Elevator emergency rescue control method and system

Abstract

The elevator emergency rescue control system is composed of a state sensing subsystem, an emergency rescue execution subsystem and an emergency dispatching control platform, and elevator running state data are collected in real time through the multi-sensor fusion technology. Multi-dimensional feature vectors of elevator running states are constructed, feature level data compression and abnormal mode preliminary screening are executed at edge computing nodes, weighted diagnosis indexes and abnormal signals are generated and uploaded to the cloud side, rescue progress is synchronized through a human-computer interaction interface, and remote intervention of external rescue workers is supported. The composite fault disposal success rate is further improved, and the manual operation risk is remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator safety, and particularly to an elevator emergency rescue control method and system. Background Art

[0002] With the continuous progress of society and the continuous improvement of people's living standards, elevators are more and more widely used. People's requirements for elevators are no longer limited to just being able to take them, but begin to pay more attention to the comfort and safety of riding. Since elevators are mechanical products, there is a situation of wear and tear during use. At the same time, elevators are also public products, and it is difficult to predict passengers' familiarity with the products. Therefore, it is inevitable that accidents occur during the daily use of elevators. And now there are also emergency systems installed in elevators, such as monitoring and one-button calling, etc. However, these still have deficiencies in actual use. For example: some elevator emergency systems do not integrate response modules for compound disasters such as fires and gas leaks, and can only passively call for help in the event of these accidents. Some systems rely on a single network link and cannot transmit distress signals when the network is disconnected. There are also some elevator systems that do not have corresponding emergency measures in case of emergencies and are prone to secondary accidents. Summary of the Invention

[0003] (1) Technical problems to be solved

[0004] Aiming at the deficiencies of the prior art, the present invention collects multi-dimensional data in real time through a state perception subsystem, classifies faults by an AI model of an emergency dispatch control platform, and drives an emergency rescue execution subsystem to perform precise rescue.

[0005] (2) Technical solutions

[0006] To achieve the above object, the present invention provides the following technical solution: An elevator emergency rescue control system includes a state perception subsystem, an emergency rescue execution subsystem, and an emergency dispatch control platform, wherein:

[0007] The state perception subsystem adopts a multi-source heterogeneous data fusion architecture and collects elevator mechanical state parameters and environmental characteristic parameters in real time;

[0008] The emergency rescue execution subsystem is configured with a variety of physical execution units and communication relay devices for performing multi-modal rescue operations according to the fault type;

[0009] The emergency dispatch control platform is used to coordinate the interaction between the monitoring and the emergency rescue execution subsystem and dynamically optimize the rescue strategy through artificial intelligence algorithms;

[0010] The state perception subsystem includes various types of sensors deployed in the elevator car, hoistway, and traction machine, a remote monitoring multimedia terminal, an edge computing terminal, and a cloud data center. Among them, the sensors include an acceleration sensor, a tension sensor, a temperature and humidity sensor, and a barometric pressure sensor.

[0011] As a preferred solution, the emergency rescue execution subsystem includes:

[0012] A dynamic balance control module that regulates the magnetorheological damper of the counterweight through a high-precision servo motor, and combines a vision positioning system to compensate for the car offset and avoid the risk of secondary tilt;

[0013] An IoT communication module that transmits fault signals and rescue instructions based on 5G and LPWAN dual links, and supports off-line emergency communication in a network-disconnected environment;

[0014] A self-rescue module that includes a mechanical device driven by a manual / electric dual mode, and realizes the micro-lifting and micro-lowering of the car through bevel gear transmission for trapped persons to operate independently;

[0015] An environmental safety linkage module that integrates a multi-spectral gas analyzer and a distributed temperature sensing network, and establishes a fire linkage protocol with the building BMS system.

[0016] As a preferred solution, the dynamic balance control module monitors the tension of the traction wire rope in real time through a tension sensor, combines the load data in the car, and uses the PID algorithm to dynamically adjust the position of the counterweight to ensure the stable docking of the car. It is also equipped with a safety verification mechanism, including a dual-position feedback verification and an overload protection circuit.

[0017] As a preferred solution, the emergency dispatch control platform includes:

[0018] A fault feature engineering unit that performs anomaly detection on sensor data based on a machine learning model and classifies the fault levels;

[0019] A rescue decision optimization engine that matches the preset rescue plan according to the fault type and dynamically adjusts the rescue steps based on real-time data;

[0020] A human-machine collaborative interface that is equipped with a voice emotion recognition algorithm and an AR augmented reality interface to achieve multi-modal emergency guidance and physiological parameter monitoring.

[0021] As a preferred solution, the fault feature engineering unit applies a temporal convolutional network (TCN) to extract the spatio-temporal features of sensor data, constructs a three-dimensional fault pattern recognition model, and uses a federated learning framework to aggregate the historical operation data of multiple elevators to train the model.

[0022] An elevator emergency rescue control method includes the following steps:

[0023] S1. Real - time collect the elevator operation status data through multi - sensor fusion technology, and construct a multi - dimensional feature vector of the elevator operation status;

[0024] S2. Execute feature - level data compression and preliminary screening of abnormal patterns at the edge computing node, generate weighted diagnostic indicators and abnormal signals, and upload them to the cloud;

[0025] S3. Use the digital twin model for virtual - reality strategy verification, dynamically generate an emergency operation list including equipment linkage sequences and timing constraints, and start the corresponding rescue mode based on the fault type, giving priority to balance adjustment and communication link self - repair;

[0026] S4. Synchronize the rescue progress through the human - machine interface and support remote intervention by external rescue personnel.

[0027] As a preferred solution, in the power outage scenario, the system automatically switches to the backup power supply and sends a positioning code to the nearest rescue station through the Internet of Things communication module. The code includes the elevator ID, floor coordinates, and vital sign data of the trapped personnel.

[0028] As a preferred solution, when an abnormal concentration gradient of dangerous gases in the car is detected, start the sound field imaging device to locate the leakage source, activate emergency ventilation by the environmental safety module. If the situation continues to deteriorate, then link with the fire protection system to open the smoke exhaust channel in the hoistway and activate the color - coded guidance of the emergency lighting system.

[0029] (III) Beneficial effects

[0030] Compared with the prior art, the present invention provides an elevator emergency rescue control method and system, which have the following beneficial effects:

[0031] First, through multi - sensor fusion and modular design, the present invention can simultaneously handle various emergency situations such as mechanical failures, environmental disasters, and communication interruptions. The state perception subsystem collects multi - dimensional data in real - time, and the AI model of the emergency dispatch control platform classifies the faults and drives the emergency rescue execution subsystem to perform precise rescue. Compared with the prior art, the present invention further improves the success rate of compound fault handling and significantly reduces the risk of human operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic flow chart of the control method of the present invention;

[0033] Figure 2 It is a schematic diagram of the control system module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] In order to better understand the purpose, structure, and function of the present invention, the following will further explain an elevator emergency rescue control method and system of the present invention in combination with the drawings and specific embodiments.

[0035] Example 1

[0036] Reference Figure 1 , a method for controlling elevator emergency rescue according to the present invention includes the following steps:

[0037] S1. Real-time collect elevator operation status data through multi-sensor fusion technology, and construct a multi-dimensional feature vector of the elevator operation status;

[0038] S2. Perform feature-level data compression and preliminary screening of abnormal patterns at the edge computing node, generate weighted diagnostic indicators and abnormal signals, and upload them to the cloud;

[0039] S3. Use a digital twin model to perform virtual-reality strategy verification, dynamically generate an emergency operation list including device linkage sequences and timing constraints, and start the corresponding rescue mode based on the fault type, giving priority to balance adjustment and communication link self-repair;

[0040] S4. Synchronize the rescue progress through the human-computer interaction interface and support remote intervention by external rescue personnel.

[0041] Specifically, multi-sensor fusion in the method of the present invention includes vibration monitoring, which configures a combination of a MEMS three-axis accelerometer (range ±16g, 5kHz sampling) and a piezoelectric ceramic vibration sensor, and also distributes a laser PM2.5 sensor (0.1μm resolution) + a non-contact infrared temperature measurement module (-20°C to 120°C) for environmental perception. In addition, a Hall effect current sensor + an insulation impedance on-line monitoring unit are used for electrical detection;

[0042] The preliminary screening of its abnormal patterns includes an abnormal detection algorithm, specifically:

[0043] First-level screening: Use Isolation Forest for rapid outlier detection;

[0044] Second-level verification: Construct an elevator health prediction model based on the GRU network;

[0045] Feature compression: Use an improved wavelet packet transform.

[0046] Example 2

[0047] The present invention proposes a method for controlling elevator emergency rescue, and its specific implementation steps are:

[0048] Step S1: Data collection and preprocessing

[0049] Collect elevator acceleration, tension, temperature and humidity and other data through a sensor array, and the edge computing terminal normalizes the original data and eliminates outliers.

[0050] Step S2: Data Acquisition and Preprocessing

[0051] Use a pre-trained deep residual network to extract features from sensor data, output the fault probability distribution, and perform secondary verification in combination with an expert rule base (such as "the wire rope slips when the tensile force fluctuates more than the threshold three times in a row").

[0052] Step S3: Rescue Mode Selection and Execution

[0053] Adopt priority measurement, such as: first priority (execute immediately): fire, gas leakage, free fall of the car; second priority (respond within 5 minutes): power outage, mechanical jamming; third priority (manual confirmation): slight vibration, sensor false alarm. When a power outage is detected, first start the backup power supply, then send the code through LPWAN, and finally activate the car emergency lighting.

[0054] Step S4: Human-Machine Collaboration and Closed-Loop Verification

[0055] The trapped person inputs their health status (such as a heart attack) through a multimedia terminal, and the system adjusts the rescue strategy accordingly. After the rescue is completed, an event report (including sensor data, operation logs, rescue timeline) is automatically generated and uploaded to the regulatory department.

[0056] More specifically, the virtual-reality strategy verification and communication link self-repair adopt a virtual verification mechanism and self-repair strategy, specifically including: a physical engine, using ADAMS dynamics simulation software to construct an elevator digital twin; fault injection, supporting 12 preset fault modes (including wire rope fracture, door lock failure, etc.); sharing of abnormal mode features among each elevator node; strategy optimization, applying the NSGA-II multi-objective genetic algorithm to optimize the rescue timing. Communication recovery, using frequency-hopping spread-spectrum (FHSS) technology to reconstruct the wireless link (dual-band 2.4GHz / 5.8GHz); power redundancy, designing a three-mode switching power supply (mains - supercapacitor - hydraulic energy storage).

[0057] The human-machine interaction terminal supports including a remote intervention protocol, building a real-time interaction channel based on WebRTC, setting 4 levels of operation permissions (passenger / property / maintenance / fire emergency mode), and setting a synchronous rescue protocol. When multiple elevators are trapped in the same hoistway, automatically coordinate the rescue order, monitor vital signs in real time, and the lighting power supplies of adjacent elevator cars are mutually backup. During the rescue process, record the key data of the rescue process.

[0058] Embodiment 3

[0059] Reference Figure 2 , an elevator emergency rescue control system of the present invention includes a state perception subsystem, an emergency rescue execution subsystem, and an emergency dispatch control platform, wherein:

[0060] The state perception subsystem adopts a multi-source heterogeneous data fusion architecture to collect elevator mechanical state parameters and environmental characteristic parameters in real time;

[0061] The emergency rescue execution subsystem is configured with a variety of physical execution units and communication relay devices to perform multi-modal rescue operations according to the fault type;

[0062] The emergency dispatch control platform is used to coordinate the interaction between the monitoring and emergency rescue execution subsystems and dynamically optimize the rescue strategy through artificial intelligence algorithms.

[0063] Specifically, its state perception subsystem includes various types of sensors, remote monitoring multimedia terminals, edge computing terminals, and cloud data centers deployed in the elevator car, hoistway, and traction machine. Among them, the sensors include acceleration sensors, tension sensors, temperature and humidity sensors, and pressure sensors.

[0064] It adopts a dynamic sampling strategy, including normal and warning modes. The normal mode is that the edge node performs basic sampling at 10Hz, and the warning mode is to trigger high-density sampling at 50Hz and continue for 30s. Specifically, in car monitoring, a pressure difference-acceleration composite impact detection model is pioneered to distinguish elevator falls and earthquake shocks. In hoistway warning, based on the temperature gradient positioning algorithm of distributed optical fiber, the fire source positioning error has been further improved compared with traditional methods.

[0065] Specifically, the emergency rescue execution subsystem includes:

[0066] The dynamic balance control module regulates the magnetorheological damper of the counterweight through a high-precision servo motor, and combines a vision positioning system to compensate for the car offset and avoid the risk of secondary tilt;

[0067] The Internet of Things communication module transmits fault signals and rescue instructions based on 5G and LPWAN dual links, supporting offline emergency communication in a network-disconnected environment;

[0068] The self-rescue module includes a mechanical device driven by a manual / electric dual mode, and realizes the micro-lifting and micro-lowering of the car through bevel gear transmission for trapped people to operate independently;

[0069] The environmental safety linkage module integrates a multi-spectral gas analyzer and a distributed temperature sensing network, and establishes a fire protection linkage protocol with the building BMS system.

[0070] Specifically, the dynamic balance control module monitors the tension of the traction steel wire rope in real time through a tension sensor, combines the load data in the car, and uses the PID algorithm to dynamically adjust the position of the counterweight to ensure the stable docking of the car. It is also equipped with a safety verification mechanism, with dual position feedback verification and an overload protection circuit

[0071] Furthermore, in the event of a power outage, the system automatically switches to the backup power supply and sends a positioning code to the nearest rescue station through the Internet of Things communication module. The code includes the elevator ID, floor coordinates, and vital sign data of the trapped personnel. When an abnormal concentration gradient of dangerous gases is detected in the car, the acoustic imaging device is activated to locate the leakage source, and the environmental safety module starts emergency ventilation. If the situation continues to deteriorate, the fire protection system is linked to open the smoke exhaust channel in the hoistway and activate the color-coded guidance of the emergency lighting system.

[0072] Furthermore, the emergency dispatch control platform of the present invention includes:

[0073] A fault feature engineering unit that performs anomaly detection on sensor data based on a machine learning model and classifies the fault levels;

[0074] A rescue decision optimization engine that matches a preset rescue plan according to the fault type and dynamically adjusts the rescue steps based on real-time data;

[0075] A human-machine collaboration interface that is equipped with a voice emotion recognition algorithm and an AR augmented reality interface to achieve multi-modal emergency guidance and physiological parameter monitoring.

[0076] Among them, the fault feature engineering unit uses a temporal convolutional network (TCN) to extract the spatio-temporal features of sensor data, constructs a three-dimensional fault pattern recognition model, and uses a federated learning framework to aggregate the historical operation data of multiple elevators to train the model.

[0077] For example: when an earthquake causes the car to get stuck

[0078] The system adopts: an acceleration sensor detects continuous vibration, triggers the emergency locking of the car, then calls the earthquake rescue plan, gives priority to keeping the car stable, starts the balance adjustment after the vibration stops, uses hardware locking to prevent secondary injuries, and uses AI to dynamically evaluate the aftershock risk.

[0079] When there is a power outage + a passenger has a heart attack

[0080] The backup power supply maintains vital sign monitoring, and the Internet of Things module sends a medical help signal. Then, the rescue priority is adjusted to coordinate the synchronous arrival of an external ambulance and elevator rescue. Among them, the dual-link communication ensures that the distress signal will surely reach, and the resource scheduling algorithm optimizes the multi-party response.

[0081] The present invention forms a complete technical closed-loop from sensor data acquisition - edge preprocessing - cloud diagnosis - rescue instruction issuance - mechanical execution - effect feedback, further improving the success rate of composite fault handling, thereby ensuring the safety of personnel.

[0082] It will be understood that the present invention is described by way of some embodiments, and those skilled in the art will know that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the present invention. Additionally, under the teaching of the present invention, these features and embodiments can be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of the present application belong to the scope protected by the present invention.

Claims

1. An elevator emergency rescue control system, characterized in that, It consists of a state perception subsystem, an emergency rescue execution subsystem, and an emergency dispatch control platform, where: The state perception subsystem adopts a multi-source heterogeneous data fusion architecture to collect elevator mechanical state parameters and environmental characteristic parameters in real time; The emergency rescue execution subsystem is configured with a variety of physical execution units and communication relay devices to perform multi-modal rescue operations according to the fault type; The emergency dispatch control platform is used to coordinate the interaction between the monitoring and emergency rescue execution subsystems and dynamically optimize the rescue strategy through artificial intelligence algorithms; The state perception subsystem includes multiple types of sensors, remote monitoring multimedia terminals, edge computing terminals, and cloud data centers deployed in the elevator car, hoistway, and traction machine. The sensors include acceleration sensors, tension sensors, temperature and humidity sensors, and air pressure sensors.

2. The elevator emergency rescue control system according to claim 1, wherein The emergency rescue execution subsystem includes: The dynamic balance control module regulates the magnetorheological damper of the counterweight through a high-precision servo motor, combines with a vision positioning system to compensate for the car offset, and avoids the risk of secondary tilt; The Internet of Things communication module transmits fault signals and rescue instructions based on 5G and LPWAN dual links, and supports offline emergency communication in a network-disconnected environment; The self-rescue module includes a mechanical device driven by a manual / electric dual mode, and realizes the micro-lifting and micro-lowering of the car through bevel gear transmission for trapped people to operate independently; The environmental safety linkage module integrates a multi-spectral gas analyzer and a distributed temperature sensing network, and establishes a fire linkage protocol with the building BMS system.

3. The elevator emergency rescue control system according to claim 2, characterized in that, The dynamic balance control module monitors the tension of the traction steel wire rope in real time through a tension sensor, combines with the load data in the car, and dynamically adjusts the position of the counterweight block using the PID algorithm to ensure the stable docking of the car. It is also equipped with a safety verification mechanism, with double position feedback verification and an overload protection circuit.

4. An elevator emergency rescue control system according to claim 1, characterized in that, The emergency dispatch control platform includes: The fault feature engineering unit performs anomaly detection on sensor data based on a machine learning model and classifies the fault levels; The rescue decision optimization engine matches the preset rescue plan according to the fault type and dynamically adjusts the rescue steps based on real-time data; The human-machine collaboration interface is equipped with a voice emotion recognition algorithm and an AR augmented reality interface to achieve multi-modal emergency guidance and physiological parameter monitoring.

5. An elevator emergency rescue control system according to claim 4, characterized in that, The fault feature engineering unit applies a temporal convolutional network (TCN) to extract the spatio-temporal features of sensor data, constructs a three-dimensional fault pattern recognition model, and uses a federated learning framework to aggregate the historical operation data of multiple elevators to train the model.

6. An elevator emergency rescue control method, comprising an elevator emergency rescue control system according to any one of claims 1-5, characterized in that, It includes the following steps: S1. Real-time collect elevator operation state data through multi-sensor fusion technology and construct a multi-dimensional feature vector of the elevator operation state; S2. Perform feature-level data compression and initial screening of abnormal patterns at the edge computing node, generate weighted diagnostic indicators and abnormal signals, and upload them to the cloud; S3. Use a digital twin model for virtual-reality strategy verification, dynamically generate an emergency operation list including device linkage sequences and temporal constraints, and start the corresponding rescue mode based on the fault type, giving priority to balance adjustment and communication link self-repair; S4. Synchronize the rescue progress through the human-machine interaction interface and support remote intervention by external rescue personnel.

7. A method for controlling elevator emergency rescue according to claim 6, characterized in that, In the event of a power outage, the system automatically switches to the backup power supply and sends the positioning code to the nearest rescue station through the Internet of Things communication module. The code includes the elevator ID, floor coordinates, and vital sign data of the trapped personnel.

8. The elevator emergency rescue control method according to claim 6, characterized in that, When an abnormal concentration gradient of dangerous gases in the car is detected, the sound field imaging device is activated to locate the leakage source. The environmental safety module starts emergency ventilation. If the situation continues to deteriorate, the fire protection system is linked to open the smoke exhaust channel in the hoistway and activate the color-coded guidance of the emergency lighting system.

Images