CN117023309A - Elevator remote monitoring method - Google Patents

Abstract

The invention discloses an elevator remote monitoring method in the technical field of elevator remote monitoring. The method includes the following steps: determining the operating status monitoring parameters of the elevator or system, obtaining influencing factors and operating status signal parameters, establishing a weak feature extraction method, and establishing elevator performance degradation. By predicting and evaluating the model, quickly finding the final fault or the most likely fault and constructing a fault diagnosis system, the present invention can ensure that the selected parameters can accurately reflect the actual operating status of the elevator, so that it can be used as a state monitoring judgment The standard engineering limit parameter range is reasonable and effective, and the maintenance quality of elevators under maintenance can be remotely monitored and supervised at any time, and potential faults can be predicted and early warned to help technicians and managers make early decisions.

Description

Elevator remote monitoring method

Technical field

The present invention relates to the technical field of elevator remote monitoring, specifically an elevator remote monitoring method.

Background technique

Elevators are widely distributed in office buildings, residential buildings, shopping malls and other places with dense population and frequent movement. Once a safety accident occurs, the involvement will be large and the impact will be severe. The safety and reliability of elevators are currently urgent research topics. To this end, the General Administration of Quality Supervision, Inspection and Quarantine has conducted an "Elevator Safety Battle" to investigate possible safety hazards in elevators across the country. It plays a certain role in the safe operation of elevators in use, but it fails to fundamentally prevent accidents from happening in the elevator itself. In view of elevator safety issues, in-depth systematic research is conducted theoretically and technically on intelligent detection and reliability evaluation of key elevator components, elevator operating status signal feature extraction and fault prediction theory, and elevator safety technology based on the Internet of Things and big data. It is very necessary.

The sensors installed on elevators today have low intelligence, and the measurement errors caused by the sensors themselves make the detection information uncertain and incomplete. In addition, current elevator prediction and diagnosis systems often make predictions based on collected sensor data, which has a certain delay and cannot achieve timely prediction of fault diagnosis. However, elevators have strong real-time requirements. When fault diagnosis symptom data is collected in real time, Finally, the elevator malfunctioned, which would greatly threaten the safety of elevator passengers. Therefore, the accuracy of elevator fault prediction and diagnosis is not only limited by the delay of the sensor's data samples, which becomes a key factor. Based on this, the present invention designs an elevator remote monitoring method to solve the above problems.

Contents of the invention

The purpose of the present invention is to provide an elevator remote monitoring method, which uses a neural network to predict the possible data of the sensor at the next moment, uses the DS data fusion theory to fuse, and obtains fault diagnosis results based on the predicted data of multiple sensors. It overcomes the problems of delay in traditional methods and the difficulty of measuring information from a single sensor to fully and accurately reflect the working status of the elevator, resulting in uncertainty and imprecision in fault diagnosis.

In order to achieve the above objectives, the present invention provides the following technical solution: an elevator remote monitoring method, which includes the following steps:

S1: Determine the operating status monitoring parameters of the elevator or system, determine the key factors and status parameters that affect the safe operation of the elevator, select the key components of the elevator, establish status monitoring points for key equipment, and collect equipment status data;

S2: Based on the systematic analysis of the elevator collected data, the information that can accurately reflect the actual operating status of the elevator is integrated with multi-source information to obtain the influencing factors and operating status signal parameters;

S3: Establish a weak feature extraction method for predictive fault diagnosis based on evidence theory, then establish an elevator performance degradation prediction and evaluation model based on multi-source information, and establish a theoretical evidence theory model for elevator fault diagnosis based on multi-source information fusion, combined with reliability Evaluate and establish dynamic failure function functions of structural systems;

S4: Carry out simulation and operation experiments on the elevator working process. During the operation, after collecting the information of the diagnosed object, the computer calls the database and application program, and after requesting the necessary information from the user, it can quickly find the final fault or the most likely cause. Fault;

S5: The constructed fault diagnosis system consists of a database, a knowledge base, a human-machine interface, and an inference engine. The fault diagnosis system can quickly and accurately analyze sensor signals sent from the wireless transmission system based on the Internet of Things, and then immediately tell the operator what should be done. measure.

Preferably, in steps S1 and S2, stm32MCU and EC20 communication modules are used to build the Internet of Things system. The fault status and diagnosis results are connected through the network and sent to the cloud; fault elevator positioning and Internet of Things node configuration can also be realized through Beidou and GPS. Various types of sensors are used to collect information on various physical and chemical states of the elevator, analyze the current operating state of the elevator, and determine whether the elevator operating state is normal.

Preferably, in step S3, the evidence theory model is as follows:

There are P sensors, Q-type status, P>1, Q>1.

make Represents the measurement data feature vector of the k-th sensor. S _ki is the i-th measurement data feature of S _k ; n _k is the total number of measurement data features provided by the k-th sensor,/>

Create a status matrix:

Among _{them ,} Represents the theoretical value of the corresponding data characteristics of the first sensor in the j-th category state; Represents the theoretical value of the data characteristics of the k-th sensor in the j-th category state. In the formula, i=1,2,…,n; j=1,2,…,Q; k=2,3,…,P.

Define the Minkowski distance between S _k and X _j :

Among them, p is a constant.

Calculate the distance between the feature vector S of all sensor measurement data and the corresponding state vector X, and establish a distance matrix:

The smaller the distance d _kj , the greater the possibility of judging that the object is in the j-th category state based on the k-th sensor information, so it is defined:

Let m _kj and d _kj be inversely correlated and normalized:

Expressed in matrix form:

Then M _k = {m _k1 , m _k2 ,…, m _kQ ,}k=1,2,…,P

It can be used as the confidence value of the k-th sensor for state identification.

Preferably, in step S4, the database usually consists of two parts: a dynamic database and a static database. The static database is a relatively stable parameter, and the dynamic database is a state parameter detected during the operation of the equipment.

Preferably, in step S5, the knowledge stored in the knowledge base is any one or combination of the system's working environment, system knowledge, equipment fault characteristic values, fault diagnosis algorithms, and inference rules, reflecting the causal relationship of the system and used for fault diagnosis. Inference, a knowledge base is a collection of expert domain knowledge.

Compared with the existing technology, the beneficial effects of the present invention are: by constructing an elevator remote monitoring method, the present invention includes sensors, MCU, EC20 full network communication module, power interface, supporting electrical components, completes the corresponding installation, and collects the corresponding operation data. Data, some of the elevator system operating status monitoring parameters are obtained, such as: elevator operating speed, vibration, wear, current, and voltage of electrical equipment. Parameters are obtained from experiments to ensure that the selected parameters can accurately reflect the actual operating status of the elevator. , making the range of engineering limit parameters used as condition monitoring judgment standards reasonable and effective.

By building a comprehensive supervision system, the elevator maintenance unit can instantly query the maintenance status of all elevators in the unit through the electronic supervision system, and establish electronic elevator maintenance management files. It can obtain real-time operation data of remote elevators in a timely and accurate manner, and remotely view elevators. With basic information, maintenance records and historical data, the quality of elevator maintenance under maintenance can be remotely monitored and supervised at any time.

Through the communication module, relevant department personnel can promptly grasp the operating status of the elevators in the monitored "elevator group" anytime and anywhere through PCs, mobile APPs, and tablets, predict and warn potential failures, and help technicians and managers make early decisions.

Description of the drawings

In order to explain the technical solutions of the embodiments of the present invention more clearly, the drawings needed to describe the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without exerting creative efforts.

Figure 1 is a flow chart of diagnosis and prediction of evidence theory based on neural network according to the present invention;

Figure 2 is a flow chart of the fusion processing based on the evidence theory of the present invention;

Figure 3 is an architecture diagram of the Internet of Things system of the present invention;

Figure 4 is a waveform diagram before and after signal fusion and denoising processing according to the present invention;

Figure 5 is an example diagram of the knowledge base of the present invention;

Figure 6 is a diagram of the fault diagnosis system of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

Please refer to Figures 1-6. The present invention provides a technical solution: since most current fault diagnosis methods are based on collected data, an elevator fault diagnosis algorithm based on DS data fusion is proposed. The project uses multi-sensor fusion to obtain predicted data. Then, the membership degree of the data corresponding to each diagnostic category is initialized to assign the initial reliability of the sensor. The data collected by each sensor is used as the body of evidence. The data fusion method is used to fuse each body of evidence to obtain the final diagnostic results. The diagnosis and prediction process of evidence theory based on neural network is shown in Figure 1, and the fusion processing process based on evidence theory is shown in Figure 2.

The present invention uses stm32MCU and Quectel EC20 communication module to build an Internet of Things system, allowing fault status and diagnosis results that were originally unable to be connected to the Internet to be connected through the network and sent to the cloud. The physical architecture is shown in Figure 3. At the same time, faulty elevator location can also be achieved through Beidou and GPS. Internet of Things nodes are equipped with various types of sensors to collect information on various physical and chemical states of the elevator, analyze the current operating state of the elevator, and determine whether the operating state of the elevator is normal. When the parameter measurement value exceeds the standard limit range, it is necessary to Further technical analysis and diagnosis include diagnosing not only parts and components, but also the entire equipment and even the system, and combined with the diagnosis of the operating process to form a comprehensive system diagnosis of the elevator system. Video collection includes face recognition, facial expression recognition, lip reading, head movement tracking, gaze tracking, gesture recognition, and body posture recognition. By processing and analyzing surveillance video data, a mapping can be established between the video and behavioral descriptions. By obtaining the content of the video, it automatically matches the pre-set rules so that the computer can "see" the video or "understand" the video, replacing the monitoring personnel to complete part of the monitoring role, saving manpower.

The elevator remote monitoring method proposed by the present invention includes the following steps:

S1: Determine the operating status monitoring parameters of the elevator or system, determine the key factors and status parameters that affect the safe operation of the elevator, select the key components of the elevator, establish status monitoring points for key equipment, and collect equipment status data;

S2: Based on the systematic analysis of the elevator collected data, starting from the perspective of predictive design, with the goal of meeting the needs of elevator power machinery status monitoring capabilities, the connotation of early monitoring is incorporated into the design, manufacturing and use of products throughout the entire life and system cycle. During the maintenance service process, the selected parameters can accurately reflect the actual operating status of the elevator, and a large amount of information such as vibration, temperature, output power, wear and electrical parameters can be integrated with multi-source information. Eliminate possible redundancies and contradictions between multi-sensor information and complement each other to reduce their uncertainty and obtain influencing factors and operating status signal parameters;

S3: Establish a weak feature extraction method for predictive fault diagnosis based on evidence theory, and then establish an elevator performance degradation prediction and evaluation model based on multi-source information such as vibration, wear, and temperature. Establish an elevator fault diagnosis theory based on multi-source information fusion, develop a predictive fault diagnosis system based on big data, and establish a dynamic failure function function of the structural system based on reliability evaluation;

The evidence theory model is as follows:

There are P sensors, Q-type status, P>1, Q>1.

make Represents the measurement data feature vector of the k-th sensor. S _ki is the i-th measurement data feature of S _k ; n _k is the total number of measurement data features provided by the k-th sensor,

Create a status matrix:

_{Among them} , Represents the theoretical value of the corresponding data characteristics of the first sensor in the j-th category state; Represents the theoretical value of the data characteristics of the k-th sensor in the j-th category state. In the formula, i=1,2,…,n; j=1,2,…,Q; k=2,3,…,P.

Define the Minkowski distance between S _k and X _j :

Among them, p is a constant.

Calculate the distance between the feature vector S of all sensor measurement data and the corresponding state vector X, and establish a distance matrix:

The smaller the distance d _kj , the greater the possibility of judging that the object is in the j-th category based on k sensor information, so it is defined:

Let m _kj and d _kj be inversely correlated and normalized:

Expressed in matrix form:

Then M _k = {m _k1 , m _k2 ,…, m _kQ ,}k=1,2,…,P

It can be used as the confidence value of the k-th sensor for state recognition.

S4: Through the simulation and operation experiment of the elevator working process, after collecting the information of the diagnosed object during the operation, the computer comprehensively uses various rules (expert experience) to conduct a series of reasoning, and can call various applications at any time when necessary. The program can quickly find the final fault or the most likely fault after requesting the necessary information from the user;

S5: The database usually consists of two parts: dynamic database and static database. The static database is relatively stable parameters, such as the design parameters of the equipment, natural frequency, etc.; the dynamic database is the status parameters detected during the operation of the equipment, such as vibration, operating speed, elevator passenger flow, traction machine operating voltage or current, etc. Figure 5 shows the established knowledge base for the large vibration peak value of the traction machine.

The knowledge stored in the knowledge base can be the system's working environment, system knowledge (reflecting the system's working mechanism and system structure knowledge), equipment fault characteristic values, fault diagnosis algorithms, reasoning rules, etc., which reflect the causal relationship of the system and are used for fault reasoning. ,Knowledge base is a collection of expert domain knowledge.

The fault diagnosis system constructed is shown in Figure 6. It consists of a database, a knowledge base, a human-machine interface, an inference engine, etc. It can quickly and accurately analyze sensor signals sent from the wireless transmission system based on the Internet of Things, and then immediately tell the operator what measures to take. .

Elevator remote monitoring technology is an emerging technology for centralized remote control detection of running elevators that has emerged with the development of computer control technology and network communication technology. It proposes a brand-new product concept and service concept, and is an important part of current elevator service management. According to the operating status of the field, the fault rate of the elevator is automatically calculated based on the fault record. Through it, the status of the elevator and the work quality of the maintenance unit can be effectively supervised, and a reliable basis can be provided for inspection and assessment.

Integrating the Internet of Things into the project, stm32MCU and Quectel EC20 communication modules are used to build an Internet of Things fault diagnosis system, so that fault status and diagnosis results that are not originally connected to the Internet can be connected through the network and sent to the cloud. At the same time, elevator positioning can be achieved through Beidou and GPS. It allows relevant departments and personnel to grasp the operating status of the elevators in the monitored "elevator group" anytime and anywhere through PCs, mobile APPs, and tablets, predict and warn potential failures, and help technicians and managers make early decisions to prevent problems before they occur; When an accident occurs in an elevator, promptly and accurately determine the type and address of the accident, call the police, notify relevant departments and personnel, and take corrective measures. It consists of IoT data collection system, transmission network and control center. The data acquisition system collects and processes the elevator's operating status and related data, packages the data and sends information packets to the remote control center through RS or TCP/IP networking according to the set plan. The remote control center analyzes the information packet after receiving it. Processing simultaneously stores information in a database. The control center provides users with a standardized monitoring window that can display the status of designated elevators in real time. When an elevator fails, it can alarm in time, allowing maintenance personnel to arrive at the scene as soon as possible to deal with the fault. At the same time, the operating data before and after the fault are automatically extracted separately and saved for a long time for fault analysis; the method of data statistics is used to analyze the type of fault. Prevent elevator failures and improve elevator operation safety; provide direct data support for daily maintenance and upkeep of elevators to improve maintenance efficiency and reduce maintenance costs; automatically record the type and occurrence time of faults during elevator operation, so as to Master the real operating status of the elevator; record the inspection and maintenance time of maintenance personnel to facilitate the supervision and verification of maintenance work by the management department, and implement the elevator supervision and evaluation system.

In the description of this specification, reference to the terms "one embodiment," "example," "specific example," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one aspect of the invention. in an embodiment or example. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

The preferred embodiments of the invention disclosed above are only intended to help illustrate the invention. The preferred embodiments do not describe all details, nor do they limit the invention to the specific implementations described. Obviously, many modifications and variations are possible in light of the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art can better understand and utilize the present invention. The invention is limited only by the claims and their full scope and equivalents.

Claims (5)

1. The elevator remote monitoring method is characterized by comprising the following steps of:

s1: determining operation state monitoring parameters of an elevator or a system, determining key factors and state parameters affecting the safe operation of the elevator, selecting key parts of the elevator, establishing state monitoring points for key equipment, and collecting equipment state data;

s2: on the basis of carrying out system analysis on elevator collected data, carrying out multi-source information fusion on information capable of accurately reflecting the actual running state of an elevator to obtain influencing factors and running state signal parameters;

s3: establishing a weak feature extraction method of predictive fault diagnosis based on an evidence theory, further establishing an elevator performance degradation prediction and evaluation model based on multi-source information, establishing an elevator fault diagnosis theory evidence theory model based on multi-source information fusion, and establishing a structural system dynamic failure function by combining reliability evaluation;

s4: the elevator working process is simulated and run, the computer in the running process calls a database and an application program after collecting the information of the diagnosed object, and after asking for necessary information from a user, the computer can quickly find out the final fault or the most likely fault;

s5: the constructed fault diagnosis system comprises a database, a knowledge base, a man-machine interface and an inference engine, and can rapidly and accurately analyze sensor signals sent by the wireless transmission system based on the Internet of things and immediately tell an operator what measures to take.

2. The elevator remote monitoring method according to claim 1, characterized in that: in the step S1 and the step S2, an Internet of things system is built by adopting the stm32MCU and the EC20 communication module, and the fault state and the diagnosis result are networked through a network and sent to a cloud; the fault elevator positioning can be realized through the Beidou and the GPS, and the nodes of the Internet of things are provided with various types of sensors for acquiring information of various physical and chemical states of the elevator, analyzing the current running state of the elevator and judging whether the running state of the elevator is normal or not.

3. The elevator remote monitoring method according to claim 1, characterized in that: in step S3, the evidence theory model is as follows:

p sensors, Q-class state, P >1, Q >1 are provided.

Order theRepresenting the measured data feature vector of the kth sensor. S is S _ki Is S _k Is the i-th measurement data feature of (a); n is n _k Is the total number of measured data features provided by the kth sensor,/for>

Establishing a state matrix:

wherein X is _j Is a vector describing the j-th class of states, x _ji > 0 represents the ith theoretical measured data characteristic representing the jth class of state;theoretical values representing corresponding data characteristics of the 1 st sensor in the j-th state;a theoretical value representing the data characteristic of the kth sensor in the jth state. Wherein i=1, 2, …, n; j=1, 2, …, Q; k=2, 3, …, P.

Definition S _k And X is _j Minkowski distance in between:

where p is a constant.

Calculating the distance between the feature vector S and the corresponding state vector X of all sensor measurement data, and establishing a distance matrix:

distance d _kj The smaller the probability of judging that the object is in the j-th state based on the k-th sensor information is, the more likely it is, thus defining:

let m _kj And d _kj And (3) performing inverse correlation and normalization:

expressed in matrix form:

m is then _k ＝{m _k1 ，m _k2 ，…，m _kQ ，}k＝1,2,…，P

Can be used as the confidence value of the kth sensor for state identification.

4. The elevator remote monitoring method according to claim 1, characterized in that: in step S4, the database is typically composed of two parts, a dynamic database, which is a relatively stable parameter, and a static database, which is a status parameter detected during operation of the device.

5. The elevator remote monitoring method according to claim 1, characterized in that: in step S5, the knowledge stored in the knowledge base is any one or combination of working environment, system knowledge, equipment fault characteristic value, fault diagnosis algorithm and reasoning rule of the system, reflects causal relationship of the system, and is used for performing fault reasoning, and the knowledge base is a set of expert domain knowledge.