A diesel engine control method and device based on the combination of LSTM algorithm and PID feedback

Abstract

A diesel engine control method and device combining feedback based on the LSTM algorithm and PID, which relates to the technical field of diesel engine speed control. The technical problem to be solved is "how to provide a diesel engine control method that can be applied to the initial state and has high accuracy". The method includes: collecting diesel engine parameters; inputting the diesel engine parameters into the LSTM network for iterative training until a predicted speed that meets the preset requirements is obtained; taking the difference between the predicted speed that meets the preset requirements and the actual speed at the previous moment and inputting it into the PID network to obtain the optimal speed. This method uses the LSTM algorithm for prediction in the feedback link and jointly obtains the optimal speed with the PID control algorithm to control the diesel engine, which can be applied to diesel engines in the initial state and has high control efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of diesel engine speed control. Background Art

[0002] In the control algorithms for diesel engines, the control index often targeted is the speed of the diesel engine. Among many control methods, the most widely used is still the traditional PID control method. During the development and improvement over the years, the traditional PID control has good reliability and practicability, but its control effect is not very ideal compared with new technologies.

[0003] In the existing public materials, most of them adopt a single control method. For example, the patent with the publication number CN114237049A discloses a method for tuning the predictive control parameters of an intelligent building system based on LSTM, which mainly tunes the parameters of the intelligent building system based on the LSTM network technology. The scheme includes the following steps: First, obtain training samples based on the MPC algorithm; Second, establish a long short-term memory artificial neural network and train the long short-term memory artificial neural network according to the training samples obtained in the first step; Third, use the trained long short-term memory artificial neural network to predict the control parameters of the intelligent building system. This method can realize the tuning of the predictive control parameters of the intelligent building system. The scheme provided by this patent does not cover the situation when a system does not have a large number of samples, that is, the initial state. At this time, due to insufficient training volume, the initial stability of the LSTM will be poor, so this scheme is not applicable to this scenario.

[0004] Another example is that the invention patent with the publication number CN106773649A discloses an intelligent control method for a gas automatic control valve based on the PSO and PID algorithms. Taking the gas automatic control valve as the research object, fully considering the characteristics of the PID control system, establishing a transfer function model of the controlled object, introducing artificial intelligence technology, and using an improved particle swarm algorithm with a convergence factor to search for the optimal solution of the PID parameters, realizing the automatic control of the gas flow. This invention patent overcomes the shortcomings of manually adjusting the parameters of the traditional PID gas control valve and realizes the self-tuning of the PID parameters. However, this scheme combines the PSO and PID algorithms for control and does not involve the influence of interference on the control parameters, resulting in problems such as human error, long time consumption, and low accuracy during the parameter tuning process, and the control effect of the diesel engine is not ideal.

[0005] Therefore, how to provide a diesel engine control method that can be applied to the initial state and has high precision has become an urgent technical problem in this field. Summary of the Invention

[0006] To solve the technical problems of the lack of a large number of samples in the initial state and the low control accuracy in the existing technology, the present invention provides a diesel engine control method and device based on the combination of an LSTM algorithm and PID in the feedback link. This method uses the LSTM algorithm for prediction in the feedback link and jointly obtains the optimal speed with the PID control algorithm to control the diesel engine. This method can be applied to the diesel engine in the initial state and has high control efficiency and accuracy.

[0007] Based on the same inventive concept, the present invention has four independent technical solutions:

[0008] 1. A diesel engine control method based on the combination of an LSTM algorithm and PID in the feedback link, including:

[0009] S1. Collect diesel engine parameters;

[0010] S2. Input the diesel engine parameters into the LSTM network for iterative training until the predicted speed that meets the preset requirements is obtained;

[0011] S3. Subtract the predicted speed that meets the preset requirements from the actual speed at the previous moment and input it into the PID network to obtain the optimal speed.

[0012] Further, the diesel engine parameters include: load torque M B , diesel engine speed N d , effective torque M of the diesel engine e , moment of inertia I of the main engine moving parts to the pump impeller of the fluid coupling e , friction torque M of the main engine moving parts f1 , fuel supply rack position F r .

[0013] Further, the diesel engine parameters are determined according to the control quasi-steady state model, and the control quasi-steady state model is specifically as follows:

[0014]

[0015]

[0016] Further, before the iterative training, it also includes: preprocessing the diesel engine parameters.

[0017] Further, step S2 includes:

[0018] S21. Input the diesel engine parameters into the LSTM network for training and obtain the predicted speed according to the trained LSTM network;

[0019] S22. Determine whether the predicted speed meets the preset requirements. If it meets, execute step S3;

[0020] Otherwise, replace the diesel engine speed N in the diesel engine parameters with the predicted speed obtained in step S21 d And repeat step S21 until the obtained predicted speed meets the preset requirements.

[0021] Further, the preset requirements are as follows: the maximum error of the mean square error prediction value between the actual speed and the predicted speed is not greater than 8%.

[0022] Further, the calculation formula in the PID network is as follows:

[0023]

[0024]

[0025] Wherein, u(k) is the output value of the PID network at the kth sampling moment, e(k) is the deviation value input at the kth sampling moment, e(k - 1) represents the deviation value input at the (k - 1)th sampling moment, N d0 (k) is the actual diesel engine speed at the kth sampling moment, is the predicted speed at the kth sampling moment, K p 、K I 、K D are all PID network coefficients.

[0026] 2. A diesel engine control device with feedback based on the combination of LSTM algorithm and PID, comprising:

[0027] An acquisition module, configured to acquire the required diesel engine parameters;

[0028] A prediction module, configured to input the diesel engine parameters into the LSTM network for iterative training until a predicted speed that meets the preset requirements is obtained;

[0029] A PID control module, configured to subtract the predicted speed that meets the preset requirements from the speed at the previous moment and input it into the PID network to obtain the optimal speed for controlling the diesel engine.

[0030] 3. A computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor executes the above method.

[0031] 4. An electronic device, comprising a processor and a storage device, wherein multiple instructions are stored in the storage device, and the processor is configured to read the multiple instructions in the storage device and execute the above method.

[0032] The diesel engine control method and device with feedback based on the combination of LSTM algorithm and PID provided by the present invention at least have the following beneficial effects:

[0033] The LSTM network is used to replace the traditional feedback link, and the LSTM algorithm is used to predict the rotational speed data that is about to be input into the PID control. Since the LSTM algorithm has a retention algorithm and a forgetting algorithm, it can effectively eliminate interference, thereby eliminating the long-term error accumulation situation, correcting the interference data, making the predicted rotational speed difference input into the PID control more accurate than without the LSTM algorithm, and achieving a better control effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0035] Figure 1 It is a flowchart of an embodiment of a diesel engine control method provided by the present invention with feedback based on the combination of the LSTM algorithm and PID;

[0036] Figure 2 It is a flowchart of an embodiment of the LSTM algorithm iteration in the diesel engine control method provided by the present invention with feedback based on the combination of the LSTM algorithm and PID;

[0037] Figure 3 It is a control block diagram of an embodiment of a diesel engine control method provided by the present invention with feedback based on the combination of the LSTM algorithm and PID. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] In order to better understand the above technical solutions, the following will describe the above technical solutions in detail in combination with the drawings in the specification and specific embodiments.

[0039] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system structures and technologies are proposed in order to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0040] It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0041] It should also be understood that the terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. As used in the specification of this application and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0042] The following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope protected by this application.

[0043] Many specific details are set forth in the following description to facilitate a thorough understanding of this application, but this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the spirit of this application, so this application is not limited by the specific embodiments disclosed below.

[0044] Embodiment 1:

[0045] See Figure 1 , in some embodiments, a diesel engine control method based on the combination of the LSTM algorithm and PID is provided, including:

[0046] S1. Collect diesel engine parameters;

[0047] S2. Input the diesel engine parameters into the LSTM network for iterative training until a predicted speed that meets the preset requirements is obtained;

[0048] S3. Subtract the predicted speed that meets the preset requirements from the actual speed at the previous moment and input it into the PID network to obtain the optimal speed.

[0049] Specifically, in step S1, the diesel engine parameters include: load torque M B , diesel engine speed N d , effective torque of the diesel engine M e , moment of inertia I of the main engine moving parts to the pump impeller of the fluid coupling e , frictional torque M of the main engine moving parts f1 , fuel supply rack position F r .

[0050] Among them, the diesel engine parameters are determined according to the control quasi-steady state model and the diesel engine dynamics model, and the control quasi-steady state model is specifically as follows:

[0051]

[0052]

[0053] Among them, M B represents the load torque, and N d represents the diesel engine speed, and M e represents the effective torque of the diesel engine, and I e represents the moment of inertia from the main engine moving parts to the pump impeller of the hydrodynamic coupling, and M f1 represents the frictional torque of the main engine moving parts.

[0054] Meanwhile, according to the diesel engine dynamics model, the diesel engine speed N d is a function of the fuel supply rack position F r and the effective torque M of the diesel engine e , and is specifically represented by the following formula:

[0055] N d = f(F r , M e ).

[0056] In some embodiments, before the iterative training, it further includes: on the basis of establishing the model, preprocessing the data to be used for training the LSTM network through the memory unit in the LSTM network.

[0057] In step S2, the experimentally collected data including the load torque M B , the diesel engine speed N d , the effective torque M of the diesel engine e , the moment of inertia I from the main engine moving parts to the pump impeller of the hydrodynamic coupling e , the frictional torque of the main engine moving parts, and the fuel supply rack position F r and other data are normalized and then used as training data to be input into the LSTM network for training. The output value of the LSTM neural network is the predicted speed

[0058] The LSTM network used in step S2 is introduced below. The LSTM neural network model is structurally divided into three layers: an input layer, a hidden layer, and an output layer. Among them, the hidden layer has a memory unit and a gate structure. The memory unit is used to remember past information, and the gate structure includes an input gate, a forget gate, and an output gate, which are used to control the use of historical information. Through the memory unit, interference can be removed from the data, and the prediction accuracy is higher compared to the input data without interference removal.

[0059] Specifically, step S2 includes:

[0060] S21. Input the diesel engine parameters into the LSTM network for training, and obtain the predicted speed according to the trained LSTM network;

[0061] S22. Determine whether the predicted rotational speed meets the preset requirements. If it does, execute step S3;

[0062] Otherwise, replace the diesel engine speed N in the diesel engine parameters with the predicted rotational speed obtained in step S21 d and repeat step S21 until the obtained predicted rotational speed meets the preset requirements.

[0063] In step S21, the LSTM network training process is shown by the following formula:

[0064] i t = σ(W xi x t + W hi h t-1 + W ci c t-1 + b i ) (1.1)

[0065] f t = σ(W xf x t + W hf h t-1 + W cf c t-1 + b f ) (1.2)

[0066] c t = f t c t-1 + i t anh(W xc x t + W hc h t-1 + b c ) (1.3)

[0067] o t = σ(W xo x t + W ho h t-1 + W co c t + b o ) (1.4)

[0068] h t = o t tanh(c t ) (1.5)

[0069] Train according to the LSTM network steps (1.1) - (1.5), and the output result is the predicted rotational speed ).

[0070] In step S22, the preset requirement is that the mean square error (MSE) between the actual speed and the predicted speed is no greater than 8%. Here, the actual speed is the diesel engine speed N collected in step S1. d Judge the output result according to the preset requirement. If the predicted speed reaches the performance target, the output result will be passed to the PID network; otherwise, continue with step S21.

[0071] It should be noted that the mean square error prediction value between the true speed and the predicted speed is the mean square error prediction value (MSE) of the LSTM prediction network. MSE is the sum of the squares of the differences between the true value and the predicted value and then averaged, and its range is [0, +∞). It is 0 when the predicted value is exactly the same as the true value. The greater the error, the greater the MSE value. Through iterative testing, it is found that when the maximum error of the predicted speed is within 8%, the approximation effect of the predicted value on the speed is good, and it can truly reflect the external characteristics of the diesel engine.

[0072] According to steps S21 - S22, the diesel engine parameters in the model include: load torque M B , diesel engine speed N d , effective torque M of the diesel engine e , moment of inertia I of the main engine moving parts to the pump impeller of the hydraulic coupling e , friction torque M of the main engine moving parts f1 , fuel supply rack position F r . Based on the above six parameters, a dynamic model of the diesel engine is constructed. Among them, the speed can be expressed by the formula dt is represented. After the diesel engine parameters are input into the LSTM network, they are preprocessed through the memory unit of the LSTM network to remove interference, and then the preprocessed diesel engine parameters are used for training to output a predicted speed that meets the requirements of the preset error range.

[0073] In step S3, the predicted speed value that meets the performance requirements and the speed N d0 (K) at the previous moment are subtracted and input into the PID network together. The proportional-integral-derivative model in the PID network is used to adjust the prediction result to obtain the output result.

[0074] The output of the PID network is expressed by the following formula:

[0075]

[0076] Among them, e(t) is defined as the difference between the speed N d0 (K) at the previous moment and the predicted speed value , and u(t) is the output diesel engine speed.

[0077] According to the characteristics of the positional PID control algorithm, it is necessary to calculate the control quantity based on the deviation at the input sampling moment. At this time, the rotational speed deviation can be expressed as:

[0078]

[0079] where N d0 (k) is the actual rotational speed of the diesel engine in the k-th group of data, is the predicted rotational speed obtained by the LSTM algorithm prediction.

[0080] In contrast, in the traditional PID network, the negative feedback result is directly used to calculate the rotational speed deviation without passing through the LSTM network prediction. The rotational speed deviation therein is expressed by the following formula:

[0081] ΔN d ’(k) = N d0 (k) - N d (k);

[0082] where ΔN d ’(k) represents the rotational speed deviation value in the traditional PID network, N d0 (k) represents the actual rotational speed of the diesel engine in the k-th group of data, N d (k) represents the rotational speed of the diesel engine in the k-th group of data that only passes through the PID negative feedback and does not pass through the LSTM network prediction.

[0083] During actual measurement of the actual rotational speed, measurement errors and other interferences will occur. Since the LSTM algorithm has its own retention algorithm and forgetting algorithm, this characteristic has a very accurate effect on removing interferences. At the same time, the predicted values obtained by predicting a large number of samples will be more accurate than the actual measurement values. Therefore, the partial differential equation constructed by the method provided in this embodiment can better fit the actual working state of the actual diesel engine.

[0084] Before using the PID output form, it is necessary to discretize the PID output form. The discretization sampling method is: using the sampling moment sequence to replace the continuous time, using the sum instead of the integral, and using the increment instead of the differential. The result after the change is as follows:

[0085] t ≈ K(T) (K = 0, 1, 2…); (2)

[0086]

[0087]

[0088] Among them, u(k) is the output value of the PID network at the k-th sampling moment, e(k) is the deviation value input at the k-th sampling moment, e(k - 1) represents the deviation value input at the (k - 1)-th sampling moment, T is the sampling period, K represents the sampling sequence number, and k = 0, 1, 2, ….

[0089] Substituting equations (2)-(4) into (1) gives the calculation formula in the PID network, which is expressed as follows:

[0090]

[0091]

[0092] Among them, u(k) is the output value of the PID network at the k-th sampling moment, e(k) is the deviation value input at the k-th sampling moment, e(k - 1) represents the deviation value input at the (k - 1)-th sampling moment, N d0 (k) is the actual rotational speed of the diesel engine at the k-th sampling moment, is the predicted rotational speed by the LSTM algorithm at the k-th sampling moment, K p 、K I 、K D are all PID network coefficients, K I represents the integral coefficient, K I = K p T / T I ; K D represents the integral coefficient, K D = K p T D / T, T I is the integral constant, T D is the differential time constant.

[0093] In a specific application scenario, first establish the control quasi-steady state model of the diesel engine; on the basis of establishing the model, normalize the data to be used for training the LSTM network; use the LSTM network to train and predict the input data; judge the output result, if the performance target is reached, then transfer the output result to the PID network, otherwise repeat the LSTM network training step until the performance target is reached; subtract the predicted rotational speed value from the rotational speed N d0 (k) at the previous moment and input the difference together into the PID network, and use the proportional-integral-differential model in the PID network to adjust the prediction result to obtain the output result; output the optimal rotational speed after PID control to the diesel engine, and at this time the diesel engine can obtain the optimal working efficiency.

[0094] Example Two:

[0095] In some embodiments, a diesel engine control device with feedback based on the combination of the LSTM algorithm and PID is provided, including:

[0096] An acquisition module for acquiring the required diesel engine parameters;

[0097] A prediction module for inputting the diesel engine parameters into an LSTM network for iterative training until a predicted speed that meets the preset requirements is obtained;

[0098] A PID control module for subtracting the predicted speed that meets the preset requirements from the speed at the previous moment and inputting it into a PID network to obtain the optimal speed for controlling the diesel engine.

[0099] Embodiment 3:

[0100] In some embodiments, a computer-readable storage medium is provided, on which executable instructions are stored. When the instructions are executed by a processor, the processor executes the above method.

[0101] Embodiment 4:

[0102] In some embodiments, an electronic device is provided, including a processor and a storage device. A plurality of instructions are stored in the storage device, and the processor is configured to read the plurality of instructions in the storage device and execute the above method.

[0103] It should be understood that in the embodiments of the present application, the so-called processor may be a central processing unit (CPU), and this processor may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or this processor may also be any conventional processor, etc.

[0104] The memory may include a read-only memory, a flash memory, and a random access memory, and provide instructions and data to the processor. A part or all of the memory may also include a non-volatile random access memory.

[0105] The diesel engine control method and device provided by this embodiment are based on the combination of the LSTM algorithm and PID. The LSTM network is used to replace the traditional feedback link, and the LSTM algorithm is used to predict the rotational speed data that is about to be input into the PID control. Since the LSTM algorithm has a retention algorithm and a forgetting algorithm, it can effectively eliminate interference, thereby eliminating the situation of long-term error accumulation, and can correct the interference data, making the predicted rotational speed difference input into the PID control more accurate than without the LSTM algorithm, and achieving a better control effect.

[0106] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications falling within the scope of the present invention. Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

[0107] It should be understood that if the above integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes of the above embodiment methods in this application, it can also be completed by a computer program instructing related hardware. The above computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above various method embodiments can be implemented. Among them, the above computer program includes computer program code, and the above computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The above computer-readable medium can include: any entity or device that can carry the above computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the above computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0108] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0109] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in connection with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0110] In the embodiments provided in the present application, it should be understood that the disclosed apparatus / terminal device and method can be implemented in other ways. For example, the apparatus / device embodiments described above are merely illustrative. For example, the above-mentioned division of modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0111] The foregoing embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A diesel engine control method combining feedback based on the LSTM algorithm and PID, characterized in that, Including: S1. Collect diesel engine parameters; S2. Input the diesel engine parameters into the LSTM network for iterative training until a predicted speed that meets the preset requirements is obtained; S3. Calculate the difference between the predicted speed that meets the preset requirements and the actual speed at the previous moment and input it into the PID network to obtain the optimal speed; The parameters of the diesel engine include: load torque M B , diesel engine speed N d , effective torque M of the diesel engine e , moment of inertia I of the main engine moving parts to the impeller of the fluid coupling e , friction torque M of the main engine moving parts f1 , fuel supply rack position F r ; The diesel engine parameters are determined according to the control quasi-steady state model, and the specific control quasi-steady state model is as follows:

2. The method according to claim 1, wherein Before iterative training, it also includes: preprocessing the diesel engine parameters.

3. The method according to claim 1, characterized in that, Step S2 includes: S21. Input the diesel engine parameters into the LSTM network for training, and obtain the predicted speed according to the trained LSTM network; S22. Determine whether the predicted speed meets the preset requirements. If it meets, execute step S3; Otherwise, replace the diesel engine speed N in the diesel engine parameters with the predicted speed obtained in step S21 d And repeat step S21 until the obtained predicted speed meets the preset requirements.

4. The method according to claim 1, characterized in that, The preset requirements are: the maximum error of the mean square error prediction value between the actual speed and the predicted speed is not greater than 8%.

5. The method according to claim 1, characterized in that The calculation formula in the PID network is as follows: Among them, u(k) is the output value of the PID network at the k-th sampling moment, e(k) is the deviation value input at the k-th sampling moment, e(k - 1) represents the deviation value input at the (k - 1)-th sampling moment, N d0 (k) is the actual rotational speed of the diesel engine at the k-th sampling moment, is the predicted rotational speed at the k-th sampling moment, K P 、K I 、K D are all integral coefficients in the PID network.

6. A diesel engine control device with feedback based on the combination of the LSTM algorithm and PID, characterized in that, Including: A collection module for collecting diesel engine parameters; A prediction module for inputting the diesel engine parameters into the LSTM network for iterative training until a predicted speed that meets the preset requirements is obtained; A PID control module for calculating the difference between the predicted speed that meets the preset requirements and the speed at the previous moment and inputting it into the PID network to obtain the optimal speed; The parameters of the diesel engine include: load torque M B , diesel engine speed N d , effective torque M of the diesel engine e , moment of inertia I of the main engine moving parts to the impeller of the hydraulic coupling e , friction torque M of the main engine moving parts f1 , fuel supply rack position F r ; The diesel engine parameters are determined according to the control quasi-steady state model, and the specific control quasi-steady state model is as follows:

7. A computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor executes the method according to any one of claims 1-5.

8. An electronic device, comprising a processor and a storage device, characterized in that, Multiple instructions are stored in the storage device, and the processor is used to read the multiple instructions in the storage device and execute the method according to claims 1-5.

Images