An interdisciplinary-based in-vehicle temperature environment control method

Abstract

The application discloses a kind of based on interdisciplinary in-car temperature environment control method, steps are S1: through sensor monitoring in-car and outside environmental parameters, output real-time environmental data;S2: processing and analyzing data, using machine learning and artificial intelligence algorithm to carry out pattern recognition and trend prediction;S3: through data analysis result and prediction model, formulate intelligent control strategy;S4: carry out integrated optimization and collaborative control, improve overall vehicle performance and energy efficiency;S5: provide user interface, and driver sets in-car temperature environment according to individual needs;S6: using machine learning algorithm continuous learning and optimization control strategy.The application realizes intelligent control, accurately adapts in-car and outside temperature environment change, and realizes close cooperation with other in-car components, optimizes overall vehicle performance and energy efficiency, improves the comfortable feeling of driver.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of in-vehicle temperature environment control, and particularly relates to an in-vehicle temperature environment control method based on interdisciplinary. BACKGROUND

[0002] In the prior art, the temperature environment in the vehicle can only provide fixed temperature control, and cannot be intelligently adjusted according to actual needs and external environmental conditions, resulting in differences in comfort under different driving conditions, which has the following disadvantages:

[0003] 1. High energy consumption: Low efficiency in balancing temperature regulation and energy consumption, especially under high or low temperature conditions, leading to reduced vehicle fuel efficiency and increased fuel costs.

[0004] 2. Slow response speed: The response speed is relatively slow, and it cannot quickly adapt to temperature changes, resulting in user comfort being affected in environments with large temperature changes.

[0005] 3. Disconnected from other controls: It is usually disconnected from other control components of the vehicle, resulting in limitations in optimizing overall vehicle performance and efficiency. SUMMARY

[0006] The present application aims to solve the above problems in the prior art by providing an in-vehicle temperature environment control method based on interdisciplinary.

[0007] An in-vehicle temperature environment control method based on interdisciplinary, comprising the following sub-steps:

[0008] S1: Monitor the environmental parameters inside and outside the vehicle through sensors and output real-time environmental data;

[0009] S2: Process and analyze data, use machine learning and artificial intelligence algorithms for pattern recognition and trend prediction;

[0010] S3: Develop intelligent control strategies based on data analysis results and prediction models;

[0011] S4: Perform integrated optimization and collaborative control to improve overall vehicle performance and energy efficiency;

[0012] S5: Provide a user interface for the driver to set the in-vehicle temperature environment according to personal needs;

[0013] S6: Use machine learning algorithms to continuously learn and optimize control strategies.

[0014] Further, an in-vehicle temperature environment control method based on interdisciplinary, the step S1 comprises the following sub-steps:

[0015] The temperature parameters inside and outside the vehicle are measured by the temperature sensor, and the real-time temperature difference inside and outside the vehicle is output.

[0016] The humidity parameter inside the vehicle is detected by the humidity sensor, and the real-time humidity inside the vehicle is output.

[0017] The external light intensity is sensed by the light sensor, and the real-time light intensity difference inside and outside the vehicle is output.

[0018] Further, a cross-disciplinary vehicle temperature environment control method, the step S2 further includes the following sub-steps:

[0019] S21: Calculate the comfort score or energy efficiency index of the vehicle temperature environment by linear regression, the formula is:

[0020] y = β0 + β1x1 + β2x2 + … β n x n + ε

[0021] Where y represents the target variable, which describes the comfort score or energy efficiency index, β0, β1... β n represent the regression coefficients of the linear regression model, x1, x2... x n represent feature variables, which describe the temperature, humidity, and light intensity inside and outside the vehicle, and ε represents the error term.

[0022] S22: Further process and analyze the data by decision tree and support vector machine to create a prediction model.

[0023] Further, a cross-disciplinary vehicle temperature environment control method, the step S3 includes the following sub-steps:

[0024] S31: Use the model's deep Q-learning algorithm to learn the vehicle temperature environment control strategy, and the update rule is:

[0025] Q(s, a)←Q(s, a) + α(r + γmax a Q(s′, a) - Q(s, a))

[0026] Where Q(s, a) represents the expected return value of action a in state s, r represents the immediate reward given by the environment, s' represents the next state after executing action a, α represents the learning rate, γ represents the discount factor, and max a Q(s′, a) represents the maximum value of the discount factor Q(s′, a).

[0027] S32: Adjust the vehicle temperature environment according to the prediction results, choose whether to adjust the vehicle temperature according to the driver's driving time, and choose the music type according to the driver's habits.

[0028] Further, a cross-disciplinary in-vehicle temperature environment control method, the step S4 comprises the following sub-steps:

[0029] S41: Cooperate with the in-vehicle entertainment system, and play music according to the mood and preference of the driver:

[0030] S42: Simultaneously optimize the comfort and energy efficiency of the in-vehicle temperature environment by using a multi-objective genetic algorithm;

[0031] S43: Define the fitness function of the MOGA as the weighted sum of the comfort index F1 and the energy efficiency index F2, and use non-dominated sorting and crowding degree to maintain the diversity of the population.

[0032] Further, a cross-disciplinary in-vehicle temperature environment control method, the step S5 comprises the following sub-steps:

[0033] S5: The driver selects the fatigue, excitement, and comfort options through the interface feedback, and the interface adjusts the in-vehicle temperature environment and music settings according to the selection feedback:

[0034] S52: According to the user interaction and feedback mechanism, use a natural language processing algorithm to analyze the feedback mood of the driver, and adjust the in-vehicle temperature environment and music settings according to the mood;

[0035] S53: Use a recurrent neural network or a convolutional neural network for text classification to realize sentiment analysis.

[0036] Further, a cross-disciplinary in-vehicle temperature environment control method, the step S6 comprises the following sub-steps:

[0037] S61: Continuously adjust the control strategy according to the feedback and behavior pattern of the driver to improve the driving experience;

[0038] S62: Use a stochastic gradient descent algorithm to realize online updating of the model parameters, and continuously learn and optimize the model, and the updating rule is:

[0039]

[0040] Wherein, θ represents the model parameters, J(θ) represents the loss function, The gradient of the loss function with respect to the parameter is represented by, and η represents the learning rate.

[0041] The beneficial effects of the present application are:

[0042] 1. Intelligent control: The cross-disciplinary algorithm and machine learning technology enable the temperature environment of the vehicle to have intelligent control capability, which can adaptively adjust the temperature parameters according to the driving conditions, external temperature, number of people in the vehicle, etc., and provide more intelligent and personalized temperature control;

[0043] 2. Precise adaptation: more accurately predict and adapt to changes in the vehicle's internal and external environment, achieve more accurate temperature control and comfort management, meet different user needs;

[0044] 3. Energy efficiency optimization: through temperature and other component integration algorithms, realize close cooperation, optimize overall vehicle performance and energy efficiency;

[0045] 4. Fast response: achieve faster temperature adjustment response, so that the indoor temperature environment can adapt to temperature changes faster, improve user comfort. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 is a method flowchart. DETAILED DESCRIPTION

[0047] In order to have a clearer understanding of the technical features, objectives and effects of the present application, the specific embodiments of the present application will be described with reference to the accompanying drawings.

[0048] As shown in the accompanying Figure 1 A cross-disciplinary in-vehicle temperature environment control method includes the following sub-steps:

[0049] S1: Monitor the environmental parameters inside and outside the vehicle through sensors, output real-time environmental data;

[0050] S2: Process and analyze data, use machine learning and artificial intelligence algorithms for pattern recognition and trend prediction;

[0051] S3: Develop intelligent control strategies through data analysis results and prediction models;

[0052] S4: Integrated optimization and collaborative control to improve overall vehicle performance and energy efficiency;

[0053] S5: Provide a user interface, allowing drivers to set the in-vehicle temperature environment according to their individual needs;

[0054] S6: Use machine learning algorithms to continuously learn and optimize control strategies.

[0055] Further, a cross-disciplinary in-vehicle temperature environment control method, the step S1 includes the following sub-steps:

[0056] Measure the temperature parameters inside and outside the vehicle through the temperature sensor, output the real-time temperature difference inside and outside the vehicle;

[0057] Detect the humidity parameters inside the vehicle through the humidity sensor, output the real-time humidity inside the vehicle;

[0058] Perceive the external light intensity through the light sensor, output the real-time difference in light intensity inside and outside the vehicle.

[0059] Further, a cross-disciplinary in-vehicle temperature environment control method, step S2 further includes the following sub-steps:

[0060] S21: Calculate the comfort score or energy efficiency index of the in-vehicle temperature environment by linear regression, the formula is:

[0061] y = β0 + β1x1 + β2x2 + … β n x n + ε

[0062] Where y represents the target variable, used to describe the comfort score or energy efficiency index, β0, β1... β n represent the regression coefficients of the linear regression model, x1, x2... x n represent feature variables, used to describe the temperature, humidity, and light intensity inside and outside the vehicle, and ε represents the error term;

[0063] S22: Further process and analyze the data by decision tree and support vector machine to create a prediction model.

[0064] Further, a cross-disciplinary in-vehicle temperature environment control method, step S3 includes the following sub-steps:

[0065] S31: Use the model's deep Q-learning algorithm to learn the in-vehicle temperature environment control strategy, with the update rule:

[0066] Q(s, a)←Q(s, a) + α(r + γmax a Q(s′, a) - Q(s, a))

[0067] Where Q(s, a) represents the expected return value of action a in state S, r represents the immediate reward given by the environment, s' represents the next state after executing action a, α represents the learning rate, γ represents the discount factor, and max a Q(s′, a) represents the maximum value of the discounted factor Q(s′, a);

[0068] S32: Adjust the in-vehicle temperature environment according to the prediction results, choose whether to adjust the in-vehicle temperature according to the driver's driving time, and choose the music type according to the driver's habits.

[0069] Further, a cross-disciplinary in-vehicle temperature environment control method, step S4 includes the following sub-steps:

[0070] S41: Collaborate with the car entertainment system to play music according to the driver's mood and preferences:

[0071] S42: Use a multi-objective genetic algorithm to simultaneously optimize the comfort and energy efficiency of the in-vehicle temperature environment;

[0072] S43: Define the fitness function of MOGA as the weighted sum of comfort index F1 and energy efficiency index F2, use non-dominated sorting and crowdedness to maintain the diversity of the population.

[0073] Further, a cross-disciplinary in-vehicle temperature environment control method, step S5 includes the following sub-steps:

[0074] S5: The driver selects the fatigue, excitement, and comfort options through the interface feedback, and the interface adjusts the in-vehicle temperature environment and music settings according to the selection feedback:

[0075] S52: According to the user interaction and feedback mechanism, use natural language processing algorithm to analyze the feedback emotion of the driver, and adjust the in-vehicle temperature environment and music settings according to the emotion;

[0076] S53: Use recurrent neural network or convolutional neural network for text classification to realize sentiment analysis.

[0077] Further, a cross-disciplinary in-vehicle temperature environment control method, step S6 includes the following sub-steps:

[0078] S61: Continuously adjust the control strategy according to the feedback and behavior pattern of the driver to improve the driving experience;

[0079] S62: Use stochastic gradient descent algorithm to realize online update of model parameters, continuous learning and optimization of model, update rule is:

[0080]

[0081] Where, θ represents the model parameters, J(θ) represents the loss function, The gradient of the loss function with respect to the parameter is represented by, and η represents the learning rate.

[0082] The present scheme realizes intelligent control of the temperature environment of the vehicle through a cross-disciplinary in-vehicle temperature environment control method, which can adaptively adjust the temperature parameters according to the driving conditions, external temperature, number of people in the vehicle, etc., and provide more intelligent and personalized temperature control; more accurately predict and adapt to changes in the internal and external environment of the vehicle, realize more accurate temperature control and comfort management, and meet the different needs of users; through temperature and other component integration algorithm, realize close cooperation, optimize the overall vehicle performance and energy efficiency; realize faster temperature regulation response, so that the in-vehicle temperature environment can adapt to temperature changes faster, and improve the user's comfort feeling.

[0083] The above shows and describes the basic principles and main features of the present application and the advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A cross-disciplinary based in-vehicle temperature environment control method, characterized by, The steps S3 include the following sub-steps: S31: learning the in-vehicle temperature environment control strategy using a deep Q-learning algorithm for the model, with the update rule being: Q(s, a)←Q(s, a)+α(r+γ maxaQ(s′,a)-Q(s,a)) where Q(s, a) represents the expected return value of action a in state s, r represents the immediate reward given by the environment, s′ represents the next state after performing action a, α represents the learning rate, γ represents the discount factor, and maxaQ(s′,a) represents the maximum value of the discounted factor Q(s′,a); S32: adjusting the in-vehicle temperature environment according to the prediction results, selecting whether to adjust the in-vehicle temperature according to the driving time of the driver, and selecting the music type according to the habits of the driver; The steps S4 include the following sub-steps: S41: coordinating with the vehicle entertainment system to play music according to the emotions and preferences of the driver: S42: using a multi-objective genetic algorithm to simultaneously optimize the comfort and energy efficiency of the in-vehicle temperature environment; S43: defining the fitness function of MOGA as the weighted sum of the comfort index F1 and the energy efficiency index F2, and using non-dominated sorting and crowding degree to maintain the diversity of the population. The steps S1 include the following sub-steps: Measuring the temperature parameters inside and outside the vehicle through temperature sensors, outputting the real-time temperature difference inside and outside the vehicle; Detecting the humidity parameters inside the vehicle through humidity sensors, outputting the real-time humidity inside the vehicle; Sensing the external light intensity through light sensors, outputting the real-time light intensity difference inside and outside the vehicle. The steps S2 further include the following sub-steps: S21: calculating the comfort score or energy efficiency index of the in-vehicle temperature environment through linear regression, with the formula being: S22: further processing and analyzing the data through decision trees and support vector machines to create a prediction model. The steps S5 include the following sub-steps:

2. The interdisciplinary in-vehicle thermal environment control method according to claim 1, characterized by, S51: the driver selects the fatigue, excitement, or comfort options through the interface feedback, and the interface adjusts the in-vehicle temperature environment and music settings according to the selection feedback; S52: using natural language processing algorithms to analyze the feedback emotions of the driver according to the user interaction and feedback mechanism, and adjusting the in-vehicle temperature environment and music settings according to the emotions; S53: using recurrent neural networks or convolutional neural networks for text classification to realize sentiment analysis. The steps S6 include the following sub-steps:

3. The interdisciplinary in-vehicle thermal environment control method of claim 1, wherein, S61: continuously adjusting the control strategy according to the feedback and behavior patterns of the driver to improve the driving experience; S62: using the stochastic gradient descent algorithm to realize online updating of the model parameters, continuously learning and optimizing the model, with the update rule being: y＝ wherein y represents a target variable for describing a comfort score or an energy efficiency indicator, a regression coefficient representing a linear regression model, representing a feature variable for describing an indoor and outdoor temperature, humidity, and light intensity, and ε represents an error term. ​ 4. The interdisciplinary in-vehicle thermal environment control method of claim 1, wherein, ​ ​ ​ ​ 5. The interdisciplinary in-vehicle thermal environment control method of claim 1, wherein, ​ ​ ​ where θ represents the model parameters, J(θ) represents the loss function, represents the gradient of the loss function with respect to the parameters, and η represents the learning rate.

Images