High-power diesel engine micro-mixing electrical system and control method thereof
By designing a high-power diesel engine micro-mixed electrical system in automobiles, optimizing engine output using the integrated and predictive model of 48-volt lithium-ion battery and 12-volt lead-acid battery, the problem of increased power consumption is solved and more efficient fuel use and Reduce exhaust emissions and improve driving experience.
Patent Information
- Application Number
- CN202411029987.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-12
AI Technical Summary
The increased demand for electricity in Hyundai cars has led to an increase in power consumption. The existing solutions, while using thicker cables and larger connectors, have increased costs and car weight, affecting the driving experience.
A high-power diesel engine micro-mixed electrical system is designed, including engine, 48-volt lithium-ion battery, 12-volt lead-acid battery and motor. The original vehicle battery system is seamlessly integrated through DC/DC transformers, and the engine and battery are optimized using prediction models. The maximum output power provides additional power support and brake energy recovery.
It effectively reduces engine load, improves fuel efficiency and exhaust emissions, reduces energy waste and production costs, and improves driving experience.
Smart Images

Figure CN118934315A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of diesel engine electrical control, and more specifically, to a high-power diesel engine mild hybrid electrical system and a control method thereof. Background Art
[0002] Against the backdrop of the rapid development of the contemporary automotive industry and the continuous innovation of technology, modern power machinery has an increasing demand for electricity. In addition to traditional starters, lights, radios, etc., the addition of new functions such as advanced driver assistance systems (ADAS) has also gradually increased the demand for electricity. For example, with the addition of more complex infotainment, electronic equipment connections, navigation systems, and dozens of assisted driving systems, the power consumption of power machinery has gradually increased. In addition, cameras, radars, sensors, and controllers all require more power than a 12V electrical system can provide.
[0003] In order to cope with the growing power load and maintain acceptable power loss, thicker cables and larger connectors are usually required. However, although this approach ensures the effectiveness of power transmission, it adds additional costs and vehicle weight, and also makes the space inside the vehicle more crowded, thus affecting the driving experience. Summary of the invention
[0004] The present invention provides a high-power diesel engine mild hybrid electrical system and a control method thereof, which solve the technical problems in the above-mentioned background technology.
[0005] The present invention provides a high-power diesel engine mild hybrid electrical system, comprising:
[0006] an engine, a first electric motor, a 48-volt lithium-ion battery, a 48-volt electric accessory, an AC / DC converter, a DC / DC transformer, a 12-volt lead-acid battery, a 12-volt electric accessory, and a second electric motor;
[0007] The output end of the engine is connected to the input end of the first motor, the output end of the first motor is respectively connected to the input end of the engine and the input end of the AC / DC converter, the output end of the AC / DC converter is connected to the input end of the 48-volt lithium-ion battery, the output end of the 48-volt lithium-ion battery is respectively connected to the input end of the 48-volt electric accessory and the input end of the DC / DC transformer, the output end of the DC / DC transformer is connected to the input end of the 12-volt lead-acid battery, the output end of the 12-volt lead-acid battery is respectively connected to the input end of the 12-volt electric accessory and the input end of the second motor, and the output end of the second motor is connected to the input end of the engine.
[0008] Furthermore, the first motor and the second motor are both engine belt driven motors.
[0009] Further, the first electric motor is powered by a 48-volt lithium-ion battery and provides additional power support when the engine stops working; the first electric motor provides additional torque to the engine to reduce the load on the engine.
[0010] Furthermore, the engine recovers kinetic energy through the first electric motor during braking, converts the kinetic energy into electrical energy and stores it in a 48-volt lithium-ion battery for use by the first electric motor.
[0011] Furthermore, the AC / DC converter is used to convert the alternating current generated by the first motor into direct current to charge the 48V lithium-ion battery.
[0012] Further, the DC / DC converter converts the power of the 48-volt lithium-ion battery into 12 volts for use by the vehicle's original 12-volt lead-acid battery and 12-volt electric accessories.
[0013] Furthermore, the present invention provides a control method for a high-power diesel engine mild hybrid electrical system, comprising the following steps:
[0014] Step S201, within a first preset time period T, vehicle data, driving instruction data and road condition data of the vehicle are collected at preset time intervals t, and a feature sequence is constructed;
[0015] Vehicle data includes: engine speed, engine coolant temperature, oil pressure, fuel level, vehicle speed and acceleration;
[0016] Driving command data includes: accelerator pedal position, brake pedal position and steering wheel rotation angle;
[0017] Road condition data includes: vehicle front image, vehicle temperature and vehicle humidity;
[0018] The i-th sequence unit of the feature sequence represents the vehicle data, driving instruction data and road condition data at the i-th time point, where 1≤i≤N, N=T / t;
[0019] Step S202, constructing and training a prediction model;
[0020] Step S203, inputting the feature sequence into the trained prediction model, and the output values respectively represent the maximum output power of the engine within the second preset time period G, the maximum output power of the first motor and the maximum output power of the 48V lithium-ion battery.
[0021] Further, the prediction model includes N hidden layers, each hidden layer includes: a first unit, an expansion unit, a concatenation unit, and a second unit, and the i-th hidden layer inputs the i-th sequence unit of the feature sequence;
[0022] The first unit is built on a convolutional neural network model;
[0023] The N first units all share weight parameters and bias parameters;
[0024] The first unit of the i-th hidden layer inputs the vehicle front image of the i-th sequence unit of the feature sequence and outputs a feature map;
[0025] The expansion unit is used to expand the feature map into a vector representation;
[0026] The concatenation unit is used to concatenate the vector representation of the feature map with the vehicle data, the driving instruction data and the road condition data to obtain a combined vector, that is, each sequence unit of the feature sequence is represented by a combined vector;
[0027] The second unit of the i-th hidden layer inputs the combination vector of the i-th sequence unit of the feature sequence and outputs the updated feature;
[0028] The updated features output by the second unit of the Nth hidden layer are respectively input to the first classifier, the second classifier and the third classifier, wherein the classification spaces of the first classifier, the second classifier and the third classifier respectively represent the maximum output power of the engine, the maximum output power of the first motor and the maximum output power of the 48V lithium-ion battery within the second preset time period G;
[0029] The calculation formula for the second unit of the i-th hidden layer includes:
[0030] The calculation formula for the i-th hidden layer includes:
[0031] F i =sigmoid(x i W f1 +h i-1 W f2 +b f );
[0032] I i =sigmoid(x i W I1 +h i-1 W I2 +b I );
[0033]
[0034] O i =sigmoid(x i W O1 +h i-1 W O2 +b O );
[0035] h i =O i ⊙tanh(Ci );
[0036] where F i ,I i , C i , O i and h i Respectively represent the forget gate, input gate, intermediate state, intermediate hidden state, output gate and update feature of the i-th hidden layer, W f1 , W f2 and b f They represent the first weight parameter, the second weight parameter and the bias parameter of the forget gate of the i-th hidden layer, respectively. I1 , W I2 and b I They represent the first weight parameter, the second weight parameter and the bias parameter of the input gate of the i-th hidden layer, respectively, and W C1 , W C2 and b C Respectively represent the first weight parameter, the second weight parameter and the bias parameter of the intermediate state of the i-th hidden layer, W O1 , W O2 and b O Respectively represent the first weight parameter, the second weight parameter and the bias parameter of the output gate of the i-th hidden layer, x i represents the frequency domain waveform of the i-th time point of the time-frequency sequence input to the i-th hidden layer, h i-1 represents the updated features of the i-1th hidden layer, C i-1 represents the intermediate hidden state of the i-1th hidden layer, ⊙ represents point-by-point multiplication, sigmoid represents the sigmoid activation function, tanh represents the hyperbolic tangent activation function, and h 0 =0, C 0 =0.
[0037] Furthermore, obtaining sample labels of training samples for training the prediction model through a simulation platform includes the following steps:
[0038] Step S301, generating a feature sequence according to step S201 through a simulation platform;
[0039] Step S302, evaluating the maximum output power of the engine, the maximum output power of the first motor, and the maximum output power of the 48V lithium-ion battery set within the second preset time period G by three relevant experts;
[0040] Step S303, recording the vehicle fuel consumption value in the second preset time period G through the simulation platform, and selecting the maximum output power of the engine, the maximum output power of the first motor and the maximum output power of the 48V lithium-ion battery in the second preset time period G corresponding to the minimum vehicle fuel consumption value as a sample label of a training sample;
[0041] Step S304, repeating steps S301 to S303 until sample labels of Q training samples are obtained, where Q is a custom parameter.
[0042] Furthermore, before the prediction model is trained, the first unit is pre-trained, and during the pre-training process, the fourth classifier is connected, and the classification space of the fourth classifier represents the situation in front of the vehicle, including: sunny, rainy, snowy and foggy.
[0043] The beneficial effects of the present invention are as follows: the present invention integrates the diesel engine and the 48-volt battery, and seamlessly integrates the original vehicle's 12-volt battery and its accessories through a DC / DC transformer. The 48-volt battery provides additional power support for the diesel engine, effectively reducing the engine load, thereby improving fuel efficiency and reducing exhaust emissions. In addition, the present invention can also be charged through the brake energy recovery function, effectively reducing energy waste during braking, and controlling the maximum output power of the diesel engine and the 48-volt battery through a predictive model, thereby effectively improving fuel efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 is a schematic diagram of a high-power diesel engine mild hybrid electrical system of the present invention;
[0045] Figure 2 is a flow chart of a control method of a high-power diesel engine mild hybrid electrical system of the present invention;
[0046] Figure 3 It is a flow chart of the present invention for obtaining sample labels of training samples for training a prediction model through a simulation platform.
[0047] In the figure: engine (101), first motor (102), 48 volt lithium-ion battery (103), 48 volt electric accessories (104), AC / DC converter (105), DC / DC transformer (106), 12 volt lead-acid battery (107), 12 volt electric accessories (108) and second motor (109). DETAILED DESCRIPTION
[0048] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.
[0049] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in one or more embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0050] like Figure 1-Figure 3 As shown, the high-power diesel engine mild hybrid electrical system includes:
[0051] Engine 101, first motor 102, 48V lithium-ion battery 103, 48V electric accessories 104, AC / DC converter 105, DC / DC transformer 106, 12V lead-acid battery 107, 12V electric accessories 108 and second motor 109;
[0052] The output end of the engine 101 is connected to the input end of the first motor 102, the output end of the first motor 102 is respectively connected to the input end of the engine 101 and the input end of the AC / DC converter 105, the output end of the AC / DC converter 105 is connected to the input end of the 48V lithium-ion battery 103, the output end of the 48V lithium-ion battery 103 is respectively connected to the input end of the 48V electric accessory 104 and the input end of the DC / DC transformer 106, the output end of the DC / DC transformer 106 is connected to the input end of the 12V lead-acid battery 107, the output end of the 12V lead-acid battery 107 is respectively connected to the input end of the 12V electric accessory 108 and the input end of the second motor 109, and the output end of the second motor 109 is connected to the input end of the engine 101.
[0053] In one embodiment of the present invention, the first motor 102 and the second motor 109 are both BSGs (Belt Driven Motors).
[0054] In one embodiment of the present invention, the first motor 102 and the second motor 109 may also both be ISGs (Integrated Drive Generators).
[0055] In one embodiment of the present invention, the first motor 102 is powered by a 48V lithium-ion battery 103 and provides additional power support when the engine 101 stops working; the first motor 102 provides additional torque to the engine 101 to reduce the load on the engine, thereby improving fuel efficiency and reducing exhaust emissions.
[0056] In one embodiment of the present invention, the engine 101 recovers kinetic energy through the first motor 102 during braking, converts the kinetic energy into electrical energy and stores it in the 48V lithium-ion battery 103 for use by the first motor 102; this energy recovery system can effectively reduce energy waste during braking and convert it into output power.
[0057] In one embodiment of the present invention, the AC / DC converter 105 is used to convert the alternating current generated by the first motor 102 into direct current to charge the 48V lithium-ion battery 103; this power conversion not only improves energy efficiency, but also ensures that the battery can effectively store and release energy.
[0058] In one embodiment of the present invention, the DC / DC converter 106 converts the electrical energy of the 48-volt lithium-ion battery 103 into 12 volts for use by the vehicle's original 12-volt lead-acid battery 107 and 12-volt electric accessories 108; this power conversion ensures that the traditional electrical system of the power machine can be seamlessly integrated with the new 48-volt system and work together.
[0059] In one embodiment of the present invention, the 12-volt lead-acid battery 107 continues to function when the engine 101 is cold started and supports the original 12-volt electric accessories 108, ensuring the reliability and stability of the power machinery.
[0060] It should be noted that the high-power diesel engine mild hybrid electrical system provided by the present invention can reduce the load of the engine during acceleration and braking, thereby improving fuel efficiency, and reduce engine exhaust emissions, thereby reducing environmental pollution, due to the support of the electric motor and the brake energy recovery function. In addition, due to the support of the electric motor, more sufficient power response can be provided at high torque, thereby improving the driving experience. Compared with the complex manufacturing process of hybrid and pure electric vehicles, for example, traditional hybrid power requires the use of high-voltage batteries, the high-power diesel engine mild hybrid electrical system provided by the present invention can use lower voltage batteries, thereby effectively reducing costs and improving production efficiency.
[0061] In one embodiment of the present invention, Figure 2 As shown, the present invention provides a control method for a high-power diesel engine mild hybrid electrical system, comprising the following steps:
[0062] Step S201, within a first preset time period T, vehicle data, driving instruction data and road condition data of the vehicle are collected at preset time intervals t, and a feature sequence is constructed;
[0063] Vehicle data includes: engine speed, engine coolant temperature, oil pressure, fuel level, vehicle speed and acceleration;
[0064] Driving command data includes: accelerator pedal position, brake pedal position and steering wheel rotation angle;
[0065] Road condition data includes: vehicle front image, vehicle temperature and vehicle humidity;
[0066] The i-th sequence unit of the feature sequence represents the vehicle data, driving instruction data and road condition data at the i-th time point, where 1≤i≤N, N=T / t;
[0067] Step S202, constructing and training a prediction model;
[0068] Step S203, inputting the feature sequence into the trained prediction model, and the output values respectively represent the maximum output power of the engine 101, the maximum output power of the first motor 102 and the maximum output power of the 48V lithium-ion battery 103 within the second preset time period G.
[0069] In one embodiment of the present invention, the driving command data may also include: battery voltage, remaining battery power and vehicle tilt angle, etc.; the driving command data may also include: engine start or shutdown instructions and current transmission gear, etc.; the road condition data may also include: vehicle rear image, real-time traffic flow and visibility, etc.
[0070] In one embodiment of the present invention, the prediction model includes N hidden layers, each hidden layer includes: a first unit, an expansion unit, a concatenation unit, and a second unit, the i-th hidden layer inputs the i-th sequence unit of the feature sequence;
[0071] The N first units all share weight parameters and bias parameters;
[0072] The first unit of the i-th hidden layer inputs the vehicle front image of the i-th sequence unit of the feature sequence and outputs a feature map;
[0073] The expansion unit is used to expand the feature map into a vector representation;
[0074] The concatenation unit is used to concatenate the vector representation of the feature map with the vehicle data, the driving instruction data and the road condition data to obtain a combined vector, that is, each sequence unit of the feature sequence is represented by a combined vector;
[0075] The second unit of the i-th hidden layer inputs the combination vector of the i-th sequence unit of the feature sequence and outputs the updated feature;
[0076] The updated features output by the second unit of the Nth hidden layer are respectively input to the first classifier, the second classifier and the third classifier, and the classification spaces of the first classifier, the second classifier and the third classifier respectively represent the maximum output power of the engine 101, the maximum output power of the first motor 102 and the maximum output power of the 48V lithium-ion battery 103 within the second preset time period G.
[0077] In one embodiment of the present invention, the first unit is constructed based on a convolutional neural network model.
[0078] In one embodiment of the present invention, the calculation formula of the second unit of the i-th hidden layer includes:
[0079] The calculation formula for the i-th hidden layer includes:
[0080] F i =sigmoid(x i W f1 +h i-1 W f2 +b f );
[0081] I i =sigmoid(x i W I1 +h i-1 W I2 +b I );
[0082]
[0083] O i =sigmoid(x i W O1 +h i-1 W O2 +b O );
[0084] h i =O i ⊙tanh(C i );
[0085] where F i ,I i , C i, O i and h i Respectively represent the forget gate, input gate, intermediate state, intermediate hidden state, output gate and update feature of the i-th hidden layer, W f1 , W f2 and b f They represent the first weight parameter, the second weight parameter and the bias parameter of the forget gate of the i-th hidden layer, respectively. I1 , W I2 and b I They represent the first weight parameter, the second weight parameter and the bias parameter of the input gate of the i-th hidden layer, respectively, and W C1 , W C2 and b C Respectively represent the first weight parameter, the second weight parameter and the bias parameter of the intermediate state of the i-th hidden layer, W O1 , W O2 and b O Respectively represent the first weight parameter, the second weight parameter and the bias parameter of the output gate of the i-th hidden layer, x i represents the frequency domain waveform of the i-th time point of the time-frequency sequence input to the i-th hidden layer, h i-1 represents the updated features of the i-1th hidden layer, C i-1 represents the intermediate hidden state of the i-1th hidden layer, ⊙ represents point-by-point multiplication, sigmoid represents the sigmoid activation function, tanh represents the hyperbolic tangent activation function, and h 0 =0, C 0 =0.
[0086] In one embodiment of the present invention, Figure 3 As shown, obtaining sample labels of training samples for training a prediction model through a simulation platform includes the following steps:
[0087] Step S301, generating a feature sequence according to step S201 through a simulation platform;
[0088] Step S302, evaluating the maximum output power of the engine 101, the maximum output power of the first motor 102, and the maximum output power of the 48V lithium-ion battery 103 set within the second preset time period G by three relevant experts;
[0089] Step S303, recording the vehicle fuel consumption value in the second preset time period G through the simulation platform, and selecting the maximum output power of the engine 101, the maximum output power of the first motor 102, and the maximum output power of the 48V lithium-ion battery 103 in the second preset time period G corresponding to the minimum vehicle fuel consumption value as a sample label of a training sample;
[0090] Step S304, repeating steps S301 to S303 until sample labels of Q training samples are obtained, where Q is a custom parameter, and preferably, Q is set to 2000.
[0091] In one embodiment of the present invention, in each iteration of training the prediction model, the difference between the output value of the prediction model and the value of the sample label of the training sample is specified as the loss function, and the weight parameters and bias parameters of the prediction model are updated by the gradient descent algorithm.
[0092] In one embodiment of the present invention, before the prediction model is trained, the first unit is pre-trained, and during the pre-training process, the fourth classifier is connected, and the classification space of the fourth classifier represents the conditions in front of the vehicle, including: sunny, rainy, snowy and foggy.
[0093] It should be noted that pre-training the first unit can better extract image features, and fine-tuning the parameters after pre-training can speed up the training of the prediction model and improve the prediction accuracy of the prediction model.
[0094] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are protected by the present embodiment.
Claims
1. High-power diesel engine mild hybrid electrical system, characterized by: include: An engine (101), a first electric motor (102), a 48-volt lithium-ion battery (103), a 48-volt electric accessory (104), an AC / DC converter (105), a DC / DC transformer (106), a 12-volt lead-acid battery (107), a 12-volt electric accessory (108), and a second electric motor (109); The output end of the engine (101) is connected to the input end of the first motor (102), the output end of the first motor (102) is respectively connected to the input end of the engine (101) and the input end of the AC / DC converter (105), the output end of the AC / DC converter (105) is connected to the input end of the 48V lithium-ion battery (103), the output end of the 48V lithium-ion battery (103) is respectively connected to the input end of the 48V electric accessory (104) and the input end of the DC / DC transformer (106), the output end of the DC / DC transformer (106) is connected to the input end of the 12V lead-acid battery (107), the output end of the 12V lead-acid battery (107) is respectively connected to the input end of the 12V electric accessory (108) and the input end of the second motor (109), and the output end of the second motor (109) is connected to the input end of the engine (101).
2. The high-power diesel engine mild hybrid electrical system according to claim 1, characterized in that: The first motor (102) and the second motor (109) are both engine belt driven motors.
3. The high-power diesel engine mild hybrid electrical system according to claim 1, characterized in that: The first electric motor (102) is powered by the 48V lithium-ion battery (103) and provides additional power support when the engine (101) stops working; the first electric motor (102) provides additional torque for the engine (101) to reduce the load of the engine.
4. The high-power diesel engine mild hybrid electrical system according to claim 1, characterized in that: The engine (101) recovers kinetic energy through the first motor (102) during braking, converts the kinetic energy into electrical energy and stores it in the 48V lithium-ion battery (103) for use by the first motor (102).
5. The high-power diesel engine mild hybrid electrical system according to claim 1, characterized in that: The AC / DC converter (105) is used to convert the alternating current generated by the first motor (102) into direct current to charge the 48V lithium-ion battery (103).
6. The high-power diesel engine mild hybrid electrical system according to claim 1, characterized in that: The DC / DC converter (106) converts the electric energy of the 48V lithium-ion battery (103) into 12V for use by the original 12V lead-acid battery (107) and 12V electric accessories (108) of the vehicle.
7. A control method for a high-power diesel engine mild hybrid electrical system, executing the high-power diesel engine mild hybrid electrical system as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: Step S201, within a first preset time period T, vehicle data, driving instruction data and road condition data of the vehicle are collected at preset time intervals t, and a feature sequence is constructed; Vehicle data includes: engine speed, engine coolant temperature, oil pressure, fuel level, vehicle speed and acceleration; Driving command data includes: accelerator pedal position, brake pedal position and steering wheel rotation angle; Road condition data includes: vehicle front image, vehicle temperature and vehicle humidity; The i-th sequence unit of the feature sequence represents the vehicle data, driving instruction data and road condition data at the i-th time point, where 1≤i≤N, N=T / t; Step S202, constructing and training a prediction model; Step S203, inputting the feature sequence into the trained prediction model, and the output values respectively represent the maximum output power of the engine (101), the maximum output power of the first motor (102) and the maximum output power of the 48V lithium-ion battery (103) within the second preset time period G.
8. The control method of the high-power diesel engine mild hybrid electrical system according to claim 7, characterized in that: The prediction model includes N hidden layers, each hidden layer includes: a first unit, an expansion unit, a concatenation unit, and a second unit, and the i-th hidden layer inputs the i-th sequence unit of the feature sequence; The first unit is built on a convolutional neural network model; The N first units all share weight parameters and bias parameters; The first unit of the i-th hidden layer inputs the vehicle front image of the i-th sequence unit of the feature sequence and outputs a feature map; The expansion unit is used to expand the feature map into a vector representation; The concatenation unit is used to concatenate the vector representation of the feature map with the vehicle data, the driving instruction data and the road condition data to obtain a combined vector, that is, each sequence unit of the feature sequence is represented by a combined vector; The second unit of the i-th hidden layer inputs the combination vector of the i-th sequence unit of the feature sequence and outputs the updated feature; The updated features output by the second unit of the Nth hidden layer are respectively input to the first classifier, the second classifier and the third classifier, wherein the classification spaces of the first classifier, the second classifier and the third classifier respectively represent the maximum output power of the engine (101), the maximum output power of the first motor (102) and the maximum output power of the 48V lithium-ion battery (103) within the second preset time period G; The calculation formula for the second unit of the i-th hidden layer includes: The calculation formula for the i-th hidden layer includes: F i =sigmoid(x i W f1 +h i-1 W f2 +b f ); I i =sigmoid(x i W I1 +h i-1 W I2 +b I ); O i =sigmoid(x i W O1 +h i-1 W O2 +b O ); h i =O i ⊙tanh(C i ); where F i ,I i , C i , O i and h i Respectively represent the forget gate, input gate, intermediate state, intermediate hidden state, output gate and update feature of the i-th hidden layer, W f1 , W f2 and b f They represent the first weight parameter, the second weight parameter and the bias parameter of the forget gate of the i-th hidden layer, respectively. I1 , W I2 and b I They represent the first weight parameter, the second weight parameter and the bias parameter of the input gate of the i-th hidden layer, respectively, and W C1 , W C2 and b C Respectively represent the first weight parameter, the second weight parameter and the bias parameter of the intermediate state of the i-th hidden layer, W O1 , W O2 and b O Respectively represent the first weight parameter, the second weight parameter and the bias parameter of the output gate of the i-th hidden layer, x i represents the frequency domain waveform of the i-th time point of the time-frequency sequence input to the i-th hidden layer, h i-1 represents the updated features of the i-1th hidden layer, C i-1 represents the intermediate hidden state of the i-1th hidden layer, ⊙ represents point-by-point multiplication, sigmoid represents the sigmoid activation function, tanh represents the hyperbolic tangent activation function, h0=0, C0=0.
9. The control method of a high-power diesel engine mild hybrid electrical system according to claim 8, characterized in that: Obtaining sample labels of training samples for training a prediction model through a simulation platform includes the following steps: Step S301, generating a feature sequence according to step S201 through a simulation platform; Step S302, evaluating the maximum output power of the engine (101), the maximum output power of the first motor (102) and the maximum output power of the 48V lithium-ion battery (103) set within a second preset time period G by three relevant experts; Step S303, recording the vehicle fuel consumption value within the second preset time period G through a simulation platform, and selecting the maximum output power of the engine (101), the maximum output power of the first motor (102), and the maximum output power of the 48V lithium-ion battery (103) within the second preset time period G corresponding to the minimum vehicle fuel consumption value as a sample label of a training sample; Step S304, repeating steps S301 to S303 until sample labels of Q training samples are obtained, where Q is a custom parameter.
10. The control method of a high-power diesel engine mild hybrid electrical system according to claim 7, characterized in that: Before the prediction model is trained, the first unit is pre-trained, and the fourth classifier is connected during the pre-training process. The classification space of the fourth classifier represents the situation in front of the vehicle, including: sunny, rainy, snowy and foggy.
