Heavy truck gradient self-adaptive predictive energy-saving control method

Abstract

The invention discloses a gradient self-adaptive predictive energy-saving control method for a heavy truck. The gradient self-adaptive predictive energy-saving control method comprises the following steps: step 1, establishing a longitudinal dynamical model of the heavy truck; step 2, establishing an engine fuel consumption characteristic model; 3, an NMPC controller of the heavy truck power system is established, a multi-objective cost function is constructed, and the multi-objective cost function comprises a speed traceable weight w1, a fuel economy weight w2 and a torque change weight w3; step 4, constructing a neural network for capturing a mapping relationship between the gradient sequence and a fuel economy weight w2, wherein the neural network outputs an optimal fuel economy weight w2 to an NMPC controller; step 5, constructing a feature data set for neural network training; step 6, establishing an automatic simulation scheduling system; step 7, the automatic simulation scheduling system generates a label; and step 8, training the neural network by using a data set formed by the label and the long-scale gradient sequence. According to the invention, the fuel efficiency under variable gradients is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of power systems, and in particular to a gradient adaptive predictive energy-saving control method for heavy trucks. Background Art

[0002] In road transportation, heavy trucks account for a significant portion of fuel consumption, and their fuel efficiency is greatly affected by road conditions, especially road slope changes. To address this issue, it is critical to obtain the road slope in advance and use this information to adjust the controller so that the vehicle's power system can respond quickly and effectively to slope changes during cruising. With the development of digital maps, the integration of slope information has increased the possibility of energy-saving driving based on slope data. However, cruise control for trucks faces new challenges, such as balancing multiple objectives (vehicle dynamics, fuel economy, and ride comfort), and maintaining control performance under different slope conditions.

[0003] In response to the above problems, experts and scholars have conducted in-depth research, and a variety of control schemes have been proposed, which can be mainly divided into the following categories: rule-based control methods, optimization-based control methods, and learning-based control methods. The rule-based method mainly abstracts the mathematical description of the law based on driving experience, and uses the law to design control strategies to achieve energy saving. However, the above rule-based strategies are highly dependent on manual experience, which limits the full realization of the energy-saving potential of vehicles. The optimization-based control method originated from the field of optimal control, and its mainstream model predictive control (MPC) is used to solve the optimal solution under the energy-saving goal. However, the computational complexity of the MPC algorithm is high, which limits the length of its rolling prediction range. This makes it difficult to obtain global information in long-scale optimization, thereby limiting the overall performance. In recent years, in the field of predictive energy-saving control, learning-based control methods are gradually emerging. Artificial intelligence technologies such as deep learning and reinforcement learning have developed rapidly. They have demonstrated powerful data feature mining capabilities and provided a new way to explore and utilize slope information for energy saving. However, these learning-based methods rely on a large amount of high-quality training data, and the data collection process is time-consuming and costly. In addition, lack of interpretability is another factor that hinders its practical application.

[0004] In summary, the MPC method has good robustness and interpretability, but it is insufficient in capturing long-scale features; while the deep learning method performs well in extracting feature information (including long-scale information) from a variety of data types, but its interpretability is poor and data collection faces challenges. In view of this, this study aims to combine these two methods to give full play to their respective advantages and achieve complementarity. In addition, in response to the difficulty of training data collection, an automatic simulation scheduling system based on a large language model (LLM) agent is introduced to replace manual simulation and data collection, demonstrating the potential of LLM in controller optimization. Summary of the invention

[0005] The purpose of the present invention is to provide a method for predictive energy-saving control of heavy trucks based on the technical defects in the predictive energy-saving control in the prior art. The controller realizes the adaptation of the long-scale slope based on the neural network, dynamically adjusts the controller parameters, and finally achieves the purpose of saving fuel. In addition, an automatic simulation scheduling system based on a large language model (LLM) intelligent agent is introduced to replace manual execution of simulation and data collection, solving the problem of difficulty and high cost of data collection for neural network training.

[0006] The technical solution adopted to achieve the purpose of the present invention is:

[0007] A method for adaptive predictive energy-saving control of a heavy truck slope with LLM-assisted data collection, comprising the following steps:

[0008] Step 1, establishing a longitudinal dynamic model of a heavy truck on a slope road;

[0009] Step 2, establish the engine fuel consumption characteristic model and obtain the brake specific fuel consumption BSFC;

[0010] Step 3, establish a NMPC controller for the heavy truck power system to accurately control the engine torque and feed it back to the control-oriented model composed of the longitudinal dynamics model in step 1 and the engine fuel consumption characteristic model in step 2, and construct a multi-objective cost function to take into account speed tracking and fuel consumption. The multi-objective cost function includes a speed traceability weight w1, a fuel economy weight w2, and a torque change weight w3;

[0011] Step 4, constructing a neural network that captures the mapping relationship between the slope sequence and the fuel economy weight w2 in step 2, and the neural network outputs the optimal fuel economy weight w2 to the NMPC controller in step 3;

[0012] Step 5, constructing a feature data set for neural network training, namely, a long-scale slope sequence;

[0013] Step 6, establish an automated simulation scheduling system consisting of an execution agent, an analysis agent, and a simulation engine. The execution agent generates an execution command in the w2 scanning direction according to the long-scale slope sequence of step 5 and the discrete scanning range of the fuel economy weight w2 set manually. The simulation engine converts the execution command into the underlying executable command of the simulation software and the specific w2 scanning simulation process, forms a multimodal simulation result and sends it to the analysis agent. The analysis agent analyzes the multimodal simulation result and generates formatted feedback to help adjust the action of the execution agent in the next round.

[0014] Step 7, the automated simulation scheduling system generates a label, where the label is the optimal fuel economy weight w2 corresponding to each long-scale slope sequence;

[0015] Step 8, training and testing the neural network using a training data set and a test data set consisting of the labels obtained in step 7 and the long-scale slope sequence obtained in step 5.

[0016] In the above technical solution, the longitudinal dynamic model in step 1 is:

[0017]

[0018] Where m is the vehicle mass, θ is the road slope, ρ is the air density, C d is the air resistance coefficient, A a is the frontal area of ​​the vehicle, v is the vehicle speed, f is the rolling resistance coefficient, F drive is the driving force that drives the vehicle to move, F aero is the air resistance, F roll is the rolling resistance, F slope is the slope resistance.

[0019] In the above technical solution, in step 2, the engine fuel consumption characteristic model is:

[0020]

[0021] Among them, α is the fuel consumption characteristic fitting coefficient, T e is the engine torque, n e is the engine speed.

[0022] In the above technical solution, in step 3, the process of establishing the NMPC controller of the heavy truck power system is as follows:

[0023]

[0024] in, is the time constant of the torque inertia link, The control quantity of the NMPC optimal control problem is the target engine torque;

[0025] The state variables of the optimal control problem are vehicle speed v and travel distance s d :

[0026]

[0027] Where i0 is the engine differential ratio, R w is the wheel radius of a heavy truck.

[0028] In the above technical solution, in step 3, the multi-objective cost function is:

[0029] J=J v +J fuel +J change (5)

[0030] J v Tracking metrics for speed:

[0031] J v =w1·(v ref -v) 2 (6)

[0032] w1 is the speed traceability weight, v ref is the vehicle cruising reference speed, v is the vehicle speed;

[0033] J fuel Fuel economy index:

[0034] J fuel =w2·f(T e ,n e ) (7)

[0035] w2 is the fuel economy weight;

[0036] Is the stability indicator:

[0037] J change =w3·ΔT e (8)

[0038] Where w3 is the torque change weight, ΔT e is the torque change.

[0039] In the above technical solution, in step 3, the engine torque variation range is constrained as follows:

[0040] T emin ≤T e ≤T emax (n e ) (10)

[0041] Where T emin is a preset constant, T emax Calculated based on engine characteristics, it is related to engine speed n e Related functions.

[0042] In the above technical solution, the neural network in step 4 is a CNN-LSTM network.

[0043] In the above technical solution, the architecture of the neural network in step 4 includes an input layer for receiving future slope sequences, two CNN layers for feature extraction, an LSTM layer for capturing time dependencies, and a fully connected layer for classifying output controller parameters w2.

[0044] In the above technical solution, in step 7, the automated simulation scheduling system performs simulation according to the discrete scanning range of the fuel economy w2 input by the user, and the discrete scanning range of the fuel economy weight w2 is manually preset based on experience;

[0045] w 2,i ∈{w 2,1 ,w 2,2 ,w 2,3 ,…,w 2,10}(11)

[0046] After completing the parameter scan (11) for each slope, the performance is evaluated by indicators such as fuel consumption and vehicle stability, as shown below:

[0047] E total =αE fuel +βE stability (12)

[0048] Among them, α and β are the weights of fuel consumption and vehicle stability index respectively. fuel and E stability The magnitudes of the indicators are different. To ensure that the value of each indicator is between 0 and 1, the normalization method is used.

[0049] The fuel consumption index is measured in terms of consumption per 100 kilometers, recorded as

[0050]

[0051] Where T is the total duration of a single simulation, is the fuel consumption rate per unit time, ρ fuel is the fuel density, s is the total driving distance in the simulation;

[0052] The vehicle stability index is expressed as follows:

[0053]

[0054] Where v is the real-time vehicle speed during the simulation.

[0055] In the above technical solution, in step 8, a cross entropy loss function is used to train the neural network.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. This paper proposes a slope adaptive predictive energy-saving control method for heavy trucks with LLM-assisted data collection, aiming to improve fuel efficiency under variable slopes. This paper uses a CNN-LSTM network to dynamically adjust the fuel economy weight w2 according to long-scale slope information, breaking through the limitation of the traditional NMPC method that only relies on short-scale slope information.

[0058] 2. Aiming at the high time and labor cost in the process of network training data collection, the present invention designs an automated simulation scheduling system based on LLM agents to automatically execute simulation and record data. This innovative LLM application enables the agent to guide the parameter scanning simulation process with the help of expert knowledge, effectively focusing on key areas and avoiding unnecessary calculations. It greatly reduces the time and labor investment in training data collection.

[0059] In summary, the LLM-assisted data collection heavy truck slope adaptive predictive energy-saving control method proposed in the present invention has good comprehensive performance, and can dynamically adjust the NMPC fuel economy weight w2 through the CNN-LSTM network, so that the controller can pay attention to the long-scale slope information, making the control effect closer to the global optimum and achieving the purpose of fuel saving. In addition, the automated simulation scheduling system composed of intelligent agents equipped with LLM can also make the collection of neural network training data faster and greatly reduce manpower investment. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flow chart of the slope adaptive predictive energy-saving control method for heavy trucks with LLM-assisted data collection.

[0061] Figure 2 It is a longitudinal dynamic model analysis.

[0062] Figure 3 It is the fitting result of engine fuel consumption characteristic BSFC.

[0063] Figure 4 is the architecture of a neural network.

[0064] Figure 5 It is an automated simulation scheduling system framework. DETAILED DESCRIPTION

[0065] The present invention is further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0066] like Figure 1 As shown, a slope adaptive predictive energy-saving control method for a heavy truck with LLM-assisted data collection includes the following steps:

[0067] Step 1: Based on Newton's second law, a longitudinal dynamic model of a heavy truck on a slope road is established. Figure 2 As shown, the longitudinal dynamics model of the vehicle is as follows:

[0068]

[0069] Where m is the vehicle mass (kg), θ is the road slope (rad), and ρ is the air density (kg / m 3 ), C d is the air resistance coefficient, A a is the frontal area of ​​the vehicle (m 2 ), v is the vehicle speed (m / s), f is the rolling resistance coefficient, F drive is the driving force that drives the vehicle to move (N), F aero is the air resistance (N), F roll is the rolling resistance (N), F slope is the slope resistance (N).

[0070] Step 2, based on the engine dynamometer test, the engine fuel consumption characteristics are obtained, and the brake specific fuel consumption (BSFC) is obtained by using a fifth-order polynomial fitting, as shown in Formula 2.

[0071]

[0072] Among them, α is the fuel consumption characteristic fitting coefficient, T e is the engine torque (N·m), n e is the engine speed (r / min).

[0073] Figure 3 The figure is the fitting result of the engine fuel consumption characteristic BSFC. The red line is the engine external characteristic torque curve, that is, the maximum torque curve of the engine at different speeds. The green line represents the constant power line, and the blue line represents the best fuel economy line, also known as the E line.

[0074] Step 3: Build the NMPC controller of the heavy truck power system to achieve precise control of the engine torque. In order to ensure the accuracy of the heavy truck power system, the inertia of the power system needs to be considered:

[0075]

[0076] in, is the time constant of the torque inertia link (s), is the control variable of the NMPC optimal control problem, i.e., the target engine torque (N·m).

[0077] The state variables of the optimal control problem are vehicle speed v and travel distance s d The state space equation of the optimal control problem can be expressed as:

[0078]

[0079] Where i0 is the engine differential ratio, R w is the wheel radius of heavy truck (m).

[0080] In the predictive cruise control system, the two performance indicators of speed tracking and fuel consumption conflict with each other, forming a multi-objective optimization problem. The multi-objective cost function is:

[0081] J=J v +J fuel +J change (5)

[0082] Among them, the designed speed tracking index is

[0083] J v =w1·(v ref -v) 2 (6)

[0084] Where w1 is the speed traceability weight, v ref is the vehicle cruising reference speed (m / s).

[0085] In vehicle cruise control, the evaluation of fuel economy requires consideration of both fuel consumption and vehicle dynamic performance, using the above-mentioned effective fuel consumption rate as an indicator of economy rather than using fuel consumption alone.

[0086] J fuel =w2·f(T e ,n e ) (7)

[0087] Where w2 is the fuel economy weight, f(T e ,n e ) is the effective fuel consumption function (g / kW·h) with respect to engine torque and engine speed, which is fitted by the above formula (2). This function is to reduce fuel consumption by keeping the engine operating point close to the optimal operating point, while ensuring that the vehicle has good power performance.

[0088] In terms of stability, in order to limit the fluctuations in vehicle acceleration caused by frequent changes in torque, a stability index is added to constrain the frequent changes in the power system.

[0089] J change =w3·ΔT e (8)

[0090] Where w3 is the torque change weight, ΔT e is the torque change (N·m).

[0091] Due to computational limitations, the predicted slope information of 150 meters ahead provided by the digital map is selected as the prediction field of NMPC. In order to adapt to the changing slope in a larger range, the CNN-LSTM network proposed in this paper sets the fuel economy weight w2 as a dynamic weight, which is updated based on the long-scale slope information.

[0092] w2=f(s long ) (9)

[0093] For this optimal control problem, there is a constraint: torque variation range constraint,

[0094] T emin ≤T e ≤T emax (n e ) (10)

[0095] Where T emin is a preset constant, T emax According to the engine characteristics, it is a function related to the speed. This constraint provides boundary conditions for the optimal control problem in this paper, ensuring that the algorithm finds the optimal solution in the feasible region.

[0096] Step 4: Construct a neural network that captures the mapping relationship between the slope sequence and the NMPC fuel consumption weight:

[0097] The NMPC in step 3 is limited in its prediction ability, and its prediction range is only 150m. This limitation causes the NMPC algorithm to essentially only obtain local optimal solutions. Applying these local optimal solutions through rolling update iterations does not ensure convergence to the global optimum. Therefore, the present invention considers using a neural network to capture longer-scale slope information to make the result closer to the global optimum.

[0098] like Figure 4 The neural network architecture includes an input layer for receiving future slope sequences, two CNN layers for feature extraction, an LSTM layer for capturing time dependencies, and a fully connected layer for classifying output controller parameters. It receives long-scale slope information within 600m ahead and outputs the optimal fuel economy weight w2 under the slope;

[0099] Since the network is used to process information about long-scale slopes hidden in the future 600m (sampled every 10m), the input layer size is set to 61. The CNN layer (both CNN layers are one-dimensional, denoted as Conv1D) is responsible for extracting local and global features from the slope sequence. The first layer uses a convolution kernel of size 1 to capture the initial slope changes and maintains the sequence length by padding; subsequent layers use larger convolution kernels to identify broader trends in the data. Specifically, the second CNN layer doubles the number of output channels to 64, with a convolution kernel size of 3 and padding of 1. In addition, each CNN layer is followed by batch normalization to stabilize the learning process, and a ReLU activation function is applied to introduce nonlinearity. After the CNN layer processes the slope data, the output is passed to an LSTM layer. This layer is critical for capturing the temporal dynamics in the slope sequence and helps accurately predict the fuel consumption weight. This LSTM component contains 32 hidden units and runs through the features output by the CNN layer to capture short-term and long-term dependencies in the slope sequence. The LSTM layer is set with a dropout rate of 0.3 to reduce overfitting and ensure that the model has good generalization ability on unseen data. The model is composed of a fully connected network (FC) at the end, which is used to compress the high-dimensional features extracted by the LSTM layer into a more manageable form for classification, that is, the selection of the optimal fuel economy weight. The fully connected layer outputs the probabilities corresponding to the preset 10 different fuel economy weights through the softmax activation function of the output layer, thereby realizing the accurate selection of fuel economy weights based on the slope sequence data according to the probability.

[0100] Step 5: Generate feature data for neural network training, namely slope sequence:

[0101] Construct a slope sequence data set containing multiple highway sections with a length of 600m. Considering the relative flatness of highways, the slope generation problem is simplified to a combination of straight line segments. Since the focus of the present invention is on predictive energy-saving strategies during uphill driving, all slopes are designed to be upward. To ensure a wide coverage of the generated data, the slope angle and horizontal distance are set as variables. Through the combination of these variables, a slope sequence composed of sufficiently diverse straight line segments can be generated;

[0102] In this embodiment, the first section is a 150m flat section with a slope of zero to allow the truck to enter a stable initial state. The second section starts after 150m and the slope rises at a certain angle, and its height is calculated by the following formula:

[0103] h(x)=x·tanθ (11)

[0104] Where h(x) represents the slope height (m), θ is the slope angle (rad), and its maximum value is limited to 0.04 rad to meet typical highway design standards. The variable x represents the horizontal distance (m) from the starting point of the slope (150m), ranging from 0 to 450m.

[0105] The third section is a flat section, starting from the end point of the grade and extending to a total length of 600 meters, maintaining the maximum height reached at the end of the grade.

[0106] The slope angle and horizontal distance are taken as variables, and the values ​​are discretized and combined, so that a set of slope sequences covering a sufficiently wide range of uphill scenarios can be generated.

[0107] Step 6: Establish an automated simulation scheduling system:

[0108] The automated simulation scheduling system consists of three parts: execution agent, analysis agent and simulation engine. Figure 5 As shown. Among them, the execution agent is responsible for generating the corresponding parameter scanning direction execution command according to the user instructions and the feedback of the analysis agent to guide the simulation engine (that is, input the next parameter w2 to be simulated into the simulation engine to obtain the multimodal results under this parameter). The analysis agent analyzes the multimodal simulation results and generates formatted feedback to help adjust the action of the execution agent in the next round. The simulation engine plays a bridging role between the two. It converts the execution command into the underlying executable command of the simulation software, the specific parameter scanning simulation process, and sends the simulation results to the analysis agent. The user only needs to provide the slope sequence (the long-scale slope sequence data set in step 5) and the parameter scanning range (that is, the artificial preset range of w2) to the execution agent to start the simulation process. During the simulation process, the execution agent determines the direction of the parameter scanning based on the feedback text of the analysis agent and the prior knowledge in the prompt word.

[0109] Specific:

[0110] In the execution agent, the function call method plays a key role in realizing simulation automation. Specifically, the semantic information of the parameter sweep simulation process is converted into a standardized function, in which the input and output of the simulation and its detailed Schema description are clearly defined and stored in the function library. Then, the large language model GLM-4.0 is deployed to identify the most appropriate simulation function by comparing user instructions and Schema descriptions. Here, the function of parameter sweep simulation is identified from the function library, and GLM-4.0 generates an execution command and passes it to the rule-based simulation engine. The following shows the system prompt words of the execution agent, defining the functions and roles:

[0111] System prompt words:

[0112] You are the executive agent, responsible for managing and executing simulation tasks. Your main responsibility is to process user instructions and analyze the agent's feedback, and convert it into executable commands for the simulation engine.

[0113] You will receive the data required for the simulation from the user, including the path to the ramp sequence file and the parameter sweep range. Your task is to perform the simulation and adjust subsequent operations based on feedback from previous results.

[0114] Since it takes too long to traverse the entire parameter scanning space, this paper adopts a prompt word-based method to incorporate the expert controller tuning experience into the large language model. By leveraging the expert knowledge in these prompt words, the LLM can prioritize the exploration of specific parameter areas that are more likely to produce the best results, thereby reducing the comprehensive search of the parameter space. The following is the design of the prompt word:

[0115] Threshold detection hints: These hints enable the model to detect critical threshold points and automatically stop further parameter exploration. For example, when the fuel consumption weight reaches a certain threshold, the engine may oscillate, causing the vehicle state to be unstable. By promptly identifying the oscillation area, the agent skips the parameter range that no longer has optimization potential and concentrates computing resources on more valuable areas.

[0116] Examples of threshold detection prompt words are as follows:

[0117] If the engine speed appears to oscillate, further parameter sweeps are stopped.

[0118] If the vehicle speed stability deteriorates rapidly, stop increasing the weight parameter.

[0119] If the vehicle speed drops to 0, stop increasing the fuel economy weighting.

[0120] Jump parameter adjustment hints: When simulation results show that the difference between adjacent parameters is very small, these hints can instruct the model to skip unnecessary parameter sweeps and directly adjust to a larger parameter step size. This jump adjustment strategy speeds up the optimization process and avoids redundant calculations.

[0121] Examples of jump parameter adjustment prompts are as follows:

[0122] If the adjacent weight w 2,1 , w 2,2 , and w 2,3 The difference between them is small, so the parameter scanning step size is increased, such as the next step weight is set to w 2,5 .

[0123] If no significant improvement is observed in the first few steps, skip the parameter w 2,i .

[0124] In the execution agent, GPT4-o with excellent multimodal capabilities is used. Its main responsibility is to analyze the results generated by each simulation run, generate text feedback, and send it to the execution agent. The overall prompt words, including standardized requirements and expert guidance suggestions for control result analysis, are as follows:

[0125] System prompt words:

[0126] You are an expert in vehicle control, focusing on the analysis and optimization of simulation results. You will receive data generated by the controller parameter sweep simulation, including the slope sequence ID, current fuel consumption weight, engine torque and speed, vehicle speed, and acceleration. You need to conduct a comprehensive analysis of the data and its corresponding curve graph.

[0127] Standardization requirement prompt words:

[0128] You need to analyze each simulation result one by one to determine whether the key performance indicators of the system meet the expected values. Please format the analysis report as follows:

[0129] Simulation ID (insert simulation ID in the format: [slope sequence ID - current fuel economy weight]).

[0130] Fuel consumption analysis (a brief description of the fuel consumption, focusing on the values ​​at the end of the simulation).

[0131] Engine oscillation analysis (indicates whether torque and speed oscillate).

[0132] Speed ​​and acceleration analysis (describe the smoothness of the speed and acceleration curves, noting any sharp changes).

[0133] Overall recommendations (parameter sweep suggestions based on analysis results).

[0134] Expert guidance and suggestions:

[0135] Pay special attention to the fuel consumption values ​​at the end of the simulation and report any unusual spikes.

[0136] Monitor the engine torque and speed during simulation for oscillations.

[0137] Carefully evaluate the smoothness of the vehicle's speed and acceleration curves, checking for sharp rises or falls.

[0138] If anomalies are found, provide detailed explanations and suggest how to adjust simulation parameters.

[0139] Finally, in the design of the simulation engine, first of all, deep integration with the simulation platform (such as Simulink) is carried out through the API (such as MATLAB API). This integration process includes configuring the API interface so that it can communicate seamlessly with the simulation platform and ensure that the instructions generated by the execution agent can be correctly transmitted to the target simulation platform. The instructions generated by the execution agent are then converted into a command format executable by the simulation platform and transmitted through the API interface. In addition, it is necessary to implement the recording and forwarding of results to ensure that the simulation platform can output the execution status or results to the analysis agent after executing the instructions.

[0140] Step 7: Generate labels for neural network training, i.e., the optimal fuel economy weight w2 corresponding to each slope sequence:

[0141] The label is determined by running the simulation engine simulation to find the optimal fuel economy weight w2 for each slope sequence. The automated simulation scheduling system in step 6 will simulate according to the parameter scan range input by the user. The discrete scan range of the fuel economy weight is pre-set based on experience to reduce the computational burden.

[0142] w 2,i ∈{w 2,1 ,w 2,2 ,w 2,3 ,…,w 2,10}(11)

[0143] After completing the parameter scan (11) for each slope, the performance is evaluated by indicators such as fuel consumption and vehicle stability, as shown below:

[0144] E total =αE fuel +αE stability (12)

[0145] Among them, α and β are the weights of fuel consumption and vehicle stability index respectively. fuel and E stability To ensure that the value of each indicator is between 0 and 1, a normalization method is used.

[0146] The fuel consumption index is measured in terms of consumption per 100 kilometers, recorded as

[0147]

[0148] Where T is the total duration of a single simulation (s), is the fuel consumption rate per unit time (kg / s), ρ fuel is the fuel density (kg / L), and s is the total driving distance in the simulation (km).

[0149] The vehicle stability index is expressed as follows:

[0150]

[0151] Where v is the real-time vehicle speed (m / s) during the simulation.

[0152] Step 8: Training of Neural Network:

[0153] The neural network is trained and tested using the training and test data sets consisting of the labels obtained in step 7 and the long-scale slope sequences obtained in step 5. During the training phase, the network uses the cross entropy loss function, which is a common and standard choice in multi-classification tasks to measure the difference between the predicted category and the actual label. The Adam optimizer is used to solve the training process because of its high efficiency in handling sparse gradients and adaptive learning rates.

[0154] In this embodiment, the batch size of the training scheme is set to 128, with a total of 500 epochs, which provides sufficient time for the network to capture the complex dependency between the slope sequence data and the optimal fuel economy weight.

[0155] The above is only a preferred embodiment of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A heavy truck slope adaptive predictive energy-saving control method with LLM-assisted data collection, characterized in that: The following steps are involved: Step 1, establish a longitudinal dynamic model of a heavy truck on a slope road; Step 2, establish the engine fuel consumption characteristic model and obtain the brake specific fuel consumption BSFC; Step 3, establish a NMPC controller for the heavy truck power system to accurately control the engine torque and feed it back to the control-oriented model composed of the longitudinal dynamics model in step 1 and the engine fuel consumption characteristic model in step 2, and construct a multi-objective cost function to take into account speed tracking and fuel consumption. The multi-objective cost function includes a speed traceability weight w1, a fuel economy weight w2, and a torque change weight w3; Step 4, constructing a neural network that captures the mapping relationship between the slope sequence and the fuel economy weight w2 in step 2, and the neural network outputs the optimal fuel economy weight w2 to the NMPC controller in step 3; Step 5, constructing a feature data set for neural network training, namely, a long-scale slope sequence; Step 6, establish an automated simulation scheduling system consisting of an execution agent, an analysis agent, and a simulation engine. The execution agent generates an execution command in the w2 scanning direction according to the long-scale slope sequence of step 5 and the discrete scanning range of the fuel economy weight w2 set manually. The simulation engine converts the execution command into the underlying executable command of the simulation software and the specific w2 scanning simulation process, forms a multimodal simulation result and sends it to the analysis agent. The analysis agent analyzes the multimodal simulation result and generates formatted feedback to help adjust the action of the execution agent in the next round. Step 7, the automated simulation scheduling system generates a label, where the label is the optimal fuel economy weight w2 corresponding to each long-scale slope sequence; Step 8, training and testing the neural network using a training data set and a test data set consisting of the labels obtained in step 7 and the long-scale slope sequence obtained in step 5.

2. The heavy truck slope adaptive predictive energy-saving control method according to claim 1, characterized in that: The longitudinal dynamic model in step 1 is: Where m is the vehicle mass, θ is the road slope, ρ is the air density, C d is the air resistance coefficient, A a is the frontal area of ​​the vehicle, v is the vehicle speed, f is the rolling resistance coefficient, F drive is the driving force that drives the vehicle to move, F aero is the air resistance, F roll is the rolling resistance, F slope is the slope resistance.

3. The heavy truck slope adaptive predictive energy-saving control method according to claim 1, characterized in that: In step 2, the engine fuel consumption characteristic model is: Among them, α is the fuel consumption characteristic fitting coefficient, T e is the engine torque, n e is the engine speed.

4. The heavy truck slope adaptive predictive energy-saving control method according to claim 1, characterized in that: In step 3, the process of establishing the NMPC controller of the heavy truck power system is as follows: in, is the time constant of the torque inertia link, The control variable of the NMPC optimal control problem is the target engine torque; The state variables of the optimal control problem are vehicle speed v and travel distance s d : Where i0 is the engine differential ratio, R w is the wheel radius of a heavy truck.

5. The heavy truck slope adaptive predictive energy-saving control method according to claim 1, characterized in that: In step 3, the multi-objective cost function is: J=J v +J fuel +J change (5) J v Tracking metrics for speed: J v =w1·(v ref -v) 2 (6) w1 is the speed traceability weight, v ref is the vehicle cruising reference speed, v is the vehicle speed; J fuel Fuel economy index: J fuel =w2·f(T e ,n e ) (7) w2 is the fuel economy weight; J change Is the stability indicator: J change =w3·ΔT e (8) Where w3 is the torque change weight, ΔT e is the torque change.

6. The heavy truck slope adaptive predictive energy-saving control method according to claim 1, characterized in that: In step 3, the engine torque variation range is constrained as follows: T emin ≤T e ≤T emax (n e ) (10) Where T emin is a preset constant, T emax Calculated based on engine characteristics, it is related to engine speed n e Related functions.

7. The heavy truck slope adaptive predictive energy-saving control method according to claim 1, characterized in that: The neural network in step 4 is a CNN-LSTM network.

8. The heavy truck slope adaptive predictive energy-saving control method according to claim 7, characterized in that: The architecture of the neural network of step 4 includes an input layer for receiving the future slope sequence, two CNN layers for feature extraction, an LSTM layer for capturing time dependencies, and a fully connected layer for classifying the output controller parameter w2.

9. The heavy truck slope adaptive predictive energy-saving control method according to claim 1, characterized in that: In step 7, the automated simulation scheduling system performs simulation according to the discrete scanning range of the fuel economy w2 input by the user, and the discrete scanning range of the fuel economy weight w2 is preset manually based on experience; In 2,i ∈{in 2,1 ,In 2,2 ,In 2,3 ,…,In 2,10 } (11) After completing the parameter scan (11) for each slope, the performance is evaluated by indicators such as fuel consumption and vehicle stability, as shown below: E total =αE fuel +βE stability (12) Among them, α and β are the weights of fuel consumption and vehicle stability index respectively. fuel and E stability The magnitudes of the indicators are different. To ensure that the value of each indicator is between 0 and 1, the normalization method is used. The fuel consumption index is measured in terms of consumption per 100 kilometers, recorded as Where T is the total duration of a single simulation, is the fuel consumption rate per unit time, ρ fuel is the fuel density, s is the total driving distance in the simulation; The vehicle stability index is expressed as follows: Where v is the real-time vehicle speed during the simulation.

10. The heavy truck slope adaptive predictive energy-saving control method according to claim 1, characterized in that: In step 8, the neural network is trained using a cross entropy loss function.

Images