Stair anti-collision prediction and early warning method for escalator

Abstract

The invention relates to the technical field of data mining, in particular to a stair anti-collision prediction and early warning method for an escalator. The method comprises the following steps that escalator operation data are obtained and analyzed, escalator abnormal operation data and escalator normal operation data are obtained, and an escalator abnormal operation detection model is constructed; analyzing the abnormal operation data of the escalator to obtain escalator guide rail vibration intensity and characteristic frequency spectrum change data; defining according to the escalator guide rail vibration intensity and the characteristic frequency spectrum change data to obtain escalator guide rail operation degradation trend data; real-time escalator operation data are obtained, trend clustering is conducted on the real-time escalator operation data, and real-time escalator operation disadvantage trend data are obtained; and carrying out escalator operation risk assessment according to the real-time escalator operation disadvantage trend data so as to obtain escalator operation risk data, and uploading the escalator operation risk data to an escalator management platform so as to execute an escalator risk early warning task. The safety performance of the escalator can be improved.

Description

Technical Field

[0001] The invention relates to the technical field of data mining, and in particular to a step anti-collision prediction and early warning method for an escalator. Background Art

[0002] The development of escalator step collision prediction and warning methods can be traced back to the continuous attention and improvement of escalator safety. At first, escalators relied on simple physical sensors, such as limit switches, to monitor the running status of the steps, but this method often failed to accurately detect small objects or irregular behaviors on the steps. Traditional escalator step collision monitoring usually relies on a single physical sensor, such as a pressure sensor or a limit switch. These sensors can only provide limited information, such as whether there is an object on the step or whether someone is standing. However, they cannot provide detailed information about specific objects or people, nor can they accurately judge potential dangerous situations, such as people falling or foreign objects stuck on the ladder. Single sensor technology often cannot accurately distinguish between real dangerous events and false alarms. For example, a pressure sensor may mistake a heavy object for a person, or fail to recognize the presence of a small object. In this case, the monitoring system may frequently issue false alarm warnings, affecting the reliability and operational efficiency of the system. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide an escalator step anti-collision prediction and warning method to solve at least one of the above technical problems.

[0004] To achieve the above object, a step anti-collision prediction and warning method for an escalator comprises the following steps: Step S1: acquiring escalator operation data, and performing escalator abnormal operation analysis on the escalator operation data, thereby obtaining escalator abnormal operation data and escalator normal operation data, and constructing an escalator abnormal operation detection model based on the escalator abnormal operation data and the escalator normal operation data; Step S2: extracting abnormal operation sensor data from the escalator abnormal operation data to obtain abnormal operation sensor data, and analyzing the escalator guide rail vibration intensity and characteristic spectrum change according to the abnormal operation sensor data to obtain escalator guide rail vibration intensity and characteristic spectrum change data; Step S3: defining the guide rail operation degradation trend according to the escalator guide rail vibration intensity and characteristic spectrum change data, thereby obtaining the escalator guide rail operation degradation trend data; Step S4: acquiring real-time escalator operation data, performing escalator abnormal operation detection on the real-time escalator operation data through an escalator abnormal operation detection model, thereby obtaining escalator abnormal detection data; clustering escalator guide rail operation trends on the natural escalator abnormal detection data according to the escalator guide rail operation disadvantage trend data, thereby obtaining real-time escalator operation disadvantage trend data; Step S5: Perform escalator operation risk assessment based on real-time escalator operation disadvantage trend data to obtain escalator operation risk data, and upload it to the escalator management platform to perform escalator risk warning tasks.

[0005] The present invention constructs an abnormal operation detection model by analyzing the abnormal operation data and normal operation data of the escalator. This model can accurately distinguish abnormal events in the operation of the escalator, such as people falling or foreign objects stuck in the escalator. Compared with the traditional single sensor, it can provide more refined and accurate alarms. The reliability and accuracy of the monitoring system are improved, the frequency of false alarms is reduced, and at the same time, it can quickly respond to real dangerous situations to ensure the safety of passengers and equipment. By extracting abnormal operation sensor data, the vibration intensity and characteristic spectrum changes of the escalator guide rail are analyzed. These data can reflect the operating status and possible degradation trend of the guide rail. Based on the analysis of vibration intensity and characteristic spectrum data, the degradation trend of the guide rail operation can be discovered early, potential mechanical failures or equipment damage can be prevented, and maintenance costs can be reduced and the service life of the equipment can be increased. According to the vibration intensity and characteristic spectrum change data, the operation degradation trend of the escalator guide rail is defined and analyzed. This trend data can help operation and maintenance personnel understand the health status of the equipment. By identifying and monitoring the degradation trend of the guide rail in advance, more accurate maintenance plans and preventive maintenance measures can be formulated, effectively reducing the downtime caused by equipment failure, and improving the reliability and safety of the escalator. The escalator abnormal operation detection model is used to detect the real-time operation data, and the trend cluster analysis is performed in combination with the rail operation degradation trend data. These analysis results can comprehensively reflect the operation status and potential risks of the escalator. The real-time perception of escalator operation risks is improved, potential problems in operation are discovered in time and measures are taken to reduce the possibility of accidents and ensure the safety of passengers. Risk assessment is performed based on the real-time escalator operation disadvantage trend data, and the assessment results are uploaded to the escalator management platform to perform risk warning tasks. Through a systematic risk assessment and warning mechanism, the operation and maintenance team can take preventive measures before problems occur, optimize equipment operation management, reduce the probability of step collision, and improve service efficiency and passenger experience. In summary, these steps not only demonstrate the application of technology in escalator safety management, but also emphasize the importance of data analysis and model building in improving equipment safety, reducing failure risks, and optimizing operation and maintenance efficiency. Through real-time monitoring, trend analysis and predictive maintenance, the reliability and safety of escalators can be effectively guaranteed, and the travel experience of passengers can be improved. At the same time, it can reduce the false alarm caused by cameras or other visual sensors being covered by foreign objects.

[0006] Optionally, step S1 specifically includes: Step S11: acquiring escalator operation data, and classifying the escalator operation data by escalator type, thereby obtaining moving walkway operation data and escalator operation data; Step S12: Analyze the abnormal operation of the horizontal escalator on the moving walk operation data, so as to obtain the abnormal operation data of the moving walk; Step S13: performing abnormal operation analysis of the inclined escalator on the escalator operation data, thereby obtaining abnormal operation data of the escalator; Step S14: Merging the abnormal operation data of the moving walk and the abnormal operation data of the escalator to obtain the abnormal operation data of the escalator, and filtering the normal operation data of the escalator operation data according to the abnormal operation data of the escalator to obtain the normal operation data of the escalator; Step S15: constructing an escalator abnormal operation detection model based on the escalator abnormal operation data and the escalator normal operation data.

[0007] The present invention can perform personalized analysis and processing on different types of escalators by acquiring the operation data of the escalator and dividing it into two categories: moving walkways and escalators. Horizontal and automatic escalators are different in design and operation, so separate processing helps to find and solve problems in a targeted manner. Anomaly analysis of the operation data of moving walkways can help detect and identify possible faults or abnormal conditions of horizontal escalators during operation. The analysis of these abnormal data can provide early warnings, which helps to timely repair and maintain to improve operating efficiency and safety. Similarly, the abnormal operation analysis of escalators is specifically targeted at the specific design and operation mode of slope escalators. This analysis helps to find possible operational abnormalities of slope escalators, such as problems in speed, inclination, etc., so that repair measures can be taken in time. Merging the abnormal operation data of moving walkways and escalators, and filtering out the normal operation data of escalators by comparing the normal operation data, can help identify and distinguish the difference between real abnormal conditions and normal operation, thereby more accurately determining the health status of the escalator. Based on the data analysis results of the previous steps, an escalator abnormal operation detection model is constructed. This model can use machine learning or other algorithmic technologies to monitor and analyze the escalator operation data collected in the future in real time to predict and prevent possible operational anomalies, further improving the reliability and safety of the escalator. In summary, these steps combine to form a complete escalator operation data processing and analysis process, thereby effectively managing and maintaining escalator equipment, improving equipment operation efficiency and user safety experience.

[0008] Optionally, step S12 is specifically: Step S121: extracting running speed features from the running data of the moving walkway, thereby obtaining running speed data of the horizontal escalator; Step S122: performing running speed reference calculation on the running speed data of the escalator, thereby obtaining the running reference speed of the escalator; Step S123: Calculating the operation cycle of the escalator operation speed data according to the escalator operation reference speed, thereby obtaining the escalator operation cycle data; Step S124: performing low-frequency operation cycle statistics on the horizontal escalator operation cycle data, thereby obtaining low-frequency operation cycle data; Step S125: clustering the low-frequency operation cycle operation data of the moving walkway according to the low-frequency operation cycle data, so as to obtain abnormal operation data of the moving walkway.

[0009] The present invention extracts the characteristics of the running speed from the running data of the moving walkway. These characteristics may include real-time speed, average speed, acceleration and deceleration, etc. These data can help identify changes and abnormal conditions in the operation process of the escalator, such as sudden changes in speed or speeds beyond the normal range. Calculate the running speed benchmark of the horizontal escalator. This benchmark can be a normal operating speed range determined according to historical data or standards specified by the manufacturer. By comparing with the benchmark, it can be identified whether the running speed is normal, thereby helping to judge the health of the escalator. According to the previously calculated benchmark speed, the running cycle of the escalator is calculated. The running cycle includes the time interval from one point to another, which can be used to detect whether the operation of the escalator is stable and whether the cycle is normal. Perform low-frequency statistics on the running cycle data of the escalator. Low-frequency running cycles generally refer to changes in the running cycle over a long period of time, such as hourly or daily changes. By counting these data, long-term running trends and periodic anomalies, such as seasonal changes or gradually increasing running cycles, can be found. Cluster analysis is performed using low-frequency running cycle data. This analysis can help identify different running modes or categories, such as normal running modes, occasional abnormal modes, or continuous abnormal modes. Through clustering, the abnormal operation of the escalator can be identified more accurately so that maintenance or adjustment measures can be taken in time.

[0010] Optionally, step S13 is specifically: Step S131: extracting the running vibration characteristics of the escalator running data, thereby obtaining the running vibration data of the inclined escalator; Step S132: performing operation amplitude calculation on the operation vibration data of the inclined escalator, thereby obtaining the operation amplitude data of the inclined escalator, and performing high-value amplitude quantity statistics on the operation amplitude data of the inclined escalator, thereby obtaining the high-value operation amplitude data; Step S133: performing amplitude interval calculation on the high-cost operation amplitude data according to the inclined escalator operation amplitude data, thereby obtaining the high-cost operation amplitude frequency data; Step S134: clustering the escalator operation data according to the high-value operation amplitude frequency data, thereby obtaining the escalator abnormal operation data.

[0011] The present invention can obtain detailed data about the running status of the escalator by extracting the running vibration characteristics of the escalator. These data can be used to analyze the running quality of the escalator and identify potential signs of faults or abnormal behaviors. For example, abnormal vibration may indicate that some parts of the escalator are seriously worn or need repair. The running amplitude of the inclined escalator is calculated and the high-value amplitude quantity is counted. The amplitude generally reflects the vibration intensity and stability of the equipment during operation. The high-value amplitude quantity statistics can help identify and quantify the situation where the vibration exceeds the normal range, so as to find potential problems in advance and perform preventive maintenance. By performing interval calculation on the amplitude data, the frequency data of the high-value amplitude can be obtained. The frequency data reflects the frequency and regularity of the occurrence of high-value amplitude events. This information is helpful for further analyzing periodic problems or abnormal phenomena in the operation of the equipment. The high-value amplitude frequency data is processed by cluster analysis, with the purpose of identifying the abnormal operation mode of the escalator. By clustering the data, different types of abnormal behaviors can be discovered and classified, such as damage to specific parts of the equipment, poor operation or other potential faults. This classification can provide guidance for the maintenance team to help them locate and solve problems accurately and quickly.

[0012] Optionally, step S2 specifically includes: Step S21: extracting abnormal operation sensor data from the escalator abnormal operation data, thereby obtaining abnormal operation sensor data; Step S22: performing guide rail vibration sensing feature extraction and photoelectric sensing feature extraction on the abnormal operation sensing data, thereby obtaining abnormal operation guide rail vibration sensing data and abnormal operation photoelectric sensing data; Step S23: Calculating the step displacement according to the abnormal operation photoelectric sensor data, thereby obtaining the abnormal operation step displacement data; Step S24: performing time series correlation on the abnormal operation guide rail vibration sensing data and the abnormal operation step displacement data, thereby obtaining abnormal operation vibration-displacement data; Step S25: performing Monte Carlo escalator vibration simulation according to the abnormal operation vibration-displacement data and the abnormal operation sensor data, thereby obtaining escalator vibration simulation data; Step S26: Performing statistics on the guide rail vibration intensity and characteristic spectrum distribution of the escalator vibration simulation data, thereby obtaining escalator guide rail vibration intensity and characteristic spectrum change data.

[0013] The present invention collects and extracts sensor data when an escalator is in abnormal operation. These data cover various parameters such as speed, acceleration, current, voltage, etc. The purpose of acquiring these data is to more deeply analyze the operating characteristics of the escalator in an abnormal state. The frequency, amplitude, periodicity and other characteristics of the vibration of the escalator guide rail are analyzed to evaluate the problems or damage that may occur in the guide rail during abnormal operation. The data recorded by the photoelectric sensor are analyzed. These data are usually used to detect whether any object (such as a person or an object) is caught in the middle of the escalator during the operation of the escalator, so as to evaluate the safety and whether the operation is normal. The operation displacement of each step (step board) of the escalator is calculated according to the photoelectric sensor data, which is very important for detecting and analyzing the movement of the steps during abnormal operation, such as whether there is step offset, unstable movement and other problems. The guide rail vibration sensor data and step displacement data extracted during abnormal operation are subjected to time series correlation analysis. This correlation analysis can reveal the relationship between vibration and displacement, and help identify the cause and consequence, such as whether the vibration causes step displacement or other problems. The Monte Carlo simulation using abnormal operation vibration-displacement data and other sensor data is to more comprehensively evaluate the vibration that may occur in the escalator in an abnormal state. This simulation can help predict the intensity and distribution of vibrations and the impact on equipment and passengers. Statistical analysis is performed on the Monte Carlo simulation data, especially the distribution of the vibration intensity and characteristic spectrum of the escalator rails. This analysis can provide key engineering data to help engineers and technicians understand the specific changes and risks of escalators under abnormal vibration conditions. In summary, these steps combine sensor data collection, feature extraction, computational analysis, and simulation to comprehensively evaluate the problems that may occur during abnormal operation of escalators, thereby helping to take appropriate maintenance and safety measures to ensure the safe operation of escalators and the safety of passengers.

[0014] Optionally, step S3 specifically includes: Step S31: Smoothing the displacement change trend of the escalator guide rail vibration intensity and characteristic spectrum change data, thereby obtaining the escalator guide rail vibration intensity and characteristic spectrum trend data; Step S32: acquiring escalator structure data, and extracting guide rail structure features from the escalator structure data, thereby obtaining guide rail structure data; Step S33: performing a risk assessment of the escalator guide rail vibration intensity and characteristic spectrum trend data according to the guide rail structure data, thereby obtaining displacement trend risk assessment data; Step S34: Defining the degradation trend of the escalator guide rail vibration intensity and characteristic spectrum trend data according to the vibration intensity and characteristic spectrum trend risk assessment data, thereby obtaining the escalator guide rail operation degradation trend data.

[0015] The present invention processes the vibration intensity and characteristic spectrum change data of the escalator guide rail, and removes the noise and instantaneous fluctuations in the data through smoothing techniques (such as sliding average, exponential smoothing, etc.), so as to obtain more stable escalator guide rail vibration intensity and characteristic spectrum trend data. This smoothing process helps to reduce the randomness of the data, making the subsequent analysis more reliable and stable. The structural data of the escalator, especially the structural data of the guide rail, is obtained. The structural characteristics of the guide rail include information such as material, size, and installation method. By extracting and analyzing these data, the system can establish a structural model of the guide rail, which is crucial for subsequent vibration intensity and characteristic spectrum trend analysis and risk assessment. According to the structural data of the guide rail, the vibration intensity and characteristic spectrum trend data of the escalator guide rail are risk assessed. This means that the system will analyze the abnormal or dangerous signals that may exist in the displacement trend data, such as vibration or displacement trends that exceed the normal operating range. Through risk assessment, possible structural problems or wear of the guide rail can be identified in advance, and preventive maintenance measures can be taken. According to the vibration intensity and characteristic spectrum trend risk assessment data, the operation degradation trend of the escalator guide rail is defined. This means analyzing and evaluating data to identify trends in the structure and operating status of the guide rails in order to provide early warning of possible structural degradation or failure. This definition can help the operation and maintenance team develop effective maintenance plans and overhaul strategies based on historical data and preset operating standards, thereby extending equipment life and improving safety.

[0016] Optionally, step S33 is specifically: Classify the displacement direction according to the vibration intensity of the escalator guide rail and the characteristic spectrum trend data, so as to obtain the left displacement trend data and the right displacement trend data; Calculate the maximum displacement of the guide rail according to the guide rail structure data, so as to obtain the guide rail peak displacement data; The guide rail peak left displacement time is calculated for the left displacement trend data according to the guide rail peak displacement data, thereby obtaining the guide rail peak left displacement time data; Calculate the guide rail peak right displacement time for the right displacement trend data according to the guide rail peak displacement data, so as to obtain the guide rail peak right displacement time data; Performing time difference calculation on the guide rail peak left displacement time data and the guide rail peak right displacement time data, thereby obtaining peak displacement time difference data; The escalator guide rail vibration intensity and characteristic spectrum trend data are risk assessed based on the peak displacement time difference data. If the peak displacement time difference data is a negative number, the corresponding escalator guide rail vibration intensity and characteristic spectrum trend data are defined as high-risk displacement trend data; if the peak displacement time difference data is a positive number, the corresponding escalator guide rail vibration intensity and characteristic spectrum trend data are defined as low-risk displacement trend data; The high-risk displacement trend data and the low-risk displacement trend data are merged to obtain displacement trend risk assessment data.

[0017] The present invention can separate the left displacement trend data and the right displacement trend data by classifying the vibration intensity and characteristic spectrum trend data of the escalator guide rail. This classification helps to further analyze and evaluate the motion characteristics of the guide rail in different directions, so as to more accurately locate the potential problem area. The peak displacement of the guide rail is calculated according to the guide rail structure data. This data can directly reflect the displacement limit of the guide rail under specific conditions and is an important basis for evaluating the health status of the guide rail structure. For the left displacement trend data and the right displacement trend data, the time point when the corresponding guide rail peak displacement occurs is calculated respectively. These time data help to determine the occurrence time and duration of the displacement event, and provide a reference for the time dimension for subsequent risk assessment. By calculating the peak displacement time difference of the left displacement and the right displacement, the displacement synchronization or asynchronism of the guide rail in different directions can be evaluated. This is crucial for judging whether there is an offset or abnormal displacement behavior in the guide rail structure. According to the peak displacement time difference data, the vibration intensity and characteristic spectrum trend data of the escalator guide rail are divided into two categories: high-risk displacement trend data and low-risk displacement trend data. Negative time differences usually indicate that the left and right displacements are out of sync, which means that the structure is unbalanced or operating abnormally; positive values ​​indicate that the displacement trend is relatively normal. This classification helps to set maintenance priorities and early warning measures at different levels to reduce the risk of structural degradation or failure. The high-risk displacement trend data and the low-risk displacement trend data are combined to obtain complete displacement trend risk assessment data. These data can be used as a basis for the operation and maintenance team to formulate maintenance plans and safety measures to ensure the safety and reliability of escalators.

[0018] Optionally, step S4 is specifically: Step S41: Acquire real-time escalator operation data; Step S42: performing escalator abnormal operation detection on the real-time escalator operation data through the escalator abnormal operation detection model, thereby obtaining escalator abnormal operation detection data, wherein the escalator abnormal operation detection data includes real-time normal operation data and real-time abnormal operation data; Step S43: performing normal operation parameter comparison on the escalator abnormality detection data, thereby obtaining a real-time operation error parameter set; Step S44: clustering the escalator guide rail operation trend of the real-time operation error parameter set according to the escalator guide rail operation disadvantage trend data, thereby obtaining the real-time escalator operation disadvantage trend data.

[0019] The present invention collects real-time escalator operation data. By acquiring data, the operation status of the escalator can be understood in real time, including key parameters such as speed, passenger capacity, and operation time. These data are the basis for the analysis of subsequent steps, ensuring the accuracy and real-time nature of the analysis. An escalator abnormal operation detection model is applied to analyze the operation data collected in real time. The model can detect abnormal conditions that are inconsistent with the normal operation mode according to the preset normal operation mode, such as abnormal speed, abnormal passenger capacity, stagnation, etc. By timely detecting abnormalities, measures can be taken quickly to prevent potential faults from further expanding and ensure the safety and reliability of the escalator. The data detected by the abnormality are compared with the preset normal operation parameters. Through this comparison process, the specific parameters or variables that cause the abnormality can be identified, such as a specific speed range, passenger capacity, etc. This helps further fault diagnosis and repair, and also provides a reference for future operation parameter adjustments. Using the disadvantage trend data of the escalator guide rail operation, trend cluster analysis is performed on the real-time operation error parameter set. This analysis can help identify long-term trends or periodic changes that lead to operation problems. By analyzing these trends, long-term maintenance plans or improvement measures can be developed to improve the overall performance and reliability of the escalator.

[0020] Optionally, step S43 is specifically: Step S431: classifying the abnormal operation data of the escalator abnormality detection data, so as to obtain real-time normal operation data and real-time abnormal operation data; Step S432: extracting operation sensing features from the real-time normal operation data to obtain normal operation sensing data, and extracting operation sensing features from the real-time abnormal operation data to obtain abnormal operation sensing data; Step S433: performing a sensor operation parameter comparison on the normal operation sensor data and the abnormal operation sensor data, so as to obtain an operation error parameter set; Step S434: extracting environmental error data from the operating error parameter set, thereby obtaining operating environment error parameter data; Step S435: performing environmental impact simulation according to the escalator structure data and the operating environment error parameter data, thereby obtaining an environmental impact parameter set; Step S436: Perform environmental impact correction on the operation error parameter set according to the environmental impact parameter set, so as to obtain a real-time operation error parameter set.

[0021] The present invention classifies and processes the data obtained from the escalator abnormal operation detection model, and divides the real-time data into two categories: normal operation data and abnormal operation data. This classification is the basis for subsequent analysis, ensuring that the operating state with possible problems has been distinguished before further processing. Operation sensing features are extracted from the normal operation data and the abnormal operation data. These features may include physical characteristics of the operating state such as speed, passenger capacity, current, vibration, etc. Through feature extraction, the data can be converted into a more informative form, which is helpful for subsequent data analysis and comparison. The extracted normal operation sensing data and abnormal operation sensing data are compared. The goal is to analyze and identify the specific operating parameters or features that cause the abnormality. In this way, potential operating problems such as equipment wear and power supply problems can be located and diagnosed, providing clues for the next step of repair and optimization. Environmental error data is extracted from the operating error parameter set. Environmental errors may include the influence of external environmental factors such as temperature, humidity, and air quality on the operation of the escalator. The extraction and analysis of these data help to understand the actual influence of the external environment on the performance of the escalator. Combined with the structural data of the escalator and the environmental error parameter data, a simulation analysis of the environmental impact is performed. Through simulation, the specific impact of external environmental factors on escalator operation can be quantified, providing a basis and direction for environmental impact correction. Based on the environmental impact parameter set obtained by simulation, the previously extracted operating error parameter set is corrected for environmental impact. This correction helps to more accurately evaluate and predict the operating status of the escalator and ensure the stability and safety of the equipment under various external environmental conditions.

[0022] Optionally, step S44 is specifically: Step S441: extracting the guide rail operation feature from the real-time operation error parameter set, thereby obtaining the guide rail operation error parameter set; Step S442: smoothing the abnormal guide rail operation trend according to the guide rail operation error parameter set, thereby obtaining real-time guide rail operation trend data; Step S443: performing degradation similarity calculation on the escalator guide rail operation disadvantage trend data and the real-time guide rail operation trend data, thereby obtaining degradation similarity data; Step S444: performing degradation trend clustering on the real-time guide rail operation trend data according to the degradation similarity data, thereby obtaining real-time escalator operation disadvantage trend data.

[0023] The present invention analyzes the real-time operation error parameter set and extracts the key features of the guide rail operation from it. These features include the vibration frequency, displacement change, noise level, etc. of the guide rail. Through feature extraction, the operation status of the guide rail can be monitored more accurately, avoiding simple reliance on raw data for judgment. Feature extraction can compress the raw data into fewer but more representative information sets, which is convenient for subsequent processing and analysis. The operation characteristics of the guide rail can be acquired and updated in real time to support timely monitoring and management. According to the guide rail operation error parameter set, the abnormal trend is processed by smoothing technology to obtain the trend data of the real-time guide rail operation. Smoothing helps to detect and identify abnormal behavior or fluctuations in the guide rail operation, and improves the accuracy and stability of abnormality detection. The smoothed trend data can more clearly show the long-term change trend of the guide rail, which helps to predict possible problems or trends. Smoothing can reduce noise and data fluctuations, making subsequent analysis more stable and reliable. By comparing the operation data of different time periods, the degradation degree and change trend of the guide rail operation status can be evaluated. Degradation similarity data can be used as an indicator to help predict the time point when the guide rail may need maintenance or replacement. Calculating degradation similarity helps identify long-term degradation trends in guide rail operation and take measures in advance to avoid serious failures. Clustering can help identify typical patterns or categories under different guide rail operating conditions, which helps analyze possible problems or risks. Clustering results can be used to develop management strategies and maintenance plans for different operating conditions. Through cluster analysis, specific problems or trends that may occur in guide rail operation can be predicted, thereby optimizing maintenance and management strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings: Figure 1 A schematic diagram of the steps of the escalator step anti-collision prediction and warning method of the present invention; Figure 2 Detailed step flow diagram of step S1 in the present invention; Figure 3 Detailed step flow diagram of step S2 in the present invention; The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0025] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0026] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0027] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0028] To achieve this, please refer to Figures 1 to 3 The present invention provides a step anti-collision prediction and warning method for an escalator, the method comprising the following steps: Step S1: acquiring escalator operation data, and performing escalator abnormal operation analysis on the escalator operation data, thereby obtaining escalator abnormal operation data and escalator normal operation data, and constructing an escalator abnormal operation detection model based on the escalator abnormal operation data and the escalator normal operation data; In this embodiment, various historical operating data such as speed, load, current, temperature, etc. are obtained through sensors or monitoring equipment on the escalator. These data are usually recorded and stored in the form of time series. For example, high-precision sensors are used to obtain the vibration frequency and vibration amplitude of the escalator, as well as the operating speed and passenger capacity of the escalator. The data acquisition frequency and storage capacity must be sufficient to support subsequent analysis and model building. Abnormal operating data generally refers to data points or data patterns that are significantly different from the normal operating mode. By analyzing historical data, various abnormal patterns can be identified, such as unusual vibration frequencies, sudden current fluctuations, or abnormal operating speed changes. Using machine learning algorithms such as support vector machines (SVM) or neural networks, an anomaly detection model can be constructed based on these abnormal operating data and normal operating data to automatically identify and classify abnormal events.

[0029] Step S2: extracting abnormal operation sensor data from the escalator abnormal operation data to obtain abnormal operation sensor data, and analyzing the escalator guide rail vibration intensity and characteristic spectrum change according to the abnormal operation sensor data to obtain escalator guide rail vibration intensity and characteristic spectrum change data; In this embodiment, sensor data such as acceleration sensor data and vibration sensor data are extracted from the escalator abnormal operation data. Usually, an acceleration sensor or a vibration sensor is used to monitor the vibration of the guide rail in real time. These data can be used to analyze the health of the escalator guide rail and detect whether the guide rail is loose or damaged. For example, the vibration signal is processed using frequency domain analysis technology, and the spectrum features are extracted to identify abnormal vibration modes, thereby obtaining data describing the vibration intensity and characteristic spectrum of the escalator guide rail. Step S3: defining the guide rail operation degradation trend according to the escalator guide rail vibration intensity and characteristic spectrum change data, thereby obtaining the escalator guide rail operation degradation trend data; In this embodiment, the guide rail operation degradation trend is defined based on the guide rail vibration intensity and characteristic spectrum change data, and time series analysis or trend prediction model can be used to describe the change of the guide rail health status. For example, the exponential smoothing method or ARIMA model is used to perform trend analysis on the vibration intensity and characteristic spectrum data to predict whether the health status of the guide rail shows a trend of gradual deterioration.

[0030] Step S4: acquiring real-time escalator operation data, performing escalator abnormal operation detection on the real-time escalator operation data through an escalator abnormal operation detection model, thereby obtaining escalator abnormal detection data; clustering escalator guide rail operation trends on the natural escalator abnormal detection data according to the escalator guide rail operation disadvantage trend data, thereby obtaining real-time escalator operation disadvantage trend data; In this embodiment, after the escalator operation data is obtained in real time through sensors or monitoring equipment on the escalator, anomaly detection is performed through the previously constructed anomaly detection model, for example, monitoring real-time speed changes or load anomalies. At the same time, combined with the guide rail operation degradation trend data, cluster analysis methods such as K-means are used to classify the detected anomalies to identify specific guide rail health problems. For example, different types of vibration intensity and characteristic spectrum anomalies are grouped to further analyze their impact and urgency.

[0031] Step S5: Perform escalator operation risk assessment based on real-time escalator operation disadvantage trend data to obtain escalator operation risk data, and upload it to the escalator management platform to perform escalator risk warning tasks.

[0032] In this embodiment, when risk assessment is performed based on real-time escalator operation disadvantage trend data, a decision tree or logistic regression model can be used to associate different types of disadvantage trends with the risk levels they may cause. The logistic regression model can be used to establish the relationship between disadvantage trends such as abnormal speed, overload and walkway sliding and the probability of accidents. For example, based on historical data and real-time monitoring results, the model can calculate the probability of accidents under different disadvantage trends, and then evaluate the contribution of each trend to safety risks. For example, high-frequency guide rail vibration intensity and characteristic spectrum indicate that the guide rail is loose or worn or there are foreign objects in the guide rail, which requires timely maintenance, while low-frequency anomalies indicate potential long-term problems. After the evaluation is completed, the escalator operation risk data is uploaded to the escalator management platform, and the corresponding risk warning tasks are performed, such as sending different alarm messages to maintenance personnel or management personnel for different types of risk levels to ensure timely processing and maintenance to ensure the safe operation of the escalator.

[0033] Optionally, step S1 specifically includes: Step S11: acquiring escalator operation data, and classifying the escalator operation data by escalator type, thereby obtaining moving walkway operation data and escalator operation data; In this embodiment, the historical operation data of the escalator can be collected by sensors installed on the escalator, and these data include speed, load, running time, etc. For moving walkways and escalators, their operation modes need to be identified and classified respectively. For example, horizontal and automatic escalators can be distinguished by detecting the inclination angle of the escalator or the structural design of the escalator. Moving walkways are usually parallel to the ground, while automatic escalators are set at an angle. This information can be confirmed by sensor measurement and equipment parameters.

[0034] Step S12: Analyze the abnormal operation of the horizontal escalator on the moving walk operation data, so as to obtain the abnormal operation data of the moving walk; In this embodiment, when performing abnormal analysis on the operating data of the moving walkway, statistical analysis or machine learning models such as decision trees or random forests can be used. For example, the speed change pattern of the escalator or the abnormal motor load can be monitored to identify abnormal operating conditions. By setting a threshold or establishing an abnormal pattern recognition algorithm, the abnormal data of the moving walkway can be accurately classified and analyzed.

[0035] Step S13: performing abnormal operation analysis of the inclined escalator on the escalator operation data, thereby obtaining abnormal operation data of the escalator; In this embodiment, when performing abnormal analysis on the operation data of the escalator, its unique operation mode and structural characteristics need to be considered. For example, the change in the inclination angle of the escalator or the abnormal distance between the escalator steps can be monitored to identify possible abnormal conditions of the escalator. The operation status of the inclined escalator can be monitored in real time through a vibration sensor or a displacement sensor to detect and analyze potential faults or abnormalities in advance.

[0036] Step S14: Merging the abnormal operation data of the moving walk and the abnormal operation data of the escalator to obtain the abnormal operation data of the escalator, and filtering the normal operation data of the escalator operation data according to the abnormal operation data of the escalator to obtain the normal operation data of the escalator; In this embodiment, when the abnormal operation data of the moving walkway and the escalator are combined and analyzed, data integration and data cleaning technology can be used to ensure the consistency and accuracy of the data. For example, data fusion technology such as database connection operation or data integration tools are used to integrate the two types of abnormal data to form a comprehensive escalator abnormal operation data set. Then, the screening conditions are set by the escalator abnormal operation data to extract the normal operation data except the abnormal operation data from the overall operation data, so as to facilitate the subsequent abnormal detection model construction and analysis.

[0037] Step S15: constructing an escalator abnormal operation detection model based on the escalator abnormal operation data and the escalator normal operation data.

[0038] In this embodiment, when constructing an escalator abnormal operation detection model based on the integrated abnormal operation data and normal operation data, supervised learning or semi-supervised learning methods can be used. For example, a support vector machine (SVM) or a neural network model is used to identify and predict different types of escalator abnormalities through training data sets. The model can automatically learn and classify new operation data based on the abnormal characteristics and operation modes analyzed previously, thereby improving the safety and reliability of escalator operation.

[0039] Optionally, step S12 is specifically: Step S121: extracting running speed features from the running data of the moving walkway, thereby obtaining running speed data of the horizontal escalator; In this embodiment, sensor equipment or data acquisition system is used to extract key running speed characteristics from the running process of the moving walkway. These characteristics include the average speed, maximum speed, minimum speed, and speed change trend in each period of time. Through data processing and analysis, a series of data sets that accurately reflect the running speed of the horizontal escalator can be obtained. Acceleration sensors and position sensor devices can be used to monitor the speed changes during the operation of the escalator in real time. By regularly collecting data and filtering, the average speed and speed change rate in each time period are extracted to establish an accurate speed feature data set.

[0040] Step S122: performing running speed reference calculation on the running speed data of the escalator, thereby obtaining the running reference speed of the escalator; In this embodiment, the goal is to determine the reference speed for the normal operation of the horizontal escalator. By analyzing historical data or preset operating conditions, the expected speed level of the escalator under normal operating conditions can be calculated. This reference speed can be used as a reference standard for abnormal detection in subsequent steps. Based on the speed data over a period of time, a statistical method (such as mean or median) is used to calculate the typical operating speed of the horizontal escalator. For example, if the speed of the escalator is different during peak and off-peak periods, the reference speed can be calculated for different time periods.

[0041] Step S123: Calculating the operation cycle of the escalator operation speed data according to the escalator operation reference speed, thereby obtaining the escalator operation cycle data; In this embodiment, the reference speed is used to calculate the operating cycle of the horizontal escalator, that is, the time required for the escalator to complete a complete operating cycle. This cycle data is one of the important bases for evaluating whether the escalator is operating normally. The operating cycle of the escalator in each period of time can be calculated based on the known reference speed. For example, if the reference speed is 1 meter per second, then a typical operating cycle may be 30 seconds, which means that the escalator completes a complete operation every 30 seconds under normal circumstances.

[0042] Step S124: performing low-frequency operation cycle statistics on the horizontal escalator operation cycle data, thereby obtaining low-frequency operation cycle data; In this embodiment, the low-frequency operation cycle data of the horizontal escalator, that is, the long-term change trend of the operation cycle, is analyzed and counted. These data are very important for detecting abnormal conditions or potential failures that may occur in the escalator. The operation cycle data of the escalator can be counted using time series analysis techniques, such as moving average method or trend analysis. For example, the average operation cycle per day, week or month can be used to identify whether there is a trend of gradual increase or decrease, which may indicate that the escalator is about to have a problem.

[0043] Step S125: clustering the low-frequency operation cycle operation data of the moving walkway according to the low-frequency operation cycle data, so as to obtain abnormal operation data of the moving walkway.

[0044] In this embodiment, a cluster analysis method is used to identify abnormal operating conditions based on low-frequency operating cycle data. This method can help engineers or maintenance personnel to promptly discover abnormal operating modes of escalators and take corresponding maintenance or adjustment measures. Clustering algorithms (such as K-means clustering) can be used to group low-frequency operating cycle data into different categories or clusters, each cluster representing a specific operating mode. These cluster labels are then applied to the actual escalator operation data. By mapping each operating cycle data point to the corresponding cluster label, it is possible to analyze whether the current operating state of the escalator belongs to a specific operating mode or abnormal situation. For example, a cluster may indicate a significant increase in the operating cycle, which indicates that the escalator is about to fail or requires maintenance.

[0045] Optionally, step S13 is specifically: Step S131: extracting the running vibration characteristics of the escalator running data, thereby obtaining the running vibration data of the inclined escalator; In this embodiment, vibration characteristics are extracted from the operation data of the escalator. The vibration characteristics may include amplitude, frequency, waveform shape, etc. Signal processing technology, such as fast Fourier transform (FFT), is usually used to analyze the vibration signal to obtain spectrum information. An acceleration sensor or a vibration sensor can be installed on the escalator to collect vibration data during operation. By performing FFT transformation on the collected signal, spectrum information is obtained, thereby analyzing the main frequency components and amplitude of the vibration.

[0046] Step S132: performing operation amplitude calculation on the operation vibration data of the inclined escalator, thereby obtaining the operation amplitude data of the inclined escalator, and performing high-value amplitude quantity statistics on the operation amplitude data of the inclined escalator, thereby obtaining the high-value operation amplitude data; In this embodiment, the running amplitude data of the inclined escalator, especially the data of the high amplitude, is calculated and counted, so as to further analyze the abnormal situation. The vibration signal can be analyzed in the time domain to calculate the amplitude in each time window. Then, according to the preset amplitude threshold, the number or duration of the high amplitude is counted. For example, the situation where the amplitude exceeds a certain threshold (such as a multiple of the standard deviation) is defined as the high amplitude.

[0047] Step S133: performing amplitude interval calculation on the high-cost operation amplitude data according to the inclined escalator operation amplitude data, thereby obtaining the high-cost operation amplitude frequency data; In this embodiment, the time intervals between the amplitudes are calculated based on the running amplitude data of the inclined escalator to determine the frequency information of the high amplitude. The time difference between adjacent high amplitude events can be calculated based on the vibration data. These time intervals can help analyze the persistence and periodicity of vibration events. For example, the frequency of occurrence of high amplitude events (per hour, per day) and their duration can be counted.

[0048] Step S134: clustering the escalator operation data according to the high-value operation amplitude frequency data, thereby obtaining the escalator abnormal operation data.

[0049] In this embodiment, a clustering algorithm is used to classify the operation data of the escalator based on the frequency data of high amplitudes to identify abnormal operation modes. An unsupervised learning clustering algorithm, such as K-means clustering or density clustering, is used to group the amplitude frequency data points into different clusters. These cluster labels are then applied to the actual escalator operation data. By mapping each operation cycle data point to the corresponding cluster label, it is possible to analyze whether the current operation state of the escalator belongs to a specific operation mode or abnormal situation. Each cluster represents a specific abnormal vibration mode, such as frequent high amplitudes or unusual vibration frequency distributions. These clustering results can help operation and maintenance personnel quickly identify and respond to abnormal operation of the escalator.

[0050] Optionally, step S2 specifically includes: Step S21: extracting abnormal operation sensor data from the escalator abnormal operation data, thereby obtaining abnormal operation sensor data; In this embodiment, sensor data of abnormal operation is extracted from the abnormal operation data of the escalator for further analysis and processing. Various sensors (such as vibration sensors, photoelectric sensors, etc.) installed on the escalator can also be used to collect data during abnormal operation. These data may include vibration intensity, position information, the state of the photoelectric sensor, etc. This information is obtained by real-time data acquisition or storage for subsequent analysis.

[0051] Step S22: performing guide rail vibration sensing feature extraction and photoelectric sensing feature extraction on the abnormal operation sensing data, thereby obtaining abnormal operation guide rail vibration sensing data and abnormal operation photoelectric sensing data; In this embodiment, characteristic information of rail vibration and photoelectric sensing is extracted from abnormal operation sensing data to further analyze the abnormal situation. Signal processing is performed on the vibration sensor data to extract information such as the spectral characteristics, amplitude, and waveform shape of the vibration. At the same time, the running displacement of the step is calculated based on the data recorded by the photoelectric sensor to evaluate the displacement change of the step during abnormal operation.

[0052] Step S23: Calculating the step displacement according to the abnormal operation photoelectric sensor data, thereby obtaining the abnormal operation step displacement data; In this embodiment, the displacement of each step during abnormal operation is calculated based on the photoelectric sensor data to obtain detailed operation status data. The displacement change of each step can be calculated based on the data recorded by the photoelectric sensor. This can be achieved by measuring the change of the sensor signal or by deducing through a mathematical model.

[0053] Step S24: performing time series correlation on the abnormal operation guide rail vibration sensing data and the abnormal operation step displacement data, thereby obtaining abnormal operation vibration-displacement data; In this embodiment, the rail vibration sensor data and step displacement data acquired during abnormal operation are time-series correlated to establish a relationship model between vibration and displacement. Time series analysis techniques, such as correlation analysis or autoregressive moving average model (ARMA), can be used to correlate vibration data with displacement data. This helps to understand how vibration changes affect the operating state of the escalator.

[0054] Step S25: performing Monte Carlo escalator vibration simulation according to the abnormal operation vibration-displacement data and the abnormal operation sensor data, thereby obtaining escalator vibration simulation data; In this embodiment, based on the vibration-displacement data and other abnormal operation sensor data, the Monte Carlo method is used to simulate the escalator vibration to predict and evaluate possible vibration conditions and step displacement changes. Monte Carlo simulation technology can be used to analyze different possible vibration modes of the escalator and their effects on step displacement by simulating random changes in vibration data and displacement data multiple times. This method can help predict potential safety risks and formulate corresponding maintenance measures.

[0055] Step S26: Performing statistics on the guide rail vibration intensity and characteristic spectrum distribution of the escalator vibration simulation data, thereby obtaining escalator guide rail vibration intensity and characteristic spectrum change data.

[0056] In this embodiment, the escalator vibration simulation data obtained by Monte Carlo simulation is statistically analyzed to obtain the change distribution data of the guide rail vibration intensity and characteristic spectrum. The probability distribution or frequency distribution of the guide rail vibration intensity and characteristic spectrum in the Monte Carlo simulation results can be statistically analyzed. These data can be used to evaluate the change trend of the guide rail vibration, identify possible abnormal operation modes, and conduct further safety assessments and formulate preventive measures.

[0057] Optionally, step S3 specifically includes: Step S31: Smoothing the displacement change trend of the escalator guide rail vibration intensity and characteristic spectrum change data, thereby obtaining the escalator guide rail vibration intensity and characteristic spectrum trend data; In this embodiment, it is necessary to process the escalator guide rail vibration intensity and characteristic spectrum change data to make its trend smoother for subsequent analysis and evaluation. Smoothing techniques, such as moving average method or Loess smoothing method, can be used to process the collected escalator guide rail vibration intensity and characteristic spectrum data. These methods can eliminate noise in the data and make the displacement change trend clearer and more stable. For example, for the Loess smoothing method, non-parametric estimation can be performed based on local weighted regression, and the smoothness can be adjusted according to demand to ensure that the core features of the trend are retained.

[0058] Step S32: acquiring escalator structure data, and extracting guide rail structure features from the escalator structure data, thereby obtaining guide rail structure data; In this embodiment, the structural data of the escalator is obtained and the structural features of the guide rail are extracted from it, which is crucial for subsequent risk assessment. The geometric shape and size of the escalator guide rail are obtained using 3D scanning technology or traditional measurement methods. Then, based on this data, key structural features are extracted, such as the shape curvature, connection method, material strength, etc. of the guide rail. These features are crucial for evaluating the safety impact of vibration intensity and characteristic spectrum trends.

[0059] Step S33: performing a risk assessment of the escalator guide rail vibration intensity and characteristic spectrum trend data according to the guide rail structure data, thereby obtaining displacement trend risk assessment data; In this embodiment, a risk assessment of the vibration intensity and characteristic spectrum trend of the escalator guide rail is performed based on the guide rail structural data to determine potential safety risks and operational problems. Risk assessment methods such as risk matrix or quantitative risk analysis can be applied in combination with structural features and smoothed vibration intensity and characteristic spectrum trend data to assess the safety of each displacement trend. This includes assessing whether the structural strength of the guide rail is sufficient to support the observed vibration intensity and characteristic spectrum changes, and whether these changes exceed design or safety standards.

[0060] Step S34: Defining the degradation trend of the escalator guide rail vibration intensity and characteristic spectrum trend data according to the vibration intensity and characteristic spectrum trend risk assessment data, thereby obtaining the escalator guide rail operation degradation trend data.

[0061] In this embodiment, the operation degradation trend of the escalator guide rail is defined according to the vibration intensity and characteristic spectrum trend risk assessment data to guide subsequent maintenance and repair work. The corresponding vibration intensity and characteristic spectrum trend can be defined according to the vibration intensity and characteristic spectrum trend risk assessed in the vibration intensity and characteristic spectrum trend risk assessment data. For example, the escalator guide rail vibration intensity and characteristic spectrum trend data corresponding to the high-risk displacement trend are defined as the escalator guide rail operation degradation trend data.

[0062] Optionally, step S33 is specifically: Classify the displacement direction according to the vibration intensity of the escalator guide rail and the characteristic spectrum trend data, so as to obtain the left displacement trend data and the right displacement trend data; In this embodiment, according to the vibration intensity of the escalator guide rail and the characteristic spectrum trend data, the displacement direction is first classified to obtain the left displacement trend data and the right displacement trend data. This requires the use of sensors or monitoring equipment to obtain the vibration data of the escalator guide rail, and to determine the direction of vibration through data processing and analysis. For example, frequency domain analysis can be used to extract the main frequency components of the vibration signal, and then determine the vibration direction (left and right displacement). By setting a threshold or using a machine learning model, the guide rail vibration data is classified to obtain left displacement trend data and right displacement trend data. The left displacement trend data indicates the vibration trend of the guide rail to the left, and the right displacement trend data indicates the vibration trend of the guide rail to the right. Acceleration sensors can be installed at different positions of the guide rail to continuously monitor the vibration data of the guide rail. By converting the acceleration data into displacement data and applying a signal processing algorithm, the left and right displacement trends of the guide rail can be analyzed in real time.

[0063] Calculate the maximum displacement of the guide rail according to the guide rail structure data, so as to obtain the guide rail peak displacement data; In this embodiment, the maximum displacement of the guide rail during the monitoring period is calculated based on the guide rail structure data to obtain the peak displacement data of the guide rail. The maximum displacement of the guide rail can reflect the maximum vibration amplitude that may occur during the operation of the guide rail and is the basic data for subsequent risk assessment. Combined with the geometric shape and material properties of the guide rail, the theoretical maximum displacement of the guide rail under different loads and operating conditions can be calculated. The actual displacement data obtained by the monitoring device can be used to find the maximum displacement of the guide rail in actual use.

[0064] The guide rail peak left displacement time is calculated for the left displacement trend data according to the guide rail peak displacement data, thereby obtaining the guide rail peak left displacement time data; In this embodiment, the obtained guide rail peak displacement data is used to analyze the left displacement trend data and calculate the time of the guide rail peak left displacement. This time represents the moment when the vibration of the guide rail offset to the left reaches the peak value, which is helpful for the subsequent time difference calculation. According to the monitored left displacement trend data, the peak point of the vibration is found and the timestamp of the moment is recorded. This can be achieved through the timing analysis of the vibration data and the peak detection algorithm.

[0065] Calculate the guide rail peak right displacement time for the right displacement trend data according to the guide rail peak displacement data, so as to obtain the guide rail peak right displacement time data; In this embodiment, the right displacement trend data is analyzed based on the guide rail peak displacement data to calculate the time of the guide rail peak right displacement. This time represents the time point when the vibration of the guide rail deflecting to the right reaches the peak value. The monitored right displacement trend data can be analyzed to find the peak point of the vibration and record the timestamp of the time point. This step can also be implemented using signal processing technology and peak detection algorithm.

[0066] Performing time difference calculation on the guide rail peak left displacement time data and the guide rail peak right displacement time data, thereby obtaining peak displacement time difference data; In this embodiment, the obtained guide rail peak left displacement time data and right displacement time data are subjected to time difference calculation to obtain peak displacement time difference data. This difference reflects the time interval between the left and right displacements of the guide rail vibration peak. The difference between the left displacement time and the right displacement time is calculated to obtain the time interval. This process needs to ensure the accuracy and consistency of the timestamp to avoid errors affecting the accuracy of risk assessment.

[0067] The escalator guide rail vibration intensity and characteristic spectrum trend data are risk assessed based on the peak displacement time difference data. If the peak displacement time difference data is a negative number, the corresponding escalator guide rail vibration intensity and characteristic spectrum trend data are defined as high-risk displacement trend data; if the peak displacement time difference data is a positive number, the corresponding escalator guide rail vibration intensity and characteristic spectrum trend data are defined as low-risk displacement trend data; In this embodiment, the escalator guide rail vibration intensity and characteristic spectrum trend data are risk assessed based on the obtained peak displacement time difference data. If the peak displacement time difference data is negative, it means that the left displacement occurs before the right displacement, and there may be a greater risk; if it is positive, it is the opposite. A threshold value can be set, such as 0 seconds, to treat negative time differences as high-risk displacement trend data and positive time differences as low-risk displacement trend data. This risk assessment can be implemented through programming logic and conditional statements to automatically process large amounts of monitoring data.

[0068] The high-risk displacement trend data and the low-risk displacement trend data are merged to obtain displacement trend risk assessment data.

[0069] In this embodiment, high-risk displacement trend data and low-risk displacement trend data are merged to obtain complete displacement trend risk assessment data. These data can help engineers and maintenance personnel effectively manage and optimize the operation and maintenance plan of escalators. The classified high-risk and low-risk data are saved in a database or file for subsequent analysis and reporting. Visual tools such as charts and reports are used to display the results to help decision makers quickly understand and take corresponding measures.

[0070] Optionally, step S4 is specifically: Step S41: Acquire real-time escalator operation data; In this embodiment, real-time escalator operation data is obtained in real time through sensors or monitoring equipment on the escalator. A sensor network can be used to collect escalator operation data in real time, such as speed, load, temperature and other parameters. Sensors can be installed in key parts of the escalator, such as motors, guide rails and transmission systems to ensure the comprehensiveness and accuracy of the data. Through network connection, the data is instantly transmitted to the monitoring center or data processing unit, providing a real-time basis for subsequent abnormality detection and analysis.

[0071] Step S42: performing escalator abnormal operation detection on the real-time escalator operation data through the escalator abnormal operation detection model, thereby obtaining escalator abnormal operation detection data, wherein the escalator abnormal operation detection data includes real-time normal operation data and real-time abnormal operation data; In this embodiment, the escalator abnormal operation detection model is used to detect the escalator abnormal operation on the real-time escalator operation data, thereby obtaining the escalator abnormal detection data, wherein the escalator abnormal detection data includes real-time normal operation data and real-time abnormal operation data. In this process, a machine learning algorithm, such as supervised learning or unsupervised learning method, can be used to analyze and classify the real-time data. For example, a support vector machine (SVM) or a neural network model is used to process the real-time data from the sensor to detect and distinguish between normal and abnormal operating states. The abnormal detection model will output the abnormal detection data for use in subsequent analysis steps.

[0072] Step S43: performing normal operation parameter comparison on the escalator abnormality detection data, thereby obtaining a real-time operation error parameter set; In this embodiment, the escalator abnormality detection data is compared with normal operation parameters to obtain a real-time operation error parameter set. In this step, the abnormal operation parameters in the abnormality detection data are compared and analyzed with the normal operation parameters using statistical analysis or rule-based methods. For example, the real-time detected abnormal operation data is compared with the normal operation data to identify and quantify the real-time operation errors of the escalator, such as speed deviation or load anomaly. These parameter sets can reflect the current operation state of the escalator and its difference from the standard operation state.

[0073] Step S44: clustering the escalator guide rail operation trend of the real-time operation error parameter set according to the escalator guide rail operation disadvantage trend data, thereby obtaining the real-time escalator operation disadvantage trend data.

[0074] In this embodiment, the escalator guide rail operation trend is clustered for the real-time operation error parameter set according to the escalator guide rail operation disadvantage trend data, thereby obtaining the real-time escalator operation disadvantage trend data. Data mining techniques, such as cluster analysis or time series analysis, can be used to further analyze and classify the real-time operation error parameter set. For example, the escalator guide rail operation disadvantage trend data is combined with the real-time operation error parameter set to identify and analyze the patterns and trends in the guide rail operation, such as the guide rail wear rate or structural deformation trend. These data can help operation and maintenance personnel predict and plan the maintenance needs of the escalator to optimize its operating efficiency and safety.

[0075] Optionally, step S43 is specifically: Step S431: classifying the abnormal operation data of the escalator abnormality detection data, so as to obtain real-time normal operation data and real-time abnormal operation data; In this embodiment, during the escalator anomaly detection process, the data is first classified by anomaly classification to distinguish between real-time normal operation data and real-time abnormal operation data. This involves using advanced machine learning algorithms, such as pattern recognition-based methods or deep learning models, to analyze the collected escalator operation data. For example, a support vector machine (SVM) is used to train a classifier to distinguish normal operation modes from abnormal operation modes. In this way, the abnormal operation of the escalator can be monitored and identified in real time so that timely maintenance and repair measures can be taken.

[0076] Step S432: extracting operation sensing features from the real-time normal operation data to obtain normal operation sensing data, and extracting operation sensing features from the real-time abnormal operation data to obtain abnormal operation sensing data; In this embodiment, once the real-time normal operation data and real-time abnormal operation data are obtained, the next task is to extract the operation sensing features. This includes extracting features of sensor data such as speed, acceleration, current, etc. from the normal operation data. For abnormal operation data, similar feature extraction is also required to facilitate subsequent analysis and comparison. For example, Fourier transform can be used to extract spectral features, or wavelet transform can be used to obtain the time-frequency characteristics of the signal, so as to better describe the operating state of the sensor under abnormal conditions.

[0077] Step S433: performing a sensor operation parameter comparison on the normal operation sensor data and the abnormal operation sensor data, so as to obtain an operation error parameter set; In this embodiment, the sensing operation parameters are compared based on the sensing data. This means that the features extracted from the normal operation data and the abnormal operation data are compared and analyzed in detail. For example, the typical sensor value range or change trend under normal operation can be calculated and the abnormal operation data can be compared with it to determine the specific type and degree of abnormal operation.

[0078] Step S434: extracting environmental error data from the operating error parameter set, thereby obtaining operating environment error parameter data; In this embodiment, after obtaining the operation error parameter set, it is necessary to further extract environmental error data. This includes identifying specific parameters that may be affected by environmental factors from the operation error parameter set. For example, the potential impact of environmental factors such as temperature, humidity, and air pressure on the operation state of the escalator will be considered, and relevant environmental error data will be extracted.

[0079] Step S435: performing environmental impact simulation according to the escalator structure data and the operating environment error parameter data, thereby obtaining an environmental impact parameter set; In this embodiment, the environmental impact simulation is then performed based on the structural data of the escalator and the obtained environmental error parameter data. This involves establishing a mathematical model or a simulation model to concretize the impact of environmental factors. For example, finite element analysis (FEA) can be used to simulate the expansion or contraction of the material of the escalator at different temperatures, or a computational fluid dynamics (CFD) model can be used to simulate the impact of air flow on operation.

[0080] Step S436: Perform environmental impact correction on the operation error parameter set according to the environmental impact parameter set, so as to obtain a real-time operation error parameter set.

[0081] In this embodiment, the environmental impact correction is performed on the operation error parameter set according to the environmental impact parameter set. This means that the environmental error data obtained previously is associated with the operation error parameter set, and the operation error parameter set is adjusted to reflect the actual impact of environmental factors. For example, the calibration coefficient of the sensor is adjusted according to the temperature change to ensure the operation stability and accuracy of the escalator under different environmental conditions.

[0082] Optionally, step S44 is specifically: Step S441: extracting the guide rail operation feature from the real-time operation error parameter set, thereby obtaining the guide rail operation error parameter set; In this embodiment, in the real-time operation error parameter set, the extraction of guide rail operation features is a key step, which aims to identify and quantify the operation status and problems of the guide rail system. This includes extracting features related to the guide rail from the real-time operation error parameter set, such as the horizontality, verticality, surface flatness, etc. of the guide rail. The geometric characteristics of the guide rail can be extracted through algorithm analysis, such as the root mean square error (RMS) of flatness or the quantification of the deviation of verticality. After these features are extracted, the operation error parameter set of the guide rail is used as an output for further analysis and processing.

[0083] Step S442: smoothing the abnormal guide rail operation trend according to the guide rail operation error parameter set, thereby obtaining real-time guide rail operation trend data; In this embodiment, once the guide rail operation error parameter set is obtained, the abnormal guide rail operation trend needs to be processed next. First, the operation mode and trend of the guide rail are analyzed according to the operation changes of the guide rail operation error parameter set. This involves smoothing the guide rail operation error parameter set collected in real time to eliminate data fluctuations caused by emergencies or sensor noise. For example, a Kalman filter or a moving average method can be used to smooth the guide rail operation error data to obtain more stable and reliable real-time guide rail operation trend data. This can effectively improve the accuracy and credibility of subsequent analysis and ensure that the system can respond to guide rail operation abnormalities in a timely manner.

[0084] Step S443: performing degradation similarity calculation on the escalator guide rail operation disadvantage trend data and the real-time guide rail operation trend data, thereby obtaining degradation similarity data; In this embodiment, after obtaining the real-time guide rail operation trend data, degradation similarity calculation needs to be performed. The purpose is to compare the similarity between the historical disadvantage trend data of the escalator guide rail operation and the current real-time guide rail operation trend. Degradation similarity calculation can use statistical methods or machine learning techniques, such as time series analysis or similarity measurement algorithms. For example, the dynamic time warping (DTW) algorithm can be used to calculate the distance between two time series and evaluate the degradation degree of the guide rail operation trend data. This can quantify the difference between the current guide rail operation status and the expected level, providing a basis for further analysis and decision-making.

[0085] Step S444: performing degradation trend clustering on the real-time guide rail operation trend data according to the degradation similarity data, thereby obtaining real-time escalator operation disadvantage trend data.

[0086] In this embodiment, the real-time escalator operation disadvantage trend data is clustered according to the degradation similarity data. By grouping similar real-time guide rail operation trend data, the guide rail operation status with similar disadvantage trends is identified. Cluster analysis can use clustering algorithms, such as k-means algorithm or hierarchical clustering algorithm, to divide the guide rail operation trend data into different groups or categories. For example, according to the features in the degradation similarity data, similar guide rail operation statuses are classified into the same category to facilitate subsequent prediction and maintenance decisions. In this way, real-time monitoring and classification of guide rail operation disadvantage trends can be achieved, providing effective fault diagnosis and prevention measures for the maintenance team.

[0087] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0088] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A step anti-collision prediction and warning method for an escalator, characterized in that: The following steps are involved: Step S1: acquiring escalator operation data, and performing escalator abnormal operation analysis on the escalator operation data, thereby obtaining escalator abnormal operation data and escalator normal operation data, and constructing an escalator abnormal operation detection model based on the escalator abnormal operation data and the escalator normal operation data; Step S2: extracting abnormal operation sensor data from the escalator abnormal operation data to obtain abnormal operation sensor data, and analyzing the escalator guide rail vibration intensity and characteristic spectrum change according to the abnormal operation sensor data to obtain escalator guide rail vibration intensity and characteristic spectrum change data; Step S3: defining the guide rail operation degradation trend according to the escalator guide rail vibration intensity and characteristic spectrum change data, thereby obtaining the escalator guide rail operation degradation trend data; Step S4: acquiring real-time escalator operation data, performing escalator abnormal operation detection on the real-time escalator operation data through an escalator abnormal operation detection model, thereby obtaining escalator abnormal detection data; clustering escalator guide rail operation trends on the natural escalator abnormal detection data according to the escalator guide rail operation disadvantage trend data, thereby obtaining real-time escalator operation disadvantage trend data; Step S5: Perform escalator operation risk assessment based on real-time escalator operation disadvantage trend data to obtain escalator operation risk data, and upload it to the escalator management platform to perform escalator risk warning tasks.

2. The escalator step anti-collision prediction and warning method according to claim 1 is characterized in that: Step S1 is specifically as follows: Step S11: acquiring escalator operation data, and classifying the escalator operation data by escalator type, thereby obtaining moving walkway operation data and escalator operation data; Step S12: Analyze the abnormal operation of the horizontal escalator on the moving walk operation data, so as to obtain the abnormal operation data of the moving walk; Step S13: performing abnormal operation analysis of the inclined escalator on the escalator operation data, thereby obtaining abnormal operation data of the escalator; Step S14: Merging the abnormal operation data of the moving walk and the abnormal operation data of the escalator to obtain the abnormal operation data of the escalator, and filtering the normal operation data of the escalator operation data according to the abnormal operation data of the escalator to obtain the normal operation data of the escalator; Step S15: constructing an escalator abnormal operation detection model based on the escalator abnormal operation data and the escalator normal operation data.

3. The escalator step anti-collision prediction and warning method according to claim 2, characterized in that: Step S12 is specifically as follows: Step S121: extracting running speed features from the running data of the moving walkway, thereby obtaining running speed data of the horizontal escalator; Step S122: performing running speed reference calculation on the running speed data of the escalator, thereby obtaining the running reference speed of the escalator; Step S123: Calculating the operation cycle of the escalator operation speed data according to the escalator operation reference speed, thereby obtaining the escalator operation cycle data; Step S124: performing low-frequency operation cycle statistics on the horizontal escalator operation cycle data, thereby obtaining low-frequency operation cycle data; Step S125: clustering the low-frequency operation cycle operation data of the moving walkway according to the low-frequency operation cycle data, so as to obtain abnormal operation data of the moving walkway.

4. The escalator step anti-collision prediction and warning method according to claim 2, characterized in that: Step S13 is specifically as follows: Step S131: extracting the running vibration characteristics of the escalator running data, thereby obtaining the running vibration data of the inclined escalator; Step S132: performing operation amplitude calculation on the operation vibration data of the inclined escalator, thereby obtaining the operation amplitude data of the inclined escalator, and performing high-value amplitude quantity statistics on the operation amplitude data of the inclined escalator, thereby obtaining the high-value operation amplitude data; Step S133: performing amplitude interval calculation on the high-cost operation amplitude data according to the inclined escalator operation amplitude data, thereby obtaining the high-cost operation amplitude frequency data; Step S134: clustering the escalator operation data according to the high-value operation amplitude frequency data, thereby obtaining the escalator abnormal operation data.

5. The escalator step anti-collision prediction and warning method according to claim 1, characterized in that: Step S2 is specifically as follows: Step S21: extracting abnormal operation sensor data from the escalator abnormal operation data, thereby obtaining abnormal operation sensor data; Step S22: performing guide rail vibration sensing feature extraction and photoelectric sensing feature extraction on the abnormal operation sensing data, thereby obtaining abnormal operation guide rail vibration sensing data and abnormal operation photoelectric sensing data; Step S23: Calculating the step displacement according to the abnormal operation photoelectric sensor data, thereby obtaining the abnormal operation step displacement data; Step S24: performing time series correlation on the abnormal operation guide rail vibration sensing data and the abnormal operation step displacement data, thereby obtaining abnormal operation vibration-displacement data; Step S25: performing Monte Carlo escalator vibration simulation according to the abnormal operation vibration-displacement data and the abnormal operation sensor data, thereby obtaining escalator vibration simulation data; Step S26: Performing statistics on the guide rail vibration intensity and characteristic spectrum distribution of the escalator vibration simulation data, thereby obtaining escalator guide rail vibration intensity and characteristic spectrum change data.

6. The escalator step anti-collision prediction and warning method according to claim 1, characterized in that: Step S3 is specifically as follows: Step S31: Smoothing the displacement change trend of the escalator guide rail vibration intensity and characteristic spectrum change data, thereby obtaining the escalator guide rail vibration intensity and characteristic spectrum trend data; Step S32: acquiring escalator structure data, and extracting guide rail structure features from the escalator structure data, thereby obtaining guide rail structure data; Step S33: performing a risk assessment of the escalator guide rail vibration intensity and characteristic spectrum trend data according to the guide rail structure data, thereby obtaining displacement trend risk assessment data; Step S34: Defining the degradation trend of the escalator guide rail vibration intensity and characteristic spectrum trend data according to the vibration intensity and characteristic spectrum trend risk assessment data, thereby obtaining the escalator guide rail operation degradation trend data.

7. The escalator step anti-collision prediction and warning method according to claim 6, characterized in that: Step S33 is specifically as follows: The displacement direction is classified according to the vibration intensity of the escalator guide rail and the characteristic spectrum trend data, thereby obtaining the left displacement trend data and the right displacement trend data; Calculate the maximum displacement of the guide rail according to the guide rail structure data, so as to obtain the guide rail peak displacement data; The guide rail peak left displacement time is calculated for the left displacement trend data according to the guide rail peak displacement data, thereby obtaining the guide rail peak left displacement time data; Calculate the guide rail peak right displacement time for the right displacement trend data according to the guide rail peak displacement data, so as to obtain the guide rail peak right displacement time data; Performing time difference calculation on the guide rail peak left displacement time data and the guide rail peak right displacement time data, thereby obtaining peak displacement time difference data; Perform risk assessment on the escalator guide rail vibration intensity and characteristic spectrum trend data based on the peak displacement time difference data. If the peak displacement time difference data is negative, the corresponding escalator guide rail vibration intensity and characteristic spectrum trend data are defined as high-risk displacement trend data. If the peak displacement time difference data is a positive number, the corresponding escalator guide rail vibration intensity and characteristic spectrum trend data are defined as low-risk displacement trend data; The high-risk displacement trend data and the low-risk displacement trend data are merged to obtain displacement trend risk assessment data.

8. The escalator step anti-collision prediction and warning method according to claim 1, characterized in that: Step S4 is specifically as follows: Step S41: Acquire real-time escalator operation data; Step S42: performing escalator abnormal operation detection on the real-time escalator operation data through the escalator abnormal operation detection model, thereby obtaining escalator abnormal operation detection data, wherein the escalator abnormal operation detection data includes real-time normal operation data and real-time abnormal operation data; Step S43: performing normal operation parameter comparison on the escalator abnormality detection data, thereby obtaining a real-time operation error parameter set; Step S44: clustering the escalator guide rail operation trend of the real-time operation error parameter set according to the escalator guide rail operation disadvantage trend data, thereby obtaining the real-time escalator operation disadvantage trend data.

9. The escalator step anti-collision prediction and warning method according to claim 8, characterized in that: Step S43 is specifically as follows: Step S431: classifying the abnormal operation data of the escalator abnormality detection data, so as to obtain real-time normal operation data and real-time abnormal operation data; Step S432: extracting operation sensing features from the real-time normal operation data to obtain normal operation sensing data, and extracting operation sensing features from the real-time abnormal operation data to obtain abnormal operation sensing data; Step S433: performing a sensor operation parameter comparison on the normal operation sensor data and the abnormal operation sensor data, so as to obtain an operation error parameter set; Step S434: extracting environmental error data from the operating error parameter set, thereby obtaining operating environment error parameter data; Step S435: performing environmental impact simulation according to the escalator structure data and the operating environment error parameter data, thereby obtaining an environmental impact parameter set; Step S436: Perform environmental impact correction on the operation error parameter set according to the environmental impact parameter set, so as to obtain a real-time operation error parameter set.

10. The escalator step anti-collision prediction and warning method according to claim 8, characterized in that: Step S44 is specifically as follows: Step S441: extracting the guide rail operation feature from the real-time operation error parameter set, thereby obtaining the guide rail operation error parameter set; Step S442: smoothing the abnormal guide rail operation trend according to the guide rail operation error parameter set, thereby obtaining real-time guide rail operation trend data; Step S443: performing degradation similarity calculation on the escalator guide rail operation disadvantage trend data and the real-time guide rail operation trend data, thereby obtaining degradation similarity data; Step S444: performing degradation trend clustering on the real-time guide rail operation trend data according to the degradation similarity data, thereby obtaining real-time escalator operation disadvantage trend data.

Images