Elevator speed anomaly detection and fault early warning method and device, equipment and medium

Abstract

The invention discloses an elevator speed anomaly detection and fault early warning method and device, equipment and a medium. The elevator speed anomaly detection and fault early warning method comprises the steps that elevator speed data and various influence data are obtained; calculating the noise degree of the elevator speed data and determining an initial K value; determining a distance set of data points in the elevator speed data by utilizing a KNN algorithm based on the elevator speed data and the various influence data; calculating the distribution degree of each distance set, and determining an improved K value of each data point based on the distribution degree and the initial K value; based on the elevator speed data, the multiple influence data and the improved K value of each data point, the abnormal elevator speed is determined through a KNN algorithm; and elevator fault early warning is conducted based on the abnormal elevator speed, the elevator speed data and the multiple influence data. The problem that when the KNN algorithm is used for elevator speed anomaly detection, if the K value is set improperly, the situation of over-fitting or under-fitting may be caused, and then the accuracy of anomaly detection is affected is solved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular, to an elevator speed anomaly detection and fault warning method, device, equipment and medium. Background Art

[0002] As an important vertical transportation tool in modern buildings, the safety and stability of elevator operation directly affect the daily lives of residents and users. Once the elevator speed is abnormal, it often indicates potential problems such as motor failure, control system failure or power supply fluctuation in the equipment. If these abnormalities are not identified and processed in a timely manner, it may not only lead to elevator shutdown and passengers being trapped, but also trigger more serious safety accidents. Therefore, building an effective speed anomaly detection and fault warning mechanism is of great significance for improving the operation reliability of elevators, ensuring the safety of passengers' lives, reducing maintenance costs and downtime.

[0003] To ensure the accuracy of elevator speed anomaly detection and fault warning, compared with relying only on single-dimensional data, it is usually more inclined to use multi-dimensional data for processing. Compared with other anomaly detection algorithms, the KNN (K-Nearest Neighbor) algorithm can identify anomalies by calculating the distances between points in a multi-dimensional space. However, the accuracy of the KNN algorithm in elevator speed anomaly detection greatly depends on the selection of the K value, and the K value needs to be set artificially in advance. If the K value is set too large, the KNN algorithm may ignore some real outliers, resulting in inaccurate detection; on the contrary, if the K value is set too small, although the sensitivity of the algorithm to anomalies is enhanced, it may also cause overfitting and be more easily misled by noise data.

[0004] The above content is only used to assist in understanding the technical solution of the present application, and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of the present application is to provide an elevator speed anomaly detection and fault warning method, device, equipment and medium, aiming to solve the technical problem that when using the KNN algorithm for elevator speed anomaly detection, if the K value is set improperly, it may lead to overfitting or underfitting, thus affecting the accuracy of anomaly detection.

[0006] To achieve the above object, the present application provides an elevator speed anomaly detection and fault warning method, including: obtaining elevator speed data and various influencing data; determining an initial K value based on the elevator speed data; determining a distance set of each data point in the elevator speed data based on the elevator speed data and various influencing data and using the KNN algorithm; determining an improved K value of each data point based on the distance set of each data point and the initial K value; determining an abnormal elevator speed based on the elevator speed data, various influencing data and the improved K value of each data point and using the KNN algorithm; and performing elevator fault warning based on the abnormal elevator speed, the elevator speed data and various influencing data.

[0007] Optionally, the determining the initial K value based on the elevator speed data includes: calculating the noise level of the elevator speed data; and determining the initial K value based on the noise level.

[0008] Optionally, the determining the initial K value based on the noise level includes: using the following formula (1) to determine the initial K value:

[0009] K = K min +(K max - K min ) × α (1)

[0010] wherein, K represents the initial K value, K max represents the preset maximum K value, K min represents the preset minimum K value, and α represents the noise level.

[0011] Optionally, the determining the improved K value of each data point based on the distance set of each data point and the initial K value includes: determining the distribution degree of each data point based on the distance set of each data point; and determining the improved K value of each data point based on the distribution degree and the initial K value.

[0012] Optionally, the determining the distribution degree of each data point based on the distance set of each data point includes: performing clustering processing on the distance set of each data point by using a clustering algorithm to obtain a plurality of clusters; calculating the DTW distance between two clusters; and determining the distribution degree of each data point based on a plurality of the DTW distances.

[0013] Optionally, the performing elevator fault warning based on the abnormal elevator speed, the elevator speed data and various influencing data includes: determining a plurality of predicted elevator speeds based on the elevator speed data and various influencing data and using the KNN algorithm; and performing elevator fault warning based on the abnormal elevator speed and the plurality of predicted elevator speeds.

[0014] Optionally, calculating the noise level of the elevator speed data includes: calculating the signal-to-noise ratio and coefficient of variation of the elevator speed data; and determining the noise level of the elevator speed data based on the signal-to-noise ratio and the coefficient of variation.

[0015] In addition, to achieve the above object, the present application further provides an elevator speed anomaly detection and fault warning device, including: a data acquisition module for acquiring elevator speed data and various influencing data; an initial K value setting module for determining an initial K value based on the elevator speed data; a model construction module for determining a distance set of each data point in the elevator speed data based on the elevator speed data and various influencing data and using the KNN algorithm; an improved K value setting module for determining an improved K value of each data point based on the distance set of each data point and the initial K value; an anomaly detection module for determining an abnormal elevator speed based on the elevator speed data, various influencing data, and the improved K value of each data point and using the KNN algorithm; and a fault warning module for performing elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influencing data.

[0016] The present application further provides an elevator speed anomaly detection and fault warning device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above-mentioned elevator speed anomaly detection and fault warning method.

[0017] The present application further provides a computer-readable storage medium, including: a computer program stored therein, and when the computer program is executed by a processor, the above-mentioned elevator speed anomaly detection and fault warning method is implemented.

[0018] A method, device, device, and medium for elevator speed anomaly detection and fault warning proposed by the present application comprehensively analyze elevator speed data and multi-dimensional influencing data, and flexibly determine an improved K value for each data point using the KNN algorithm, improving the accuracy and robustness of elevator speed anomaly detection. Compared with the method of using a fixed K value in the traditional KNN algorithm, the present application can dynamically adjust the K value according to the distance set of each data point and data differences, making the model more sensitive when there is more noise and able to capture local anomalies; while when the data is more complex or there is less noise, the K value is larger, enhancing the stability of the model and reducing the over-response to outliers. The present application not only effectively avoids the overfitting problem caused by the traditional algorithm for elevator speed anomaly detection, but also ensures the accurate detection of different types of anomalies such as sudden anomalies, persistent anomalies, and periodic anomalies, thereby improving the accuracy and reliability of elevator fault warning. Description of the Drawings

[0019] Figure 1 Flow chart of an elevator speed anomaly detection and fault warning method according to an embodiment of the present application;

[0020] Figure 2 Structural block diagram of an elevator speed anomaly detection and fault warning device according to an embodiment of the present application;

[0021] Figure 3 Schematic structural diagram of an elevator speed anomaly detection and fault warning device according to an embodiment of the present application.

[0022] The realization of the purpose of the present application, functional features and advantages will be further described in conjunction with embodiments with reference to the accompanying drawings. Specific embodiments

[0023] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] In the prior art, the steps of using the KNN algorithm for elevator speed anomaly detection are usually as follows:

[0025] Obtain multi-dimensional operation data of the elevator;

[0026] Calculate the distance between each data point and other data points;

[0027] Select K nearest neighbor data points for each data point based on a preset K value and distance value;

[0028] Determine whether each data point is an abnormal point based on the K nearest neighbor data points of each data point.

[0029] Among them, since the K value is preset, if the K value is set to be relatively large, more nearest neighbor data points are selected, and thus it is easy to ignore some local abnormal points. For example, occasional instantaneous fluctuations in the elevator speed will be considered normal values. On the contrary, if the K value is set to be relatively small, the model is more sensitive to the local environment around each data point and can capture local anomalies more accurately. However, the model is extremely vulnerable to noise. If there are many noise points in the multi-dimensional operation data of the elevator, some normal data points will be misled by the noise data and thus be determined as abnormal points.

[0030] To solve the above problems, the present application provides an elevator speed anomaly detection and fault warning method, device, equipment and medium. The solutions of the present application will be introduced in detail below.

[0031] Figure 1FIG. 0 is a flowchart of an elevator speed anomaly detection and fault warning method according to an embodiment of the present application. The elevator speed anomaly detection and fault warning method can be executed by an anomaly detection and fault warning device with data processing capabilities. The anomaly detection and fault warning device can be, for example, an elevator speed anomaly detection and fault warning device. Referring to Figure 1 , the elevator speed anomaly detection and fault warning method may include the following steps:

[0032] S1. Obtain elevator speed data and various influencing data.

[0033] Among them, the various influencing data represent data that affect the elevator speed. The various influencing data can be elevator load data, elevator acceleration data, floor gap data, etc.

[0034] In a specific implementation process, obtain the elevator speed data and various influencing data of a preset time length in the elevator control system. It should be noted that the timestamps of the elevator speed data and the various influencing data are the same.

[0035] S2. Determine an initial K value based on the elevator speed data.

[0036] In this step, the elevator speed anomaly detection and fault warning device can first calculate the noise level of the elevator speed data, and then determine the initial K value based on the noise level.

[0037] Among them, calculating the noise level of data is to measure the part of the data that does not conform to the true signal, that is, to calculate the number of noise points. Noise points are usually generated by sensor failures, signal interference, or external environmental factors (such as electromagnetic interference); the initial K value is the preset K value in the KNN algorithm. It can be understood that in traditional algorithms, the initial K value is set artificially in advance, and in this embodiment, the initial K value is determined based on the noise level of the data.

[0038] In one embodiment, in step S2, calculating the noise level of the elevator speed data may specifically include:

[0039] S21. Calculate the signal-to-noise ratio and coefficient of variation of the elevator speed data;

[0040] S22. Determine the noise level of the elevator speed data based on the signal-to-noise ratio and the coefficient of variation.

[0041] In a specific implementation process, first calculate the signal-to-noise ratio and coefficient of variation of the elevator speed data. It should be noted that the signal-to-noise ratio is the ratio of the square of the mean value of the elevator speed data to the square of the standard deviation of the elevator speed data, and the coefficient of variation is the ratio of the standard deviation of the elevator speed data to the mean value of the elevator speed data. The higher the signal-to-noise ratio, the smaller the noise component in the data, and the higher the coefficient of variation, the more noise.

[0042] Further, the coefficient of variation and the signal-to-noise ratio are normalized, and the ratio of the normalized coefficient of variation to the signal-to-noise ratio is used as the noise level of the elevator data. Then, the initial K value is determined based on the noise level.

[0043] Specifically, the initial K value K can be determined using the following formula (1):

[0044] K = K min +(K max - K min ) × α (1)

[0045] Wherein, K max represents the preset maximum K value, K min represents the preset minimum K value, and α represents the noise level. Exemplarily, according to the setting of the K value when the KNN algorithm is usually used for elevator speed anomaly detection, K min can be 3, and K max can be 10.

[0046] It can be understood that in the KNN algorithm, the size of the K value directly affects the sensitivity and stability of the model. When there are many noise points in the data, in this embodiment, the K value is set to be small, which helps to improve the sensitivity of the model to local details. On the contrary, when the data is more complex or there are fewer noise points, in this embodiment, the K value is set to be large, so as to enhance the stability of the model and reduce the overreaction to outliers.

[0047] S3. Based on the elevator speed data and various influence data, and using the KNN algorithm, determine the distance set of each data point in the elevator speed data;

[0048] S4. Based on the distance set of each data point and the initial K value, determine the improved K value of each data point;

[0049] S5. Based on the elevator speed data, various influence data, and the improved K value of each data point, and using the KNN algorithm, determine the abnormal elevator speed.

[0050] Wherein, the distance set of each data point is the set of distances between each data point and all other data points.

[0051] In the specific implementation process, first, based on the elevator speed data and various influence data, and using the KNN algorithm, calculate the distance set of each data point in the elevator speed data.

[0052] It should be noted that since the abnormal elevator speed includes not only sudden changes but also periodic and persistent anomalies, and different types of anomalies are manifested differently in elevator speed data. For example, sudden anomalies only show data differences at specific mutation points, while periodic anomalies exhibit similarities with past data. Therefore, if the same K value is used for anomaly detection, when the K value is small, periodic anomalies may be misjudged as normal points; conversely, when the K value is large, sudden anomalies may be ignored.

[0053] Based on this, in this embodiment, based on the data differences in the distance sets of each data point, the K value is flexibly set for different data points to ensure that various abnormal types in the elevator speed, that is, whether it is a sudden anomaly, a persistent anomaly or a periodic anomaly, can be accurately detected, thus significantly improving the robustness of the detection.

[0054] In one embodiment, in step S4, determining the improved K value of each data point based on the distance set of each data point and the initial K value may specifically include:

[0055] S41. Determine the distribution degree of each data point based on the distance set of each data point;

[0056] S42. Determine the improved K value of each data point based on the distribution degree and the initial K value.

[0057] Among them, the distribution degree can represent the magnitude of the differences between the data in the distance set of each data point. For example, when a certain data point is a sudden anomaly point or a minor fluctuation point, there are differences between this data point and multiple adjacent data points, but the change range between these differences is small and the degree of difference is relatively consistent; while when the elevator has a periodic anomaly or a persistent anomaly, there are still differences between this data point and multiple adjacent data points, but the sizes of these differences are different, with some adjacent data points having larger differences and others having smaller differences.

[0058] In one embodiment, in step S41, determining the distribution degree of each data point based on the distance set of each data point may specifically include:

[0059] S411. Use a clustering algorithm to perform clustering processing on the distance set of each data point to obtain multiple clusters;

[0060] S412. Calculate the DTW distance between every two of the clusters;

[0061] S413. Determine the distribution degree of each data point based on multiple DTW distances.

[0062] In a specific implementation process, taking any data point as an example for illustration, clustering algorithms are used to cluster all the data in the distance set of this data point, resulting in multiple clusters. It can be understood that the clustering algorithm classifies similar data in the distance set into one cluster.

[0063] Further, calculate the DTW distances between pairwise clusters and sum up the multiple DTW distances to form the distribution degree of this data point. It should be noted that the DTW distance can measure the similarity between two sequences, and the number of elements between the two sequences can be different. Thus, when the DTW distance between two clusters is smaller, it indicates that the similarity between these two clusters is higher, that is, the difference between the data contained in the two clusters is smaller; conversely, when the DTW distance between two clusters is larger, it indicates that the similarity between these two clusters is lower, that is, the difference between the data contained in the two clusters is larger.

[0064] Further, linearly normalize the distribution degrees of each data point, and use the product of the initial K value and the distribution degree of each data point as the improved K value of each data point.

[0065] Thus, after obtaining the improved K values of each data point, the KNN algorithm can be used and based on the improved K values of each data point to perform anomaly detection on elevator speed data and various influencing data, obtaining multiple abnormal elevator speeds.

[0066] It can be understood that in the traditional KNN algorithm, the value of K is fixed, that is, the K value is the same for all data points. However, it is usually difficult to achieve a good balance between the sensitivity and robustness of the anomaly detection model by using a fixed K value. Specifically, when the K value is small, the model is prone to being overly sensitive to noise and local anomalies, while when the K value is large, some key local anomalies may be ignored, making it difficult to accurately capture the subtle fluctuations and sudden anomalies of elevator speed. Therefore, the strategy of fixing the K value cannot adapt to the diverse requirements in different scenarios, and thus has a negative impact on the accuracy and reliability of detection.

[0067] Based on this, in this embodiment, the initial K value is first set according to the noise level of the data. When there are many noise points in the data, the initial K value is set to be small, which can improve the sensitivity of the anomaly detection model to local details. Conversely, when the data is more complex or there are fewer noise points, the initial K value is set to be large to enhance the stability of the model and reduce the overreaction to outliers. On the basis of the initial K value, this embodiment further improves the initial K value according to the distribution degree of each data point to obtain the improved K value of each data point, that is, the adaptive K value, so as to accurately detect different types of anomalies in elevator speed, such as sudden anomalies, persistent anomalies, and periodic anomalies, and significantly improve the robustness of detection.

[0068] S6. Perform elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influence data.

[0069] It should be noted that the KNN algorithm can not only be used for anomaly detection but also for data prediction. Therefore, the KNN algorithm can be used in combination with the abnormal elevator speed obtained in the above steps to perform elevator fault warning.

[0070] In one embodiment, in step S6, performing elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influence data may specifically include:

[0071] S61. Determine multiple predicted elevator speeds based on the elevator speed data and various influence data using the KNN algorithm;

[0072] S62. Perform elevator fault warning based on the abnormal elevator speed and the multiple predicted elevator speeds.

[0073] In the specific implementation process, first, use the KNN algorithm to process the elevator speed data and various influence data to obtain multiple predicted elevator speeds.

[0074] Furthermore, determine the normal elevator speed range according to the abnormal elevator speed, that is, take the minimum abnormal elevator speed as the lower limit of the normal elevator speed range and the maximum abnormal elevator speed as the upper limit of the normal elevator speed range.

[0075] It can be understood that after obtaining the normal elevator speed range, the predicted elevator speeds can be compared with the normal elevator speed range to determine whether each predicted elevator speed is abnormal, and then perform elevator fault warning according to the determination result.

[0076] Furthermore, compare each predicted elevator speed with the normal elevator speed range. If the predicted elevator speed is not within the normal elevator speed range, trigger elevator fault warning; otherwise, no subsequent processing is performed.

[0077] In this embodiment, after triggering the elevator fault warning, technicians can analyze and confirm the cause of the elevator fault and decide whether maintenance or other measures are needed. At the same time, the fault warning event is recorded and archived to provide reference for subsequent data analysis and elevator maintenance, so as to improve the warning accuracy of the system and the safety of long-term elevator operation.

[0078] An elevator speed anomaly detection and fault warning method proposed by an embodiment of the present application comprehensively analyzes elevator speed data and multi-dimensional influencing data, and flexibly determines an improved K value for each data point by using the KNN algorithm, improving the accuracy and robustness of elevator speed anomaly detection. Compared with the method of fixing the K value in the traditional KNN algorithm, this embodiment can dynamically adjust the K value according to the distance set and data difference of each data point, making the model more sensitive when there is more noise and able to capture local anomalies; while when the data is more complex or there is less noise, the K value is larger, enhancing the stability of the model and reducing the over-response to outliers. This embodiment not only effectively avoids the overfitting problem caused by the traditional algorithm for elevator speed anomaly detection, but also ensures the accurate detection of different types of anomalies such as sudden anomalies, persistent anomalies, and periodic anomalies, thus improving the accuracy and reliability of elevator fault warning.

[0079] Based on the above embodiment, Figure 2 As shown in the structural block diagram of an elevator speed anomaly detection and fault warning device according to an embodiment of the present application, Figure 2 as shown, the elevator speed anomaly detection and fault warning device may include: a data acquisition module 210, an initial K value setting module 220, a model construction module 230, an improved K value setting module 240, an anomaly detection module 250, and a fault warning module 260, where,

[0080] The data acquisition module 210 is used to obtain elevator speed data and various influencing data;

[0081] The initial K value setting module 220 is used to determine an initial K value based on the elevator speed data;

[0082] The model construction module 230 is used to determine the distance set of each data point in the elevator speed data based on the elevator speed data and various influencing data and by using the KNN algorithm;

[0083] The improved K value setting module 240 is used to determine the improved K value of each data point based on the distance set of each data point and the initial K value;

[0084] The anomaly detection module 250 is used to determine the abnormal elevator speed based on the elevator speed data, various influencing data, and the improved K value of each data point and by using the KNN algorithm;

[0085] The fault warning module 260 is used to perform elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influencing data.

[0086] In an exemplary embodiment, the initial K value setting module 220 may also be used to calculate the noise degree of the elevator speed data; determine the initial K value based on the noise degree.

[0087] In an exemplary embodiment, the initial K-value setting module 220 may also determine the initial K-value using the following formula (1):

[0088] K = K min +(K max -K min )×α (1)

[0089] Where K represents the initial K-value, K max represents the preset maximum K-value, K min represents the preset minimum K-value, and α represents the degree of noise.

[0090] In an exemplary embodiment, the improved K-value setting module 240 may also be used to determine the distribution degree of each data point based on the distance set of each data point; and determine the improved K-value of each data point based on the distribution degree and the initial K-value.

[0091] In an exemplary embodiment, the improved K-value setting module 240 may also be used to perform clustering processing on the distance set of each data point using a clustering algorithm to obtain multiple clusters; calculate the DTW distance between every two of the clusters; and determine the distribution degree of each data point based on multiple DTW distances

[0092] In an exemplary embodiment, the fault warning module 260 may also be used to determine multiple predicted elevator speeds based on the elevator speed data and multiple types of the influence data and using the KNN algorithm; and perform elevator fault warning based on the abnormal elevator speed and the multiple predicted elevator speeds.

[0093] In an exemplary embodiment, the initial K-value setting module 220 may also be used to calculate the signal-to-noise ratio and the coefficient of variation of the elevator speed data; and determine the degree of noise of the elevator speed data based on the signal-to-noise ratio and the coefficient of variation.

[0094] Those skilled in the art should understand that the division of each module in the embodiment is only a logical function division. In actual application, it can be fully or partially integrated into one or more actual carriers, and these modules can all be implemented in the form of software called by a processing unit, or all be implemented in the form of hardware, or be implemented in a form combining software and hardware. It should be noted that each module in an elevator speed anomaly detection and fault warning device in this embodiment corresponds one by one to each step in an elevator speed anomaly detection and fault warning method in the foregoing embodiment. Therefore, the specific implementation manner of this embodiment may refer to the implementation manner of the foregoing elevator speed anomaly detection and fault warning method, which will not be elaborated here.

[0095] Based on the above embodiments, Figure 3As shown in the structural schematic diagram of an elevator speed anomaly detection and fault warning device according to an embodiment of the present application, Figure 3 as shown, the elevator speed anomaly detection and fault warning device may include: a processor 310, a communications interface 320, a memory 330, and a communication bus 340. Among them, the processor 310, the communications interface 320, and the memory 330 complete mutual communication through the communication bus 340. The processor 310 may call logical instructions in the memory 330 to execute an elevator speed anomaly detection and fault warning method, and the method includes: obtaining elevator speed data and various influence data; determining an initial K value based on the elevator speed data; determining a distance set of each data point in the elevator speed data based on the elevator speed data and various influence data and using the KNN algorithm; determining an improved K value for each data point based on the distance set of each data point and the initial K value; determining an abnormal elevator speed based on the elevator speed data, various influence data, and the improved K value of each data point and using the KNN algorithm; and performing elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influence data.

[0096] In addition, when the logical instructions in the above-mentioned memory 330 are implemented in the form of software function units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0097] On the basis of the above embodiments, on the other hand, the present invention further provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the elevator speed anomaly detection and fault warning method provided by the above-mentioned various methods. The method includes: obtaining elevator speed data and various influencing data; determining an initial K value based on the elevator speed data; determining a distance set of each data point in the elevator speed data based on the elevator speed data and various influencing data and using the KNN algorithm; determining an improved K value for each data point based on the distance set of each data point and the initial K value; determining an abnormal elevator speed based on the elevator speed data, various influencing data, and the improved K value of each data point and using the KNN algorithm; and performing elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influencing data.

[0098] On the basis of the above embodiments, on another aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the elevator speed anomaly detection and fault warning method provided by the above-mentioned various methods. The method includes: obtaining elevator speed data and various influencing data; determining an initial K value based on the elevator speed data; determining a distance set of each data point in the elevator speed data based on the elevator speed data and various influencing data and using the KNN algorithm; determining an improved K value for each data point based on the distance set of each data point and the initial K value; determining an abnormal elevator speed based on the elevator speed data, various influencing data, and the improved K value of each data point and using the KNN algorithm; and performing elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influencing data.

[0099] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be included in the patent protection scope of the present application by the same token.

Claims

1. An elevator speed anomaly detection and fault warning method, characterized in that Including: Obtain elevator speed data and various influencing data; Determine an initial K value based on the elevator speed data; Determine a distance set of each data point in the elevator speed data based on the elevator speed data and various influencing data and using the KNN algorithm; Determine an improved K value for each data point based on the distance set of each data point and the initial K value; Determine an abnormal elevator speed based on the elevator speed data, various influencing data, and the improved K value of each data point and using the KNN algorithm; Perform elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influencing data.

2. The elevator speed anomaly detection and fault warning method according to claim 1, wherein The determining the initial K value based on the elevator speed data includes: Calculate the noise level of the elevator speed data; Determine the initial K value based on the noise level.

3. The elevator speed anomaly detection and fault warning method according to claim 2, characterized in that, The determining the initial K value based on the noise level includes: Use the following formula (1) to determine the initial K value: K = K min +(K max -K min )×α (1) Among them, K represents the initial K value, and K max represents the preset maximum K value, and K min represents the preset minimum K value, and α represents the degree of noise.

4. The elevator speed anomaly detection and fault warning method according to claim 1, characterized in that, The determining the improved K value for each data point based on the distance set of each data point and the initial K value includes: Determine the distribution degree of each data point based on the distance set of each data point; Determine the improved K value for each data point based on the distribution degree and the initial K value.

5. The elevator speed abnormal detection and fault warning method according to claim 4, characterized in that, The determining the distribution degree of each data point based on the distance set of each data point includes: Use a clustering algorithm to perform clustering processing on the distance set of each data point to obtain multiple clusters; Calculate the DTW distance between every two clusters; Determine the distribution degree of each data point based on multiple DTW distances.

6. The elevator speed anomaly detection and fault warning method according to claim 1, characterized in that The performing elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influencing data includes: Determine multiple predicted elevator speeds based on the elevator speed data and various influencing data and using the KNN algorithm; Perform elevator fault warning based on the abnormal elevator speed and multiple predicted elevator speeds.

7. The elevator speed anomaly detection and fault warning method according to claim 2, characterized in that, The calculating the noise level of the elevator speed data includes: Calculate the signal-to-noise ratio and coefficient of variation of the elevator speed data; Determine the noise level of the elevator speed data based on the signal-to-noise ratio and the coefficient of variation.

8. An elevator speed anomaly detection and fault warning device, characterized in that, Including: A data acquisition module for obtaining elevator speed data and various influencing data; An initial K value setting module for determining an initial K value based on the elevator speed data; A model construction module for determining a distance set of each data point in the elevator speed data based on the elevator speed data and various influencing data and using the KNN algorithm; An improved K value setting module for determining an improved K value for each data point based on the distance set of each data point and the initial K value; An anomaly detection module for determining an abnormal elevator speed based on the elevator speed data, various influencing data, and the improved K value of each data point and using the KNN algorithm; A fault warning module for performing elevator fault warning based on the abnormal elevator speed, the elevator speed data, and various influencing data.

9. An elevator speed anomaly detection and fault warning device, characterized in that Including: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the elevator speed anomaly detection and fault warning method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the elevator speed anomaly detection and fault warning method according to any one of claims 1 to 7.

Images