A method and system for monitoring fault data of physical therapy equipment

Abstract

The present invention discloses a method and system for monitoring fault data of physiotherapy equipment, and relates to the field of data processing technology. A physiotherapy equipment fault data monitoring system includes: a physiotherapy equipment fault data extraction module, a physiotherapy equipment fault data identification module and a physiotherapy equipment fault data analysis module. The present invention can timely discover whether the equipment has a fault by periodically acquiring equipment data and performing fault judgment at preset time intervals, thereby quickly responding; in-depth analysis is performed based on the data set to be analyzed based on the faulty physiotherapy equipment data, and the obtained fault judgment results and processing solutions are more accurate, which helps to solve the problem in a targeted manner; timely discovery and processing of faults can reduce equipment downtime and improve equipment reliability and service life; through physiotherapy equipment fault data analysis, fault data with similar characteristics can be classified into one category, thereby more accurately identifying the type and cause of the fault.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for monitoring fault data of physical therapy equipment. Background Art

[0002] Physiotherapy equipment promotes the rehabilitation and treatment of patients through physical means, and plays an important role in healthcare. However, due to its high frequency of use and long usage time, traditional fault monitoring methods rely on manual inspection and regular maintenance, which may lead to maintenance before or after the actual failure of the equipment, and fail to achieve timely detection and handling of faults; traditional methods often lack real-time monitoring and data analysis of the equipment's operating status, and fail to detect small changes in equipment performance in a timely manner, which may miss the best time to prevent failures, thereby affecting the output data and treatment effects of the physiotherapy equipment, resulting in unsatisfactory treatment results. Summary of the invention

[0003] The present invention aims to provide a method and system for monitoring fault data of a physical therapy device, so as to monitor the operation data of the physical therapy device in real time and respond in time.

[0004] A method for monitoring fault data of a physical therapy device, comprising:

[0005] S1. Obtain the data of the physiotherapy device to be tested at preset time intervals; input the data of the physiotherapy device to be tested into the physiotherapy device data fault identification model to judge the data of the physiotherapy device to be tested, and obtain the fault data judgment result; if the fault data judgment result is yes, store the data of the physiotherapy device to be tested as the fault physiotherapy device data; otherwise, continue to obtain the data of the physiotherapy device to be tested at preset time intervals;

[0006] S2. If the number of faulty physiotherapy equipment data collected within the preset time range reaches N, the N pieces of faulty physiotherapy equipment data within the preset time range are combined to obtain a data set of faulty physiotherapy equipment data to be analyzed; the data set of faulty physiotherapy equipment data to be analyzed is input into the physiotherapy equipment data fault analysis model for analysis to obtain a physiotherapy equipment fault judgment result and a corresponding fault handling plan; the physiotherapy equipment is fault-handled according to the fault handling plan; if the number of faulty physiotherapy equipment data collected within the preset time range does not reach N, then the storage of the faulty physiotherapy equipment data is continued according to step S1.

[0007] As a preferred technical solution of the present invention, the physical therapy equipment data fault identification model in step S1 includes a data preprocessing layer, a fault identification layer and a result output layer;

[0008] The data preprocessing layer is used to preprocess the physical therapy equipment data to be inspected, and obtain preprocessed physical therapy equipment data to be inspected;

[0009] The fault identification layer is used to identify the faults of the pre-processed physiotherapy equipment data to be inspected and obtain the fault data judgment result;

[0010] The result output layer is used to output the fault data judgment results.

[0011] As a preferred technical solution of the present invention, the specific steps of training the fault identification layer include:

[0012] Collecting several groups of physical therapy equipment fault training samples, each group of physical therapy equipment fault training samples includes a pre-processed physical therapy equipment data and a device state label value; combining several groups of physical therapy equipment fault training samples to obtain a physical therapy equipment fault training set;

[0013] The physiotherapy equipment fault training set is input into the physiotherapy equipment data fault identification model to train the fault identification layer with the equipment status label value as the target to obtain the initial fault identification layer; the initial fault identification layer is model evaluated to obtain the initial fault identification layer model evaluation result; if the initial fault identification layer model evaluation result is passed, the initial fault identification layer is used as the fault identification layer in the physiotherapy equipment data fault identification model; otherwise, the physiotherapy equipment fault training set is used to continue the model training.

[0014] As a preferred technical solution of the present invention, the specific steps of determining the value of N in step S2 include:

[0015] Acquire all physiotherapy equipment data within a historical time range to obtain a physiotherapy equipment data set to be analyzed;

[0016] Construct G time sliding windows M g , g = 1, 2, ..., G; each time sliding window M g The sizes in are different, and the sliding step size is the same; using the time sliding window M g Perform mobile fault analysis on the physiotherapy equipment dataset to be analyzed and obtain the mean fault frequency V of the time sliding window g ; Output time sliding window fault frequency mean V g The maximum corresponding time sliding window M g , which is the optimal time sliding window;

[0017] The average number of all faulty physiotherapy equipment data for mobile fault analysis in the physiotherapy equipment data set to be analyzed by the optimal time sliding window is the value of N.

[0018] As a preferred technical solution of the present invention, the physiotherapy equipment data fault grouping model in step S2 includes a data clustering layer, a correlation calculation layer, a fault result evaluation layer and a result output layer;

[0019] The data clustering layer is used to cluster all the faulty physiotherapy equipment data X in the data set to be analyzed. n Perform data clustering analysis to obtain the faulty physiotherapy equipment data clustering identification result S n , n=1, 2, ..., N; N represents the number of all faulty physiotherapy equipment data in the data set of faulty physiotherapy equipment data to be analyzed;

[0020] In the data clustering analysis process, there are K fault therapy data clusters D k , each fault therapy data cluster D k The center point of is the fault treatment data cluster center point C k ; Each fault therapy data cluster D k Corresponding fault cause L k ;

[0021] The correlation calculation layer is used for all the faulty physiotherapy equipment data in the data set to be analyzed. n and the corresponding fault therapy data cluster center point C k Calculate the correlation and obtain the correlation result R of the faulty physiotherapy equipment data. n ; If the faulty physiotherapy equipment data correlation result R n If the preset correlation threshold is reached, the data of the faulty physiotherapy equipment will be retained. n And the corresponding faulty physiotherapy equipment data clustering identification result S n Otherwise, the faulty physical therapy equipment data X n Input into the outlier physiotherapy equipment failure data set, and use the outlier physiotherapy equipment failure data set to update the data clustering layer;

[0022] The fault result evaluation layer is used to cluster and identify the results S based on all faulty physiotherapy equipment data. n Perform feature fusion to obtain a comprehensive judgment result of the physiotherapy equipment fault; match the comprehensive judgment result of the physiotherapy equipment fault in the fault data processing library to obtain the physiotherapy equipment fault judgment result and the corresponding fault handling solution;

[0023] The result output layer is used to output the fault judgment results of the physical therapy equipment and the corresponding fault handling solutions.

[0024] As a preferred technical solution of the present invention, K fault therapy data clusters D are determined in the data clustering layer. k and the corresponding fault therapy data cluster center point C k The specific steps include:

[0025] Construct I fault therapy data clustering individuals P i , each fault therapy data cluster individual P i W iCluster classification of fault therapy data Wi , different fault therapy data clustering individual P i W i The values ​​of are different; cluster I fault therapy data into individual P i Combine to obtain the fault therapy data clustering iterative population; set the maximum number of iterations;

[0026] Collecting a physiotherapy equipment failure anomaly data set, where the physiotherapy equipment failure anomaly data set includes a number of different physiotherapy equipment failure data;

[0027] The steps for simulation calculation using the physiotherapy equipment failure anomaly data set are:

[0028] Step 1: Cluster the fault therapy data according to the fault therapy data clustering iteration population P i Randomly select W in the physiotherapy equipment failure anomaly dataset i locations, as the initial cluster center point H Wi ;

[0029] Step 2: Calculate the distance from each physical therapy equipment failure data in the physical therapy equipment failure analysis data set to the initial cluster center point H Wi The distance between each physical therapy equipment fault data is divided into the fault physical therapy data cluster classification F to which the nearest initial cluster center belongs. Wi middle;

[0030] Step 3: Cluster classification of fault therapy data F Wi , recalculate the clustering classification of fault therapy data F Wi The center point of the cluster is updated by Wi ;

[0031] Step 4: Calculation formula , U represents the clustering criterion function, y∈F Wi The y in the formula represents the cluster classification of fault therapy data F Wi All physiotherapy equipment failure data in T wi Represented as updating the cluster center point T Wi Repeat steps 2 and 3 until the clustering criterion function converges, and the fault therapy data clustering individual P i Cluster classification of all fault therapy data in F Wi Complete the construction; calculate the clustering individual P of the fault physiotherapy data at this time i Cluster classification of all fault therapy data in F Wi The DB index of the fault therapy data clustering individual P is taken as the inverse of the DB index. i The fitness Q i ;

[0032] When the maximum number of iterations is reached, the fault physiotherapy data clustering individual P corresponding to the current maximum fitness is output i , which is the optimal fault therapy data clustering individual; at this time, the W in the optimal fault therapy data clustering individual i The value of is taken as the value of K; W i Cluster classification of fault therapy data Wi As a fault treatment data clustering D k ; Cluster and classify fault therapy data F Wi The cluster center point in is taken as the corresponding fault treatment data cluster center point C k .

[0033] As a preferred technical solution of the present invention, the specific steps of updating the data clustering layer using the outlier physiotherapy equipment fault data set include:

[0034] Combining the outlier physiotherapy equipment failure data set with the physiotherapy equipment failure anomaly data set to obtain a new physiotherapy equipment failure anomaly data set;

[0035] Use the new physiotherapy equipment fault anomaly data set to perform simulation calculations and repeat steps 1-4. When the maximum number of iterations is reached, output the new K fault physiotherapy data clusters D corresponding to the optimal fault physiotherapy data cluster individuals. k and the corresponding fault therapy data cluster center point C k ;

[0036] Using the new K fault therapy data clustering D k and the corresponding fault therapy data cluster center point C k Update the data clustering layer.

[0037] A physical therapy equipment fault data monitoring system, comprising:

[0038] A physiotherapy equipment fault data extraction module, used to obtain the data of the physiotherapy equipment to be inspected at preset time intervals;

[0039] The physiotherapy equipment fault data identification module is used to deploy the physiotherapy equipment data fault identification model, input the physiotherapy equipment data to be inspected into the physiotherapy equipment data fault identification model to judge the physiotherapy equipment data to be inspected, and obtain the fault data judgment result; if the fault data judgment result is yes, the physiotherapy equipment data to be inspected is stored as the fault physiotherapy equipment data; otherwise, continue to obtain the physiotherapy equipment data to be inspected at a preset time interval;

[0040] The physiotherapy equipment fault data analysis module is used to deploy the physiotherapy equipment data fault analysis model. If the amount of fault physiotherapy equipment data collected within a preset time range reaches N, the N pieces of fault physiotherapy equipment data within the preset time range are combined to obtain a data set of fault physiotherapy equipment data to be analyzed; the data set of fault physiotherapy equipment data to be analyzed is input into the physiotherapy equipment data fault analysis model for analysis to obtain a physiotherapy equipment fault judgment result and a corresponding fault handling plan; the physiotherapy equipment is fault-handled according to the fault handling plan; if the amount of fault physiotherapy equipment data collected within the preset time range does not reach N, the storage of the fault physiotherapy equipment data is continued according to step S1.

[0041] The present invention has the following advantages:

[0042] 1. The present invention can timely discover whether the equipment has a fault by periodically acquiring equipment data and performing fault judgment at preset time intervals, thereby quickly responding; in-depth analysis is performed based on the data set to be analyzed of the faulty physiotherapy equipment data, and the obtained fault judgment results and processing solutions are more accurate, which helps to solve the problem in a targeted manner; timely discovery and processing of faults can reduce equipment downtime and improve equipment reliability and service life; through data clustering analysis, fault data with similar characteristics can be classified into one category, thereby more accurately identifying the type and cause of the fault; through correlation calculation, the degree of correlation between the fault data and the specific cluster center point can be determined, which helps to more accurately match the fault processing solution.

[0043] 2. The present invention can predict the types of possible equipment failures by performing data clustering and correlation calculation in the physical therapy equipment data fault analysis model, thereby performing preventive maintenance in advance; by continuously updating and optimizing the data clustering layer, the equipment can better adapt to different failure modes and changes, and improve the ability of self-diagnosis and self-repair; through matching of the fault data processing library, it can provide data support for the diagnosis of equipment failures, and enhance the accuracy and traceability of fault cause analysis; the data clustering layer is updated using the outlier physical therapy equipment fault data set, so that the model can dynamically adapt to new data features, thereby improving the dynamic adaptability of clustering. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of the structure of a physical therapy equipment fault data monitoring system used in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0046] Embodiment 1, a method for monitoring fault data of a physical therapy device, comprising:

[0047] S1. Obtain the data of the physiotherapy device to be tested at preset time intervals; input the data of the physiotherapy device to be tested into the physiotherapy device data fault identification model to judge the data of the physiotherapy device to be tested, and obtain the fault data judgment result; if the fault data judgment result is yes, store the data of the physiotherapy device to be tested as the fault physiotherapy device data; otherwise, continue to obtain the data of the physiotherapy device to be tested at preset time intervals;

[0048] The preset time interval is set by professional technicians according to actual conditions. The data of the physiotherapy equipment to be inspected is obtained from the sensor in the physiotherapy equipment, which can reflect the working status of the physiotherapy equipment;

[0049] The physical therapy equipment data fault identification model in step S1 includes a data preprocessing layer, a fault identification layer and a result output layer;

[0050] The data preprocessing layer is used to preprocess the physical therapy equipment data to be inspected, and obtain preprocessed physical therapy equipment data to be inspected;

[0051] The fault identification layer is used to identify the faults of the pre-processed physiotherapy equipment data to be inspected and obtain the fault data judgment result;

[0052] The result output layer is used to output the fault data judgment results;

[0053] The specific steps for training the fault identification layer include:

[0054] Collect several groups of physiotherapy equipment fault training samples, each group of physiotherapy equipment fault training samples includes a pre-processed physiotherapy equipment data and a device status label value; combine several groups of physiotherapy equipment fault training samples to obtain a physiotherapy equipment fault training set; the device status label value includes yes and no, yes represents faulty device data, and no represents non-faulty device data, and the device status label value is annotated by professional technicians;

[0055] Input the physical therapy equipment fault training set into the physical therapy equipment data fault identification model to train the fault identification layer with the equipment status label value as the target to obtain the initial fault identification layer; perform model evaluation on the initial fault identification layer to obtain the initial fault identification layer model evaluation result; if the initial fault identification layer model evaluation result is passed, the initial fault identification layer is used as the fault identification layer in the physical therapy equipment data fault identification model; otherwise, continue to perform model training using the physical therapy equipment fault training set;

[0056] The BP neural network model is used as the basic model of the physical therapy equipment data fault identification model. Through professional data preprocessing and training of the fault identification layer, the model can more accurately identify the fault status of the physical therapy equipment and reduce misjudgment and missed judgment. Traditional fault data detection often relies on manual inspection, while the automated fault data identification model can reduce the dependence on professional technicians and save time. The model can provide data support for the diagnosis of equipment faults by training and learning the normal and faulty states of the equipment, and enhance the accuracy and traceability of fault cause analysis.

[0057] S2. If the number of faulty physiotherapy equipment data collected within the preset time range reaches N, the N pieces of faulty physiotherapy equipment data within the preset time range are combined to obtain a data set of faulty physiotherapy equipment data to be analyzed; the data set of faulty physiotherapy equipment data to be analyzed is input into the physiotherapy equipment data fault analysis model for analysis to obtain a physiotherapy equipment fault judgment result and a corresponding fault handling plan; the physiotherapy equipment is fault handled according to the fault handling plan; if the number of faulty physiotherapy equipment data collected within the preset time range does not reach N, the storage of the faulty physiotherapy equipment data is continued according to step S1;

[0058] The starting time of the preset time range starts from the receipt of the first piece of data of the faulty physiotherapy device. At this time, the faulty data of the physiotherapy device has just appeared, which cannot indicate that the physiotherapy device is in a faulty state. If the number of data collected within the preset time range does not exceed N, misjudgment may occur, wasting time. When the number of data of the faulty physiotherapy device collected within the preset time range reaches N, data processing of the data of the faulty physiotherapy device begins;

[0059] The specific steps of determining the value of N in step S2 include:

[0060] Acquire all physiotherapy equipment data within a historical time range to obtain a physiotherapy equipment data set to be analyzed; the physiotherapy equipment data set to be analyzed is continuous data within the historical time range;

[0061] Construct G time sliding windows M g , g = 1, 2, ..., G; each time sliding window M g The sizes of the sliding window M are different, and the sliding step length is the same; the sliding step length is set by professional technicians according to the actual situation; g Perform mobile fault analysis on the physiotherapy equipment dataset to be analyzed and obtain the mean fault frequency V of the time sliding window g ; Output time sliding window fault frequency mean V g The maximum corresponding time sliding window M g , which is the optimal time sliding window; the mean fault frequency of the time sliding window is V gThe calculation steps are as follows: when the faulty physiotherapy equipment data is collected within the time sliding window, the number of faulty physiotherapy equipment data is calculated, and the number of faulty physiotherapy equipment data is divided by the time sliding window M. g The size in is the fault frequency, and the mean of all fault frequencies that the time sliding window slides in the physiotherapy equipment data set to be analyzed is calculated;

[0062] The average number of all faulty physiotherapy equipment data for mobile fault analysis in the physiotherapy equipment data set to be analyzed by the optimal time sliding window is the value of N; after the value of N is determined, it can be used after being evaluated by professional technicians to see if it is feasible;

[0063] Through the analysis of the time sliding window, the best detection time can be determined, so that the fault can be identified and processed at the early stage of occurrence, reducing the impact of the fault on the operation of the equipment; by reasonably setting the size and step of the time sliding window, false alarms due to accidental fault data can be avoided, and the accuracy of fault detection can be improved; by optimizing the fault detection time and improving the accuracy of fault detection, unnecessary maintenance and inspection can be reduced, thereby reducing maintenance costs;

[0064] The physical therapy equipment data fault grouping model in step S2 includes a data clustering layer, a correlation calculation layer, a fault result evaluation layer and a result output layer;

[0065] The data clustering layer is used to cluster all the faulty physiotherapy equipment data X in the data set to be analyzed. n Perform data clustering analysis to obtain the faulty physiotherapy equipment data clustering identification result S n , n=1, 2, ..., N; N represents the number of all faulty physiotherapy equipment data in the data set to be analyzed; the clustering recognition result S of the faulty physiotherapy equipment data n For data on physiotherapy equipment at fault X n After data clustering analysis, the fault therapy data cluster D k Corresponding fault cause L k ;

[0066] In the data clustering analysis process, there are K fault therapy data clusters D k , each fault therapy data cluster D k The center point of is the fault treatment data cluster center point C k ; Each fault therapy data cluster D k Corresponding fault cause L k ; Fault cause L k Professional technicians will cluster all fault therapy data according to the actual situation. k Conduct evaluation and analysis, and mark the causes, such as equipment voltage being too high or too low;

[0067] The correlation calculation layer is used for all the faulty physiotherapy equipment data in the data set to be analyzed. n and the corresponding fault therapy data cluster center point C k Calculate the correlation and obtain the correlation result R of the faulty physiotherapy equipment data. n ; If the faulty physiotherapy equipment data correlation result R n If the preset correlation threshold is reached, the data of the faulty physiotherapy equipment will be retained. n And the corresponding faulty physiotherapy equipment data clustering identification result S n Otherwise, the faulty physical therapy equipment data X n The data is input into the outlier physiotherapy equipment fault data set, and the outlier physiotherapy equipment fault data set is used to update the data clustering layer; the correlation calculation layer is used to detect outlier data. If there is data that exceeds all fault physiotherapy data clusters D during the fault data monitoring process, k , it means that the current classification needs to be updated and an unknown equipment failure has occurred; the preset correlation threshold is set by professional technicians according to the actual situation; the corresponding fault physiotherapy data cluster center point C k is the faulty physiotherapy equipment data clustering identification result S obtained after data clustering n Corresponding fault therapy data cluster D k The center point of is the fault treatment data cluster center point C k ;

[0068] The fault result evaluation layer is used to cluster and identify the results S based on all faulty physiotherapy equipment data. n Perform feature fusion to obtain a comprehensive judgment result of the physiotherapy equipment fault; match the comprehensive judgment result of the physiotherapy equipment fault in the fault data processing library to obtain the physiotherapy equipment fault judgment result and the corresponding fault handling solution;

[0069] The result output layer is used to output the fault judgment results of the physical therapy equipment and the corresponding fault handling solutions;

[0070] The fault data processing library contains several fault results and corresponding solutions, which are continuously updated by professional technicians based on actual experience;

[0071] Through data clustering analysis, fault data with similar characteristics can be grouped together to more accurately identify the fault type and cause. Through correlation calculation, the degree of correlation between fault data and a specific cluster center point can be determined, which helps to more accurately match fault handling solutions. Through feature fusion and comprehensive judgment, fault data can be analyzed more comprehensively, providing more in-depth fault diagnosis and handling suggestions. Through automated fault identification and matching of handling solutions, manual intervention can be reduced, and the speed and efficiency of fault handling can be improved. By continuously updating and optimizing the data clustering layer, the equipment can better adapt to different fault modes and changes, and improve its self-diagnosis and self-repair capabilities.

[0072] Determine K fault therapy data clusters D in the data clustering layer k and the corresponding fault therapy data cluster center point C k The specific steps include:

[0073] Construct I fault therapy data clustering individuals P i , each fault therapy data cluster individual P i W i Cluster classification of fault therapy data Wi , different fault therapy data clustering individual P i W i The values ​​of are different; cluster I fault therapy data into individual P i combination, obtain the fault physiotherapy data clustering iteration population; set the maximum number of iterations; the maximum number of iterations is set by professional and technical personnel according to actual conditions; construct the fault physiotherapy data clustering individual P i When , we first establish a preliminary clustering classification of fault therapy data, W i The specific data fluctuates within the preset fluctuation range, which is consistent with the actual situation of fault classification;

[0074] Collecting a data set of abnormal physiotherapy equipment failures, which contains several pieces of different physiotherapy equipment failure data; the data set of abnormal physiotherapy equipment failures is integrated and constructed by professionals, and contains various data when physiotherapy equipment fails for subsequent analysis;

[0075] The steps for simulation calculation using the physiotherapy equipment failure anomaly data set are:

[0076] Step 1: Cluster the fault therapy data according to the fault therapy data clustering iteration population P i Randomly select W in the physiotherapy equipment failure anomaly dataset i locations, as the initial cluster center point H Wi ;

[0077] Step 2: Calculate the distance from each physical therapy equipment failure data in the physical therapy equipment failure analysis data set to the initial cluster center point H Wi The distance between each physical therapy equipment fault data is divided into the fault physical therapy data cluster classification F to which the nearest initial cluster center belongs. Wi middle;

[0078] Step 3: Cluster classification of fault therapy data F Wi , recalculate the clustering classification of fault therapy data F Wi The center point of the cluster is updated by Wi ;

[0079] Step 4: Calculation formula , U represents the clustering criterion function, y∈F Wi The y in the formula represents the cluster classification of fault therapy data F Wi All physiotherapy equipment failure data in T wi Represented as updating the cluster center point T Wi；重复步骤2和步骤3，直到聚类准则函数收敛时，故障理疗数据聚类个体Pi Cluster classification of all fault therapy data in F Wi Complete the construction; calculate the clustering individual P of the fault physiotherapy data at this time i Cluster classification of all fault therapy data in F Wi The DB index of the fault therapy data clustering individual P is taken as the inverse of the DB index. i The fitness Q i ; The DB indicator is mainly used to measure the compactness and separation of clusters. The smaller the DB value, the better the clustering effect. The minimum value is 0, and the value close to zero indicates a better partition.

[0080] When the maximum number of iterations is reached, the fault physiotherapy data clustering individual P corresponding to the current maximum fitness is output i , which is the optimal fault therapy data clustering individual; at this time, the W in the optimal fault therapy data clustering individual i The value of is taken as the value of K; W i Cluster classification of fault therapy data Wi As a fault treatment data clustering D k ; Cluster and classify fault therapy data F Wi The cluster center point in is taken as the corresponding fault treatment data cluster center point C k ;

[0081] By iteratively optimizing cluster center points, fault data can be more accurately divided into corresponding clusters, improving the compactness and separation of clusters; accurate calculation of cluster center points helps to more accurately identify and classify different types of fault data, thereby improving the accuracy of fault identification; using DB indicators as fitness functions, it can handle clusters of different sizes and densities, and enhance the robustness of clustering algorithms to noise and outliers; through cluster analysis, the types of possible equipment failures can be predicted, so that preventive maintenance can be carried out in advance and unexpected downtime can be reduced; through continuous cluster optimization and fitness evaluation, the effects of fault identification and processing can be continuously optimized, supporting the continuous improvement and upgrading of equipment fault handling technology;

[0082] The specific steps for updating the data clustering layer using the outlier physiotherapy equipment failure dataset include:

[0083] Combining the outlier physiotherapy equipment failure data set with the physiotherapy equipment failure anomaly data set to obtain a new physiotherapy equipment failure anomaly data set;

[0084] Use the new physiotherapy equipment fault anomaly data set to perform simulation calculations and repeat steps 1-4. When the maximum number of iterations is reached, output the new K fault physiotherapy data clusters D corresponding to the optimal fault physiotherapy data cluster individuals. k and the corresponding fault therapy data cluster center point C k ;

[0085] Using the new K fault therapy data clustering D k and the corresponding fault therapy data cluster center point C k Update the data clustering layer;

[0086] By introducing outlier data, the clustering model can better capture abnormal patterns in the data and improve the accuracy of the clustering results; updating the clustering model to include outlier data makes the clustering results more flexible to adapt to data changes and enhances the ability to identify different fault modes; dynamically updating the clustering model enables the model to continuously learn and adapt to new data features and improve the model's self-adaptability; through continuous clustering optimization and fitness evaluation, the effect of fault identification and processing can be continuously optimized, supporting the continuous improvement and upgrading of equipment fault handling technology.

[0087] Example 2, a physical therapy equipment fault data monitoring system, see Figure 1 As shown, including:

[0088] A physiotherapy equipment fault data extraction module, used to obtain the data of the physiotherapy equipment to be inspected at preset time intervals;

[0089] The physiotherapy equipment fault data identification module is used to deploy the physiotherapy equipment data fault identification model, input the physiotherapy equipment data to be inspected into the physiotherapy equipment data fault identification model to judge the physiotherapy equipment data to be inspected, and obtain the fault data judgment result; if the fault data judgment result is yes, the physiotherapy equipment data to be inspected is stored as the fault physiotherapy equipment data; otherwise, continue to obtain the physiotherapy equipment data to be inspected at a preset time interval;

[0090] The physiotherapy equipment fault data analysis module is used to deploy the physiotherapy equipment data fault analysis model. If the amount of fault physiotherapy equipment data collected within a preset time range reaches N, the N pieces of fault physiotherapy equipment data within the preset time range are combined to obtain a data set of fault physiotherapy equipment data to be analyzed; the data set of fault physiotherapy equipment data to be analyzed is input into the physiotherapy equipment data fault analysis model for analysis to obtain a physiotherapy equipment fault judgment result and a corresponding fault handling plan; the physiotherapy equipment is fault-handled according to the fault handling plan; if the amount of fault physiotherapy equipment data collected within the preset time range does not reach N, the storage of the fault physiotherapy equipment data is continued according to step S1.

[0091] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A method for monitoring fault data of a physical therapy device, characterized in that: include: S1. Obtain the data of the physiotherapy equipment to be tested at preset time intervals; Input the to-be-tested physiotherapy equipment data into the physiotherapy equipment data fault identification model to judge the to-be-tested physiotherapy equipment data and obtain the fault data judgment result; If the result of the fault data judgment is yes, the data of the physical therapy device to be inspected is stored as the data of the faulty physical therapy device; otherwise, the data of the physical therapy device to be inspected continues to be obtained at a preset time interval; S2. If the number of faulty physiotherapy equipment data collected within a preset time range reaches N, the N pieces of faulty physiotherapy equipment data within a preset time range are combined to obtain a data set of faulty physiotherapy equipment data to be analyzed; Inputting the data set of the faulty physiotherapy equipment data to be analyzed into the physiotherapy equipment data fault analysis model for analysis, obtaining the physiotherapy equipment fault judgment result and the corresponding fault handling plan; and handling the fault of the physiotherapy equipment according to the fault handling plan; If the amount of data collected from the faulty physiotherapy device within the preset time range does not reach N, then continue to store the data from the faulty physiotherapy device according to step S1; The specific steps of determining the value of N in step S2 include: Acquire all physiotherapy equipment data within a historical time range to obtain a physiotherapy equipment data set to be analyzed; Construct G time sliding windows M g , g = 1, 2, ..., G; each time sliding window M g The sizes in are different, and the sliding step size is the same; using the time sliding window M g Perform mobile fault analysis on the physiotherapy equipment dataset to be analyzed and obtain the mean fault frequency V of the time sliding window g ; Output time sliding window fault frequency mean V g The maximum corresponding time sliding window M g , which is the optimal time sliding window; The average number of all faulty physiotherapy equipment data for mobile fault analysis in the physiotherapy equipment data set to be analyzed by the optimal time sliding window is the value of N; The physical therapy equipment data fault grouping model in step S2 includes a data clustering layer, a correlation calculation layer, a fault result evaluation layer and a result output layer; The data clustering layer is used to cluster all the faulty physiotherapy equipment data X in the data set to be analyzed. n Perform data clustering analysis to obtain the faulty physiotherapy equipment data clustering identification result S n , n=1, 2, ..., N; N represents the number of all faulty physiotherapy equipment data in the data set of faulty physiotherapy equipment data to be analyzed; In the data clustering analysis process, there are K fault therapy data clusters D k , each fault therapy data cluster D k The center point of is the fault treatment data cluster center point C k ; Each fault therapy data cluster D k Corresponding fault cause L k ; The correlation calculation layer is used for all the faulty physiotherapy equipment data in the data set to be analyzed. n and the corresponding fault therapy data cluster center point C k Calculate the correlation and obtain the correlation result R of the faulty physiotherapy equipment data. n ; If the faulty physiotherapy equipment data correlation result R n If the preset correlation threshold is reached, the data of the faulty physiotherapy equipment will be retained. n And the corresponding faulty physiotherapy equipment data clustering identification result S n Otherwise, the faulty physical therapy equipment data X n Input into the outlier physiotherapy equipment failure data set, and use the outlier physiotherapy equipment failure data set to update the data clustering layer; The fault result evaluation layer is used to cluster and identify the results S based on all faulty physiotherapy equipment data. n Perform feature fusion to obtain a comprehensive judgment result of the physiotherapy equipment fault; match the comprehensive judgment result of the physiotherapy equipment fault in the fault data processing library to obtain the physiotherapy equipment fault judgment result and the corresponding fault handling solution; The result output layer is used to output the fault judgment results of the physical therapy equipment and the corresponding fault handling solutions.

2. A method for monitoring fault data of a physical therapy device according to claim 1, characterized in that: The physical therapy equipment data fault identification model in step S1 includes a data preprocessing layer, a fault identification layer and a result output layer; The data preprocessing layer is used to preprocess the physical therapy equipment data to be inspected, and obtain preprocessed physical therapy equipment data to be inspected; The fault identification layer is used to identify the faults of the pre-processed physiotherapy equipment data to be inspected and obtain the fault data judgment result; The result output layer is used to output the fault data judgment results.

3. A method for monitoring fault data of a physical therapy device according to claim 2, characterized in that: The specific steps for training the fault identification layer include: Collecting several groups of physical therapy equipment fault training samples, each group of physical therapy equipment fault training samples includes a pre-processed physical therapy equipment data and a device state label value; combining several groups of physical therapy equipment fault training samples to obtain a physical therapy equipment fault training set; The physiotherapy equipment fault training set is input into the physiotherapy equipment data fault identification model to train the fault identification layer with the equipment status label value as the target to obtain the initial fault identification layer; the initial fault identification layer is model evaluated to obtain the initial fault identification layer model evaluation result; if the initial fault identification layer model evaluation result is passed, the initial fault identification layer is used as the fault identification layer in the physiotherapy equipment data fault identification model; otherwise, the physiotherapy equipment fault training set is used to continue the model training.

4. A method for monitoring fault data of a physical therapy device according to claim 3, characterized in that: Determine K fault therapy data clusters D in the data clustering layer k and the corresponding fault therapy data cluster center point C k The specific steps include: Construct I fault therapy data clustering individuals P i , each fault therapy data cluster individual P i W i Cluster classification of fault therapy data Wi , different fault therapy data clustering individual P i W i The values ​​of are different; cluster I fault therapy data into individual P i Combine to obtain the fault therapy data clustering iterative population; set the maximum number of iterations; Collecting a physiotherapy equipment failure anomaly data set, where the physiotherapy equipment failure anomaly data set includes a number of different physiotherapy equipment failure data; The steps for simulation calculation using the physiotherapy equipment failure anomaly data set are: Step 1: Cluster the fault therapy data according to the fault therapy data clustering iteration population P i Randomly select W in the physiotherapy equipment failure anomaly dataset i locations, as the initial cluster center point H Wi ; Step 2: Calculate the distance from each physical therapy equipment failure data in the physical therapy equipment failure analysis data set to the initial cluster center point H Wi The distance between each physical therapy equipment fault data is divided into the fault physical therapy data cluster classification F to which the nearest initial cluster center belongs. Wi middle; Step 3: Cluster classification F for fault therapy data Wi , recalculate the clustering classification of fault therapy data F Wi The center point of the cluster is updated by Wi ; Step 4: Calculation formula , U represents the clustering criterion function, y∈F Wi The y in the formula represents the cluster classification of fault therapy data F Wi All physiotherapy equipment failure data in T wi Represented as updating the cluster center point T Wi；重复步骤2和步骤3，直到聚类准则函数收敛时，故障理疗数据聚类个体Pi Cluster classification of all fault therapy data in F Wi Complete the construction; calculate the clustering individual P of the fault physiotherapy data at this time i Cluster classification of all fault therapy data in F Wi The DB index of the fault therapy data clustering individual P is taken as the inverse of the DB index. i The fitness Q i ; When the maximum number of iterations is reached, the fault physiotherapy data clustering individual P corresponding to the current maximum fitness is output i , which is the optimal fault therapy data clustering individual; at this time, the W in the optimal fault therapy data clustering individual i The value of is taken as the value of K; W i Cluster classification of fault therapy data Wi As a fault treatment data clustering D k ; Cluster and classify fault therapy data F Wi The cluster center point in is taken as the corresponding fault treatment data cluster center point C k .

5. A method for monitoring fault data of a physical therapy device according to claim 4, characterized in that: The specific steps for updating the data clustering layer using the outlier physiotherapy equipment failure dataset include: Combining the outlier physiotherapy equipment failure data set with the physiotherapy equipment failure anomaly data set to obtain a new physiotherapy equipment failure anomaly data set; Use the new physiotherapy equipment fault anomaly data set to perform simulation calculations and repeat steps 1-4. When the maximum number of iterations is reached, output the new K fault physiotherapy data clusters D corresponding to the optimal fault physiotherapy data cluster individuals. k and the corresponding fault therapy data cluster center point C k ; Using the new K fault therapy data clustering D k and the corresponding fault therapy data cluster center point C k Update the data clustering layer.

6. A physical therapy equipment fault data monitoring system, characterized in that: The system applies a method for monitoring fault data of a physical therapy device according to any one of claims 1 to 5, comprising: A physiotherapy equipment fault data extraction module, used to obtain the data of the physiotherapy equipment to be inspected at preset time intervals; The physiotherapy equipment fault data identification module is used to deploy the physiotherapy equipment data fault identification model, input the physiotherapy equipment data to be inspected into the physiotherapy equipment data fault identification model to judge the physiotherapy equipment data to be inspected, and obtain the fault data judgment result; if the fault data judgment result is yes, the physiotherapy equipment data to be inspected is stored as the fault physiotherapy equipment data; otherwise, continue to obtain the physiotherapy equipment data to be inspected at a preset time interval; The physiotherapy equipment fault data analysis module is used to deploy the physiotherapy equipment data fault analysis model. If the amount of fault physiotherapy equipment data collected within a preset time range reaches N, the N pieces of fault physiotherapy equipment data within the preset time range are combined to obtain a data set of fault physiotherapy equipment data to be analyzed; the data set of fault physiotherapy equipment data to be analyzed is input into the physiotherapy equipment data fault analysis model for analysis to obtain a physiotherapy equipment fault judgment result and a corresponding fault handling plan; the physiotherapy equipment is fault-handled according to the fault handling plan; if the amount of fault physiotherapy equipment data collected within the preset time range does not reach N, the storage of the fault physiotherapy equipment data is continued according to step S1.

Images