CN210558707U - Elevator operation safety detection and early warning identification system based on intelligent hardware - Google Patents
Elevator operation safety detection and early warning identification system based on intelligent hardware Download PDFInfo
- Publication number
- CN210558707U CN210558707U CN201920875728.7U CN201920875728U CN210558707U CN 210558707 U CN210558707 U CN 210558707U CN 201920875728 U CN201920875728 U CN 201920875728U CN 210558707 U CN210558707 U CN 210558707U
- Authority
- CN
- China
- Prior art keywords
- elevator
- car
- wireless network
- network bridge
- early warning
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Indicating And Signalling Devices For Elevators (AREA)
- Maintenance And Inspection Apparatuses For Elevators (AREA)
Abstract
The utility model relates to an elevator operation safety inspection and early warning identification system based on intelligent hardware, include: the intelligent camera, the microphone, the loudspeaker and the emergency call button are placed in the car; the system comprises a wireless network bridge, a computing unit, an acceleration sensor and an air pressure sensor, wherein the wireless network bridge, the computing unit, the acceleration sensor and the air pressure sensor are placed on the car roof; a photoelectric sensing sensor placed at the car door; a wireless network bridge placed in the elevator machine room; a wireless network bridge placed on the roof; a wireless bridge located at a property office; the cloud server is placed at the cloud end; and any networked computer or mobile terminal. The wireless network bridge of the elevator car roof, the wireless network bridge of the machine room, the wireless network bridge of the roof and the wireless network bridge of the property office are networked through wireless connection and connected with the cloud server to perform data interaction; the utility model has the advantages that, the security is high, with low costs, intelligent degree is high, convenient stable.
Description
Technical Field
The utility model belongs to the technical field of elevator safety, concretely relates to elevator operation safety inspection and early warning identification system based on intelligent hardware.
Background
With the development of the internet of things and the IOT technology in recent years, more hardware starts to access the internet, and great convenience and safety are brought to the life of people. However, in the elevator industry and field, the intelligent internet of things technology is not applied and played. Elevator safety accidents frequently occur, technical barriers of various major brands are high, equipment faults can not be early warned in advance in time, passenger dangerous behaviors can not be stopped in advance in time, monitoring is not comprehensive, elevator inspection is inconvenient, the elevator talkback communication system is high in limitation, maintenance work is not in place, elevator safety supervision is not strong, a series of problems such as shortage of elevator industry employees are solved before the elevator safety supervision is carried out, in addition, the newly-installed elevator maintenance amount in China still continuously increases every year, the old elevator maintenance amount also continuously increases, and therefore the elevator safety brings huge challenges to the whole industry.
The whole elevator industry including elevator manufacturers and corresponding industry companies wants to solve the above pain points in the elevator industry at present through a set of service or system, so as to improve the ecology of the elevator industry. The company deposits the elevator industry for many years and accumulates profound technology and resources. The intelligent elevator safety detection and early warning identification system is developed based on intelligent hardware.
The existing Chinese patent document CN201510299754.6 discloses a detection and early warning method of an elevator online safety detection and early warning device, and the main technical scheme is as follows: a detection and early-warning method of an elevator online safety detection and early-warning device comprises the following steps in sequence: (1) acquiring collected data of a characteristic parameter of an elevator state; (2) extracting a maximum value M1 of the characteristic parameter in the process of starting and stopping a continuous elevator, judging whether an early warning threshold value T1 is generated or not, if the early warning threshold value T1 of the parameter is not generated, adding M1 into a data sample library D1, judging whether the number of samples in the data sample library D1 reaches a set number V1 or not, and if the number of samples does not reach the set number V1, returning to the first step to continuously acquire acquired data; if the set number V1 is reached, establishing an elevator state probability neural network model, generating an early warning threshold value T1, and returning to the first step to obtain the acquired data; (3) if the early warning threshold T1 is generated, adding M1 into a data updating sample library D2, judging whether the number of samples in the data updating sample library D2 reaches a set number V2, wherein V2 is more than 0 and less than V1, and returning to the first step to obtain the acquired data if the number of the samples does not reach the set number V2; if the set number V2 is reached, counting the number K exceeding T1, wherein K is more than or equal to 0, and entering the next step; (4) judging whether K is larger than or equal to V2/2, if K is larger than or equal to V2/2, judging that the data is early-warning, sending early-warning information through the near-distance communication unit, and successively replacing the earliest V2-K data in the data sample library D1 with the remaining V2-K normal data; if K is less than V2/2, judging that the data are normal, and successively replacing the earliest V2 data in a data sample library D1 by the V2 data, wherein K abnormal data are added into D1 to serve as training samples of the probability model; (5) reestablishing an elevator state probability neural network model of the characteristic parameter, and updating an early warning threshold T1; (6) returning to the first step to obtain the collected data; the early warning device comprises a car early warning device and a machine room early warning device, the car early warning device comprises a first sensing unit, a first acquisition unit, a first processing unit and a first near field communication unit, the machine room early warning device comprises a second sensing unit, a second acquisition unit, a second processing unit, a second near field communication unit and a remote communication gateway, wireless communication is carried out between the first near field communication unit and the second near field communication unit, and the second processing unit is in remote communication with a remote platform through the remote communication gateway. Different from the technical proposal of the utility model.
SUMMERY OF THE UTILITY MODEL
The utility model aims at solving the above problem, providing an elevator operation safety inspection and early warning identification system based on intelligent hardware, it can realize: the intelligent elevator system has the intelligent functions of monitoring elevator car pictures in real time, monitoring whether people exist, monitoring up and down of an elevator, detecting floors where the elevator is located, detecting an opening and closing door of the elevator, monitoring the running speed of the elevator, having a voice stop reporting function, automatically alarming for sleepers, identifying and early warning strangers, statistically analyzing elevator running data, intelligently monitoring and early warning elevator faults, intelligently monitoring and early warning abnormal passenger behaviors, visually talkback for cross-platform emergency faults, positioning for receiving alarms of maintenance personnel, automatically monitoring maintenance and managing intelligent properties and the like.
In order to achieve the above object, the utility model provides a following technical scheme:
an elevator operation safety detection and early warning identification system based on intelligent hardware comprises an intelligent camera, a microphone, a loudspeaker and an emergency call button which are arranged in an elevator car, wherein the elevator car top is provided with a car top wireless bridge, a computing unit, an acceleration sensor and an air pressure sensor, the position of the elevator car door is provided with a photoelectric sensing sensor, and an elevator machine room is provided with a wireless bridge; the computing unit is internally provided with an information sending/receiving module, a judging module and an information processing module;
the car roof wireless network bridge is sequentially connected with a machine room wireless network bridge, a roof wireless network bridge and a property roof wireless network bridge in a wireless mode, the property roof wireless network bridge is arranged on the roof of a property office, the property office is provided with a property computer end, a cloud server is deployed at the cloud end, and the cloud server is connected with the property roof wireless network bridge through property optical fibers or broadband; the property computer terminal and the mobile phone terminal are connected with the cloud server in a wired or wireless mode;
the intelligent camera is connected with the computing unit through a car roof wireless network bridge, the acceleration sensor, the air pressure sensor, the photoelectric sensing sensor, the microphone, the loudspeaker and the emergency call button are all connected with the computing unit and transmit signals to the computing unit, and the computing unit is connected with the cloud server through the car roof wireless network bridge, the machine room wireless network bridge, the roof wireless network bridge and the property and property roof wireless network bridge;
the computing unit is internally provided with a deep neural network image recognition algorithm and a face recognition algorithm, and the face characteristic values of each person are extracted, marked and stored by analyzing video signals and comparing the face characteristic values;
the system is also provided with a UPS emergency power supply, and the system can still work normally under the condition of power failure.
Furthermore, the intelligent camera is used for acquiring a video picture in the car, identifying the opening and closing state of the car door of the elevator, identifying a floor display panel in the car of the elevator and transmitting information to the computing unit;
the acceleration sensor is used for identifying and monitoring the up-down and static states of the elevator and feeding back signals to the computing unit;
the air pressure sensor is used for identifying the floor where the elevator is located and feeding back a signal to the computing unit;
the photoelectric sensing sensor is combined with the calculating unit to realize the identification and monitoring of whether the elevator car door is opened or closed.
By combining an intelligent camera, a deep neural network image recognition algorithm and a face recognition algorithm, strangers are automatically recognized and early warned;
the intelligent camera and the computing unit are combined, so that the 'abnormal door opening and closing' fault and 'non-flat floor stopping' fault of the elevator are recognized and early-warned;
the fault of ' car shaking ' and the fault of abnormal running speed ' of the elevator are recognized and early-warned by combining the computing unit and the acceleration sensor;
the intelligent camera and the deep neural network image recognition algorithm are combined, so that the recognition, monitoring and alarming of abnormal behaviors of passengers in the elevator car are realized.
Further, the microphone and the speaker are independently disposed in the car.
Furthermore, the car top wireless bridge, the machine room wireless bridge, the roof wireless bridge and the property roof wireless bridge are all high-speed bridges.
Further, the acceleration sensor is a vector acceleration sensor.
Compared with the prior art, the beneficial effects of the utility model reside in that:
the utility model realizes the identification of whether people are in the elevator, the elevator is trapped, abnormal passenger behavior, stranger identification and elevator fault identification based on the intelligent camera and the deep neural network image identification algorithm; the data acquisition of the up-down movement, the running speed, the floor where the elevator is located and the door opening and closing is realized through a plurality of sensors arranged in the elevator; the elevator emergency fault visible intercom system realized by hardware expansion expands the elevator emergency fault rescue information in a large range (from one fixed telephone to a mobile phone, a computer and other clients), and improves the processing efficiency of elevator rescue; through the calculation unit installed in the elevator, the calculation, sensing and triggering of various complex scenes are realized, the analysis and the arrangement of sensor data and the result output are realized, and the accurate elevator running speed output and the door opening and closing result output are realized. The cloud server and the client realize the transmission, storage and visual analysis of data and data management; the elevator fault emergency visible intercom system is realized by hardware expansion; the functions of up-down movement, running speed, floor and door opening and closing states of the elevator are realized through a hardware scheme; the intelligent elevator fault monitoring and early warning function, the passenger abnormal behavior monitoring and early warning function, the elevator trapping automatic warning function and the stranger recognizing and early warning function are realized through hardware and software schemes. The safety of the whole is high, the complexity of installation and the cost of products are reduced to the greatest extent, the installation universality is high, the compatibility is high, the operation is convenient and humanized, the manual operation intensity and difficulty are reduced, and the security efficiency is greatly improved; the system has high stability and reliability, can be remotely and directly updated, upgraded and maintained, has the functions of disaster tolerance and self-learning, and has high intelligent degree, time saving, labor saving and worry saving.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only for the purpose of more clearly illustrating the embodiments of the present invention or the technical solutions in the prior art, and for those skilled in the art, other drawings can be obtained according to these drawings without any creative efforts.
Fig. 1 is a schematic diagram of the overall hardware layout structure of the present invention.
Detailed Description
In order to make the technical solution of the present invention better understood by those skilled in the art, the present invention will be further described with reference to the following specific embodiments, which are provided as illustration only and are not intended to limit the present invention.
As shown in fig. 1, the elevator operation safety detection and early warning identification system based on intelligent hardware comprises an intelligent camera 4, a microphone 9, a loudspeaker 10 and an emergency call button 16 which are arranged in an elevator car, wherein a car roof wireless bridge 3, a calculation unit 5, an acceleration sensor 6 and an air pressure sensor 7 are arranged on a car roof above the elevator car, and a photoelectric induction sensor 8 is arranged at the position of a car door 17 of the elevator; the computing unit 5 is internally provided with an information sending/receiving module, a judging module and an information processing module.
The car roof wireless network bridge 3 is wirelessly connected with the machine room wireless network bridge 2, the machine room wireless network bridge 2 is wirelessly connected with the roof wireless network bridge 1, the roof wireless network bridge 1 is wirelessly connected with the property roof wireless network bridge 11 arranged on the roof of the property office 12, the property office is provided with a property computer end 15, a cloud server 13 is arranged at the cloud end, the cloud server 13 is connected with the property roof wireless network bridge 11 through property optical fibers or broadband, and the property computer end 15 and the mobile phone terminal 14 are connected with the cloud server 13 in a wired or wireless mode;
the car top wireless bridge 3 is respectively connected with an intelligent camera 4 and a computing unit 5, the acceleration sensor 6, the air pressure sensor 7, the photoelectric sensing sensor 8, the microphone 9, the loudspeaker 10 and the emergency call button 16 are all connected with the computing unit 5 and transmit signals to the computing unit 5, and the computing unit 5 is connected with a cloud server 13 through the car top wireless bridge 3, the machine room wireless bridge 2, the roof wireless bridge 1 and the property and building top wireless bridge 11;
the calculation unit 5 is internally provided with a deep neural network image recognition algorithm and a face recognition algorithm, the deep neural network image recognition algorithm has self-learning and identification marking functions, and the face characteristic values of each person are extracted, stored and marked by comparing the face characteristic values through analyzing video signals.
The system is also provided with a UPS emergency power supply, and the system can still work normally under the condition of power failure.
Through the utility model discloses can realize: the intelligent elevator system comprises an elevator car picture real-time monitoring function, a voice stop reporting function, an automatic sleepers alarm function, a stranger identification and early warning function, an elevator operation data statistic analysis function (such as elevator use frequency analysis, stop frequency analysis of each floor, elevator taking number analysis and the like), an elevator fault intelligent monitoring and early warning function (such as faults of elevator shake, abnormal operation speed, non-flat floor stop and the like), an intelligent passenger abnormal behavior monitoring and early warning function (such as behaviors of children taking the elevator alone, old people falling down, jumping and the like), a cross-platform emergency fault visible talkback function, an alarm receiving and positioning function of maintenance personnel, an automatic maintenance supervision function, an intelligent property management function and the like.
Throughout the market, no more than 3 products are first marketed and released for commercial use. Through the analysis to its product and system, with the utility model discloses a contrast, can see out the technical defect that its product key scheme exists, perhaps the disappearance of whole core function, and the utility model discloses an advantage and unique place.
The utility model discloses the function that can realize and the realization principle do:
1. cross-platform visual intercom function for emergency fault
Background: in order to facilitate the elevator passengers and maintenance workers to get in contact with the outside in emergency, emergency rescue is carried out. At present, all elevators in China have to be provided with five-way talkback systems according to the requirements of the national quality inspection bureau. The system is characterized in that a set of telephones is respectively deployed in a machine room, a car roof, a car, a pit and a property office, and because the brands of manufacturers are different, the manufacturers are connected in a wired/wireless mode, and in the five parties, as long as any party initiates a call, all other parties ring, so that the conversation is realized. However, this solution has obvious drawbacks: one is an application scene, particularly, a property side needs to keep a telephone in a property duty room all the time, and if an emergency fault telephone exists, the condition of missing connection due to unattended operation may exist; the second is limited by distance and high product cost, if the wireless talkback is adopted, the effective transmission distance is about 3 kilometers, a plurality of sets of the wireless talkback are required to be arranged for a large cell, and if the wireless talkback is adopted, the wireless talkback means that long-distance wiring is required, so that the construction is difficult. In addition, most importantly, the selling price of a set of five-party talkback products on the market is more than 2000 yuan, which is expensive.
The existing scheme is as follows: synthesize most elevator thing networking system products on the market, to this pain point and scene, mainly have two kinds of schemes: the first one is to continue to use the five-way intercom system of the elevator without any change; the second is that the audio acquisition/playing and transmission are realized by arranging a loudspeaker and a microphone in a monitoring camera of the car. However, this solution can only be used for one-way call initiation, i.e. only the mobile phone terminal 14 can call the car, and the call cannot be initiated from the car. If the car is trapped, the person of the car cannot notify the outside person through it. Therefore, the proposal is more chicken rib and can not solve the practical problem.
The utility model discloses scheme and principle: the utility model discloses at first independently disposed high-fidelity loudspeaker 10 and microphone 9 in the car, guaranteed speech quality. Then, a triggering key emergency call button 16 and a high-definition intelligent camera 4 are arranged in the car and used for initiating a call and acquiring a car video picture. The realization principle is as follows:
when an emergency fault occurs:
1) the car passenger clicks the emergency call button 16 to trigger a call signal;
2) after the call signal is triggered, the computing unit 5 receives the call signal and triggers a call request;
3) the call request signal sends a call request to the cloud server 13 through the car roof wireless network bridge 3 → the machine room wireless network bridge 2 → the roof wireless network bridge 1 → the property roof wireless network bridge 11;
4) the cloud server 13 receives the call request signal, firstly searches the property elevator manager and the related maintainer corresponding to the elevator initiating the call request, and initiates a voice call to the corresponding mobile phone/computer after finding the property elevator manager and the related maintainer;
5) after receiving the call, the property elevator manager and the related maintenance personnel can communicate with the help seeking passenger in the elevator car through the property computer terminal 15 or the mobile phone terminal 14, and the 9 microphone and the 10 loudspeaker connected to the computing unit 5 realize the sound playing and the passenger sound collection;
6) during voice communication, the property computer terminal 15 or the mobile phone terminal 14 can see real-time pictures in the car through the 4 cameras, know the field environment and perform rescue scheme preparation as soon as possible;
7) people with related authority, such as property managers/maintenance staff, can also initiate a call to the car through the property computer terminal 15 or the mobile phone terminal 14, so that real bidirectional calling is realized;
the whole system also adopts a high-speed network bridge and a UPS emergency power supply, so that the system can still be continuously used under the condition of power failure.
The advantages are that:
1) the range of emergency calls is expanded, and the conventional emergency calls can only be directionally initiated to property. The utility model expands the calling range to all people with elevator management authority, such as property managers, maintenance personnel, and the like, and increases the reliability of rescue;
2) the limit of the original five-party intercom system is broken through, the original five-party intercom system can only call one fixed telephone of the property monitoring room, and people 7 x 24h are needed to watch the system. The utility model breaks the limitation, all people do not need to stay in the duty room, and can answer the emergency call from the elevator at any time and any place only by a mobile phone;
3) the prior five-party intercom system can only hear but cannot see. It is inconvenient for rescue workers to judge the current environment and deploy a rescue scheme in advance. Through the utility model discloses, not only can talk with the people in the car, can also know the on-the-spot condition through the visual interface of cell-phone in real time, be convenient for arrange in advance and know the rescue condition.
Alternative scheme: the utility model discloses a more mature urgent five sides intercom system based on the fixed telephone has been gone out in the market, and other schemes can be replaced temporarily.
2. Elevator up-down and floor detection function
Background: in the running process of the elevator, an up-down running state and a floor state are provided, and the elevator inspection is very important, namely the elevator runs upwards or downwards at present, and the elevator runs on several floors at present. In the prior inspection and monitoring process, a property elevator manager or other persons with authority cannot know the current running state of the elevator unless the manager or other persons are checked on site. If the running state of the elevator can be remotely seen through the mobile phone terminal 14 or the computer terminal 15, remote inspection is realized, and the elevator monitoring system is undoubtedly the largest bright spot for the elevator industry.
The existing scheme is as follows: through the analysis to the product on the market, the product of finding only one family has realized the monitoring of elevator upper and lower line and floor, and the scheme is the magnetic induction switch, and concrete implementation scheme is: because in the elevator design, each layer of lift car can both stop, so at the well position that each layer of lift car stopped, paste a magnetic induction piece, can burn in advance the floor number in the magnetic induction piece. Also on the top of the car, there is a magnetic induction sensor for sensing with the magnetic induction sheet. When the elevator car stops or passes by, the floor where the elevator is located at present is sensed, and the elevator can be known to go up or down through the comparison of two adjacent floors. However, this solution presents a great challenge to installation and deployment, increasing the installation cost and installation risk of the equipment, for example, if the building has 33 floors, magnetic induction sheets need to be attached to the well-way position of each floor of the 33 floors. Then if the building is a building with 100 floors, the workload will be high, the efficiency will be low, and the cost will be high.
The utility model discloses scheme and principle: the utility model discloses a complexity of at utmost reduction installation and the cost of reduction product have used the mode that vector acceleration sensor 6 and baroceptor 7 combine to realize elevator up-and-down and the floor monitoring of locating. The realization principle is as follows:
1) identification principle of up and down elevator:
when the object moves towards different directions, the acceleration sensor 6 can feed back different signals and can also feed back the acceleration of the motion of the object;
when the elevator car moves up and down, the acceleration sensor 6 has a signal feedback;
the signals of the acceleration sensor 6 are fed back to the calculation unit 5 through a connecting line, and the calculation unit 5 analyzes the signals, so that the upward/downward/static identification and monitoring of the elevator are realized.
2) Identification principle of the floor where the elevator is located:
along with the physical principle that the air pressure is reduced when the altitude is increased, if the current air pressure is known, the current altitude can be known;
when the elevator moves up and down, the current air pressure value can be monitored in real time through the air pressure sensor 7;
the air pressure sensor 7 feeds a monitored air pressure value back to the computing unit 5 through a connecting line, the computing unit 5 converts the air pressure value into an altitude, and the current floor of the elevator can be known by matching according to the floor height of each floor of the elevator (for example, the floor height 1 is 0-3 m, and the floor height 2 is 3-6 m);
after the floor is known, the running condition of the elevator can be known remotely through the data transmission of the computing unit 5 and the property computer terminal 15 or the mobile phone terminal 14, so that remote inspection is realized.
The advantages are that:
1) the cost and the complicated flow of the installation of the elevator Internet of things hardware equipment are reduced, and the functions can be realized only by placing the corresponding sensor at the top of the elevator car and configuring the parameters;
2) the floor identification and uplink and downlink monitoring functions are more stable.
Alternative scheme: the distance between the elevator car and a fixed position (such as a machine room or a pit) is measured in a laser ranging mode, and the functions of identifying the floor where the elevator is located and identifying the up-down movement can be realized.
3. Automatic alarming function for people trapped
Background: one of the most serious faults in elevators today is elevator jamming. Because the elevator runs in a fault, the lift car does not stop at a flat floor (for example, the lift car stops between the 2 th floor and the 3 rd floor) or the door is not normally opened after the lift car stops, passengers in the lift car are trapped, and if the passengers are not rescued in time, very serious consequences can be caused. The elevator is stuck by people for a plurality of reasons, such as mechanical failure, electrical failure or artificial reason, and the like, and cannot be fundamentally stopped, so that the stuck information is diffused outwards in the first time of the stuck, and the elevator is rescued in time, but is the most critical. If the trapped person is alert aware, an alert may be triggered, but if the trapped person is not alert (e.g., a child or an elderly person is trapped), the consequences are very serious. Therefore, if a set of system can automatically monitor that the elevator is trapped and can automatically trigger alarm at the first time, the system brings extremely high rescue efficiency and reduces risks to the greatest extent.
The existing scheme is as follows: at present, no existing scheme in the market can realize the automatic alarming function of elevator trapping.
The utility model discloses scheme and principle: the utility model discloses a neural network image recognition algorithm and to the discernment of elevator car door 17 on-off state based on intelligent camera 4 to and the discernment of floor display panel in the elevator car, the combination has realized the discernment of the stranded trouble of elevator and automatic alarm's triggering. The realization principle is as follows:
1) the intelligent camera 4 installed in the elevator car transmits real-time video signals to the computing unit 5 through the car top wireless network bridge 3, the computing unit 5 is internally provided with a developed deep neural network image recognition algorithm, whether people and several people exist in the car can be known through analysis of real-time pictures, and whether the car door 17 faces a wall after being opened (if non-flat-layer stopping occurs, namely if the elevator stops between a 3-th building and a 4-th building, the wall is seen after the door is opened).
2) Through installing the photoelectric sensing sensor 8 in the car door 17 position, can realize light and electric signal conversion, when car door 17 closed, can block photoelectric sensing sensor 8's light signal, can produce a signal, when car door 17 opened, can not block photoelectric sensing sensor 8's light signal, can produce the signal of another type. The monitoring of whether the elevator is opened or closed can be realized by sending the signal to the computing unit 5 through a connecting line.
3) Based on 1) and 2) implementation: if the system identifies that the elevator is occupied, but after 30 seconds, the car door 17 is not opened, the calculation unit 5 considers that the elevator is in a trapped state and fails; if the system recognizes that someone is in the elevator, the door is also open, but the elevator door position is a wall (non-flat landing), the calculation unit 5 considers that the elevator has a trapped fault;
4) after the computing unit 5 successfully identifies, the people trapping fault information is sent to the cloud server 13 through the car roof wireless bridge 3 → the machine room wireless bridge 2 → the roof wireless bridge 1 → the property roof wireless bridge 11;
5) after receiving the trouble information fault push of the trapped person, the cloud server 13 firstly searches the property elevator manager and the related maintenance personnel corresponding to the trouble elevator, and sends the fault push to the property computer terminal 15 or the mobile phone terminal 14 corresponding to the trouble elevator after finding the property elevator manager and the related maintenance personnel;
6) after receiving the early warning of the elevator fault trapping through the property computer terminal 15 or the mobile phone terminal 14, the system is turned on, so that the corresponding fault information can be seen, the trapping situation in the elevator car can be seen through the intelligent camera 4, the situation can be known as soon as possible, and a rescue scheme is deployed;
the advantages are that:
1) the identification of people in the elevator car is realized through a complex image neural network algorithm;
2) the passive mode is changed into the active mode, and the efficiency of safety precaution and fault rescue of the elevator is improved in one go from finding, identifying and early warning notification, so that the risk of serious consequences caused by long-time sleepiness is greatly reduced.
Alternative scheme: there are no other viable alternatives available on the market to achieve this functionality.
4. Stranger automatic identification and early warning function
Background: at present, security protection of all buildings including residential areas, office buildings, business centers and the like is an absolute point of attention of the property of the buildings. The existing security schemes in the market are all the combination of multiple schemes, so that the security measures are gradually enhanced, for example: the entrance guard of district + entrance guard of cell door + surveillance camera head etc.. Therefore, intelligent security and protection cannot be achieved, strangers can enter a cell and enter a cell gate along with owners, and the monitoring probe only plays a role in tracking afterwards. Real intelligent security protection should be early warning in advance, save worry laborsaving labour saving and time saving cost, should be close to and a better control scene of fitting. The elevator plays an increasingly prominent role in modern society, and can not be left at work, and if owners and strangers can be automatically identified in the elevator and alarms are immediately sent to corresponding persons of property after the strangers are identified, the security efficiency is undoubtedly and greatly improved.
The existing scheme is as follows: products which can automatically distinguish owners from strangers and actively trigger early warning are not available in the elevator car in the market.
The utility model discloses scheme and principle: the utility model discloses a nuclear core point is based on install intelligent camera 4 in the car and the deep neural network learning algorithm who develops all by oneself realize. The intelligent camera 4 samples the human faces and human figures of passengers in the car, and the deep neural network image recognition algorithm realizes the extraction of sampling characteristic values and the calculation of the occurrence frequency of the characteristic values. The specific realization principle is as follows:
after the equipment is installed, the intelligent camera 4 starts to work, and based on a built-in face recognition algorithm, once a person in the car is recognized, a real-time video signal is transmitted to the computing unit 5 through a line;
the calculation unit 5 is internally provided with a self-developed deep neural network image recognition algorithm and a human face recognition algorithm, and human face characteristic values (human face recognition is realized by comparing the human face characteristic values, the face shapes, the positions of five sense organs and the like of each person are different, and the characteristic values are naturally different) can be extracted and stored by analyzing video signals;
and then the equipment carries out self-learning, wherein the self-learning process is to collect a large number of face characteristic values in the lift car. If the system finds a user A and takes an elevator 5 or 6 times a day, the system marks the user as the owner identity, and the safety is high. After the study in a period of time, the system can mark more than 95% of the resident people of the building;
the system then begins normal monitoring operation. If the computing unit 5 identifies that a person never appears in the elevator, the stranger is considered to appear, and a stranger early warning signal is triggered;
the stranger early warning signal sends a call request to the cloud server 13 through the car top wireless bridge 3 → the machine room wireless bridge 2 → the roof wireless bridge 1 → the property roof wireless bridge 11;
after receiving and identifying the information push of the strangers, the cloud server 13 searches for the property elevator manager and the related maintenance personnel corresponding to the elevator triggering the signal, and sends an early warning push of 'identification to strangers' to the property computer terminal 15 or the mobile phone terminal 14 corresponding to the property elevator manager and the related maintenance personnel after finding;
after receiving the early warning through the property computer terminal 15 or the mobile phone terminal 14, the system is turned on, and the pictures or videos of the stranger shot through the intelligent camera 4 can be seen. The person may also be re-tagged, such as by being a courier or other person. The next time the system re-identifies the person, no further alarm will be issued.
The advantages are that:
1) the stranger early warning system based on the self-learning of the deep neural network image recognition algorithm is combined with an elevator car scene, aiming at the recognition and monitoring of strangers, the passive state is changed into the active state, the monitoring scene is enlarged, and the security and protection efficiency is greatly improved;
2) the early warning is directly pushed to the mobile phones of the property related personnel anytime and anywhere, and the property is not limited by time and space, so that the property can be found and processed at the first time, and the risk is reduced;
3) one camera not only realizes the real-time monitoring of the elevator car picture, but also realizes the identification and alarm of strangers, and additional hardware cost is not needed, thereby saving time, labor and cost.
Alternative scheme: at present, the market does not have products which can realize the automatic identification and early warning of strangers of the car within the acceptable cost range of the property industry in a commercialized way. There is currently no good alternative.
5. Intelligent monitoring and early warning function for elevator faults
Background: the maintenance and safety situation of the elevator industry is more and more tense, the quantity of the elevators in the whole country reaches nearly 700 thousands of elevators by 2018 according to statistics, and the elevators are increased at the speed of 50-70 thousands of elevators per year. The more advanced the installation time, the higher the failure frequency of the elevator (old elevator) along with the time, and the fewer and fewer people engaged in the maintenance industry, so that the maintenance and safety situation of the elevator is more tense. If we can realize the elevator networking comprehensively, use sensor and intelligent neural network real time monitoring elevator's machinery and electrical structure to establish unified control and early warning center, the elevator just can carry out the early warning in advance when not breaking down yet, even when the elevator breaks down, in time reports to the police, the unity of the elevator resource of being convenient for and the fastest allocation, will lose and fall to minimumly.
The existing scheme is as follows: at present, no commercial-grade elevator Internet of things early warning system exists in China, and comprehensive, systematic and stable monitoring and early warning can be carried out on frequent faults of an elevator.
The utility model discloses scheme and principle: the utility model discloses detection to elevator trouble mainly comprises through photoelectric sensor 8, acceleration sensor 6, baroceptor 7, intelligent camera 4, computational element 5 and degree of depth neural network image recognition algorithm. The realization principle is as follows:
whether the elevator is monitored by a person can be monitored by the intelligent camera 4 and the computing unit 5 (the realization principle is described in the automatic alarming function 1) of the trapped person in the front); the opening/closing of the elevator car door 17 can be monitored by the car door 17 photoelectric sensor 8 (the principle of the detection is described in the above-mentioned automatic alarming function 2). In combination with the above, if it is found that the elevator is empty, but the elevator car door 17 is frequently opened and closed, the calculation unit 5 will consider that the elevator has a "door open and close abnormality" fault;
whether the elevator stops on the flat floor or not (the realization principle is described in the automatic alarming function 3 for people trapping in the front) can be realized through the intelligent camera 4 and the calculating unit 5, and if the elevator does not stop on the flat floor, the calculating unit 5 can think that the elevator has the fault of 'non-flat floor stop';
whether the elevator car shakes can be monitored through the acceleration sensor 6 and the calculation unit 5, and if the fact that the elevator car instantaneously has displacement data of up, down, left and right is calculated, the calculation unit 5 can think that the elevator has a car shaking fault;
the real-time running speed of the elevator car in the moving process can be realized through the acceleration sensor 6 and the computing unit 5, and if the running speed curve of the elevator car is not smooth or meets the requirements of an elevator manufacturer, the computing unit 5 can consider that the elevator has a fault of abnormal running speed;
various early warning signals identified by the computing unit 5 are sent to the 13 cloud server through the car roof wireless network bridge 3 → the machine room wireless network bridge 2 → the roof wireless network bridge 1 → the property roof wireless network bridge 11;
after receiving the elevator fault early warning information push, the cloud server 13 searches for a property elevator manager and a related maintenance worker corresponding to the elevator triggering the signal, and sends a corresponding elevator fault early warning push to a property computer terminal 15 or a mobile phone terminal 14 corresponding to the property elevator manager and the related maintenance worker after finding the property elevator manager and the related maintenance worker;
after receiving the early warning through the property computer terminal 15 or the mobile phone terminal 14, the system is turned on, and the corresponding details of the elevator fault and the picture or video of the corresponding fault shot through the intelligent camera 4 can be seen. The elevator fault diagnosis system is convenient for maintenance personnel to diagnose the elevator fault in advance, make a maintenance scheme and take measures.
The advantages are that:
1) the installation and construction are simple, and the original mechanical and electrical structures of the elevator do not need to be damaged;
2) the equipment cost is reduced linearly, and due to the application of the deep neural network algorithm based on image and video recognition, the investment of hardware equipment is greatly reduced, so that the equipment cost is reduced, and the popularization and the promotion of products on the market are facilitated;
3) installation universality and high compatibility. Technical barriers among all large elevator brands are high, one set of system can be compatible with elevators of all brands and models on the market, and the challenge is huge. The utility model adopts the modular assembly design, is not coupled with the mechanical and electrical structure of the elevator, and can be compatible with 99 percent of elevators on the market;
4) the system has high stability and reliability, the main core identification and algorithm module is deployed at the cloud end, can be remotely and directly updated, upgraded and maintained, has disaster recovery and self-learning functions, and several main sensors installed on the elevator all use military levels;
alternative scheme: in the elevator industry, the biggest problems are that the elevator brands are too many, the technical barriers among the brands are high, and various mechanical and electrical parameters and protocols of the elevator are not unified at all. One of the most intuitive manifestations is the engineer who will repair the brand a elevator, not the brand B elevator. Therefore, in the market, products for realizing fault monitoring and early warning of the elevator are available.
6. Passenger abnormal behavior monitoring and early warning function
Background: the elevator malfunctions if there is no passenger, which affects at most the normal use of people, but the situation is more serious if there is a malfunction if there is a passenger. In common faults and damages of the elevator, a large part of the common faults and damages are caused by passengers in the elevator car, and the elevator is not used and taken according to a certain safety rule in a normalized mode, so that the faults are caused. Therefore, if the analysis and capturing of passenger behavior is implemented within the elevator car, passengers can be stopped before a hazard occurs, avoiding more serious consequences.
The existing scheme is as follows: through the analysis investigation on the market, no product or scheme can realize the monitoring and early warning of elevator passenger's unusual action at present.
The utility model discloses scheme and principle: the utility model discloses to passenger's unusual action discernment and monitoring in the elevator car environment, be through installing at the inside intelligent camera 4 of car and based on image and video degree of depth neural network recognition algorithm constitute, can realize: bouncing in the car, smoking in the car, falling down of passengers, riding of children alone, moving/shielding of the camera, blocking of the car door 17 to close, forced opening of the car door, dependence on abnormal behaviors of passengers and the like. The specific realization principle is as follows:
now, the passenger is just a faint example, and the realization principle of the abnormal behavior recognition of other passengers is consistent with the above principle:
in the development stage, a large amount of pictures and video resources related to the falling of passengers in the elevator car are collected through the intelligent camera 4, the pictures and the video resources are self-learned to a computer, the purpose is to collect characteristic values of the behavior of falling of the passengers (for example, the passengers lie on the elevator car when falling, and some parts of bodies contact the bottom of the elevator car), then a data model base is established, a deep learning recognition algorithm of falling of the elevator is analyzed, and the algorithm is deployed into the computing unit 5;
after the equipment is installed, the video stream collected by the intelligent camera 4 can reach the computing unit 5 through a connecting line;
the calculation unit 5 monitors and analyzes the video streams in real time, and if a certain behavior characteristic in the video streams is found to be matched with the behavior of 'passengers falling down' in the algorithm model library, the calculation unit 5 considers that the passengers fall down and immediately triggers an early warning signal;
the early warning signal is sent to the 13 cloud server through the car roof wireless network bridge 3 → the machine room wireless network bridge 2 → the roof wireless network bridge 1 → the property roof wireless network bridge 11;
after receiving the elevator early warning information push, the cloud server 13 searches for a property elevator manager and a related maintenance worker corresponding to the elevator triggering the signal, and sends a corresponding elevator early warning push to a property computer terminal 15 or a mobile phone terminal 14 corresponding to the property elevator manager and the related maintenance worker after finding the property elevator manager and the related maintenance worker;
after receiving the early warning through the property computer terminal 15 or the mobile phone terminal 14, the corresponding property elevator manager and the related maintenance personnel open the system, and can see the corresponding elevator early warning details and the pictures or videos shot by the intelligent camera 4 and showing the passengers falling down. The elevator fault diagnosis system is convenient for maintenance personnel to diagnose the elevator fault in advance, make a maintenance scheme and take measures.
The abnormal behavior analysis recognition rate of the elevator can reach more than 80% at present, and the recognition rate can continuously rise through continuous optimization of an algorithm in the later period. Just like the automatic driving technology of automobiles, with continuous updating and algorithm upgrading, the monitoring of the peripheral environment can be more and more accurate, and the corresponding processing can be more and more accurate.
The advantages are that:
1) the analysis and identification of passenger behaviors in the elevator car are realized, the harm to the safe operation of the elevator caused by the behaviors is prevented and prevented in advance, and the occurrence of safety accidents is avoided;
2) the hardware cost of the whole product is low, the installation and the deployment are convenient, and the marketization popularization is convenient;
3) the product function can be continuously upgraded and perfected through the cloud, and the cost of later maintenance upgrading is reduced.
Alternative scheme: currently, there is no better alternative for monitoring and identifying abnormal behavior of passengers in a car in an elevator.
The present invention is not described in detail in the prior art.
The above description is only a preferred embodiment of the present invention, and should not be taken as limiting the invention, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. An elevator operation safety detection and early warning identification system based on intelligent hardware is characterized by comprising an intelligent camera (4), a microphone (9), a loudspeaker (10) and an emergency call button (16) which are arranged in an elevator car, wherein the elevator car top is provided with a car top wireless bridge (3), a calculation unit (5), an acceleration sensor (6) and an air pressure sensor (7), and a photoelectric induction sensor (8) is arranged at the position of an elevator car door (17); the computing unit (5) is internally provided with an information sending/receiving module, a judging module and an information processing module;
the car roof wireless network bridge (3) is sequentially in wireless connection with a machine room wireless network bridge (2), a roof wireless network bridge (1) and a property roof wireless network bridge (11), the property roof wireless network bridge (11) is arranged on the roof of a property office (12), the property office (12) is provided with a property computer end (15), a cloud server (13) is deployed at the cloud end, the cloud server (13) is connected with the property roof wireless network bridge (11) through optical fibers or broadband, and the property computer end (15) and a mobile phone terminal (14) are connected with the cloud server (13) in a wireless or wired mode;
the intelligent camera (4) is connected with a computing unit (5) through a car roof wireless network bridge (3), the acceleration sensor (6), the air pressure sensor (7), the photoelectric sensing sensor (8), the microphone (9), the loudspeaker (10) and the emergency call button (16) are all connected with the computing unit (5) and transmit signals to the computing unit (5), and the computing unit (5) is connected with a cloud server (13) through the car roof wireless network bridge (3), the machine room wireless network bridge (2), the roof wireless network bridge (1) and the property building roof wireless network bridge (11);
the system is also provided with a UPS emergency power supply, and the system can still work normally under the condition of power failure.
2. An elevator operation safety detection and early warning identification system based on intelligent hardware as claimed in claim 1, characterized in that the intelligent camera (4) is used for obtaining video pictures in the elevator car, identifying the opening and closing states of the elevator car door (17), identifying the floor display panel in the elevator car, and transmitting the information to the computing unit (5);
the acceleration sensor (6) is used for identifying and monitoring the up-down and static states of the elevator and feeding back signals to the computing unit (5);
the air pressure sensor (7) is used for identifying the floor where the elevator is located and feeding back a signal to the computing unit (5);
the photoelectric induction sensor (8) is combined with the computing unit (5) to realize the identification and monitoring of whether the elevator car door (17) is opened or closed.
3. The elevator operation safety detection and early warning identification system based on intelligent hardware as claimed in claim 1, characterized in that the identification of the abnormal door opening and closing fault, the non-flat landing fault and the early warning of the elevator are realized by combining the intelligent camera (4) and the computing unit (5);
the calculation unit (5) and the acceleration sensor (6) are combined, so that the ' car shaking ' fault ' and ' abnormal running speed ' fault of the elevator can be recognized, and early warning is realized.
4. An intelligent hardware-based elevator operation safety detection and early warning identification system according to claim 1, wherein the microphone (9) and the speaker (10) are independently deployed in the car.
5. The elevator operation safety detection and early warning identification system based on intelligent hardware as claimed in claim 1, wherein the car top wireless bridge (3), the machine room wireless bridge (2), the roof wireless bridge (1) and the property roof wireless bridge (11) are all high-speed bridges.
6. An intelligent hardware-based elevator operation safety detection and early warning identification system according to any one of claims 1 to 5, wherein the acceleration sensor (6) is a vector acceleration sensor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920875728.7U CN210558707U (en) | 2019-06-11 | 2019-06-11 | Elevator operation safety detection and early warning identification system based on intelligent hardware |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920875728.7U CN210558707U (en) | 2019-06-11 | 2019-06-11 | Elevator operation safety detection and early warning identification system based on intelligent hardware |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN210558707U true CN210558707U (en) | 2020-05-19 |
Family
ID=70625687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201920875728.7U Expired - Fee Related CN210558707U (en) | 2019-06-11 | 2019-06-11 | Elevator operation safety detection and early warning identification system based on intelligent hardware |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN210558707U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112408140A (en) * | 2020-12-08 | 2021-02-26 | 湖北惠天下建工科技股份有限公司 | System and method applied to elevator control reminding and fault diagnosis |
| CN114655804A (en) * | 2022-02-25 | 2022-06-24 | 深圳市励科机电科技工程有限公司 | Elevator monitoring method, device and system |
-
2019
- 2019-06-11 CN CN201920875728.7U patent/CN210558707U/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112408140A (en) * | 2020-12-08 | 2021-02-26 | 湖北惠天下建工科技股份有限公司 | System and method applied to elevator control reminding and fault diagnosis |
| CN114655804A (en) * | 2022-02-25 | 2022-06-24 | 深圳市励科机电科技工程有限公司 | Elevator monitoring method, device and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110155842A (en) | A kind of lift running safety detection and early warning identifying system based on Intelligent hardware | |
| CN104627769A (en) | Safe operation internet-of-things monitoring system for elevator | |
| CN206375523U (en) | A kind of elevator faults monitoring and warning system | |
| CN105314485B (en) | Elevator faults monitoring system and its method based on multimedia delivery system | |
| CN105540377A (en) | Internet of things remote elevator monitoring system with human face matching function | |
| CN205230242U (en) | Intelligent residential district security protection system | |
| CN103231956A (en) | Multifunctional intelligent remote alarm media machine for elevator | |
| CN112520528B (en) | Automatic monitoring elevator fault detection system and detection method thereof | |
| CN103640944A (en) | Intelligent and remote automatic alarming system for elevator faults | |
| CN102344069A (en) | Elevator emergency management technology supporting system | |
| CN210558707U (en) | Elevator operation safety detection and early warning identification system based on intelligent hardware | |
| CN206502479U (en) | A kind of security system applied on box elevator | |
| CN206767346U (en) | A kind of high definition visual type elevator internet-of-things terminal | |
| CN208732382U (en) | Elevator signal acquisition device and system | |
| CN103896114A (en) | Intelligent elevator state remote monitoring system | |
| CN106976777A (en) | One kind is used for elevator long distance fault monitoring system and its detection method | |
| CN105366477A (en) | Safe operation monitoring system for intelligent elevator | |
| CN114638455A (en) | Emergency disposal plan generation pushing method and comprehensive command control management system | |
| CN202346532U (en) | Elevator emergency management technology supporting system | |
| CN110155845A (en) | The elevator remote control system that failure and disaster monitor simultaneously | |
| CN103130056A (en) | Cross-domain elevator monitoring method | |
| CN207632306U (en) | A kind of escalator Monitoring System | |
| KR20170133595A (en) | A method for determining the elevator car layer by the sensor value | |
| CN207390720U (en) | Elevator wireless intercommunication warning device | |
| CN109809281B (en) | Manual unlocking monitoring method for elevator landing door |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200519 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |