Method and system for predicting compression ratio of linear internal combustion power generation system

Abstract

The invention provides a compression ratio prediction method and system of a linear internal combustion power generation system, and relates to the technical field of performance prediction of a novel hybrid power system.The method comprises the steps that a multi-physical field coupling simulation model of the linear internal combustion power generation system is built; extracting a plurality of key design parameters influencing the change of the compression ratio in the stable operation process of the model; determining a numerical value input range of each key design parameter through parameter quantitative analysis, and forming a key design parameter sample set through Latin hypercube sampling; based on the key design parameter sample set, compression ratio responses under different design parameter combinations are obtained through a multi-physics field coupling simulation model; combining the key design parameter sample set and each compression ratio response to form a simulation data set; constructing a compression ratio prediction model based on the hybrid neural network, and performing model optimization according to the simulation data set and an adam optimizer; and carrying out compression ratio prediction through the optimized compression ratio prediction model.

Description

Technical Field

[0001] The present invention relates to the technical field of performance prediction of a new hybrid power system, and particularly to a method and system for predicting the compression ratio of a linear internal combustion power generation system. Background Art

[0002] A linear internal combustion power generation system (Free Piston Linear Generator, FPLG) is a distributed and modular hybrid power system with great application prospects. It mainly consists of a free piston internal combustion engine, a high-performance linear motor, an integrated control system, and an auxiliary system, etc. The FPLG highly integrates and couples the free piston internal combustion engine and the linear motor. Through the continuous and alternating expansion work of the heat engine, the mover of the motor is pushed to reciprocate and cut the magnetic field lines, thereby realizing the continuous output of electricity. Since the piston of the FPLG is not restricted by a mechanism, the piston top dead center position and the compression ratio of the FPLG heat engine are variable, so it can be applicable to various types of fuels such as gasoline and diesel. In addition, the FPLG includes energy storage subsystems such as a storage battery and a supercapacitor, which can decouple the heat engine-linear motor from the electrical load, enabling the system to operate at a working condition with relatively better fuel consumption and emissions.

[0003] Current research mostly focuses on the design and performance optimization of the FPLG, including the control of free piston movement, the regulation of the combustion process, and the efficient coupling between the motor and the heat engine, etc. Existing technologies have adopted various control strategies to adjust the piston position and compression ratio, so as to improve the system efficiency and reduce fuel consumption and emissions. Through the optimization and prediction of different design parameters of the system, the best cooperation between the heat engine-linear motor can be achieved to a certain extent, and the stability and reliability of the FPLG in practical applications can be improved.

[0004] However, the existing methods lack a prediction model for the change of the compression ratio under different design parameters and operating conditions, making it difficult to accurately evaluate and predict the performance of the FPLG system under various operating conditions. Summary of the Invention

[0005] In order to solve the technical problem that the existing methods lack a prediction model for the change of the compression ratio under different design parameters and operating conditions, making it difficult to accurately evaluate and predict the performance of the FPLG system under various operating conditions, the present invention provides a method and system for predicting the compression ratio of a linear internal combustion power generation system.

[0006] The technical solutions provided by the embodiments of the present invention are as follows:

[0007] First aspect:

[0008] A method for predicting the compression ratio of a linear internal combustion power generation system provided by an embodiment of the present invention includes:

[0009] S1: Construct a multi - physical - field coupling simulation model of the linear internal combustion power generation system to describe the stable operation process of the linear internal combustion power generation system;

[0010] S2: Based on the multi - physical - field coupling simulation model, extract multiple key design parameters that affect the change of compression ratio during the stable operation process;

[0011] S3: Through parametric quantitative analysis, determine the numerical input range of each of the key design parameters;

[0012] S4: Through Latin - hypercube sampling, uniformly sample from the numerical input range of each of the key design parameters to form a key design parameter sample set;

[0013] S5: Based on the key design parameter sample set, through the multi - physical - field coupling simulation model, obtain the compression ratio response under different combinations of design parameters during the stable power generation process;

[0014] S6: Combine the key design parameter sample set and each of the compression ratio responses to form a simulation data set;

[0015] S7: Construct a compression ratio prediction model based on a hybrid neural network;

[0016] S8: According to the simulation data set, optimize the compression ratio prediction model through the adam optimizer;

[0017] S9: Obtain real - time key design parameters;

[0018] S10: According to the real - time key design parameters, through the optimized compression ratio prediction model, perform compression ratio prediction.

[0019] Second aspect:

[0020] A compression ratio prediction system for a linear internal combustion power generation system provided by an embodiment of the present invention includes:

[0021] A processor;

[0022] A memory, on which computer - readable instructions are stored, and when the computer - readable instructions are executed by the processor, the compression ratio prediction method for the linear internal combustion power generation system as described in the first aspect is implemented.

[0023] Third aspect:

[0024] A computer - readable storage medium provided by an embodiment of the present invention, on which a computer program is stored, and when the program is executed by a processor, the compression ratio prediction method for the linear internal combustion power generation system as described in the first aspect is implemented.

[0025] The beneficial effects brought by the technical solutions provided in the embodiments of the present invention at least include:

[0026] (1) In the embodiments of the present invention, by constructing a multi-physical-field coupling simulation model, extracting and quantifying key design parameters, and using the Latin hypercube sampling technique to uniformly sample the design parameter range, the compression ratio response under different design conditions is obtained, thereby effectively improving the accuracy of compression ratio prediction. By constructing a compression ratio prediction model with a hybrid neural network and using the Adam optimizer to optimize the compression ratio prediction model, the accuracy of compression ratio prediction is further improved, enabling the performance of the FPLG system under various operating conditions to be accurately evaluated and predicted. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0028] Figure 1 It is a schematic flow chart of a method for predicting the compression ratio of a linear internal combustion power generation system provided in the embodiments of the present invention;

[0029] Figure 2 It is a schematic structural diagram of a system for predicting the compression ratio of a linear internal combustion power generation system provided in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The following will describe the technical solutions in the present invention with reference to the drawings.

[0031] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of the word "example" aims to present concepts in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0032] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meaning they express is the same. "(of)", "corresponding", and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meaning they express is the same.

[0033] In the embodiments of the present invention, sometimes subscripts such as W 1 may be written in a non-subscript form such as W1. When the difference is not emphasized, their intended meanings are the same.

[0034] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0035] Referring to the attached Figure 1 figures, a schematic flow diagram of a method for predicting the compression ratio of a linear internal combustion power generation system provided by an embodiment of the present invention is shown.

[0036] An embodiment of the present invention provides a method for predicting the compression ratio of a linear internal combustion power generation system. This method can be implemented by a compression ratio prediction device of the linear internal combustion power generation system, and the compression ratio prediction device of the linear internal combustion power generation system can be a terminal or a server. The processing flow of the method for predicting the compression ratio of the linear internal combustion power generation system can include the following steps:

[0037] S1: Construct a multi-physics field coupling simulation model of the linear internal combustion power generation system to describe the stable operation process of the linear internal combustion power generation system.

[0038] In a possible implementation manner, S1 is specifically:

[0039] Based on the Matlab mathematical calculation software and the Simulink graphical modeling tool, construct a multi-physics field coupling simulation model of the linear internal combustion power generation system.

[0040] Among them, Matlab is a highly efficient mathematical calculation software widely used in mathematical modeling, data analysis, algorithm development, signal processing, image processing, control system design, robotics, and financial analysis fields.

[0041] Among them, Simulink is a graphical modeling and simulation tool based on Matlab, widely used in system modeling, simulation, control design, signal processing, communication, mechanical systems, power systems and other fields. Simulink provides a graphical interface where users can build system models by dragging icons and connecting modules without writing a large amount of code. It is especially suitable for the modeling and simulation of dynamic systems, especially the modeling and optimization of multi-field systems such as control systems, electronic systems, and mechanical systems.

[0042] In a possible implementation manner, the multi-physics field coupling simulation model includes: a dynamics model and a thermodynamics model.

[0043] Among them, the kinetic model is mainly used to describe the motion laws of various components (such as pistons, motors, etc.) in the system and helps us understand how these components interact with each other. It focuses on the relationships among force, motion, and acceleration.

[0044] Among them, the thermodynamic model is mainly used to describe the energy conversion process in the system, especially the conversion between thermal energy and mechanical energy. In an internal combustion power generation system, the compression and expansion processes of the gas involve the mutual conversion of thermal energy (from the combustion process and the heat exchange between the gas and the cylinder wall) and mechanical energy (the force exerted by the gas on the piston).

[0045] In a possible implementation, the kinetic model is specifically:

[0046]

[0047] Among them, F p_left represents the force of the gas in the left cylinder, F p_right represents the force of the gas in the right cylinder, F e represents the electromagnetic resistance of the linear motor, F f represents the system frictional force, x represents the displacement of the moving component, t represents time, p represents the pressure in the cylinder, A represents the piston bottom area, F p represents the force of the gas in the cylinder.

[0048] In a possible implementation, the thermodynamic model is specifically:

[0049]

[0050] Among them, P represents the current pressure in the cylinder, t represents time, γ represents the adiabatic coefficient, Q c represents the heat released during the combustion process, Q ht represents the heat lost due to heat transfer between the gas in the cylinder and the cylinder wall, V represents the current volume in the cylinder, i represents the intake process, e represents the exhaust process, l represents the leakage process of the gas through the piston ring, m n represents the gas mass during the intake process or the exhaust process, m l represents the leaked gas mass, m air represents the gas mass in the current cylinder, LHV represents the lower heating value of the fuel, AFR represents the air-fuel ratio, a, b represent empirical parameters, t c represents the combustion duration, t 0 represents the start time of combustion, exp represents the exponential function, T represents the current temperature in the cylinder, represents the average speed of the piston moving component, A cyl represents the area in contact with the high-temperature gas, T w represents the wall reference temperature, m jRepresents the in-cylinder gas mass involved in the intake process, exhaust process, or leakage process, C d Represents the flow coefficient, A d Represents the reference area of the fluid, P u Represents the gas pressure on the high-pressure side, R represents the gas constant, T u Represents the high-pressure side gas temperature, P d Represents the gas pressure on the low-pressure side.

[0051] In the present invention, through the powerful tools provided by Matlab and Simulink, the dynamic response and heat transfer process of the internal combustion power generation system can be accurately simulated. Especially in complex nonlinear systems, these tools can effectively capture the interactions of different physical processes. At the same time, by establishing a multi-physics field coupling simulation model, detailed system performance analysis can be carried out at the design stage, predicting the performance of the system under different working conditions, and then helping to optimize the design, improving the system efficiency, stability, and reliability.

[0052] Furthermore, the combined use of the thermodynamic model and the kinetic model can help analyze the response of the system under different working conditions and provide data support for the design of the control system.

[0053] S2: Based on the multi-physics field coupling simulation model, extract multiple key design parameters that affect the change of the compression ratio during the stable operation process.

[0054] Optionally, the key design parameters specifically include: air-fuel ratio, ignition position, intake pressure, starting compression ratio, electromagnetic damping coefficient, and the mass of the moving components.

[0055] It should be noted that the key design parameters should include both engine design parameters and linear motor design parameters, and the change of the variable values needs to have an obvious impact on the compression ratio response.

[0056] In the present invention, by focusing on the key design parameters, the design optimization process can be more accurately simulated, and then a more precise prediction of the compression ratio change can be made. At the same time, clarifying which design parameters have a significant impact on the compression ratio can provide a basis for the design of the control system. Thus, during the actual operation process, according to the changes of these key parameters, the control strategy can be adjusted, such as by automatically adjusting the ignition position or intake pressure, so that the system is always in the best working state.

[0057] S3: Through parametric quantitative analysis, determine the numerical input range of each key design parameter.

[0058] Among them, parameter quantification analysis is a process used to analyze and determine the impact of system parameters on system behavior or performance within a specific range. It analyzes the numerical values of each key design parameter in the system to determine its reasonable value range to ensure the stable and efficient operation of the system.

[0059] It should be noted that in order to ensure the stable power generation of the FPLG system during actual operation and not exceed the instability critical point of the system, when determining the value range of the numerical values of the key design parameters, it should be ensured that under this set of key design parameters, the energy generated by the thermal cycle should match the system load, neither causing insufficient energy generated by the thermal cycle to prevent the system from overcoming the motor load and frictional load to reach the preset ignition position, resulting in system shutdown, nor causing excessive energy generated by the thermal cycle to lead to phenomena such as detonation, misfire, and cylinder knocking.

[0060] In the present invention, through a reasonable design parameter range, it is ensured that the energy generated by the thermal cycle matches the system load, avoiding system overload or insufficient load, thereby ensuring the stable operation of the system.

[0061] S4: Through Latin hypercube sampling, uniform sampling is carried out from the numerical input ranges of each key design parameter to form a key design parameter sample set.

[0062] Among them, Latin hypercube sampling (LHS) is a probability sampling method commonly used in high-dimensional parameter spaces, especially suitable for fields such as uncertainty analysis, Monte Carlo simulation, and optimization. Its main advantage lies in being able to sample uniformly in a multi-dimensional space, thereby effectively covering the entire design space and avoiding the uneven distribution of data that may be caused by traditional random sampling methods.

[0063] Specifically, the multi-dimensional key design parameters are stratified sampled through Latin hypercube sampling, so that the collected samples are evenly distributed in the input vector space and the arrangement order is random.

[0064] In the present invention, through stratified sampling within each dimension, each interval can represent a sample, thereby avoiding the lack of samples in some areas and making the sampling more comprehensive. At the same time, Latin hypercube sampling can explore the entire design space more comprehensively with a given number of samples, thereby reducing the number of samples and avoiding a large amount of unnecessary calculations. This significantly improves the calculation efficiency when performing simulation and optimization in a high-dimensional space.

[0065] S5: Based on the key design parameter sample set, through a multi-physics field coupling simulation model, obtain the compression ratio response under different design parameter combinations during stable power generation.

[0066] Among them, the compression ratio response describes the dynamic response or variation law of the system's compression ratio with respect to changes in design parameters (such as air-fuel ratio, ignition position, intake pressure, etc.). In the multi-physics field coupling simulation model, the compression ratio response is obtained through simulation calculations, which helps us understand and predict the behavior of the compression ratio under different operating conditions.

[0067] In a possible implementation, the calculation formula for the compression ratio response is specifically:

[0068]

[0069] Among them, CR L represents the compression ratio of the left cylinder, V 0 represents the total volume of the cylinder, A represents the piston bottom area, L represents the half stroke of the cylinder starting from the midpoint of the designed stroke, x c represents the cylinder length corresponding to the remaining volume, x TDC represents the top dead center, CR R represents the compression ratio of the right cylinder, x BDC represents the bottom dead center.

[0070] Optionally, the identification method for the top dead center x TDC and the bottom dead center x BDC is as follows:

[0071] After the start of the stable power generation working process of the multi-physics field coupling simulation model, calculate according to the real-time position, movement speed and direction of the moving components. When the system is in the compression process and the speed becomes 0, record the position of the moving components at this time as the top dead center x TDC ; when the system is in the expansion process and the speed becomes 0, record the position of the moving components at this time as the bottom dead center x BDC .

[0072] In the present invention, by simulating and calculating the positions of the top dead center and the bottom dead center, the compression ratio can be calculated more accurately, thereby ensuring a higher matching degree between the simulation results and the actual working conditions. At the same time, since the top dead center and the bottom dead center of the piston are affected by multiple factors such as combustion pressure, intake pressure, load, etc., by calculating these parameters in real time, more accurate data can be provided for the control system, helping to optimize the combustion process, reduce energy loss, and improve system stability.

[0073] S6: Combine the key design parameter sample set and each compression ratio response to form a simulation data set.

[0074] It should be noted that the simulated data set is shaped into a seven-dimensional shape, including six key design parameter inputs: air-fuel ratio, ignition position, intake pressure, starting compression ratio, electromagnetic damping coefficient, mass of moving components, and one output of the compression ratio during the stable operation process. The data is saved as an Excel file, with the first six columns as inputs and the seventh column as the output, and the number of rows is the sample size.

[0075] S7: Construct a compression ratio prediction model based on a hybrid neural network.

[0076] In a possible implementation manner, the compression ratio prediction model based on the hybrid neural network in S7 specifically includes: a CNN layer, an unfolding layer, a smoothing layer, an LSTM layer, and a fully connected layer.

[0077] The CNN layer also includes: a convolutional layer, an activation layer, and a pooling layer.

[0078] In a possible implementation manner, S7 specifically includes:

[0079] S701: Randomly shuffle the simulated data set and divide it into a training data set and a test data set according to a preset ratio.

[0080] S702: Perform normalization processing on the training data set and the test data set:

[0081]

[0082] Among them, y represents the normalized data, x represents the original data, y max represents the maximum value of the target normalization, y min represents the minimum value of the target normalization, x max represents the maximum value of the original data, x min represents the minimum value of the original data.

[0083] S703: Take the normalized training data set as the input, perform a convolution operation through the convolutional layer, and obtain the first target feature map.

[0084] S704: Through the activation layer, perform an activation operation on the first target feature map.

[0085] S705: Input the first target feature map after the activation operation into the pooling layer, and reduce the size of the first target feature map after the activation operation by taking the average value to obtain the second target feature map.

[0086] In the present invention, an average pooling layer is adopted to reduce the size of the feature map, making it easier for the model to learn global information, while reducing the amount of calculation and improving the training speed.

[0087] S706: Use the expansion layer to expand the second target feature map into a form that can be serialized and processed, and expand it into a one-dimensional vector through the smoothing layer.

[0088] S707: Input the one-dimensional vector into the LSTM layer to extract the time-dependent features of the one-dimensional vector.

[0089] S708: Input the time-dependent features into the fully connected layer, output the compression ratio prediction result, and complete the construction of the compression ratio prediction model.

[0090] Specifically, the hybrid neural network is CNN-LSTM. The CNN layer includes an input layer, a folding layer, a convolutional layer, an activation layer, and an average pooling layer. The number of filters in the convolutional layer is 40, and the ReLU activation function is used in the activation layer; the pooling layer reduces the size of the feature map by taking the average value, making the model more tolerant to changes in local patterns in the feature map; between the CNN layer and the LSTM layer, the convolutional output is expanded into a form suitable for serialized processing through the expansion layer; the smoothing layer is used to expand the high-dimensional feature map into a one-dimensional vector; it includes two LSTM layers. The first LSTM layer contains 50 hidden units, and the second LSTM layer contains 58 hidden units; a Dropout layer is added to the output layer to prevent overfitting by randomly discarding 20% of the neurons.

[0091] In the present invention, through the convolutional operation of 40 filters, the CNN can effectively capture the local change trend of the compression ratio without the need for artificial feature design. At the same time, the CNN is responsible for local feature learning, and the LSTM is responsible for extracting time-dependent relationships, which can effectively improve the compression ratio prediction accuracy and provide reliable support for optimizing engine parameter control, improving combustion efficiency, and enhancing system stability and energy efficiency.

[0092] S8: Optimize the compression ratio prediction model according to the simulation dataset through the adam optimizer.

[0093] In a possible implementation manner, S8 specifically includes:

[0094] S801: Determine the mean square error loss function of the compression ratio prediction model:

[0095]

[0096] where RMSE represents the mean square error loss function, represents the predicted value of the compression ratio prediction model, and y i represents the true value of the compression ratio obtained by simulation, and n represents the number of test samples.

[0097] S802: Input the normalized test dataset into the compression ratio prediction model and perform optimization processing through the Adam optimizer.

[0098] S803: Stop the optimization when the function value of the loss function is less than the preset function value, and obtain the optimized compression ratio prediction model.

[0099] It should be noted that those skilled in the art can set the size of the preset function value according to actual needs, and the present invention does not limit it here.

[0100] In the present invention, RMSE amplifies larger errors through squaring, is more sensitive to samples with larger prediction errors, which helps the optimizer adjust the model weights faster and make it more accurately fit the change trend of the compression ratio. At the same time, since Adam calculates gradients more efficiently, it usually converges faster than the traditional SGD (Stochastic Gradient Descent) method and can achieve better optimization effects under the same training time.

[0101] S9: Obtain real-time key design parameters.

[0102] S10: Predict the compression ratio according to the real-time key design parameters through the optimized compression ratio prediction model.

[0103] The beneficial effects brought by the technical solution provided by the embodiment of the present invention at least include:

[0104] (1) In the embodiment of the present invention, by constructing a multi-physical field coupling simulation model, extracting and quantifying key design parameters, and using the Latin hypercube sampling technique to uniformly sample the design parameter range, the compression ratio response under different design conditions is obtained, effectively improving the accuracy of compression ratio prediction. By constructing a compression ratio prediction model through a hybrid neural network and using the Adam optimizer to optimize the compression ratio prediction model, the accuracy of compression ratio prediction is further improved, so that the performance of the FPLG system under various operating conditions can be accurately evaluated and predicted.

[0105] Refer to the attached Figure 2 illustrates the structural schematic diagram of a compression ratio prediction system for a linear internal combustion power generation system provided by the present invention.

[0106] The present invention also provides a compression ratio prediction system 20 for a linear internal combustion power generation system, which is applied to the compression ratio prediction method of the above-mentioned linear internal combustion power generation system, including:

[0107] A processor 201.

[0108] A memory 202, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor 201, the compression ratio prediction method of the linear internal combustion power generation system as in the method embodiment is implemented.

[0109] The compression ratio prediction system 20 of the linear internal combustion power generation system provided by the present invention can execute the compression ratio prediction method of the above-mentioned linear internal combustion power generation system and achieve the same or similar technical effects. To avoid repetition, the present invention will not be described in detail herein.

[0110] The beneficial effects brought by the technical solutions provided in the embodiments of the present invention at least include:

[0111] (1) In the embodiments of the present invention, by constructing a multi-physical field coupling simulation model, extracting and quantifying key design parameters, and using the Latin hypercube sampling technique to uniformly sample the design parameter range, the compression ratio response under different design conditions is obtained, thereby effectively improving the accuracy of compression ratio prediction. A compression ratio prediction model is constructed by a hybrid neural network, and the compression ratio prediction model is optimized by an Adam optimizer, further improving the accuracy of compression ratio prediction, so that the performance of the FPLG system under various operating conditions can be accurately evaluated and predicted.

[0112] It should be understood that the processor in the embodiments of the present invention may be a central processing unit (CPU), and the 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 the processor may also be any conventional processor, etc.

[0113] It should also be understood that the memory in the embodiments of the present invention can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0114] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0115] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be understood specifically by referring to the context.

[0116] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or plural.

[0117] It should be understood that in various embodiments of the present invention, the magnitudes of the sequence numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0118] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination 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. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0119] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices, apparatuses, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0120] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0121] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] In addition, the functional units in each embodiment of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0123] If the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0124] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the compression ratio prediction method of the linear internal combustion power generation system as described in the method embodiment.

[0125] The computer-readable storage medium provided by the present invention can implement the steps and effects of the compression ratio prediction method of the linear internal combustion power generation system in the above method embodiment. To avoid repetition, the present invention will not elaborate further.

[0126] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0127] (1) In the embodiment of the present invention, by constructing a multi-physical field coupling simulation model, extracting and quantifying key design parameters, and using the Latin hypercube sampling technique to uniformly sample the design parameter range, the compression ratio response under different design conditions is obtained, thereby effectively improving the accuracy of compression ratio prediction. By constructing a compression ratio prediction model through a hybrid neural network and using the Adam optimizer to optimize the compression ratio prediction model, the accuracy of compression ratio prediction is further improved, enabling the performance of the FPLG system under various operating conditions to be accurately evaluated and predicted.

[0128] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

[0129] The following points need to be explained:

[0130] (1) The drawings in the embodiments of the present invention only relate to the structures involved in the embodiments of the present invention, and other structures can refer to the usual designs.

[0131] (2) For clarity, in the drawings used to describe the embodiments of the present invention, the thickness of layers or regions is enlarged or reduced, that is, these drawings are not drawn to actual scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element can be "directly" on or under the other element or there can be intervening elements.

[0132] (3) Without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0133] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for predicting the compression ratio of a linear internal combustion power generation system, characterized in that: include: S1: construct a multi-physics field coupling simulation model of a linear internal combustion power generation system to describe a stable working process of the linear internal combustion power generation system; S2: extracting multiple key design parameters that affect the change of compression ratio during stable operation based on the multi-physics field coupling simulation model; S3: Determine the numerical input range of each of the key design parameters through parameter quantitative analysis; S4: uniformly sampling from the numerical input range of each of the key design parameters through Latin hypercube sampling to form a key design parameter sample set; S5: Based on the key design parameter sample set, the compression ratio response under different design parameter combinations in a stable power generation process is obtained through the multi-physics field coupling simulation model; S6: combining the key design parameter sample set and each of the compression ratio responses to form a simulation data set; S7: Construct a compression ratio prediction model based on hybrid neural network; S8: Optimizing the compression ratio prediction model through an adam optimizer according to the simulated data set; S9: Obtain real-time key design parameters; S10: According to the real-time key design parameters, compression ratio prediction is performed using an optimized compression ratio prediction model.

2. The compression ratio prediction method of the linear internal combustion power generation system according to claim 1, characterized in that: The S1 is specifically: Based on Matlab mathematical calculation software and Simulink graphical modeling tool, the multi-physical field coupling simulation model of the linear internal combustion power generation system is constructed.

3. The compression ratio prediction method of the linear internal combustion power generation system according to claim 2, characterized in that: The multi-physics field coupling simulation model includes: a kinetic model and a thermodynamic model.

4. The compression ratio prediction method of the linear internal combustion power generation system according to claim 3, characterized in that: The kinetic model is specifically: Among them, F p_left Indicates the force of the gas in the left cylinder, F p_right Indicates the force of the gas in the right cylinder, F e Indicates the electromagnetic resistance of the linear motor, F f represents the friction of the system, x represents the displacement of the moving component, t represents the time, p represents the pressure in the cylinder, A represents the bottom area of ​​the piston, and F p Indicates the force of the gas in the cylinder.

5. The compression ratio prediction method of the linear internal combustion power generation system according to claim 3, characterized in that: The thermodynamic model is specifically: Among them, P represents the current cylinder pressure, t represents time, γ represents the adiabatic coefficient, Q c Indicates the heat released during the combustion process, Q ht represents the heat lost by heat transfer between the gas in the cylinder and the cylinder wall, V represents the current volume in the cylinder, i represents the intake process, e represents the exhaust process, l represents the leakage process of the gas through the piston ring, and m n Indicates the gas mass during the intake or exhaust process, m l Indicates the leaked gas mass, m air represents the gas mass in the current cylinder, LHV represents the lower heating value of the fuel, AFR represents the air-fuel ratio, a and b represent empirical parameters, t c represents the duration of combustion, t0 represents the start time of combustion, exp represents the exponential function, T represents the current cylinder temperature, represents the average speed of the piston moving component, A cyl Indicates the area of ​​high temperature gas contact, T w represents the wall reference temperature, m j Indicates the mass of the in-cylinder gas involved in the intake process, exhaust process or leakage process, C d Indicates the flow coefficient, A d represents the reference area of ​​the fluid, P u represents the gas pressure on the high pressure side, R represents the gas constant, T u Indicates the high pressure side gas temperature, P d Indicates the gas pressure on the low-pressure side.

6. The compression ratio prediction method of a linear internal combustion power generation system according to claim 1, characterized in that: The calculation formula of the compression ratio response is specifically: Among them, CR L represents the compression ratio of the left cylinder, V0 represents the total volume of the cylinder, A represents the piston bottom area, L represents the half stroke of the cylinder starting from the midpoint of the design stroke, x c represents the length of the cylinder corresponding to the remaining volume, x TDC Indicates top dead center, CR R represents the compression ratio of the right cylinder, x BDC Indicates bottom dead center.

7. The compression ratio prediction method of a linear internal combustion power generation system according to claim 1, characterized in that: The compression ratio prediction model based on the hybrid neural network in S7 specifically includes: a CNN layer, an expansion layer, a smoothing layer, an LSTM layer and a fully connected layer; The CNN layer also includes: a convolutional layer, an activation layer and a pooling layer.

8. The compression ratio prediction method of the linear internal combustion power generation system according to claim 7, characterized in that: The S7 specifically includes: S701: Randomly shuffle the simulated data set, and divide the training data set and the test data set into a preset ratio; S702: Normalize the training data set and the test data set: Among them, y represents the normalized data, x represents the original data, and y max represents the maximum value of target normalization, y min represents the minimum value of target normalization, x max Indicates the maximum value of the original data, x min Indicates the minimum value of the original data; S703: taking the normalized training data set as input, performing a convolution operation through the convolution layer to obtain a first target feature map; S704: performing an activation operation on the first target feature map through the activation layer; S705: Inputting the first target feature map after the activation operation into the pooling layer, and reducing the size of the first target feature map after the activation operation by taking an average value to obtain a second target feature map; S706: using the expansion layer, expanding the second target feature map into a form that can be processed serially, and expanding it into a one-dimensional vector through the smoothing layer; S707: Input the one-dimensional vector into the LSTM layer to extract the time-dependent features of the one-dimensional vector; S708: Input the time-dependent feature into the fully connected layer, output the compression ratio prediction result, and complete the construction of the compression ratio prediction model.

9. The compression ratio prediction method of the linear internal combustion power generation system according to claim 8, characterized in that: The S8 specifically includes: S801: Determine the mean square error loss function of the compression ratio prediction model: Among them, RMSE represents the mean square error loss function, represents the predicted value of the compression ratio prediction model, y i represents the true value of the compression ratio obtained by simulation, and n represents the number of test samples; S802: Input the normalized test data set into the compression ratio prediction model, and perform optimization processing through the Adam optimizer; S803: When the function value of the loss function is less than a preset function value, stop the optimization to obtain an optimized compression ratio prediction model.

10. A compression ratio prediction system for a linear internal combustion power generation system, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, a compression ratio prediction method for a linear internal combustion power generation system as claimed in any one of claims 1 to 9 is implemented.

Images