CN105731209A - Intelligent prediction, diagnosis and maintenance method for elevator faults on basis of Internet of Things - Google Patents

Abstract

The invention relates to the field of elevator control and the field of the Internet of Things. It provides an intelligent prediction, diagnosis and maintenance system for elevator faults, realizes effectively shortening the elevator downtime caused by elevator faults, improves the reliability of elevator operation and elevator maintenance efficiency, and protects personal safety. and property safety. The technical solution adopted by the present invention is, based on the elevator failure intelligent prediction and diagnosis and maintenance method of the Internet of Things, the elevator data is collected in real time through the data acquisition module, and then the collected data is transmitted to the server through WIFI or 4G wireless communication technology , and apply big data technology on the server side to analyze the data, so as to predict the possible failure of the elevator; if the elevator fails, use the fault diagnosis system on the server side to analyze the failure, and then give the location and cause of the failure , and then provide a solution to the fault through the intelligent maintenance system on the server side. The present invention is mainly applied to elevator control occasions.

Description

Intelligent prediction and diagnosis and maintenance method of elevator fault based on Internet of Things

technical field

The present invention relates to the field of elevator control, the field of the Internet of Things, the field of embedded technology, the field of Wi-Fi wireless data transmission, the field of mobile network communication, and the field of data acquisition technology, and especially relates to an intelligent prediction, diagnosis and maintenance of elevator faults based on the Internet of Things system.

Background technique

As of the end of 2015, the total number of elevators in the country exceeded 4 million, and the number of elevators in my country, the annual output, and the annual growth rate ranked first in the world. With the continuous expansion of the domestic elevator market, people have put forward higher requirements for the service quality of elevators, among which safety and reliability have become more and more important issues that people attach great importance to. In daily life, problems such as elevator jamming, failure to stop the elevator, and abnormal operation frequently occur. These failures have brought great troubles to people's lives and property safety. In recent years, there have been about 50 elevator accidents in our country every year on average. At 13:26 on September 13, 2012, a manned elevator fell from the 30th floor of a residence under construction in Wuhan City, Hubei Province. According to the Propaganda Department of the Wuhan Municipal Party Committee, a total of 19 people were killed; April 10, 2013 In a clothing factory on Fangzhou Road, Liuhe North Gate, Nanjing, a porter wanted to take the elevator down from the 4th floor to the 1st floor. When the elevator door opened, the elevator did not reach the 4th floor. The porter stepped on the ground and was seriously injured. , died; on March 14, 2014, an elevator maintenance worker in Ganzhou, Jiangxi Province was working, and his foot was stuck in the dismantled hole at the bottom of the elevator. After nearly an hour of cracking by the firefighters, the injured were finally rescued; on July 26, 2015, an escalator accident occurred in a shopping mall in Jingzhou, Hubei, resulting in the tragic death of a young mother with a child. Afterwards, a one-year-old child in Sun Plaza in Wuzhou, Guangxi accidentally fell down on an escalator, causing his arm to get stuck; the child was rescued and sent to the hospital for examination and found that his left hand was fractured. The frequent occurrence of various elevator accidents has sounded the alarm for people. However, the inspection by the quality supervision department found that 20% of elevators in my country have potential safety hazards. People living in high-rise buildings often discuss issues such as elevator shutdown, failure to stop, abnormal operation, and failure to repair in time. Failure to diagnose and maintain elevator faults in time poses a great threat to passenger safety. Elevator failure has become a hidden danger to urban security, and it is an urgent problem to be solved in urban security. At the same time, it is another hidden danger of elevator safety in my country to fail to decide the maintenance method in time after grasping the cause of the failure. Therefore, it is imminent to improve the level of elevator security management, fault diagnosis and maintenance efficiency and effect in our country. The elevator remote monitoring fault diagnosis and intelligent maintenance system came into being under such circumstances.

A relatively advanced function of the remote elevator monitoring system is "failure prediction", that is, when the elevator system is about to fail but has not yet failed, the control system has already predicted the impending failure through the method of model reference comparison, and through the monitoring The system informs the service center that "somewhere is going to fail". This function has already appeared in some imported elevator monitoring systems (such as the REM system of Otis Corporation). As a mature elevator control system, its remote monitoring part should form a whole with the signal control system and drag control system. and other data from multiple sampling links to comprehensively evaluate the operating status of the system.

In addition, after the elevator breaks down, due to the lack of sufficient information, the maintenance personnel cannot quickly analyze the cause of the failure and eliminate it, which greatly prolongs the maintenance time and brings great inconvenience to people's life and work. Therefore, it is of great practical significance to study the fault diagnosis of the elevator system. It can not only timely and accurately notify the elevator maintenance personnel through various alarm methods when the elevator fails, so that the maintenance personnel can rush to the scene to repair the elevator in time, and solve the regional restrictions of elevator fault diagnosis to a certain extent; at the same time, It can help maintenance personnel to quickly determine the location and cause of the failure after the elevator fails.

When the cause of the failure is known, the intelligent maintenance system provides the maintenance personnel with a fault handling strategy, thereby effectively shortening the downtime caused by the elevator failure, improving the reliability of the elevator operation and the efficiency of elevator maintenance.

Therefore, the research on the elevator local fault prediction, diagnosis and maintenance system is conducive to improving the safety and reliability of the elevator, improving the maintenance service level of the elevator system, and meeting the needs of the domestic market in elevator fault prediction, diagnosis and maintenance.

The purpose of the remote elevator monitoring system is to monitor the operation and fault information of the elevator in real time to ensure the safety of the elevator. However, due to the scattered and wide distribution of elevators, the remote data transmission of elevators based on wired networks has great limitations, such as difficult wiring and high wiring costs. Therefore, it can be seen that the development from wired network to wireless network is the only way to monitor the field. For this reason, it is of great significance to introduce wireless communication technologies such as Wi-Fi, 4G and even future 5G into the elevator monitoring and alarm system.

To sum up, the benefits of remote service to users are obvious. The remote monitoring of elevators not only enables users to obtain a component, but also enables users to enjoy a complete set of services. The remote monitoring and maintenance centers always monitor the elevators they maintain, and can know the operating status and failure of the elevators at any time. The maintenance personnel already know the maintenance items, tools and spare parts that need to be carried before maintenance, reducing maintenance. The cost and time of service, this kind of pre-maintenance after-sales service has won the trust of users abroad, and this will also be an important direction for the development of my country's elevator industry technology. It can be seen that elevator remote monitoring refers to the real-time operation status monitoring of any elevator within its jurisdiction by the elevator service center through wireless communication technology. The implementation of remote monitoring can get the fault information of the elevator at the first time, and deal with it in time, changing passive maintenance into active maintenance, greatly reducing the downtime due to faults. Through remote monitoring, remote fault diagnosis and fault analysis can also be carried out in the service center to assist elevator on-site maintenance personnel to troubleshoot elevator faults.

Contents of the invention

In order to overcome the deficiencies in the prior art, the present invention aims to provide an intelligent elevator failure prediction, diagnosis and maintenance system, which can effectively shorten the downtime caused by elevator failure, improve the reliability of elevator operation and elevator maintenance efficiency, and protect personal safety. and property security. The technical solution adopted in the present invention is, based on the intelligent prediction, diagnosis and maintenance method of the elevator failure of the Internet of Things, the elevator data is collected in real time through the data acquisition module, and then the collected data is transmitted to the server through WIFI or 4G wireless communication technology , and apply big data technology on the server side to analyze the data, so as to predict the possible failure of the elevator; if the elevator fails, use the fault diagnosis system on the server side to analyze the failure, and then give the location and cause of the failure , and then provide a solution to the fault through the intelligent maintenance system on the server side.

Data acquisition and sending: The data acquisition module collects the real-time running data of the elevator through the Can bus, and performs simple processing, including filtering and format conversion, and then uploads the data to the server through WIFI or 4G module using TCP/IP protocol and Socket .

Data receiving and storage method: the server is responsible for receiving data and analyzing and storing data packets: using MySq15.1 as the database server, using Socket to receive data, running a TCP/IP listening program on the server, When sending to the listening port, the server will receive the data. After receiving the data, analyze the data according to the agreed agreement, read the data packet header, and get the verification information. If you do not register on this platform, you will not pass the verification and immediately lose the data; otherwise, continue to analyze the following data, and then Stored in the database; all data tables related to elevators are pre-established in the MySql database.

Server construction: The elevator failure intelligent prediction, diagnosis and maintenance system based on the Internet of Things mainly adopts the Struts+Hibernate+Spring architecture on the server side to realize a scalable elevator network service and management platform;

Big data processing and analysis: the big data analysis of the present invention adopts Hadoop distributed infrastructure to carry out, and Hadoop has realized a distributed file system (Hadoop DistributedFileSystem), be called for short HDFS;

Fault prediction system based on big data analysis: use big data technology to analyze the elevator status data collected in the server database, predict the future operation status of the elevator, and then use the predicted results to guide the future operation of the elevator; through Hadoop technology Use multiple computers to build distributed clusters for big data analysis and data mining; the overall process of fault prediction: information analysis and preprocessing, selection of feature quantities, state evolution analysis, fault type matching, estimated time of fault occurrence or remaining availability Life RUL.

Set up a remote elevator monitoring fault diagnosis system on the server side: When an elevator fails, the remote elevator monitoring fault diagnosis system will find out the location and cause of the fault in time, and then list the possible fault locations and reasons according to the probability, thereby assisting maintenance personnel to diagnose the elevator For timely repair, the server push technology is used in the system to push the monitoring data. When the status of the background server changes, it is hoped that the result will be immediately reflected to the client. We use the HTTP persistent connection method to push the data from the server background to the foreground.

An intelligent maintenance system is set up on the server side: according to the location and cause of the fault, using database matching technology, a fault solution is given; when the local case library cannot provide a solution, the system analyzes the cause of the fault based on machine learning and deep learning Analysis, and then provide solutions to users. Once the problem is successfully solved, the solution will be automatically stored in the local case library to provide solutions for future maintenance. In the intelligent maintenance system, the elevator remote location service subsystem is designed. The geographic location information coordinates of the maintenance personnel can be obtained on the electronic map. This subsystem uses JSON as the data exchange format. When the server sends a request from the front desk, the background will The latitude and longitude of the location is encapsulated into data in JSON format. The front desk uses JavaScript to easily parse the latitude and longitude of the maintenance personnel and send it through the asynchronous http communication framework, and display it on the map page.

Handheld mobile terminal: The implementation of the elevator remote monitoring fault diagnosis and intelligent maintenance system based on the Internet of Things also adopts the MVC design method on the mobile terminal, and the access and operation of the platform database by the mobile terminal are first sent to the server by sending instructions, and then checked by the server. After the verification is valid, the server transfers to the database for access and operation. There are three main ways to control the software interface of the mobile terminal: using xml layout files, using Java code, and a combination of the two; Get the information pushed by the elevator failure in a short time, so that the maintenance personnel can arrive at the accident scene as quickly as possible, and use the function of the intelligent maintenance system on the spot to get the solution strategy conveniently and quickly, and reduce the downtime caused by the elevator failure; through The design of the handheld terminal service software can get the predicted trend, maintenance strategy, and fault location.

Features and beneficial effects of the present invention are:

The invention combines the elevator networking technology, adds the elevator to the Internet, and forms an elevator-centered service and management system. This system can monitor the operating data and geographical location of the elevator in real time through a computer or mobile phone, which brings great convenience to maintenance personnel and management personnel.

The present invention has following functional characteristics:

Fault prediction system based on big data analysis:

Using the intelligent prediction model based on big data analysis, it dynamically predicts the probability of failure in the future or predicts the failure trend in the future, so as to achieve the purpose of predicting failure before failure and eliminating hidden dangers.

①Data receiving and preprocessing module: mainly responsible for receiving and screening the observation data, the data is sent to the fault prediction system through the network, each piece of data is received according to the pre-set protocol, and the system will process the data after receiving the data The validity of the data is judged, and the data is discarded when the data is invalid.

②Fault prediction module: It is mainly responsible for building the prediction model and pre-storing relevant experience, and selecting key feature quantities that can reflect system status and faults, using all available knowledge, combining models, expert experience, and data mining to analyze key features. The evolution of feature quantities, and finally the matching of fault types or fault trend types, realizes the prediction of faults and fault trends.

③Prediction result query module: It is mainly responsible for providing the way for users to query the prediction results. Users can query the fault prediction results according to the fault type, or query the fault prediction results according to the probability priority method.

④Report printing module: Print out the system results in the form of reports, which include some performance indicators and charts, etc., for reporting to superiors, further discussion, etc.

Remote elevator monitoring fault diagnosis system:

①Network communication module: mainly responsible for receiving the real-time operation data of the elevator through the network, including the operation status and fault information of the elevator, and then saving these data to the corresponding database for later data analysis module and elevator historical operation data Inquire.

②Setting and query module: It can set the number of elevators to be monitored, the mode of screen switching, switching time, monitoring mode, personalized interface, etc. At the same time, users can also query historical records, fault information records, historical operating status records, etc. through this module. These data All information provides a good data foundation for big data analysis functions.

③Video monitoring module: including multi-elevator monitoring animation, single-elevator monitoring animation and door animation, with the characteristics of large amount of information and clear at a glance, it is an indispensable function of the monitoring system.

④Performance index analysis module: mainly responsible for reading the real-time data of the elevator, and obtaining various data of the currently running elevator by calling the performance analysis algorithm, including: the average waiting time of passengers, the average ride time of passengers, the average number of starts and stops of elevators, etc. Performance.

⑤ Fault diagnosis module: The diagnosis method used is based on fault tree and expert rules. And record each failure information into the database.

⑥Performance display module: It is mainly responsible for presenting the calculated performance indicators to users in various intuitive ways, for example, the above three performance indicators can be displayed through curves, and bar charts, pie charts, and line charts can also be used and so on represent the running status of each elevator.

⑦Report printing module: It can regularly print out the events (especially faults) and analysis results of some performance indicators of the monitored elevators, so as to report to the superior, keep for backup or analysis, etc.

⑧Video monitoring module: It can visually monitor the internal condition of the elevator, which is helpful for fault analysis.

Among them, in the video monitoring module, an elevator networked digital video monitoring subsystem is specially designed to strengthen the function of video monitoring.

Intelligent maintenance system:

① Data reception and display: Receive and preprocess the elevator operating parameters sent through the 4G or WIFI network, and display various operating parameters in the software interface written with SSH as the framework, and display them in the form of curves, pie charts, Histograms and other methods are displayed in the dynamic screen of the interface.

②Database storage: The remote intelligent maintenance software interacts with the database server, and stores the elevator operation status information transmitted by the local data acquisition system in the defined database in real time, so as to facilitate the query of historical records, data mining of fault trends and fault case system case information.

③Software interface and animation display: Embed animation in the software interface and display the elevator animation according to the feedback elevator running status information. At the same time, it can also draw historical curves and data charts based on the fault information and case information extracted from the database server. Display the health status and running status of the elevator.

④Troubleshooting module: The remote server software retrieves and matches the existing case information in the database through the collected quantitative information and qualitative information, and obtains the similarity and similarity trend with the existing cases, and provides corresponding solutions for maintenance personnel Method. When the case information cannot be matched in the case library, this module provides a fault analysis method based on machine learning and deep learning to provide maintenance personnel with troubleshooting solutions. Once the fault result is eliminated, the case will be automatically stored in the database for future maintenance.

⑤Mark the GPS location information provided by the local system: Obtain the latitude and longitude information of the faulty elevator from the local system through 4G/WIFI and other networks. The collected information is displayed graphically on the software interface of the remote maintenance center to provide supervisors with more intuitive elevator fault conditions and statistics.

⑥Position positioning function: This function needs to collect the location information of the elevator and the location information of the maintenance personnel at the same time. The location information of the maintenance personnel is mainly sent by the smart mobile terminal such as a mobile phone and other devices carried around. Then calculate the position of the idle maintenance personnel closest to the faulty elevator through the distance algorithm in the software.

⑦ Intelligent maintenance scheduling function based on optimal distance: After calculating the idle maintenance personnel nearest to the faulty elevator, the remote service center transmits the maintenance information to the intelligent terminal carried by the maintenance personnel through 4G/WIFI and other networks. Complete specific task scheduling.

Handheld terminal: The present invention adds the design of the client terminal to the handheld terminal, so that maintenance and management personnel can enjoy the convenience brought by the network anytime and anywhere.

The present invention does not only include the functions mentioned above, but only enumerates some common functions. This platform is scalable, and it can expand all elevator-centric functions. This platform provides a good elevator service and management platform framework, and all functions realized on the basis of this framework should be protected.

Description of drawings:

Figure 1. Intelligent elevator fault prediction, diagnosis and maintenance system based on the Internet of Things.

Figure 2 is the flow chart of elevator data collection and transmission program.

Figure 3 Data receiving and storage software flow chart.

Figure 4 is a structural diagram of the remote elevator monitoring fault diagnosis system.

Figure 5 is a flow chart of fault prediction based on big data analysis.

Figure 6 is a structural diagram of intelligent maintenance software.

Figure 7 Link access between the mobile terminal and the server terminal. In the figure, (1) the access structure diagram of the mobile terminal and the server terminal, (2) the flow chart of the mobile terminal and the server terminal access.

detailed description

The purpose of the present invention is to collect real-time operating data from multiple monitored elevators, send these data to the server through WIFI or 4G technology, and perform fault prediction through big data technology; when a fault occurs, the fault location can be diagnosed through the fault diagnosis system and the cause of the failure, the intelligent maintenance system analyzes the solution to the failure, and the maintenance personnel access the relevant services provided by the server through the mobile client or the Web, thereby effectively shortening the downtime caused by the elevator failure and improving the reliability and reliability of the elevator operation. Elevator maintenance efficiency, to ensure the safety of personal and property.

Function and characteristics of the present invention are as follows:

(1) The present invention includes three subsystems, which are respectively a fault prediction system based on big data analysis, a remote elevator monitoring fault diagnosis system and an intelligent maintenance system, and their development on the Web end and the handheld mobile end.

(2) Fault prediction system based on big data analysis

Big data can be summarized as 5 Vs, which are volume, velocity, variety, value, and veracity. With the advent of the era of big data, big data analysis also came into being. Applying big data analysis to the present invention is also a feature of the present invention. Fault prediction is different from fault diagnosis. Fault diagnosis is to find fault causes and fault components according to the faults that occur. Fault prediction is based on existing data. These data are all information about elevators stored in the database. The prediction model predicts the probability of failure in a period of time in the future, so as to eliminate the hidden danger of failure before the failure occurs.

(3) Remote elevator monitoring fault diagnosis system

The remote elevator monitoring fault diagnosis system is an all-round and multi-functional remote elevator monitoring module integrating network communication module, setting and query module, animation monitoring module, performance index analysis module, fault diagnosis module, performance display module, report printing module and video monitoring module. Elevator monitoring fault diagnosis system. The system also integrates the elevator networked digital video monitoring subsystem. After acquiring and transmitting video images to the network through the network camera, the video management software developed under the Windows platform is used to achieve acquisition, storage, management and application of video images, etc. Features.

(4) Intelligent maintenance system

When the remote monitoring fault diagnosis system diagnoses the cause of the fault, in order to facilitate the maintenance personnel to make decisions, this project will develop a remote elevator maintenance service system to realize the remote intelligent maintenance function of the elevator. Maintenance personnel log in to the remote maintenance server through a PC or mobile phone, input the cause of the failure, and the remote maintenance service system will inform the maintenance personnel of the troubleshooting method of the failure based on the experience in the database, which is convenient for the maintenance personnel to work. And we combined the intelligent maintenance system with the electronic map, and designed the location service module, so that the remote management personnel can grasp the geographical location information between the maintenance personnel and the fault location in real time, which is convenient for the management personnel to dispatch the maintenance personnel. Because the number of elevators in my country has increased sharply in recent years, and the number of elevator maintenance personnel is far behind the demand for elevators in use, one maintenance personnel often needs to repair nearly a thousand elevators within a month. In order to improve the overall maintenance efficiency from two aspects of scheduling efficiency and on-site maintenance efficiency, on the basis of solving the on-site maintenance efficiency through the case system, Amap is introduced to improve scheduling efficiency.

(5) Handheld service software

With the rapid development of the Internet industry, various network services have been transformed from desktop to mobile. Many mobile APPs have brought a lot of convenience to people's production and life. Previously, various online activities could only be performed on computers. With the development of the mobile Internet, we can all use mobile phones instead, and the development of mobile terminals for mobile phones is increasingly showing its advantages. In addition, with the advent of the 4G era, using mobile phones to surf the Internet has become faster and more convenient. Combining this system with handheld devices allows maintenance and management personnel to manage and grasp the status of the elevator anytime, anywhere, and does not limit elevator monitoring and other functions to the computer. When the elevator encounters a fault, the maintenance and management personnel's The mobile phone can get the push information of the elevator failure at the first time, so that the maintenance personnel can arrive at the accident scene as quickly as possible, and use the function of the intelligent maintenance system on the spot to get the solution strategy conveniently and quickly, reducing the outage caused by the elevator failure time. Through the design of the hand-held terminal service software, it is possible to obtain the predicted trend, maintenance strategy, and fault location, which greatly improves the benefits brought by the elevator network.

The present invention uses WIFI and 4G wireless communication technology, and combines embedded software and hardware technology, data acquisition technology, etc. to invent an intelligent elevator fault prediction, diagnosis and maintenance system based on the Internet of Things. The present invention collects elevator data in real time through the data collection module, then transmits the collected data to the server through WIFI or 4G wireless communication technology, and analyzes the data by applying big data technology on the server side, thereby predicting that the elevator may failure; if the elevator fails, the fault diagnosis system on the server side will analyze the fault, and then give the location and cause of the fault, and then provide a solution to the fault through the intelligent maintenance system on the server side. In addition, remote video monitoring of the elevator can be carried out on the server side, so that the supervision department can understand the situation in the elevator car in real time.

Data acquisition and transmission method: Each elevator that needs to join the elevator failure intelligent prediction and diagnosis and maintenance system based on the Internet of Things is equipped with a data acquisition module and a WIFI or 4G wireless communication module. Collect and perform simple processing, including filtering and format conversion, etc., and then upload the data to the server through WIFI or 4G module. Define a certain rule by yourself to add a header to the data, then arrange different data in a certain order to form a data packet, and finally send it by the WIFI or 4G module, and use the TCP/IP protocol to write the Socket. The flow of the whole program is: the initialization of the collection and transmission module, the collection and processing of data, the packaging and transmission of data.

Data receiving and storage method: the receiving terminal is mainly responsible for receiving data and analyzing and storing data packets. The present invention adopts MySq15.1 as a database server, and utilizes Socket to realize data reception. There is a TCP/IP listening program running on the server side, and when the data is sent to the listening port, the server side will receive the data. After receiving the data, analyze the data according to the agreed agreement, read the data packet header, and get the verification information. If you do not register on this platform, you will not pass the verification and the data will be lost immediately. Otherwise, continue to parse the following data, and then store it in the database. All data tables related to elevators are pre-established in the MySql database. The entire method flow is: receiving and storing the initialization of the listening program, receiving the request, performing data analysis and judgment, and discarding or storing the data.

Server construction: The intelligent prediction, diagnosis and maintenance system for elevator faults based on the Internet of Things mainly adopts the Struts+Hibernate+Spring architecture on the server side to realize a scalable elevator network service and management platform.

Struts provides a complete MVC development framework, which is mainly responsible for process control, that is, page turning, and data processing operations. The purpose of using Struts is to help us reduce the time to develop Web applications using the MVC design model. If we want to mix the advantages of Servlets and JSP to build scalable applications, Struts is a good choice. And it almost does not directly access the database itself, but calls the related methods of the database operation class (the general name contains DAO).

Hibernate is an open source object-relational mapping framework. It encapsulates JDBC with a very lightweight object, so that Java programmers can use object programming thinking to manipulate the database as they like. It can be used in both Java client programs and Servlet/JSP web applications. In this project, it is mainly responsible for data persistence operations, including data addition, deletion, checking, and modification.

Spring is an open source framework created to address the complexities of enterprise application development. One of the main advantages of the framework is its layered architecture, which allows users to choose which components to use while providing an integrated framework for J2EE application development. However, the use of Spring is not limited to server-side development. Any Java application can benefit from Spring in terms of simplicity, testability, and loose coupling. Its core is IoC control inversion and AOP aspect-oriented programming, which allows programmers to program against interfaces, which is more convenient when modifying functional modules to implement code. And because of dependency injection, programmers don't have to instantiate objects themselves.

Database establishment: We use MySq15.1 as the database server, and use Socket to receive data. Run the TCP/IP listening program on the server side, and when receiving the data sent to the listening port, the server side will receive the data. After receiving the data, analyze the data according to the agreed agreement, read the data packet header, get the verification information, store the data in the database after the verification is valid, otherwise discard the data. All data tables related to the elevator are pre-established in the MySql database, the static information is entered through the software platform, and the dynamic information is stored in real time by the network. The entire method flow is: receiving and storing the initialization of the listening program, receiving the request, performing data analysis and judgment, and discarding or storing the data. And we store the knowledge of elevator faults in the MySQL database, and then realize the elevator fault diagnosis method based on fault trees and expert rules by designing fault trees, knowledge bases, and inference engines.

Big data processing and analysis: the big data analysis of the present invention adopts Hadoop distributed infrastructure to carry out, and Hadoop has realized a distributed file system (Hadoop Distributed File System), referred to as HDFS. HDFS has the characteristics of high fault tolerance and is designed to be deployed on low-cost hardware; and it provides high throughput to access application data, suitable for applications with very large data sets. HDFS relaxes the requirements of POSIX and can access data in the file system in the form of streams. The core design of the Hadoop framework is: HDFS and MapReduce. HDFS provides storage for massive data, while MapReduce provides calculation for massive data. It performs data processing in a reliable, efficient, and scalable manner.

Fault prediction system based on big data analysis: use big data technology to analyze the elevator status data collected in the server database, and predict the future operating status of the elevator. The predicted results are then used to guide future elevator operations. The big data prediction of the present invention is not carried out on a single computer, because a single computer performs big data analysis, which requires extremely high performance of the computer. The present invention uses Hadoop technology to use multiple computers to build a distributed cluster to carry out big data analysis and Data mining, improve operating efficiency, and realize fault prediction function. The overall process of fault prediction: information analysis and preprocessing, selection of feature quantities, state evolution analysis, fault type matching, estimated fault occurrence time or RUL (remaining usable life).

Remote elevator monitoring fault diagnosis system: Through the elevator diagnosis method mentioned above, when the elevator fails, the remote elevator monitoring fault diagnosis system will find the location and cause of the fault in time, and then list the location and cause of the possible fault according to the probability, Thereby assisting the maintenance personnel to repair the elevator in time. In this system, the server push technology is used to push the monitoring data. When the status of the background server changes, it is hoped that the result will be immediately reflected to the client. We use the HTTP persistent connection method to push the data from the server background to the foreground.

Intelligent maintenance system: According to the location and cause of the fault, using the database matching technology, a fault solution is given. When the local case library cannot provide a solution, the system analyzes the cause of the failure based on machine learning and deep learning, and then provides a solution to the user. Once the problem is successfully solved, the solution will be automatically stored in the local case library. Provide solutions for future maintenance. In the intelligent maintenance system, the elevator remote location service subsystem is designed. The geographic location information coordinates of the maintenance personnel can be obtained on the electronic map. This subsystem uses JSON as the data exchange format, which is a subset of javascript syntax and does not require specific The parsing method can easily encapsulate and transmit data, and is very suitable for transmitting geographic location coordinate data. When the server sends a request from the front desk, the backend encapsulates the latitude and longitude of the maintenance personnel’s position into data in JSON format, and the front desk uses JavaScript to parse the latitude and longitude of the maintenance personnel very conveniently and send it to the present invention through the asynchronous http communication framework, thereby displaying it on the map page .

Handheld mobile terminal: The implementation of the elevator remote monitoring fault diagnosis and intelligent maintenance system based on the Internet of Things also adopts the MVC design method on the mobile terminal, and the access and operation of the platform database by the mobile terminal are first sent to the server by sending instructions, and then checked by the server. After the verification is valid, the server transfers to the database for access and operation, which ensures the reliability and security of the information. The mobile terminal also designs a local sqlite lightweight database, and establishes corresponding database tables based on the existing database tables on the server side. Database tables, and according to the relationship between various types of data, the constraint relationship between data tables is obtained. There are mainly three ways to control the software interface of the mobile terminal: using xml layout files, using Java codes, and a combination of the two. In addition, the powerful graphics processing capabilities of the Android system (including 2D and 3D) can be used to optimize the design of the interface, special effects processing of graphics, and texture rendering of images, and the unique 3D effects can also be developed with the help of OpenGLES.

The present invention will be described in further detail below in conjunction with the accompanying drawings.

Fig. 1 The present invention is a set of intelligent elevator fault prediction, diagnosis and maintenance system based on the Internet of Things that integrates functions such as fault alarm, remote monitoring, remote fault diagnosis, remote maintenance service, and big data analysis and prediction. The elevator failure intelligent prediction, diagnosis and maintenance system based on the Internet of Things can be divided into three levels, namely, the perception layer, the network layer, and the application layer. Perception layer: The main function is to use sensors and other means to obtain the required data. Network layer: It mainly uses the network to realize data and instruction transmission functions. Application layer: It is mainly a functional layer for processing and processing the data collected at the bottom layer.

The real-time running data of the elevator is transmitted to the remote server through the network and the real-time data is stored in the database. The database not only stores the dynamic data of the elevator, but also stores all the information of the elevator from being put into use to being scrapped and replaced. Your own identity information file, which contains: address, elevator number, brand, age, property company name and its contact information, maintenance company name and its contact information, SIM card number of the 4G module, etc. Remote workstations can use this system for remote monitoring and diagnosis of fault causes. The fault diagnosis method of the remote monitoring workstation is a fault diagnosis method based on fault tree and expert rules. After diagnosing the cause of the elevator failure, the manager dispatches personnel through the elevator remote maintenance location service subsystem, and the maintenance personnel can use a computer or mobile phone to log in to the remote maintenance server to obtain maintenance methods and experience, and realize the remote intelligent maintenance function of the elevator. And the server platform where this system is located provides the function based on big data analysis, which can predict the fault and fault trend of the elevator, and remind the elevator management personnel to repair and maintain the elevator in time. In addition, this system can also collect images inside the elevator car through the network camera, and transmit them to the server side through the network to realize video monitoring. Finally, an elevator-centered service and management platform is formed through this system, which improves the information service and management level of elevators.

Figure 2 shows the program flow chart of elevator data collection and transmission. By installing data acquisition equipment on the elevator controller, the real-time data of the elevator can be collected and sent to the server. This is the data source for this invention. Its software part mainly includes program initialization module, communication interface setting module, real-time data acquisition module, data processing module and data sending module. When the elevator is running, the data acquisition device initializes the data acquisition and transmission data, including various interfaces and communication modules. After opening and during the elevator running, the data acquisition equipment collects the data of each interface continuously and in real time at a specified fixed time interval, and performs simple data processing, including conversion and adding headers and packaging, and finally through the network module. The data is sent out.

Figure 3 shows the software flow chart for data reception and storage. This part of the software mainly includes a monitoring module, a data receiving and verification module, and an analysis and storage module. This part of the program runs cyclically on the server side and is used to monitor the data storage request of the elevator client.

Figure 4 shows the functional structure diagram of the remote elevator monitoring fault diagnosis system. This service system is functionally expandable, and the ones mentioned in the figure are just some examples. The service and management software platform adopts the Struts+Hibernate+Spring architecture, and uses the MVC development model for modular development, which has strong scalability. Under this architecture, according to the development sequence of persistence layer, DAO layer, business layer to WEB layer, MySql is used as the database as shown in Figure 1. For the persistence layer, analyze the tables involved in the module, determine the relationship between the tables, and then write HibernateMapping files and persistence classes. For the DAO layer, determine the corresponding method according to the various persistence operations required by the module, and then write the DAO interface and implementation class. For the business layer, write business logic to call DAO operations, and then write the Service method interface and implementation class. For the WEB layer, it handles the interaction between the page and the business, and performs necessary verification and conversion, writes Action and Form, and finally uses JSP to display to the user. As introduced above, in fact, the present invention modularizes each system and then integrates them on the Internet of Things platform, which facilitates later maintenance and expansion work.

Figure 5 shows the flow chart of the fault prediction function based on big data analysis. The present invention first acquires and screens the observed data, and sends the data to the fault prediction system through the network, and each piece of data follows a pre-set Protocol to receive, the system will judge the validity of the data after receiving the data, and discard the data when the data is invalid. In the present invention, the fault prediction model is pre-designed and the relevant experience is stored, and then the key feature quantities that can reflect the system state and fault are selected, and all available knowledge is used, combined with the model, environmental stress, and operating data Analyze the evolution of key feature quantities based on the experience of experts, etc., and finally match the type of fault or the type of fault trend to realize the prediction of fault and fault trend.

Figure 6 shows the structure diagram of intelligent maintenance software. It is divided into fault handling and service module and location service module. The fault handling and service module provides fault resolution based on big data and deep learning for elevator faults. The location module combined with the electronic map realizes the intelligent scheduling method of maintenance personnel. The concept of Ping An Cloud has been better implemented.

Figure 7 shows the connection and access between the mobile terminal and the server, and Figure 7 (1) is an overall structural diagram of the mobile terminal and the server. They access and transfer data through the http protocol to achieve data sharing. The main function of the mobile terminal is to enable maintenance personnel and managers to manage and grasp the elevator situation within their jurisdiction in real time, and display it through the app to make the operation more convenient. The interface is more beautiful and the user experience is improved. The flow chart of data transfer is shown in Figure 7(2). The request sent by the mobile terminal needs to be checked by the Web server every time, and then the corresponding data table is accessed through the Web server to obtain data to realize the corresponding mobile terminal functions.

Claims (7)

1. the elevator faults intelligent predicting based on Internet of Things and diagnosis and maintaining method, it is characterized in that, in real time elevator data is collected by data acquisition module, then pass through WIFI or 4G wireless communication technology and the data collected are transferred to server, and at server end application big data technique, data are analyzed, thus predicting the fault that elevator occurs possibly；If elevator there occurs fault, then utilizing the fault diagnosis system of server end that fault is analyzed, the position and reason that then giving is out of order produces, the intelligent maintenance system again through server end gives out of order solution.

2. the elevator faults intelligent predicting based on Internet of Things as claimed in claim 1 and diagnosis and maintaining method, it is characterized in that, data acquisition and transmission: elevator real-time running data is acquired by data acquisition module by Can bus, and simply process, including filtering and form conversion, then pass through WIFI or 4G module and utilize ICP/IP protocol to adopt Socket to upload the data to server end.

3. the elevator faults intelligent predicting based on Internet of Things as claimed in claim 1 and diagnosis and maintaining method, it is characterized in that, the reception of data and storage method: server end is responsible for the reception of data and carries out parsing and the storage of packet: adopt MySql5.1 as database server, Socket is utilized to realize the reception of data, run at server end and have TCP/IP oracle listener, when receive be sent to listening port time, service end can by data receiver.Data are resolved by agreement good according to a preconcerted arrangement after receiving data, read data packet head, are verified information, if not registering at this platform, then will not pass through to verify, at once by loss of data；Otherwise, continue to resolve data below, be then stored in data base；MySql data base pre-builds all tables of data relevant to elevator.

4. the elevator faults intelligent predicting based on Internet of Things as claimed in claim 1 and diagnosis and maintaining method, it is characterized in that, server is built: the elevator faults intelligent predicting based on Internet of Things mainly adopts Struts+Hibernate+Spring framework with diagnosis and maintenance system at server end, it is achieved the terraced the Internet services of multifunction and management platform；

Big data process&analysis: the big data analysis of the present invention adopts Hadoop distributed basis framework to carry out, and Hadoop achieves a distributed file system (HadoopDistributedFileSystem), is called for short HDFS；

Failure prediction system based on big data analysis: utilize the elevator status data collected in big data technique Analysis server data base, is predicted the running status that elevator is following, then utilizes the result the predicted operation to instruct elevator following；Use multiple computers to build distributed type assemblies by Hadoop technology, carry out big data analysis and data mining；Failure predication overall flow: time that information analysis mates with pretreatment, the selection of characteristic quantity, state evolution analysis, fault type, estimate fault occurs or residue probable life RUL.

5. the elevator faults intelligent predicting based on Internet of Things as claimed in claim 1 and diagnosis and maintaining method, it is characterized in that, remote elevator monitoring fault diagnosis system is set at server end: during when Lift out of order, remote elevator monitoring fault diagnosis system can find location of fault and reason in time, then position and the reason of possible breakdown is listed by probability, thus elevator is repaired by auxiliary maintaining personnel timely, server push technology is used to realize the propelling movement of monitoring data within the system, when background server state changes, wish result immediate reaction to client.We use the mode that HTTP length connects to realize the server background data-pushing to foreground.

6. the elevator faults intelligent predicting based on Internet of Things as claimed in claim 1 and diagnosis and maintaining method, it is characterized in that, server end is provided with intelligent maintenance system: according to abort situation and failure cause, utilize database matching techniques, provide fault solution；When local case Kuku can not provide solution, system learns failure cause is analyzed based on machine learning and the degree of depth, is provided to user solution, once problem successfully solves, this solution can be automatically credited in local case library, provides solution for later maintenance.Intelligent maintenance system devises elevator long distance position service sub system, the geographical location information coordinate of maintainer can obtain on electronic chart, this subsystem adopts JSON as data interchange format, when server sends request from foreground, the position longitude and latitude of maintainer is packaged into the data of JSON form by backstage, foreground utilizes JavaScript parse the longitude and latitude of maintainer very easily and sent by asynchronous http communications framework, and is shown in the map page.

7. the elevator faults intelligent predicting based on Internet of Things as claimed in claim 1 and diagnosis and maintaining method, it is characterized in that, hand-held mobile terminal: be also adopted by the design of MVC based on elevator long-distance monitorng fault diagnosis and the intelligent maintenance system realization in mobile terminal of Internet of Things, and access and the operation of platform database are all first pass through to send instructions to reach server by mobile terminal, after server verification effectively, conducted interviews by transit server to data base again and operate, and the control of mobile terminal software interface mainly has three kinds of modes: use xml topology file, use Java code and both combinations；Hand-held mobile terminal is for obtaining, by the mobile phone very first time, the information that elevator faults pushes, make maintainer can arrive at the scene of the accident with the fastest speed, and use the function of intelligent maintenance system at the scene, conveniently obtain resolution policy, reduce the idle time that elevator faults brings；Design by handheld terminal service software, it is possible to obtain the trend of prediction, the strategy of maintenance, location of fault.