Elevator safety monitoring and fault prediction system based on Internet of Things technology

Abstract

The invention discloses an elevator safety monitoring and fault prediction system based on the Internet of Things technology. The elevator safety monitoring and fault prediction system comprises a data acquisition module, a data processing module, a fault early warning module, a fault prediction module and an information display and interaction module based on a system architecture. Compared with the prior art, the system has the advantages that real-time monitoring and early warning are realized; accurate prediction and prevention are realized; intelligent processing and display are realized; the maintenance efficiency is improved; the maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator safety monitoring, and in particular to an elevator safety monitoring and fault prediction system based on Internet of Things technology. Background Art

[0002] With the acceleration of urbanization, elevators have become an indispensable means of vertical transportation in modern buildings. However, elevators often fail during operation due to various reasons, such as mechanical wear, electrical failure, environmental factors, etc. These failures not only affect people's daily life and work, but may also cause personal injury and property loss. Traditional elevator monitoring systems mainly rely on manual inspections and post-maintenance, which have problems such as slow response speed, high maintenance costs, and poor preventiveness. Therefore, it is particularly important to develop an intelligent system that can monitor the operating status of elevators in real time, predict potential failures, and provide early warnings. Summary of the invention

[0003] The present invention aims to solve the problems of slow response speed, high maintenance cost, poor preventiveness, etc. in the existing elevator monitoring system. By integrating Internet of Things technology, data processing technology and prediction algorithm, real-time monitoring, fault warning and fault prediction of the elevator operation status can be realized, thereby improving the safety and reliability of the elevator and reducing maintenance costs.

[0004] In order to solve the above technical problems, the technical solution provided by the present invention is: an elevator safety monitoring and fault prediction system based on the Internet of Things technology, including a data acquisition module, a data processing module, a fault warning module, a fault prediction module, and an information display and interaction module based on the system architecture, wherein the data acquisition module collects various physical quantity data during the operation of the elevator, including but not limited to the vibration information of the elevator car, the tension information of the wire rope, the air pressure information, the acceleration information and the switch status of the elevator door; the data processing module receives the data transmitted by the data acquisition module, performs preprocessing, feature extraction and data analysis to identify abnormal conditions in the operation of the elevator; the fault warning module warns of possible faults in the operation of the elevator according to the analysis results of the data processing module, and sends a warning signal to the monitoring center; the fault prediction module predicts possible future faults of the elevator based on historical data and real-time data using a prediction algorithm, and generates a prediction report; the information display and interaction module provides a user interface for displaying the operation status of the elevator, fault warning and prediction information, and supports functions such as user query of historical data, setting of system parameters and receiving of system notifications.

[0005] Compared with the prior art, the advantages of the present invention are: 1) Real-time monitoring and early warning: The real-time monitoring of the elevator operation status is realized through the Internet of Things technology. Combined with the fault early warning module, potential fault hazards can be discovered and warned in time, thereby improving the safety of elevator operation;

[0006] 2) Accurate prediction and prevention: Use machine learning or deep learning algorithms to predict possible future elevator failures and provide detailed prediction reports and recommended measures, which helps to formulate maintenance plans in advance and reduce maintenance costs and unplanned downtime;

[0007] 3) Intelligent processing and display: The data processing module uses advanced signal processing technology and data analysis algorithms to efficiently process the collected data; the information display and interaction module provides an intuitive and easy-to-use user interface, allowing users to understand the operation status and system status of the elevator at any time;

[0008] 4) Improve maintenance efficiency: Through real-time monitoring, early warning and prediction functions, maintenance personnel can quickly locate and handle faults, improving maintenance efficiency and response speed;

[0009] 5) Reduce maintenance costs: By providing early warning and predicting failures, unplanned downtime and maintenance times can be reduced, thereby reducing maintenance costs.

[0010] Furthermore, the data acquisition module includes: 1) a vibration sensor for collecting vibration information of the elevator car; 2) a tension sensor for collecting tension information of the elevator wire rope; 3) an air pressure sensor for collecting air pressure information at the location of the elevator car; 4) an acceleration sensor for detecting the acceleration of the elevator car; 5) a door status sensor for detecting the switch status of the elevator door.

[0011] Furthermore, the data processing module includes a signal processing circuit and a data analysis algorithm for filtering, denoising, feature extraction and pattern recognition of the collected data to extract key features reflecting the operating status of the elevator.

[0012] Furthermore, the fault warning module judges the abnormal data output by the data processing module according to the preset warning rules. Once the warning condition is triggered, it immediately sends a warning signal to the monitoring center and displays the warning information to the user through the information display and interaction module.

[0013] Furthermore, the fault prediction module uses machine learning or deep learning algorithms, combined with historical data and real-time data, to predict possible future elevator failures and generate a detailed prediction report, including the fault type, predicted time, possible causes and recommended measures.

[0014] Furthermore, the filtering process of the data processing module is implemented by a digital filtering algorithm to remove or weaken the noise component while retaining the useful information in the signal. The specific steps include:

[0015] 1) Design the filter: First, design the coefficients of the FIR filter according to the cutoff frequency, passband and stopband attenuation, which can be done by the window function method or the frequency sampling method. Assume that the designed filter coefficient vector is h[n], where n ranges from 0 to N-1, and N is the order of the filter;

[0016] 2) Initialization: Initialize a buffer to store the most recent N input samples. This buffer is usually initialized to zero or a specific initial condition at the beginning of the algorithm, and the output y[m] is initialized to zero;

[0017] 3) Filtering process: put the new input sample x[m] at the front of the buffer and remove the oldest sample in the buffer, that is, shift the buffer left by one bit, and calculate the current output y[m], which is the convolution of the current input and the filter coefficient:

[0018]

[0019] Where, h[n]: coefficient of the FIR filter, where n is the index of the coefficient, from 0 to N-1;

[0020] x[m]: input sample at time m;

[0021] y[m]: The output sample at time m, which is the convolution result of the input sample and the filter coefficient;

[0022] N: the order of the filter, i.e. the number of coefficients;

[0023] When mn < 0, x[mn] should be considered zero. This is usually achieved by ensuring that the buffer contains enough zeros or initial conditions at the beginning of the algorithm.

[0024] 4) Output result: Output the calculated y[m] and repeat steps 3 and 4 for each new input sample x[m+1], x[m+2],…

[0025] Furthermore, in the denoising process, fast median filtering is used for image denoising, while empirical mode decomposition is used for signal denoising. The fast median filtering steps include: 1) selecting the window size: determining the size of the filter window, 3x3 or 5x5; 2) sliding the window: traversing each pixel in the image and placing the filter window with the pixel as the center; 3) sorting and taking the median: sorting the pixel values ​​in the window and taking the median as the new value of the center pixel, which is expressed as:

[0026]

[0027] The median function is used to calculate the median value, and the get_window function is used to extract the pixel value within the window from the image;

[0028] The steps of EMD include:

[0029] 1) Initialization: Let the original signal be x(t), the first IMF component be c1(t), and the residual be r1(t)=x(t);

[0030] 2) Extract IMF: Find r n All local maxima and minima of (t) are fitted with cubic spline interpolation to form the upper envelope u n (t) and the lower envelope l n (t), calculate the average envelope Calculate the first IMF component: c n (t) = r n (t)-m n (t), check c n (t) Whether the IMF condition is met, that is, the number of extreme points and zero-crossing points does not differ by more than 1 and is locally symmetrical; if not, c n (t) as a new r n (t), repeat the above steps;

[0031] 3) Calculate the residual: r n+1 (t) = r n (t)-c n (t);

[0032] 4) Repeat steps 2 and 3 until the residual r n (t) becomes small enough or becomes a monotonic function, and the IMF can no longer be extracted;

[0033] 5) Reconstructed signal: The original signal can be expressed as the sum of all IMF components and the final residual: BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the architecture of the patent of this invention.

[0035] Figure 2 It is a schematic diagram of the work flow of the patent of this invention. DETAILED DESCRIPTION

[0036] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0037] Embodiment 1

[0038] like Figure 1 and Figure 2 As shown in the figure, the workflow of this system includes:

[0039] 1) System initialization

[0040] Hardware configuration: Install all necessary sensors (such as vibration sensors, tension sensors, air pressure sensors, acceleration sensors, door status sensors, etc.) at key locations of the elevator. At the same time, ensure the correct connection and configuration of hardware devices such as data processing units, communication modules, and information display terminals.

[0041] Software deployment: Deploy data acquisition software, data processing algorithms, fault warning and prediction algorithms, and information display and interaction software on the data processing unit. Ensure that all software modules can operate normally and communicate with each other.

[0042] Parameter setting: According to the specific model and operating environment of the elevator, set the key parameters of the data acquisition module, such as sampling frequency, warning threshold, and prediction algorithm parameters.

[0043] 2) Data collection

[0044] Real-time monitoring: The data acquisition module collects various physical quantity data of the elevator in real time through various sensors during operation, including vibration information, wire rope tension, air pressure, acceleration, and elevator door switch status.

[0045] Data transmission: The collected data is transmitted to the data processing unit via wired or wireless means to ensure the real-time and accuracy of data transmission.

[0046] 3) Data processing

[0047] Preprocessing: Perform preprocessing operations such as filtering and denoising on the received raw data to improve data quality.

[0048] Feature extraction: Extract key features reflecting the elevator's operating status from the preprocessed data, such as vibration frequency, acceleration change rate, wire rope tension fluctuation, etc.

[0049] Data analysis: Use data analysis algorithms to further process the extracted features and identify abnormal conditions in elevator operation, such as abnormal vibrations and sudden changes in tension.

[0050] 4) Fault warning and prediction

[0051] Fault warning: According to the preset warning rules, the data analysis results are judged. If the abnormal situation reaches the warning threshold, the warning mechanism is triggered immediately, and a warning signal is sent to the monitoring center, and the warning information is displayed to the user through the information display and interaction module.

[0052] Fault prediction: Based on historical data and real-time data, prediction algorithms such as machine learning or deep learning are used to predict possible future elevator faults. The prediction results include fault type, predicted time, possible causes and recommended measures, and a detailed prediction report is generated.

[0053] 5) Information display and interaction

[0054] Status display: The elevator's operating status, warning information and forecast reports are displayed in real time on the information display terminal for users to monitor and query.

[0055] User interaction: Provides a user interaction interface that allows users to query historical data, set system parameters, receive system notifications, etc. At the same time, supports users to input feedback information through the interface to continuously optimize and improve the system.

[0056] 6) System maintenance

[0057] Data backup: Regularly back up important data in the system to prevent data loss or damage.

[0058] Software upgrade: According to system operation status and user needs, the software is regularly upgraded and maintained to ensure system stability and security.

[0059] Hardware maintenance: Regularly inspect and replace sensors and other hardware equipment to ensure the accuracy and reliability of data collection.

[0060] Embodiment 2

[0061] Based on the first embodiment, Figure 1 and Figure 2 As shown, firstly, the raw data from each sensor is received, and the appropriate filtering algorithm is selected according to the sensor type and data characteristics to filter the raw data. For image data or signal data that needs further denoising, the corresponding denoising technology is used for processing. The filtered and denoised data are integrated to form a preprocessed data set. The preprocessed data is input into the data analysis, fault warning and prediction modules for further processing. Extracting key features reflecting the elevator operation status from the preprocessed data is an important part of the elevator safety monitoring and fault prediction system. These features can directly reflect the elevator operation status and provide an important basis for subsequent fault warning and prediction. In terms of threshold detection, one or more thresholds are set. When the characteristic value exceeds or falls below these thresholds, it is considered that an abnormal situation has occurred. It is applicable to features with a clear normal range such as vibration frequency and tension. By setting reasonable thresholds through historical data and empirical knowledge, abnormal vibration and tension mutations in elevator operation can be monitored in real time. Specifically: 1) set the normal range thresholds of vibration frequency, tension and other characteristics; 2) collect elevator operation data in real time; 3) check the data at each time point to determine whether it exceeds the set threshold; 4) if it exceeds the threshold, it is determined to be an abnormal situation. Assuming the vibration frequency threshold is T and the current vibration frequency is f, the abnormal situation determination condition is:

[0062] |f-μ|>T or f>Upper_Bound or f<Lower_Bound

[0063] Among them, μ is the average value of the vibration frequency (optional), and Upper_Bound and Lower_Bound are directly set upper and lower bounds. The execution code is:

[0064] importnumpyasnp

[0065] #Assume that data is a NumPy array containing vibration frequencies

[0066] data = np.random.randn(100)*0.5+10 # Generate simulated data with a mean of 10 and a standard deviation of 0.5

[0067] #Set threshold

[0068] threshold=1.0

[0069] # Anomaly Detection

[0070] anomalies=np.where(np.abs(data-np.mean(data))>threshold)[0]

[0071] print("Anomaly index:", anomalies)

[0072] #If you need to process real-time data, you can apply this detection logic in real time when the data arrives.

[0073] Embodiment 3

[0074] Based on the second embodiment, Figure 1 and Figure 2 As shown, in the machine learning algorithm of this scheme, the anomaly detection algorithm uses the isolation forest algorithm model to detect anomalies in the data set. The isolation forest algorithm divides the data space by randomly selecting features until all data points are isolated. Because anomalies are different from most data points, they are usually isolated faster and therefore have a higher anomaly score. Specifically: 1) prepare training data, including normal and (optional) abnormal samples; 2) use the isolation forest algorithm to train the model; 3) use the trained model to score the anomaly of real-time data; 4) determine whether it is an abnormal situation based on the anomaly score. The execution code is:

[0075]

[0076]

[0077] The above shows and describes the basic principles and main features of the present invention and the advantages of the invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An elevator safety monitoring and fault prediction system based on Internet of Things technology, including a data acquisition module, a data processing module, a fault warning module, a fault prediction module, and an information display and interaction module based on the system architecture, characterized in that: The data acquisition module collects various physical quantity data during the operation of the elevator, including but not limited to the vibration information of the elevator car, the tension information of the wire rope, the air pressure information, the acceleration information and the switch status of the elevator door; the data processing module receives the data transmitted by the data acquisition module, performs preprocessing, feature extraction and data analysis to identify abnormal conditions in the operation of the elevator; The fault warning module warns of possible faults that may occur during elevator operation based on the analysis results of the data processing module, and sends a warning signal to the monitoring center; the fault prediction module uses a prediction algorithm based on historical data and real-time data to predict possible future elevator faults and generate a prediction report; the information display and interaction module provides a user interface for displaying the elevator operation status, fault warning and prediction information, and supports users to query historical data, set system parameters, and receive system notifications.

2. According to claim 1, an elevator safety monitoring and fault prediction system based on Internet of Things technology is characterized in that The data acquisition module includes: 1) a vibration sensor for collecting vibration information of the elevator car; 2) a tension sensor for collecting tension information of the elevator wire rope; 3) an air pressure sensor for collecting air pressure information at the location of the elevator car; 4) an acceleration sensor for detecting the acceleration of the elevator car; 5) a door state sensor for detecting the switch state of the elevator door.

3. The elevator safety monitoring and fault prediction system based on Internet of Things technology according to claim 1 is characterized in that: The data processing module includes a signal processing circuit and a data analysis algorithm, which is used to perform filtering, denoising, feature extraction, pattern recognition and other processing on the collected data to extract key features reflecting the operating status of the elevator.

4. The elevator safety monitoring and fault prediction system based on Internet of Things technology according to claim 1 is characterized in that: The fault warning module judges the abnormal data output by the data processing module according to the preset warning rules. Once the warning condition is triggered, it immediately sends a warning signal to the monitoring center and displays the warning information to the user through the information display and interaction module.

5. The elevator safety monitoring and fault prediction system based on Internet of Things technology according to claim 1 is characterized in that: The fault prediction module uses machine learning or deep learning algorithms, combined with historical data and real-time data, to predict possible future elevator faults and generate a detailed prediction report, including fault type, predicted time, possible causes and recommended measures.

6. The elevator safety monitoring and fault prediction system based on Internet of Things technology according to claim 3 is characterized in that The filtering process of the data processing module is implemented by a digital filtering algorithm to remove or weaken the noise component while retaining the useful information in the signal. The specific steps include: 1) Design the filter: First, design the coefficients of the FIR filter according to the cutoff frequency, passband and stopband attenuation, which can be done by the window function method or the frequency sampling method. Assume that the designed filter coefficient vector is h[n], where n ranges from 0 to N-1, and N is the order of the filter; 2) Initialization: Initialize a buffer to store the most recent N input samples. This buffer is usually initialized to zero or a specific initial condition at the beginning of the algorithm, and the output y[m] is initialized to zero; 3) Filtering process: put the new input sample x[m] at the front of the buffer and remove the oldest sample in the buffer, that is, shift the buffer left by one bit, and calculate the current output y[m], which is the convolution of the current input and the filter coefficient: Where, h[n]: coefficient of the FIR filter, where n is the index of the coefficient, from 0 to N-1; x[m]: input sample at time m; y[m]: The output sample at time m, which is the convolution result of the input sample and the filter coefficient; N: the order of the filter, i.e. the number of coefficients; When mn < 0, x[mn] should be considered zero. This is usually achieved by ensuring that the buffer contains enough zeros or initial conditions at the beginning of the algorithm. 4) Output result: Output the calculated y[m] and repeat steps 3 and 4 for each new input sample x[m+1], x[m+2],… 7. The elevator safety monitoring and fault prediction system based on Internet of Things technology according to claim 3 is characterized in that In the denoising process, fast median filtering is used for image denoising, while empirical mode decomposition is used for signal denoising. The fast median filtering step includes: 1) selecting the window size: determining the size of the filter window, 3x3 or 5x5; 2) sliding the window: traversing each pixel in the image and placing the filter window with the pixel as the center; 3) sorting and taking the median: sorting the pixel values ​​in the window and taking the median as the new value of the center pixel, which is expressed as: for each pixel(i,j)in image: window = get_window (image, i, j, window_size) # Assume that window_size is the window size median_value = median(window) image[i,j]=median_value The median function is used to calculate the median value, and the get_window function is used to extract the pixel value within the window from the image; The empirical mode decomposition step comprises: 1) Initialization: Let the original signal be x(t), the first IMF component be c1(t), and the residual be r1(t)=x(t); 2) Extract IMF: Find r n All local maxima and minima of (t) are fitted with cubic spline interpolation to form the upper envelope u n (t) and the lower envelope l n (t), calculate the average envelope Calculate the first IMF component: c n (t) = r n (t)-m n (t), check c n (t) Whether the IMF condition is met, that is, the number of extreme points and zero-crossing points does not differ by more than 1 and is locally symmetrical; if not, c n (t) as a new r n (t), repeat the above steps; 3) Calculate the residual: r n+1 (t) = r n (t)-c n (t); 4) Repeat steps 2 and 3 until the residual r n (t) becomes small enough or becomes a monotonic function, and the IMF can no longer be extracted; 5) Reconstructed signal: The original signal can be expressed as the sum of all IMF components and the final residual:

Images