Method for evaluating degradation state of vehicle power system based on thermal efficiency

Abstract

The invention belongs to the technical field of vehicle power systems, and particularly relates to a vehicle power system degradation state evaluation method based on heat efficiency, which comprises the following steps: establishing a power system heat flow formation and dissipation mechanism model, and determining a power system heat efficiency characterization parameter set; establishing a correlation analysis model, and optimizing a thermal efficiency characterization parameter set of the power system; establishing a CNN-LSTM thermal efficiency prediction neural network model; analyzing the test parameter values of the health power system, and establishing reference thermal efficiency; analyzing the test parameter value of the degraded power system, and determining the current thermal efficiency; and comparing the current thermal efficiency with the reference thermal efficiency to determine the system degradation state. According to the method, a heat flow forming and dissipation mechanism of the vehicle power system is analyzed, correlation analysis is combined, a representation parameter set is established, the difference between the heat efficiency of a degraded power system and the heat efficiency of a healthy power system is analyzed through a neural network model, and the degradation state of the vehicle power system is determined.

Description

Technical Field

[0001] This invention belongs to the technical field of vehicle power systems, specifically a method for assessing the degradation state of a vehicle power system based on thermal efficiency. This method analyzes the heat flow formation and dissipation mechanisms of the vehicle power system to establish a set of thermal efficiency characterization parameters. This set of characterization parameters is then optimized through correlation analysis. A neural network model is used to establish a thermal efficiency benchmark for typical evaluation conditions. The thermal efficiency of the degraded power system is then compared with the benchmark thermal efficiency to determine the degradation state of the vehicle power system. This method can be applied to assessing the degradation state of any fuel-powered vehicle power system. Background Art

[0002] With the continuous advancement of vehicle intelligence and the influx of electronic information products, vehicles have become increasingly high-tech products, visible yet intangible. Monitoring powertrain performance degradation has become a major bottleneck in troubleshooting potential vehicle faults and improving operational safety. Powertrain performance prediction currently focuses on component-level testing. Component-level testing primarily operates at a microscopic level, utilizing component control and detection information for diagnosis. For example, in the engine, this primarily involves collecting sensor-based test parameters such as engine speed, oil pressure, oil temperature, intake air temperature, and output torque for fault diagnosis and status prediction. This diagnostic approach suffers from several critical flaws: The vehicle powertrain status prediction relies on isolated diagnosis, failing to consider inter-system coupling. It only monitors and diagnoses the operating status and performance of a single component, ignoring the interplay of performance degradation across different components. Furthermore, the performance status assessment derived from a single powertrain component health test yields limited information and lacks diagnostic information. This can easily lead to inaccurate fault diagnosis and status prediction in complex fault mechanisms. In order to realize the gradual transition from component-level diagnosis based on the micro level to system-level diagnosis based on the macro-global level, research on system-level degradation state assessment methods based on thermal efficiency is carried out, which is of great significance to improving the degradation state assessment capabilities of vehicle power systems. Summary of the Invention

[0003] (1) Technical issues to be solved

[0004] The technical problem to be solved by the present invention is: how to provide a vehicle power system degradation state assessment method based on thermal efficiency, which can be applied to the degradation state assessment of any fuel vehicle power system, realize the system-level degradation state assessment of the vehicle power system, and improve the vehicle operation state monitoring capability.

[0005] (2) Technical solution

[0006] To solve the above technical problems, the present invention provides a method for evaluating the degradation state of a vehicle power system based on thermal efficiency, the method comprising the following steps:

[0007] Step S1: Establish a model of the heat flow formation and dissipation mechanism of the power system and determine a parameter set representing the thermal efficiency of the power system;

[0008] Step S2: Establish a correlation analysis model to optimize the power system thermal efficiency characterization parameter set;

[0009] Step S3: Establishing a CNN-LSTM thermal efficiency prediction neural network model;

[0010] Step S4: Analyze the test parameter values ​​of the healthy power system and establish a baseline thermal efficiency;

[0011] Step S5: Analyze the test parameter values ​​of the degraded power system to determine the current thermal efficiency;

[0012] Step S6: Compare the current thermal efficiency with the benchmark thermal efficiency to determine the system degradation state.

[0013] Wherein, in said step S1, a power system heat flow formation and dissipation mechanism model is established to determine a power system thermal efficiency characterization parameter set;

[0014] During the heat flow conversion and dissipation process of the power system, the total heat released by the engine burning fuel flows out through three channels: effective output work, engine heat, and exhaust heat. The total heat and effective output work, engine heat, and exhaust heat have energy conservation characteristics.

[0015] 1) Principle of Conservation of Energy

[0016] The principle of conservation of energy is expressed in the following equation:

[0017] Q fuel =P e +Q S +Q R

[0018] Where Q fuel To burn calories, P e is the effective output power, Q S is the body heat, Q R is the exhaust heat;

[0019] 2) The thermodynamic model of fuel combustion heat is shown as follows:

[0020]

[0021] Where, is the fuel mass flow rate, Hv is the lower heating value of the fuel;

[0022] 3) The thermodynamic model of effective output work is shown as follows:

[0023] Pe =T tq ×n / 9550

[0024] Where, T tq is the engine torque, n is the engine speed;

[0025] 4) The body's thermal thermodynamic model is shown below:

[0026]

[0027] Where, is the coolant mass flow rate, c s is the specific heat capacity of the cooling medium, T1 and T2 are the inlet and outlet temperatures of the cooling liquid respectively;

[0028] 5) The exhaust heat thermodynamic model is shown as follows:

[0029]

[0030] Where, is the exhaust mass flow rate, c pr is the constant pressure specific heat capacity of exhaust gas, c pk is the specific heat capacity of air at constant pressure, T r is the engine exhaust temperature, T k is the engine intake air temperature;

[0031] 6) The thermodynamic model of power system thermal efficiency is shown as follows:

[0032]

[0033] 7) Determination of parameters characterizing thermal efficiency of power system

[0034] From the thermal efficiency thermodynamic model, we can see that the thermal efficiency of the power system is a comprehensive evaluation indicator, which is affected by multiple variable parameters. The relationship between these influencing factors and thermal efficiency is expressed by the following general function:

[0035]

[0036] Based on the general function expression of the thermal efficiency of the power system, it is preliminarily determined that the parameter set characterizing the thermal efficiency of the power system is composed of eight characterizing parameters: oil quantity, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature, and exhaust flow rate.

[0037] Wherein, in said step S2, a correlation analysis model is established to optimize the thermal efficiency characterization parameter set;

[0038] Clean the test data, extract the characterization parameters, select the characterization parameters with strong correlation with thermal efficiency, and establish the optimal characterization parameter set;

[0039] 1) Acquisition of health status test data: Conduct characteristic tests on healthy power systems to obtain health status test data;

[0040] 2) Thermal efficiency characterization parameter extraction: The collected test data is cleaned, denoised, and other data preprocessing is performed to extract eight characterization parameters: oil volume, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature, and exhaust flow rate;

[0041] 3) Correlation analysis and calculation: With eight characterization parameters (torque, circulating oil volume, intake temperature, exhaust temperature, speed, coolant temperature, intake pressure, and oil temperature) as input and the power system thermal efficiency as output, the Pearson correlation analysis algorithm is used to perform a correlation analysis between the eight characterization parameters and the power system thermal efficiency. The Pearson correlation coefficient calculation formula is as follows:

[0042]

[0043] In the formula, r is the Pearson correlation coefficient, n is the number of samples, and are the average values ​​of the two samples, σ X and σ Y X i and Y i Standard score of

[0044] 4) Determination of the optimized characterization parameter set for thermal flow characteristics: Based on the correspondence between the Pearson correlation coefficient value and the degree of correlation, with |r| ≥ 0.4, the four characterization parameters of torque, circulating oil volume, intake temperature, and exhaust temperature are selected to form the optimal thermal efficiency characterization parameter set for power system thermal efficiency evaluation.

[0045] Wherein, in said step S3, a CNN-LSTM thermal efficiency prediction neural network model is established;

[0046] The CNN-LSTM thermal efficiency prediction neural network model takes the test data as input, the optimized thermal efficiency characterization parameter set as the feature vector, and the thermal efficiency value as output to achieve data-based thermal efficiency prediction;

[0047] 1) Input layer establishment: Create time series data consisting of four parameters: fuel volume, torque, intake temperature, and exhaust temperature. Each set of series represents a different heat flow characterization parameter of the power system;

[0048] 2) CNN layer establishment: Convolutional neural network is established. The convolutional layer uses one-dimensional convolution operation (1D-CNN) to extract parameter features in each time series. Max pooling is used to reduce data dimensions, extract key features, and improve computational efficiency.

[0049] 3) LSTM layer establishment: Establish an LSTM layer to capture the long-range dependencies and dynamic change trends of time series and analyze time series characteristics;

[0050] 4) Fully connected layer: A fully connected layer is established to aggregate the extracted features to ultimately generate the thermal efficiency prediction results;

[0051] 5) Output layer: Output the predicted value of thermal efficiency of the power system.

[0052] Wherein, in said step S4, the healthy power system test parameter values ​​are analyzed to establish a benchmark thermal efficiency;

[0053] The experimental test data of the healthy power system is input into the CNN-LSTM thermal efficiency prediction neural network model to obtain the long-term series thermal efficiency benchmark value under the healthy state.

[0054] Wherein, in said step S5, the test parameter value of the degraded power system is analyzed to determine the current thermal efficiency;

[0055] The experimental test data of the degraded power system is input into the CNN-LSTM thermal efficiency prediction neural network model to obtain the long-term series thermal efficiency evaluation value under the degraded state.

[0056] Wherein, in said step S6, the current thermal efficiency is compared with the reference thermal efficiency to determine the degradation state of the system;

[0057] In the same time dimension, the difference between the long-term series thermal efficiency evaluation value in the degraded state and the long-term series thermal efficiency benchmark value in the healthy state is compared to evaluate the system degradation state. The system degradation state evaluation algorithm is as follows:

[0058]

[0059] Where n is the number of test data points, η i is the thermal efficiency evaluation value of point i, η i0 is the benchmark value for thermal efficiency evaluation at point i.

[0060] (3) Beneficial effects

[0061] Compared with the prior art, the present invention has the following effects:

[0062] This vehicle powertrain degradation assessment method analyzes the heat flow formation and dissipation mechanisms of the vehicle powertrain, combines this with correlation analysis to establish a characterization parameter set. Using a neural network model, it analyzes the difference in thermal efficiency between a degraded powertrain and a healthy one, determining the degradation state of the vehicle powertrain. This method can be applied to assess the degradation state of any fuel-powered vehicle powertrain. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flow chart of the power system degradation state assessment based on thermal efficiency according to the present invention;

[0064] Figure 2 This is a flow chart of the method for optimizing the thermal efficiency characterization parameters of a power system according to the present invention;

[0065] Figure 3 Schematic diagram of the thermal efficiency evaluation neural network model structure of the present invention. DETAILED DESCRIPTION

[0066] In order to make the purpose, content, and advantages of the present invention more clear, the specific implementation methods of the present invention are further described in detail below with reference to the accompanying drawings and examples.

[0067] To solve the above technical problems, the present invention provides a method for evaluating the degradation state of a vehicle power system based on thermal efficiency, the method comprising the following steps:

[0068] Step S1: Establish a model of the heat flow formation and dissipation mechanism of the power system and determine a parameter set representing the thermal efficiency of the power system;

[0069] Step S2: Establish a correlation analysis model to optimize the power system thermal efficiency characterization parameter set;

[0070] Step S3: Establishing a CNN-LSTM thermal efficiency prediction neural network model;

[0071] Step S4: Analyze the test parameter values ​​of the healthy power system and establish a baseline thermal efficiency;

[0072] Step S5: Analyze the test parameter values ​​of the degraded power system to determine the current thermal efficiency;

[0073] Step S6: Compare the current thermal efficiency with the benchmark thermal efficiency to determine the system degradation state.

[0074] Wherein, in said step S1, a power system heat flow formation and dissipation mechanism model is established to determine a power system thermal efficiency characterization parameter set;

[0075] During the heat flow conversion and dissipation process of the power system, the total heat released by the engine burning fuel flows out through three channels: effective output work, engine heat, and exhaust heat. The total heat and effective output work, engine heat, and exhaust heat have energy conservation characteristics.

[0076] 1) Principle of Conservation of Energy

[0077] The principle of conservation of energy is expressed in the following equation:

[0078] Q fuel =P e +Q S +QR

[0079] Where Q fuel To burn calories, P e is the effective output power, Q S is the body heat, Q R is the exhaust heat;

[0080] 2) The thermodynamic model of fuel combustion heat is shown as follows:

[0081]

[0082] Where, is the fuel mass flow rate, Hv is the lower heating value of the fuel;

[0083] 3) The thermodynamic model of effective output work is shown as follows:

[0084] P e =T tq ×n / 9550

[0085] Where, T tq is the engine torque, n is the engine speed;

[0086] 4) The body's thermal thermodynamic model is shown below:

[0087]

[0088] Where, is the coolant mass flow rate, c s is the specific heat capacity of the cooling medium, T1 and T2 are the inlet and outlet temperatures of the cooling liquid respectively;

[0089] 5) The exhaust heat thermodynamic model is shown as follows:

[0090]

[0091] Where, is the exhaust mass flow rate, c pr is the constant pressure specific heat capacity of exhaust gas, c pk is the specific heat capacity of air at constant pressure, T r is the engine exhaust temperature, T k is the engine intake air temperature;

[0092] 6) The thermodynamic model of power system thermal efficiency is shown as follows:

[0093]

[0094] 7) Determination of parameters characterizing thermal efficiency of power system

[0095] From the thermal efficiency thermodynamic model, we can see that the thermal efficiency of the power system is a comprehensive evaluation indicator, which is affected by multiple variable parameters. The relationship between these influencing factors and thermal efficiency is expressed by the following general function:

[0096]

[0097] Based on the general function expression of the thermal efficiency of the power system, it is preliminarily determined that the parameter set characterizing the thermal efficiency of the power system is composed of eight characterizing parameters: oil quantity, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature, and exhaust flow rate.

[0098] Wherein, in said step S2, a correlation analysis model is established to optimize the thermal efficiency characterization parameter set;

[0099] Clean the test data, extract the characterization parameters, select the characterization parameters with strong correlation with thermal efficiency, and establish the optimal characterization parameter set;

[0100] 1) Acquisition of health status test data: Conduct characteristic tests on healthy power systems to obtain health status test data;

[0101] 2) Thermal efficiency characterization parameter extraction: The collected test data is cleaned, denoised, and other data preprocessing is performed to extract eight characterization parameters: oil volume, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature, and exhaust flow rate;

[0102] 3) Correlation analysis and calculation: With eight characterization parameters (torque, circulating oil volume, intake temperature, exhaust temperature, speed, coolant temperature, intake pressure, and oil temperature) as input and the power system thermal efficiency as output, the Pearson correlation analysis algorithm is used to perform a correlation analysis between the eight characterization parameters and the power system thermal efficiency. The Pearson correlation coefficient calculation formula is as follows:

[0103]

[0104] In the formula, r is the Pearson correlation coefficient, n is the number of samples, and are the average values ​​of the two samples, σ X and σ Y X i and Y i Standard score of

[0105] 4) Determination of the optimized characterization parameter set for thermal flow characteristics: Based on the correspondence between the Pearson correlation coefficient value and the degree of correlation, with |r| ≥ 0.4, the four characterization parameters of torque, circulating oil volume, intake temperature, and exhaust temperature are selected to form the optimal thermal efficiency characterization parameter set for power system thermal efficiency evaluation.

[0106] Wherein, in said step S3, a CNN-LSTM thermal efficiency prediction neural network model is established;

[0107] The CNN-LSTM thermal efficiency prediction neural network model takes the test data as input, the optimized thermal efficiency characterization parameter set as the feature vector, and the thermal efficiency value as output to achieve data-based thermal efficiency prediction;

[0108] 1) Input layer establishment: Create time series data consisting of four parameters: fuel volume, torque, intake temperature, and exhaust temperature. Each set of series represents a different heat flow characterization parameter of the power system;

[0109] 2) CNN layer establishment: Convolutional neural network is established. The convolutional layer uses one-dimensional convolution operation (1D-CNN) to extract parameter features in each time series. Max pooling is used to reduce data dimensions, extract key features, and improve computational efficiency.

[0110] 3) LSTM layer establishment: Establish an LSTM layer to capture the long-range dependencies and dynamic change trends of time series and analyze time series characteristics;

[0111] 4) Fully connected layer: A fully connected layer is established to aggregate the extracted features to ultimately generate the thermal efficiency prediction results;

[0112] 5) Output layer: Output the predicted value of thermal efficiency of the power system.

[0113] Wherein, in said step S4, the healthy power system test parameter values ​​are analyzed to establish a benchmark thermal efficiency;

[0114] The experimental test data of the healthy power system is input into the CNN-LSTM thermal efficiency prediction neural network model to obtain the long-term series thermal efficiency benchmark value under the healthy state.

[0115] Wherein, in said step S5, the test parameter value of the degraded power system is analyzed to determine the current thermal efficiency;

[0116] The experimental test data of the degraded power system is input into the CNN-LSTM thermal efficiency prediction neural network model to obtain the long-term series thermal efficiency evaluation value under the degraded state.

[0117] Wherein, in said step S6, the current thermal efficiency is compared with the reference thermal efficiency to determine the degradation state of the system;

[0118] In the same time dimension, the difference between the long-term series thermal efficiency evaluation value in the degraded state and the long-term series thermal efficiency benchmark value in the healthy state is compared to evaluate the system degradation state. The system degradation state evaluation algorithm is as follows:

[0119]

[0120] Where n is the number of test data points, η i is the thermal efficiency evaluation value of point i, η i0 is the benchmark value for thermal efficiency evaluation at point i.

[0121] Example 1

[0122] In order to better understand the present invention, the following is a detailed description of the present invention. Figure 1 Power system degradation state assessment process based on thermal efficiency, Figure 2 Optimization method of thermal efficiency characterization parameters of power system, Figure 3 The thermal efficiency evaluation neural network model structure is described in detail for the present invention.

[0123] (1) Establish a model of the heat flow formation and dissipation mechanism of the power system and determine the parameter set characterizing the thermal efficiency of the power system; the heat flow conversion and dissipation process of the power system is mainly that the total heat released by the engine burning fuel flows out through three channels: effective output power, body heat, and exhaust heat, and the total heat and effective output power, body heat, and exhaust heat have energy conservation characteristics.

[0124] (101) Principle of conservation of energy. The principle of conservation of energy is expressed in the following equation:

[0125] Q fuel =P e +Q S +Q R

[0126] Where Q fuel To burn calories, P e is the effective output power, Q S is the body heat, Q R Exhaust heat.

[0127] (102) The thermodynamic model of fuel combustion heat is shown as follows:

[0128]

[0129] Where, is the fuel mass flow rate, and Hv is the lower heating value of the fuel.

[0130] The fuel combustion heat thermodynamic model can determine that the thermal efficiency of the power system is related to the fuel quantity parameters;

[0131] (103) The thermodynamic model of effective output work is shown as follows:

[0132] P e =T tq ×n / 9550

[0133] Where, Ttq is the engine torque, and n is the engine speed.

[0134] The effective output work thermodynamic model can be used to determine that the thermal efficiency of the power system is related to the speed and torque parameters;

[0135] (104) The body's thermal thermodynamic model is shown below:

[0136]

[0137] Where, is the coolant mass flow rate, c s is the specific heat capacity of the cooling medium, T1 and T2 are the inlet and outlet temperatures of the coolant respectively.

[0138] The thermal efficiency of the power system can be determined to be related to the oil temperature and coolant temperature parameters based on the heat thermodynamic model of the engine body;

[0139] (105) The exhaust heat thermodynamic model is shown as follows:

[0140]

[0141] Where, is the exhaust mass flow rate, c pr is the constant pressure specific heat capacity of exhaust gas, c pk is the specific heat capacity of air at constant pressure, T r is the engine exhaust temperature, T k is the engine intake air temperature.

[0142] The exhaust heat thermodynamic model can determine that the thermal efficiency of the power system is related to the exhaust flow rate, intake temperature and exhaust temperature parameters;

[0143] 6) The thermodynamic model of power system thermal efficiency is shown as follows:

[0144]

[0145] 7) Determination of parameters characterizing thermal efficiency of power system

[0146] From the thermal efficiency thermodynamic model, we know that the thermal efficiency of the power system is a comprehensive evaluation indicator, which is affected by multiple variable parameters. The relationship between these influencing factors and thermal efficiency can be expressed by the following general function:

[0147]

[0148] Based on the general function expression of the thermal efficiency of the power system, it is preliminarily determined that the parameter set characterizing the thermal efficiency of the power system is composed of eight characterizing parameters, including oil quantity, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature, and exhaust flow rate.

[0149] (2) Establish a correlation analysis model to optimize the thermal efficiency characterization parameter set; clean the test data, extract the characterization parameters, select the characterization parameters with strong correlation with thermal efficiency, and establish the optimal characterization parameter set;

[0150] (201) Acquisition of health status test data: Conduct characteristic tests on healthy power systems to obtain health status test data;

[0151] (202) Thermal efficiency characterization parameter extraction: The collected test data is cleaned, denoised and other data preprocessed to extract eight characterization parameters including oil volume, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature and exhaust flow rate;

[0152] (203) Correlation analysis calculation: With the eight characterization parameters, including oil quantity, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature, and exhaust flow, as input and the power system thermal efficiency as output, the Pearson correlation analysis algorithm is used to perform a correlation analysis on the eight characterization parameters and the power system thermal efficiency. The Pearson correlation coefficient calculation formula is as follows:

[0153]

[0154] In the formula, r is the Pearson correlation coefficient, n is the number of samples, and are the average values ​​of the two samples, σ X and σ Y X i and Y i The standard score of .

[0155] (204) Determination of the optimal characterization parameter set for thermal flow characteristics: Based on the correspondence between the value of the Pearson correlation coefficient and the degree of correlation, |r|≥0.4 is taken, and the four characterization parameters such as oil quantity, torque, intake temperature, and exhaust temperature are preferably selected to form the optimal thermal efficiency characterization parameter set for the thermal efficiency evaluation of the power system.

[0156] (3) Establish a CNN-LSTM thermal efficiency prediction neural network model; use the test data as input, the optimal thermal efficiency characterization parameter set as the feature vector, and the thermal efficiency value as the output of the CNN-LSTM thermal efficiency prediction neural network model to achieve data-based thermal efficiency prediction;

[0157] (301) Input layer establishment: Establish time series data consisting of four parameters: oil volume, torque, intake temperature, and exhaust temperature. Each set of series represents different heat flow characterization parameters of the power system;

[0158] (302) CNN layer establishment: Convolutional neural network is established. The convolution layer uses one-dimensional convolution operation (1D-CNN) to extract parameter features in each time series; maximum pooling is used to reduce data dimensions, extract key features and improve computational efficiency;

[0159] (303) LSTM layer establishment: Establish an LSTM layer to capture the long-range dependency and dynamic change trend of the time series and analyze the time series characteristics;

[0160] (304) Fully connected layer: Establish a fully connected layer to aggregate the extracted features to ultimately generate the prediction results of thermal efficiency;

[0161] (305) Output layer: output the predicted value of thermal efficiency of the power system;

[0162] (4) Analyze the test parameter values ​​of the healthy power system and establish the benchmark thermal efficiency; input the test data of the healthy power system into the CNN-LSTM thermal efficiency prediction neural network model to obtain the long-term series thermal efficiency benchmark value under the healthy state;

[0163] (5) Analyze the test parameter values ​​of the degraded power system to determine the current thermal efficiency; input the test data of the degraded power system into the CNN-LSTM thermal efficiency prediction neural network model to obtain the long-term series thermal efficiency evaluation value under the degraded state;

[0164] (6) Compare the current thermal efficiency with the benchmark thermal efficiency to determine the degradation state of the system; in the same time dimension, compare the difference between the long-term series thermal efficiency evaluation value in the degradation state and the long-term series thermal efficiency benchmark value in the healthy state to evaluate the degradation state of the system. The system degradation state evaluation algorithm is as follows:

[0165]

[0166] Where n is the number of test data points, η i is the thermal efficiency evaluation value of point i, η i0 The benchmark value for thermal efficiency evaluation at point i

[0167] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for evaluating the degradation state of a vehicle power system based on thermal efficiency, characterized in that: The method comprises the following steps: Step S1: Establish a model of the heat flow formation and dissipation mechanism of the power system and determine a parameter set representing the thermal efficiency of the power system; Step S2: Establish a correlation analysis model to optimize the power system thermal efficiency characterization parameter set; Step S3: Establishing a CNN-LSTM thermal efficiency prediction neural network model; Step S4: Analyze the test parameter values ​​of the healthy power system and establish a baseline thermal efficiency; Step S5: Analyze the test parameter values ​​of the degraded power system to determine the current thermal efficiency; Step S6: Compare the current thermal efficiency with the benchmark thermal efficiency to determine the system degradation state.

2. The method for evaluating the degradation state of a vehicle power system based on thermal efficiency according to claim 1, wherein: In the step S1, a power system heat flow formation and dissipation mechanism model is established to determine a power system thermal efficiency characterization parameter set; During the heat flow conversion and dissipation process of the power system, the total heat released by the engine burning fuel flows out through three channels: effective output work, engine heat, and exhaust heat. The total heat and effective output work, engine heat, and exhaust heat have energy conservation characteristics. 1) Principle of Conservation of Energy The principle of conservation of energy is expressed in the following equation: Q fuel =P e +Q S +Q R Where Q fuel To burn calories, P e is the effective output power, Q S is the body heat, Q R is the exhaust heat; 2) The thermodynamic model of fuel combustion heat is shown as follows: Where, is the fuel mass flow rate, Hv is the lower heating value of the fuel; 3) The thermodynamic model of effective output work is shown as follows: P e =T tq ×n9550 Where, T tq is the engine torque, n is the engine speed; 4) The body's thermal thermodynamic model is shown below: Where, is the coolant mass flow rate, c s is the specific heat capacity of the cooling medium, T1 and T2 are the inlet and outlet temperatures of the cooling liquid respectively; 5) The exhaust heat thermodynamic model is shown as follows: Where, is the exhaust mass flow rate, c pr is the constant pressure specific heat capacity of exhaust gas, c pk is the specific heat capacity of air at constant pressure, T r is the engine exhaust temperature, T k is the engine intake air temperature; 6) The thermodynamic model of power system thermal efficiency is shown as follows: 7) Determination of parameters characterizing thermal efficiency of power system From the thermal efficiency thermodynamic model, we can see that the thermal efficiency of the power system is a comprehensive evaluation indicator, which is affected by multiple variable parameters. The relationship between these influencing factors and thermal efficiency is expressed by the following general function: Based on the general function expression of the thermal efficiency of the power system, it is preliminarily determined that the parameter set characterizing the thermal efficiency of the power system is composed of eight characterizing parameters: oil quantity, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature, and exhaust flow rate.

3. The method for evaluating the degradation state of a vehicle power system based on thermal efficiency according to claim 2, wherein: In step S2, a correlation analysis model is established to optimize the thermal efficiency characterization parameter set; Clean the test data, extract the characterization parameters, select the characterization parameters with strong correlation with thermal efficiency, and establish the optimal characterization parameter set; 1) Acquisition of health status test data: Conduct characteristic tests on healthy power systems to obtain health status test data; 2) Thermal efficiency characterization parameter extraction: The collected test data is cleaned, denoised, and other data preprocessing is performed to extract eight characterization parameters: oil volume, oil temperature, speed, torque, coolant temperature, intake temperature, exhaust temperature, and exhaust flow rate; 3) Correlation analysis and calculation: With eight characterization parameters (torque, circulating oil volume, intake temperature, exhaust temperature, speed, coolant temperature, intake pressure, and oil temperature) as input and the power system thermal efficiency as output, the Pearson correlation analysis algorithm is used to perform a correlation analysis between the eight characterization parameters and the power system thermal efficiency. The Pearson correlation coefficient calculation formula is as follows: In the formula, r is the Pearson correlation coefficient, n is the number of samples, and are the average values ​​of the two samples, σ X and σ Y X i and Y i Standard score of 4) Determination of the optimized characterization parameter set for thermal flow characteristics: Based on the correspondence between the Pearson correlation coefficient value and the degree of correlation, with |r| ≥ 0.4, the four characterization parameters of torque, circulating oil volume, intake temperature, and exhaust temperature are selected to form the optimal thermal efficiency characterization parameter set for power system thermal efficiency evaluation.

4. The method for evaluating the degradation state of a vehicle power system based on thermal efficiency according to claim 3, wherein: In step S3, a CNN-LSTM thermal efficiency prediction neural network model is established; The CNN-LSTM thermal efficiency prediction neural network model takes the test data as input, the optimized thermal efficiency characterization parameter set as the feature vector, and the thermal efficiency value as output to achieve data-based thermal efficiency prediction; 1) Input layer establishment: Create time series data consisting of four parameters: fuel volume, torque, intake temperature, and exhaust temperature. Each set of series represents a different heat flow characterization parameter of the power system; 2) CNN layer establishment: Convolutional neural network is established. The convolutional layer uses one-dimensional convolution operation 1D-CNN to extract parameter features in each time series; Max pooling is used to reduce data dimensions, extract key features and improve computational efficiency; 3) LSTM layer establishment: Establish an LSTM layer to capture the long-range dependencies and dynamic change trends of time series and analyze time series characteristics; 4) Fully connected layer: A fully connected layer is established to aggregate the extracted features to ultimately generate the thermal efficiency prediction results; 5) Output layer: Output the predicted value of thermal efficiency of the power system.

5. The method for evaluating the degradation state of a vehicle power system based on thermal efficiency according to claim 4, wherein: In step S4, the healthy power system test parameter values ​​are analyzed to establish a benchmark thermal efficiency; The experimental test data of the healthy power system is input into the CNN-LSTM thermal efficiency prediction neural network model to obtain the long-term series thermal efficiency benchmark value under the healthy state.

6. The method for evaluating the degradation state of a vehicle power system based on thermal efficiency according to claim 5, wherein: In step S5, the degraded power system test parameter values ​​are analyzed to determine the current thermal efficiency; The experimental test data of the degraded power system is input into the CNN-LSTM thermal efficiency prediction neural network model to obtain the long-term series thermal efficiency evaluation value under the degraded state.

7. The method for evaluating the degradation state of a vehicle power system based on thermal efficiency according to claim 6, wherein: In step S6, the current thermal efficiency is compared with the reference thermal efficiency to determine the degradation state of the system; In the same time dimension, the difference between the long-term series thermal efficiency evaluation value in the degraded state and the long-term series thermal efficiency benchmark value in the healthy state is compared to evaluate the system degradation state. The system degradation state evaluation algorithm is as follows: Where n is the number of test data points, η i is the thermal efficiency evaluation value of point i, η i0 is the benchmark value for thermal efficiency evaluation at point i.

8. The method for evaluating the degradation state of a vehicle power system based on thermal efficiency according to claim 7, wherein: The method analyzes the heat flow formation and dissipation mechanism of the vehicle power system to establish a thermal efficiency characterization parameter set. Combined with correlation analysis, the characterization parameter set is optimized. A thermal efficiency benchmark for typical evaluation conditions is established through a neural network model. The thermal efficiency of the degraded power system state is compared with the benchmark thermal efficiency to determine the degradation state of the vehicle power system.

9. The method for evaluating the degradation state of a vehicle power system based on thermal efficiency according to claim 7, wherein: The method can be applied to the degradation state assessment of any fuel vehicle power system.

10. The method for evaluating vehicle power system degradation state based on thermal efficiency according to claim 7, wherein: The method enables the evaluation of the system-level degradation state of the vehicle power system and improves the vehicle operation state monitoring capability.

Images