel quantity and the lower and upper limits of the main injection advance angle corresponding to the first training engine, that is, main injection fuel quantity min ≤ main injection fuel quantity ≤ main injection fuel quantity max; main injection advance angle min ≤ main injection advance angle ≤ main injection advance angle max.

[0146] An electronic device needs to ensure normal engine operation and good performance according to the operating conditions of the first training engine, such as engine speed and torque, within a specific range. Based on the design specifications and actual application scenarios of the first training engine, the lower and upper limits of the engine speed are determined, i.e., engine speed min ≤ engine speed ≤ engine speed max. For engine torque, there are similar constraints, engine torque min ≤ engine torque ≤ engine torque max. In addition, environmental conditions such as intake air temperature and intake air pressure also affect the operation of the engine. According to the adaptability of the engine and the actual working environment, corresponding constraint conditions are set, such as intake air temperature min ≤ intake air temperature ≤ intake air temperature max.

[0147] Step a2643, construct a target optimization model based on the objective function and the target constraint conditions.

[0148] Specifically, the electronic device constructs a target optimization model based on the objective function and the target constraint conditions.

[0149] Step a2644, solve the target optimization model to determine the first training injection strategy of the first training engine corresponding to the training working condition information and the training working environment information.

[0150] Specifically, the above step a2644 may include the following steps:

[0151] Step a26441, detect whether each group of test injection strategies meets the target constraint conditions.

[0152] Specifically, the electronic device can compare each parameter included in the test injection strategy with the corresponding injection strategy constraint to detect whether each group of test injection strategies meets the injection strategy constraint. In addition, the electronic device can also obtain the engine operation data corresponding to the test injection strategy, and then compare the engine operation data corresponding to the test injection strategy with the corresponding engine operation condition constraint, so as to detect whether each group of test injection strategies meets the engine operation condition constraint.

[0153] Step a26442, if the test injection strategy meets the target constraint conditions, calculate the objective function value corresponding to the test injection strategy based on the objective function.

[0154] Specifically, if the test injection strategy meets the target constraint conditions, the electronic device substitutes the engine combustion noise data and engine-related data corresponding to the test injection strategy into the objective function to calculate the objective function value corresponding to the test injection strategy.

[0155] Step a26443, update each test injection strategy according to the objective function value corresponding to each test injection strategy to generate a new injection strategy.

[0156] Specifically, the electronic device can collect and organize the objective function values corresponding to each test fuel injection strategy and arrange them in a certain order. Then, based on the objective function values corresponding to each test fuel injection strategy, the common characteristics of the optimal test fuel injection strategy in terms of fuel injection parameters are determined. For example, if the smaller the objective function value, the better, the electronic device determines the test fuel injection strategies with objective function values less than the preset function value threshold. Then, it is determined that the test fuel injection strategies with lower objective function values generally have smaller pilot injection advance angles and moderate main injection fuel amounts. These common characteristics may provide an important reference direction for the generation of new fuel injection strategies.

[0157] Optionally, based on the important reference direction provided by the above optimal test fuel injection strategy, the electronic device generates a new fuel injection strategy. For example, if the objective function value of a certain strategy is optimal and its pilot injection fuel amount is 1.5 mg, on this basis, the pilot injection fuel amount can be increased or decreased in smaller steps (such as 0.1 mg) to generate new pilot injection fuel amount values (such as 1.4 mg, 1.6 mg), while keeping other parameters unchanged, thereby obtaining a new fuel injection strategy. Similar fine-tuning can also be performed on parameters such as the main injection advance angle and the main injection fuel amount.

[0158] Optionally, the electronic device can also select multiple test fuel injection strategies with optimal objective function values and recombine their fuel injection parameters. For example, if the pilot injection advance angle and the main injection fuel amount of Strategy A have good effects, and the pilot injection fuel amount and the main injection advance angle of Strategy B are excellent, then the pilot injection advance angle and the main injection fuel amount of Strategy A can be combined with the pilot injection fuel amount and the main injection advance angle of Strategy B to generate a new fuel injection strategy. This method can fully integrate the advantages of different strategies.

[0159] Optionally, the electronic device can also use optimization algorithms such as genetic algorithms and particle swarm optimization algorithms to generate new fuel injection strategies. Taking the genetic algorithm as an example, the test fuel injection strategies are regarded as chromosomes, and the objective function values are used as fitness. Through operations such as selection, crossover, and mutation, new strategies are generated from the current population of test fuel injection strategies. The selection operation tends to retain the strategies with better objective function values; the crossover operation exchanges and combines the parameters of two or more strategies; the mutation operation randomly changes the parameters of some strategies slightly, thereby generating fuel injection strategies with new parameter combinations.

[0160] Step a26444, detect whether each new fuel injection strategy meets the target constraint conditions, and calculate the objective function value corresponding to the new fuel injection strategy based on the objective function.

[0161] Then, the electronic device detects whether each new fuel injection strategy meets the target constraint conditions, and calculates the objective function value corresponding to the new fuel injection strategy based on the objective function.

[0162] Step a26445, until the new fuel injection strategy meets the target constraint conditions and the objective function value corresponding to the new fuel injection strategy is optimal, then determine the new fuel injection strategy as the first training fuel injection strategy of the first training engine corresponding to the training condition information and the training working environment information.

[0163] Specifically, starting from generating a new fuel injection strategy, the electronic device enters a continuously repeating cycle of detection, evaluation, and adjustment. Each time a new set of fuel injection strategies is generated, it needs to be comprehensively inspected and analyzed to determine whether it meets the established goals and conditions. This is a process of gradually approaching the optimal fuel injection strategy, and each cycle is improved based on the results of the previous one.

[0164] For example, after the first generation of a new fuel injection strategy, it is detected. If it is found that the strategy does not meet the target constraint conditions or its objective function value is not optimal, the strategy needs to be updated again according to certain rules and methods, and then detected and evaluated again, and so on, until the requirements are met.

[0165] Specifically, the electronic device can carefully check the fuel injection parameters in the new fuel injection strategy to ensure that they comply with the physical limitations of the fuel injector and the logical relationships of the fuel injection strategy. For example, the fuel injection pressure must be within the range that the fuel injector can withstand, neither too high nor too low. If the fuel injection pressure is too high, it may damage the fuel injector; if it is too low, the normal injection and atomization effect of the fuel cannot be guaranteed. The fuel injection quantity and the fuel injection duration also need to be within a reasonable range, otherwise it will affect the combustion process of the engine. In addition, the time interval between the pre-injection and the main injection, as well as the sequence and number of multiple injections and other logical relationships must be correct to ensure the stability and efficiency of combustion.

[0166] In addition, the electronic device can also monitor the operating state of the engine in real time during the implementation of the new fuel injection strategy to ensure that the operating parameters such as the engine speed, load, and temperature are within the specified range. For example, the engine speed cannot be lower than the minimum stable speed, otherwise it may cause the engine to shake or even stall; nor can it be higher than the maximum safe speed to avoid damaging the engine components. The engine load needs to match the fuel injection strategy to avoid situations of excessive or insufficient load. At the same time, the engine temperature also needs to be maintained within the normal operating range, and too high or too low temperature will affect the performance and reliability of the engine. The intake air volume and the air-fuel ratio also need to meet the operating requirements of the engine to ensure the full combustion of the fuel and the compliance of the emissions.

[0167] After the new fuel injection strategy meets the target constraint conditions, compare the objective function value of the new fuel injection strategy with those of all previous test fuel injection strategies and the current existing new fuel injection strategies. If the objective function value of the new fuel injection strategy is the best among all the compared strategies (which may be minimization or maximization according to the optimization direction of the objective function), then the strategy is considered to perform best in terms of the performance indicators considered by the objective function. For example, if the optimization direction of the objective function is to minimize combustion noise and fuel consumption rate while maximizing engine power and torque, then the new fuel injection strategy with the smallest objective function value is the current optimal strategy.

[0168] When the new fuel injection strategy simultaneously meets the target constraint conditions and its objective function value is optimal, it can be determined that the new fuel injection strategy is the first training fuel injection strategy for the first training engine under the specific training condition information and training working environment information. This finally determined fuel injection strategy is obtained by comprehensively considering various limiting conditions and performance optimization objectives of the engine, enabling the engine to achieve the best operating state under the current working conditions and environment, and realizing optimizations in multiple aspects such as reducing combustion noise, improving fuel economy, enhancing power performance, and meeting emission standards. For example, after multiple iterations and optimizations, a set of new fuel injection strategies is obtained. The fuel injection parameters of these strategies meet the physical limitations of the fuel injectors and the fuel injection logic relationship. At the same time, when the engine operates under this strategy, operating parameters such as speed, load, and temperature are within the normal range, and the objective function value of this strategy is the smallest among all the compared strategies, that is, indicators such as combustion noise and fuel consumption rate are effectively reduced, the engine power and torque also remain at a good level, and the emissions meet the standards. Then it can be determined that this set of new fuel injection strategies is the best fuel injection strategy for the first training engine under the current training conditions and environment.

[0169] Step a27, for the second training engine, determine the second training fuel injection strategy corresponding to the training condition information and training working environment information of the second training engine.

[0170] Among them, the second training fuel injection strategy only includes the training main injection strategy.

[0171] Specifically, based on the training attribute information corresponding to the second training engine, generate multiple test main injection strategies. Obtain the engine combustion noise data and engine-related data generated after fuel injection based on each group of test main injection strategies under the training condition information and training working environment information of the second training engine. Among them, the engine-related data includes at least one of engine power, engine torque, engine fuel consumption rate, and engine emissions.

[0172] Then, based on the engine combustion noise data and engine-related data corresponding to each group of test main injection strategies, the electronic device constructs an objective function. The objective function is generated based on at least one of the engine combustion noise, fuel consumption rate, nitrogen oxide emissions, and particulate matter emissions corresponding to the second training engine.

[0173] In addition, the electronic device constructs objective constraint conditions based on the training attribute information corresponding to the second training engine. The objective constraint conditions include injection strategy constraints and engine operating condition constraints. The electronic device constructs an objective optimization model based on the objective function and the objective constraint conditions.

[0174] The electronic device detects whether each group of test main injection strategies meets the objective constraint conditions. If the test main injection strategy meets the objective constraint conditions, the electronic device calculates the objective function value corresponding to the test main injection strategy based on the objective function. Then, the electronic device updates each test main injection strategy according to the objective function value corresponding to each test main injection strategy to generate a new main injection strategy. The electronic device detects whether each new main injection strategy meets the objective constraint conditions and calculates the objective function value corresponding to the new main injection strategy based on the objective function. Until the new main injection strategy meets the objective constraint conditions and the objective function value corresponding to the new main injection strategy is optimal, the new main injection strategy is determined as the second training injection strategy corresponding to the second training engine under the training condition information and the training working environment information.

[0175] Step a3: Generate a training data set based on the training attribute information, training condition information, training working environment information, and training injection strategy corresponding to each training engine.

[0176] Specifically, the electronic device generates a training data set based on the training attribute information, training condition information, training working environment information, and training injection strategy corresponding to each training engine. The training injection strategy is label information.

[0177] Step a4: Train the initial injection strategy determination network based on the training data set to obtain a preset injection strategy determination model.

[0178] The initial injection strategy determination network can be a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN) and its variants (such as long short-term memory network LSTM, gated recurrent unit GRU), etc. The embodiments of the present application do not make specific limitations on the initial injection strategy determination network.

[0179] Specifically, the electronic device can initialize the parameters of the initial fuel injection strategy determination network, usually using a random initialization method to make the network have a certain degree of randomness at the beginning of training. Common initialization methods include uniform distribution initialization, normal distribution initialization, etc. Reasonable parameter initialization can help the network converge faster and avoid problems such as gradient disappearance or explosion.

[0180] Then, the electronic device determines an appropriate loss function to measure the difference between the predicted fuel injection strategy of the model and the actual fuel injection strategy (or the corresponding performance index). For example, the mean squared error (MSE) loss function can be used to measure the error between the predicted fuel consumption rate and the actual fuel consumption rate, or the cross-entropy loss function can be used to handle classification problems (such as determining whether the fuel injection strategy meets certain conditions).

[0181] Finally, the electronic device can determine an optimizer to update the parameters of the network to minimize the loss function. Common optimizers include Stochastic Gradient Descent (SGD), Adagrad, Adadelta, RMSProp, Adam, etc. Different optimizers have different characteristics and application scenarios, and need to be selected according to specific problems.

[0182] The specific training process is as follows:

[0183] The electronic device inputs the training set of the training data set into the initial fuel injection strategy determination network and calculates the prediction result of the model through forward propagation. Then, the value of the loss function is calculated based on the prediction result and the actual label (i.e., the labeled performance index). The optimizer calculates the gradient through the backpropagation algorithm and updates the parameters of the network to reduce the value of the loss function. Repeat the above process to perform multiple rounds of training on the training set until the loss function converges or reaches the preset number of training times. During the training process, the validation set can be used regularly to evaluate the performance of the model and adjust the hyperparameters to prevent overfitting.

[0184] Step S203, based on the target fuel injection strategy, control the fuel injection mode of the target engine to reduce the combustion noise corresponding to the target engine.

[0185] For this step, please refer to the introduction of step S103 above and will not be elaborated here.

[0186] The engine combustion noise control method provided by the embodiments of the present application inputs attribute information, current operating condition information, and current working environment information into a preset fuel injection strategy determination model. The preset fuel injection strategy determination model extracts features from the attribute information, current operating condition information, and current working environment information, and based on the extracted features, outputs the target fuel injection strategy corresponding to the target engine. The preset fuel injection strategy determination model can quickly process the input target engine attribute information, current operating condition information, and current working environment information. Compared with manual analysis or traditional simple rule judgment, through algorithms and data training, the model can accurately extract key features in a short time and generate appropriate target fuel injection strategies based on these features, ensuring the accuracy of the generated target fuel injection strategies. The preset fuel injection strategy determination model after sufficient training has strong adaptability and generalization ability. It can process the attribute information of various different types of engines, as well as the operating conditions and working environment information within a wide range. Whether it is engines of different brands and models, or in various extreme working environments (such as high-temperature and high-cold regions) and complex operating conditions (such as frequent start-stop, rapid acceleration and deceleration), the model can generate appropriate fuel injection strategies according to the input information. This makes the solution have a wider application scenario and can adapt to diverse user needs and actual usage situations. In addition, the above method reduces the influence of human factors on the determination of fuel injection strategies and reduces the errors caused by misjudgment of human factors.

[0187] Among them, the training process of the preset fuel injection strategy determination model is as follows: Obtain the training attribute information, training working condition information, and training working environment information corresponding to multiple training engines. Then, obtain the initial pre-injection strategy and the initial main injection strategy. For each training engine, obtain the first fuel consumption trend line and the first noise trend line generated after fuel injection based on the initial pre-injection strategy and the initial main injection strategy under the training working condition information and the training working environment information. Obtain the second fuel consumption trend line and the second noise trend line generated after fuel injection based on the initial main injection strategy under the training working condition information and the training working environment information. The electronic device compares the first fuel consumption trend line corresponding to the training engine with the second fuel consumption trend line, and compares the first noise trend line with the second noise trend line; according to the comparison results, determine the first training engine suitable for turning on the initial pre-injection strategy and the second training engine not suitable for turning on the initial pre-injection strategy. Thus, it can accurately judge which training engines are suitable for turning on the pre-injection strategy and which are not. This comparative analysis based on actual operation data can fully consider the performance differences of different training engines under specific working conditions and environments, avoiding the adverse effects that may be brought about by blindly adopting the pre-injection strategy, such as fuel consumption deterioration in some engines, etc., and improving the pertinence and accuracy of fuel injection strategy formulation. For the first training engine, generate a variety of test pre-injection strategies and test main injection strategies based on the training attribute information corresponding to the first training engine, so as to comprehensively explore the possible combination space of fuel injection strategies. Then, based on each test pre-injection strategy and each test main injection strategy, generate multiple groups of test fuel injection strategies. Obtain the engine combustion noise data and engine-related data generated after fuel injection based on each group of test fuel injection strategies under the training working condition information and the training working environment information of the first training engine. This multi-dimensional data collection and evaluation method can more comprehensively understand the comprehensive impact of different fuel injection strategies on engine performance. For example, not only can we focus on reducing combustion noise, but also consider fuel economy, power output, and environmental protection performance at the same time, which helps to determine a fuel injection strategy that can achieve a better balance in all aspects and improve the overall performance of the engine. Based on the engine combustion noise data and engine-related data corresponding to each group of test fuel injection strategies, construct an objective function. Thus, the goal of engine performance optimization can be scientifically quantified. Instead of vaguely pursuing engine performance improvement, an objective function is clearly constructed with specific indicators such as combustion noise, fuel consumption rate, and emissions, making the optimization direction clearer and more specific. Based on the training attribute information corresponding to the first training engine, construct objective constraint conditions; the objective constraint conditions include fuel injection strategy constraints and engine operation condition constraints. All kinds of limiting factors in the actual operation of the engine are comprehensively considered.The injection strategy constraints ensure that the injection parameters are within the range allowed by the engine hardware and technology, avoiding unreasonable injection settings; the engine operating condition constraints ensure that the engine operates under safe and stable conditions. This comprehensive constraint setting makes the optimization results more practically feasible and avoids situations that are theoretically feasible but actually unimplementable. Based on the objective function and the objective constraint conditions, an objective optimization model is constructed. The objective optimization model can accurately describe the relationship between the injection strategy and the engine performance, as well as the limitations of various constraint conditions on the injection strategy. Detecting whether each group of test injection strategies meets the objective constraint conditions can timely exclude those injection strategies that do not conform to the actual operating limitations of the engine during the optimization process. If the test injection strategy meets the objective constraint conditions, then based on the objective function, calculate the objective function value corresponding to the test injection strategy, so as to accurately quantify the comprehensive impact of each test injection strategy on the engine performance. According to the objective function values corresponding to each test injection strategy, update each test injection strategy to generate a new injection strategy; detect whether each new injection strategy meets the objective constraint conditions, and based on the objective function, calculate the objective function value corresponding to the new injection strategy. By continuously adjusting the injection strategy parameters and re-evaluating whether they meet the constraint conditions and calculating the objective function values, the optimal solution can be gradually approached. This iterative optimization method can fully explore better injection strategies and continuously improve the engine performance, such as further reducing combustion noise and improving fuel economy. Until the new injection strategy meets the objective constraint conditions and the objective function value corresponding to the new injection strategy is the best, then the new injection strategy is determined as the first training injection strategy of the first training engine corresponding to the training condition information and the training working environment information. The finally determined first training injection strategy is obtained under the condition of meeting the objective constraint conditions and having the best objective function value. This means that this injection strategy can not only theoretically optimize the comprehensive performance of the engine to the best, but also ensure the stable and reliable operation of the engine in actual operation. A suitable injection strategy can make the combustion process of the engine more stable, reduce the occurrence of abnormal combustion phenomena, thereby extending the service life of the engine and reducing the maintenance cost. In addition, this optimization process takes into account various factors and constraint conditions and can generate injection strategies adapted to different training condition information and training working environment information. Whether the engine operates at different speeds and loads, or under different environmental temperatures, pressures, etc., the injection strategies obtained through this optimization process can better adapt to these complex situations and ensure that the engine maintains good performance in various working conditions and environments. For the second training engine that is not suitable for starting the pilot injection strategy, only the main injection strategy for training is adopted, avoiding the negative impact brought by inappropriate pilot injection, ensuring that the engine operates in the best state under this condition and environment, and maintaining good performance.

[0188] Then, based on the training attribute information, training operating condition information, training working environment information, and training fuel injection strategies corresponding to each training engine, a training data set is generated. This enables the training of the preset fuel injection strategy determination model to be based on a large amount of real and rich data. These data cover the situations of different engines under various operating conditions and environments. The preset fuel injection strategy determination model can learn the mapping relationship between different conditions and the optimal fuel injection strategy. Thus, in practical applications, for the specific situation of the target engine, a more accurate fuel injection strategy can be output, effectively reducing combustion noise and improving engine performance. Based on the training data set, the initial fuel injection strategy determination network is trained to obtain the preset fuel injection strategy determination model. During the training process, the model can adjust its own parameters such as weights and biases according to the feedback of the training data to minimize the error between the predicted fuel injection strategy and the actual training fuel injection strategy. Through multiple iterative trainings, the performance of the model is gradually improved, and the processing ability for different input information and the output accuracy of the fuel injection strategy will be improved, thereby more effectively reducing the engine combustion noise and enhancing the comprehensive performance of the engine.

[0189] In this embodiment, an engine combustion noise control device is also provided. This device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that can achieve 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 contemplated.

[0190] This embodiment provides an engine combustion noise control device, as Figure 3 shown, including:

[0191] An acquisition module 301, configured to acquire the attribute information, current operating condition information, and current working environment information corresponding to the target engine;

[0192] A determination module 302, configured to determine the target fuel injection strategy corresponding to the target engine based on the attribute information, current operating condition information, and current working environment information; the target fuel injection strategy includes a target pre-injection strategy and a target main injection strategy or only includes the target main injection strategy; the target pre-injection strategy includes the pre-injection fuel quantity and the pre-injection advance angle; the target main injection strategy includes the main injection fuel quantity, the main injection advance angle, and the main injection duration;

[0193] A control module 303, configured to control the fuel injection mode of the target engine based on the target fuel injection strategy to reduce the combustion noise corresponding to the target engine.

[0194] In some alternative embodiments, the determination module 302 is specifically configured to input the attribute information, the current operating condition information, and the current working environment information into a preset fuel injection strategy determination model; the preset fuel injection strategy determination model extracts features from the attribute information, the current operating condition information, and the current working environment information, and outputs a target fuel injection strategy corresponding to the target engine based on the extracted features.

[0195] In some alternative embodiments, the determination module 302 is specifically configured to obtain the training attribute information, the training operating condition information, and the training working environment information corresponding to a plurality of training engines; determine the training fuel injection strategy corresponding to the training engines based on the training attribute information, the training operating condition information, and the training working environment information; generate a training data set based on the training attribute information, the training operating condition information, the training working environment information, and the training fuel injection strategy corresponding to each training engine; and train an initial fuel injection strategy determination network based on the training data set to obtain a preset fuel injection strategy determination model.

[0196] In some alternative embodiments, the determination module 302 is specifically configured to obtain an initial pilot injection strategy and an initial main injection strategy; for each training engine, obtain a first fuel consumption trend line and a first noise trend line generated after fuel injection based on the initial pilot injection strategy and the initial main injection strategy under the training operating condition information and the training working environment information; obtain a second fuel consumption trend line and a second noise trend line generated after fuel injection based on the initial main injection strategy under the training operating condition information and the training working environment information; compare the first fuel consumption trend line corresponding to the training engine with the second fuel consumption trend line, and compare the first noise trend line with the second noise trend line; determine a first training engine suitable for activating the initial pilot injection strategy and a second training engine not suitable for activating the initial pilot injection strategy according to the comparison results; for the first training engine, determine a first training fuel injection strategy corresponding to the first training engine under the training operating condition information and the training working environment information; the first training fuel injection strategy includes a training pilot injection strategy and a training main injection strategy; for the second training engine, determine a second training fuel injection strategy corresponding to the second training engine under the training operating condition information and the training working environment information; the second training fuel injection strategy only includes a training main injection strategy.

[0197] In some alternative embodiments, the determining module 302 is specifically configured to generate a variety of test pre-injection strategies and test main injection strategies based on the training attribute information corresponding to the first training engine; generate multiple groups of test fuel injection strategies based on each test pre-injection strategy and each test main injection strategy; each group of test fuel injection strategies includes a test pre-injection strategy and a test main injection strategy; obtain the engine combustion noise data and engine-related data generated after fuel injection based on each group of test fuel injection strategies under the training condition information and training working environment information of the first training engine, and the engine-related data includes at least one of engine power, engine torque, engine fuel consumption rate, and engine emissions; determine the first training fuel injection strategy of the first training engine corresponding to the training condition information and training working environment information based on each group of test fuel injection strategies and their corresponding engine combustion noise data and engine-related data.

[0198] In some alternative embodiments, the determining module 302 is specifically configured to construct an objective function based on the engine combustion noise data and engine-related data corresponding to each group of test fuel injection strategies; the objective function is generated based on at least one of the engine combustion noise, fuel consumption rate, nitrogen oxide emissions, and particulate matter emissions corresponding to the first training engine; construct objective constraint conditions based on the training attribute information corresponding to the first training engine; the objective constraint conditions include fuel injection strategy constraints and engine operating condition constraints; construct an objective optimization model based on the objective function and the objective constraint conditions; solve the objective optimization model to determine the first training fuel injection strategy of the first training engine corresponding to the training condition information and training working environment information.

[0199] In some alternative embodiments, the determining module 302 is specifically configured to detect whether each group of test fuel injection strategies meets the objective constraint conditions; if the test fuel injection strategy meets the objective constraint conditions, calculate the objective function value corresponding to the test fuel injection strategy based on the objective function; update each test fuel injection strategy according to the objective function values corresponding to each test fuel injection strategy to generate a new fuel injection strategy; detect whether each new fuel injection strategy meets the objective constraint conditions and calculate the objective function value corresponding to the new fuel injection strategy based on the objective function; until the new fuel injection strategy meets the objective constraint conditions and the objective function value corresponding to the new fuel injection strategy is optimal, determine the new fuel injection strategy as the first training fuel injection strategy of the first training engine corresponding to the training condition information and training working environment information.

[0200] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding embodiments above, and will not be repeated here.

[0201] The engine combustion noise control device in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0202] An embodiment of the present invention further provides an electronic device having the above Figure 3 shown engine combustion noise control device.

[0203] Please refer to Figure 4 , Figure 4 which is a schematic structural diagram of an electronic device provided by an alternative embodiment of the present invention. As Figure 4 shown, the electronic device includes: one or more processors 10, a memory 20, and an interface for connecting each component, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the electronic device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (such as an array of servers, a set of blade servers, or a multi-processor system). Figure 4 One processor 10 is taken as an example in

[0204] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The above programmable logic device can be a complex programmable logic device, a field programmable gate array, a general array logic, or any combination thereof.

[0205] Among them, the memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0206] The memory 20 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely provided with respect to the processor 10, and these remote memories may be connected to the electronic device through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0207] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive; the memory 20 may further include a combination of the above types of memories.

[0208] The electronic device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 may be connected through a bus or other means. Figure 4 Taking the connection through the bus as an example.

[0209] The input device 30 may receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the electronic device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (such as an LED), and a haptic feedback device (such as a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light-emitting diode, a display, and a plasma display. In some alternative embodiments, the display device may be a touch screen.

[0210] Embodiments of the present invention also provide a computer-readable storage medium. The methods according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium. Thus, the methods described herein can be stored as such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by the computer, the processor, or the hardware, the methods shown in the above embodiments are implemented.

[0211] A part of the present invention can be applied as a computer program product, such as computer program instructions. When executed by a computer, through the operation of the computer, the methods and / or technical solutions according to the present invention can be invoked or provided. Those skilled in the art should understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Herein, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible by the computer.

[0212] Although the embodiments of the present invention are described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for controlling engine combustion noise, characterized in that: The method comprises: Obtaining the property information, current working condition information and current working environment information corresponding to the target engine; Based on the attribute information, the current operating condition information and the current working environment information, a target injection strategy corresponding to the target engine is determined; the target injection strategy includes a target pre-injection strategy and a target main injection strategy or only includes a target main injection strategy; the target pre-injection strategy includes a pre-injection amount and a pre-injection advance angle; the target main injection strategy includes a main injection amount, a main injection advance angle and a main injection duration; Based on the target fuel injection strategy, the fuel injection mode of the target engine is controlled to reduce the combustion noise corresponding to the target engine.

2. The method according to claim 1, characterized in that The determining, based on the attribute information, the current operating condition information and the current working environment information, a target fuel injection strategy corresponding to the target engine includes: Inputting the attribute information, the current operating condition information and the current working environment information into a preset fuel injection strategy determination model; The preset fuel injection strategy determination model extracts features from the attribute information, the current operating condition information, and the current working environment information, and outputs the target fuel injection strategy corresponding to the target engine based on the extracted features.

3. The method according to claim 2, characterized in that The training process of the preset fuel injection strategy determination model includes: Obtaining training attribute information, training condition information, and training working environment information corresponding to multiple training engines; Determining a training fuel injection strategy corresponding to the training engine based on the training attribute information, the training operating condition information, and the training working environment information; Generate a training data set based on the training attribute information, the training operating condition information, the training working environment information, and the training injection strategy corresponding to each of the training engines; Based on the training data set, the initial fuel injection strategy determination network is trained to obtain the preset fuel injection strategy determination model.

4. The method according to claim 3, characterized in that The determining, based on the training attribute information, the training operating condition information, and the training working environment information, a training injection strategy corresponding to the training engine includes: Obtaining an initial pre-spray strategy and an initial main spray strategy; For each of the training engines, obtaining a first fuel consumption trend line and a first noise trend line generated after the training engine performs fuel injection based on the initial pilot injection strategy and the initial main injection strategy under the training operating condition information and the training working environment information; Obtaining a second fuel consumption trend line and a second noise trend line generated by the training engine after fuel injection based on the initial main injection strategy under the training operating condition information and the training working environment information; Comparing the first fuel consumption trend line and the second fuel consumption trend line corresponding to the training engine, and comparing the first noise trend line and the second noise trend line; According to the comparison result, determining a first training engine suitable for starting the initial pilot injection strategy and a second training engine unsuitable for starting the initial pilot injection strategy; For the first training engine, determining a first training injection strategy of the first training engine corresponding to the training operating condition information and the training working environment information; the first training injection strategy includes a training pre-injection strategy and a training main injection strategy; For the second training engine, a second training injection strategy of the second training engine corresponding to the training operating condition information and the training working environment information is determined; the second training injection strategy only includes a training main injection strategy.

5. The method according to claim 4, characterized in that The determining of the first training fuel injection strategy of the first training engine corresponding to the training operating condition information and the training working environment information includes: generating a plurality of test pilot injection strategies and test main injection strategies based on the training attribute information corresponding to the first training engine; Based on each of the test pilot injection strategies and each of the test main injection strategies, a plurality of groups of test injection strategies are generated; each group of the test injection strategies includes one of the test pilot injection strategies and one of the test main injection strategies; Obtaining engine combustion noise data and engine-related data generated by the first training engine after fuel injection based on each group of the test fuel injection strategies under the training operating condition information and the training working environment information, wherein the engine-related data includes at least one of engine power, engine torque, engine fuel consumption rate, and engine emissions; Based on each group of the test injection strategies and the corresponding engine combustion noise data and engine-related data, the first training injection strategy corresponding to the training operating condition information and the training working environment information of the first training engine is determined.

6. The method according to claim 5, characterized in that The determining, based on each group of the test injection strategies and the corresponding engine combustion noise data and engine-related data, the first training injection strategy corresponding to the training operating condition information and the training working environment information of the first training engine comprises: constructing an objective function based on the engine combustion noise data and engine-related data corresponding to each group of the test injection strategies; the objective function is generated based on at least one of the engine combustion noise, fuel consumption rate, nitrogen oxide emissions, and particulate matter emissions corresponding to the first training engine; Based on the training attribute information corresponding to the first training engine, constructing target constraint conditions; the target constraint conditions include injection strategy constraints and engine operating condition constraints; Constructing a target optimization model based on the target function and the target constraint condition; The target optimization model is solved to determine the first training injection strategy of the first training engine corresponding to the training operating condition information and the training working environment information.

7. The method according to claim 6, characterized in that The solving the target optimization model to determine the first training injection strategy of the first training engine corresponding to the training operating condition information and the training working environment information includes: Detecting whether the test fuel injection strategies of each group meet the target constraint conditions; If the test fuel injection strategy satisfies the target constraint condition, then based on the target function, calculating the target function value corresponding to the test fuel injection strategy; updating each of the test injection strategies according to the objective function value corresponding to each of the test injection strategies to generate a new injection strategy; Detecting whether each of the new injection strategies satisfies the target constraint condition, and calculating the target function value corresponding to the new injection strategy based on the target function; Until the new injection strategy satisfies the target constraint condition and the objective function value corresponding to the new injection strategy is optimal, the new injection strategy is determined as the first training injection strategy corresponding to the training operating condition information and the training working environment information of the first training engine.

8. An engine combustion noise control device, characterized in that: The device comprises: An acquisition module is used to obtain the property information, current working condition information and current working environment information corresponding to the target engine; A determination module, configured to determine a target injection strategy corresponding to the target engine based on the attribute information, the current operating condition information, and the current working environment information; the target injection strategy includes a target pre-injection strategy and a target main injection strategy or only includes a target main injection strategy; the target pre-injection strategy includes a pre-injection amount and a pre-injection advance angle; the target main injection strategy includes a main injection amount, a main injection advance angle, and a main injection duration; A control module is used to control the fuel injection mode of the target engine based on the target fuel injection strategy to reduce the combustion noise corresponding to the target engine.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the engine combustion noise control method according to any one of claims 1 to 7 by executing the computer instructions.

10. A target engine, characterized in that: The invention comprises an electronic device and an engine body, wherein the electronic device is used to execute the engine combustion noise control method according to any one of claims 1 to 7.

Images