Multi-sensor fusion vehicle-mounted adaptive temperature control method and system

Abstract

This invention relates to the field of intelligent cockpit and vehicle thermal management technology, specifically to a multi-sensor fusion in-vehicle adaptive temperature control method and system. The method includes: synchronously acquiring multi-source heterogeneous signals within the vehicle; preprocessing them through filtering and format standardization; using Kalman filtering to perform time synchronization and state estimation, generating a fused feature vector; identifying the current operating condition label, calculating the predicted average voting value, and outputting the target predicted average voting value; calculating the basic control quantity through a feedback controller and the disturbance compensation control quantity through a feedforward controller, and adding them to obtain the total control quantity; decomposing the total control quantity into compressor speed commands, fan airflow commands, damper position commands, and expansion valve opening commands, driving actuator actions, and outputting the temperature control execution result. This invention solves the problems of air conditioning control lag, large temperature fluctuations, and unintelligent mode switching, achieving high-precision temperature control and seamless adaptive adjustment, significantly improving driving and passenger thermal comfort.

Description

Technical Field

[0001] This invention relates to the field of intelligent cockpit and vehicle thermal management technology, specifically to a multi-sensor fusion in-vehicle adaptive temperature control method and system. Background Technology

[0002] As cars evolve from mere transportation tools to a third living space, cabin thermal comfort has become a core indicator for evaluating the driving and riding experience. However, traditional in-vehicle air conditioning control systems face several inherent challenges:

[0003] First, there are limitations in the control target. Most automotive air conditioning systems still rely solely on a single temperature sensor within the passenger cabin as their control target. However, human thermal comfort depends not only on air temperature but also on a combination of factors including humidity, wind speed, solar radiation, clothing thermal resistance, and activity level. This often results in automatic air conditioning systems failing to match the set temperature with the actual feeling of comfort, requiring users to frequently adjust them manually.

[0004] Second, the delayed response to disturbances. A vehicle is a moving, open environment. Disturbances such as opening and closing doors, entering tunnels, sunlight being blocked by clouds, and sudden changes in vehicle speed can rapidly alter the cabin's thermal environment. Traditional proportional-integral-derivative (PID) control based on temperature negative feedback can only respond after the temperature sensor detects a significant change, resulting in noticeable fluctuations in the perceived temperature for occupants.

[0005] Third, the mechanical switching of operating conditions and modes. Traditional air conditioners either rely entirely on manual operation by the user for switching between internal and external circulation, and between blowing air onto the face and feet, or are based on simple logic. They lack intelligent perception of the current operating conditions and cannot automatically select the optimal airflow organization and ventilation strategy.

[0006] Fourth, the parameters of the control algorithm are fixed. The factory-set proportional-integral-derivative parameters are a compromise and cannot adapt to the different comfort needs of different regions, seasons, or even different users.

[0007] Among existing technologies, the closest solution is an automotive automatic air conditioning system based on a single sensor and fixed proportional-integral-derivative (PID) parameters. Specifically, it involves installing one or two temperature sensors on the dashboard or headliner; an electronic controller reads this temperature value and calculates the difference between it and the user-set target temperature; then, using a set of factory-calibrated PID parameters, it calculates the target outlet air temperature and controls the compressor displacement, fan speed, and mixing damper opening. This solution is simple in structure and low in cost, achieving basic constant temperature control, but it suffers from technical drawbacks such as a single sensing dimension, control lag, mismatch between mode and operating conditions, and fixed parameters lacking adaptive capability. Summary of the Invention

[0008] To address the aforementioned technical problems, this invention provides a multi-sensor fusion vehicle-mounted adaptive temperature control method and system, aiming to establish a perception and target system centered on human thermal comfort, realize feedforward and feedback composite control, improve response speed, achieve intelligent mode management based on operating condition recognition, and enable the controller to be adaptive and personalized.

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] On one hand, embodiments of the present invention provide a multi-sensor fusion vehicle-mounted adaptive temperature control method, the method comprising the following steps:

[0011] S100: Acquire multi-source heterogeneous signals collected by the multi-source heterogeneous sensor array, as well as a trained working condition classifier and thermal comfort model. The multi-source heterogeneous signals include multi-point temperature signals inside the vehicle, multi-point humidity signals inside the vehicle, solar radiation intensity signals, and vehicle speed signals.

[0012] S200, the multi-source heterogeneous signal is input into the data acquisition and preprocessing unit, filtering and format standardization are performed, and the preprocessed multi-source heterogeneous signal is output.

[0013] S300, the preprocessed multi-source heterogeneous signal is input into the multi-source data fusion and feature extraction module, and time synchronization and state estimation are performed through Kalman filtering to generate a fused feature vector;

[0014] S400, the fused feature vector is input into the working condition identification and thermal comfort calculation module, the working condition identification process is performed by the working condition classifier, the current working condition label is output, and the predicted average voting value is calculated by the thermal comfort model, and the target predicted average voting value is output.

[0015] S500, the target prediction average voting value is input into the adaptive feedforward-feedback composite controller, the basic control quantity is calculated by the feedback controller, the disturbance compensation control quantity is calculated by the feedforward controller, and the basic control quantity and the disturbance compensation control quantity are added together to obtain the total control quantity.

[0016] S600 inputs the total control quantity into the actuator drive module, decomposes it into compressor speed command, fan air volume command, damper position command and expansion valve opening command, drives the corresponding actuator to act, and outputs the temperature control execution result.

[0017] On the other hand, embodiments of the present invention provide a multi-sensor fusion vehicle-mounted adaptive temperature control system, including:

[0018] At least one processor;

[0019] At least one memory for storing at least one program;

[0020] When the at least one program is executed by the at least one processor, the at least one processor performs the method described above.

[0021] On the other hand, embodiments of the present invention provide a computer-readable storage medium storing a processor-executable program, which, when executed by a processor, is used to perform the above-described method.

[0022] On the other hand, embodiments of the present invention provide a computer program product, including a computer program or computer instructions, the computer program or computer instructions being stored in a memory, a processor of a computer device reading the computer program or computer instructions from the memory, and the processor executing the computer program or computer instructions to cause the computer device to perform the above-described method.

[0023] The embodiments of the present invention have the following beneficial effects:

[0024] This invention establishes a perception and target system centered on human thermal comfort, integrates multi-source sensor data, introduces a thermal comfort model based on predicted average votes, and elevates the control target from air temperature to the thermal comfort of occupants, thus realizing a paradigm shift from temperature control to comfort control.

[0025] This invention achieves combined feedforward and feedback control. It utilizes sensors to detect major disturbances in advance and outputs compensation signals in advance through a feedforward controller, which greatly reduces temperature fluctuations and significantly reduces the amplitude of temperature fluctuations caused by disturbances such as opening doors and entering / exiting tunnels.

[0026] This invention realizes intelligent mode management based on operating condition recognition. It automatically identifies the vehicle's driving environment and operating conditions by using fused features, and intelligently and proactively switches between internal and external circulation, air outlet mode and air volume distribution strategy, upgrading from passive response to proactive service.

[0027] This invention enables the controller to be adaptive and personalized by introducing fuzzy logic to tune the proportional, integral, and derivative parameters online, making it adaptable to different operating conditions. It also achieves personalized self-adaptation of the target temperature curve by recording the user's historical adjustment behavior.

[0028] This invention avoids excessive cooling and ineffective heating through more precise and faster control, thereby saving energy consumption of the air conditioning system under comprehensive operating conditions and indirectly improving the range of new energy vehicles. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a flowchart illustrating the multi-sensor fusion vehicle adaptive temperature control method in an embodiment of the present invention.

[0031] Figure 2 This is a schematic diagram of the overall architecture of the multi-sensor fusion vehicle adaptive temperature control system in this embodiment of the invention. Detailed Implementation

[0032] The following will provide a clear and complete description of the concept, specific structure, and technical effects of the present invention in conjunction with embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of the present invention. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of this invention; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this invention as detailed in the appended claims.

[0034] It is understood that the terms "first," "second," etc., used in this invention can be used to describe various concepts, but unless specifically stated otherwise, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of embodiments of this invention, first information can also be referred to as second information, and similarly, second information can also be referred to as first information. Depending on the context, words such as "if" and "if" as used herein can be interpreted as "at" or "when" or "in response to determination."

[0035] The terms used in this invention, such as at least one, multiple, each, any, etc., include at least one, two or more, multiple, each, each, and any, where at least one includes one, two, or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0036] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this invention is for descriptive purposes only and is not intended to limit the invention.

[0037] refer to Figure 1 ,like Figure 1 The image shows a multi-sensor fusion vehicle-mounted adaptive temperature control method provided by an embodiment of the present invention. The method includes the following steps:

[0038] S100: Acquire multi-source heterogeneous signals collected by the multi-source heterogeneous sensor array, as well as a trained working condition classifier and thermal comfort model. The multi-source heterogeneous signals include multi-point temperature signals inside the vehicle, multi-point humidity signals inside the vehicle, solar radiation intensity signals, and vehicle speed signals.

[0039] S200, the multi-source heterogeneous signal is input into the data acquisition and preprocessing unit, filtering and format standardization are performed, and the preprocessed multi-source heterogeneous signal is output.

[0040] S300, the preprocessed multi-source heterogeneous signal is input into the multi-source data fusion and feature extraction module, and time synchronization and state estimation are performed through Kalman filtering to generate a fused feature vector;

[0041] S400, the fused feature vector is input into the working condition identification and thermal comfort calculation module, the working condition identification process is performed by the working condition classifier, the current working condition label is output, and the predicted average voting value is calculated by the thermal comfort model, and the target predicted average voting value is output.

[0042] S500, the target prediction average voting value is input into the adaptive feedforward-feedback composite controller, the basic control quantity is calculated by the feedback controller, the disturbance compensation control quantity is calculated by the feedforward controller, and the basic control quantity and the disturbance compensation control quantity are added together to obtain the total control quantity.

[0043] S600 inputs the total control quantity into the actuator drive module, decomposes it into compressor speed command, fan air volume command, damper position command and expansion valve opening command, drives the corresponding actuator to act, and outputs the temperature control execution result.

[0044] This invention provides a multi-sensor fusion in-vehicle adaptive temperature control method and system. It constructs a panoramic perception network of the cabin thermal environment through a multi-source heterogeneous sensor array, overcoming the limitations of traditional single-point temperature sensing. A multi-source data fusion and feature extraction module achieves spatiotemporal alignment and feature-level fusion of multi-source heterogeneous signals, providing high-quality input features for subsequent operating condition identification and thermal comfort calculation. The operating condition identification and thermal comfort calculation module elevates the control target from a single air temperature to a human thermal comfort index, realizing intelligent temperature control centered on occupant comfort. An adaptive feedforward-feedback composite controller, combined with feedforward disturbance compensation and feedback fuzzy proportional-integral-derivative online tuning, significantly improves the system's response speed to sudden disturbances and steady-state control accuracy. An actuator drive module precisely maps the control quantity to each actuator in the air conditioning system, achieving coordinated adjustment of the compressor, fan, damper, and expansion valve. Personalized learning and parameter library recording and analysis of user manual intervention behaviors enable personalized self-adaptation of the temperature control target curve, becoming increasingly user-friendly with use.

[0045] refer to Figure 2 , Figure 2 This is a schematic diagram of the overall system architecture in an embodiment of the present invention. The system includes: a multi-source heterogeneous sensor array, a data acquisition and preprocessing unit, a multi-source data fusion and feature extraction module, a working condition identification and thermal comfort calculation module, an adaptive feedforward-feedback composite controller, an actuator drive module, and a personalized learning and parameter library.

[0046] The core of this embodiment lies in the multi-sensor fusion vehicle adaptive temperature control system architecture, which achieves a paradigm shift from mechanical constant temperature to intelligent comfort through five functional levels: multi-source perception, intelligent fusion, operating condition recognition, adaptive control, and personalized learning.

[0047] The multi-source heterogeneous sensor array includes cabin environment sensors, external environment sensors, vehicle status sensors, and air conditioning system status sensors. The cabin environment sensors include: an integrated temperature and humidity sensor installed in each of the driver's seat, front passenger seat, left rear seat, and right rear seat to collect multi-point temperature and humidity signals within the vehicle; and an infrared thermopile sensor installed in the center console for non-contact monitoring of occupant facial and hand skin temperature, outputting occupant skin temperature signals. The external environment sensors include: ambient temperature and humidity sensors installed in the exterior rearview mirrors to collect external temperature and humidity signals; and a dual-channel solar radiation sensor installed on the inside of the windshield to measure solar radiation intensity and direction. The vehicle status sensors read vehicle speed, engine speed or motor speed, air conditioning blower voltage, internal / external air circulation damper position, and air vent temperature signals via the controller area network bus. The air conditioning system status sensors include an evaporator outlet temperature sensor, a condenser pressure sensor, and a compressor speed sensor, which respectively collect evaporator temperature signals, condenser pressure signals, and compressor speed signals.

[0048] In some embodiments, S200, the step of inputting the multi-source heterogeneous signal into the data acquisition and preprocessing unit, performing filtering and format normalization processing, and outputting the preprocessed multi-source heterogeneous signal includes:

[0049] S210, input the multi-point temperature signal and the multi-point humidity signal inside the vehicle into the median filter, remove outliers, and output the filtered cabin environment signal.

[0050] S220, inputs the solar radiation intensity signal, the vehicle speed signal, the vehicle exterior temperature signal, and the vehicle exterior humidity signal into the analog-to-digital converter, performs the conversion of analog signals to digital signals, and outputs digital vehicle exterior environment signals and vehicle status signals;

[0051] S230, unify the format of the digitized external environment signal and the vehicle status signal, and output the preprocessed multi-source heterogeneous signal.

[0052] Specifically, all analog signals are sampled by an analog-to-digital converter, and digital signals are input at a preset frequency through the controller area network interface. Median filtering is performed on the temperature signal to remove outliers, and the filtered temperature signal is output. The same median filtering is performed on the humidity signal. All signals are standardized in format, unifying data types, units, and numerical ranges, and the pre-processed multi-source heterogeneous signal is output.

[0053] In some embodiments, in S300, the step of inputting the preprocessed multi-source heterogeneous signal into the multi-source data fusion and feature extraction module, performing time synchronization and state estimation through Kalman filtering, and generating a fused feature vector includes:

[0054] S310, add a hardware timestamp to the preprocessed multi-source heterogeneous signal, align all the preprocessed multi-source heterogeneous signals to a unified time base, and output the time-synchronized multi-source heterogeneous signal.

[0055] S320, combine the multi-point temperature signal and multi-point humidity signal inside the vehicle from the time-synchronized multi-source heterogeneous signal with the occupant position information to calculate the average heat load value on the left side, the average heat load value on the right side and the temperature difference value between the front and rear rows, and output the spatial distribution characteristics.

[0056] S330, input the time-synchronized multi-source heterogeneous signal into the Kalman filter, perform state estimation and noise suppression, and output the filtered state estimate value;

[0057] S340, based on the filtered state estimate, calculate the cabin temperature change rate, the temperature difference between inside and outside the vehicle, the solar radiation change rate, and the vehicle speed change rate, and output the time-varying characteristics;

[0058] S350, The spatial distribution features and the time-varying features are concatenated to generate the fused feature vector.

[0059] Specifically, in terms of time synchronization, hardware timestamps are added to all preprocessed multi-source heterogeneous signals to align all signals to a unified time base. A Kalman filter is used to perform state estimation and noise suppression on the time-synchronized multi-source heterogeneous signals, outputting filtered state estimates.

[0060] In terms of spatial integration, the system combines temperature and humidity data from the four seats, solar radiation data from the left and right channels, and occupant position information to calculate the average heat load value on the left side, the average heat load value on the right side, and the temperature difference between the front and rear rows, outputting the spatial distribution characteristics. Occupant position information is obtained through seat occupancy sensors or infrared sensors.

[0061] For feature extraction, based on the filtered state estimates, the cabin temperature change rate, the temperature difference between the inside and outside of the vehicle, the solar radiation change rate, and the vehicle speed change rate are calculated, and time-varying features are output. The spatial distribution features and time-varying features are then concatenated to generate a fused feature vector.

[0062] In some embodiments, S400, the step of inputting the fused feature vector into the working condition identification and thermal comfort calculation module, performing working condition identification processing through the working condition classifier, outputting the current working condition label, calculating the predicted average voting value through the thermal comfort model, and outputting the target predicted average voting value includes:

[0063] S410, the fused feature vector is input into the working condition classifier. The working condition classifier performs a decision tree classification operation based on vehicle speed features, external light intensity features, external temperature features, and windshield wiper status features, and outputs the current working condition label. The current working condition label includes high-speed cruising working condition, urban congestion working condition, tunnel entry working condition, underground parking exit working condition, rainy day working condition, and summer sun exposure working condition.

[0064] S420, input the air temperature value, average radiation temperature value, relative humidity value and wind speed value from the fused feature vector into the thermal comfort model. The thermal comfort model calculates the predicted average vote based on the human metabolic rate and the thermal resistance value of clothing, and outputs the predicted average vote value.

[0065] S430, compare the predicted average voting value with the preset comfort range, and when the predicted average voting value exceeds the preset comfort range, generate a temperature control adjustment command and output the target predicted average voting value.

[0066] Specifically, a lightweight decision tree model is used for the driving condition classifier. Input features include vehicle speed, external light intensity, external temperature, and windshield wiper status. Output driving condition labels include highway cruising, urban congestion, tunnel entry, underground parking garage exit, rain, fog / haze, and summer sun exposure. If navigation map data is available, tunnel information ahead is used as an auxiliary input feature to improve the predictability of driving condition identification.

[0067] For thermal comfort calculations, a simplified predicted average vote model is used. Input parameters include air temperature, average radiant temperature, relative humidity, and wind speed. The average radiant temperature is estimated from solar radiation intensity and in-vehicle temperature. The wind speed is estimated based on the fan speed setting. The human metabolic rate is assumed to be the metabolic rate under sedentary conditions. Clothing thermal resistance is estimated based on the outside temperature. The output is a predicted average vote value, ranging from -3 (cold) to 3 (hot), with 0 being thermally neutral. The system's target predicted average vote value is set within the comfort range of -0.5 to 0.5. When the predicted average vote value exceeds the comfort range, a temperature control adjustment command is generated, and the target predicted average vote value is output.

[0068] In some embodiments, S500, the step of inputting the target predicted average voting value into an adaptive feedforward-feedback composite controller, calculating a basic control quantity through the feedback controller, calculating a disturbance compensation control quantity through the feedforward controller, and adding the basic control quantity and the disturbance compensation control quantity to obtain the total control quantity includes:

[0069] S510, the difference between the target predicted average vote value and the actual predicted average vote value is calculated to obtain the predicted average vote deviation value, and the rate of change of the predicted average vote deviation value is calculated to obtain the deviation change rate value.

[0070] S520, the predicted average vote deviation value and the deviation change rate value are input to the feedback controller. The feedback controller includes a fuzzy proportional-integral-derivative controller, which performs online tuning of the proportional coefficient, integral coefficient and derivative coefficient through a fuzzy rule base, and outputs the basic control quantity.

[0071] S530, the solar radiation intensity signal, the vehicle speed signal, and the temperature difference between the inside and outside of the vehicle are input to the feedforward controller. The feedforward controller performs a static mapping operation based on the changes in solar radiation intensity, vehicle speed, and temperature difference between the inside and outside of the vehicle, and outputs the disturbance compensation control quantity.

[0072] S540, perform an addition operation on the basic control quantity and the disturbance compensation control quantity to generate the total control quantity.

[0073] Specifically, in terms of controller architecture, it includes a feedback controller and a feedforward controller. The feedback controller employs a fuzzy proportional-integral-derivative (PID) controller. The feedforward controller constructs a static mapping model between the main disturbances and the compensation values ​​of the control input.

[0074] In feedback control, the difference between the target predicted average vote value and the actual predicted average vote value is used as the predicted average vote deviation value, and the rate of change of the predicted average vote deviation value is calculated as the deviation change rate value. The predicted average vote deviation value and the deviation change rate value are input into the fuzzy proportional-integral-derivative (PID) controller. The fuzzy rule base contains multiple conditional conclusion-type fuzzy rules. Each fuzzy rule uses the deviation linguistic variable and the deviation change rate linguistic variable as conditions, and the proportional coefficient correction value, integral coefficient correction value, and derivative coefficient correction value as conclusions. Fuzzy inference operations are performed, outputting the proportional coefficient correction value, integral coefficient correction value, and derivative coefficient correction value. After defuzzification processing, the precise proportional coefficient correction values, integral coefficient correction values, and derivative coefficient correction values ​​are obtained. The precise proportional coefficient correction values ​​are added to the basic proportional coefficient to obtain the online-tuned proportional coefficient, the precise integral coefficient correction values ​​are added to the basic integral coefficient to obtain the online-tuned integral coefficient, and the precise derivative coefficient correction values ​​are added to the basic derivative coefficient to obtain the online-tuned derivative coefficient. The proportional coefficient, integral coefficient, and derivative coefficient after online tuning are input into the proportional-integral-derivative (PID) arithmetic unit. The predicted average vote deviation is used as input to perform proportional, integral, and derivative operations, and output the basic control quantity.

[0075] In some embodiments, in S520, the step of inputting the predicted average vote deviation value and the deviation change rate value into the feedback controller, wherein the feedback controller includes a fuzzy proportional-integral-derivative (PID) controller, performs online tuning of the proportional coefficient, integral coefficient, and derivative coefficient through a fuzzy rule base, and outputs the basic control quantity, including:

[0076] S521, Input the predicted average vote deviation value and the deviation change rate value into the fuzzification interface, perform fuzzification processing, and output the fuzzified deviation linguistic variable and deviation change rate linguistic variable;

[0077] S522, the fuzzified deviation linguistic variable and the deviation change rate linguistic variable are input into the fuzzy rule base. The fuzzy rule base contains multiple condition-conclusion type fuzzy rules. Each fuzzy rule takes the deviation linguistic variable and the deviation change rate linguistic variable as conditions, and takes the proportional coefficient correction amount, integral coefficient correction amount and differential coefficient correction amount as conclusions. It performs fuzzy inference operations and outputs the proportional coefficient correction amount value, integral coefficient correction amount and differential coefficient correction amount.

[0078] S523, input the proportional coefficient correction value, the integral coefficient correction value and the differential coefficient correction value into the defuzzification interface, perform defuzzification processing, and output accurate proportional coefficient correction value, integral coefficient correction value and differential coefficient correction value;

[0079] S524, the precise proportional coefficient correction amount is added to the basic proportional coefficient to obtain the online tuned proportional coefficient, the precise integral coefficient correction amount is added to the basic integral coefficient to obtain the online tuned integral coefficient, and the precise differential coefficient correction amount is added to the basic differential coefficient to obtain the online tuned differential coefficient.

[0080] S525, the online tuned proportional coefficient, the online tuned integral coefficient, and the online tuned derivative coefficient are input into the proportional-integral-derivative operation unit, and the predicted average vote deviation value is used as input to perform proportional operation, integral operation, and derivative operation, and output the basic control quantity.

[0081] Specifically, in terms of feedforward control, a static mapping model is constructed between solar radiation intensity, vehicle speed, and the temperature difference between the inside and outside of the vehicle, and the control compensation value. The static mapping model is obtained through offline experimental fitting or online identification. When solar radiation intensity suddenly increases, the feedforward controller immediately outputs a positive disturbance compensation control value, instructing the system to increase cooling in advance to compensate for the impending thermal shock. When vehicle speed changes drastically, the feedforward controller outputs a corresponding disturbance compensation control value based on the speed change to compensate for the changes in airflow demand and heat load caused by the speed change. When the temperature difference between the inside and outside of the vehicle changes abruptly, the feedforward controller outputs a corresponding disturbance compensation control value based on the temperature difference change to compensate for the changes in heat transfer load caused by the temperature difference change.

[0082] For the final control quantity calculation, the basic control quantity and the disturbance compensation control quantity are added together to generate the total control quantity. The total control quantity includes the target outlet air temperature value or the compressor base speed value.

[0083] In terms of actuator driving, the total control quantity is decomposed into specific instructions: the electric compressor speed or variable displacement valve current is controlled via pulse width modulation signals; the blower pulse width modulation duty cycle is controlled via a local area network bus; and the positions of the mixing damper, mode damper, and internal / external circulation damper are controlled via stepper motors. After the actuators are activated, the air conditioning system outputs the temperature control execution result, the cabin thermal environment changes accordingly, and the multi-source heterogeneous sensor array collects new multi-source heterogeneous signals, forming a closed-loop control circuit.

[0084] In some embodiments, after S600, the method further includes:

[0085] S700 inputs user manual intervention behavior into the personalized learning and parameter library, records user correction operations on the system's automatic temperature control output, analyzes historical correction data, calculates the personalized predicted average vote offset, adds the personalized predicted average vote offset to the target predicted average vote value, and outputs the personalized corrected target predicted average vote value.

[0086] Specifically, the system continuously records users' manual interventions, including actions such as lowering or raising the target temperature under a specific predicted average vote value. It periodically analyzes historical correction data to calculate the user's personalized predicted average vote offset. This personalized offset is then added to the target predicted average vote value, outputting the personalized corrected target predicted average vote value. For example, if the system learns that the user prefers a cooler environment, the default target will be adjusted from zero to -0.3.

[0087] This invention also provides a multi-sensor fusion vehicle adaptive temperature control system, comprising:

[0088] At least one processor;

[0089] At least one memory for storing at least one program;

[0090] When the at least one program is executed by the at least one processor, the at least one processor performs the method described above.

[0091] The content of the above method embodiments is applicable to this embodiment. The specific functions implemented in this embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments. Therefore, they will not be repeated here.

[0092] This invention also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0093] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0094] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0095] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0096] This invention also provides a computer program product, including a computer program or computer instructions, which are stored in a memory. A processor of a computer device reads the computer program or computer instructions from the memory and executes the computer program or computer instructions, causing the computer device to perform the above-described method.

[0097] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0098] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0099] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0100] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

Claims

1. A multi-sensor fusion method for vehicle-mounted adaptive temperature control, characterized in that, The method includes the following steps: S100: Acquire multi-source heterogeneous signals collected by the multi-source heterogeneous sensor array, as well as a trained working condition classifier and thermal comfort model. The multi-source heterogeneous signals include multi-point temperature signals inside the vehicle, multi-point humidity signals inside the vehicle, solar radiation intensity signals, and vehicle speed signals. S200, the multi-source heterogeneous signal is input into the data acquisition and preprocessing unit, filtering and format standardization are performed, and the preprocessed multi-source heterogeneous signal is output. S300, the preprocessed multi-source heterogeneous signal is input into the multi-source data fusion and feature extraction module, and time synchronization and state estimation are performed through Kalman filtering to generate a fused feature vector; S400, the fused feature vector is input into the working condition identification and thermal comfort calculation module, the working condition identification process is performed by the working condition classifier, the current working condition label is output, and the predicted average voting value is calculated by the thermal comfort model, and the target predicted average voting value is output. S500, the target prediction average voting value is input into the adaptive feedforward-feedback composite controller, the basic control quantity is calculated by the feedback controller, the disturbance compensation control quantity is calculated by the feedforward controller, and the basic control quantity and the disturbance compensation control quantity are added together to obtain the total control quantity. S600 inputs the total control quantity into the actuator drive module, decomposes it into compressor speed command, fan air volume command, damper position command and expansion valve opening command, drives the corresponding actuator to act, and outputs the temperature control execution result.

2. The method according to claim 1, characterized in that, In S200, the step of inputting the multi-source heterogeneous signal into the data acquisition and preprocessing unit, performing filtering and format normalization processing, and outputting the preprocessed multi-source heterogeneous signal includes: S210, input the multi-point temperature signal and the multi-point humidity signal inside the vehicle into the median filter, remove outliers, and output the filtered cabin environment signal. S220, inputs the solar radiation intensity signal, the vehicle speed signal, the vehicle exterior temperature signal, and the vehicle exterior humidity signal into the analog-to-digital converter, performs the conversion of analog signals to digital signals, and outputs digital vehicle exterior environment signals and vehicle status signals; S230, unify the format of the digitized external environment signal and the vehicle status signal, and output the preprocessed multi-source heterogeneous signal.

3. The method according to claim 1, characterized in that, In S300, the step of inputting the preprocessed multi-source heterogeneous signal into the multi-source data fusion and feature extraction module, performing time synchronization and state estimation through Kalman filtering, and generating a fused feature vector includes: S310, add a hardware timestamp to the preprocessed multi-source heterogeneous signal, align all the preprocessed multi-source heterogeneous signals to a unified time base, and output the time-synchronized multi-source heterogeneous signal. S320, combine the multi-point temperature signal and multi-point humidity signal inside the vehicle from the time-synchronized multi-source heterogeneous signal with the occupant position information to calculate the average heat load value on the left side, the average heat load value on the right side and the temperature difference value between the front and rear rows, and output the spatial distribution characteristics. S330, input the time-synchronized multi-source heterogeneous signal into the Kalman filter, perform state estimation and noise suppression, and output the filtered state estimate value; S340, based on the filtered state estimate, calculate the cabin temperature change rate, the temperature difference between inside and outside the vehicle, the solar radiation change rate, and the vehicle speed change rate, and output the time-varying characteristics; S350, The spatial distribution features and the time-varying features are concatenated to generate the fused feature vector.

4. The method according to claim 1, characterized in that, In S400, the step of inputting the fused feature vector into the working condition identification and thermal comfort calculation module, performing working condition identification processing through the working condition classifier, outputting the current working condition label, calculating the predicted average voting value through the thermal comfort model, and outputting the target predicted average voting value includes: S410, the fused feature vector is input into the working condition classifier. The working condition classifier performs a decision tree classification operation based on vehicle speed features, external light intensity features, external temperature features, and windshield wiper status features, and outputs the current working condition label. The current working condition label includes high-speed cruising working condition, urban congestion working condition, tunnel entry working condition, underground parking exit working condition, rainy day working condition, and summer sun exposure working condition. S420, input the air temperature value, average radiation temperature value, relative humidity value and wind speed value from the fused feature vector into the thermal comfort model. The thermal comfort model calculates the predicted average vote based on the human metabolic rate and the thermal resistance value of clothing, and outputs the predicted average vote value. S430, compare the predicted average voting value with the preset comfort range, and when the predicted average voting value exceeds the preset comfort range, generate a temperature control adjustment command and output the target predicted average voting value.

5. The method according to claim 1, characterized in that, In S500, the step of inputting the target predicted average voting value into an adaptive feedforward-feedback composite controller, calculating the basic control quantity through the feedback controller, calculating the disturbance compensation control quantity through the feedforward controller, and adding the basic control quantity and the disturbance compensation control quantity to obtain the total control quantity includes: S510, the difference between the target predicted average vote value and the actual predicted average vote value is calculated to obtain the predicted average vote deviation value, and the rate of change of the predicted average vote deviation value is calculated to obtain the deviation change rate value. S520, the predicted average vote deviation value and the deviation change rate value are input to the feedback controller. The feedback controller includes a fuzzy proportional-integral-derivative controller, which performs online tuning of the proportional coefficient, integral coefficient and derivative coefficient through a fuzzy rule base, and outputs the basic control quantity. S530, the solar radiation intensity signal, the vehicle speed signal, and the temperature difference between the inside and outside of the vehicle are input to the feedforward controller. The feedforward controller performs a static mapping operation based on the changes in solar radiation intensity, vehicle speed, and temperature difference between the inside and outside of the vehicle, and outputs the disturbance compensation control quantity. S540, perform an addition operation on the basic control quantity and the disturbance compensation control quantity to generate the total control quantity.

6. The method according to claim 5, characterized in that, In S520, the predicted average vote deviation value and the deviation change rate value are input into the feedback controller. The feedback controller includes a fuzzy proportional-integral-derivative (PID) controller, which performs online tuning of the proportional coefficient, integral coefficient, and derivative coefficient through a fuzzy rule base, and outputs the basic control quantity, including: S521, Input the predicted average vote deviation value and the deviation change rate value into the fuzzification interface, perform fuzzification processing, and output the fuzzified deviation linguistic variable and deviation change rate linguistic variable; S522, the fuzzified deviation linguistic variable and the deviation change rate linguistic variable are input into the fuzzy rule base. The fuzzy rule base contains multiple condition-conclusion type fuzzy rules. Each fuzzy rule takes the deviation linguistic variable and the deviation change rate linguistic variable as conditions, and takes the proportional coefficient correction amount, integral coefficient correction amount and differential coefficient correction amount as conclusions. It performs fuzzy inference operations and outputs the proportional coefficient correction amount value, integral coefficient correction amount and differential coefficient correction amount. S523, input the proportional coefficient correction value, the integral coefficient correction value and the differential coefficient correction value into the defuzzification interface, perform defuzzification processing, and output accurate proportional coefficient correction value, integral coefficient correction value and differential coefficient correction value; S524, the precise proportional coefficient correction amount is added to the basic proportional coefficient to obtain the online tuned proportional coefficient, the precise integral coefficient correction amount is added to the basic integral coefficient to obtain the online tuned integral coefficient, and the precise differential coefficient correction amount is added to the basic differential coefficient to obtain the online tuned differential coefficient. S525, the online tuned proportional coefficient, the online tuned integral coefficient, and the online tuned derivative coefficient are input into the proportional-integral-derivative operation unit, and the predicted average vote deviation value is used as input to perform proportional operation, integral operation, and derivative operation, and output the basic control quantity.

7. The method according to claim 1, characterized in that, Following S600, the method further includes: S700 inputs user manual intervention behavior into the personalized learning and parameter library, records user correction operations on the system's automatic temperature control output, analyzes historical correction data, calculates the personalized predicted average vote offset, adds the personalized predicted average vote offset to the target predicted average vote value, and outputs the personalized corrected target predicted average vote value.

8. A multi-sensor fusion vehicle-mounted adaptive temperature control system, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor performs the method as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images