Hybrid power system health management method and system

Abstract

The invention discloses a hybrid power system health management method and system, and belongs to the field of power system health management. Characteristic signals of an engine, a battery and a vehicle in the hybrid power system are selected to establish the health degree of the hybrid power system; when the hybrid power system breaks down, the health degree of the hybrid power system based on the grey correlation analysis method GRA is calculated in real time by obtaining a characteristic signal and conducting normalization processing on the characteristic signal, and misjudgment and missed judgment of the fault are reduced through a fault confirmation method based on counting. After the fault of the hybrid power system is confirmed, through a power redistribution health management strategy under the system fault based on the LSTM-MPC, the changes of the vehicle speed, the battery temperature and the diesel engine exhaust temperature of the hybrid power system in a limited time domain range are predicted; and predictive maintenance is carried out on the health degree of the hybrid power system in combination with the change prediction result of the partial state of the hybrid power system, so that the hybrid power system can still operate with faults under the condition that constraint conditions are met.

Description

Technical Field

[0001] The present invention relates to a hybrid power system health management method and system, belonging to the technical field of power system health management. Background Art

[0002] The hybrid power system is one of the main power systems to achieve the "dual carbon" goal. In particular, in recent years, hybrid power systems have coexisted in various types for different application scenarios. Although the systems are different, the increasingly complex composition and operation mode of hybrid systems have become a common technical trend. For complex systems, it is necessary to manage the health of the system.

[0003] Especially from the system level. Considering the characteristics of multiple power sources in the hybrid system, that is, each power source can work independently and can also complement each other as backup, when the system performance degrades or deteriorates, a health management strategy that meets the constraints can be formulated to achieve the purpose of improving system reliability, and to achieve real-time health status assessment of the system and predictive maintenance after system failures. That is, by reconfiguring the system's control strategy, the system can still achieve fault-tolerant operation while meeting the constraints, which is of great significance to the control of the hybrid system. Summary of the invention

[0004] The purpose of the present invention is to provide a hybrid power system health management method and system, select the characteristic signals of the engine, battery and vehicle in the hybrid power system to establish the health of the hybrid power system, when the hybrid power system fails, by obtaining the characteristic signals and normalizing them, the health of the hybrid power system based on the gray correlation analysis method GRA is calculated in real time, and the misjudgment and missed judgment of the fault are reduced by the fault confirmation method based on counting, according to the degree of deviation of the health after the fault from the normal value, accurately judge whether the hybrid power system fault occurs, when the hybrid power system fault is confirmed, the power redistribution health management strategy under the system fault based on LSTM-MPC is used to predict the changes of the hybrid power system speed, battery temperature and diesel engine exhaust temperature within a limited time domain, and the health of the hybrid power system is predictively maintained in combination with the prediction results of the changes in the partial states of the hybrid power system, so that the hybrid power system can still operate with the fault under the constraint conditions. The partial states of the hybrid power system refer to the vehicle speed, battery temperature and diesel engine temperature.

[0005] The objective of the present invention is achieved through the following technical solutions:

[0006] A hybrid power system health management method disclosed in the present invention comprises the following steps:

[0007] Step 1: Obtain the diesel engine instantaneous analog signal, the battery instantaneous analog signal, and the vehicle instantaneous analog signal through the signal acquisition and processing module to be tested, and obtain the corresponding hybrid powertrain status signal, normalize the hybrid powertrain status signal, and obtain the normalized hybrid powertrain diesel engine instantaneous status signal, battery instantaneous status signal, and vehicle instantaneous status signal.

[0008] Step 1.1: Obtain the original instantaneous analog signal of the vehicle component through the signal acquisition and processing module to be tested;

[0009] Step 1.2: performing signal conversion and calibration on the original instantaneous analog signal obtained in step 1.1 to obtain the relationship between the original analog signal and the original digital signal, and performing filtering processing to obtain the instantaneous state signal of the vehicle;

[0010] Step 1.3: Based on the original instantaneous analog signal obtained in step 1.1 and the instantaneous state signal obtained in step 1.2, the hybrid power system state signal is obtained, and the state signal includes the following three types of signals: diesel engine instantaneous state signal, battery instantaneous state signal, and vehicle instantaneous state signal: the diesel engine instantaneous state signal includes intake temperature, intake pressure, exhaust temperature, diesel engine speed and accelerator pedal position; the battery instantaneous state signal includes SOC, battery bus voltage, battery bus current, and battery temperature; the vehicle instantaneous state signal includes target vehicle speed, actual vehicle speed, and vehicle operation mode.

[0011] Step 1.4: Perform signal processing on the hybrid powertrain state signal obtained in step 1.3, perform normalization on it, eliminate the influence of different dimensions, and obtain a normalized hybrid powertrain state signal.

[0012] Step 2: According to the unified dimension state signal input by the signal acquisition and processing module to be tested, combined with the fault form and the actual vehicle sensor signal, select some state signals of the engine, battery and vehicle in the hybrid system as characteristic signals. The characteristic signals of the selected area include diesel engine intake pressure, diesel engine intake temperature, diesel engine exhaust temperature, accelerator pedal position, battery bus voltage, battery bus current, battery temperature and vehicle speed difference. Determine the weight of the characteristic signal through principal component analysis, and calculate the health of the hybrid system based on the grey correlation analysis method GRA in real time. Confirm the fault through the counting-based fault confirmation method to reduce the fault misjudgment and missed judgment caused by the error and noise of the characteristic signal, and obtain the fault confirmation signal.

[0013] Step 3: Based on the LSTM-MPC health management strategy for power redistribution under system failure, the hybrid powertrain signal obtained in step 1 is received as a predictive maintenance indicator. After receiving the fault confirmation signal obtained in step 2, a partial prediction model for MPC is obtained through LSTM training. The LSTM-MPC prediction model is used to predict the changes in the partial state of the system within a limited time domain, and the health of the hybrid power system is predictively maintained in combination with the prediction results of the partial state changes of the hybrid power system. When a hybrid power system fails, the conversion relationship between power consumption and fuel consumption is given, the battery energy consumption is equivalent to the fuel energy consumption, the comprehensive energy consumption is calculated, and the target vehicle speed following effect is improved by sacrificing the comprehensive energy consumption. At the same time, the battery is guaranteed to work in its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, ensuring that the hybrid power system can still operate with faults while meeting the constraints.

[0014] The power redistribution health management strategy under system failure based on LSTM-MPC, the rules of the MPC control strategy are as follows:

[0015] The current time is k. The LSTM-based prediction model obtains the current k-time input and historical time information of the controlled object to obtain the future time domain k+t. p The optimal control sequence of the optimal control sequence is retained, and only the first control variable u of the optimal control sequence is retained p (k+1) and applies it to the controlled object, and all other control variables are ignored, and the above optimization process is repeated at the next sampling time. At time k, the optimization objective is expressed by formula (1):

[0016]

[0017] In formula (1), J k is the cost function, t p is the prediction duration, x(t) is the state quantity at time t, and u(t) is the control quantity at time t.

[0018] The constraint condition is expressed by formula (2):

[0019]

[0020] In formula (2), x min (t) and x max (t) are the upper and lower limits of the state variable at time t, u min (t) and u max (t) are the upper and lower limits of the control variable at time t.

[0021] According to the multi-power source characteristics of the hybrid power system, the optimization goal is to minimize the vehicle speed following error by sacrificing comprehensive energy consumption. The constraints include diesel engine exhaust temperature constraints, battery temperature constraints, and theoretical physical constraints of other components. At time k, in the prediction time domain, the cost function of the hybrid power system is expressed by formula (3):

[0022]

[0023] In formula (3), t p is the prediction duration, k p is the weight coefficient of vehicle speed difference, k e is the weight coefficient of comprehensive energy consumption, u(t) is the control quantity, here is the accelerator pedal position correction coefficient, v a (u(t),t) and v t (t) are the actual vehicle speed and target vehicle speed, km / h;

[0024] is the comprehensive energy consumption, g / s.

[0025] The constraint condition is expressed by formula (4):

[0026]

[0027] In formula (4), T e is the diesel engine exhaust temperature, T e_max is the diesel engine exhaust temperature limit value; T b is the battery temperature,

[0028] [T b_smin ,T b_smax ] is the optimal operating temperature range of lithium-ion batteries, ranging from [15,35], ℃; SOC min , SOC max They are the upper and lower limits of the power battery charge and discharge SOC; P dem , P e and P m They are total required power, diesel engine power and motor power, kW respectively.

[0029] The input of the LSTM prediction model is the health status assessment result, vehicle operation mode, accelerator pedal position, battery power, actual vehicle speed, battery temperature and diesel engine exhaust temperature. The output is the vehicle speed, diesel engine exhaust temperature and battery temperature of the hybrid system. The vehicle speed, battery temperature and diesel engine exhaust temperature of the system at future moments are predicted by rolling based on the input information of the historical time period and the current moment. The mean absolute error (MAE) is selected as the loss function to optimize the parameters of the prediction model and measure the accuracy of the prediction model. MAE can be expressed by formula (5):

[0030]

[0031] In formula (5), Y is the true value, f(x) is the predicted value, and n is the number of samples.

[0032] Preferably, according to the conversion relationship between power consumption and fuel consumption given in GB / T 19754-2021 "Heavy-duty hybrid electric vehicle energy consumption test method", the battery energy consumption is equivalent to fuel energy consumption, and the comprehensive energy consumption is calculated. The target vehicle speed following effect is improved by sacrificing the comprehensive energy consumption. At the same time, the battery is ensured to operate within its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, thereby ensuring that the hybrid system can still operate with faults while meeting the constraints.

[0033] The present invention also discloses a hybrid power system health management system, which is used to implement the hybrid power system health management method. The hybrid power system health management system mainly consists of a host computer PC, a hybrid powertrain, a vehicle controller lower computer VCU, a test signal acquisition and processing module, a health status evaluation module, and a power distribution correction module based on LSTM-MPC. The host computer PC and the vehicle controller lower computer VCU are connected through a signal line. The hybrid powertrain is connected to the vehicle controller lower computer VCU through a signal bus. The host computer PC controls the vehicle controller lower computer VCU to receive and process signals from the hybrid powertrain, and performs real-time signal transmission interaction. The hybrid power system health status is evaluated based on the test signal acquisition and processing module, health status evaluation module, and LSTM-MPC-based power distribution correction module written in the vehicle controller lower computer VCU, and the power distribution correction of the hybrid power system is performed according to the health status evaluation result.

[0034] The signal acquisition and processing module receives and processes the hybrid power system status signal, which includes three types of signals: diesel engine characteristic signal, battery characteristic signal, and vehicle characteristic signal. Diesel engine characteristic signals include intake temperature, intake pressure, exhaust temperature, diesel engine speed, and accelerator pedal position; battery characteristic signals include SOC, battery bus voltage, battery bus current, and battery temperature; vehicle characteristic signals include target vehicle speed, actual vehicle speed, and vehicle operation mode. After acquiring the signal, the status signal is normalized to eliminate the influence of different dimensions, and preprocessing is performed for the subsequent health status assessment module. By unifying the status signal dimensions, the efficiency of the hybrid power system health status assessment is improved on the basis of ensuring the accuracy of characteristic signals at different orders of magnitude.

[0035] According to formula (6), the time series data such as intake pressure, intake temperature, exhaust temperature, accelerator pedal position, bus current, bus voltage, battery temperature and vehicle speed are normalized:

[0036]

[0037] Where i = 1, 2, 3, ..., n, k = 1, 2, 3, ..., m, n is the number of samples, m is the number of features

[0039] The health status assessment module obtains the status signal from the signal acquisition and processing module, calculates the health status through the vehicle controller lower computer VCU according to the health status, determines the health status of the hybrid powertrain, analyzes the hybrid power operating conditions and diagnoses whether the system has a fault.

[0040] The health degree is constructed based on the above-mentioned normalized characteristic signals, that is, based on the intake pressure, intake temperature, exhaust temperature, accelerator pedal position, bus current, bus voltage, battery temperature and vehicle speed, and the constructed health degree is embedded in the health status assessment module.

[0041] The health degree construction method is as follows:

[0042] The diesel-electric hybrid system has the characteristics of high coupling, nonlinearity and multiple power sources, and is comprehensively judged through multiple dimensions and multiple standards. Combined with the multi-dimensional state signals collected by the diesel-electric hybrid system, the correlation between the operating state and the healthy state (normal state) of the hybrid system is calculated using gray correlation analysis. The correlation is used to reflect the degree of deviation between the operating state and the healthy state of the hybrid system. The greater the deviation, the smaller the correlation.

[0043] The details are as follows: Health:

[0044] 1) Using the normalized data output by the signal acquisition and processing module to calculate the grey correlation coefficient one by one;

[0045]

[0046] In formula (7), k = 1, 2, 3, ..., m, i = 1, 2, 3, ..., n, ρ represents the discrimination coefficient, ρ∈(0, 1), the smaller ρ is, the greater the difference in correlation coefficients is, the stronger the resolution is, and it is preferably 0.5.

[0047] 2) Calculate the correlation degree one by one according to formula (8).

[0048]

[0049] In formula (8), ω i (k) represents the weight coefficient of the feature signal, k = 1, 2, 3, ..., m, i = 1, 2, 3, ..., n.

[0050] The health status assessment of hybrid systems based on GRA requires the calculation of correlation. Since different characteristic signals in hybrid systems have different sensitivities to faults, directly taking the average value may lead to inaccurate correlation calculation. The weight values ​​of characteristic signals of different dimensions are determined by using the PCA method.

[0051] The LSTM-MPC-based power distribution correction module obtains the output and test signal from the health status assessment module, obtains the hybrid power system status in the self-health status assessment module, calculates the LSTM-MPC-based power redistribution health management method through the lower computer VCU of the vehicle controller, and calculates the accelerator pedal position correction coefficient in advance. When the fault is confirmed, the output correction coefficient is used to correct the accelerator pedal position of the vehicle controller, that is, in the case of a hybrid power system fault, the target speed following effect is improved by sacrificing comprehensive energy consumption, while ensuring that the battery operates in its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, ensuring that the hybrid power system can still operate with faults under the constraints.

[0052] Beneficial effects:

[0053] 1. The present invention discloses a hybrid power system health management method and system, which analyzes hybrid power system faults, determines characteristic signals, and uses PCA to confirm the weight coefficients of each characteristic signal to obtain the health of the hybrid power system. Compared with directly setting the average value as the weight coefficient of each characteristic signal, it can accurately evaluate the health status of the hybrid power system.

[0054] 2. The present invention discloses a hybrid power system health management method and system, which utilizes a health management strategy of power redistribution under system failure based on LSTM-MPC, giving full play to the advantages of the hybrid system with multiple power sources, allowing the hybrid power system to operate with failures while satisfying constraints, thereby increasing the tolerance of the hybrid power system. At the same time, the vehicle's original vehicle controller lower computer VCU can meet the computing requirements. Compared with the health management strategy based on MPC, it has the advantages of fast response and strong practicality.

[0055] 3. The present invention discloses a hybrid power system health management method and system, which integrates the constructed health level into the health status assessment module. When the hybrid power system is evaluated, the health level is directly called to perform the health status assessment, which has a faster assessment speed and can improve the efficiency of the health status assessment.

[0056] 4. The present invention discloses a hybrid power system health management method and system, which is based on the LSTM-MPC health management strategy for power redistribution under system failure, receives hybrid powertrain signals as predictive maintenance indicators, obtains partial prediction models for MPC through LSTM training after receiving the fault confirmation signal, predicts the changes of partial system states within a limited time domain based on the LSTM-MPC prediction model, and performs predictive maintenance on the health of the hybrid power system in combination with the prediction results of partial state changes of the hybrid power system. When a hybrid power system fails, the conversion relationship between power consumption and fuel consumption is given, the battery energy consumption is equivalent to fuel energy consumption, the comprehensive energy consumption is calculated, and the target vehicle speed following effect is improved by sacrificing the comprehensive energy consumption, while ensuring that the battery operates within its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, ensuring that the hybrid power system can still operate with faults while meeting the constraints.

[0057] 5. The present invention discloses a hybrid power system health management method and system, which obtains characteristic signals and normalizes them, calculates the health of the hybrid power system based on the grey correlation analysis method GRA in real time, reduces misjudgment and missed judgment of faults through a counting-based fault confirmation method, and accurately determines whether a hybrid power system fault occurs according to the degree to which the health degree deviates from the normal value after the fault, thereby improving the accuracy and efficiency of the hybrid power system health management.

[0058] 6. A hybrid power system health management method and system disclosed in the present invention converts battery energy consumption into fuel energy consumption according to the conversion relationship between power consumption and fuel consumption given in GB / T 19754-2021 "Test Method for Energy Consumption of Heavy-Duty Hybrid Electric Vehicles", and unifies the calculated comprehensive energy consumption into fuel consumption, which facilitates the energy consumption calculation and statistics of the hybrid power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a schematic diagram of a hybrid power system health management system of the present invention;

[0060] Figure 2 is a flow chart of a hybrid power system health management method of the present invention;

[0061] Figure 3 It is a schematic diagram of the health status and fault confirmation of the battery under the fault of increased internal resistance of the present invention;

[0062] Figure 4 It is a flow chart of the algorithm of the power redistribution control strategy based on LSTM-MPC of the present invention;

[0063] Figure 5 It is a vehicle speed following diagram with or without power correction under battery internal resistance fault of the present invention;

[0064] Figure 6 It is a diagram of battery temperature and exhaust temperature with and without power correction under battery internal resistance fault of the present invention.

[0065] Figure 7 It is a comprehensive fuel consumption diagram with and without power correction under battery internal resistance failure of the present invention. DETAILED DESCRIPTION

[0066] In order to better illustrate the purpose and advantages of the present invention, the invention is further described below with reference to the accompanying drawings and examples.

[0067] Embodiment 1:

[0068] This embodiment is based on a certain type of P2 parallel hybrid electric vehicle, and a hybrid power system health assessment method is composed of a test signal acquisition and processing module, a health status assessment module and a power distribution correction module. Combined with the hybrid power system health assessment system published in the invention patent, a specific health assessment is performed by interacting with the hybrid power system in real time. The system health assessment system of the present invention can quickly and accurately determine the health status of the hybrid power system, and at the same time can effectively utilize the advantages of the hybrid system, and redistribute the output power ratio of the engine and the battery through the LSTM-MPC-based method, so that the system can operate with faults within the range of certain constraints.

[0069] like Figure 1 As shown, the hybrid power system health management system disclosed in this embodiment is composed of a host computer PC, a hybrid powertrain, a vehicle controller slave computer VCU, a test signal acquisition and analysis module, a health status assessment module, and a power distribution correction module. The host computer PC and the vehicle controller slave computer VCU are connected via an RS485 signal line. The hybrid powertrain is connected to the vehicle controller slave computer VCU via the CANs bus. The host computer PC controls the vehicle controller slave computer VCU to receive and process signals from the hybrid powertrain, and performs real-time signal transmission interaction. The health status of the hybrid powertrain is evaluated based on the test signal acquisition and analysis module, health status assessment module, and power distribution correction module written in the vehicle controller slave computer VCU, and the power distribution of the hybrid powertrain is corrected according to the health status obtained by the assessment. The workflow of the hybrid power system health assessment method is as follows: Figure 2As shown in the figure, first, the signal acquisition and analysis module obtains the driving state signals of the engine and battery in the vehicle through the CANs bus, and normalizes them to eliminate the influence of different dimensions, and preprocesses them for the subsequent health status assessment module; the health status assessment module obtains the state signal from the signal acquisition and analysis module, calculates the health degree through the lower computer VCU of the vehicle controller, analyzes the hybrid working condition and determines whether the system is in a healthy state. The power distribution correction module obtains the output from the health status assessment module and the hybrid system state in the signal acquisition and processing module, and calculates the power distribution correction control strategy based on LSTM-MPC through the lower computer VCU of the vehicle controller, continuously reducing the power proportion of the hybrid powertrain fault subsystem to achieve the optimization goal. The power gap reduced by the faulty subsystem is supplemented by other normal subsystems in the hybrid powertrain to obtain a new power distribution under the fault of the hybrid powertrain, giving full play to the advantages of the hybrid powertrain and making the system run with faults within the specified constraints.

[0070] The signal acquisition and processing module receives and processes the hybrid power system status signal, which includes the following three types of signals: diesel engine characteristic signal, battery characteristic signal, and vehicle characteristic signal: diesel engine characteristic signal includes intake temperature, intake pressure, exhaust temperature, diesel engine speed and accelerator pedal position; battery characteristic signal includes SOC, battery bus voltage, battery bus current, and battery temperature; vehicle characteristic signal includes target vehicle speed, actual vehicle speed and vehicle operation mode. After acquiring the signal, the status signal is normalized to eliminate the influence of different dimensions, and preprocessed for the subsequent health status assessment module. By unifying the status signal dimension, the efficiency of hybrid power system health status assessment is improved on the basis of ensuring the accuracy of characteristic signals at different orders of magnitude.

[0071] According to formula (6), the time series data such as intake pressure, intake temperature, exhaust temperature, accelerator pedal position, bus current, bus voltage, battery temperature and vehicle speed are normalized:

[0072]

[0073] Where i = 1, 2, 3, ..., n, k = 1, 2, 3, ..., m, n is the number of samples, m is the number of features

[0075] The health status assessment module obtains the status signal from the signal acquisition and processing module, calculates the health status through the vehicle controller lower computer VCU according to the health status, determines the health status of the hybrid powertrain, analyzes the hybrid power operating conditions and diagnoses whether the system has a fault.

[0076] The internal resistance of the battery in the hybrid powertrain is set to increase to simulate a real fault condition, and the health level is used to quickly and accurately determine whether the hybrid powertrain has a fault;

[0077] The details are as follows: Health:

[0078] 1) Using the normalized data output by the signal acquisition and processing module to calculate the grey correlation coefficient one by one;

[0079]

[0080] In formula (7), k = 1, 2, 3, ..., m, i = 1, 2, 3, ..., n, ρ represents the discrimination coefficient, ρ∈(0, 1), the smaller ρ is, the greater the difference in correlation coefficients is, the stronger the discrimination ability is, and the general value is 0.5.

[0081] 2) Calculate the correlation one by one.

[0082]

[0083] In formula (8), ω i (k) represents the weight coefficient of the feature signal, k = 1, 2, 3, ..., m, i = 1, 2, 3, ..., n.

[0084] The PCA method is used to determine the weight values ​​of feature signals of different dimensions.

[0085] The characteristic signal data under normal cycle and hybrid system fault cycle are collected as sample sets, and the characteristic signal weight values ​​under different modes are obtained by PCA method as follows:

[0086] Battery Drive:

[0087] 1) Bus current: 15.7

[0088] 2) Bus voltage: 4.7

[0089] 3) Battery temperature: 42.3

[0090] 4) Speed ​​difference: 37.3

[0091] Diesel engine drive:

[0092] 1) Inlet pressure: 18.5

[0093] 2) Inlet air temperature: 19.7

[0094] 3) Exhaust temperature: 33.4

[0095] 4) Accelerator pedal: 12.1

[0096] 5) Speed ​​difference: 16.3

[0097] Joint drive:

[0098] 1) Inlet pressure: 15.6

[0099] 2) Inlet air temperature: 16.7

[0100] 3) Exhaust temperature: 29.1

[0101] 4) Accelerator pedal: 10.4

[0102] 5) Bus current: 6.6

[0103] 6) Bus voltage: 1.6

[0104] 7) Battery temperature: 11.3

[0105] 8) Speed ​​difference: 8.7

[0106] The fault status of the hybrid powertrain is determined by the obtained health status, and the fault status is accumulated by counting mode to reduce the misjudgment and missed judgment of faults. The upper threshold is set to 127 and the lower threshold is -128. The counter is increased by 1 when the current fault status is 1 (faulty), and decreased by 1 when it is 0 (no fault). When the limit is reached, the counter will no longer accumulate. When the count reaches the set upper threshold, the fault is confirmed.

[0107] When the battery internal resistance in the hybrid powertrain increases, the health status and fault confirmation are as follows: Figure 3 shown.

[0108] The LSTM-MPC-based power distribution correction module obtains the output and test signal from the health status assessment module, obtains the hybrid power system status in the health status assessment module, calculates the LSTM-MPC-based power redistribution health management method through the lower computer VCU of the vehicle controller, and calculates the accelerator pedal position correction coefficient in advance. When the fault is confirmed, the output correction coefficient is used to correct the accelerator pedal position of the vehicle controller, that is, in the case of system failure, the target vehicle speed following effect is improved by sacrificing a certain amount of comprehensive energy consumption, while ensuring that the battery operates in its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, ensuring that the system can still operate with faults while meeting the constraints. The algorithm flow chart is as follows: Figure 4 shown.

[0109] When the hybrid powertrain is in a fault condition of increased battery internal resistance, the vehicle speed following condition becomes worse and the battery is out of the optimal operating temperature range. After the power distribution correction, the vehicle speed following condition (power) is improved at the expense of a certain comprehensive fuel consumption (economy), and the battery is guaranteed to work in the optimal temperature range and the diesel engine does not exceed the exhaust temperature limit. Figure 5 , as shown in Figures 6 and 7.

[0110] This embodiment can evaluate the health status of the hybrid power system and give full play to the advantages of the hybrid power system. When a fault occurs, power is redistributed through a power redistribution strategy based on LSTM-MPC, so that the hybrid power system can operate with faults under the conditions of the optimal operating temperature range of the battery and the exhaust temperature of the diesel engine not exceeding the limit, thereby increasing the tolerance of the hybrid power system. At the same time, a hybrid power system health management method and system disclosed in the present invention embed the constructed health into a health status evaluation module. When evaluating the health of the hybrid power system, the health index is directly called for, and the evaluation speed is fast, which improves the efficiency of the health evaluation. At the same time, the vehicle's original vehicle controller lower computer VCU can meet the computing requirements, and has the characteristics of fast response and strong practicality.

[0111] like Figure 2 As shown, the embodiment discloses a hybrid power system health management method, and the specific implementation steps are as follows:

[0112] Step 1: Obtain the diesel engine instantaneous analog signal, the battery instantaneous analog signal, and the vehicle instantaneous analog signal through the signal acquisition and processing module to be tested, and obtain the corresponding hybrid powertrain status signal, normalize the hybrid powertrain status signal, and obtain the normalized hybrid powertrain diesel engine instantaneous status signal, battery instantaneous status signal, and vehicle instantaneous status signal.

[0113] Step 1.1: Obtain the original instantaneous analog signal of the vehicle component through the signal acquisition and processing module to be tested;

[0114] Step 1.2: performing signal conversion and calibration on the original instantaneous analog signal obtained in step 1.1 to obtain the relationship between the original analog signal and the original digital signal, and performing filtering processing to obtain the instantaneous state signal of the vehicle;

[0115] Step 1.3: Based on the original instantaneous analog signal obtained in step 1.1 and the instantaneous state signal obtained in step 1.2, the hybrid power system state signal is obtained, and the state signal includes the following three types of signals: diesel engine instantaneous state signal, battery instantaneous state signal, and vehicle instantaneous state signal: the diesel engine instantaneous state signal includes intake temperature, intake pressure, exhaust temperature, diesel engine speed and accelerator pedal position; the battery instantaneous state signal includes SOC, battery bus voltage, battery bus current, and battery temperature; the vehicle instantaneous state signal includes target vehicle speed, actual vehicle speed, and vehicle operation mode.

[0116] Step 1.4: Perform signal processing on the hybrid powertrain state signal obtained in step 1.3, perform normalization on it, eliminate the influence of different dimensions, and obtain a normalized hybrid powertrain state signal.

[0117] Step 2: According to the unified dimension state signal input by the signal acquisition and processing module to be tested, combined with the fault form and the actual vehicle sensor signal, select some state signals of the engine, battery and vehicle in the hybrid system as characteristic signals. The characteristic signals of the selected area include diesel engine intake pressure, diesel engine intake temperature, diesel engine exhaust temperature, accelerator pedal position, battery bus voltage, battery bus current, battery temperature and vehicle speed difference. Determine the weight of the characteristic signal through principal component analysis, and calculate the health of the hybrid system based on the grey correlation analysis method GRA in real time. Confirm the fault through the counting-based fault confirmation method to reduce the fault misjudgment and missed judgment caused by the error and noise of the characteristic signal, and obtain the fault confirmation signal.

[0118] Step 3: Based on the LSTM-MPC health management strategy for power redistribution under system failure, the hybrid powertrain signal obtained in step 1 is received as a predictive maintenance indicator. After receiving the fault confirmation signal obtained in step 2, a partial prediction model for MPC is obtained through LSTM training. The LSTM-MPC prediction model is used to predict the changes in the partial state of the system within a limited time domain, and the health of the hybrid power system is predictively maintained in combination with the prediction results of the partial state changes of the hybrid power system. When a hybrid power system fails, the conversion relationship between power consumption and fuel consumption is given, the battery energy consumption is equivalent to the fuel energy consumption, the comprehensive energy consumption is calculated, and the target vehicle speed following effect is improved by sacrificing the comprehensive energy consumption. At the same time, the battery is guaranteed to work in its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, ensuring that the hybrid power system can still operate with faults while meeting the constraints.

[0119] The power redistribution health management strategy under system failure based on LSTM-MPC, the rules of the MPC control strategy are as follows:

[0120] The current time is k. The LSTM-based prediction model obtains the current k-time input and historical time information of the controlled object to obtain the future time domain k+t. p The optimal control sequence of the optimal control sequence is retained, and only the first control variable u of the optimal control sequence is retained p (k+1) and applies it to the controlled object, and all other control variables are ignored, and the above optimization process is repeated at the next sampling time. At time k, the optimization objective is expressed by formula (1):

[0121]

[0122] In formula (1), J k is the cost function, t p is the prediction duration, x(t) is the state quantity at time t, and u(t) is the control quantity at time t.

[0123] The constraint condition is expressed by formula (2):

[0124]

[0125] In formula (2), x min (t) and x max (t) are the upper and lower limits of the state variable at time t, u min (t) and u max (t) are the upper and lower limits of the control variable at time t.

[0126] According to the multi-power source characteristics of the hybrid power system, the optimization goal is to minimize the vehicle speed following error by sacrificing comprehensive energy consumption. The constraints include diesel engine exhaust temperature constraints, battery temperature constraints, and theoretical physical constraints of other components. At time k, in the prediction time domain, the cost function of the hybrid power system is expressed by formula (3):

[0127]

[0128] In formula (3), t p is the prediction duration, k p is the weight coefficient of vehicle speed difference, k e is the weight coefficient of comprehensive energy consumption, u(t) is the control quantity, here is the accelerator pedal position correction coefficient, v a (u(t),t) and v t (t) are the actual vehicle speed and target vehicle speed, km / h; is the comprehensive energy consumption, g / s.

[0129] The constraint condition is expressed by formula (4):

[0130]

[0131] In formula (4), T e is the diesel engine exhaust temperature, T e_max is the exhaust temperature limit of the diesel engine, and its exhaust temperature limit comes from the characteristic index provided by the diesel engine manufacturer; T b is the battery temperature, [T b_smin ,T b_smax ] is the optimal operating temperature range of lithium-ion batteries, ranging from [15,35], ℃; SOC min , SOC max are the upper and lower limits of the power battery charge and discharge SOC, which are 0.36 and 0.6 respectively; P dem , P e and P m They are total required power, diesel engine power and motor power, kW respectively.

[0132] The input and output of the LSTM prediction model are selected according to the common application scenarios of hybrid power systems. The input is the health status assessment result, vehicle operation mode, accelerator pedal position, battery power, actual vehicle speed, battery temperature and diesel engine exhaust temperature. The output is the vehicle speed, diesel engine exhaust temperature and battery temperature of the hybrid system. The system's vehicle speed, battery temperature and diesel engine exhaust temperature at future moments are rolled out through the input information of the historical time period and the current moment. The first 85% of the offline data is selected as the training set, and the last 15% is used as the test set for training. The mean absolute error (MAE) is selected as the loss function to optimize the parameters of the prediction model and measure the accuracy of the prediction model. MAE can be expressed by formula (5):

[0133]

[0134] In formula (5), Y is the true value, f(x) is the predicted value, and n is the number of samples.

[0135] Preferably, according to the conversion relationship between power consumption and fuel consumption given in GB / T 19754-2021 "Heavy-duty hybrid electric vehicle energy consumption test method", the battery energy consumption is equivalent to fuel energy consumption, and the comprehensive energy consumption is calculated. The target vehicle speed following effect is improved by sacrificing the comprehensive energy consumption. At the same time, the battery is ensured to operate within its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, thereby ensuring that the hybrid system can still operate with faults while meeting the constraints.

[0136] The specific description above further illustrates the purpose, technical solutions and beneficial effects of the invention in detail. It should be understood that the above is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A hybrid power system health management method, characterized by: The following steps are included: Step 1: Obtain the diesel engine instantaneous analog signal, the battery instantaneous analog signal, and the vehicle instantaneous analog signal through the signal acquisition and processing module to be tested, and obtain the corresponding hybrid powertrain state signal, and perform normalization processing on the hybrid powertrain state signal to obtain the normalized diesel engine instantaneous state signal, battery instantaneous state signal, and vehicle instantaneous state signal of the hybrid powertrain; Step 2: According to the state signal of uniform dimension input by the signal acquisition and processing module to be tested, combined with the fault form and the actual vehicle sensor signal, select some state signals of the engine, battery and vehicle in the hybrid system as characteristic signals, and the characteristic signals of the selected area include diesel engine intake pressure, diesel engine intake temperature, diesel engine exhaust temperature, accelerator pedal position, battery bus voltage, battery bus current, battery temperature and vehicle speed difference; determine the characteristic signal weight through principal component analysis, and calculate the health of the hybrid system based on the grey correlation analysis method GRA in real time; confirm the fault through a fault confirmation method based on counting, reduce the fault misjudgment and missed judgment caused by the error and noise of the characteristic signal, and obtain a fault confirmation signal; Step 3: A health management strategy for power redistribution under system faults based on LSTM-MPC is used to receive the hybrid powertrain signal obtained in step 1 as a predictive maintenance indicator. After receiving the fault confirmation signal obtained in step 2, a partial prediction model for MPC is obtained through LSTM training. The change of partial state of the system within a limited time domain is predicted based on the LSTM-MPC prediction model, and predictive maintenance of the health of the hybrid powertrain is performed in combination with the prediction result of partial state change of the hybrid powertrain system. When a hybrid power system fails, the conversion relationship between power consumption and fuel consumption is given, the battery energy consumption is equivalent to fuel energy consumption, the comprehensive energy consumption is calculated, and the target vehicle speed following effect is improved by sacrificing the comprehensive energy consumption. At the same time, the battery is guaranteed to operate within its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, ensuring that the hybrid power system can still operate with a fault while meeting the constraints.

2. A hybrid power system health management method as claimed in claim 1, characterized in that: The implementation method of step one is: Step 1.1: Obtain the original instantaneous analog signal of the vehicle component through the signal acquisition and processing module to be tested; Step 1.2: performing signal conversion and calibration on the original instantaneous analog signal obtained in step 1.1 to obtain the relationship between the original analog signal and the original digital signal, and performing filtering processing to obtain the instantaneous state signal of the vehicle; Step 1.3: Based on the original instantaneous analog signal obtained in step 1.1 and the instantaneous state signal obtained in step 1.2, a hybrid power system state signal is obtained, wherein the state signal includes the following three types of signals: diesel engine instantaneous state signal, battery instantaneous state signal, and vehicle instantaneous state signal: the diesel engine instantaneous state signal includes intake air temperature, intake air pressure, exhaust temperature, diesel engine speed, and accelerator pedal position; the battery instantaneous state signal includes SOC, battery bus voltage, battery bus current, and battery temperature; the vehicle instantaneous state signal includes target vehicle speed, actual vehicle speed, and vehicle operation mode; Step 1.4: Perform signal processing on the state signal of the hybrid powertrain obtained in step 1.3, perform normalization on the state signal of the hybrid powertrain, eliminate the influence of different dimensions, and obtain a normalized state signal of the hybrid powertrain.

3. A hybrid power system health management method as claimed in claim 2, characterized in that: In step three, The power redistribution health management strategy under system failure based on LSTM-MPC, the rules of the MPC control strategy are as follows: The current time is k. The LSTM-based prediction model obtains the current k-time input and historical time information of the controlled object to obtain the future time domain k+t. p The optimal control sequence of the optimal control sequence only retains the first control quantity u of the optimal control sequence p (k+1) and applies it to the controlled object, and all other control variables are ignored. The above optimization process is repeated at the next sampling time. At time k, the optimization objective is expressed by formula (1): In formula (1), J k is the cost function, t p is the prediction duration, x(t) is the state quantity at time t, and u(t) is the control quantity at time t; The constraint condition is expressed by formula (2): In formula (2), x min (t) and x max (t) are the upper and lower limits of the state variable at time t, u min (t) and u max (t) are the upper and lower limits of the control variable at time t; According to the multi-power source characteristics of the hybrid power system, the optimization goal is to minimize the vehicle speed following error by sacrificing comprehensive energy consumption; the constraints include diesel engine exhaust temperature constraints, battery temperature constraints and theoretical physical constraints of other components; at time k, in the prediction time domain, the cost function of the hybrid power system is expressed by formula (3): In formula (3), t p is the prediction duration, k p is the weight coefficient of vehicle speed difference, k e is the weight coefficient of comprehensive energy consumption, u(t) is the control quantity, here is the accelerator pedal position correction coefficient, v a (u(t),t) and v t (t) are the actual vehicle speed and target vehicle speed, km / h; is the comprehensive energy consumption, g / s; The constraint condition is expressed by formula (4): In formula (4), T e is the diesel engine exhaust temperature, T e_max is the diesel engine exhaust temperature limit value; T b is the battery temperature, [T b_smin ,T b_smax ] is the optimal operating temperature range of lithium-ion batteries, ranging from [15,35], ℃; SOC min , SOC max They are the upper and lower limits of the power battery charge and discharge SOC; P dem , P e and P m are the total required power, diesel engine power and motor power, kW respectively; The input of the LSTM prediction model is the health status assessment result, vehicle operation mode, accelerator pedal position, battery power, actual vehicle speed, battery temperature and diesel engine exhaust temperature; the output is the vehicle speed, diesel engine exhaust temperature and battery temperature of the hybrid system; the vehicle speed, battery temperature and diesel engine exhaust temperature of the system at future moments are predicted by rolling based on the input information of the historical time period and the current moment; the mean absolute error (MAE) is selected as the loss function to optimize the parameters of the prediction model and measure the accuracy of the prediction model; MAE is expressed by formula (5): In formula (5), Y is the true value, f(x) is the predicted value, and n is the number of samples.

4. A hybrid power system health management method as claimed in claim 4, characterized in that: In step three, according to the conversion relationship between power consumption and fuel consumption given in GB / T 19754-2021 "Heavy-duty hybrid electric vehicle energy consumption test method", the battery energy consumption is equivalent to fuel energy consumption, and the comprehensive energy consumption is calculated. By sacrificing the comprehensive energy consumption, the target vehicle speed following effect is improved. At the same time, the battery is ensured to operate within its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, ensuring that the hybrid system can still operate with faults while meeting the constraints.

5. A hybrid power system health management system, used to implement a hybrid power system health management method as claimed in claim 1, 2, 3 or 4, characterized in that: It is mainly composed of a host computer PC, a hybrid powertrain, a vehicle controller lower computer VCU, a test signal acquisition and processing module, a health status assessment module, and a power distribution correction module based on LSTM-MPC; the host computer PC and the vehicle controller lower computer VCU are connected through a signal line; the hybrid powertrain is connected to the vehicle controller lower computer VCU through a signal bus, and the host computer PC controls the vehicle controller lower computer VCU to receive and process signals from the hybrid powertrain, and conducts real-time signal transmission interaction. Based on the test signal acquisition and processing module, health status assessment module, and LSTM-MPC-based power distribution correction module written in the vehicle controller lower computer VCU, the health status of the hybrid powertrain is evaluated, and the power distribution of the hybrid powertrain is corrected according to the health status assessment results; The signal acquisition and processing module receives and processes the hybrid power system status signal, which includes three types of signals: diesel engine characteristic signal, battery characteristic signal, and vehicle characteristic signal; the diesel engine characteristic signal includes intake air temperature, intake air pressure, exhaust temperature, diesel engine speed, and accelerator pedal position; the battery characteristic signal includes SOC, battery bus voltage, battery bus current, and battery temperature; The vehicle characteristic signals include the target speed, actual speed and vehicle operation mode. After acquiring the signal, the state signal is normalized to eliminate the influence of different dimensions and pre-process the subsequent health status assessment module. By unifying the state signal dimensions, the health status assessment efficiency of the hybrid system is improved while ensuring the accuracy of characteristic signals at different orders of magnitude. According to formula (6), the time series data such as intake pressure, intake temperature, exhaust temperature, accelerator pedal position, bus current, bus voltage, battery temperature and vehicle speed are normalized: Where i = 1, 2, 3, ..., n, k = 1, 2, 3, ..., m, n is the number of samples, m is the number of features The health status assessment module obtains the status signal from the signal acquisition and processing module, calculates the health status through the vehicle controller lower computer VCU according to the health status, determines the health status of the hybrid powertrain, analyzes the hybrid powertrain working conditions and diagnoses whether the system has any faults; The health degree is constructed according to the above normalized characteristic signals, that is, according to the intake pressure, intake temperature, exhaust temperature, accelerator pedal position, bus current, bus voltage, battery temperature and vehicle speed, and the constructed health degree is embedded in the health status assessment module; The health degree construction method is as follows: The diesel-electric hybrid system has the characteristics of high coupling, nonlinearity and multiple power sources. It is comprehensively judged through multiple dimensions and multiple standards. It combines the multi-dimensional state signals collected by the diesel-electric hybrid system and uses grey correlation analysis to calculate the correlation between the operating state and the health state of the hybrid system. The correlation degree is used to reflect the degree of deviation between the hybrid system's operating state and healthy state. The greater the deviation, the smaller the correlation degree. Health is determined based on the following method: 1) Using the normalized data output by the signal acquisition and processing module to calculate the grey correlation coefficient one by one; In formula (7), k = 1, 2, 3, ..., m, i = 1, 2, 3, ..., n, ρ represents the discrimination coefficient, ρ∈(0, 1), the smaller ρ is, the greater the difference in correlation coefficients is, and the stronger the discrimination ability is; 2) Calculate the correlation degree one by one according to formula (8); In formula (8), ω i (k) represents the weight coefficient of the characteristic signal, k = 1, 2, 3, ..., m, i = 1, 2, 3, ..., n; The health status assessment of the hybrid system based on GRA requires the calculation of correlation. Since different characteristic signals in the hybrid system have different sensitivities to faults, directly taking the average value may lead to inaccurate correlation calculation. The weight values ​​of characteristic signals of different dimensions are determined by using the PCA method. The LSTM-MPC-based power distribution correction module obtains the output and test signal from the health status assessment module, obtains the hybrid power system status in the health status assessment module, calculates the LSTM-MPC-based power redistribution health management method through the vehicle controller lower computer VCU, and calculates the accelerator pedal position correction coefficient in advance; when the fault is confirmed, the output correction coefficient is used to correct the accelerator pedal position of the vehicle controller, that is, in the case of a hybrid power system fault, the target vehicle speed following effect is improved by sacrificing the comprehensive energy consumption, while ensuring that the battery operates in its optimal temperature range and the diesel engine does not exceed the exhaust temperature limit, ensuring that the hybrid power system can still operate with a fault while meeting the constraints.

6. A hybrid power system health management system as claimed in claim 5, characterized in that: ρ is chosen to be 0.5.

Images