CN113581961B - Automatic fault identification method for elevator hall door - Google Patents
Automatic fault identification method for elevator hall door Download PDFInfo
- Publication number
- CN113581961B CN113581961B CN202110911375.3A CN202110911375A CN113581961B CN 113581961 B CN113581961 B CN 113581961B CN 202110911375 A CN202110911375 A CN 202110911375A CN 113581961 B CN113581961 B CN 113581961B
- Authority
- CN
- China
- Prior art keywords
- door
- elevator
- hall door
- fault
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0018—Devices monitoring the operating condition of the elevator system
- B66B5/0025—Devices monitoring the operating condition of the elevator system for maintenance or repair
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0087—Devices facilitating maintenance, repair or inspection tasks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B50/00—Energy efficient technologies in elevators, escalators and moving walkways, e.g. energy saving or recuperation technologies
Landscapes
- Elevator Door Apparatuses (AREA)
- Maintenance And Inspection Apparatuses For Elevators (AREA)
Abstract
The invention relates to the field of nondestructive testing, in particular to an automatic fault identification method for an elevator hall door. The method comprises the following steps: s1: constructing a hoistway door fault detection model based on a machine learning algorithm; s2: and collecting the motion state data of the elevator hall door and drawing a corresponding motion state curve. S3: and taking the characteristic data corresponding to the marked motion state curve and the data of the fault type as a data set for model training, namely a training set. S4: and training the elevator hall door fault detection model through the training set, and verifying the training effect of the model by using the test set. S5: when a hall door switch command is received every time, acquiring characteristic data of a hall door motion state corresponding to an elevator layer, and identifying through an elevator hall door fault detection model; and the corresponding real-time motion state curve and the corresponding recognition result are reserved. The invention solves the problems that the existing elevator hall door state monitoring depends on manual work, the monitoring effect is poor, the efficiency is low and the real-time performance is insufficient.
Description
Technical Field
The invention relates to the field of nondestructive testing, in particular to an automatic fault identification method for an elevator hall door.
Background
The hoistway door is one of the most important parts of an elevator, and is also the most frequently used part that is most vulnerable to external impact or extrusion in a working environment. The national standard GB-7588 Elevator manufacturing and installation safety Specification separately provides design and installation requirements for the safety performance of the landing door. However, according to incomplete statistics, about 70% of casualty accidents of the elevator are caused by elevator hall doors, so that the hall doors play an extremely important role in protecting the safety of people and the normal operation of the elevator.
In daily use, the landing door frequently fails due to a plurality of reasons, such as foreign matters on the doorsill, the loss of an emergency guide device, external impact, the failure of a device for preventing people from being clamped by the door, and the like. In order to solve the problem, the conventional technical scheme only can continuously shorten the maintenance period of the elevator and improve the inspection frequency of managers. Although maintenance personnel can regularly maintain the elevator, the maintenance personnel cannot monitor the state of the elevator hall door in real time. And for the high-rise high-speed group control elevator group, the number of the landing doors of a single group of elevators can reach thirty or more, and maintenance personnel are difficult to master the running state of each landing door at any time and any place.
Disclosure of Invention
Based on the situation, the problem that the existing elevator hall door state monitoring depends on manual work, the monitoring effect is poor, the efficiency is low, the real-time performance is insufficient, and further the elevator faults are frequent is solved; provided is a method for automatically identifying a failure of a hoistway door.
The invention provides a method for automatically identifying the fault of an elevator hall door, which comprises the following steps:
s1: constructing a hoistway door fault detection model based on a machine learning algorithm; the input of the elevator hall door fault detection model is the characteristic data of the motion state of the elevator hall door in a switching period, and the output of the elevator hall door fault detection model is the judgment result of the fault type of the current elevator hall door.
S2: and collecting the motion state data of the elevator hall door in a normal state and various different fault type states, and drawing a corresponding motion state curve.
S3: manually marking the states of the hall doors represented by the drawn motion state curves; in the sampling data, the number ratio of the fault state to the normal state is 1:1, and the number ratio of various fault types is the same; and forming a data set by using the characteristic data corresponding to the marked motion state curve and the data of the fault type, and taking the data set as the data set of model training.
S4: the data set is divided into a training set and a test set according to the same sampling proportion as the previous step. Wherein, the ratio of the data volume in the training set and the test set is 8:2. Training the constructed elevator hall door fault detection model through a training set, and verifying the training effect of the model by using a test set; until the elevator hall door fault detection model reaches the requirements of the training phase.
S5: when a landing door switch command is received every time, acquiring characteristic data of the motion state of a landing door corresponding to an elevator layer, drawing a corresponding real-time motion state curve, and simultaneously inputting the characteristic data into a trained elevator landing door fault detection model to obtain the identification result of the landing door state of the layer; and the corresponding real-time motion state curve and the corresponding recognition result are reserved.
In order to solve the problem of identifying the faults of the elevator hall door, firstly, the characteristic data reflecting the change of the door state in the elevator hall door is obtained, the related data is drawn into curves of different types, meanwhile, the characteristic data is used for training a network based on a machine learning algorithm, and then a trained network model is used as a fault detection model of the elevator hall door required by the invention for automatically identifying the state of the elevator hall door.
Compared with the traditional manual detection method, the method has the advantages of higher efficiency and capability of synchronously detecting the states of a plurality of elevator hall doors simultaneously, and effectively solves the problem of insufficient real-time monitoring of the states of the elevator hall doors. Meanwhile, the index significance of the state data acquired by the invention is stronger, and the autonomous learning capability of the neural network is stronger, so that the reliability of the fault automatic identification result is better, and the requirement of fault monitoring on the elevator hall door can be met. Even under certain conditions, the elevator hall door fault detection model provided by the invention can identify faults which are difficult to find by manual detection, or can find hidden dangers in time at the initial stage of fault occurrence.
In the invention, the result and the characteristic data of each recognition and the motion state curve are saved, so that the recognition result can be manually rechecked in combination with the later elevator maintenance process, the related data can be saved for the recognition error result, and the manual mark of the recognition result is modified to be used as the training data for subsequently improving the performance of the elevator hall door fault detection model. The elevator hall door fault detection model provided by the invention has stronger identification performance.
As a further improvement of the present invention, in step S2, the fault types include: the locking state that threshold foreign matter led to, the not hard up state of door system change gear, the disappearance state of door guide shoe. The three faults are the most frequent hall door fault types in practical application, data acquisition can be performed on other types of hall door faults, and then the hall door faults can be used for training the model in the invention.
As a further improvement of the present invention, in step S2, the collected motion state data of the hall door includes three-axis velocity, angular velocity, and acceleration, and displacement in three axial directions. The three-axis speed, the angular speed and the acceleration are obtained through an inertia measuring unit, and the displacement in the three axial directions is obtained through a displacement sensor.
As a further improvement of the invention, in step S3, typical motion state curves of the hoistway door in various different state types are drawn according to the acquired feature data of the motion state of the hoistway door in a normal state and in various different fault type states; then comparing a sample motion state curve drawn by the collected sample data with a typical motion state curve, and determining the state type of the elevator hall door corresponding to each sample motion state curve; and completing manual marking according to the judgment result of the state type.
As a further improvement of the invention, in step S5, when a failure of a hoistway door of a certain floor is identified, an alarm signal is also sent to a manager; and sending a control instruction for stopping the operation of the landing door to the elevator control system until the corresponding fault is removed. In the invention, when the corresponding hoistway door is detected to have a fault, the corresponding hoistway door needs to be closed in time to avoid casualty events, and meanwhile, an alarm needs to be sent to a manager to wait for a technician to maintain and process the corresponding fault.
As a further improvement of the invention, in step S5, when receiving any one hall door switch command, the hall door motion state data of the rest layers are also obtained, and whether the hall door has a change in speed, acceleration or displacement in each axial direction is detected, if yes, a control command for stopping the operation of the hall door in the layer is sent to the elevator control system, and meanwhile, an early warning signal is sent to a manager to remind the manager to check.
As a further improvement of the invention, the generation method of the early warning signal comprises the following steps: setting a threshold value of the axial speed, the acceleration, the angular speed and the axial displacement of the landing door, wherein the threshold values of the axial speed, the acceleration, the angular speed and the axial displacement are respectively the minimum values of the speed, the acceleration, the angular speed and the displacement change of the landing door of the elevator in a non-switching state, and when the speed, the acceleration, the angular speed or the displacement change of any one layer of the landing door of the elevator in the non-switching state is greater than the threshold values of the axial speed, the acceleration, the angular speed or the axial displacement, judging that the landing door of the layer in the elevator is in a fault early warning state, and waiting for the confirmation of a manager.
Normally, the elevator will only stop at one of the floors at the same time, and when the elevator is in an open or closed state at one of the landing doors, the remaining landing doors should remain closed. However, if any one of the other elevator hall doors is detected to have a displacement larger than the safety limit, the hall door at the floor is considered to be possibly broken or damaged by external force, and at the moment, a control instruction for stopping the operation of the hall door at the floor is sent to an elevator control system, and meanwhile, an early warning signal is sent to a manager to wait for a technician to perform fault checking and problem processing.
As a further improvement of the present invention, in step S5, after the corresponding real-time motion state curve and the identification result thereof are retained each time, a fault probability database for indicating the frequency of various faults occurring in the whole life cycle of the hoistway door is updated, and the fault probability database is used for guiding the production and maintenance process of the elevator.
The method for automatically identifying the faults of the elevator hall door has the following beneficial effects:
1. the elevator hall door sensor is arranged on the elevator hall door, the running state of the elevator hall door is monitored in real time, and whether the elevator hall door breaks down or not is judged according to the acquired characteristic data reflecting the motion state of the elevator hall door. The monitoring mode can replace manual work, realizes 24h uninterrupted monitoring, has better real-time performance, and can find and process in time at the beginning of a fault. The personal safety or property loss caused by the failure of the elevator hall door is reduced. In addition, the scheme of the invention can realize centralized management of large-scale group control elevators and obviously improve the efficiency of elevator safety monitoring.
2. The elevator hall door state recognition method based on the machine learning algorithm completes the recognition of the elevator hall door state through the network model based on the machine learning algorithm, and can remarkably improve the training effect of the network model based on a large amount of real data. Meanwhile, in practical application, more effective training samples can be accumulated through manual rechecking of the recognition result, and the training effect and the recognition accuracy of the network model are further improved.
3. Besides the operation fault of the elevator hall door, the invention can also timely detect the external force impact on the elevator hall door, timely provide a processing scheme according to the damage degree of the impact on the elevator hall door, close the hall door on the related layer and inform technical personnel to arrange maintenance.
4. The invention can also accumulate the data of the elevator in the running process and provide prospective feedback and suggestions for the production and manufacture of the elevator hall door according to the problems reflected in the accumulated data. And meanwhile, technical support is provided for the daily operation management of the elevator.
Drawings
Fig. 1 is a flowchart of a method for automatically identifying a failure of a hoistway door according to embodiment 1 of the present invention;
fig. 2 is a picture of an apparatus for an elevator hall door in a state of a threshold foreign matter in embodiment 1 of the present invention;
fig. 3 is a picture of the device of the hoistway door in the roller-released state in embodiment 1 of the present invention;
fig. 4 is a picture of the device for the hoistway door in the state that the hoistway door guide shoe is loosened in embodiment 1 of the present invention;
fig. 5 is a schematic block diagram of a failure monitoring system for hoistway doors provided in embodiment 2 of the present invention;
fig. 6 is a schematic view of a topology of a failure monitoring system for hoistway doors provided in embodiment 2 of the present invention;
fig. 7 is a schematic view of a topology of a failure monitoring system for hoistway doors provided in embodiment 3 of the present invention;
fig. 8 is a comparison graph of the moving state curves of the hoistway door in the threshold foreign matter and normal state in embodiment 3 of the present invention;
fig. 9 is a comparison graph of the moving state curves of the hoistway door in the roller releasing and normal states in embodiment 3 of the present invention;
fig. 10 is a comparison graph of the movement state curve of the elevator hall door in the loose and normal state of the hall guide shoes in embodiment 3 of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "or/and" includes any and all combinations of one or more of the associated listed items.
Example 1
The method for automatically identifying the faults of the elevator hall door provided by the embodiment comprises the following steps:
s1: constructing a hoistway door fault detection model based on a machine learning algorithm; the input of the hoistway door fault detection model is characteristic data of a motion state of the hoistway door in one switching cycle, and specifically, in the present embodiment, the input of the hoistway door fault detection model is three axial speeds, angular speeds and accelerations of the elevator door. And the output of the elevator hall door fault detection model is the judgment result of the fault type of the current elevator hall door.
S2: and collecting motion state data of the elevator hall door in a normal state and various different fault type states, wherein the motion state data is input required by the elevator hall door fault detection model in a training or identification stage. The related characteristic data are also plotted into corresponding motion state curves in the embodiment. The purpose of drawing the motion state curve is mainly to facilitate manual marking of the motion state of the elevator hall door according to the motion state curve and distinguish different types of motion faults.
The types of faults that can be identified in this embodiment mainly include: the locking state caused by threshold foreign bodies, the loosening state of a door system change gear, the missing state of a door guide shoe and the like. Fig. 2 shows a picture of the device in the threshold foreign matter state; FIG. 3 shows a picture of the device with the roller released; fig. 4 shows a picture of the apparatus in a state where the hall shoe is loosened.
The three faults are the most frequently occurring types of hall door faults in practical application, and data collection can be performed on other types of hall door faults, so that the three faults are used for training the model in the embodiment. And for elevator hall door fault types except the three types of fault types, the addition is completed according to the proportion of the fault types in practical application. That is, when the probability of occurrence of a new fault type is higher, the proportion of the fault type in the training data set is properly increased, and the optimal recognition effect on the problem is further ensured.
The motion state data of the hall door collected in this embodiment includes three-axis velocity, angular velocity, acceleration, and displacement in three axial directions. The three-axis speed, the angular speed and the acceleration are obtained through an inertia measuring unit, and the displacement in the three axial directions is obtained through a displacement sensor. The data acquired by the inertia measuring unit is characteristic data for identifying the failure of the hall door, and the data measured by the displacement sensor is used for judging whether the hall door is deformed due to the external force when the hall door is in a static state, and the failure of the hall door is a precursor when the structure is deformed. In many cases, just because the hoistway door is damaged by external force, the switching state of the hoistway door is abnormal, and further, a safety accident is caused.
S3: manually marking the states of the hall doors represented by the drawn motion state curves; in the sampling data, the number ratio of the fault state to the normal state is 1:1, the number ratio of various fault types is the same, and the characteristic data corresponding to the marked motion state curve and the data of the fault type form a data set which is used as a data set for model training. The data set in the embodiment can well reflect the proportion of various faults in the actual situation, and further achieve a better model training effect.
The manual labeling process for the data set is as follows: drawing typical motion state curves of the elevator hall door under various different state types according to the acquired feature data of the motion state of the elevator hall door under the normal state and various different fault type states; then comparing a sample motion state curve drawn by the collected sample data with a typical motion state curve, and determining the state type of the elevator hall door corresponding to each sample motion state curve; and completing manual marking according to the judgment result of the state type.
S4: the data set is divided into a training set and a test set according to the same sampling proportion as the previous step. Wherein, the ratio of the data volume in the training set and the test set is 8:2. Training the constructed elevator hall door fault detection model through a training set, and verifying the training effect of the model by using a test set; until the elevator hall door fault detection model reaches the requirements of the training phase.
S5: when a landing door switch command is received every time, acquiring characteristic data of the motion state of a landing door corresponding to an elevator layer, drawing a corresponding real-time motion state curve, and simultaneously inputting the characteristic data into a trained elevator landing door fault detection model to obtain the identification result of the landing door state of the layer; and the corresponding real-time motion state curve and the corresponding recognition result are reserved.
When the failure of the elevator hall door at a certain floor is identified, the embodiment also sends an alarm signal to the manager; and sending a control instruction for stopping the operation of the landing door to the elevator control system until the corresponding fault is removed.
According to the scheme in the embodiment, when the corresponding hoistway door is detected to be in fault, the corresponding elevator hoistway door can be closed in time, so that casualty events are avoided, meanwhile, an alarm can be sent to managers, and technicians are waited to maintain and process the corresponding fault.
The real-time motion curve of the elevator hall door in each identification and detection process is drawn, the purpose of drawing the curve is to facilitate manual rechecking of the identification result of the network model when necessary, and retrain the machine-learned network model according to the rechecking result at the later stage, wherein the result of the network model identification error can be supplemented into a training set after being manually marked, and the network model is retrained, so that the training effect of the model is improved.
In this embodiment, when receiving any one hall door switch instruction, the hall door motion state data of the rest of layers is also acquired, whether the hall door has speed, acceleration or displacement change in each axial direction is detected, if yes, a control instruction for stopping the operation of the hall door on the layer is sent to an elevator control system, and meanwhile, an early warning signal is sent to a manager to remind the manager to check.
The generation method of the early warning signal comprises the following steps: setting a threshold value of the axial speed, the acceleration, the angular speed and the axial displacement of the landing door, wherein the threshold values of the axial speed, the acceleration, the angular speed and the axial displacement are respectively the minimum values of the speed, the acceleration, the angular speed and the displacement change of the landing door of the elevator in a non-switching state, and when the speed, the acceleration, the angular speed or the displacement change of any one layer of the landing door of the elevator in the non-switching state is greater than the threshold values of the axial speed, the acceleration, the angular speed or the axial displacement, judging that the landing door of the layer in the elevator is in a fault early warning state, and waiting for the confirmation of a manager.
Normally, the elevator will only stop at one of the floors at the same time, and when the elevator is in the open or closed position at one of the landing doors, the remaining landing doors should remain closed. However, if any one of the other hoistway doors is detected to have a displacement larger than the safety limit, the hoistway door on the floor is considered to be possibly broken or damaged by external force, and the hoistway door on the floor must be closed and locked in time, and a technician is waited for fault checking and problem processing.
In order to solve the problem of identifying the failure of the hoistway door, the method comprises the steps of firstly obtaining characteristic data reflecting the change of the door state in the hoistway door, drawing the related data into different types of curves, simultaneously training a network based on a machine learning algorithm by using the characteristic data, and further taking a trained network model as a failure detection model of the hoistway door required by the invention for automatically identifying the state of the hoistway door.
Compared with the traditional manual detection method, the method has the advantages of higher efficiency and capability of synchronously detecting the states of a plurality of elevator hall doors simultaneously, and effectively solves the problem of insufficient real-time monitoring of the states of the elevator hall doors. Meanwhile, the index significance of the state data acquired by the embodiment is strong, and the autonomous learning capability of the neural network is strong, so that the reliability of the fault automatic identification result is better, and the requirement of fault monitoring on the elevator hall door can be met. Even in some cases, the elevator hall door fault detection model provided by the embodiment can identify faults which are difficult to find by manual detection, or find hidden dangers in time at the initial stage of fault occurrence.
In addition, in step S5, after the corresponding real-time motion state curve and the identification result thereof are retained each time, a fault probability database for indicating the frequency of various faults occurring in the whole life cycle of the hoistway door is updated, and the fault probability database is used for guiding the elevator production and maintenance process. After the elevator manufacturer or the maintenance enterprise obtains the data in the corresponding fault probability database, the most common faults encountered in the operation process of the elevator hall door can be known, and further, the performance of the corresponding elevator component can be enhanced, and the effect of reducing the fault rate is achieved. Meanwhile, the maintenance company can increase the maintenance frequency of corresponding parts of the elevator according to various frequent faults and the safe operation of the elevator is guaranteed.
Example 2
The embodiment provides a fault monitoring system for elevator hall doors, which adopts the automatic fault identification method for the elevator hall doors as in embodiment 1 to identify the motion state of the hall doors in each layer of elevator, so as to realize the timely monitoring of various types of faults. As shown in fig. 5, the fault monitoring system includes: the system comprises a hall sensor, a signal transmission module, a data relay processor, a communication module and a cloud server. The topology of the system is shown in fig. 6.
Wherein the hall sensor includes an inertia measuring unit and a displacement measuring unit. The inertial measurement unit comprises three single-axis acceleration sensors and three single-axis gyroscopes; the inertia measurement unit is used for measuring speed signals, acceleration signals and angular speed signals of the independent three axes. The displacement measurement unit comprises three single-axis displacement sensors and is used for measuring displacement signals of independent three axes. The hall sensors are respectively arranged on the hall doors of each floor of the elevator.
The signal transmission module is used for transmitting the detection signal of the hall sensor to a data relay processor. The signal transmission module is mainly used to transmit the detection result of the hall sensor to the data relay processor, and thus the module should be able to realize data transmission between a large number of modules (door state sensors) and a single module (data relay processor). In this embodiment, data transmission is realized by using a one-to-many master-slave bluetooth module, and considering that the hall sensor does not need to receive data, the bluetooth slave modules of the master-slave bluetooth module in this embodiment are installed at the hall sensor side, and the bluetooth master module is installed at the data relay processor side. The master and slave bluetooth modules operate in a one-way data transfer mode, i.e. data is only allowed to be transferred from the bluetooth slave to the bluetooth master.
Of course, in other embodiments, other data transmission modules may be added to solve the problem of real-time data transmission. The characteristic data for identifying the failure state of the elevator hall door in the embodiment only needs to be collected and transmitted when the elevator car reaches the corresponding hall door, and the bluetooth module in the embodiment just can meet the transmission requirement of the data. In addition, the bluetooth module adopted in this embodiment also has the characteristics of wireless transmission, strong confidentiality (the matching connection between the master module and the slave module needs to be completed in advance), and low power consumption.
The data relay processor is installed on the top of the elevator car and is in communication with the elevator control system. The data relay processing is a black box for installing the elevator car roof. The data relay processor in this embodiment is configured to:
(1) After the switching control command of any one landing door sent by the elevator control system is inquired, the detection signal of the landing door sensor in the landing door on the floor is obtained, and the corresponding detection signal is converted into characteristic data reflecting the movement state change of the elevator landing door. And drawing a motion state curve reflecting the change of the motion state of the elevator hall door according to the characteristic data, and sending out the characteristic data of the motion state.
(2) After receiving the fault type of a certain floor of elevator hall door sent by a cloud server; and sending a control command for stopping the operation of the landing door of the floor and an alarm signal for indicating the stop of the operation of the landing door of the floor to an elevator control system.
(3) When the elevator control system responds to the switch control command of any one elevator hall door, the displacement signals detected in the rest elevator hall doors are obtained. And when the displacement in any direction of any other hall door exceeds the preset displacement limit, sending an early warning signal for representing the hall door fault to be checked.
Specifically, the data relay processor comprises an instruction transceiving unit and a computing unit; the command transceiving unit is used for inquiring the switching control command of any landing door generated by the elevator control system or sending a control command and an alarm signal for stopping the operation of the landing door on a certain floor to the elevator control system. And the computing unit is used for converting the acquired sensing signals of the hall door sensor into characteristic data reflecting the motion state change of the elevator hall door. The data relay processor also comprises a storage unit which is used for storing the receiving and transmitting signals of the command receiving and transmitting unit and the calculation result of the calculation unit as the data backup of the elevator running state.
The communication module is used for realizing bidirectional communication between the data relay processing and a cloud server. Specifically, the communication module in the present embodiment adopts a communication module based on a communication network of a mobile operator. The communication module supports 2G, 3G, 4G or 5G networks of three major operators in the mainstream. And the communication module sends the characteristic data obtained from the data relay processor during the movement of the elevator hall door to the cloud server end for fault identification, and sends the fault type identified by the cloud server end back to the data relay processor.
The cloud server comprises a trained elevator hall door fault detection model; the model is the hoistway door fault detection model which is trained in the embodiment 1. And the cloud server is used for inputting the acquired characteristic data of the motion state of the hall door on any layer into the elevator hall door fault detection model to obtain a fault state detection result of the hall door on the layer. And after identifying the type of failure of the hall door, sends out the identification result of the type of failure.
Example 3
The present embodiment provides a failure monitoring system for a hoistway door, which is different from embodiment 2 in that: as shown in fig. 7, the system in this embodiment further includes a local data server and a local application server; all data received and generated in the cloud server are stored in the local data server; the local application server is used for responding to the request of the manager and providing the manager with the access service of all data in the local data server.
Specifically, in this embodiment, the manager opens a browser in the local application server to access the server with the fixed domain name, or inputs a password in software to log in a corresponding account, so as to obtain data in the local data server, and browse the movement state of the hall door of each elevator. And at the local application server end, the real-time motion states of all elevators are displayed in a motion state curve mode. The manager can not only check the real-time status data of the current elevator hall door, but also inquire the historical data of any hall door. And manually rechecking the identification result of the hall door fault state according to the hall door motion state curve of the browsing interface. In order to facilitate the manual review process, a curve of the movement state of the hoistway door in the normal state can be generated in the local data server and used as a reference. Specifically, fig. 8 to 10 show graphs comparing a curve of a state of motion of the hall door with a curve of a normal state in three states of a threshold foreign matter, a roller release, and a looseness of the hall door shoe, respectively.
The above examples are merely illustrative of several embodiments of the present invention, and the description thereof is more specific and detailed, but not to be construed as limiting the scope of the invention. It should be noted that various changes and modifications can be made by those skilled in the art without departing from the spirit of the invention, and these changes and modifications are all within the scope of the invention. Therefore, the protection scope of the present invention should be subject to the appended claims.
Claims (7)
1. A failure automatic identification method of elevator hall door is characterized in that: which comprises the following steps:
s1: constructing a hoistway door fault detection model based on a machine learning algorithm; the input of the elevator hall door fault detection model is characteristic data of the motion state of the elevator hall door in a switching period, and the output of the elevator hall door fault detection model is a fault type judgment result of the current elevator hall door;
s2: collecting characteristic data of the motion states of the elevator hall door in a normal state and various different fault type states, and drawing a corresponding motion state curve; the collected motion state data of the hall door comprise three-axis angular velocity, acceleration and speed and three axial displacements; the three-axis angular velocity, the acceleration and the speed are obtained through an inertial measurement unit, and the three axial displacements are obtained through a displacement sensor;
the fault types include: the locking state caused by the threshold foreign matter, the loosening state of a door system change gear and the missing state of a door guide shoe; the proportion of the three parts in the data set for model training is the same;
s3: manually marking the states of the hall doors represented by the drawn motion state curves; in the sampled data, the quantity ratio of the fault state to the normal state is 1:1, the quantity ratios of various fault types are the same, and a data set formed by the marked feature data of the hoistway door motion state and the corresponding fault types is used as a data set for model training;
the process of manually marking the data set is as follows: drawing typical motion state curves of the elevator hall door under various different state types according to the acquired feature data of the motion states of the elevator hall door under the normal state and various different fault type states; then comparing a sample motion state curve drawn by the collected sample data with a typical motion state curve, and determining the state type of the elevator hall door corresponding to each sample motion state curve; and completing manual marking according to the judgment result of the state type;
s4: dividing the data set into a training set and a testing set according to the same sampling proportion as the previous step; training the constructed elevator hall door fault detection model through a training set, and verifying the training effect of the model by using a test set; until the elevator hall door fault detection model meets the requirements of a training phase;
s5: when a landing door switch command is received every time, acquiring characteristic data of a landing door motion state corresponding to an elevator, drawing a corresponding real-time motion state curve, inputting the characteristic data into the trained elevator landing door fault detection model, and obtaining an identification result of the landing door state of the elevator layer; and the corresponding real-time motion state curve and the recognition result thereof are reserved.
2. The method of automatically recognizing a failure of a hall door according to claim 1, wherein: the fault types further include other fault states besides a stuck state caused by a threshold foreign body, a loose state of a door system change gear, or a missing state of a door guide shoe, and samples of the other fault states are added to the data set in proportion to the occurrence frequency of all the fault states.
3. The method of automatically recognizing a failure of a hall door according to claim 1, wherein: in step S4, the ratio of the data volumes in the training set and the test set is 8:2.
4. The method of automatically recognizing a failure of a hall door according to claim 1, wherein: in step S5, when the failure of the elevator hall door at a certain floor is identified, an alarm signal is also sent to a manager; and sends a control instruction for stopping the operation of the landing door on the floor to the elevator control system until the corresponding fault is removed.
5. The method of automatically recognizing a failure of a hoistway door of claim 1, wherein: in step S5, when any hall door switch instruction is received, the motion state data of the hall doors of other layers are also obtained, and whether the hall doors of other layers have the change of speed, acceleration or displacement in each axial direction is detected; if so, a control instruction for stopping the operation of the landing door on the floor is sent to the elevator control system, and an early warning signal is sent to the manager to remind the manager to check.
6. The method of automatically recognizing a failure of a hall door according to claim 5, wherein: the generation method of the early warning signal comprises the following steps: and when the change of the speed, the acceleration, the angular speed or the displacement of any one layer of the landing door in the non-switching state is greater than the threshold value of the axial speed, the acceleration, the angular speed or the displacement, the landing door in the layer in the elevator is judged to be in a fault early warning state, and the confirmation of a manager is waited.
7. The method of automatically recognizing a failure of a hall door according to claim 1, wherein: in step S5, after a corresponding real-time motion state curve and an identification result thereof are reserved each time, a fault probability database for representing the occurrence frequency of various faults in the whole life cycle of the elevator hall door is updated, and the fault probability database is used for guiding the production and maintenance process of the elevator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110911375.3A CN113581961B (en) | 2021-08-10 | 2021-08-10 | Automatic fault identification method for elevator hall door |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110911375.3A CN113581961B (en) | 2021-08-10 | 2021-08-10 | Automatic fault identification method for elevator hall door |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113581961A CN113581961A (en) | 2021-11-02 |
| CN113581961B true CN113581961B (en) | 2023-03-28 |
Family
ID=78256741
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110911375.3A Active CN113581961B (en) | 2021-08-10 | 2021-08-10 | Automatic fault identification method for elevator hall door |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113581961B (en) |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110498314B (en) * | 2019-08-28 | 2020-11-10 | 上海电气集团股份有限公司 | Health assessment method and system for elevator door system, electronic device and storage medium |
| CN110894038B (en) * | 2019-11-20 | 2021-06-11 | 上海电气集团股份有限公司 | Method and device for predicting running state of elevator door system |
| CN110817636B (en) * | 2019-11-20 | 2021-09-21 | 上海电气集团股份有限公司 | Elevator door system fault diagnosis method, device, medium and equipment |
| CN110790105B (en) * | 2019-11-20 | 2021-11-16 | 上海电气集团股份有限公司 | Elevator door system diagnosis and decline time prediction method and diagnosis and prediction system |
| CN113128322A (en) * | 2020-01-16 | 2021-07-16 | 宁波微科光电股份有限公司 | Elevator sill detection method |
| CN112478975A (en) * | 2020-12-09 | 2021-03-12 | 浙江新再灵科技股份有限公司 | Elevator door fault detection method based on audio features |
| CN112938684B (en) * | 2021-03-22 | 2022-05-17 | 大连奥远电子股份有限公司 | Elevator operation track analysis system |
| CN112960503B (en) * | 2021-03-22 | 2022-09-13 | 大连奥远电子股份有限公司 | Elevator car door running track modeling method |
-
2021
- 2021-08-10 CN CN202110911375.3A patent/CN113581961B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113581961A (en) | 2021-11-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113581962B (en) | Fault monitoring system of elevator hall door | |
| CN111650919B (en) | Multi-dimensional monitoring escalator fault prediction and health management method and system | |
| CN111562096B (en) | Real-time online health state monitoring system of escalator | |
| CN204280939U (en) | A kind of elevator operational monitoring forewarn system | |
| CN109019216B (en) | Intelligent diagnosis system and method for elevator faults | |
| WO2020147711A1 (en) | Elevator operation status monitoring method and device | |
| CN110143504A (en) | A kind of elevator rescue channel based on fault pre-alarming mechanism | |
| EP3674242B1 (en) | Enhancing elevator sensor operation for improved maintenance | |
| US20220388810A1 (en) | Method, device, and early warning system for monitoring elevator health state | |
| US20200087111A1 (en) | Sensor-based shutdown detection of elevator system | |
| CN110015601A (en) | Analyze the tele-control system and method for elevator faults reason | |
| JP2020104611A (en) | Platform door state diagnosis system, platform door and platform door state diagnosis method | |
| CN113581961B (en) | Automatic fault identification method for elevator hall door | |
| CN112499418B (en) | Magnetic induction elevator operation data acquisition system and acquisition method thereof | |
| CN109823932A (en) | Active safety elevator | |
| CN110255310B (en) | Remote elevator monitoring system and method for cooperatively monitoring property data | |
| CN112225027B (en) | Photoinduction elevator operation data acquisition system and acquisition method thereof | |
| CN106081834A (en) | A kind of mine vertical shaft elevator unmanned control system and control method thereof | |
| CN211741892U (en) | Multi-dimensional state online monitoring system for escalator | |
| CN111302176A (en) | Lifting device monitored control system for building | |
| CN220664574U (en) | High-speed elevator safety clamp brake parameter measuring device | |
| KR102545137B1 (en) | Control board for escalator and remote integrated control system including the same | |
| CN212623713U (en) | Shuttle vehicle monitoring device and system and shuttle vehicle | |
| CN112093605B (en) | Elevator car running state recognition system and recognition method thereof | |
| CN108516432A (en) | A kind of elevator reparing maintenance system based on Internet of Things |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |