A machine learning oriented equipment component health management method and system

Abstract

The application discloses a kind of equipment component health management methods and systems for machine learning, it is related to the equipment component health management technical field for machine learning, the application includes historical maintenance data acquisition, maintenance component screening, reference roller data analysis, reference screen data analysis, robot operation monitoring, component state data determination analysis and fault maintenance display, the operation of the robot wheel and screen is monitored, and the robot wheel is not agile and the robot screen is not sensitive Warning, solve the deficiency in the prior art, realize the intelligent and automated monitoring and management of robot, ensure the timeliness of robot wheel and screen maintenance, improve the operation efficiency of robot, effectively reduce the workload of medical staff, also improve the experience of patient in hospital.

Description

Technical Field

[0001] This invention relates to the field of machine health technology, and more specifically to a method and system for health management of equipment components for machine learning. Background Technology

[0002] Robots are indispensable equipment in hospitals and have a wide range of applications. They can provide services such as patient navigation and medical treatment, thereby improving the patient's medical experience. The agility of the robot's wheels is a prerequisite for providing navigation, and the sensitivity of the screen can improve the speed of patients' information retrieval. Therefore, in order to ensure the operation of the robot, it is necessary to manage the agility of the robot's wheels and the sensitivity of the screen to ensure the normal operation of the robot.

[0003] Current technology mainly relies on maintenance personnel to conduct regular inspections or repair malfunctioning robots when they occur. However, neither regular maintenance nor repairs on the spot can guarantee the timeliness of robot maintenance, which affects the operational effectiveness of robots in hospitals, increases the workload of medical staff to some extent, and also reduces the patient's experience when seeking medical treatment in the hospital. Summary of the Invention

[0004] In view of the above-mentioned technical shortcomings, the purpose of this invention is to provide a method and system for health management of equipment components based on machine learning.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: In the first aspect, the present invention provides a machine learning-oriented equipment component health management method, including: Step 1, historical maintenance information acquisition: acquiring maintenance information of the corresponding robot during each maintenance in the hospital;

[0006] Step 2, Repair Parts Screening: Based on the repair information of the corresponding robot during each repair, screen the corresponding robots for each repair to obtain the robots corresponding to each roller repair and the robots corresponding to each screen repair.

[0007] Step 3: Reference Roller Data Analysis: Extract the accuracy of the roller turning speed and turning angle of the robot corresponding to each roller repair. Then, calculate the average roller turning speed of the robot corresponding to each roller repair by averaging the roller turning speed of each roller repair, and use it as the reference roller turning speed of the robot. Similarly, obtain the accuracy of the robot's reference turning angle.

[0008] Step 4: Reference screen data analysis: Extract the screen response time and screen response strength of the robot corresponding to each screen repair, and then calculate the average screen response time of the robot corresponding to each screen repair by averaging the average screen response time of the robot, which is used as the reference screen response time of the robot. Similarly, the reference screen response strength of the robot is obtained.

[0009] Step 5: Robot operation monitoring: Monitor each robot in the hospital to obtain the accuracy of the corresponding roller turning speed, turning angle, screen response time, and screen response click force for each robot;

[0010] Step 6: Component status data judgment and analysis: Based on the accuracy of the turning speed and turning angle of each robot's corresponding roller, the screen response time and screen response force, analyze the roller agility and screen sensitivity of each robot, and then judge the roller agility and screen sensitivity of each robot, and select each target roller repair robot and each target screen repair robot.

[0011] Step 7, Fault Repair Display: Displays the repair robots for each target roller robot and each target screen robot.

[0012] Preferably, the maintenance information includes the maintenance parts, roller steering speed, steering angle accuracy, screen response time, and screen response strength.

[0013] Preferably, the process of screening the robots corresponding to each maintenance operation involves the following steps:

[0014] Obtain the repair parts of the robot corresponding to each repair. If the repair part of the robot corresponding to a certain repair is a roller, then the robot corresponding to that repair is taken as the robot corresponding to roller repair. In this way, the robot corresponding to each roller repair is obtained. If the repair part of the robot corresponding to a certain repair is a screen, then the robot corresponding to that repair is taken as the robot corresponding to screen repair. In this way, the robot corresponding to each screen repair is obtained.

[0015] Preferably, the monitoring of each robot in the hospital is carried out in the following specific process:

[0016] Several cameras are deployed in the hospital, and the operation videos of each robot in the hospital are collected through each camera.

[0017] The actual turning angle of each robot during each turn is obtained from the operation video of each robot, and the preset turning angle of each robot during each turn is also obtained. Then, the accuracy of the turning angle of each robot is calculated by the turning angle calculation formula.

[0018] Speed ​​sensors are installed in the rollers of each robot. The turning speed of each robot during each turn is collected by the speed sensors in the rollers of each robot. The average turning speed of each robot is obtained by averaging the values, and this average turning speed is used as the turning speed of each robot.

[0019] The screen response time of each robot is obtained from the screen usage of each robot in each use, and then the screen response time of each robot is calculated by averaging.

[0020] Force sensors are installed on the screens of each robot. The force sensors on the screens of each robot collect the reaction force of each robot when it is clicked. The average reaction force of each robot is calculated by averaging the values, and this average reaction force is used as the reaction force of each robot.

[0021] Preferably, the calculation includes the accuracy of obtaining the turning angle corresponding to each robot, and the specific calculation process is as follows:

[0022] By analyzing the formula The analysis yielded the accuracy of the steering angles for each robot. Where θ if θ i ′ f Wherein, Δθ represents the preset turning angle and actual turning angle corresponding to the i-th robot during the f turns, σ is the set permissible turning angle difference, σ is the set turning angle compensation factor, i represents the number corresponding to each robot, i = 1, 2, ..., n, and f represents the number corresponding to each turn, f = 1, 2, ..., m.

[0023] Preferably, the specific calculation process for analyzing the agility of the rollers corresponding to each robot is as follows:

[0024] Through calculation formula The analysis yielded the roller agility of each robot, where β i Let r1 and r2 represent the agility of the i-th robot's roller, where r1 and r2 are the set weighting factors for speed and agility, respectively, and v represents the robot's reference roller turning speed. v represents the accuracy of the robot's reference steering angle. i This represents the turning speed of the roller corresponding to the i-th robot. This represents the accuracy of the roller steering angle corresponding to the i-th robot.

[0025] Preferably, the analysis of the screen sensitivity of each robot yields the following specific calculation results:

[0026] Through calculation formula The analysis yielded the screen sensitivity for each robot, among which... This represents the screen sensitivity corresponding to the i-th robot, t1 and t2 are the set force and sensitivity weighting factors, respectively, T represents the reference screen reaction time, and F represents the robot's reference screen reaction force. i F represents the screen response time for the i-th robot. i This indicates that the i-th robot corresponds to the screen's reaction force.

[0027] Preferably, the process includes judging the agility of the rollers and the sensitivity of the screen for each robot, with the specific judgment results as follows:

[0028] The agility of the rollers corresponding to each robot is compared with the set threshold for roller agility. If the roller agility of a robot is greater than or equal to the threshold, the roller corresponding to that robot is determined to be agile; otherwise, the roller corresponding to that robot is determined to be insensitive and the robot is recorded as the target roller repair robot. The roller agility of each robot is judged in this way, and the target roller repair robots are obtained.

[0029] The sensitivity of the screen corresponding to each robot is compared with the set screen sensitivity. If the sensitivity of the screen corresponding to a robot is greater than or equal to the screen sensitivity threshold, the screen corresponding to that robot is determined to be sensitive. Otherwise, the screen corresponding to that robot is determined to be insensitive, and the robot is recorded as the target screen repair robot. The screen sensitivity of each robot is judged in this way, and the target screen repair robots are obtained.

[0030] In a second aspect, the present invention provides a machine learning-oriented equipment component health management system, comprising: a historical maintenance information acquisition module for acquiring maintenance information of the corresponding robot during each maintenance in a hospital;

[0031] The repair parts filtering module is used to filter the corresponding robots for each repair based on the repair information of the corresponding robots during each repair, so as to obtain the robots corresponding to each roller repair and the robots corresponding to each screen repair.

[0032] The reference roller data analysis module is used to extract the accuracy of the roller turning speed and turning angle of the robot corresponding to each roller repair. Then, the average roller turning speed of the robot corresponding to each roller repair is calculated by averaging the roller turning speed of the robot corresponding to the roller repair, which is used as the reference roller turning speed of the robot. Similarly, the accuracy of the reference turning angle of the robot is obtained.

[0033] The reference screen data analysis module is used to extract the screen response time and screen response strength of the robot corresponding to each screen repair. Then, the average screen response time of the robot corresponding to each screen repair is calculated by averaging the average screen response time of the robot, which is used as the reference screen response time of the robot. Similarly, the reference screen response strength of the robot is obtained.

[0034] The robot operation monitoring module is used to monitor each robot in the hospital and obtain the accuracy of the corresponding roller turning speed, turning angle, screen response time and screen response click force for each robot.

[0035] The component status data judgment and analysis module is used to analyze the agility of the corresponding rollers and the sensitivity of the screen of each robot based on the accuracy of the turning speed and turning angle of the corresponding rollers, the screen response time and the screen response force. Then, it judges the agility of the corresponding rollers and the sensitivity of the screen of each robot, and selects each target roller repair robot and each target screen repair robot.

[0036] The fault display terminal is used to display the faults of each target roller robot and each target screen repair robot.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention provides a machine learning-oriented equipment component health management method and system, which monitors the operation of the robot's wheels and screen, and provides early warnings for robots with unresponsive wheels and unresponsive screens. This solves the shortcomings of the prior art, realizes intelligent and automated monitoring and management of robots, ensures the timeliness of robot wheel and screen maintenance, improves robot operating efficiency, effectively reduces the workload of medical staff, and also improves the patient experience when seeking medical treatment in the hospital. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention.

[0040] Figure 2 This is a schematic diagram of the system module structure connection of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Please see Figure 1 As shown, the present invention provides a machine learning-oriented equipment component health management method, including: Step 1, historical maintenance information acquisition: acquiring maintenance information of the corresponding robot during each maintenance in the hospital;

[0043] The repair information mentioned above includes the repaired parts, the scroll wheel speed, the accuracy of the turning angle, the screen response time, and the screen response strength.

[0044] It should be noted that the screen response force refers to the amount of force required for the screen to react and change when the screen is clicked.

[0045] It should also be noted that the repairable parts include the robot's wheels, screen, and shell.

[0046] The above describes obtaining maintenance information for the corresponding robot during each maintenance session in the hospital from the robot management center.

[0047] Step 2, Repair Parts Screening: Based on the repair information of the corresponding robot during each repair, screen the corresponding robots for each repair to obtain the robots corresponding to each roller repair and the robots corresponding to each screen repair.

[0048] As an optional implementation method, the robots corresponding to each maintenance are screened. The specific screening steps are as follows:

[0049] Obtain the repair parts of the robot corresponding to each repair. If the repair part of the robot corresponding to a certain repair is a roller, then the robot corresponding to that repair is taken as the robot corresponding to roller repair. In this way, the robot corresponding to each roller repair is obtained. If the repair part of the robot corresponding to a certain repair is a screen, then the robot corresponding to that repair is taken as the robot corresponding to screen repair. In this way, the robot corresponding to each screen repair is obtained.

[0050] Step 3: Reference Roller Data Analysis: Extract the accuracy of the roller turning speed and turning angle of the robot corresponding to each roller repair. Then, calculate the average roller turning speed of the robot corresponding to each roller repair by averaging the roller turning speed of each roller repair, and use it as the reference roller turning speed of the robot. Similarly, obtain the accuracy of the robot's reference turning angle.

[0051] Step 4: Reference screen data analysis: Extract the screen response time and screen response strength of the robot corresponding to each screen repair, and then calculate the average screen response time of the robot corresponding to each screen repair by averaging the average screen response time of the robot, which is used as the reference screen response time of the robot. Similarly, the reference screen response strength of the robot is obtained.

[0052] Step 5: Robot operation monitoring: Monitor each robot in the hospital to obtain the accuracy of the corresponding roller turning speed, turning angle, screen response time, and screen response click force for each robot;

[0053] As an optional implementation method, the robots in the hospital are monitored. The specific monitoring process is as follows:

[0054] Several cameras are deployed in the hospital, and the operation videos of each robot in the hospital are collected through each camera.

[0055] The actual turning angle of each robot during each turn is obtained from the operation video of each robot, and the preset turning angle of each robot during each turn is also obtained. Then, the accuracy of the turning angle of each robot is calculated by the turning angle calculation formula.

[0056] Speed ​​sensors are installed in the rollers of each robot. The turning speed of each robot during each turn is collected by the speed sensors in the rollers of each robot. The average turning speed of each robot is obtained by averaging the values, and this average turning speed is used as the turning speed of each robot.

[0057] The screen response time of each robot is obtained from the screen usage of each robot in each use, and then the screen response time of each robot is calculated by averaging.

[0058] Force sensors are installed on the screens of each robot. The force sensors on the screens of each robot collect the reaction force of each robot when it is clicked. The average reaction force of each robot is calculated by averaging the values, and this average reaction force is used as the reaction force of each robot.

[0059] Step 6: Component status data judgment and analysis: Based on the accuracy of the turning speed and turning angle of each robot's corresponding roller, the screen response time and screen response force, analyze the roller agility and screen sensitivity of each robot, and then judge the roller agility and screen sensitivity of each robot, and select each target roller repair robot and each target screen repair robot.

[0060] As an optional implementation method, this includes calculating the accuracy of the steering angle corresponding to each robot. The specific calculation process is as follows:

[0061] By analyzing the formula The analysis yielded the accuracy of the steering angles for each robot. Where θ if θ i ′ f Wherein, Δθ represents the preset turning angle and actual turning angle corresponding to the i-th robot during the f turns, σ is the set permissible turning angle difference, σ is the set turning angle compensation factor, i represents the number corresponding to each robot, i = 1, 2, ..., n, and f represents the number corresponding to each turn, f = 1, 2, ..., m.

[0062] As an optional implementation method, the agility of the rollers for each robot is analyzed, and the specific calculation process is as follows:

[0063] Through calculation formula The analysis yielded the roller agility of each robot, where β i Let r1 and r2 represent the agility of the i-th robot's roller, where r1 and r2 are the set weighting factors for speed and agility, respectively, and v represents the robot's reference roller turning speed. v represents the accuracy of the robot's reference steering angle. i This represents the turning speed of the roller corresponding to the i-th robot. This represents the accuracy of the roller steering angle corresponding to the i-th robot.

[0064] As an optional implementation method, the screen sensitivity of each robot was analyzed, and the specific calculation results are as follows:

[0065] Through calculation formula The analysis yielded the screen sensitivity for each robot, among which... This represents the screen sensitivity corresponding to the i-th robot, t1 and t2 are the set weighting factors for force and sensitivity, respectively, T represents the reference screen reaction time, and F represents the robot's reference screen reaction force. i F represents the screen response time for the i-th robot. i This indicates that the i-th robot corresponds to the screen's reaction force.

[0066] As an optional implementation, this includes judging the agility of the rollers and the sensitivity of the screen for each robot, with the specific judgment results as follows:

[0067] The agility of the rollers corresponding to each robot is compared with the set threshold for roller agility. If the roller agility of a robot is greater than or equal to the threshold, the roller corresponding to that robot is determined to be agile; otherwise, the roller corresponding to that robot is determined to be insensitive and the robot is recorded as the target roller repair robot. The roller agility of each robot is judged in this way, and the target roller repair robots are obtained.

[0068] The sensitivity of the screen corresponding to each robot is compared with the set screen sensitivity. If the sensitivity of the screen corresponding to a robot is greater than or equal to the screen sensitivity threshold, the screen corresponding to that robot is determined to be sensitive. Otherwise, the screen corresponding to that robot is determined to be insensitive, and the robot is recorded as the target screen repair robot. The screen sensitivity of each robot is judged in this way, and the target screen repair robots are obtained.

[0069] Step 7, Fault Repair Display: Displays the repair robots for each target roller robot and each target screen robot.

[0070] Please see Figure 2 As shown, a machine learning-oriented equipment component health management system includes a historical maintenance information acquisition module, a maintenance component screening module, a reference roller data analysis module, a reference screen data analysis module, a robot operation monitoring module, a component status data judgment and analysis module, and a fault display terminal.

[0071] The historical maintenance information acquisition module is connected to the maintenance component screening module and the component status data judgment and analysis module, respectively. The maintenance component screening module is connected to the reference roller data analysis module and the reference screen data analysis module, respectively. The robot operation monitoring module is connected to the component status data judgment and analysis module and the fault maintenance display module, respectively.

[0072] The historical maintenance information acquisition module is used to acquire maintenance information of the corresponding robot during each maintenance in the hospital;

[0073] The repair parts filtering module is used to filter the corresponding robots for each repair based on the repair information of the corresponding robots during each repair, so as to obtain the robots corresponding to each roller repair and the robots corresponding to each screen repair.

[0074] The reference roller data analysis module is used to extract the accuracy of the roller turning speed and turning angle of the robot corresponding to each roller repair. Then, the average roller turning speed of the robot corresponding to each roller repair is calculated by averaging the roller turning speed of the robot corresponding to the roller repair, which is used as the reference roller turning speed of the robot. Similarly, the accuracy of the reference turning angle of the robot is obtained.

[0075] The reference screen data analysis module is used to extract the screen response time and screen response strength of the robot corresponding to each screen repair. Then, the average screen response time of the robot corresponding to each screen repair is calculated by averaging the average screen response time of the robot, which is used as the reference screen response time of the robot. Similarly, the reference screen response strength of the robot is obtained.

[0076] The robot operation monitoring module is used to monitor each robot in the hospital and obtain the accuracy of the corresponding roller turning speed, turning angle, screen response time and screen response click force for each robot.

[0077] The component status data judgment and analysis module is used to analyze the agility of the corresponding rollers and the sensitivity of the screen of each robot based on the accuracy of the turning speed and turning angle of the corresponding rollers, the screen response time and the screen response force. Then, it judges the agility of the corresponding rollers and the sensitivity of the screen of each robot, and selects each target roller repair robot and each target screen repair robot.

[0078] The fault display terminal is used to display the faults of each target roller robot and each target screen repair robot.

[0079] This invention addresses the shortcomings of existing technologies by monitoring the operation of the robot's wheels and screen, and issuing warnings for robots with unresponsive wheels and screens. This enables intelligent and automated monitoring and management of the robot, ensuring timely repair of the robot's wheels and screen, improving the robot's operating efficiency, effectively reducing the workload of medical staff, and enhancing the patient experience in the hospital.

[0080] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A machine learning-oriented method for equipment component health management, characterized in that, include: Step 1: Obtain historical maintenance information: Obtain maintenance information for the corresponding robot during each maintenance in the hospital; Step 2, Repair Parts Screening: Based on the repair information of the corresponding robot during each repair, screen the corresponding robots for each repair to obtain the robots corresponding to each roller repair and the robots corresponding to each screen repair. Step 3: Reference Roller Data Analysis: Extract the accuracy of the roller turning speed and turning angle of the robot corresponding to each roller repair. Then, calculate the average roller turning speed of the robot corresponding to each roller repair by averaging the roller turning speed of each roller repair, and use it as the reference roller turning speed of the robot. Similarly, obtain the accuracy of the robot's reference turning angle. Step 4: Reference screen data analysis: Extract the screen response time and screen response strength of the robot corresponding to each screen repair, and then calculate the average screen response time of the robot corresponding to each screen repair by averaging the average screen response time of the robot, which is used as the reference screen response time of the robot. Similarly, the reference screen response strength of the robot is obtained. Step 5: Robot operation monitoring: Monitor each robot in the hospital to obtain the accuracy of the corresponding roller turning speed, turning angle, screen response time, and screen response click force for each robot; Step 6: Component status data judgment and analysis: Based on the accuracy of the turning speed and turning angle of the corresponding rollers of each robot, the screen response time and screen response force are analyzed to determine the roller agility and screen sensitivity of each robot. Then, the roller agility and screen sensitivity of each robot are judged, and the target roller repair robots and target screen repair robots are selected. Step 7, Fault Repair Display: Displays the repair robots for each target roller robot and each target screen; The accuracy of the steering angle is calculated as follows: By analyzing the formula The analysis yields the accuracy ∂i for the steering angle of each robot, where θ if θ i ′ f Where Δ represents the preset turning angle and the actual turning angle of the i-th robot during the f-th turning operation. θ For the set permissible steering angle difference, σ Here, is the set steering angle compensation factor, and 'i' represents the corresponding number of each robot. i = 1,2...... n f represents the number corresponding to each turn. f = 1,2...... m ; The analysis of the roller agility of each robot is specifically calculated as follows: Through calculation formula The analysis yielded the roller agility of each robot, among which... β i represents the agility of the roller corresponding to the i-th robot, r1 and r2 are the set speed and agility weight factors respectively, v represents the reference roller turning speed of the robot, ∂ represents the reference turning angle accuracy of the robot, vi represents the roller turning speed corresponding to the i-th robot, and ∂i represents the roller turning angle accuracy corresponding to the i-th robot. The analysis of the screen sensitivity of each robot yielded the following specific calculation results: Through calculation formula The analysis yielded the screen sensitivity for each robot, among which... φ i represents the screen sensitivity corresponding to the i-th robot, t1 and t2 are the set weighting factors of force and sensitivity, respectively, T represents the reference screen reaction time, and F represents the robot's reference screen reaction force. i F represents the screen response time for the i-th robot. i This indicates that the i-th robot corresponds to the screen's reaction force; The agility of the rollers and the sensitivity of the screen for each robot are judged, and the specific judgment results are as follows: The agility of the rollers corresponding to each robot is compared with the set threshold for roller agility. If the roller agility of a robot is greater than or equal to the threshold, the roller corresponding to that robot is determined to be agile; otherwise, the roller corresponding to that robot is determined to be insensitive and the robot is recorded as the target roller repair robot. The roller agility of each robot is judged in this way, and the target roller repair robots are obtained. The sensitivity of the screen corresponding to each robot is compared with the set screen sensitivity. If the sensitivity of the screen corresponding to a robot is greater than or equal to the screen sensitivity threshold, the screen corresponding to that robot is determined to be sensitive. Otherwise, the screen corresponding to that robot is determined to be insensitive, and the robot is recorded as the target screen repair robot. The screen sensitivity of each robot is judged in this way, and the target screen repair robots are obtained.

2. The equipment component health management method based on machine learning as described in claim 1, characterized in that, The maintenance information includes the maintenance parts, roller steering speed, steering angle accuracy, screen response time, and screen response strength.

3. The equipment component health management method based on machine learning as described in claim 2, characterized in that, The specific steps for selecting the robot corresponding to each maintenance session are as follows: Obtain the repair parts of the robot corresponding to each repair. If the repair part of the robot corresponding to a certain repair is a roller, then the robot corresponding to that repair is taken as the robot corresponding to roller repair. In this way, the robot corresponding to each roller repair is obtained. If the repair part of the robot corresponding to a certain repair is a screen, then the robot corresponding to that repair is taken as the robot corresponding to screen repair. In this way, the robot corresponding to each screen repair is obtained.

4. The equipment component health management method based on machine learning as described in claim 1, characterized in that, The monitoring process for the various robots in the hospital is as follows: Several cameras are deployed in the hospital, and the operation videos of each robot in the hospital are collected through each camera. The actual turning angle of each robot during each turn is obtained from the operation video of each robot, and the preset turning angle of each robot during each turn is also obtained. Then, the accuracy of the turning angle of each robot is calculated by the turning angle calculation formula. Speed ​​sensors are installed in the rollers of each robot. The turning speed of each robot during each turn is collected by the speed sensors in the rollers of each robot. The average turning speed of each robot is obtained by averaging the values, and this average turning speed is used as the turning speed of each robot. The screen response time of each robot is obtained from the screen usage of each robot in each use, and then the screen response time of each robot is calculated by averaging. Force sensors are installed on the screens of each robot. The force sensors on the screens of each robot collect the reaction force of each robot when it is clicked. The average reaction force of each robot is calculated by averaging the values, and this average reaction force is used as the reaction force of each robot.

5. A machine learning-oriented equipment component health management system that performs the method as described in claim 1, characterized in that, include: The historical maintenance information acquisition module is used to acquire maintenance information of the corresponding robot during each maintenance in the hospital; The repair parts filtering module is used to filter the corresponding robots for each repair based on the repair information of the corresponding robots during each repair, so as to obtain the robots corresponding to each roller repair and the robots corresponding to each screen repair. The reference roller data analysis module is used to extract the accuracy of the roller turning speed and turning angle of the robot corresponding to each roller repair. Then, the average roller turning speed of the robot corresponding to each roller repair is calculated by averaging the roller turning speed of the robot corresponding to the roller repair, which is used as the reference roller turning speed of the robot. Similarly, the accuracy of the reference turning angle of the robot is obtained. The reference screen data analysis module is used to extract the screen response time and screen response strength of the robot corresponding to each screen repair. Then, the average screen response time of the robot corresponding to each screen repair is calculated by averaging the average screen response time of the robot, which is used as the reference screen response time of the robot. Similarly, the reference screen response strength of the robot is obtained. The robot operation monitoring module is used to monitor each robot in the hospital and obtain the accuracy of the corresponding roller turning speed, turning angle, screen response time and screen response click force for each robot. The component status data judgment and analysis module is used to analyze the agility of the corresponding rollers and the sensitivity of the screen of each robot based on the accuracy of the turning speed and turning angle of the corresponding rollers, the screen response time and the screen response force. Then, it judges the agility of the corresponding rollers and the sensitivity of the screen of each robot, and selects each target roller repair robot and each target screen repair robot. The fault display terminal is used to display the faults of each target roller robot and each target screen repair robot.

Images