Intelligent partition ventilation control system and method for seat

Through the intelligent partition ventilation control system, using data acquisition and model calculation, personalized and real-time adjustment of the seat ventilation system is achieved, solving the problem that traditional systems cannot adapt to personalized needs, and improving riding comfort and ventilation effect.

CN118906937APending Publication Date: 2024-11-08XINZHENG TECH (YONGZHOU) CO LTD
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411183322.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The traditional seat ventilation system cannot adapt to the personalized needs and real-time environmental changes of different riders, resulting in the inability to provide the best comfortable experience.

Method used

Design an intelligent partition ventilation control system for seats, obtain the fitting parameters between the rider and the seat through the data acquisition module, build a human fit model, automatically partition and obtain the seat-related parameters of each partition, and combine the internal environmental parameters of the vehicle. Calculate the comprehensive parameters and enter the ventilation control model to obtain and execute the partition ventilation control parameters.

Benefits of technology

It realizes real-time adaptation to environmental changes, improves partition accuracy and comfort, ensures that each area is within a comfortable range, accurately controls ventilation temperature and wind speed, and improves seat ventilation effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118906937A_ABST
    Figure CN118906937A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of ventilation control, and discloses an intelligent partition ventilation control system and method for a seat. The system comprises a data acquisition module used for presetting an acquisition period, obtaining fitting parameters of a passenger and a seat in the acquisition period, and constructing a human body fitting model through the fitting parameters; the region division module is used for automatically partitioning the human body fitting model so as to obtain relevant seat parameters of each partition; the data integration module is used for obtaining environmental parameters in the vehicle and comfort level parameters of all the subareas, and comprehensively collecting environmental parameters in a period, the comfort level parameters of all the subareas and seat related parameters of all the subareas to obtain comprehensive parameters of all the subareas; according to the invention, the seat can be automatically partitioned, and the partitioning accuracy is improved; the environment change can be adapted in real time; by outputting the ventilation control parameters of all the subareas, the ventilation temperature and wind speed can be accurately controlled, the comfort level of passengers is improved, and the ventilation effect of the seat can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of ventilation control technology, and more specifically, to a seat intelligent zoning ventilation control system and method. Background Art

[0002] With the continuous development and popularization of automobile technology, the automobile industry is undergoing unprecedented changes. Consumers' expectations for automobile products are no longer limited to basic transportation functions, but are more concerned about the all-round comfort, high-level safety protection and highly intelligent operation experience they bring. Among them, comfort has become one of the important factors in passenger experience. The seat ventilation system is a key technology to improve the comfort in the car, and its performance directly affects the passengers' riding experience.

[0003] However, traditional seat ventilation systems are usually based on fixed ventilation modes, which are difficult to adapt to the personalized needs of different passengers and real-time environmental changes, and thus cannot provide passengers with the best comfort experience.

[0004] In view of this, the present invention proposes a seat intelligent zoning ventilation control system and method to solve the above problems. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a seat intelligent partition ventilation control system, comprising:

[0006] A data acquisition module is used to preset a collection period, obtain the fitting parameters between the occupant and the seat during the collection period, and construct a human body fitting model based on the fitting parameters;

[0007] The area division module is used to automatically divide the human body fitting model into different areas, and then obtain the seat-related parameters of each area;

[0008] The data integration module is used to obtain the environmental parameters inside the vehicle and the comfort parameters of each partition, and to obtain the comprehensive parameters of each partition by integrating the environmental parameters, the comfort parameters of each partition and the seat-related parameters of each partition within the collection period;

[0009] The output control module is used to input the comprehensive parameters of each partition into the ventilation control model, thereby obtaining the ventilation control parameters of each partition and executing ventilation control.

[0010] Furthermore, the step of obtaining the fitting parameters between the occupant and the seat during the collection period and constructing a human body fitting model based on the fitting parameters includes:

[0011] Acquire seat specification data, and construct a three-dimensional seat model based on the seat specification data;

[0012] A scatter plot is set at the back and buttocks of the three-dimensional seat model against the wall, and the scatter plot is composed of R evenly distributed scattered points;

[0013] Construct a scatter plot coordinate system, pre-process the pressure value, temperature value and humidity value corresponding to the same scatter point in the acquisition period, and obtain the fitting parameters;

[0014] The scatter points with fitting parameters on the scatter plot are marked to form a human body fitting model.

[0015] Furthermore, the method of preprocessing the pressure value, temperature value and humidity value corresponding to the same scattered point in the collection period includes:

[0016] The pressure values, temperature values ​​and humidity values ​​corresponding to the same scattered point are subjected to outlier removal, denoising and standardization;

[0017] The pressure value, temperature value and humidity value corresponding to the same scattered point are weighted and fused. The formula for weighted fusion is: TH = Q 1 ·W 1 +Q 2 ·W 2 +Q 3 ·W 3 , where TH is the fitting parameter, Q 1 is the pressure value, Q 2 is the humidity value, Q 3 is the temperature value, W 1 is the weight coefficient corresponding to the pressure value, W 2 is the weight coefficient corresponding to the humidity value, W 3 is the weight coefficient corresponding to the temperature value.

[0018] Further, obtaining the W 1 , W 2 and W 3 The steps include:

[0019] Initialize the population and randomly generate m groups of W by pre-set conditions. 1 , W 2 and W 3 The weight coefficient combination formed is used as the initial population, and each group of weight coefficient W 1 , W 2 and W 3 As an individual genome, the preset condition is 1>W 3 >W 2 >W 1 >0 and W 1 +W 2 +W 3 =1;

[0020] Define a fitness function for evaluating the fitness of each individual, the fitness function is fitness = -(W 3 ·|Q 3 -Q' 3 |+W 2 ·|Q 2 -Q' 2 |+W 1 ·|Q 1 -Q' 1 |), where fitness is the fitness, |Q 3 -Q' 3 | is the temperature value Q 3 and the ideal temperature value Q' 3 The absolute difference between |Q 2 -Q' 2 | is the humidity value Q 2 and ideal humidity value Q' 2 The absolute difference between |Q 1 -Q' 1 | is the pressure value Q 1 and ideal pressure value Q' 1 The absolute difference between them is used to calculate the fitness of each individual in the initial population;

[0021] Sort the fitness of individuals from high to low, and select the individuals with the top K% fitness as the parents of the next generation through the tournament;

[0022] Perform crossover and mutation on the selected parent individuals to obtain offspring individuals;

[0023] The offspring individuals and the initial population are combined into a new population, and the iteration is repeated until the preset iteration stop condition is met. The individual with the highest fitness is taken as the optimal solution. The weight coefficient combination corresponding to the optimal solution is W 1 , W 2 and W 3 , where the iteration stopping condition is that the maximum number of iterations or the improvement in fitness is less than u%.

[0024] Furthermore, the step of automatically partitioning the human body fitting model and obtaining seat-related parameters of each partition includes:

[0025] Obtain the fitting parameters of each scattered point in the human body fitting model, and use the fitting parameters of each scattered point as the feature vector G;

[0026] Use the feature vector G of all scattered points in the human body fitting model to build a KD tree data structure, automatically partition the scattered points on the human body fitting model through DBSCAN, define the domain radius ∈ and the minimum number of scattered points minpts, and record the scattered points with a number of neighbors greater than or equal to minPts as core points;

[0027] Initialize all scattered points to an unvisited state, traverse each unvisited scattered point, and use the KD tree to find all scattered points within the domain radius ∈ as neighbor scattered points; if the number of neighbor scattered points of the current scattered point is greater than or equal to minPts, the current scattered point is a core point, and the cluster is continued to be expanded. Recursively check whether each neighbor scattered point is also a core point. If so, add the neighbor scattered point to the cluster; if the number of neighbor scattered points of the current scattered point is less than minPts, but belongs to the domain radius ∈ of a core point, then add the current scattered point to the existing cluster;

[0028] Gradually reduce the domain radius ∈ and increase the minimum number of scattered points minpts to obtain H clustering results, calculate the silhouette coefficient of the H clustering results, and take the clustering result corresponding to the highest silhouette coefficient as the final clustering result;

[0029] The scattered points contained in each cluster in the final clustering result are taken as a partition, and the average pressure value, average temperature value and average humidity value corresponding to all the scattered points in each partition are taken as the seat-related parameters of this partition.

[0030] Furthermore, the step of obtaining the comprehensive parameters of each partition includes:

[0031] Obtaining the vehicle's interior temperature and humidity as environmental parameters;

[0032] Obtain the maximum temperature difference and the maximum humidity difference of each zone as the comfort parameters of the zone;

[0033] The seat-related parameters of each partition are used as the seat-related parameters of the partition;

[0034] The environmental parameters inside the vehicle, the comfort parameters of each partition and the seat-related parameters of each partition during the collection period are used as the comprehensive parameters of each partition.

[0035] Furthermore, the step of obtaining the maximum temperature difference and the maximum humidity difference of each partition includes:

[0036] Define the comfortable temperature and humidity values ​​inside the vehicle;

[0037] Calculate the difference between the highest temperature value and the comfortable temperature value among all scattered points in each partition, and record this difference as the maximum temperature difference;

[0038] Calculate the difference between the highest humidity value and the comfortable humidity value among all scattered points in each partition, and record this difference as the maximum humidity difference.

[0039] Furthermore, the step of obtaining the ventilation control parameters of each partition includes:

[0040] The comprehensive parameters of each partition are input into the pre-trained ventilation control model, so as to obtain the ventilation control parameters of each partition, the ventilation control parameters including ventilation temperature and ventilation wind speed;

[0041] The training step of the ventilation control model comprises:

[0042] Collecting historical comprehensive parameters and historical ventilation control parameters corresponding to the historical comprehensive parameters, and taking the historical comprehensive parameters and the historical ventilation control parameters corresponding to the historical comprehensive parameters as a sample set;

[0043] Number the historical ventilation control parameters, construct the feature vector P of the historical comprehensive parameters, use TensorFlow to define the input layer and output layer and build the model framework; use the feature vector P as the input layer and the historical ventilation control parameter number as the output layer;

[0044] The sample set is divided into a training set and a test set according to the ratio of J% and (1-J)%, a ventilation control model is constructed, the ventilation control model is trained to obtain an initial ventilation control model, the initial ventilation control model is tested using the test set, and the initial ventilation control model that meets the preset accuracy is output, wherein the ventilation control model is one of a logistic regression model, a naive Bayes model or a support vector machine model.

[0045] Furthermore, the step of obtaining ventilation control parameters of each partition and performing ventilation control includes:

[0046] The ventilation control parameters of each zone are input into the MCU in the vehicle, and the ventilation system in the seat is regulated by the MCU.

[0047] A seat intelligent zoning ventilation control method comprises the following steps:

[0048] S1. Preset a collection period, obtain the fitting parameters between the occupant and the seat during the collection period, and build a human body fitting model based on the fitting parameters;

[0049] S2, automatically partitioning the human body fitting model, and then obtaining seat-related parameters of each partition;

[0050] S3, obtaining the environmental parameters inside the vehicle and the comfort parameters of each partition, and obtaining the comprehensive parameters of each partition by comprehensively collecting the environmental parameters within a period, the comfort parameters of each partition and the seat-related parameters of each partition;

[0051] S4. Input the comprehensive parameters of each partition into the ventilation control model, thereby obtaining the ventilation control parameters of each partition and executing ventilation control.

[0052] The technical effects and advantages of the seat intelligent zoning ventilation control system and method of the present invention are as follows:

[0053] 1. By integrating pressure, temperature and humidity values ​​into fitting parameters through weighted fusion, more representative fitting parameters can be generated, thereby improving the accuracy of partitioning; by optimizing weight coefficients using genetic algorithms, excellent weight coefficient combinations can be selected; by using KD trees to build data structures and combining DBSCAN algorithms for automatic partitioning, the accuracy of partitioning can be improved;

[0054] 2. By presetting the collection cycle, the system can adapt to environmental changes in real time; by calculating the maximum temperature difference and maximum humidity difference of each partition to objectively evaluate the comfort, it can ensure that each area is within the comfort range; by outputting the ventilation control parameters of each partition, the ventilation temperature and wind speed can be accurately controlled, the passenger comfort can be improved, and the ventilation effect of the seat can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a structural schematic diagram of a seat intelligent zoned ventilation control system of the present invention;

[0056] Figure 2 It is a flow chart of a seat intelligent zoning ventilation control method of the present invention. DETAILED DESCRIPTION

[0057] 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.

[0058] Example 1

[0059] See also Figure 1 As shown, the seat intelligent zoned ventilation control system described in this embodiment includes:

[0060] A data acquisition module is used to preset a collection period, obtain the fitting parameters between the occupant and the seat during the collection period, and construct a human body fitting model based on the fitting parameters;

[0061] The steps of obtaining the fitting parameters between the occupant and the seat during the collection period and constructing a human body fitting model based on the fitting parameters include:

[0062] Acquire the seat specification data (length, width, height and other dimensions) within the collection period. The duration of the collection period is set by relevant personnel according to actual conditions. Construct a three-dimensional seat model based on the seat specification data.

[0063] A scatter plot is set at the back of the three-dimensional seat model against the wall (in contact with the back of the passenger) and the buttocks against the wall (in contact with the buttocks and legs of the passenger), and the scatter plot is composed of R evenly distributed scattered points;

[0064] Construct a scatter plot coordinate system, pre-process the pressure value, temperature value and humidity value corresponding to the same scatter point in the acquisition period, and obtain the fitting parameters;

[0065] Among them, after the passenger fastens the seat belt or the vehicle starts to move, the pressure value corresponding to each scattered point is obtained through the pressure pad sensor (pressure distribution pad) or the pressure distribution sensor array on the seat; the temperature value corresponding to each scattered point is obtained through the thermocouple array or thermistor array on the seat; the humidity value corresponding to each scattered point is obtained through the capacitive humidity sensor array or the resistive humidity sensor array on the seat;

[0066] Scatter points on the scatter plot that have fitting parameters (scatter points corresponding to any data of pressure, humidity different from the vehicle interior, and temperature different from the vehicle interior are considered to have fitting parameters) are marked, and the marked scatter points then form a human body fitting model.

[0067] The methods for preprocessing the pressure value, temperature value and humidity value corresponding to the same scattered point in the collection period include:

[0068] The pressure values, temperature values ​​and humidity values ​​corresponding to the same scattered point are subjected to outlier removal, denoising and standardization;

[0069] Detect and eliminate abnormal values ​​and noise in the pressure values, temperature values, and humidity values ​​corresponding to the same scattered point in the acquisition period. The abnormal values ​​in the pressure values, temperature values, and humidity values ​​corresponding to the same scattered point in the acquisition period are identified and eliminated by the standard deviation method. Then, the pressure values, temperature values, and humidity values ​​corresponding to the same scattered point in the acquisition period are denoised by the median filter. Finally, the pressure values, temperature values, and humidity values ​​corresponding to the same scattered point are standardized.

[0070] The pressure value, temperature value and humidity value corresponding to the same scattered point are weighted and integrated. The weighted integration formula is: TH = Q 1 ·W 1 +Q 2 ·W 2 +Q 3 ·W 3 , where TH is the fitting parameter, Q 1 is the pressure value, Q 2 is the humidity value, Q 3 is the temperature value, W 1 is the weight coefficient corresponding to the pressure value, W 2 is the weight coefficient corresponding to the humidity value, W 3 is the weight coefficient corresponding to the temperature value.

[0071] Get W 1 , W 2 and W 3 The steps include:

[0072] Initialize the population and randomly generate m groups of W by pre-set conditions. 1 , W 2 and W 3 The weight coefficient combination formed is used as the initial population, and each group of weight coefficient W 1 , W 2 and W 3 As an individual genome, the default condition is 1>W 3 >W 2 >W 1 >0 and W 1 +W 2 +W 3 =1;

[0073] Define the fitness function used to evaluate the fitness of each individual, the fitness function is fitness = -(W 3 ·|Q 3 -Q' 3 |+W 2 ·|Q 2 -Q' 2 |+W 1 ·|Q 1 -Q' 1 |), where fitness is the fitness, |Q 3 -Q' 3 | is the temperature value Q 3 and the ideal temperature value Q' 3 The absolute difference between |Q 2 -Q' 2 | is the humidity value Q 2 and ideal humidity value Q' 2 The absolute difference between |Q 1 -Q' 1 | is the pressure value Q 1 and ideal pressure value Q' 1 The absolute difference between the actual value and the ideal value is used to calculate the fitness of each individual in the initial population (the value of the fitness function). The fitness function is designed to be a negative value because the goal is to minimize the difference between the actual value and the ideal value, so that a smaller difference corresponds to a higher fitness value.

[0074] Specifically, the ideal temperature value is preset to a relatively low value, such as 20℃-22℃, so that the temperature difference can be more obvious and easier to optimize; the ideal humidity value should be preset to a relatively low value, such as 40%-50%, so that the humidity difference can be more obvious and easier to optimize; the ideal pressure value should be preset to a relatively high value, such as 100mmHg-120mmHg, so that the pressure value difference can be smaller and less easy to optimize; through such preset, the fitting information can better highlight the temperature and humidity, and finally the pressure value; or the ideal temperature value, ideal humidity value and ideal pressure value are set by relevant personnel according to actual conditions.

[0075] Sort the fitness of individuals from high to low, and select the individuals with the top K% fitness as the parents of the next generation through the tournament;

[0076] Perform crossover and mutation on the selected parent individuals to obtain offspring individuals;

[0077] The offspring individuals and the initial population are combined into a new population, and the iteration is repeated until the preset iteration stop condition is met. The individual with the highest fitness is taken as the optimal solution. The weight coefficient combination corresponding to the optimal solution is W 1 , W 2 and W 3 Among them, the iteration stopping condition is that the maximum number of iterations or the improvement value of fitness is less than u%, and K% and u% are set by relevant personnel according to actual conditions.

[0078] It should be noted that by obtaining the specification data of the seat and constructing a three-dimensional seat model, the actual shape and characteristics of the seat can be accurately reflected, and the high accuracy and reliability of the human body fitting model can be ensured; by eliminating outliers, denoising and standardizing the scattered pressure values, temperature values ​​and humidity values, the data quality can be significantly improved, which helps to obtain more accurate and consistent fitting parameters and improves the robustness of data processing; by using the weighted fusion method to integrate the pressure values, temperature values ​​and humidity values ​​into fitting parameters, the influence of different data sources can be fully considered, and more representative fitting parameters can be generated, which is helpful for the accuracy of subsequent partition divisions; by using genetic algorithms to optimize the combination of weight coefficients, the optimal solution can be found in a complex data space to ensure that the weight coefficient combination corresponding to the optimal solution finally selected can minimize the difference between the actual value (temperature value, humidity value, pressure value) and the ideal value (ideal temperature value, ideal humidity value, ideal pressure value).

[0079] The area division module is used to automatically divide the human body fitting model into different areas, and then obtain the seat-related parameters of each area;

[0080] The steps of automatically partitioning the human body fitting model and obtaining seat-related parameters of each partition include:

[0081] Obtain the fitting parameters of each scattered point in the human body fitting model, and use the fitting parameters of each scattered point as the feature vector G;

[0082] Use the feature vector G of all scattered points in the human body fitting model to build a KD tree data structure, automatically partition the scattered points on the human body fitting model through DBSCAN, define the domain radius ∈ and the minimum number of scattered points minpts, and record the scattered points with a number of neighbors greater than or equal to minPts as core points;

[0083] Initialize all scattered points to an unvisited state, traverse each unvisited scattered point, and use the KD tree to find all scattered points within the domain radius ∈ as neighbor scattered points; if the number of neighbor scattered points of the current scattered point is greater than or equal to minPts, the current scattered point is a core point, and the cluster is continued to be expanded. Recursively check whether each neighbor scattered point is also a core point. If so, add the neighbor scattered point to the cluster; if the number of neighbor scattered points of the current scattered point is less than minPts, but belongs to the domain radius ∈ of a core point, then add the current scattered point to the existing cluster;

[0084] Among them, in the DBSCAN algorithm, the "current scatter point" refers to the point currently being considered or processed during the execution of the algorithm;

[0085] Gradually reduce the domain radius ∈ and increase the minimum number of scattered points minpts to obtain H clustering results, calculate the silhouette coefficient of the H clustering results, and take the clustering result corresponding to the highest silhouette coefficient as the final clustering result;

[0086] The scattered points contained in each cluster in the final clustering result are taken as a partition, and the average pressure value, average temperature value and average humidity value corresponding to all the scattered points in each partition are taken as the seat-related parameters of this partition.

[0087] It should be noted that by using the fitting parameters of each scattered point in the human body fitting model to construct a KD tree data structure and combining it with the DBSCAN algorithm for automatic partitioning, the seat area can be efficiently and accurately divided into multiple partitions, which is beneficial to the subsequent partition ventilation control; by using the KD tree for efficient neighborhood query, the calculation speed of the DBSCAN algorithm can be significantly accelerated, and the processing efficiency is improved; by dynamically adjusting the domain radius ∈ and the minimum number of scattered points minPts, and calculating the silhouette coefficient of multiple clustering results, the optimal clustering result can be selected; by using the silhouette coefficient as an evaluation indicator, the separation and compactness of the cluster can be effectively evaluated, thereby improving the quality and reliability of the partition; by using the KD tree to find all the scattered points in the domain, the core points can be quickly determined, and the neighbor scattered points can be recursively checked during the cluster expansion process, which can efficiently identify the core points and expand the cluster, thereby ensuring the accuracy and completeness of the partition; when the number of neighbors of the scattered point is insufficient, it is marked as a boundary point or a noise point and classified into an existing cluster, which can effectively reduce the misclassification and ensure the accuracy of the final partition; by selecting the result with the highest silhouette coefficient as the final partition, the superiority of the final clustering result is ensured.

[0088] In this embodiment, by obtaining the specification data of the seat and constructing a three-dimensional seat model, the high precision and reliability of the human body fitting model are ensured, which helps to improve the accuracy of the fitting parameters and the reliability of subsequent partitioning; by integrating the pressure values, temperature values ​​and humidity values ​​into fitting parameters through weighted fusion, the influence of different data sources can be fully considered, and more representative fitting parameters can be generated, thereby improving the accuracy of partitioning; by optimizing the weight coefficients using a genetic algorithm, an excellent combination of weight coefficients can be selected; by using a KD tree to construct a data structure and combining it with the DBSCAN algorithm for automatic partitioning, the accuracy of partitioning can be improved; by selecting the clustering result with the highest silhouette coefficient as the final partition, the superiority of the final clustering result can be ensured.

[0089] Example 2

[0090] See also Figure 1 As shown, this embodiment provides a seat intelligent zoned ventilation control system, including:

[0091] The data integration module is used to obtain the environmental parameters inside the vehicle and the comfort parameters of each partition, and to obtain the comprehensive parameters of each partition by integrating the environmental parameters, the comfort parameters of each partition and the seat-related parameters of each partition within the collection period;

[0092] The steps of obtaining comprehensive parameters of each partition include:

[0093] Obtaining an interior temperature value and an interior humidity value of the vehicle as environmental parameters, wherein the interior temperature value is an average interior temperature value, and the interior humidity value is an average interior humidity value;

[0094] Obtain the maximum temperature difference and the maximum humidity difference of each zone as the comfort parameters of the zone;

[0095] The seat-related parameters of each partition are used as the seat-related parameters of the partition;

[0096] The environmental parameters inside the vehicle, the comfort parameters of each partition and the seat-related parameters of each partition during the collection period are used as the comprehensive parameters of each partition.

[0097] The steps for obtaining the maximum temperature difference and the maximum humidity difference of each zone include:

[0098] Define the comfortable temperature and humidity values ​​inside the vehicle;

[0099] Specifically, the comfort temperature value is the temperature value inside the vehicle that is considered comfortable during riding, such as 22°C or 24°C. The comfort temperature value is generally obtained based on user feedback and environmental test data.

[0100] Specifically, the comfortable humidity value is a humidity value that is considered comfortable inside the vehicle during riding, such as 50% RH or 55% RH. The comfortable humidity value is generally obtained based on user feedback and environmental test data.

[0101] Calculate the difference between the highest temperature value and the comfortable temperature value among all scattered points in each partition, and record this difference as the maximum temperature difference;

[0102] Calculate the difference between the highest humidity value and the comfortable humidity value among all scattered points in each partition, and record this difference as the maximum humidity difference.

[0103] It should be noted that by collecting data regularly, it is possible to monitor and record changes in the vehicle's internal environment and seat-related parameters in real time, thereby improving the timeliness and completeness of the comprehensive parameters and providing accurate basic data support for subsequent ventilation control; by calculating the maximum temperature difference and the maximum humidity difference, it is possible to objectively evaluate the comfort of each partition and ensure that each area is within a comfortable range, thereby improving the overall riding experience of the passengers.

[0104] The output control module is used to input the comprehensive parameters of each partition into the ventilation control model, thereby obtaining the ventilation control parameters of each partition and executing ventilation control.

[0105] The steps of obtaining ventilation control parameters of each partition include:

[0106] The comprehensive parameters of each partition are input into the pre-trained ventilation control model, so as to obtain the ventilation control parameters of each partition, the ventilation control parameters including ventilation temperature and ventilation wind speed;

[0107] The training steps of the ventilation control model include:

[0108] Collecting historical comprehensive parameters and historical ventilation control parameters corresponding to the historical comprehensive parameters, and taking the historical comprehensive parameters and the historical ventilation control parameters corresponding to the historical comprehensive parameters as a sample set;

[0109] Specifically, the samples in the sample set are historical environmental parameters, historical comfort parameters, historical seat-related parameters, and corresponding historical ventilation control parameters extracted from experiments or historical records, in which passengers have a high evaluation or good experience with seat ventilation;

[0110] Number the historical ventilation control parameters, construct the feature vector P of the historical comprehensive parameters, use TensorFlow to define the input layer and output layer and build the model framework; use the feature vector P as the input layer and the historical ventilation control parameter number as the output layer;

[0111] Divide the sample set into a training set and a test set according to the ratio of J% and (1-J)%, construct a ventilation control model, train the ventilation control model to obtain an initial ventilation control model, test the initial ventilation control model using the test set, and output an initial ventilation control model that meets a preset accuracy, wherein the ventilation control model is one of a logistic regression model, a naive Bayes model, or a support vector machine model;

[0112] The calculation formula for prediction accuracy is Z f =(A f -W f ) 2 , where is the historical comprehensive parameter number, Z f is the prediction accuracy of the fth group, A f is the prediction number corresponding to the fth group of historical comprehensive parameters, W f is the actual number corresponding to the fth group of historical comprehensive parameters.

[0113] The steps of obtaining ventilation control parameters of each partition and performing ventilation control include:

[0114] The ventilation control parameters of each partition are input into the MCU (control center or micro control unit) in the vehicle, and the ventilation system in the seat (including ventilation components, heating components and cooling components) is regulated by the MCU;

[0115] Among them, the seat is equipped with a ventilation component (such as a micro fan array for ventilation, heat dissipation or auxiliary heating / cooling), a heating component (an electric heating wire network or a heating pad for providing heating) and a cooling component (a micro thermoelectric cooler array for providing cooling). The above three components are embedded in the seat structure in an array form to ensure that each position of the seat is covered by the three components, and the positions of the above three components correspond or are associated with the scattered coordinates to achieve precise adjustment of the ventilation temperature and ventilation speed.

[0116] It should be noted that by outputting the ventilation control parameters of each partition, more effective seat ventilation control can be achieved, passenger comfort can be improved, and the in-vehicle environment can be optimized; by training the model with historical comprehensive parameters, ventilation control can be optimized based on actual conditions, thereby improving the accuracy and reliability of the model; using TensorFlow to build a model framework can improve the efficiency of model training and optimization, and can reduce development time; through accuracy evaluation, the effectiveness of the ventilation control model in actual applications can be ensured, and the system's predictive ability and control accuracy can be improved; by inputting the ventilation control parameters of each partition into the MCU, the MCU uses the ventilation control parameters of each partition as the target parameters of the ventilation system. Based on this target parameter, the ventilation system regulates its fan, heating wire, and cooler, thereby improving the comfort and functionality of the seat and improving the passenger experience.

[0117] In this embodiment, by presetting the collection cycle, accurate basic data support can be provided for subsequent ventilation control, so that the system can adapt to environmental changes in real time; by calculating the maximum temperature difference and maximum humidity difference of each partition to objectively evaluate the comfort, it can ensure that each area is within the comfort range, thereby improving the overall riding experience of the passengers, and improving the accuracy of riding comfort through scientific evaluation; by outputting the ventilation control parameters of each partition, the ventilation temperature and wind speed can be accurately controlled, thereby improving the comfort of passengers and improving the ventilation effect of the seats.

[0118] Example 3

[0119] See also Figure 2 As shown, the present embodiment provides a seat intelligent zone ventilation control method, comprising the following steps:

[0120] S1. Preset a collection period, obtain the fitting parameters between the occupant and the seat during the collection period, and build a human body fitting model based on the fitting parameters;

[0121] S2, automatically partitioning the human body fitting model, and then obtaining seat-related parameters of each partition;

[0122] S3, obtaining the environmental parameters inside the vehicle and the comfort parameters of each partition, and obtaining the comprehensive parameters of each partition by comprehensively collecting the environmental parameters within a period, the comfort parameters of each partition and the seat-related parameters of each partition;

[0123] S4. Input the comprehensive parameters of each partition into the ventilation control model, thereby obtaining the ventilation control parameters of each partition and executing ventilation control.

[0124] In this embodiment, the pressure value, temperature value and humidity value are integrated into fitting parameters through weighted fusion, which can fully consider the influence of different data sources and generate more representative fitting parameters; by using the KD tree to construct the data structure and combining the DBSCAN algorithm for automatic partitioning, the accuracy of the partitioning can be improved; by presetting the collection cycle, it can provide accurate basic data support for subsequent ventilation regulation, so that the system can adapt to environmental changes in real time; by outputting the ventilation control parameters of each partition, the ventilation temperature and wind speed can be accurately controlled, the passenger comfort can be improved, and the ventilation effect of the seat can be improved.

[0125] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0126] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0127] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0128] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A seat intelligent zoning ventilation control system, characterized in that: include: A data acquisition module is used to preset a collection period, obtain the fitting parameters between the occupant and the seat during the collection period, and construct a human body fitting model based on the fitting parameters; The area division module is used to automatically divide the human body fitting model into different areas, and then obtain the seat-related parameters of each area; The data integration module is used to obtain the environmental parameters inside the vehicle and the comfort parameters of each partition, and to obtain the comprehensive parameters of each partition by integrating the environmental parameters, the comfort parameters of each partition and the seat-related parameters of each partition within the collection period; The output control module is used to input the comprehensive parameters of each partition into the ventilation control model, thereby obtaining the ventilation control parameters of each partition and executing ventilation control.

2. The seat intelligent zoning ventilation control system according to claim 1 is characterized in that: The step of obtaining the fitting parameters between the occupant and the seat during the collection period and constructing a human body fitting model based on the fitting parameters comprises: Acquire seat specification data, and construct a three-dimensional seat model based on the seat specification data; A scatter plot is set at the back and buttocks of the three-dimensional seat model against the wall, and the scatter plot is composed of R evenly distributed scattered points; Construct a scatter plot coordinate system, pre-process the pressure value, temperature value and humidity value corresponding to the same scatter point in the acquisition period, and obtain the fitting parameters; The scatter points with fitting parameters on the scatter plot are marked to form a human body fitting model.

3. The seat intelligent zoning ventilation control system according to claim 2 is characterized in that: The method of preprocessing the pressure value, temperature value and humidity value corresponding to the same scattered point in the acquisition period includes: The pressure values, temperature values ​​and humidity values ​​corresponding to the same scattered point are subjected to outlier removal, denoising and standardization; The pressure value, temperature value and humidity value corresponding to the same scattered point are weightedly fused, and the formula for weighted fusion is: TH=Q1·W1+Q2·W2+Q3·W3, where TH is the fitting parameter, Q1 is the pressure value, Q2 is the humidity value, Q3 is the temperature value, W1 is the weight coefficient corresponding to the pressure value, W2 is the weight coefficient corresponding to the humidity value, and W3 is the weight coefficient corresponding to the temperature value.

4. The seat intelligent zoning ventilation control system according to claim 3 is characterized in that: The steps of obtaining W1, W2 and W3 include: Initialize the population, randomly generate m groups of weight coefficient combinations consisting of W1, W2 and W3 as the initial population according to preset conditions, and take each group of weight coefficients W1, W2 and W3 as the genome of an individual, wherein the preset conditions are 1>W3>W2>W1>0 and W1+W2+W3=1; Define a fitness function for evaluating the fitness of each individual, the fitness function is fitness = -(W3·|Q3-Q'3|+W2·|Q2-Q'2|+W1·|Q1-Q'1|), where fitness is fitness, |Q3-Q3'| is the absolute difference between the temperature value Q3 and the ideal temperature value Q3', |Q2-Q'2| is the absolute difference between the humidity value Q2 and the ideal humidity value Q'2, |Q1-Q'1| is the absolute difference between the pressure value Q1 and the ideal pressure value Q'1, and calculate the fitness of each individual in the initial population; Sort the fitness of individuals from high to low, and select the individuals with the top K% fitness as the parents of the next generation through the tournament; Perform crossover and mutation on the selected parent individuals to obtain offspring individuals; The offspring individuals are combined with the initial population to form a new population, and the iteration is repeated until the preset iteration stop condition is met. The individual with the highest fitness is taken as the optimal solution. The weight coefficient combination corresponding to the optimal solution is W1, W2 and W3, where the iteration stop condition is the maximum number of iterations or the fitness improvement value is less than u%.

5. The seat intelligent zoning ventilation control system according to claim 3 is characterized in that: The step of automatically partitioning the human body fitting model and obtaining seat-related parameters of each partition includes: Obtain the fitting parameters of each scattered point in the human body fitting model, and use the fitting parameters of each scattered point as the feature vector G; Use the feature vector G of all scattered points in the human body fitting model to build a KD tree data structure, automatically partition the scattered points on the human body fitting model through DBSCAN, define the domain radius ∈ and the minimum number of scattered points minpts, and record the scattered points with a number of neighbors greater than or equal to minPts as core points; Initialize all scattered points to an unvisited state, traverse each unvisited scattered point, and use the KD tree to find all scattered points within the domain radius ∈ as neighbor scattered points; if the number of neighbor scattered points of the current scattered point is greater than or equal to minPts, the current scattered point is a core point, and the cluster is continued to be expanded. Recursively check whether each neighbor scattered point is also a core point. If so, add the neighbor scattered point to the cluster; if the number of neighbor scattered points of the current scattered point is less than minPts, but belongs to the domain radius ∈ of a core point, then add the current scattered point to the existing cluster; Gradually reduce the domain radius ∈ and increase the minimum number of scattered points minpts to obtain H clustering results, calculate the silhouette coefficient of the H clustering results, and take the clustering result corresponding to the highest silhouette coefficient as the final clustering result; The scattered points contained in each cluster in the final clustering result are taken as a partition, and the average pressure value, average temperature value and average humidity value corresponding to all the scattered points in each partition are taken as the seat-related parameters of this partition.

6. The seat intelligent zoning ventilation control system according to claim 3 is characterized in that: The step of obtaining the comprehensive parameters of each partition includes: Obtaining the vehicle's interior temperature and humidity as environmental parameters; Obtain the maximum temperature difference and the maximum humidity difference of each zone as the comfort parameters of the zone; The seat-related parameters of each partition are used as the seat-related parameters of the partition; The environmental parameters inside the vehicle, the comfort parameters of each partition and the seat-related parameters of each partition during the collection period are used as the comprehensive parameters of each partition.

7. The seat intelligent zoning ventilation control system according to claim 6, characterized in that: The step of obtaining the maximum temperature difference and the maximum humidity difference of each partition comprises: Define the comfortable temperature and humidity values ​​inside the vehicle; Calculate the difference between the highest temperature value and the comfortable temperature value among all scattered points in each partition, and record this difference as the maximum temperature difference; Calculate the difference between the highest humidity value and the comfortable humidity value among all scattered points in each partition, and record this difference as the maximum humidity difference.

8. The seat intelligent zoning ventilation control system according to claim 1, characterized in that: The step of obtaining the ventilation control parameters of each partition includes: The comprehensive parameters of each partition are input into the pre-trained ventilation control model, so as to obtain the ventilation control parameters of each partition, the ventilation control parameters including ventilation temperature and ventilation wind speed; The training step of the ventilation control model comprises: Collect historical comprehensive parameters and historical ventilation control parameters corresponding to the historical comprehensive parameters, and use the historical comprehensive parameters and the historical ventilation control parameters corresponding to the historical comprehensive parameters as a sample set; Number the historical ventilation control parameters, construct the feature vector P of the historical comprehensive parameters, use TensorFlow to define the input layer and output layer and build the model framework; use the feature vector P as the input layer and the historical ventilation control parameter number as the output layer; The sample set is divided into a training set and a test set according to the ratio of J% and (1-J)%, a ventilation control model is constructed, the ventilation control model is trained to obtain an initial ventilation control model, the initial ventilation control model is tested using the test set, and the initial ventilation control model that meets the preset accuracy is output, wherein the ventilation control model is one of a logistic regression model, a naive Bayes model or a support vector machine model.

9. The seat intelligent zoning ventilation control system according to claim 1, characterized in that: The step of obtaining ventilation control parameters of each partition and performing ventilation control includes: The ventilation control parameters of each zone are input into the MCU in the vehicle, and the ventilation system in the seat is regulated by the MCU.

10. A seat intelligent zoning ventilation control method according to claim 1, used to implement a seat intelligent zoning ventilation control system according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1. Preset a collection period, obtain the fitting parameters between the occupant and the seat during the collection period, and build a human body fitting model based on the fitting parameters; S2, automatically partitioning the human body fitting model, and then obtaining seat-related parameters of each partition; S3, obtaining the environmental parameters inside the vehicle and the comfort parameters of each partition, and obtaining the comprehensive parameters of each partition by comprehensively collecting the environmental parameters within a period, the comfort parameters of each partition and the seat-related parameters of each partition; S4. Input the comprehensive parameters of each partition into the ventilation control model, thereby obtaining the ventilation control parameters of each partition and executing ventilation control.

Citation Information

Cited By

  • Intelligent control method and system for automobile seat ventilation

    CN119734620A

  • Automobile seat ventilation parameter determination method and device and computer

    CN120197559A

  • Method, device and computer for determining vehicle seat ventilation parameters

    CN120197559B