A bed frame processing optimization method and system based on model analysis

Abstract

The present invention discloses a bed frame processing optimization method and system based on model analysis. Pressure and load tests are performed on the target bed frame, and the pressure and temperature control data of different parts of the bed frame are collected through a preset sensor layout scheme. According to different load conditions, pressure and temperature distribution analysis is performed, and the values ​​are mapped with color depth to generate pressure distribution maps and temperature distribution maps. The pressure and temperature control distribution maps under the expected state are generated, and the characteristic difference trends of the pressure and temperature distribution maps and the expected maps are evaluated through HOG feature extraction and difference calculation to generate the pressure and temperature control expected difference curves. The difference curves are forward and reverse predicted using a linear regression model to evaluate the distribution abnormality trend, and the abnormal area is screened in combination with the characteristic difference, and the sensor layout abnormality traceability analysis is performed based on the abnormal area. The present invention can accurately evaluate the performance of the intelligent bed frame and provide an effective basis for optimizing design and maintenance.

Description

Technical Field

[0001] The present invention relates to the field of intelligent production of bed frames, and more specifically, to a bed frame processing optimization method and system based on model analysis. Background Art

[0002] With breakthroughs in technologies like the Internet of Things, big data, and artificial intelligence, the smart home industry is entering a period of rapid development. As a core category of smart home products, smart electric beds integrate sensors, motor drives, wireless communications, and other technologies to enable functions like bed angle adjustment, health monitoring, and intelligent connectivity.

[0003] During the production of smart bed frames, the connected sensors need to undergo more testing processes and steps to ensure the reasonable layout of the sensors on the bed frame. The layout of sensors in traditional technology is often based on general needs, and it is difficult to effectively test the sensors, resulting in poor production quality of smart bed frames. In addition, the traditional bed frame sensor layout and sensor load testing process are often based on a one-to-one comparison of single values, analyzing the numerical differences of each sensor and adjusting the layout based on manual experience. The process is time-consuming and labor-intensive, making it difficult to achieve intelligent testing and analysis, and is inefficient, further hindering the production and application of smart bed frames. Summary of the Invention

[0004] The present invention overcomes the defects of the prior art and proposes a bed frame processing optimization method and system based on model analysis.

[0005] A first aspect of the present invention provides a bed frame processing optimization method based on model analysis, comprising:

[0006] S1: In the first test cycle, the target bed frame is subjected to pressure and load tests. The pressure and temperature control data of different bed frame parts are collected through the preset sensor layout plan;

[0007] S2: Under different load conditions, pressure and temperature distribution analysis is performed based on pressure data and temperature control data, and distribution values ​​are mapped with different color depths to generate pressure distribution maps and temperature distribution maps of the target bed frame;

[0008] S3: Generate a first distribution map and a second distribution map based on the expected pressure and temperature control conditions according to different load conditions. Perform regional HOG feature extraction and feature difference calculation on the pressure distribution map and the first distribution map. Evaluate the feature difference trend based on the order of each monitoring area of ​​the distribution map to generate an expected pressure difference curve. Analyze the temperature distribution map and the second distribution map based on the HOG feature to obtain an expected temperature control difference curve.

[0009] S4: Based on the expected difference curve, a linear regression model is introduced for forward and reverse prediction to evaluate the abnormal distribution trend of pressure and temperature control. The abnormal areas under different load conditions are screened out based on the characteristic difference. The abnormal sensor layout is traced and analyzed based on the abnormal areas.

[0010] In the present invention, the S1 is specifically:

[0011] In the production test of the intelligent bed frame, the first test cycle and the preset sensor layout plan are set based on the target bed frame;

[0012] In the preset sensor layout scheme, different bed frame locations include no less than a preset number of temperature sensors and pressure sensors;

[0013] Based on various load conditions, pressure data, temperature control data and test performance data of different bed frame parts are collected, and the pressure data and temperature control data are cleaned and outliers are eliminated.

[0014] In the present invention, S2 is specifically:

[0015] A blank distribution map is set based on the sensor layout scheme of the bed frame, mapped to a preset color range [0-255] based on the pressure range, and an associated mapping table is obtained;

[0016] Under a load pressure condition, the pressure sensor position and the corresponding pressure monitoring value are marked in the blank distribution map according to the pressure data. An association mapping table is introduced to convert the pressure monitoring data value into a color value and fill it in the blank distribution map to obtain the initial distribution map.

[0017] The pressure sensor position is used as the monitoring point. In the initial distribution map, the area outside the monitoring point is filled with color. The filling color value is set according to the distance between the point in the area to be filled and the nearest monitoring point, and a pressure distribution map is generated.

[0018] Perform corresponding distribution analysis based on the temperature control data and generate a temperature distribution map.

[0019] In the present invention, the S3 is specifically:

[0020] Based on a load pressure condition, setting the numerical distribution of the pressure sensor and the temperature sensor in an expected state, and performing color value mapping based on the numerical distribution to generate a first distribution graph and a second distribution graph;

[0021] Divide the distribution map into multiple monitoring areas;

[0022] In the pressure distribution map and the first distribution map, HOG feature extraction is performed for the same monitoring area, and the HOG features are vectorized to form a first feature vector and a second feature vector;

[0023] Calculate the difference between the first eigenvector and the second eigenvector based on the Manhattan distance, and obtain the feature difference;

[0024] Analyze each monitoring area to obtain multiple feature differences;

[0025] Sort the density of pressure sensors in the monitoring area by region, and map the sorting results to corresponding multiple feature differences to obtain a difference sequence;

[0026] Perform curve fitting on the difference degree sequence to obtain the pressure expected difference curve;

[0027] Based on the HOG feature, the characteristic difference of the monitoring area of ​​the temperature distribution map and the second distribution map is compared, and a temperature control expected difference curve is generated.

[0028] In the present invention, the S4 is specifically:

[0029] The pressure expected difference curve and the temperature control expected difference curve are respectively imported into the linear regression model for linear regression fitting analysis. Based on the fitting results, forward and reverse predictions are performed to evaluate the abnormal change trends of the sensor in high-density and low-density areas.

[0030] For the pressure expected difference curve, a first predicted difference value and a second predicted difference value are generated. If the difference mean between the first predicted difference value and the second predicted difference value is greater than a preset difference value, the characteristic difference degree of each monitoring area is compared with the difference mean to screen out abnormal pressure areas;

[0031] For the temperature control expected difference curve, generate a third predicted difference value and a fourth predicted difference value. If the difference mean between the third predicted difference value and the fourth predicted difference value is greater than the preset difference value, compare the characteristic difference of each monitoring area with the difference mean to screen out abnormal temperature control areas;

[0032] Based on the abnormal pressure areas and abnormal temperature control areas, the pressure sensors and temperature sensors are analyzed for abnormalities, and the preset sensor layout plan is adjusted.

[0033] In the present invention, in S2, each load condition includes a corresponding pressure distribution diagram and temperature distribution diagram.

[0034] A second aspect of the present invention further provides a bed frame processing optimization system based on model analysis, the system comprising: a memory and a processor, wherein the memory includes a bed frame processing optimization program based on model analysis, and when the bed frame processing optimization program based on model analysis is executed by the processor, the following steps are implemented:

[0035] S1: In the first test cycle, the target bed frame is subjected to pressure and load tests. The pressure and temperature control data of different bed frame parts are collected through the preset sensor layout plan;

[0036] S2: Under different load conditions, pressure and temperature distribution analysis is performed based on pressure data and temperature control data, and distribution values ​​are mapped with different color depths to generate pressure distribution maps and temperature distribution maps of the target bed frame;

[0037] S3: Generate a first distribution map and a second distribution map based on the expected pressure and temperature control conditions according to different load conditions. Perform regional HOG feature extraction and feature difference calculation on the pressure distribution map and the first distribution map. Evaluate the feature difference trend based on the order of each monitoring area of ​​the distribution map to generate an expected pressure difference curve. Analyze the temperature distribution map and the second distribution map based on the HOG feature to obtain an expected temperature control difference curve.

[0038] S4: Based on the expected difference curve, a linear regression model is introduced for forward and reverse prediction to evaluate the abnormal distribution trend of pressure and temperature control. The abnormal areas under different load conditions are screened out based on the characteristic difference. The abnormal sensor layout is traced and analyzed based on the abnormal areas.

[0039] The third aspect of the present invention also provides a computer-readable storage medium, which includes a bed frame processing optimization program based on model analysis. When the bed frame processing optimization program based on model analysis is executed by a processor, the steps of the bed frame processing optimization method based on model analysis as described in any one of the above items are implemented.

[0040] The present invention discloses a bed frame processing optimization method and system based on model analysis. Pressure and load tests are performed on the target bed frame, and the pressure and temperature control data of different parts of the bed frame are collected through a preset sensor layout scheme. According to different load conditions, pressure and temperature distribution analysis is performed, and the values ​​are mapped with color depth to generate pressure distribution maps and temperature distribution maps. The pressure and temperature control distribution maps under the expected state are generated, and the characteristic difference trends of the pressure and temperature distribution maps and the expected maps are evaluated through HOG feature extraction and difference calculation to generate the pressure and temperature control expected difference curves. The difference curves are forward and reverse predicted using a linear regression model to evaluate the distribution abnormality trend, and the abnormal area is screened in combination with the characteristic difference, and the sensor layout abnormality traceability analysis is performed based on the abnormal area. The present invention can accurately evaluate the performance of the intelligent bed frame and provide an effective basis for optimizing design and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flow chart of a bed frame processing optimization method based on model analysis of the present invention is shown;

[0042] Figure 2A block diagram of a bed frame processing optimization system based on model analysis of the present invention is shown. DETAILED DESCRIPTION

[0043] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0045] Figure 1 A flow chart of a bed frame processing optimization method based on model analysis of the present invention is shown.

[0046] like Figure 1 As shown, the first aspect of the present invention provides a bed frame processing optimization method based on model analysis, comprising:

[0047] S1: In the first test cycle, the target bed frame is subjected to pressure and load tests. The pressure and temperature control data of different bed frame parts are collected through the preset sensor layout plan;

[0048] S2: Under different load conditions, pressure and temperature distribution analysis is performed based on pressure data and temperature control data, and distribution values ​​are mapped with different color depths to generate pressure distribution maps and temperature distribution maps of the target bed frame;

[0049] S3: Generate a first distribution map and a second distribution map based on the expected pressure and temperature control conditions according to different load conditions. Perform regional HOG feature extraction and feature difference calculation on the pressure distribution map and the first distribution map. Evaluate the feature difference trend based on the order of each monitoring area of ​​the distribution map to generate an expected pressure difference curve. Analyze the temperature distribution map and the second distribution map based on the HOG feature to obtain an expected temperature control difference curve.

[0050] S4: Based on the expected difference curve, a linear regression model is introduced for forward and reverse prediction to evaluate the abnormal distribution trend of pressure and temperature control. The abnormal areas under different load conditions are screened out based on the characteristic difference. The abnormal sensor layout is traced and analyzed based on the abnormal areas.

[0051] It should be noted that the HOG feature, namely the Histogram of Oriented Gradients, is an effective method for image feature extraction. In the present invention, it is applied to image color distribution analysis and effectively reflects the distribution state analysis of the sensor.

[0052] According to an embodiment of the present invention, S1 specifically includes:

[0053] In the production test of the intelligent bed frame, the first test cycle and the preset sensor layout plan are set based on the target bed frame;

[0054] In the preset sensor layout scheme, different bed frame locations include no less than a preset number of temperature sensors and pressure sensors;

[0055] Based on various load conditions, pressure data, temperature control data and test performance data of different bed frame parts are collected, and the pressure data and temperature control data are cleaned and outliers are eliminated.

[0056] It should be noted that the target bed frame is an intelligent bed frame, which includes multiple types of sensors, such as temperature and pressure sensors, which are used for the smart bed to monitor the user's pressure and temperature control monitoring of the motor and circuit areas. At the same time, based on the monitoring data, personalized sleep monitoring and analysis, individualized adjustment of the smart bed and other functions can be realized.

[0057] The first test cycle included multiple load conditions, including those exceeding certain rated parameter ranges, to evaluate the pressure and temperature control of the target bed frame and further optimize the sensor layout analysis.

[0058] According to an embodiment of the present invention, S2 specifically includes:

[0059] A blank distribution map is set based on the sensor layout scheme of the bed frame, mapped to a preset color range [0-255] based on the pressure range, and an associated mapping table is obtained;

[0060] Under a load pressure condition, the pressure sensor position and the corresponding pressure monitoring value are marked in the blank distribution map according to the pressure data. An association mapping table is introduced to convert the pressure monitoring data value into a color value and fill it in the blank distribution map to obtain the initial distribution map.

[0061] The pressure sensor position is used as the monitoring point. In the initial distribution map, the area outside the monitoring point is filled with color. The filling color value is set according to the distance between the point in the area to be filled and the nearest monitoring point, and a pressure distribution map is generated.

[0062] Perform corresponding distribution analysis based on the temperature control data and generate a temperature distribution map.

[0063] It should be noted that when filling the numerical distribution in the area outside the monitoring point, the color value is inversely proportional to the distance between the point in the area to be filled and the nearest monitoring point. The greater the distance, the smaller the color value, and the closer the point in the area to be filled is to the monitoring point, the closer its value is to the corresponding value of the monitoring point. In the temperature distribution diagram and the pressure distribution diagram, the color settings can be changed and set according to the actual visualization requirements. The numerical settings are consistent with the associated mapping table, and the analysis and distribution diagram generation process of the two are consistent. Under each load condition, including the corresponding pressure distribution diagram and temperature distribution diagram, the load conditions include pressure load, motor operation load, etc., for different conditions and based on different working conditions, different functional tests are performed corresponding load tests.

[0064] According to an embodiment of the present invention, S3 specifically includes:

[0065] Based on a load pressure condition, setting the numerical distribution of the pressure sensor and the temperature sensor in an expected state, and performing color value mapping based on the numerical distribution to generate a first distribution graph and a second distribution graph;

[0066] Divide the distribution map into multiple monitoring areas;

[0067] In the pressure distribution map and the first distribution map, HOG feature extraction is performed for the same monitoring area, and the HOG features are vectorized to form a first feature vector and a second feature vector;

[0068] Calculate the difference between the first eigenvector and the second eigenvector based on the Manhattan distance, and obtain the feature difference;

[0069] Analyze each monitoring area to obtain multiple feature differences;

[0070] Sort the density of pressure sensors in the monitoring area by region, and map the sorting results to corresponding multiple feature differences to obtain a difference sequence;

[0071] Perform curve fitting on the difference degree sequence to obtain the pressure expected difference curve;

[0072] Based on the HOG feature, the characteristic difference of the monitoring area of ​​the temperature distribution map and the second distribution map is compared, and a temperature control expected difference curve is generated.

[0073] It should be noted that under the expected state, that is, the ideal condition of the sensor, under a certain load pressure, the corresponding sensor measurement data is consistent with the test data, and there is no numerical deviation. The first distribution diagram and the second distribution diagram correspond to the pressure and temperature distribution under the ideal state, respectively. A monitoring area corresponds to a part of the bed frame, which can be freely divided based on product requirements or functional requirements. It is used to fine-tune the rationality of the sensor layout and regional control anomalies in different parts of the bed. For example, multiple areas can be divided based on the upper and lower order, corresponding to pressure monitoring of different parts of the human body. Here, sensor density is used as the order to sequentially analyze the different functional areas and different demand areas of the smart bed, determine the difference deviation curve, and subsequently perform deviation prediction, which can further trace the abnormal sensor area. The regional sorting of sensor density is generally from small to large. Curve fitting can be based on general linear fitting.

[0074] The expected difference curve of pressure and temperature control reflects the change in sensor monitoring accuracy deviation based on the two dimensions of pressure and temperature control in the monitoring area when the bed frame is in the current sensor layout state, from simple to complex requirements.

[0075] It is worth mentioning here that the traditional bed frame sensor layout and sensor load testing process is often based on a one-to-one comparison of single values, analyzing the numerical differences of each sensor and adjusting the layout based on manual experience. The process is time-consuming and labor-intensive, and it is difficult to achieve intelligent testing and analysis. It is difficult to analyze the expected differences and predict abnormal difference trends of the target bed frame sensors from the overall layout form and overall test state. There is also a lack of regionalized sensor abnormality trend analysis process, making it difficult to capture potential abnormal sensor layout areas and difficult to trace the production process with problems, resulting in low production efficiency of the bed frame in the smart bed.

[0076] According to an embodiment of the present invention, the S4 is specifically:

[0077] The pressure expected difference curve and the temperature control expected difference curve are respectively imported into the linear regression model for linear regression fitting analysis. Based on the fitting results, forward and reverse predictions are performed to evaluate the abnormal change trends of the sensor in high-density and low-density areas.

[0078] For the pressure expected difference curve, a first predicted difference value and a second predicted difference value are generated. If the difference mean between the first predicted difference value and the second predicted difference value is greater than a preset difference value, the characteristic difference degree of each monitoring area is compared with the difference mean to screen out abnormal pressure areas;

[0079] For the temperature control expected difference curve, generate a third predicted difference value and a fourth predicted difference value. If the difference mean between the third predicted difference value and the fourth predicted difference value is greater than the preset difference value, compare the characteristic difference of each monitoring area with the difference mean to screen out abnormal temperature control areas;

[0080] Based on the abnormal pressure areas and abnormal temperature control areas, the pressure sensors and temperature sensors are analyzed for abnormalities, and the preset sensor layout plan is adjusted.

[0081] It should be noted that the two mean differences in the analysis of abnormal pressure and temperature control areas represent different data. In the forward and reverse predictions, forward prediction is used to analyze the rationality and anomalies of sensor layout in high-density areas, while reverse prediction is used to analyze the rationality and anomalies of sensor layout in low-density areas. High- and low-density areas represent areas of high and low sensor density. Forward and reverse predictions are sequential predictions, representing different directions of regional trend analysis and abnormal change prediction, respectively.

[0082] According to an embodiment of the present invention, in S2, each load condition includes a corresponding pressure distribution diagram and temperature distribution diagram.

[0083] According to an embodiment of the present invention, the sequence of monitoring areas in the distribution map and the trend of evaluation characteristics differences further include:

[0084] For each monitoring area, obtain the density value and monitoring proportion;

[0085] Generate the feature vector of the monitoring area based on the density value and the monitoring ratio;

[0086] The feature vector of each monitoring area is used as sample data, and the sample data is clustered based on the Kmeans clustering algorithm. The number of groups K is set, and K area groups are generated.

[0087] Based on the average monitoring proportion of the monitoring areas in each regional group, the regional groups are sorted, and the analysis order of the monitoring areas is determined based on the sorting results.

[0088] It should be noted that the feature vector is a two-dimensional vector, and the monitoring proportion is based on the demand corresponding to the monitoring area. Generally, the monitoring proportion of different monitoring areas is evaluated based on the functionality of different areas of the bed frame. In the present invention, sorting and predicting the monitoring areas based on density can effectively predict the differences in areas with different demand levels. For bed frames with multiple functions and more complex sensor layouts, the analysis order of sensors can be evaluated and sorted based on the two dimensions of density value and monitoring proportion. Kmeans clustering is introduced here to sort the sensor areas, which can subsequently generate a difference curve that better fits the actual bed frame layout characteristics and effectively discover abnormal layout areas. K can be set based on the number of functional areas of the smart bed frame.

[0089] According to an embodiment of the present invention, the further embodiment includes:

[0090] Set the same color map for the pressure distribution map and the temperature distribution map for analysis;

[0091] For the pressure distribution map and temperature distribution map, a monitoring area is selected, a 3×3 convolution kernel is set with the Sobel operator, and the edge strength of each window in the area is calculated. The calculated edge strength is serialized to obtain two edge detection sequences;

[0092] Determine the correlation between the two edge detection sequences by using a grey correlation method, and if there is a correlation, mark the one monitoring area as a correlation area;

[0093] Perform correlation detection on all monitoring areas, judge the rationality of sensor layout based on the correlation areas, and adjust the sensor deployment plan.

[0094] It should be noted that during the operational test, the bed pressure test is correlated with the motor's operating process, and there is a certain correlation between the pressure point and the motor's temperature control point. Effectively mining this correlation during the test process allows for efficient sensor layout analysis and adjustment. The window size is the size of the 3×3 convolution kernel. The edge strength is the gradient amplitude.

[0095] The distribution maps used in correlation analysis are the pressure and temperature distribution maps corresponding to a specific load condition. For the correlation region, the deviation consistency and correlation of the data collected by various sensors can be determined. The sensor layout in the correlation region can then be optimized to ensure that the sensors meet functional and monitoring requirements. The two edge detection sequences correspond to the pressure and temperature distribution map sequences, respectively.

[0096] Figure 2 A block diagram of a bed frame processing optimization system based on model analysis of the present invention is shown.

[0097] A second aspect of the present invention further provides a bed frame processing optimization system 2 based on model analysis, the system comprising: a memory 21 and a processor 22, wherein the memory 21 includes a bed frame processing optimization program based on model analysis, and when the bed frame processing optimization program based on model analysis is executed by the processor 22, the following steps are implemented:

[0098] S1: In the first test cycle, the target bed frame is subjected to pressure and load tests. The pressure and temperature control data of different bed frame parts are collected through the preset sensor layout plan;

[0099] S2: Under different load conditions, pressure and temperature distribution analysis is performed based on pressure data and temperature control data, and distribution values ​​are mapped with different color depths to generate pressure distribution maps and temperature distribution maps of the target bed frame;

[0100] S3: Generate a first distribution map and a second distribution map based on the expected pressure and temperature control conditions according to different load conditions. Perform regional HOG feature extraction and feature difference calculation on the pressure distribution map and the first distribution map. Evaluate the feature difference trend based on the order of each monitoring area of ​​the distribution map to generate an expected pressure difference curve. Analyze the temperature distribution map and the second distribution map based on the HOG feature to obtain an expected temperature control difference curve.

[0101] S4: Based on the expected difference curve, a linear regression model is introduced for forward and reverse prediction to evaluate the abnormal distribution trend of pressure and temperature control. The abnormal areas under different load conditions are screened out based on the characteristic difference. The abnormal sensor layout is traced and analyzed based on the abnormal areas.

[0102] It should be noted that the HOG feature, namely the Histogram of Oriented Gradients, is an effective method for image feature extraction. In the present invention, it is applied to image color distribution analysis and effectively reflects the distribution state analysis of the sensor.

[0103] According to an embodiment of the present invention, S1 specifically includes:

[0104] In the production test of the intelligent bed frame, the first test cycle and the preset sensor layout plan are set based on the target bed frame;

[0105] In the preset sensor layout scheme, different bed frame locations include no less than a preset number of temperature sensors and pressure sensors;

[0106] Based on various load conditions, pressure data, temperature control data and test performance data of different bed frame parts are collected, and the pressure data and temperature control data are cleaned and outliers are eliminated.

[0107] It should be noted that the target bed frame is an intelligent bed frame, which includes multiple types of sensors, such as temperature and pressure sensors, which are used for the smart bed to monitor the user's pressure and temperature control monitoring of the motor and circuit areas. At the same time, based on the monitoring data, personalized sleep monitoring and analysis, individualized adjustment of the smart bed and other functions can be realized.

[0108] The first test cycle included multiple load conditions, including those exceeding certain rated parameter ranges, to evaluate the pressure and temperature control of the target bed frame and further optimize the sensor layout analysis.

[0109] According to an embodiment of the present invention, S2 specifically includes:

[0110] A blank distribution map is set based on the sensor layout scheme of the bed frame, mapped to a preset color range [0-255] based on the pressure range, and an associated mapping table is obtained;

[0111] Under a load pressure condition, the pressure sensor position and the corresponding pressure monitoring value are marked in the blank distribution map according to the pressure data. An association mapping table is introduced to convert the pressure monitoring data value into a color value and fill it in the blank distribution map to obtain the initial distribution map.

[0112] The pressure sensor position is used as the monitoring point. In the initial distribution map, the area outside the monitoring point is filled with color. The filling color value is set according to the distance between the point in the area to be filled and the nearest monitoring point, and a pressure distribution map is generated.

[0113] Perform corresponding distribution analysis based on the temperature control data and generate a temperature distribution map.

[0114] It should be noted that when filling the numerical distribution in the area outside the monitoring point, the color value is inversely proportional to the distance between the point in the area to be filled and the nearest monitoring point. The greater the distance, the smaller the color value, and the closer the point in the area to be filled is to the monitoring point, the closer its value is to the corresponding value of the monitoring point. In the temperature distribution diagram and the pressure distribution diagram, the color settings can be changed and set according to the actual visualization requirements. The numerical settings are consistent with the associated mapping table, and the analysis and distribution diagram generation process of the two are consistent. Under each load condition, including the corresponding pressure distribution diagram and temperature distribution diagram, the load conditions include pressure load, motor operation load, etc., for different conditions and based on different working conditions, different functional tests are performed corresponding load tests.

[0115] According to an embodiment of the present invention, S3 specifically includes:

[0116] Based on a load pressure condition, setting the numerical distribution of the pressure sensor and the temperature sensor in an expected state, and performing color value mapping based on the numerical distribution to generate a first distribution graph and a second distribution graph;

[0117] Divide the distribution map into multiple monitoring areas;

[0118] In the pressure distribution map and the first distribution map, HOG feature extraction is performed for the same monitoring area, and the HOG features are vectorized to form a first feature vector and a second feature vector;

[0119] Calculate the difference between the first eigenvector and the second eigenvector based on the Manhattan distance, and obtain the feature difference;

[0120] Analyze each monitoring area to obtain multiple feature differences;

[0121] Sort the density of pressure sensors in the monitoring area by region, and map the sorting results to corresponding multiple feature differences to obtain a difference sequence;

[0122] Perform curve fitting on the difference degree sequence to obtain the pressure expected difference curve;

[0123] Based on the HOG feature, the characteristic difference of the monitoring area of ​​the temperature distribution map and the second distribution map is compared, and a temperature control expected difference curve is generated.

[0124] It should be noted that under the expected state, that is, the ideal condition of the sensor, under a certain load pressure, the corresponding sensor measurement data is consistent with the test data, and there is no numerical deviation. The first distribution diagram and the second distribution diagram correspond to the pressure and temperature distribution under the ideal state, respectively. A monitoring area corresponds to a part of the bed frame, which can be freely divided based on product requirements or functional requirements. It is used to fine-tune the rationality of the sensor layout and regional control anomalies in different parts of the bed. For example, multiple areas can be divided based on the upper and lower order, corresponding to pressure monitoring of different parts of the human body. Here, sensor density is used as the order to sequentially analyze the different functional areas and different demand areas of the smart bed, determine the difference deviation curve, and subsequently perform deviation prediction, which can further trace the abnormal sensor area. The regional sorting of sensor density is generally from small to large. Curve fitting can be based on general linear fitting.

[0125] The expected difference curve of pressure and temperature control reflects the change in sensor monitoring accuracy deviation based on the two dimensions of pressure and temperature control in the monitoring area when the bed frame is in the current sensor layout state, from simple to complex requirements.

[0126] It is worth mentioning here that the traditional bed frame sensor layout and sensor load testing process is often based on a one-to-one comparison of single values, analyzing the numerical differences of each sensor and adjusting the layout based on manual experience. The process is time-consuming and labor-intensive, and it is difficult to achieve intelligent testing and analysis. It is difficult to analyze the expected differences and predict abnormal difference trends of the target bed frame sensors from the overall layout form and overall test state. There is also a lack of regionalized sensor abnormality trend analysis process, making it difficult to capture potential abnormal sensor layout areas and difficult to trace the production process with problems, resulting in low production efficiency of the bed frame in the smart bed.

[0127] According to an embodiment of the present invention, the S4 is specifically:

[0128] The pressure expected difference curve and the temperature control expected difference curve are respectively imported into the linear regression model for linear regression fitting analysis. Based on the fitting results, forward and reverse predictions are performed to evaluate the abnormal change trends of the sensor in high-density and low-density areas.

[0129] For the pressure expected difference curve, a first predicted difference value and a second predicted difference value are generated. If the difference mean between the first predicted difference value and the second predicted difference value is greater than a preset difference value, the characteristic difference degree of each monitoring area is compared with the difference mean to screen out abnormal pressure areas;

[0130] For the temperature control expected difference curve, generate a third predicted difference value and a fourth predicted difference value. If the difference mean between the third predicted difference value and the fourth predicted difference value is greater than the preset difference value, compare the characteristic difference of each monitoring area with the difference mean to screen out abnormal temperature control areas;

[0131] Based on the abnormal pressure areas and abnormal temperature control areas, the pressure sensors and temperature sensors are analyzed for abnormalities, and the preset sensor layout plan is adjusted.

[0132] It should be noted that the two mean differences in the analysis of abnormal pressure and temperature control areas represent different data. In the forward and reverse predictions, forward prediction is used to analyze the rationality and anomalies of sensor layout in high-density areas, while reverse prediction is used to analyze the rationality and anomalies of sensor layout in low-density areas. High- and low-density areas represent areas of high and low sensor density. Forward and reverse predictions are sequential predictions, representing different directions of regional trend analysis and abnormal change prediction, respectively.

[0133] According to an embodiment of the present invention, in S2, each load condition includes a corresponding pressure distribution diagram and temperature distribution diagram.

[0134] The third aspect of the present invention also provides a computer-readable storage medium, which includes a bed frame processing optimization program based on model analysis. When the bed frame processing optimization program based on model analysis is executed by a processor, the steps of the bed frame processing optimization method based on model analysis as described in any one of the above items are implemented.

[0135] The present invention discloses a bed frame processing optimization method and system based on model analysis. Pressure and load tests are performed on the target bed frame, and the pressure and temperature control data of different parts of the bed frame are collected through a preset sensor layout scheme. According to different load conditions, pressure and temperature distribution analysis is performed, and the values ​​are mapped with color depth to generate pressure distribution maps and temperature distribution maps. The pressure and temperature control distribution maps under the expected state are generated, and the characteristic difference trends of the pressure and temperature distribution maps and the expected maps are evaluated through HOG feature extraction and difference calculation to generate the pressure and temperature control expected difference curves. The difference curves are forward and reverse predicted using a linear regression model to evaluate the distribution abnormality trend, and the abnormal area is screened in combination with the characteristic difference, and the sensor layout abnormality traceability analysis is performed based on the abnormal area. The present invention can accurately evaluate the performance of the intelligent bed frame and provide an effective basis for optimizing design and maintenance.

[0136] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0137] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0138] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0139] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0140] Alternatively, if the integrated units described above are implemented as software modules and sold or used as standalone products, they can also be stored on a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product, stored on a storage medium, includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute all or part of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as removable storage devices, ROM, RAM, magnetic disks, or optical disks.

[0141] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. A bed frame processing optimization method based on model analysis, characterized in that: include: S1: In the first test cycle, the target bed frame is subjected to pressure and load tests. The pressure and temperature control data of different bed frame parts are collected through the preset sensor layout plan; S2: Under different load conditions, pressure and temperature distribution analysis is performed based on pressure data and temperature control data, and distribution values ​​are mapped with different color depths to generate pressure distribution maps and temperature distribution maps of the target bed frame; S3: Generate a first distribution map and a second distribution map based on the expected pressure and temperature control conditions according to different load conditions. Perform regional HOG feature extraction and feature difference calculation on the pressure distribution map and the first distribution map. Evaluate the feature difference trend based on the order of each monitoring area of ​​the distribution map to generate an expected pressure difference curve. Analyze the temperature distribution map and the second distribution map based on the HOG feature to obtain an expected temperature control difference curve. S4: Based on the expected difference curve, a linear regression model is introduced for forward and reverse prediction to evaluate the abnormal distribution trend of pressure and temperature control. The abnormal areas under different load conditions are screened out based on the characteristic difference. The abnormal sensor layout is traced and analyzed based on the abnormal areas.

2. A bed frame processing optimization method based on model analysis according to claim 1, characterized in that: The S1 is specifically: In the production test of the intelligent bed frame, the first test cycle and the preset sensor layout plan are set based on the target bed frame; In the preset sensor layout scheme, different bed frame locations include no less than a preset number of temperature sensors and pressure sensors; Based on various load conditions, pressure data, temperature control data and test performance data of different bed frame parts are collected, and the pressure data and temperature control data are cleaned and outliers are eliminated.

3. The bed frame processing optimization method based on model analysis according to claim 1 is characterized in that: The S2 is specifically: A blank distribution map is set based on the sensor layout scheme of the bed frame, mapped to a preset color range [0-255] based on the pressure range, and an associated mapping table is obtained; Under a load pressure condition, the pressure sensor position and the corresponding pressure monitoring value are marked in the blank distribution map according to the pressure data. An association mapping table is introduced to convert the pressure monitoring data value into a color value and fill it in the blank distribution map to obtain the initial distribution map. The pressure sensor position is used as the monitoring point. In the initial distribution map, the area outside the monitoring point is filled with color. The filling color value is set according to the distance between the point in the area to be filled and the nearest monitoring point, and a pressure distribution map is generated. Perform corresponding distribution analysis based on the temperature control data and generate a temperature distribution map.

4. A bed frame processing optimization method based on model analysis according to claim 3, characterized in that: The S3 is specifically: Based on a load pressure condition, setting the numerical distribution of the pressure sensor and the temperature sensor in an expected state, and performing color value mapping based on the numerical distribution to generate a first distribution graph and a second distribution graph; Divide the distribution map into multiple monitoring areas; In the pressure distribution map and the first distribution map, HOG feature extraction is performed for the same monitoring area, and the HOG features are vectorized to form a first feature vector and a second feature vector; Calculate the difference between the first eigenvector and the second eigenvector based on the Manhattan distance, and obtain the feature difference; Analyze each monitoring area to obtain multiple feature differences; Sort the density of pressure sensors in the monitoring area by region, and map the sorting results to corresponding multiple feature differences to obtain a difference sequence; Perform curve fitting on the difference degree sequence to obtain the pressure expected difference curve; Based on the HOG feature, the characteristic difference of the monitoring area of ​​the temperature distribution map and the second distribution map is compared, and a temperature control expected difference curve is generated.

5. The bed frame processing optimization method based on model analysis according to claim 1 is characterized in that: The S4 is specifically: The pressure expected difference curve and the temperature control expected difference curve are respectively imported into the linear regression model for linear regression fitting analysis. Based on the fitting results, forward and reverse predictions are performed to evaluate the abnormal change trends of the sensor in high-density and low-density areas. For the pressure expected difference curve, a first predicted difference value and a second predicted difference value are generated. If the difference mean between the first predicted difference value and the second predicted difference value is greater than a preset difference value, the characteristic difference degree of each monitoring area is compared with the difference mean to screen out abnormal pressure areas; For the temperature control expected difference curve, generate a third predicted difference value and a fourth predicted difference value. If the difference mean between the third predicted difference value and the fourth predicted difference value is greater than the preset difference value, compare the characteristic difference of each monitoring area with the difference mean to screen out abnormal temperature control areas; Based on the abnormal pressure areas and abnormal temperature control areas, the pressure sensors and temperature sensors are analyzed for abnormalities, and the preset sensor layout plan is adjusted.

6. The bed frame processing optimization method based on model analysis according to claim 1, characterized in that: In S2, each load condition includes a corresponding pressure distribution diagram and temperature distribution diagram.

7. A bed frame processing optimization system based on model analysis, characterized in that: The system includes: a memory and a processor. The memory includes a bed frame processing optimization program based on model analysis. When the bed frame processing optimization program based on model analysis is executed by the processor, the following steps are implemented: S1: In the first test cycle, the target bed frame is subjected to pressure and load tests. The pressure and temperature control data of different bed frame parts are collected through the preset sensor layout plan; S2: Under different load conditions, pressure and temperature distribution analysis is performed based on pressure data and temperature control data, and distribution values ​​are mapped with different color depths to generate pressure distribution maps and temperature distribution maps of the target bed frame; S3: Generate a first distribution map and a second distribution map based on the expected pressure and temperature control conditions according to different load conditions. Perform regional HOG feature extraction and feature difference calculation on the pressure distribution map and the first distribution map. Evaluate the feature difference trend based on the order of each monitoring area of ​​the distribution map to generate an expected pressure difference curve. Analyze the temperature distribution map and the second distribution map based on the HOG feature to obtain an expected temperature control difference curve. S4: Based on the expected difference curve, a linear regression model is introduced for forward and reverse prediction to evaluate the abnormal distribution trend of pressure and temperature control. The abnormal areas under different load conditions are screened out based on the characteristic difference. The abnormal sensor layout is traced and analyzed based on the abnormal areas.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a bed frame processing optimization program based on model analysis. When the bed frame processing optimization program based on model analysis is executed by a processor, the steps of the bed frame processing optimization method based on model analysis as described in any one of claims 1 to 6 are implemented.

Images