Energy consumption management method and device for controller, equipment and computer program product

Abstract

The invention relates to the technical field of intelligent control, and discloses an energy consumption management method and device for an electric tricycle controller. The method comprises the steps of generating driving preference data based on historical driving behavior data so as to obtain a power control parameter initial value and an energy recovery control parameter initial value; optimizing and adjusting motor power output according to the current running state data and the real-time environment data of the vehicle; inputting the real-time energy consumption characteristic data into the energy consumption optimization model to obtain a prediction result; and adjusting the power control parameter and the energy recovery control parameter in real time according to the prediction result. Compared with an energy consumption management method depending on fixed parameters in the prior art, optimization and balance of energy efficiency are difficult to realize when a dynamically changing driving environment is processed, especially under a complex driving condition. According to the method, the driving behavior analysis and the deep learning model are combined, the energy consumption of the electric tricycle is predicted and adjusted in real time, and the cruising ability and the energy efficiency management level of the electric tricycle are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to an energy consumption management method and system for an electric tricycle controller. Background Art

[0002] As a means of transportation with batteries as power source and controllers as core control devices, electric tricycles have been widely used in urban short-distance transportation and logistics distribution in recent years. The controller of electric tricycles plays a vital role in the start, operation, advance and retreat, speed adjustment, braking and energy recovery of the whole vehicle. Traditional electric tricycle controllers mainly rely on preset control parameters to manage energy consumption, and control motor output and battery use in a static way. In the prior art, the controllers of electric tricycles have the following problems in energy consumption management: First, the controller is usually unable to dynamically adjust the control parameters according to real-time driving conditions and user driving habits, which leads to poor adaptability of energy consumption management and difficulty in achieving accurate energy consumption optimization. Secondly, the existing energy consumption management methods lack in-depth analysis of user driving behavior, and cannot effectively combine user driving preferences to optimize energy recovery and power output; in order to improve the efficiency of energy consumption management, there is an urgent need for a dynamic adjustment method that can combine historical driving data, real-time environmental data and user driving behavior, and accurately manage the energy consumption of electric tricycles through intelligent control algorithms. The new energy consumption management method should be able to respond to changes in driving conditions in real time and provide personalized control strategies to adapt to different driving scenarios and user needs, thereby improving the overall performance and user experience of the electric tricycle. Summary of the invention

[0003] The main purpose of the present invention is to provide an energy consumption management method, device, equipment and computer program product for a controller, aiming to solve the technical problem that the energy consumption management method in the prior art relies on fixed parameters and static control, and is difficult to optimize and balance energy efficiency when dealing with dynamically changing driving environments, especially under complex driving conditions.

[0004] To achieve the above object, the present invention adopts the following technical solution: The present invention provides an energy consumption management method and system method for a controller,

[0005] Preferably, the method comprises:

[0006] Step S10: generating driving preference data based on historical driving behavior data, and generating initial values ​​of power control parameters and energy recovery control parameters according to the driving preference data;

[0007] Step S20: Optimizing and adjusting the motor power output according to the vehicle's current operating state data and real-time environmental data, and recording real-time energy consumption data;

[0008] Step S30: performing data preprocessing on the real-time energy consumption data to obtain real-time energy consumption characteristic data, and inputting the real-time energy consumption characteristic data into an energy consumption optimization model to obtain a prediction result;

[0009] Step S40: According to the prediction results, the power control parameters and the energy recovery control parameters are adjusted in real time. The adjusted power control parameters act on the motor power output control in real time to reduce the frequency of instantaneous high energy consumption. The adjusted energy recovery control parameters act on the braking energy recovery control in real time to improve the energy recovery efficiency when the vehicle decelerates.

[0010] Preferably, before the step of inputting the real-time energy consumption characteristic data into the energy consumption optimization model to obtain the prediction result, the method further includes:

[0011] Build a training dataset based on historical driving behavior data and energy consumption data;

[0012] Selecting and initializing an energy consumption optimization model according to the training data set;

[0013] The Adam optimization algorithm is used to optimize the weight parameters in the energy consumption optimization model, the back propagation algorithm is used to calculate the loss function value, and the weight parameters are updated according to the gradient of the loss function, where the loss function includes: where y i is the actual energy consumption value, To predict the energy consumption value, N is the number of samples and i represents the i-th sample.

[0014] Preferably, after the step of completing the energy consumption optimization model training, the following step is also included:

[0015] The prediction accuracy of the model is evaluated through the early stopping method, and the trained energy consumption optimization model is deployed to the vehicle system.

[0016] Preferably, the step of optimizing and adjusting the motor power output according to the current running state data of the vehicle and the real-time environmental data includes:

[0017] Perform real-time analysis based on the vehicle's current speed, acceleration and load data to dynamically adjust the motor's output power;

[0018] Comprehensively analyze the vehicle's current operating status data and real-time environmental data to generate a real-time power output command signal.

[0019] Preferably, the step of performing data preprocessing on the real-time energy consumption data to obtain real-time energy consumption characteristic data includes:

[0020] Clean and denoise the collected real-time energy consumption data;

[0021] The processed data is subjected to feature extraction to obtain a feature data set, and the feature data set is normalized to obtain real-time energy consumption feature data.

[0022] Preferably, the real-time environmental data includes air temperature, humidity, wind speed and road wetness.

[0023] Preferably, the generated power control parameters and energy recovery control parameters can be further adjusted according to the driving mode manually selected by the user, the steps comprising:

[0024] receiving a driving mode signal manually selected by a user, wherein the driving mode includes an energy-saving mode, a standard mode, and a sports mode;

[0025] In energy-saving mode, the maximum motor power output in the power control parameters is automatically reduced, and the recovery intensity in the energy recovery control parameters is increased;

[0026] In standard mode, the default values ​​of power control parameters and energy recovery control parameters are maintained;

[0027] In sport mode, the acceleration response and maximum motor power output in the power control parameters are improved, while the recovery intensity in the energy recovery control parameters is weakened.

[0028] Preferably, the energy consumption management device for the electric tricycle controller comprises:

[0029] a control parameter generation module, configured to generate driving preference data based on historical driving behavior data, and to generate power control parameters and energy recovery control parameters according to the driving preference data;

[0030] The power optimization and adjustment module is used to optimize and adjust the motor power output according to the vehicle's current operating status data and real-time environmental data, and record real-time energy consumption data;

[0031] A model prediction module, used for performing data preprocessing on the real-time energy consumption data to obtain real-time energy consumption characteristic data, and inputting the real-time energy consumption characteristic data into an energy consumption optimization model to obtain a prediction result;

[0032] A control parameter adjustment module is used to adjust the power control parameter and the energy recovery control parameter in real time according to the prediction result.

[0033] The beneficial effect of the present invention is that compared with the energy consumption management method in the prior art that relies on fixed parameters and static control, it is difficult to achieve energy efficiency optimization and balance when dealing with dynamically changing driving environments, especially under complex driving conditions. The present invention combines driving behavior analysis with a deep learning model to predict and adjust the energy consumption of the vehicle in real time, thereby being able to efficiently and accurately manage the energy consumption of the electric tricycle, avoiding the limitations of the prior art due to poor adaptability and excessive energy consumption, and significantly improving the endurance and energy efficiency management level of the electric tricycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 FIG. 4 is a flow chart of a first embodiment of an energy consumption management method for a controller according to the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Embodiment 1: Figure 1 FIG. 1 is a flow chart of a first embodiment of an energy consumption management method for a controller according to the present invention.

[0038] Step S10: generating driving preference data based on historical driving behavior data, and generating power control parameters and energy recovery control parameters according to the driving preference data;

[0039] It should be noted that the driving preference data in the present invention not only includes the driver's speed preference, acceleration and deceleration habits, etc., but also covers the driver's driving behavior patterns in different environments. These data are obtained by analyzing and processing historical driving data, and can more comprehensively reflect the driver's personalized needs.

[0040] It can be understood that in the process of generating power control parameters and energy recovery control parameters, it not only relies on static historical data, but also combines real-time environmental data, such as current road conditions, weather conditions, etc., to ensure that the generated parameters can better adapt to the current driving environment.

[0041] It should be understood that the dynamic adjustment of the power control parameters and energy recovery control parameters is achieved through an intelligent algorithm. The algorithm can automatically optimize the parameter settings according to the driving preference data and real-time environmental data to achieve the best energy efficiency in different driving scenarios. This dynamic adjustment process not only improves the energy utilization rate, but also effectively extends the cruising range of the electric tricycle. The specific implementation of the intelligent algorithm can be based on the combination of reinforcement learning and neural networks, in which the environmental state and action reward mechanism are used for continuous optimization. Electric tricycles often need to operate in complex urban traffic environments. Faced with frequent starts and stops, sudden acceleration, and driving challenges under various road conditions, intelligent algorithms must be able to respond to these changes in real time. Specifically, the reinforcement learning algorithm makes decisions by observing the real-time state of the electric tricycle (such as speed, acceleration, road slope) and driving behavior (such as braking, acceleration), and adjusts the power output and energy recovery strategies. The energy efficiency performance after each adjustment is fed back to the algorithm as a reward for optimizing future decisions. The optimization formula of the intelligent algorithm of this scheme can be expressed as:

[0042] Q(s,a)=Q(s,a)+α[r+γmaxQ(s′,a′)-Q(s,a)]

[0043] Among them, Q(s, a) represents the utility value obtained by performing action a in state s; α is the learning rate, which determines the degree of influence of new information on old information; r is the immediate reward obtained by performing action a in state s (such as improved energy efficiency); γ is the discount factor, which indicates the degree of influence of future rewards on current decisions;

[0044] For example, suppose an electric tricycle is driving on a city road and encounters a road section with frequent traffic lights. The intelligent algorithm determines the current state s (for example: low speed, about to brake) based on real-time environmental data and driving behavior data. At this time, action a may be to reduce power output and enhance energy recovery. Through algorithm calculation and update, the Q value will be adjusted according to the performance of the vehicle in different states (such as the actual energy recovery effect). For example, when the vehicle detects that it is about to stop and slows down at a red light, the algorithm will choose a low power output, high energy recovery strategy to maximize the amount of energy recovery. Through repeated training, the algorithm gradually learns how to adjust the power and recovery parameters under different road conditions to achieve optimal energy efficiency.

[0045] Step S20: Optimizing and adjusting the motor power output according to the vehicle's current operating state data and real-time environmental data, and recording real-time energy consumption data;

[0046] It should be noted that this step not only takes into account the vehicle operating status data such as the current speed, acceleration and load of the electric tricycle, but also fully combines the real-time external environmental data, such as road slope, traffic conditions, weather conditions, etc., to dynamically adjust the power output of the motor to ensure that the vehicle can achieve optimal energy efficiency under various driving conditions.

[0047] It is understandable that the acquisition of real-time environmental data can be achieved through a variety of sensors and external data sources, such as GPS data, road sensors, meteorological data, etc. After pre-processing, these data are combined with vehicle operating status data and input into the control algorithm for real-time optimization of motor power output.

[0048] It should be understood that in this step, the recorded real-time energy consumption data will be used for subsequent energy consumption analysis and optimization model training. These data can not only reflect the current energy consumption situation, but also serve as a basis for future decision-making. By continuously accumulating and analyzing these data, the system can gradually improve the accuracy and adaptability of energy consumption management and further optimize the overall performance and endurance of the electric tricycle.

[0049] Step S30: performing data preprocessing on the real-time energy consumption data to obtain real-time energy consumption characteristic data, and inputting the real-time energy consumption characteristic data into an energy consumption optimization model to obtain a prediction result;

[0050] It should be noted that real-time energy consumption data refers to the raw data related to energy consumption collected by on-board sensors and monitoring systems during the operation of electric tricycles. The data includes but is not limited to the vehicle's instantaneous current, voltage, power consumption, energy recovery, driving speed, acceleration, etc. Real-time energy consumption data can reflect the current energy consumption status and operating conditions of the vehicle, and is an important basis for subsequent analysis and optimization.

[0051] It should be noted that in traditional energy consumption management methods, energy consumption optimization models are mostly predicted based on rules or simple linear regression. These methods usually only consider basic parameters such as the current speed and acceleration of the vehicle, and fail to comprehensively consider complex driving behaviors, real-time environmental data (such as temperature, slope, road conditions, etc.) and nonlinear changes in energy consumption characteristic data, resulting in low prediction accuracy and inability to make real-time dynamic adjustments.

[0052] It should be understood that the specific formula of the energy consumption optimization model in the present invention can be:

[0053] E mod (t)=β1V(t)+β2A(t)+β3I(t)+β4U(t)+β5θ(t)+β6T(t)+β7R(t)+∈

[0054] Among them, E mod(t) is the predicted energy consumption at time t, V(t) is the speed at time t, A(t) is the acceleration at time t, I(t) is the current at time t, U(t) is the voltage at time t, θ(t) is the road slope, T(t) is the ambient temperature, R(t) is the energy recovery amount, β1 to β7 are all regression coefficients, and ∈ is the error term; by introducing more variables and nonlinear features, the model can predict energy consumption more accurately, especially in complex driving environments.

[0055] It is understandable that real-time energy consumption characteristic data refers to the representative and recognizable key features extracted from the above real-time energy consumption data. These characteristic data include processed power consumption rate, energy recovery efficiency, energy consumption change rate per unit time, etc. After preprocessing and optimization, these characteristic data can effectively summarize the energy consumption performance of the vehicle under specific conditions and provide reliable data support for subsequent energy consumption optimization.

[0056] It should be understood that the step of performing data preprocessing on the real-time energy consumption data to obtain the real-time energy consumption characteristic data includes the following specific steps:

[0057] Clean and denoise the collected real-time energy consumption data: First, clean the raw data to remove noise and outliers to ensure the accuracy and consistency of the data. Noise may come from sensor errors or interference from the external environment, so it is necessary to use filtering algorithms or other denoising techniques to process it. Feature extraction of the processed data: Extract the most representative energy consumption features from the cleaned data. These features may include the rate of change of instantaneous power, average energy recovery, energy consumption increment per unit time, etc. This step usually uses feature engineering techniques to identify and extract the most critical indicators for energy consumption optimization. Normalize the feature data set: In order to make the feature data suitable for the optimization model, it is usually necessary to normalize it. By scaling the feature values ​​to a uniform range (such as 0 to 1), the impact of dimensional differences between different features can be eliminated, thereby improving the training effect and prediction accuracy of the model.

[0058] Step S40: adjusting the power control parameter and the energy recovery control parameter in real time according to the prediction result;

[0059] It should be noted that the core of this step is to use the prediction results generated by the energy consumption optimization model to dynamically adjust the power output and energy recovery strategy of the electric tricycle. This adjustment process is carried out in real time and aims to optimize the energy efficiency performance of the electric tricycle according to different driving conditions and environmental changes. Specifically, the adjustment of power control parameters may include the adjustment of motor output power, while the adjustment of energy recovery control parameters involves the energy recovery strength and strategy during braking.

[0060] Embodiment 2:

[0061] Based on the first embodiment, in this embodiment, before the step of inputting the real-time energy consumption characteristic data into the energy consumption optimization model to obtain the prediction result, the following steps are further included:

[0062] Build a training dataset based on historical driving data and energy consumption data;

[0063] Selecting and initializing a deep learning model based on the training data set;

[0064] The Adam optimization algorithm is used to optimize the model parameters, the back propagation algorithm is used to calculate the loss function value, and the model parameters are updated according to the gradient of the loss function, where the loss function includes: where y i is the actual energy consumption value, is the predicted energy consumption value, and N is the number of samples;

[0065] It should be noted that when training deep learning models, choosing appropriate historical driving data and energy consumption data is crucial to the accuracy of the model. In order to ensure that the model can work effectively under a variety of driving conditions, the training data set should include energy consumption performance under different weather, road conditions, and traffic conditions. These data include not only the energy consumption of the vehicle during stable driving, but also the energy consumption changes in complex scenarios such as frequent acceleration, deceleration, and uphill and downhill driving. By training on diverse data, the model can generalize better and provide accurate energy consumption predictions.

[0066] It is understandable that in the current application scenario, the creativity of this method is reflected in the following aspects: First, the use of deep learning models for energy consumption prediction can capture complex nonlinear relationships, while traditional statistical methods have difficulty processing such complex data patterns. Secondly, the Adam optimization algorithm is particularly suitable for real-time application scenarios due to its fast convergence and insensitivity to hyperparameters, allowing the model to quickly adapt to the changing driving environment. Finally, the model parameters are optimized through the back-propagation algorithm so that the minimization process of the loss function can accurately reflect the error of the energy consumption prediction, which can significantly improve the accuracy of energy efficiency management in practical applications and effectively extend the range of electric tricycles.

[0067] It should be understood that the deep learning model is initially trained using offline data, but its design can adapt to the needs of online learning, that is, as the vehicle continues to operate, new energy consumption data can be continuously acquired and updated in real time. This dynamic model update method ensures the long-term effectiveness and adaptability of the model, especially when the vehicle's operating conditions and driving behavior change over time, it can continue to provide high-precision energy consumption predictions. In addition, the modular design of the system enables it to be flexibly applied to different types of electric tricycles, with wide adaptability and scalability.

[0068] The rest is the same as that of the first embodiment.

[0069] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An energy consumption management method for an electric tricycle controller, characterized in that: The method comprises: Step S10: generating driving preference data based on historical driving behavior data, and generating initial values ​​of power control parameters and energy recovery control parameters according to the driving preference data; Step S20: Optimizing and adjusting the motor power output according to the vehicle's current operating state data and real-time environmental data, and recording real-time energy consumption data; Step S30: performing data preprocessing on the real-time energy consumption data to obtain real-time energy consumption characteristic data, and inputting the real-time energy consumption characteristic data into an energy consumption optimization model to obtain a prediction result; Step S40: According to the prediction results, the power control parameters and the energy recovery control parameters are adjusted in real time. The adjusted power control parameters act on the motor power output control in real time to reduce the frequency of instantaneous high energy consumption. The adjusted energy recovery control parameters act on the braking energy recovery control in real time to improve the energy recovery efficiency when the vehicle decelerates.

2. The energy consumption management method for an electric tricycle controller according to claim 1, characterized in that: Before the step of inputting the real-time energy consumption characteristic data into the energy consumption optimization model to obtain the prediction result, the method further includes: Build a training dataset based on historical driving behavior data and energy consumption data; Selecting and initializing an energy consumption optimization model according to the training data set; The Adam optimization algorithm is used to optimize the weight parameters in the energy consumption optimization model, the back propagation algorithm is used to calculate the loss function value, and the weight parameters are updated according to the gradient of the loss function, where the loss function includes: where y i is the actual energy consumption value, To predict the energy consumption value, N is the number of samples and i represents the i-th sample.

3. The energy consumption management method for an electric tricycle controller as claimed in claim 2, characterized in that: After the energy consumption optimization model training step is completed, it also includes The prediction accuracy of the model is evaluated through the early stopping method, and the trained energy consumption optimization model is deployed to the vehicle system.

4. The energy consumption management method for an electric tricycle controller according to claim 1, characterized in that: The step of optimizing and adjusting the motor power output according to the vehicle's current operating state data and real-time environmental data includes: Perform real-time analysis based on the vehicle's current speed, acceleration and load data to dynamically adjust the motor's output power; Comprehensively analyze the vehicle's current operating status data and real-time environmental data to generate a real-time power output command signal.

5. The energy consumption management method for an electric tricycle controller according to claim 1, characterized in that: The step of preprocessing the real-time energy consumption data to obtain the real-time energy consumption characteristic data includes: Clean and denoise the collected real-time energy consumption data; The processed data is subjected to feature extraction to obtain a feature data set, and the feature data set is normalized to obtain real-time energy consumption feature data.

6. The energy consumption management method for an electric tricycle controller according to claim 1, characterized in that: The real-time environmental data includes air temperature, humidity, wind speed and road wetness.

7. The energy consumption management method for an electric tricycle controller as claimed in claim 6, characterized in that: The generated power control parameters and energy recovery control parameters can be further adjusted according to the driving mode manually selected by the user, the steps comprising: receiving a driving mode signal manually selected by a user, wherein the driving mode includes an energy-saving mode, a standard mode, and a sports mode; In energy-saving mode, the maximum motor power output in the power control parameters is automatically reduced, and the recovery intensity in the energy recovery control parameters is increased; In standard mode, the default values ​​of power control parameters and energy recovery control parameters are maintained; In sport mode, the acceleration response and maximum motor power output in the power control parameters are improved, while the recovery intensity in the energy recovery control parameters is weakened.

8. An energy consumption management device for an electric tricycle controller, characterized in that: The energy consumption management device for the electric tricycle controller comprises: a control parameter generation module, configured to generate driving preference data based on historical driving behavior data, and to generate power control parameters and energy recovery control parameters according to the driving preference data; The power optimization and adjustment module is used to optimize and adjust the motor power output according to the vehicle's current operating status data and real-time environmental data, and record real-time energy consumption data; A model prediction module, used for performing data preprocessing on the real-time energy consumption data to obtain real-time energy consumption characteristic data, and inputting the real-time energy consumption characteristic data into an energy consumption optimization model to obtain a prediction result; A control parameter adjustment module is used to adjust the power control parameter and the energy recovery control parameter in real time according to the prediction result.

9. An energy consumption management device for an electric tricycle controller, characterized in that: The energy consumption management device for the electric tricycle controller includes: A memory, a processor, and an energy consumption management program for an electric tricycle controller stored in the memory and executable on the processor, wherein the energy consumption management program for an electric tricycle controller, when executed by the processor, implements the energy consumption management method for an electric tricycle controller as described in any one of claims 1 to 7.

10. A computer program product, characterized in that The computer program product comprises an energy consumption management program for an electric tricycle controller, and when the energy consumption management program for an electric tricycle controller is executed by a processor, the energy consumption management method for an electric tricycle controller as claimed in any one of claims 1 to 7 is implemented.

Images