Characteristic value selection method, detection method and system for detecting lifting equipment
Through the feature value selection method and gradient enhancement tree algorithm, the feature value screening and threshold applicability problems in elevator fault diagnosis are solved, and accurate detection and fault warning of elevator operating status are realized, and it is suitable for a variety of equipment types.
Patent Information
- Application Number
- CN202110276763.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-03-15
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2041-03-15
AI Technical Summary
In the diagnosis of elevator faults, there is a detrend processing that cannot completely remove the impact of acceleration and deceleration processes. The lack of characteristic value screening steps and cannot objectively and accurately reflect the health of elevator operation. The threshold is derived from a single elevator type and is not suitable for different elevator types. multiplex.
The eigenvalue selection method is used to obtain the individual and general eigenvalue thresholds of each device type, select the eigenvalue combination and threshold through the comprehensive accuracy index, feature extraction and zero-drifting based on vibration sensor data, and contribute using the gradient boost tree algorithm. Rate sorting to obtain detection criteria suitable for multiple device types.
It realizes accurate reflection of the operating status of the elevator and provides unified detection standards. It can detect problems in advance before the elevator car fails, avoiding the occurrence of shutdown failures. It is suitable for a variety of equipment types.
Smart Images

Figure CN113269222B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of state monitoring and diagnosis of fixed lifting equipment, and in particular to a characteristic value selection method, a detection method and a system for detecting lifting equipment. Background Art
[0002] With the increase in high-rise buildings, elevators are becoming an indispensable fixed lifting equipment. After years of development, the number of elevators in my country has increased significantly and is becoming saturated. As the elevator runs longer, the probability of failure increases significantly. Elevator failures are prone to serious consequences, but there are many types of elevator failures, and the equipment parts that cause failures are different. After the elevator fails, it is time-consuming and laborious for maintenance personnel to conduct on-site inspections, resulting in huge waste of resources. Therefore, fault diagnosis and on-demand maintenance of elevator equipment, and taking measures to avoid failures have become issues that need to be faced in theoretical research and practical engineering.
[0003] At present, the health status assessment of elevator equipment is mainly based on the car acceleration signal and threshold division. The specific implementation method is as follows:
[0004] (1) Selection of normal elevator operation data: The acceleration signal corresponding to the operation process without any fault label is selected from the elevator history database, and the selected acceleration signal is preprocessed by outlier cleaning and other preprocessing operations to obtain the preprocessed acceleration signal a(t); where t is the time parameter;
[0005] (2) Acceleration signal denoising: The preprocessed acceleration signal a(t) is denoised using a denoising method based on the wavelet modulus maximum.
[0006] (3) Vibration signal extraction: Although the elevator acceleration signal after noise reduction has removed a large amount of noise components, it still retains the initial acceleration process and the deceleration process inherent in the elevator operation process. In order to prevent the significant changes in acceleration caused by these two processes from affecting subsequent analysis, it is necessary to detrend the acceleration signal and further extract the vibration signal that is closely related to the elevator operation performance;
[0007] (4) Establishment of the elevator operation health evaluation model, which mainly includes determining the evaluation index, calculating the operation health evaluation index value, determining the evaluation index threshold, determining the evaluation function, and obtaining the elevator operation health score;
[0008] (5) When applied online, the acceleration data of the operation process is collected in real time and an operation health evaluation is given.
[0009] The above evaluation method has the following main disadvantages:
[0010] (1) After using the wavelet modulus maximum method to reduce noise, the data is detrended to eliminate the impact of the acceleration and deceleration processes during the elevator operation. This detrending method cannot completely eliminate the impact of the acceleration and deceleration processes during the elevator operation, which will have a negative impact on subsequent modeling;
[0011] (2) Based on experience, three characteristic values, namely peak value Xpv, root mean square Xrms, and waveform factor Sf, are directly selected as evaluation indicators of the health model. There is no step for screening the characteristic values, and it is impossible to objectively and accurately reflect the operation health of the elevator;
[0012] (3) When applied online, the health score is calculated using the historical threshold determined during the modeling process. However, the establishment of this threshold is based only on the historical operation data of a single ladder type, which makes it difficult to reuse this method for different ladder types in practical applications.
[0013] In summary, detrending the data cannot completely remove the impact of the acceleration and deceleration process, which will have a negative impact on subsequent modeling. There is a lack of steps to screen the eigenvalues, and it is impossible to objectively and accurately reflect the health of the elevator operation. The threshold value comes from a single ladder type and is not suitable for reuse of other different ladder types in practical applications. Summary of the invention
[0014] The technical problem to be solved by the present invention is to overcome the defects that the detrending processing of data in the prior art cannot completely remove the influence of the acceleration and deceleration process, which will have a negative impact on subsequent modeling, lacks the step of screening the characteristic values, cannot objectively and accurately reflect the health of the elevator operation, and the threshold value is derived from a single ladder type, which is not suitable for reuse of other different ladder types in practical applications. A characteristic value selection method, a detection method and a system for detecting lifting equipment are provided.
[0015] The present invention solves the above technical problems through the following technical solutions:
[0016] The present invention provides a method for selecting characteristic values for detecting the operating state of a fixed lifting device. The method is for fixed lifting devices of multiple device types, and the method comprises:
[0017] Obtain the type and value of the characteristic value of each device type in the normal state of the device;
[0018] For each individual characteristic value type of each device type, obtaining an individual upper threshold value and an individual lower threshold value for characterizing that a device of the device type is in a normal state;
[0019] For each device type, performing a health assessment on devices of the device type based on the individual upper threshold and the individual lower threshold to obtain an individual accuracy result;
[0020] Obtain a common characteristic value type according to each individual characteristic value type of each device type;
[0021] Obtaining a universal upper threshold and a universal lower threshold of each universal characteristic value type, wherein the universal upper threshold and the universal lower threshold are used to indicate that devices of all device types are in a normal state;
[0022] For each device type, performing health assessment on devices of the device type based on the general upper threshold and the general lower threshold to obtain a general accuracy result;
[0023] Obtaining a comprehensive accuracy index based on the individual accuracy result and the general accuracy result, wherein a value of the comprehensive accuracy index is between the individual accuracy result and the general accuracy result;
[0024] A combination of eigenvalues is selected, and the comprehensive accuracy index is used as the objective function to obtain a combination of eigenvalues that meets the requirements of the comprehensive accuracy index and a threshold value corresponding to each eigenvalue.
[0025] Preferably, the selection method further comprises:
[0026] The feature value combination includes a separate feature value combination for each device type and a common feature value combination;
[0027] and / or,
[0028] For each individual characteristic value type of each device type, obtaining an individual upper threshold value and an individual lower threshold value according to a statistically based data selection criterion;
[0029] and / or,
[0030] The step of obtaining the type and value of the characteristic value of each device type in the normal state of the device specifically includes:
[0031] For each equipment type, obtaining vibration data generated by the fixed lifting equipment during operation through a vibration sensor;
[0032] Acquire the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to acquire classified vibration data;
[0033] Acquire target data, wherein the target data is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting device, and the uniform speed segment data is vibration data when the fixed lifting device is in a uniform speed operating state;
[0034] Performing feature extraction on the target data to obtain the type and value of the specified feature value;
[0035] and / or,
[0036] The step of obtaining the type and value of the characteristic value of each device type in the normal state of the device specifically includes:
[0037] For each equipment type, the vibration data generated by the fixed lifting equipment during operation is obtained through a vibration sensor, and the vibration data is normal vibration data of a complete cycle, wherein the fixed lifting equipment is a complete cycle from docking at the starting floor to running and docking at the target floor;
[0038] Acquire the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to acquire classified vibration data;
[0039] Acquire target data, wherein the target data is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting device, and the uniform speed segment data is vibration data when the fixed lifting device is in a uniform speed operating state;
[0040] Extracting static segment data from the vibration data, wherein the static segment data is vibration data when the fixed lifting device is in a static state;
[0041] Acquire a zero drift value of a vibration sensor based on the static segment data, wherein the target data is specifically zero drift-removed data obtained after removing the zero drift value of the vibration sensor from the uniform speed segment data;
[0042] Performing feature extraction on the target data to obtain the type and value of the specified feature value;
[0043] The feature values are sorted by contribution rate using a gradient boosting tree algorithm, and the first m principal component features whose cumulative contribution rates are greater than a predetermined percentage threshold are extracted to form a data set, where m is a positive integer.
[0044] Preferably, the characteristic value under normal equipment status is expressed by the following formula:
[0045] x i0 =(f i1 ,…,f i j,…,f im );
[0046] Where m represents the total number of eigenvalues, f ij The unit characteristic value of type j represents the characteristic value of device type i, where I represents the set of device types to be analyzed. J represents the set of eigenvalue types to be selected,
[0047] The single upper threshold is denoted as UT ij , the single lower threshold is denoted as LT ij ;
[0048] The general upper threshold is expressed by the following formula:
[0049]
[0050] The general lower threshold is expressed by the following formula:
[0051]
[0052] The individual accuracy result is expressed as A iJ , the general accuracy result is denoted as UA iJ , the comprehensive accuracy index is expressed by the following formula:
[0053]
[0054] The eigenvalue combination is represented by J * , the threshold value of each eigenvalue of the eigenvalue combination is expressed by the following formula:
[0055] {UT ij ,LT ij ,UT Ij ,LT Ij |i∈I,j∈J *}.
[0056] Preferably, the step of obtaining a combination of feature values that meets the comprehensive accuracy index requirement and a threshold value corresponding to each feature value specifically includes:
[0057] Obtain the eigenvalue combination that meets the comprehensive accuracy index requirements and corresponds to the maximum accuracy result and the threshold value corresponding to each eigenvalue.
[0058] The present invention also provides a detection method for a fixed lifting device, the detection method comprising:
[0059] Acquire the data to be detected of the fixed lifting device to be detected, wherein the data to be detected is the data obtained after processing the fixed lifting device to be detected by using the aforementioned method for selecting characteristic values for detecting the operating state of the fixed lifting device;
[0060] Extracting the combination data to be detected corresponding to the characteristic value combination from the data to be detected;
[0061] The combination data to be detected is compared with the threshold value corresponding to the characteristic value combination. If the combination data to be detected is within the threshold interval of the corresponding equipment type in the threshold value, the fixed lifting equipment is in a healthy state; otherwise, the fixed lifting equipment is in a fault state.
[0062] Preferably, the step of comparing the combination data to be detected with the threshold value corresponding to the characteristic value combination specifically includes:
[0063] If the equipment type of the fixed lifting equipment to be detected has a corresponding individual characteristic value combination and threshold, then comparing the combination data to be detected with the individual characteristic value combination and threshold;
[0064] If the equipment type to which the fixed lifting equipment to be detected belongs does not have a corresponding individual characteristic value combination and threshold value, the combination data to be detected is compared with the universal characteristic value combination and threshold value.
[0065] The present invention also provides a system for selecting characteristic values for detecting the operating status of a fixed lifting device, the selection system is for fixed lifting devices of multiple equipment types, and the selection system comprises: a characteristic value acquisition module, a separate threshold acquisition module, a separate accuracy acquisition module, a general characteristic value acquisition module, a general threshold acquisition module, a general accuracy acquisition module, a comprehensive accuracy acquisition module and a characteristic value selection module;
[0066] The characteristic value acquisition module is used to obtain the type and value of the characteristic value of each device type when the device is in a normal state;
[0067] The separate threshold value acquisition module is used to acquire, for each separate characteristic value type of each device type, a separate upper threshold value and a separate lower threshold value for characterizing that the device of the device type is in a normal state;
[0068] The individual accuracy acquisition module is used for performing health assessment on devices of the device type based on the individual upper threshold and the individual lower threshold for each device type to obtain an individual accuracy result;
[0069] The universal feature value acquisition module is used to acquire the universal feature value type according to each individual feature value type of each device type;
[0070] The universal threshold acquisition module is used to acquire a universal upper threshold and a universal lower threshold of each universal feature value type, wherein the universal upper threshold and the universal lower threshold are used to indicate that devices of all device types are in a normal state;
[0071] The general accuracy acquisition module is used for performing health assessment on devices of each device type based on the general upper threshold and the general lower threshold to obtain a general accuracy result;
[0072] The comprehensive accuracy acquisition module is used to acquire a comprehensive accuracy index based on the individual accuracy result and the general accuracy result, and the value of the comprehensive accuracy index is between the individual accuracy result and the general accuracy result;
[0073] The feature value selection module is used to select a combination of feature values, take the comprehensive accuracy index as the objective function, and obtain a combination of feature values that meets the requirements of the comprehensive accuracy index and a threshold value corresponding to each feature value.
[0074] Preferably, the feature value combination includes a separate feature value combination for each device type and a common feature value combination;
[0075] and / or,
[0076] The separate threshold value acquisition module is used to acquire a separate upper threshold value and a separate lower threshold value for each separate characteristic value type of each device type according to a statistically based data selection criterion;
[0077] and / or,
[0078] The feature value acquisition module includes: a vibration data acquisition unit, a vibration data classification unit, a target data acquisition unit and a feature extraction unit;
[0079] The vibration data acquisition unit is used to acquire vibration data generated by the fixed lifting equipment during operation through a vibration sensor for each equipment type;
[0080] The vibration data classification unit is used to obtain the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to obtain classified vibration data;
[0081] The target data acquisition unit is used to acquire target data, wherein the target data is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting device, and the uniform speed segment data is vibration data when the fixed lifting device is in a uniform speed operating state;
[0082] The feature extraction unit is used to perform feature extraction on the target data to obtain the type and value of the specified feature value;
[0083] and / or,
[0084] The characteristic value acquisition module includes: a vibration data acquisition unit, a vibration data classification unit, a target data acquisition unit, a static segment data extraction unit, a zero drift value acquisition unit, a characteristic extraction unit and a contribution rate sorting unit;
[0085] The vibration data acquisition unit is used to acquire, for each equipment type, vibration data generated by the fixed lifting equipment during operation through a vibration sensor, wherein the vibration data is normal vibration data of a complete cycle, wherein a complete cycle is when the fixed lifting equipment stops at a starting floor and then runs and stops at a target floor;
[0086] The vibration data classification unit is used to obtain the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to obtain classified vibration data;
[0087] The target data acquisition unit is used to acquire target data, wherein the target data is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting device, and the uniform speed segment data is vibration data when the fixed lifting device is in a uniform speed operating state;
[0088] The static segment data extraction unit is used to extract static segment data from the vibration data, wherein the static segment data is vibration data when the fixed lifting device is in a static state;
[0089] The zero drift value acquisition unit is used to acquire the zero drift value of the vibration sensor based on the static segment data, and the target data is specifically the zero drift-removed data acquired after removing the zero drift value of the vibration sensor from the uniform speed segment data;
[0090] The feature extraction unit is used to perform feature extraction on the target data to obtain the type and value of the specified feature value;
[0091] The contribution rate sorting unit is used to sort the contribution rates of the features using a gradient boosting tree algorithm on the feature values, and extract the first m principal component features whose cumulative contribution rates are greater than a predetermined percentage threshold to form a data set, where m is a positive integer.
[0092] Preferably, the characteristic value under normal equipment status is expressed by the following formula:
[0093] x i0 =(f i1 ,…,f i j,…,f im );
[0094] Where m represents the total number of eigenvalues, f ijThe unit characteristic value of type j represents the characteristic value of device type i, where I represents the set of device types to be analyzed. J represents the set of eigenvalue types to be selected,
[0095] The single upper threshold is denoted as UT ij , the single lower threshold is denoted as LT ij ;
[0096] The general upper threshold is expressed by the following formula:
[0097]
[0098] The general lower threshold is expressed by the following formula:
[0099]
[0100] The individual accuracy result is expressed as A iJ , the general accuracy result is denoted as UA iJ , the comprehensive accuracy index is expressed by the following formula:
[0101]
[0102] The eigenvalue combination is represented by J * , the threshold value of each eigenvalue of the eigenvalue combination is expressed by the following formula:
[0103] {UT ij ,LT ij ,UT Ij ,LT Ij |i∈I,j∈J *}.
[0104] Preferably, the eigenvalue selection module is used to obtain the eigenvalue combination that meets the comprehensive accuracy index requirements and corresponds to the maximum accuracy result, as well as the threshold value corresponding to each eigenvalue.
[0105] The present invention also provides a detection system for fixed lifting equipment, the detection system comprising: a to-be-detected data acquisition module, a to-be-detected combined data extraction module and a comparison module;
[0106] The to-be-detected data acquisition module is used to acquire the to-be-detected data of the fixed lifting device to be detected, wherein the to-be-detected data is data obtained after processing the fixed lifting device to be detected by using the aforementioned feature value selection system for detecting the operating state of the fixed lifting device;
[0107] The to-be-detected combined data extraction module is used to extract the to-be-detected combined data corresponding to the characteristic value combination from the to-be-detected data;
[0108] The comparison module is used to compare the combination data to be detected with the threshold value corresponding to the characteristic value combination. If the combination data to be detected is within the threshold interval of the corresponding equipment type in the threshold value, the fixed lifting equipment is in a healthy state; otherwise, the fixed lifting equipment is in a fault state.
[0109] Preferably, if the equipment type to which the fixed lifting equipment to be detected belongs has a corresponding individual characteristic value combination and threshold value, the comparison module compares the combination data to be detected with the individual characteristic value combination and threshold value;
[0110] If the equipment type to which the fixed lifting equipment to be detected belongs does not have a corresponding separate feature value combination and threshold, the comparison module compares the combination data to be detected with the general feature value combination and threshold.
[0111] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the processor implements the aforementioned method for selecting characteristic values for detecting the operating status of a fixed lifting device, or implements the aforementioned method for detecting the operating status of a fixed lifting device.
[0112] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program implements the steps of the aforementioned method for selecting characteristic values for detecting the operating status of a fixed lifting device, or implements the steps of the aforementioned method for detecting the operating status of a fixed lifting device.
[0113] The positive and progressive effect of the present invention is that accurate data is obtained by removing zero drift and extracting features from the uniform speed segment data that can accurately reflect the operating status of the fixed lifting equipment, and the accuracy results of each equipment type are obtained based on the accurate data. Not only the accuracy results are provided for a single equipment type, but also the universal accuracy results are calculated by integrating all equipment types, and then a comprehensive accuracy index applicable to all equipment types is obtained, and a unified detection standard is established for equipment of all equipment types. The thresholds corresponding to the characteristic value combinations make the evaluation standards more universal and more effective, and can objectively and accurately reflect the operating health of the elevator. It is applicable to a variety of equipment types and can detect abnormal conditions before a shutdown failure occurs in the elevator car, helping maintenance personnel to discover problems in advance and avoid shutdown failures. BRIEF DESCRIPTION OF THE DRAWINGS
[0114] Figure 1This is a flow chart of a method for selecting characteristic values for detecting the operating status of a fixed lifting device according to embodiment 1 of the present invention.
[0115] Figure 2 This is a flowchart of a specific implementation of step S11 in the method for selecting characteristic values for detecting the operating status of a fixed lifting device according to embodiment 1 of the present invention.
[0116] Figure 3 This is a flowchart of another specific implementation of step S11 in the method for selecting characteristic values for detecting the operating status of a fixed lifting device according to embodiment 1 of the present invention.
[0117] Figure 4 This is a flow chart of a detection method for a fixed lifting device according to Embodiment 2 of the present invention.
[0118] Figure 5 This is a flowchart of a specific implementation of step S23 in the detection method for fixed lifting equipment according to embodiment 2 of the present invention.
[0119] Figure 6 This is a module diagram of a system for selecting characteristic values for detecting the operating status of a fixed lifting device according to Embodiment 3 of the present invention.
[0120] Figure 7 It is a module diagram of a specific implementation of the feature value acquisition module 101 in the feature value selection system for detecting the operating status of a fixed lifting device according to Example 3 of the present invention.
[0121] Figure 8 This is a module diagram of another specific implementation of the feature value acquisition module 101 in the feature value selection system for detecting the operating status of a fixed lifting device according to Example 3 of the present invention.
[0122] Fig. 9 This is a module schematic diagram of a detection system for a fixed lifting device according to Embodiment 4 of the present invention.
[0123] Fig.10 This is a schematic diagram of the structure of an electronic device according to Embodiment 5 of the present invention. DETAILED DESCRIPTION
[0124] The present invention is further described below by way of examples, but the present invention is not limited to the scope of the examples.
[0125] Example 1
[0126] This embodiment provides a method for selecting characteristic values for detecting the operating status of a fixed lifting device. The selection method is for fixed lifting devices of multiple device types, wherein the fixed lifting device may be an elevator car. Figure 1 , the selection methods include:
[0127] S11, obtaining the type and value of the characteristic value of each device type in the normal state of the device, wherein the characteristic value type includes peak value, kurtosis, etc.
[0128] S12. For each individual characteristic value type of each device type, obtain an individual upper threshold value and an individual lower threshold value for characterizing that a device of the device type is in a normal state.
[0129] S13. For each device type, perform a health assessment on the devices of the device type based on a separate upper threshold and a separate lower threshold to obtain a separate accuracy result.
[0130] S14. Obtain a common feature value type according to each individual feature value type of each device type.
[0131] S15. Obtain a universal upper threshold and a universal lower threshold of each universal feature value type, wherein the universal upper threshold and the universal lower threshold are used to indicate that all device types are in a normal state. When the value is between the universal upper threshold and the universal lower threshold (including the case where the value is equal to the universal upper threshold or the universal lower threshold), it indicates that the device is in a normal state.
[0132] S16. For each device type, perform a health assessment on the devices of the device type based on a common upper threshold and a common lower threshold to obtain a common accuracy result.
[0133] S17, obtaining a comprehensive accuracy index based on the individual accuracy result and the general accuracy result, wherein the value of the comprehensive accuracy index is between the individual accuracy result and the general accuracy result. The value of the comprehensive accuracy index may be equal to the individual accuracy result or the general accuracy result.
[0134] S18, select a combination of eigenvalues, take the comprehensive accuracy index as the objective function, and obtain a combination of eigenvalues that meets the requirements of the comprehensive accuracy index and a threshold value corresponding to each eigenvalue. Specifically, a particle swarm algorithm is used for selection.
[0135] During specific implementation, the selection method also includes:
[0136] The feature value combination includes a separate feature value combination for each device type and a common feature value combination. The common feature value combination has the common feature value type in the separate feature value combination of each device type. For example, there are three device types in total. The separate feature value combination of device type 1 is A, B and C, the separate feature value combination of device type 2 is B, C and E, and the separate feature value combination of device type 3 is A, B, C and F. Then the common feature value combination is B and C.
[0137] and / or,
[0138] Step S12 specifically includes:
[0139] S121. For each individual characteristic value type of each device type, obtain an individual upper threshold and an individual lower threshold according to a statistically based data selection criterion. Preferably, the statistically based data selection criterion may be a 3σ criterion or a 2σ criterion.
[0140] and / or,
[0141] Reference Figure 2 , step S11 specifically includes:
[0142] S1101. For each equipment type, obtain vibration data generated by the fixed lifting equipment during operation through a vibration sensor, wherein the vibration data includes vibration acceleration data in the directions of the X-axis, Y-axis and Z-axis generated by the elevator car during operation.
[0143] S1102: Acquire the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to acquire classified vibration data.
[0144] S1103, obtaining target data, the target data is the uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting equipment, and the uniform speed segment data is the vibration data of the fixed lifting equipment when it is in a uniform speed operating state. Among them, the operating state of the fixed lifting equipment includes an upward operation acceleration segment, an upward operation uniform speed segment, an upward operation deceleration segment, a downward operation acceleration segment, a downward operation uniform speed segment, a downward operation deceleration segment and a stationary segment at a floor stop, and the uniform speed segment data includes the vibration data corresponding to the upward operation uniform speed segment and the downward operation uniform speed segment. Compared with other segment data, the uniform speed segment data can more accurately reflect the operating state of the fixed lifting equipment, improve the quality of the data, and improve the accuracy of the data.
[0145] S1104. Perform feature extraction on the target data to obtain the type and value of the specified feature value.
[0146] and / or,
[0147] Reference Figure 3 , step S11 specifically includes:
[0148] S1111. For each equipment type, the vibration data generated by the fixed lifting equipment during operation is obtained through a vibration sensor. The vibration data is normal vibration data of a complete cycle, wherein the fixed lifting equipment is a complete cycle from docking at the starting floor to running and docking at the target floor. The normal vibration data is data that can truthfully reflect the operating status of the fixed lifting equipment. The vibration data includes vibration acceleration data in the X-axis, Y-axis and Z-axis directions generated by the elevator car during operation. In the specific implementation process, the original vibration data can be removed by removing noise data, large deviation data and incomplete cycle data to obtain normal vibration data of a complete cycle. Noise data is meaningless data, and incomplete cycle data is data that cannot be identified as belonging to a complete cycle. Those skilled in the art can set the judgment conditions and parameters of large deviation data according to actual needs.
[0149] S1112. Obtain the device status corresponding to the vibration data, and classify the vibration data according to the normal status of the device and the type of device failure to obtain classified vibration data.
[0150] S1113, obtaining target data, the target data is the uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting equipment, and the uniform speed segment data is the vibration data of the fixed lifting equipment when it is in a uniform speed operating state. Among them, the operating state of the fixed lifting equipment includes an upward operation acceleration segment, an upward operation uniform speed segment, an upward operation deceleration segment, a downward operation acceleration segment, a downward operation uniform speed segment, a downward operation deceleration segment and a stationary segment at a floor stop, and the uniform speed segment data includes the vibration data corresponding to the upward operation uniform speed segment and the downward operation uniform speed segment. Compared with other segment data, the uniform speed segment data can more accurately reflect the operating state of the fixed lifting equipment, improve the quality of the data, and improve the accuracy of the data.
[0151] S1114. Extract static segment data from the vibration data, where the static segment data is vibration data when the fixed lifting equipment is in a static state.
[0152] S1115. Obtain the zero drift value of the vibration sensor based on the static segment data, wherein the target data is specifically the zero drift-removed data obtained after removing the zero drift value of the vibration sensor from the uniform speed segment data. Removing the zero drift value of the vibration sensor from the uniform speed segment data can eliminate the influence of different elevator types, different fault types, and different vibration sensor installation positions on the original vibration data.
[0153] S1116. Perform feature extraction on the target data to obtain the type and value of the specified feature value.
[0154] S1117, using the gradient boosting tree algorithm to sort the contribution rates of the features, extracting the first m principal component features whose cumulative contribution rates are greater than a predetermined percentage threshold to form a data set, where m is a positive integer. The predetermined percentage threshold can be set according to actual needs.
[0155] In specific implementation, the characteristic value under normal equipment status is expressed by the following formula:
[0156] x i0 =(f i1 ,…,f i j,…,f im ).
[0157] Where m represents the total number of eigenvalues, f ij The unit characteristic value of type j represents the characteristic value of device type i, where I represents the set of device types to be analyzed. J represents the set of eigenvalue types to be selected,
[0158] The upper threshold alone is denoted as UT ij , the lower threshold is denoted as LT ij .
[0159] The general upper threshold is expressed by the following formula:
[0160]
[0161] The general lower bound threshold is expressed by the following formula:
[0162]
[0163] The individual accuracy result is expressed as A iJ , the general accuracy result is expressed as UA iJ , the comprehensive accuracy index is expressed by the following formula:
[0164]
[0165] The eigenvalue combination is denoted as J * , the threshold of each eigenvalue of the eigenvalue combination is expressed by the following formula:
[0166] {UT ij ,LT ij ,UT Ij ,LT Ij |i∈I,j∈J *}.
[0167] In specific implementation, step S18 specifically includes:
[0168] S181. Select a combination of eigenvalues, use the comprehensive accuracy index as the objective function, obtain the eigenvalue combination that meets the requirements of the comprehensive accuracy index and corresponds to the maximum accuracy result, and the threshold value corresponding to each eigenvalue.
[0169] This embodiment obtains precise data by removing zero drift and extracting features from the uniform speed segment data that can accurately reflect the operating status of the fixed lifting equipment, and obtains the accuracy results of each equipment type based on the precise data. It not only provides accuracy results for a single equipment type, but also integrates all equipment types to calculate a universal accuracy result, and then obtains a comprehensive accuracy index applicable to all equipment types, establishes a unified detection standard for equipment of all equipment types, and the thresholds corresponding to the characteristic value combinations make the evaluation standard more universal and more effective, and can objectively and accurately reflect the operating health of the elevator, and is applicable to a variety of equipment types.
[0170] Example 2
[0171] This embodiment provides a detection method for a fixed lifting device. Figure 4 , detection methods include:
[0172] S21. Acquire the data to be detected of the fixed lifting device to be detected, where the data to be detected is data obtained after processing the fixed lifting device to be detected using the aforementioned method for selecting characteristic values for detecting the operating status of the fixed lifting device.
[0173] S22, extracting the combination data to be detected corresponding to the characteristic value combination from the data to be detected.
[0174] S23. Compare the combination data to be detected with the threshold value corresponding to the characteristic value combination. If the combination data to be detected is within the threshold interval of the corresponding equipment type in the threshold value, the fixed lifting equipment is in a healthy state; otherwise, the fixed lifting equipment is in a fault state.
[0175] When implementing it, refer to Figure 5 , step S23 specifically includes:
[0176] S231. If the equipment type of the fixed lifting equipment to be detected has a corresponding individual characteristic value combination and threshold, the combination data to be detected is compared with the individual characteristic value combination and the threshold. If the combination data to be detected is within the threshold range of the corresponding equipment type in the threshold, the fixed lifting equipment is in a healthy state; otherwise, the fixed lifting equipment is in a faulty state.
[0177] S232. If the equipment type of the fixed lifting equipment to be detected does not have a corresponding individual characteristic value combination and threshold, the combination data to be detected is compared with the general characteristic value combination and threshold. If the combination data to be detected is within the threshold range of the corresponding equipment type in the threshold, the fixed lifting equipment is in a healthy state; otherwise, the fixed lifting equipment is in a faulty state.
[0178] This embodiment detects fixed lifting equipment of various equipment types through thresholds corresponding to characteristic value combinations, making the evaluation criteria more universal and effective, and can objectively and accurately reflect the health of the elevator operation. It is applicable to a variety of equipment types and can detect abnormal conditions before an elevator car stops functioning, helping maintenance personnel to discover problems in advance and avoid shutdown failures.
[0179] Example 3
[0180] This embodiment provides a system for selecting characteristic values for detecting the operating status of a fixed lifting device. The system is for multiple types of fixed lifting devices, wherein the fixed lifting device may be an elevator car. Figure 6 The selection system includes: a feature value acquisition module 101, a separate threshold acquisition module 102, a separate accuracy acquisition module 103, a general feature value acquisition module 104, a general threshold acquisition module 105, a general accuracy acquisition module 106, a comprehensive accuracy acquisition module 107 and a feature value selection module 108.
[0181] The characteristic value acquisition module 101 is used to acquire the type and value of the characteristic value of each device type in the normal state of the device. The characteristic value type includes peak value, kurtosis, etc.
[0182] The individual threshold value acquisition module 102 is used to acquire, for each individual characteristic value type of each device type, an individual upper threshold value and an individual lower threshold value for indicating that the device of the device type is in a normal state.
[0183] The individual accuracy acquisition module 103 is used to perform health assessment on devices of each device type based on an individual upper threshold and an individual lower threshold to obtain an individual accuracy result.
[0184] The general feature value acquisition module 104 is used to acquire a general feature value type according to each individual feature value type of each device type.
[0185] The general threshold acquisition module 105 is used to obtain the general upper threshold and the general lower threshold of each general feature value type, and the general upper threshold and the general lower threshold are used to indicate that the devices of all device types are in a normal state. Among them, when the value is between the general upper threshold and the general lower threshold (including the case where it is equal to the general upper threshold or the general lower threshold), it means that it is in a normal state.
[0186] The general accuracy acquisition module 106 is used to perform health assessment on devices of each device type based on a general upper threshold and a general lower threshold to obtain a general accuracy result.
[0187] The comprehensive accuracy acquisition module 107 is used to acquire a comprehensive accuracy index based on the individual accuracy result and the general accuracy result, and the value of the comprehensive accuracy index is between the individual accuracy result and the general accuracy result. The value of the comprehensive accuracy index can be equal to the individual accuracy result or the general accuracy result.
[0188] The eigenvalue selection module 108 is used to select a combination of eigenvalues, using the comprehensive accuracy index as the objective function, to obtain a combination of eigenvalues that meets the requirements of the comprehensive accuracy index and a threshold value corresponding to each eigenvalue. Specifically, a particle swarm algorithm is used for selection.
[0189] In specific implementation, the feature value combination includes a separate feature value combination for each device type and a common feature value combination. Among them, the common feature value combination has the common feature value type in the separate feature value combination of each device type. For example, there are 3 device types in total, the separate feature value combination of device type 1 is A, B and C, the separate feature value combination of device type 2 is B, C and E, and the separate feature value combination of device type 3 is A, B, C and F, then the common feature value combination is B and C.
[0190] and / or,
[0191] The individual threshold acquisition module 102 is used to acquire an individual upper threshold and an individual lower threshold for each individual characteristic value type of each device type according to a statistically based data selection criterion. Preferably, the statistically based data selection criterion may be a 3σ criterion or a 2σ criterion.
[0192] and / or,
[0193] Reference Figure 7 The feature value acquisition module 101 includes: a vibration data acquisition unit 109 , a vibration data classification unit 110 , a target data acquisition unit 111 and a feature extraction unit 112 .
[0194] The vibration data acquisition unit 109 is used to acquire vibration data generated by the fixed lifting equipment during operation through a vibration sensor for each equipment type, wherein the vibration data includes vibration acceleration data in the directions of the X-axis, Y-axis and Z-axis generated by the elevator car during operation.
[0195] The vibration data classification unit 110 is used to obtain the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to obtain classified vibration data.
[0196] The target data acquisition unit 111 is used to acquire target data, which is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting equipment. The uniform speed segment data is the vibration data of the fixed lifting equipment when it is in a uniform speed operating state. The operating state of the fixed lifting equipment includes an upward operating acceleration segment, an upward operating uniform speed segment, an upward operating deceleration segment, a downward operating acceleration segment, a downward operating uniform speed segment, a downward operating deceleration segment, and a stationary segment at a floor stop. The uniform speed segment data includes vibration data corresponding to the upward operating uniform speed segment and the downward operating uniform speed segment. The uniform speed segment data can more accurately reflect the operating state of the fixed lifting equipment than other segment data, thereby improving the quality of the data and the accuracy of the data.
[0197] The feature extraction unit 112 is used to extract features from the target data to obtain the type and value of the specified feature value.
[0198] and / or,
[0199] Reference Figure 8 The feature value acquisition module 101 includes: a vibration data acquisition unit 109, a vibration data classification unit 110, a target data acquisition unit 111, a static segment data extraction unit 113, a zero drift value acquisition unit 114, a feature extraction unit 112 and a contribution rate sorting unit 115.
[0200] The vibration data acquisition unit 109 is used to obtain the vibration data generated by the fixed lifting equipment during operation through the vibration sensor for each equipment type. The vibration data is the normal vibration data of a complete cycle, wherein the fixed lifting equipment is a complete cycle from docking at the starting floor to running and docking at the target floor. The normal vibration data is data that can truthfully reflect the operating status of the fixed lifting equipment. The vibration data includes the vibration acceleration data in the X-axis, Y-axis and Z-axis directions generated by the elevator car during operation. In the specific implementation process, the original vibration data can be removed by removing noise data, large deviation data and incomplete cycle data to obtain normal vibration data of a complete cycle. Noise data is meaningless data, and incomplete cycle data is data that cannot be identified as belonging to a complete cycle. Those skilled in the art can set the judgment conditions and parameters of large deviation data according to actual needs.
[0201] The vibration data classification unit 110 is used to obtain the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to obtain classified vibration data.
[0202] The target data acquisition unit 111 is used to acquire target data, which is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting equipment. The uniform speed segment data is the vibration data of the fixed lifting equipment when it is in a uniform speed operating state. The operating state of the fixed lifting equipment includes an upward operating acceleration segment, an upward operating uniform speed segment, an upward operating deceleration segment, a downward operating acceleration segment, a downward operating uniform speed segment, a downward operating deceleration segment, and a stationary segment at a floor stop. The uniform speed segment data includes vibration data corresponding to the upward operating uniform speed segment and the downward operating uniform speed segment. The uniform speed segment data can more accurately reflect the operating state of the fixed lifting equipment than other segment data, thereby improving the quality of the data and the accuracy of the data.
[0203] The static segment data extraction unit 113 is used to extract static segment data from the vibration data, where the static segment data is vibration data when the fixed lifting device is in a static state.
[0204] The zero drift value acquisition unit 114 is used to acquire the zero drift value of the vibration sensor based on the static segment data, and the target data is specifically the zero drift-free data obtained after removing the zero drift value of the vibration sensor from the uniform speed segment data. Among them, removing the zero drift value of the vibration sensor from the uniform speed segment data can eliminate the influence of different elevator types, different fault types, and different vibration sensor installation positions on the original vibration data.
[0205] The feature extraction unit 112 is used to extract features from the target data to obtain the type and value of the specified feature value.
[0206] The contribution rate sorting unit 115 is used to sort the contribution rates of the features using the gradient boosting tree algorithm, extract the first m principal component features whose cumulative contribution rates are greater than a predetermined percentage threshold to form a data set, where m is a positive integer. The predetermined percentage threshold can be set according to actual needs.
[0207] In specific implementation, the characteristic value under normal equipment status is expressed by the following formula:
[0208] x i0 =(f i1 ,…,f i j,…,f im ).
[0209] Where m represents the total number of eigenvalues, f ij The unit characteristic value of type j represents the characteristic value of device type i, where I represents the set of device types to be analyzed. J represents the set of eigenvalue types to be selected,
[0210] The upper threshold alone is denoted as UT ij , the lower threshold is denoted as LT ij .
[0211] The general upper threshold is expressed by the following formula:
[0212]
[0213] The general lower bound threshold is expressed by the following formula:
[0214]
[0215] The individual accuracy result is expressed as A iJ , the general accuracy result is expressed as UA iJ , the comprehensive accuracy index is expressed by the following formula:
[0216]
[0217] The eigenvalue combination is denoted as J * , the threshold of each eigenvalue of the eigenvalue combination is expressed by the following formula:
[0218] {UT ij ,LT ij ,UT Ij ,LT Ij |i∈I,j∈J *}.
[0219] In specific implementation, the feature value selection module 108 is used to obtain a feature value combination that meets the comprehensive accuracy index requirements and has the largest corresponding accuracy result, as well as a threshold value corresponding to each feature value.
[0220] This embodiment obtains precise data by removing zero drift and extracting features from the uniform speed segment data that can accurately reflect the operating status of the fixed lifting equipment, and obtains the accuracy results of each equipment type based on the precise data. It not only provides accuracy results for a single equipment type, but also integrates all equipment types to calculate a universal accuracy result, and then obtains a comprehensive accuracy index applicable to all equipment types, establishes a unified detection standard for equipment of all equipment types, and the thresholds corresponding to the characteristic value combinations make the evaluation standard more universal and more effective, and can objectively and accurately reflect the operating health of the elevator, and is applicable to a variety of equipment types.
[0221] Example 4
[0222] This embodiment is a detection system for a fixed lifting device. Fig. 9 The detection system includes: a to-be-detected data acquisition module 201, a to-be-detected combined data extraction module 202 and a detection module 203.
[0223] The data acquisition module 201 is used to acquire the data to be detected of the fixed lifting equipment to be detected. The data to be detected is the data obtained after the fixed lifting equipment to be detected is processed by the aforementioned feature value selection system for detecting the operating status of the fixed lifting equipment.
[0224] The to-be-detected combined data extraction module 202 is used to extract the to-be-detected combined data corresponding to the characteristic value combination from the to-be-detected data.
[0225] The detection module 203 is used to compare the combination data to be detected with the threshold value corresponding to the characteristic value combination. If the combination data to be detected is within the threshold interval of the corresponding equipment type in the threshold, the fixed lifting equipment is in a healthy state; otherwise, the fixed lifting equipment is in a fault state.
[0226] In a specific implementation, if the equipment type of the fixed lifting equipment to be detected has a corresponding individual feature value combination and threshold, the detection module 203 compares the combination data to be detected with the individual feature value combination and the threshold.
[0227] If the equipment type of the fixed lifting equipment to be detected does not have a corresponding individual feature value combination and threshold, the detection module 203 compares the combination data to be detected with the general feature value combination and threshold.
[0228] This embodiment detects fixed lifting equipment of various equipment types through thresholds corresponding to characteristic value combinations, making the evaluation criteria more universal and effective, and can objectively and accurately reflect the health of the elevator operation. It is applicable to a variety of equipment types and can detect abnormal conditions before an elevator car stops functioning, helping maintenance personnel to discover problems in advance and avoid shutdown failures.
[0229] Example 5
[0230] Fig.10 A schematic diagram of the structure of an electronic device provided in Example 5 of the present invention. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the method for selecting characteristic values for detecting the operating state of a fixed lifting device in Example 1 is implemented, or the method for detecting a fixed lifting device in Example 2 is implemented. Fig.10 The electronic device 30 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0231] The electronic device 30 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including the memory 32 and the processor 31).
[0232] The bus 33 includes a data bus, an address bus, and a control bus.
[0233] The memory 32 may include a volatile memory, such as a random access memory (RAM) 321 and / or a cache memory 322 , and may further include a read-only memory (ROM) 323 .
[0234] The memory 32 may also include a program / utility 325 having a set (at least one) of program modules 324, such program modules 324 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0235] The processor 31 executes various functional applications and data processing by running the computer program stored in the memory 32, such as the method for selecting characteristic values for detecting the operating status of a fixed lifting device in Example 1 of the present invention, or implementing the detection method for a fixed lifting device in Example 2.
[0236] The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 35. Furthermore, the model-generated device 30 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 36. As shown, the network adapter 36 communicates with other modules of the model-generated device 30 via a bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the model-generated device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (RAID) systems, tape drives, and data backup storage systems.
[0237] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, such division is only exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more units / modules described above can be embodied in one unit / module; conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules for embodiment.
[0238] Example 6
[0239] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the program implements the steps of the method for selecting characteristic values for detecting the operating status of a fixed lifting device in Embodiment 1, or implements the steps of the method for detecting a fixed lifting device in Embodiment 2.
[0240] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.
[0241] In a possible implementation manner, the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps of the method for selecting characteristic values for detecting the operating status of a fixed lifting device in Example 1, or to implement the steps of the method for detecting a fixed lifting device in Example 2.
[0242] The program code for executing the present invention may be written in any combination of one or more programming languages, and may be executed entirely on a user device, partially on a user device, as an independent software package, partially on a user device and partially on a remote device, or entirely on a remote device.
[0243] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that this is only for illustration and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.
[0244] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that this is only for illustration and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.
Claims
1. A method for selecting characteristic values for detecting the operating status of fixed lifting equipment, It is characterized in that The selection method is for fixed lifting equipment of multiple equipment types, and the selection method includes: Obtain the type and value of the characteristic value of each device type in the normal state of the device; For each individual characteristic value type of each device type, obtaining an individual upper threshold value and an individual lower threshold value for characterizing that a device of the device type is in a normal state; For each device type, performing a health assessment on devices of the device type based on the individual upper threshold and the individual lower threshold to obtain an individual accuracy result; Obtain a common characteristic value type according to each individual characteristic value type of each device type; Obtaining a universal upper threshold and a universal lower threshold of each universal characteristic value type, wherein the universal upper threshold and the universal lower threshold are used to indicate that devices of all device types are in a normal state; For each device type, performing health assessment on devices of the device type based on the general upper threshold and the general lower threshold to obtain a general accuracy result; Obtaining a comprehensive accuracy index based on the individual accuracy result and the general accuracy result, wherein a value of the comprehensive accuracy index is between the individual accuracy result and the general accuracy result; Select a combination of feature values, take the comprehensive accuracy index as the objective function, and obtain a combination of feature values that meets the requirements of the comprehensive accuracy index and a threshold value corresponding to each feature value; The step of obtaining the type and value of the characteristic value of each device type in the normal state of the device specifically includes: For each equipment type, the vibration data generated by the fixed lifting equipment during operation is obtained through a vibration sensor, and the vibration data is normal vibration data of a complete cycle, wherein the fixed lifting equipment is a complete cycle from docking at the starting floor to running and docking at the target floor; Acquire the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to acquire classified vibration data; Acquire target data, wherein the target data is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting device, and the uniform speed segment data is vibration data when the fixed lifting device is in a uniform speed operating state; Extracting static segment data from the vibration data, wherein the static segment data is vibration data when the fixed lifting device is in a static state; Acquire a zero drift value of a vibration sensor based on the static segment data, wherein the target data is specifically zero drift-removed data obtained after removing the zero drift value of the vibration sensor from the uniform speed segment data; Performing feature extraction on the target data to obtain the type and value of the specified feature value; The feature values are sorted by contribution rate using a gradient boosting tree algorithm, and the first m principal component features whose cumulative contribution rates are greater than a predetermined percentage threshold are extracted to form a data set, where m is a positive integer.
2. The method for selecting characteristic values for detecting the operating status of a fixed lifting device according to claim 1, It is characterized in that The selection method also includes: The feature value combination includes a separate feature value combination for each device type and a common feature value combination; and / or, For each individual characteristic value type of each device type, obtaining an individual upper threshold value and an individual lower threshold value according to a statistically based data selection criterion; and / or, The step of obtaining the type and value of the characteristic value of each device type in the normal state of the device specifically includes: For each equipment type, obtaining vibration data generated by the fixed lifting equipment during operation through a vibration sensor; Acquire the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to acquire classified vibration data; Acquire target data, wherein the target data is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting device, and the uniform speed segment data is vibration data when the fixed lifting device is in a uniform speed operating state; Feature extraction is performed on the target data to obtain the type and value of the specified feature value.
3. The method for selecting characteristic values for detecting the operating status of a fixed lifting device according to claim 2, It is characterized in that The characteristic value under normal equipment condition is expressed by the following formula: x i0 =(f i1 ,…,f ij ,…,f im ); Where m represents the total number of eigenvalues, f ij The unit characteristic value of type j represents the characteristic value of device type i, where I represents the set of device types to be analyzed. J represents the set of eigenvalue types to be selected, The single upper threshold is denoted as UT ij , the single lower threshold is denoted as LT ij ; The general upper threshold is expressed by the following formula: The general lower threshold is expressed by the following formula: The individual accuracy result is expressed as A iJ , the general accuracy result is denoted as UA iJ , the comprehensive accuracy index is expressed by the following formula: The eigenvalue combination is represented by J * , the threshold value of each eigenvalue of the eigenvalue combination is expressed by the following formula: {UT ij, LT ij ,UT Ij ,LT Ij |i∈I,j∈J * }。 4. The method for selecting characteristic values for detecting the operating status of a fixed lifting device according to claim 2, It is characterized in that The step of obtaining a combination of feature values that meets the comprehensive accuracy index requirements and a threshold value corresponding to each feature value specifically includes: Obtain the eigenvalue combination that meets the comprehensive accuracy index requirements and corresponds to the maximum accuracy result and the threshold value corresponding to each eigenvalue.
5. A method for detecting a fixed lifting device, It is characterized in that The detection method comprises: Acquire data to be detected of the fixed lifting equipment to be detected, wherein the data to be detected is data obtained after processing the fixed lifting equipment to be detected using the method for selecting characteristic values for detecting the operating state of the fixed lifting equipment according to any one of claims 1 to 4; Extracting the combination data to be detected corresponding to the characteristic value combination from the data to be detected; The combination data to be detected is compared with a threshold value corresponding to the characteristic value combination. If the combination data to be detected is within a threshold interval of the corresponding equipment type in the threshold value, the fixed lifting equipment is in a healthy state; otherwise, the fixed lifting equipment is in a fault state.
6. The detection method of the fixed lifting equipment according to claim 5, It is characterized in that The step of comparing the combination data to be detected with the threshold value corresponding to the characteristic value combination specifically includes: If the equipment type of the fixed lifting equipment to be detected has a corresponding individual characteristic value combination and threshold, then comparing the combination data to be detected with the individual characteristic value combination and threshold; If the equipment type to which the fixed lifting equipment to be detected belongs does not have a corresponding individual characteristic value combination and threshold value, the combination data to be detected is compared with the universal characteristic value combination and threshold value.
7. A system for selecting characteristic values for detecting the operating status of fixed lifting equipment, It is characterized in that The selection system is for fixed lifting equipment of multiple equipment types, and the selection system includes: a feature value acquisition module, a separate threshold acquisition module, a separate accuracy acquisition module, a general feature value acquisition module, a general threshold acquisition module, a general accuracy acquisition module, a comprehensive accuracy acquisition module and a feature value selection module; The characteristic value acquisition module is used to obtain the type and value of the characteristic value of each device type when the device is in a normal state; The separate threshold value acquisition module is used to acquire, for each separate characteristic value type of each device type, a separate upper threshold value and a separate lower threshold value for characterizing that the device of the device type is in a normal state; The individual accuracy acquisition module is used for performing health assessment on devices of the device type based on the individual upper threshold and the individual lower threshold for each device type to obtain an individual accuracy result; The universal feature value acquisition module is used to acquire the universal feature value type according to each individual feature value type of each device type; The universal threshold acquisition module is used to acquire a universal upper threshold and a universal lower threshold of each universal feature value type, wherein the universal upper threshold and the universal lower threshold are used to indicate that devices of all device types are in a normal state; The general accuracy acquisition module is used for performing health assessment on devices of each device type based on the general upper threshold and the general lower threshold to obtain a general accuracy result; The comprehensive accuracy acquisition module is used to acquire a comprehensive accuracy index based on the individual accuracy result and the general accuracy result, and the value of the comprehensive accuracy index is between the individual accuracy result and the general accuracy result; The feature value selection module is used to select a combination of feature values, taking the comprehensive accuracy index as the objective function, and obtaining a combination of feature values that meets the requirements of the comprehensive accuracy index and a threshold value corresponding to each feature value; The characteristic value acquisition module includes: a vibration data acquisition unit, a vibration data classification unit, a target data acquisition unit, a static segment data extraction unit, a zero drift value acquisition unit, a characteristic extraction unit and a contribution rate sorting unit; The vibration data acquisition unit is used to acquire, for each equipment type, vibration data generated by the fixed lifting equipment during operation through a vibration sensor, wherein the vibration data is normal vibration data of a complete cycle, wherein a complete cycle is when the fixed lifting equipment stops at a starting floor and then runs and stops at a target floor; The vibration data classification unit is used to obtain the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to obtain classified vibration data; The target data acquisition unit is used to acquire target data, wherein the target data is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting device, and the uniform speed segment data is vibration data when the fixed lifting device is in a uniform speed operating state; The static segment data extraction unit is used to extract static segment data from the vibration data, wherein the static segment data is vibration data when the fixed lifting device is in a static state; The zero drift value acquisition unit is used to acquire the zero drift value of the vibration sensor based on the static segment data, and the target data is specifically the zero drift-removed data acquired after removing the zero drift value of the vibration sensor from the uniform speed segment data; The feature extraction unit is used to perform feature extraction on the target data to obtain the type and value of the specified feature value; The contribution rate sorting unit is used to sort the contribution rates of the features using a gradient boosting tree algorithm on the feature values, and extract the first m principal component features whose cumulative contribution rates are greater than a predetermined percentage threshold to form a data set, where m is a positive integer.
8. The system for selecting characteristic values for detecting the operating status of a fixed lifting device according to claim 7, It is characterized in that The feature value combination includes a separate feature value combination for each device type and a common feature value combination; and / or, The separate threshold value acquisition module is used to acquire a separate upper threshold value and a separate lower threshold value for each separate characteristic value type of each device type according to a statistically based data selection criterion; and / or, The feature value acquisition module includes: a vibration data acquisition unit, a vibration data classification unit, a target data acquisition unit and a feature extraction unit; The vibration data acquisition unit is used to acquire vibration data generated by the fixed lifting equipment during operation through a vibration sensor for each equipment type; The vibration data classification unit is used to obtain the device state corresponding to the vibration data, and classify the vibration data according to the normal state of the device and the type of device failure to obtain classified vibration data; The target data acquisition unit is used to acquire target data, wherein the target data is uniform speed segment data obtained by segmenting and extracting the classified vibration data according to the operating state of the fixed lifting device, and the uniform speed segment data is vibration data when the fixed lifting device is in a uniform speed operating state; The feature extraction unit is used to perform feature extraction on the target data to obtain the type and value of the specified feature value.
9. The system for selecting characteristic values for detecting the operating state of a fixed lifting device according to claim 8, It is characterized in that The characteristic value under normal equipment condition is expressed by the following formula: x i0 =(f i1 ,…,f ij ,…,f im ); Where m represents the total number of eigenvalues, f ij The unit characteristic value of type j represents the characteristic value of device type i, where I represents the set of device types to be analyzed. J represents the set of eigenvalue types to be selected, The single upper threshold is denoted as UT ij , the single lower threshold is denoted as LT ij ; The general upper threshold is expressed by the following formula: The general lower threshold is expressed by the following formula: The individual accuracy result is expressed as A iJ , the general accuracy result is denoted as UA iJ , the comprehensive accuracy index is expressed by the following formula: The eigenvalue combination is represented by J * , the threshold value of each eigenvalue of the eigenvalue combination is expressed by the following formula: {UT ij, LT ij ,UT Ij ,LT Ij |i∈I,j∈J * }。 10. The system for selecting characteristic values for detecting the operating status of a fixed lifting device according to claim 8, It is characterized in that The feature value selection module is used to obtain the feature value combination that meets the comprehensive accuracy index requirements and corresponds to the maximum accuracy result, as well as the threshold value corresponding to each feature value.
11. A detection system for fixed lifting equipment, It is characterized in that The detection system comprises: a to-be-detected data acquisition module, a to-be-detected combined data extraction module and a comparison module; The to-be-detected data acquisition module is used to acquire the to-be-detected data of the fixed lifting device to be detected, wherein the to-be-detected data is data obtained after being processed by the feature value selection system for detecting the operating state of the fixed lifting device according to any one of claims 7 to 10 for the fixed lifting device to be detected; The to-be-detected combined data extraction module is used to extract the to-be-detected combined data corresponding to the characteristic value combination from the to-be-detected data; The comparison module is used to compare the combination data to be detected with the threshold value corresponding to the characteristic value combination. If the combination data to be detected is within the threshold interval of the corresponding equipment type in the threshold value, the fixed lifting equipment is in a healthy state; otherwise, the fixed lifting equipment is in a fault state.
12. The detection system for fixed lifting equipment according to claim 11, It is characterized in that If the equipment type to which the fixed lifting equipment to be detected belongs has a corresponding individual characteristic value combination and threshold, the comparison module compares the combination data to be detected with the individual characteristic value combination and threshold; If the equipment type to which the fixed lifting equipment to be detected belongs does not have a corresponding separate feature value combination and threshold, the comparison module compares the combination data to be detected with the general feature value combination and threshold.
13. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the method for selecting characteristic values for detecting the operating status of a fixed lifting device described in any one of claims 1-4 is implemented, or the method for detecting a fixed lifting device described in any one of claims 5-6 is implemented.
14. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method for selecting characteristic values for detecting the operating status of a fixed lifting device described in any one of claims 1-4 are implemented, or the steps of the method for detecting a fixed lifting device described in any one of claims 5-6 are implemented.
