Elevator operation state detection method and system
Through the collaborative verification of multi-source features of mobile terminal clusters and cloud-based big data analysis, the high cost, limited coverage, lack of real-time performance and privacy risks of elevator monitoring technology have been solved, and low-cost, highly reliable real-time accurate perception of elevator operating status and fault warning have been achieved.
Patent Information
- Application Number
- CN202511053764.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-30
- Publication Date
- 2025-10-17
- Estimated Expiration
- 2045-07-30
AI Technical Summary
Existing elevator monitoring technology has the disadvantages of high cost, limited coverage, insufficient real-time performance, low data reliability, lack of positioning capability and privacy risks, making it difficult to achieve low-cost, wide-coverage and highly reliable real-time and accurate perception of elevator operating status and fault warning.
Elevator environment data is collected using mobile terminal clusters. Through lightweight detection models and multi-source feature collaborative verification, combined with cloud-based big data analysis, anonymized elevator status processing and high-confidence fault diagnosis are achieved. Multi-sensor fusion technology is used for real-time monitoring and anomaly detection.
It significantly reduces deployment costs, expands coverage, improves the real-time and accuracy of monitoring, enhances the ability to detect elevator equipment status, and protects user privacy.
Smart Images

Figure CN120553525B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of Internet of Things, and particularly relates to an elevator operation state detection method and system. BACKGROUND
[0002] As the core equipment of modern urban vertical transportation, the safe operation and state monitoring of elevators have always been the focus of technical research. However, the traditional elevator monitoring scheme and the existing mobile device-based technology still have significant limitations, as follows.
[0003] 1) Limitations of traditional elevator monitoring technology
[0004] High cost and complex deployment: relying on special sensors (such as vibration, acceleration, and video equipment) and wired / wireless data acquisition systems, the equipment procurement, installation, wiring, and maintenance costs are high, especially the old elevator modification is difficult and economically inefficient.
[0005] Limited coverage: only the elevators with deployed systems can be monitored, making it difficult to achieve comprehensive coverage of a large number of in-service elevators (especially unconnected or low-end elevators), and it is difficult to obtain regional or nationwide big data, which restricts macro safety management decisions.
[0006] Insufficient real-time performance: some systems rely on periodic manual inspection or fault reporting, and cannot monitor the elevator state in real time, making it difficult to timely warn potential risks.
[0007] 2) Limitations of mobile device-based monitoring technology
[0008] Low data reliability: single phone sensors (such as accelerometers and barometers) are easily affected by device differences, user holding posture, and environmental interference, resulting in large feature extraction errors and high misjudgment rates (such as misjudging walking and vehicle bumps as elevator movement).
[0009] Insufficient monitoring depth: existing technologies can only identify elevator running / starting and stopping states or rough floor changes, and cannot accurately identify uniform speed, acceleration, and deceleration running states, making it difficult to detect potential mechanical faults such as flat layer precision, abnormal vibration, and steel wire rope shaking and guide rail abnormalities.
[0010] Lack of positioning capability: lacking multi-device coordination mechanism, it is difficult to accurately locate abnormal elevator equipment and implement large-scale monitoring.
[0011] Privacy risks: user data collection and use lack effective privacy protection design, limiting technology promotion and application.
[0012] Although the traditional scheme and mobile device technology have made attempts, they have not solved the core problems of cost, coverage, real-time, accuracy and privacy. Therefore, a low-cost, wide-coverage and high-reliability elevator monitoring technology is urgently needed to realize real-time and accurate perception of elevator operation state and fault warning through multi-source data fusion, intelligent algorithm optimization and privacy protection design, while meeting the needs of large-scale deployment.
[0013] Based on the above situation, the application provides an elevator operation state detection method and system. SUMMARY
[0014] The application provides a simple and efficient elevator operation state detection method and system to overcome the defects of the prior art.
[0015] The application is implemented by the following technical solutions:
[0016] An elevator operation state detection method comprises the following steps:
[0017] Step S1, mobile terminal cluster MTC (Mobile Terminal Cluster) data processing;
[0018] Step S1.1, using a mobile terminal as a sensor to collect elevator environment data and obtaining original feature values of the elevator operation state from the data, including motion state, height, environmental sound and geographic location;
[0019] Step S1.2, after filtering and denoising the extracted data, extracting key features, including acceleration waveform, air pressure change amplitude and sound spectrum;
[0020] Step S1.3, using a local lightweight detection model, a threshold-based state machine or a small decision tree, to preliminarily judge the elevator state, including the start-stop state and floor change of the elevator, and generate a preliminary event to realize single-point event recognition;
[0021] Step S1.4, removing personal identity information, blurring building-level positioning, retaining a geographic fence area, adding differential privacy noise, and performing anonymization processing to generate an anonymous feature data packet, and uploading it to a coordination center;
[0022] In step S1.1, an acceleration and angular velocity of the elevator operation is collected by a motion sensor to determine the vibration mode and motion state of the elevator, and the motion state includes static, uplink, downlink, acceleration and deceleration;
[0023] An air pressure sensor is used to detect air pressure, and the vertical displacement of the elevator is determined by detecting the change in air pressure;
[0024] The microphone is used to collect the sound of the running environment, which is used to identify the abnormal sound features, including elevator abnormal sound, friction sound and alarm sound.
[0025] In the step S1.1, the magnetic field sensor is also used to collect the magnetic field signal, which is combined with the floor sensor to assist in determining the position of the elevator in the shaft.
[0026] The global positioning system GPS, wireless network WiFi or Bluetooth module is used to collect the global positioning system GPS signal, wireless network WiFi signal or Bluetooth signal, which is used to realize the preliminary coarse-grained positioning of the elevator relative to the building floor or the shaft entrance.
[0027] Step S2, the collaborative localization and verification center CLVC processes.
[0028] Step S2.1, time-space correlation grouping.
[0029] According to the fuzzy position and the timestamp, the device data in the same time-space window is grouped, the device data packets in the same custom position range and the same custom time period are clustered, and the candidate device group is formed.
[0030] Step S2.2, multi-source feature collaborative verification.
[0031] Verify whether the acceleration waveform, air pressure change trend and abnormal sound feature of the devices in the same candidate device group are consistent; according to the verification result, remove the device data with feature difference exceeding 10% from the candidate device group to form a collaborative device group.
[0032] The different device data in the same collaborative device group is integrated to determine the elevator motion state and the floor change sequence, and the elevator motion state includes the elevator motion direction, speed and start-stop time.
[0033] Step S2.3, elevator positioning and confidence calculation.
[0034] The center point matching based on the fuzzy position of the device group is used to determine the elevator position, including the specific building unit or the car area.
[0035] The confidence of positioning and motion state recognition is calculated combined with the number of devices, feature consistency and sensor accuracy, and the confidence is evaluated.
[0036] Step S2.4, generate high-confidence events.
[0037] Filtering out high-confidence events with confidence greater than a custom threshold from the cooperative device group, generating a cooperative status event including elevator position, time, motion state, and detected abnormal features including strong vibration frequency points, abnormal sound types, and floor deviation;
[0038] In the step S2.2, the dynamic time warping (DTW) distance algorithm or correlation coefficient is used to compare the same dimension data in the same candidate device group, and the device data with a feature difference exceeding 10% is removed.
[0039] In the step S2.3, the elevator position is determined based on the signal fingerprint matching of the cooperative device group, and the signal fingerprint includes the magnetic field signal strength, the wireless network WiFi signal strength, and the Bluetooth signal strength.
[0040] Or map the fuzzy position of the cooperative device group to the elevator number through the property management system.
[0041] Step S3, cloud data analysis and diagnostic platform (CDAP) process;
[0042] Step S3.1, receiving the cooperative status event uploaded by the verification center, and constructing an elevator operation history database;
[0043] Step S3.2, identifying potential problems including flat layer error and speed anomaly by comparing with the standard operation curve, and realizing pattern recognition;
[0044] Diagnosing the elevator fault through the custom detection mechanism and the machine learning model, and realizing anomaly detection;
[0045] Calculating the health score of the elevator and the regional network, and realizing health degree evaluation;
[0046] Step S3.3, alarm and visualization;
[0047] Generating a visualization report and pushing alarm information including fault type, severity level, and elevator position to maintenance, property, and regulatory departments;
[0048] Step S3.4, system optimization feedback;
[0049] Iteratively optimizing the local algorithm, cooperative rules, and cloud model according to the artificial diagnosis result.
[0050] In the step S3.2, the custom detection mechanism includes a custom detection threshold and an abnormal sound alarm; the detection threshold includes a vibration threshold, an environmental sound B threshold, an ascending speed threshold, and a descending speed threshold.
[0051] Detecting start-stop shock anomalies or specific frequency vibrations caused by rail gaps using a time series classification model to diagnose elevator faults, identify potential faults from collaborative state events, including door machine jams, traction sheave wear, guide system anomalies, and inverter faults;
[0052] The time series classification model is trained using accumulated labeled data.
[0053] Step S4, user interaction process with management terminal UIMP (User Interface & Management Portal);
[0054] Passengers customize data collection preferences through the interface, including but not limited to turning off sound collection and adjusting the location blur range;
[0055] Maintenance, property, and regulatory departments view alarm information and handle work orders through the management terminal, updating elevator records;
[0056] Regulatory departments monitor regional elevator safety status and obtain compliance reports.
[0057] An elevator operation state detection system for implementing the above method, including a terminal layer, a collaborative layer, a cloud layer, and an interaction layer;
[0058] The terminal layer is responsible for using mobile terminals as sensors to collect elevator environment data, filtering and denoising the extracted data, extracting key features, fuzzy building-level positioning, retaining geographic fence areas, anonymizing processing, generating anonymous feature data packets, and uploading to the collaborative center;
[0059] The collaborative layer is responsible for clustering device data packets within the same custom location range and the same custom time period to form candidate device groups; removing device data with feature differences exceeding 10% (such as non-elevator devices) from the candidate device groups through consistency checking to form collaborative device groups; determining elevator positions, filtering high-confidence events with confidence greater than a custom threshold from the collaborative device groups, and generating collaborative state events;
[0060] The cloud layer is responsible for building an elevator operation history database, diagnosing elevator faults, and iteratively optimizing local algorithms, collaborative rules, and cloud models based on human diagnosis results;
[0061] The interaction layer is responsible for helping passengers customize data collection preferences, implementing privacy control, and responding to alarms and managing elevator systems through the management terminal.
[0062] An elevator operation state detection device, including a memory and a processor; the memory is used to store a computer program, and the processor is used to execute the computer program to implement the method steps described above.
[0063] A readable storage medium, a computer program is stored on the readable storage medium, the computer program is executed by a processor to realize the method steps described above.
[0064] The elevator operation state detection method and system have the advantages of low deployment threshold, wide coverage, strong real-time and continuity of monitoring, significantly reduced single device misjudgment rate, improved accuracy and confidence of elevator equipment, operation state and abnormality detection, and are suitable for popularization and application. BRIEF DESCRIPTION OF DRAWINGS
[0065] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0066] FIG. 1 is a schematic diagram of the overall architecture of the elevator detection network of the mobile terminal sensing and cooperative positioning of the present application. Figure 1 FIG. 2 is a schematic diagram of the local processing and data uploading process of the mobile terminal of the present application.
[0067] Figure 2 FIG. 3 is a schematic diagram of the core workflow of the cooperative positioning and verification center (CLVC) of the present application.
[0068] FIG. 4 is a schematic diagram of the workflow of the cloud big data analysis and diagnosis platform CDAP of the present application. Figure 3 FIG. 5 is a schematic diagram of the workflow of the cloud big data analysis and diagnosis platform CDAP of the present application.
[0069] Figure 4 FIG. 6 is a schematic diagram of the workflow of the cloud big data analysis and diagnosis platform CDAP of the present application. DETAILED DESCRIPTION
[0070] In order to make the person skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described in detail below. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.
[0071] The elevator operation state detection method comprises the following steps:
[0072] Step S1, mobile terminal cluster MTC (Mobile Terminal Cluster) data processing;
[0073] Step S1.1, using a mobile terminal as a sensor to collect elevator environment data and obtain original feature values of the elevator running state from the data, including motion state, height, environmental sound, and geographic location;
[0074] Step S1.2, after filtering and denoising the extracted data, key features are extracted, including acceleration waveform, air pressure change amplitude, and sound spectrum;
[0075] Step S1.3, using a local lightweight detection model, a threshold-based state machine or a small decision tree, to preliminarily determine the elevator state, including the start-stop state and floor change of the elevator, and generate a preliminary event, realizing single-point event recognition;
[0076] Step S1.4, removing personal identity information, blurring building-level positioning, retaining a geographic fence area, adding differential privacy noise, anonymizing, generating an anonymous trace feature data packet, and uploading to a collaborative center;
[0077] In the step S1.1, an acceleration and angular velocity of the elevator running is collected by a motion sensor, which is used to determine the vibration mode and motion state of the elevator, including static, uplink, downlink, acceleration, and deceleration;
[0078] An air pressure sensor is used to detect air pressure, and by detecting the change in air pressure, the vertical displacement of the elevator is determined;
[0079] A microphone is used to collect the running environment sound, which is used to identify abnormal sound features, including elevator abnormal sound, friction sound, and alarm sound.
[0080] In the step S1.1, a magnetic field sensor is also used to collect magnetic field signals, combined with a leveling inductor, to assist in determining the position of the elevator in the shaft;
[0081] A global positioning system GPS, wireless network WiFi, or Bluetooth module is used to collect global positioning system GPS signals, wireless network WiFi signals, or Bluetooth signals, which are used to realize preliminary coarse-grained positioning of the elevator relative to the building level or shaft entrance.
[0082] Step S2, collaborative positioning and verification center CLVC (Collaborative Localization & Verification Center) processing;
[0083] Step S2.1, time-space correlation grouping;
[0084] Grouping device data in the same spatio-temporal window according to fuzzy location and timestamp (e.g. same WiFi network key identifier BSSID, adjacent Bluetooth signal strength, same cell or close geographic fence area are likely to be a device cluster in the same elevator or a device group in the same shaft, experiencing the same elevator operation process, which are grouped into the same device group), clustering device data packets in the same custom location range and the same custom time period to form a candidate device group;
[0085] Step S2.2, multi-source feature collaborative verification;
[0086] Verifying whether the acceleration waveform, air pressure change trend and abnormal sound features of devices in the same candidate device group are consistent; according to the verification result, removing device data with feature difference exceeding 10% (e.g. non-elevator device) from the candidate device group to form a collaborative device group;
[0087] Determining elevator motion state and floor change sequence by integrating different device data in the same collaborative device group, the elevator motion state including elevator motion direction, speed and start-stop time;
[0088] Step S2.3, elevator positioning and confidence calculation;
[0089] Determining elevator position based on the center point matching of device group fuzzy location, including specific building unit or car area;
[0090] Combining device quantity, feature consistency and sensor accuracy to calculate the confidence of positioning and motion state recognition, and performing confidence evaluation;
[0091] Step S2.4, generating high-confidence event;
[0092] Filtering out high-confidence events with confidence greater than a custom threshold from the collaborative device group to generate collaborative state events, including elevator position, time, motion state and detected abnormal features, including strong vibration frequency point, abnormal sound type and floor change deviation;
[0093] In the step S2.2, dynamic time warping (DTW) distance algorithm or correlation coefficient is used to compare two-by-two the same dimension data in the same candidate device group, and remove device data with feature difference exceeding 10%.
[0094] In the step S2.3, the elevator position is determined based on the signal fingerprint matching of the collaborative device group, the signal fingerprint including magnetic field signal strength, wireless network WiFi signal strength and Bluetooth signal strength;
[0095] Or interfacing with the property management system, mapping the fuzzy location of the collaborative device group to the elevator number through the property management system.
[0096] Step S3, cloud data analysis & diagnostic platform (CDAP) process;
[0097] Step S3.1, receive the collaborative state events uploaded by the verification center, and build an elevator operation history database;
[0098] Step S3.2, identify potential problems including floor error and speed anomaly by comparing with standard operation curve, and realize pattern recognition;
[0099] Diagnose elevator faults through self-defined detection mechanism and machine learning model, and realize anomaly detection;
[0100] Calculate the health score of the elevator and regional network (such as 0-100 points), and realize health degree evaluation;
[0101] Step S3.3, alarm and visualization;
[0102] Generate visual reports (dashboard, trend chart), and push alarm information including fault type, severity level and elevator location to maintenance, property and regulatory departments;
[0103] Step S3.4, system optimization feedback;
[0104] Iteratively optimize local algorithms, collaborative rules and cloud models according to artificial diagnosis results.
[0105] In step S3.2, the self-defined detection mechanism includes self-defined detection threshold and abnormal sound alarm; the detection threshold includes vibration threshold, environmental sound threshold, ascending speed threshold and descending speed threshold;
[0106] Adopt time series classification model to detect start-stop impact anomaly or specific frequency vibration caused by guide rail gap, diagnose elevator faults, and identify potential faults from collaborative state event data, including door machine blocking, traction sheave wear, guide system anomaly and frequency converter failure;
[0107] The time series classification model is trained using accumulated labeled data.
[0108] Step S4, user interface & management portal (UIMP) interaction process;
[0109] Passengers can customize data collection preferences through the interface, including but not limited to turning off sound collection and adjusting position blur range;
[0110] Maintenance, property and regulatory departments can view alarm information and handle work orders through the management portal, and update elevator archives;
[0111] The regulatory department supervises the regional elevator safety state and obtains a compliance report.
[0112] The elevator operation state detection system is used to realize the above method, which includes a terminal layer, a collaborative layer, a cloud layer and an interactive layer.
[0113] The terminal layer is responsible for using a mobile terminal as a sensor to collect elevator environment data, filtering and denoising the extracted data, extracting key features, fuzzy building-level positioning, retaining a geographic fence area, anonymizing processing, generating an anonymous feature data packet, and uploading it to the collaborative center.
[0114] The collaborative layer is responsible for clustering device data packets within the same custom location range and the same custom time period to form a candidate device group; removing device data with more than 10% feature difference (such as non-elevator devices) from the candidate device group through consistency checking to form a collaborative device group; determining the elevator position, filtering out high-confidence events with a confidence greater than a custom threshold from the collaborative device group, and generating a collaborative state event.
[0115] The cloud layer is responsible for building an elevator operation history database, diagnosing elevator faults, and iteratively optimizing local algorithms, collaborative rules and cloud models based on artificial diagnosis results.
[0116] The interactive layer is responsible for helping passengers to customize data collection preferences, implementing privacy control, and responding to alarms and managing elevator systems through the management end.
[0117] The elevator operation state detection device includes a memory and a processor; the memory is used to store a computer program, and the processor is used to execute the computer program to realize the method steps described above.
[0118] The readable storage medium stores a computer program, which is executed by the processor to realize the method steps described above.
[0119] Compared with the prior art, the elevator operation state detection method and system have the following characteristics:
[0120] First, the cost is revolutionarily reduced: using existing massive mobile devices as sensing nodes, almost no additional hardware investment is needed in the elevator body, greatly reducing the deployment threshold.
[0121] Second, the coverage is greatly expanded: in theory, it can realize the monitoring of the state of all mobile terminal users equipped with corresponding sensors who ride elevators, covering various elevators from high-end buildings to old communities.
[0122] Third, the monitoring real-time and continuity are enhanced: mobile phone users use elevators frequently, providing high-frequency sampling, near real-time monitoring, and covering time periods that maintenance personnel cannot reach outside working hours.
[0123] Fourth, monitoring accuracy and reliability improvement (core advantage): Through multi-device cooperative positioning and feature consistency verification, the single-device misjudgment rate is significantly reduced, and the accuracy and confidence of detecting elevator equipment, running state (such as leveling precision) and abnormalities (such as abnormal vibration mode) are improved.
[0124] Fifth, deep abnormal diagnosis capability: Cloud big data analysis and AI model can mine potential fault patterns from complex multi-sensor fusion data, providing more in-depth diagnostic information.
[0125] Sixth, establish macroscopic elevator operation big data: Build an unprecedented wide-area elevator operation state database to serve maintenance optimization, quality supervision, and smart city management decision-making.
[0126] Seventh, privacy protection design: Through local preprocessing, anonymization, and fuzzy positioning techniques, valuable information can be obtained while protecting users' personal privacy to the greatest extent.
[0127] Eighth, easy to promote and implement: Users only need to install (or pre-install in the background) the APP to contribute data (participation can be set on and off), and the promotion resistance is small.
[0128] The above-described embodiments are only one of the specific implementations of the present application, and the usual changes and replacements made by those skilled in the art within the scope of the technical solutions of the present application should be included in the protection scope of the present application.
Claims
1. A method for detecting an elevator operating state, characterized in that: The following steps are involved: Step S1: processing MTC data of a mobile terminal cluster; Step S1.1: Use a mobile terminal as a sensor to collect elevator environmental data and obtain the original characteristic values of the elevator's operating status, including movement status, altitude, ambient sound, and geographic location; Step S1.2: After filtering and denoising the extracted data, key features are extracted, including acceleration waveform, air pressure variation amplitude, and sound spectrum; Step S1.3: Use a local lightweight detection model, a threshold-based state machine or a small decision tree to preliminarily determine the elevator status, including the start and stop status and floor changes of the elevator, and generate preliminary events to achieve single-point event recognition; Step S1.4: Remove personal identification information, blur the building-level positioning, retain the geo-fenced area, add differential privacy noise, perform anonymization, generate an anonymous feature data package, and upload it to the collaboration center; Step S2, collaborative location and verification center CLVC processing; Step S2.1, spatiotemporal correlation grouping; According to the fuzzy location and timestamp, the device data within the same spatiotemporal window are grouped, and the device data packets within the same custom location range and the same custom time period are clustered to form candidate device groups; Step S2.2, multi-source feature collaborative verification; Verify whether the acceleration waveforms, air pressure change trends, and abnormal sound characteristics of devices within the same candidate device group are consistent; Based on the verification results, device data with feature differences exceeding 10% are removed from the candidate device group to form a collaborative device group; Determine the elevator motion state and floor change sequence by integrating data from different devices within the same collaborative device group. The elevator motion state includes the direction of movement, speed, and start and stop times. Step S2.3, elevator positioning and confidence calculation; Determine the elevator location based on the center point matching of the fuzzy position of the equipment group, including the specific building unit or car area; Calculate the confidence of positioning and motion state recognition by combining the number of devices, feature consistency, and sensor accuracy, and perform confidence assessment; Step S2.4, generating a high confidence event; High-confidence events with a confidence level greater than a custom threshold are selected from the collaborative device group to generate collaborative status events, including elevator location, time, motion status, and detected abnormal features, including strong vibration frequency points, abnormal sound types, and floor change deviations. Step S3, cloud-based big data analysis and diagnosis platform CDAP process; Step S3.1: Receive the collaborative status events uploaded by the verification center and build an elevator operation history database; Step S3.2: Identify potential problems, including leveling errors and speed anomalies, by comparing with standard operating curves to achieve pattern recognition; Diagnose elevator failures and detect anomalies through custom detection mechanisms and machine learning models; Calculate the health scores of elevators and regional networks to achieve health assessment; Step S3.3, alarm and visualization; Generate visual reports and push alarm information to maintenance, property management, and regulatory departments, including fault type, severity level, and elevator location; Step S3.4, system optimization feedback; Iteratively optimize local algorithms, collaborative rules, and cloud models based on manual diagnosis results; Step S4: User and management UIMP interaction process; Passengers can customize data collection preferences through the interface, including but not limited to turning off sound collection and adjusting the location blur range; Maintenance, property management, and regulatory departments can view alarm information and process work orders through the management terminal, and update elevator files; Regulatory authorities monitor regional elevator safety status and obtain compliance reports.
2. The elevator operation status detection method according to claim 1, characterized in that: In step S1.1, a motion sensor is used to collect the acceleration and angular velocity of the elevator to determine the vibration mode and motion state of the elevator. The motion state includes stationary, upward, downward, accelerating, and decelerating. An air pressure sensor is used to detect air pressure, and the vertical displacement of the elevator is determined by detecting changes in air pressure; A microphone is used to collect operating environment sounds to identify abnormal sound characteristics, including elevator noises, friction sounds, and alarm sounds.
3. The elevator operation status detection method according to claim 2, characterized in that: In step S1.1, a magnetic field sensor is also used to collect magnetic field signals, which are combined with a leveling sensor to assist in determining the position of the elevator in the shaft; A global positioning system (GPS), wireless network (WiFi) or Bluetooth module is used to collect GPS signals, wireless network (WiFi) signals or Bluetooth signals to achieve preliminary coarse-grained positioning of the elevator relative to the building level or shaft entrance.
4. The elevator operation status detection method according to claim 1, characterized in that: In step S2.2, a dynamic time warping (DTW) distance algorithm or a correlation coefficient is used to perform pairwise comparisons on data of the same dimension within the same candidate device group, and device data with feature differences exceeding 10% are eliminated.
5. The elevator operation status detection method according to claim 1, characterized in that: In step S2.3, the elevator position is determined based on the signal fingerprint matching of the collaborative device group, where the signal fingerprint includes the magnetic field signal strength, the wireless network WiFi signal strength, and the Bluetooth signal strength; Or it can be connected to the property management system, and the fuzzy location of the collaborative equipment group can be mapped to the elevator number through the property management system.
6. The elevator operation status detection method according to claim 1, characterized in that: In step S3.2, the custom detection mechanism includes a custom detection threshold and an abnormal sound alarm; the detection threshold includes a vibration threshold, an ambient sound threshold, an ascending speed threshold, and a descending speed threshold; A time series classification model is used to detect abnormal vibration frequencies associated with faults caused by abnormal start-stop impact or guide rail gaps, diagnose elevator faults, and identify potential faults from coordinated state event data, including door machine jams, traction sheave wear, guide system anomalies, and inverter failures. The time series classification model is trained using the accumulated labeled data.
7. An elevator operation status detection system, characterized in that: Used to implement the method according to any one of claims 1 to 6, comprising a terminal layer, a collaboration layer, a cloud layer and an interaction layer; The terminal layer is responsible for using mobile terminals as sensors to collect elevator environmental data, filtering and denoising the extracted data, extracting key features, blurring building-level positioning, retaining geo-fenced areas, performing anonymization, generating anonymous feature data packets, and uploading them to the collaboration center. The collaboration layer is responsible for clustering device data packets within the same custom location range and the same custom time period to form candidate device groups. Device data with feature differences exceeding 10% are removed from the candidate device groups through consistency checks to form collaborative device groups. Determine the elevator location, filter out high-confidence events with confidence greater than a custom threshold from the collaborative device group, and generate collaborative status events; The cloud layer is responsible for building a database of elevator operation history, diagnosing elevator faults, and iteratively optimizing local algorithms, collaborative rules, and cloud models based on manual diagnosis results. The interaction layer is responsible for helping passengers customize data collection preferences, implement privacy control, and implement alarm response and elevator system management through the management end.
8. An elevator operation status detection device, characterized in that: The method comprises a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the method steps according to any one of claims 1 to 6 when executing the computer program.
9. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps according to any one of claims 1 to 6 are implemented.
Citation Information
Patent Citations
Elevator stopping detection method based on multi-sensor confidence vote mechanism
CN104816992A
Elevator running state multi-source sensing Internet of Things inspection system
CN120288599A