Escalator operation environment monitoring and risk assessment method

Abstract

The invention discloses an escalator operation environment monitoring and risk assessment method, and relates to the related field of escalator safety monitoring, and the method comprises the following steps: cooperatively carrying out escalator monitoring data acquisition; real-time receiving and transmission are carried out based on the Internet of Things; performing data association mining on the time sequence environment parameters and the time sequence operation parameters, and outputting environment risk association data; loading to an environment risk assessment model for operation risk assessment, and constructing an integrated assessment system from the aspects of structural stability, durability, safety protection and the like; and escalator operation and maintenance decision making is carried out according to the risk assessment result, multi-dimensional risk prevention and control and response are achieved, performance assessment, risk grade division and early warning are included, and preventive maintenance, environmental adaptability transformation and emergency response mechanisms are established. The technical problems that existing escalator risk assessment is incomplete in monitoring, not real-time, inaccurate in risk identification and insufficient in assessment result reliability are solved, and the technical effect of improving the risk identification accuracy and the assessment result reliability is achieved.

Description

Technical Field

[0001] The present application relates to the field of escalator safety monitoring, and in particular to a method for monitoring and risk assessment of the operating environment of an escalator. Background Art

[0002] The safety and stability of the escalator operating environment are crucial for ensuring passenger safety and maintaining the normal operation of public places. Monitoring and risk assessment of the operating environment are key links in ensuring the safe operation of escalators. Currently, solutions to the monitoring and risk assessment of the escalator operating environment mainly rely on traditional regular inspections and monitoring methods based on simple threshold judgments. These methods manually check the escalator's operating status regularly and set fixed environmental parameter thresholds. When a parameter exceeds the threshold, an alert is triggered. Due to the low monitoring frequency and incomplete data collection, this method is unable to accurately and accurately obtain dynamic changes in the escalator's operating environment and equipment status in real time. This leads to insufficiently timely and accurate identification of potential risks, and the easy omission of hidden risk factors. This in turn makes the risk assessment results unreliable and unable to provide effective support for operation and maintenance decisions. This can lead to escalator failures not being handled promptly, impacting passenger safety and the normal order of public places.

[0003] Among the current related technologies, there are technical problems in the monitoring and risk assessment of the escalator operating environment, such as incomplete and non-real-time monitoring, inaccurate risk identification, and insufficient reliability of assessment results. Summary of the Invention

[0004] This application provides an escalator operating environment monitoring and risk assessment method, which adopts the collaborative collection of time-series environment and operating parameters by an environmental parameter acquisition module and an equipment status monitoring module. The data transmission module transmits these parameters to the data analysis and processing module in real time through the Internet of Things. The data analysis and processing module performs correlation mining on the parameters and filters out environmental risk-related data. The risk assessment and early warning module uses a built-in model to assess risks and outputs operating risk levels and early warning signals. The decision support module formulates real-time operation and maintenance strategies based on the assessment results. These technical means achieve the technical effect of improving the accuracy of risk identification and the reliability of assessment results through comprehensive and real-time monitoring.

[0005] The present application provides an escalator operation environment monitoring and risk assessment method, comprising: a collaborative operation environment parameter acquisition module and an equipment status monitoring module to collect monitoring data of the escalator to obtain time series environment parameters and time series operation parameters; a data transmission module receives and transmits the time series environment parameters and time series operation parameters to a data analysis and processing module in real time based on the Internet of Things; the data analysis and processing module performs data association mining on the time series environment parameters and time series operation parameters, and screens and outputs environmental risk association data; after receiving the environmental risk association data, the risk assessment and early warning module loads the environmental risk association data into a built-in environmental risk assessment model to perform escalator operation risk assessment, and outputs an operation risk assessment result, wherein the operation risk assessment result includes an operation risk level and a risk early warning signal; a decision support module makes escalator operation and maintenance decisions based on the operation risk assessment result, and outputs a real-time operation and maintenance strategy.

[0006] In a possible implementation, the environment parameter acquisition module and the equipment status monitoring module are operated in a coordinated manner to collect monitoring data of the escalator, obtain time-series environment parameters and time-series operation parameters, and perform the following processing: configure the environment parameter acquisition module on the escalator, wherein the environment parameter acquisition module is integrated with a temperature sensor, a humidity sensor, a wind speed sensor, and a dust concentration sensor; configure the equipment status monitoring module on the escalator, wherein the equipment status monitoring module is integrated with a step vibration sensor, a step displacement sensor, a handrail vibration sensor, a handrail displacement sensor, a drive device vibration sensor, and a drive device displacement sensor; while running the environment parameter acquisition module to collect the environment data of the escalator and obtain the time-series environment parameters, synchronously run the equipment status monitoring module to collect the operation data of the escalator and obtain the time-series operation parameters.

[0007] In a possible implementation, the data analysis and processing module performs data association mining on the time series environment parameters and the time series operation parameters, filters and outputs the environmental risk association data, and performs the following processing: aligning the time series environment parameters and the time series operation parameters based on timestamps to obtain a multidimensional time series data set; using a preset segmentation scale to perform sliding segmentation on the multidimensional time series data set to obtain multiple environment-operation two-dimensional number sets; using the Pearson correlation coefficient algorithm, calculating the correlation degree of the multiple environment-operation two-dimensional number sets to filter out M environment-operation two-dimensional number sets with a correlation degree exceeding a preset threshold as M association parameter combinations; performing data association mining on the M association parameter combinations to filter and output the environmental risk association data.

[0008] In a possible implementation, data association mining is performed on the M association parameter combinations, the environmental risk association data is screened and output, and the following processing is performed: based on a gradient boosting decision tree model, feature importance evaluation is performed on the M association parameter combinations to extract M key environmental risk factors; the M key environmental risk factors are aggregated according to the consistency of the risk factors to obtain N key environmental risk factors; based on the N key environmental risk factors, the M association parameter combinations are reversely aggregated to obtain N groups of association parameter combinations; the N key environmental risk factors and the N groups of association parameter combinations are stored in the form of a structured database and output as the environmental risk association data to the risk assessment and early warning module.

[0009] In a possible implementation, the following processing is performed: the time series environmental parameters include an environmental temperature parameter, an environmental humidity parameter, an environmental wind speed parameter, and an environmental dust concentration parameter.

[0010] In a possible implementation, the following processing is also performed: interactively obtaining multiple sample-related parameter sets of multiple key environmental risk factors; performing risk assessment backtracking on the multiple sample-related parameter sets to obtain multiple sample risk levels and multiple sample warning signals; using the multiple sample-related parameter sets, multiple sample risk levels and multiple sample warning signals as training data to perform parameter optimization of the standard environmental risk assessment network to obtain multiple environmental risk assessment sub-models; after using the multiple key environmental risk factors to mark the multiple environmental risk assessment sub-models, completing the construction of the environmental risk assessment model by connecting the multiple environmental risk assessment sub-models in parallel; and loading the environmental risk assessment model into the risk assessment and warning module.

[0011] In a possible implementation, the following processing is performed: the time sequence operation parameters include step vibration data, step displacement data, handrail vibration data, handrail displacement data, drive device vibration data and drive device displacement data.

[0012] In a possible implementation, after receiving the environmental risk association data, the risk assessment and warning module loads the environmental risk association data into a built-in environmental risk assessment model to perform escalator operation risk assessment, outputs an operation risk assessment result, and performs the following processing: directionally activates N environmental risk assessment sub-models in the environmental risk assessment model according to the N key environmental risk factors; loads the N groups of associated parameter combinations into the N environmental risk assessment sub-models according to the mapping relationship between the N key environmental risk factors and the N groups of associated parameter combinations, synchronously performs escalator operation risk assessment, and outputs N single-dimensional risk levels and N single-dimensional warning signals; performs frequency statistics on the N single-dimensional risk levels and N single-dimensional warning signals to locate the main risk level and the main warning signal; if the main risk level meets the preset warning threshold, the main risk level and the main warning signal are respectively used as the operation risk level and the risk warning signal in the operation risk assessment result.

[0013] The present application proposes a method for monitoring and assessing the operating environment of an escalator. First, the operating environment parameter acquisition module and the equipment status monitoring module are used to collect monitoring data for the escalator to obtain time-series environmental parameters and time-series operating parameters. Then, the data transmission module receives and transmits the time-series environmental parameters and time-series operating parameters to the data analysis and processing module in real time based on the Internet of Things. The data analysis and processing module performs data association mining on the time-series environmental parameters and time-series operating parameters, filters and outputs environmental risk-related data. After receiving the environmental risk-related data, the risk assessment and early warning module loads the environmental risk-related data into a built-in environmental risk assessment model to assess the escalator's operating risk and outputs an operating risk assessment result, wherein the operating risk assessment result includes an operating risk level and a risk early warning signal. Finally, the decision support module makes an escalator operation and maintenance decision based on the operating risk assessment result and outputs a real-time operation and maintenance strategy. The method achieves the technical effect of improving the accuracy of risk identification and the reliability of assessment results through comprehensive and real-time monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the methods according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0015] Figure 1 A flow chart of an escalator operating environment monitoring and risk assessment method provided in an embodiment of the present application.

[0016] Figure 2 A schematic diagram of a process for screening and outputting environmental risk-related data in an escalator operating environment monitoring and risk assessment method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0018] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to “some embodiments” which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict. The terms “including” and “having” and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0020] The embodiment of the present application provides a method for monitoring and risk assessment of the operating environment of an escalator. Figure 1 As shown, the method includes:

[0021] Step S100 : The environment parameter acquisition module and the equipment status monitoring module are operated in coordination to collect monitoring data of the escalator to obtain time sequence environment parameters and time sequence operation parameters.

[0022] Specifically, various sensors, such as temperature, humidity, wind speed, and dust concentration sensors, are installed to collect real-time environmental data such as temperature, humidity, wind speed, and dust concentration in the escalator's operating environment. These sensors output environmental data via analog or digital signals. Time-series environmental parameters refer to environmental parameters that change over time.

[0023] Vibration sensors and displacement sensors are installed to monitor the operating status of escalator components, such as the vibration and displacement changes of steps, handrails, and drive units. These sensors output equipment status data through accelerometers or displacement meters.

[0024] Use a data acquisition card or microcontroller (such as an Arduino or Raspberry Pi) to convert the sensor's analog signal into a digital signal. Use a built-in clock module to timestamp the collected data, creating time series data. Time series operating parameters refer to device operating parameters that change over time.

[0025] For example, a temperature sensor can use a thermistor (NTC), whose resistance changes with temperature. This resistance is converted into a voltage signal through a voltage divider circuit and then converted into a digital signal through an analog-to-digital converter (ADC). A humidity sensor can use a capacitive humidity sensor, whose capacitance changes with humidity. A dedicated humidity sensor module (such as the DHT11) directly outputs a digital signal. A vibration sensor can use an accelerometer (such as the MPU6050), which transmits acceleration data to a microcontroller via an I2C interface.

[0026] In one possible implementation, a collaborative operating environment parameter acquisition module and an equipment status monitoring module are used to collect monitoring data for the escalator to obtain time-series environment parameters and time-series operating parameters. Step S100 further includes step S110, configuring the environment parameter acquisition module on the escalator, wherein the environment parameter acquisition module is integrated with a temperature sensor, a humidity sensor, a wind speed sensor, and a dust concentration sensor. Specifically, a high-precision digital temperature sensor, such as DS18B20, is used and installed at key locations (such as the machine room, entrances and exits, etc.) of the escalator to monitor the ambient temperature in real time. A capacitive humidity sensor, such as DHT22, is used and installed around the escalator to monitor air humidity. A three-cup wind speed sensor is used and installed at the entrances and exits or vents of the escalator to measure wind speed. A dust sensor based on the laser scattering principle, such as PMS5003, is used and installed at the air inlet or key locations of the escalator to monitor dust concentration. For example, a DS18B20 temperature sensor connected to a microcontroller (such as an Arduino) via a single-wire bus interface can measure temperature with 0.5°C accuracy and transmit the data in real time to a data acquisition system. A DHT22 humidity sensor connected to a microcontroller via a GPIO interface can simultaneously measure temperature and humidity and output a digital signal. A wind speed sensor outputs wind speed data via an analog signal, which is converted to a digital signal via an analog-to-digital converter (ADC) and then transmitted to the microcontroller. The PMS5003 dust concentration sensor outputs particulate matter concentration data via a serial communication interface (UART). See Table 1 for detailed information on each sensor.

[0027] Table 1: Environmental parameter sensor selection and installation examples

[0028]

[0029] Step S120: Install the equipment status monitoring module on the escalator. The equipment status monitoring module integrates a step vibration sensor, a step displacement sensor, a handrail vibration sensor, a handrail displacement sensor, a drive unit vibration sensor, and a drive unit displacement sensor. Specifically, a triaxial accelerometer, such as the ADXL345, is installed on the escalator steps to monitor step vibration. A displacement sensor, such as an LVDT (Linear Variable Differential Transformer), is installed below the steps to monitor step displacement changes. A triaxial accelerometer, such as the MPU6050, is installed on the handrail support structure to monitor handrail vibration. A photoelectric displacement sensor is installed on the handrail guide rail to monitor handrail displacement changes. A high-precision vibration sensor, such as a PCBPiezotronics accelerometer, is installed on the drive unit to monitor drive unit vibration. A displacement sensor, such as an LVDT, is installed at key locations on the drive unit to monitor drive unit displacement changes. For example, the ADXL345 step vibration sensor connects to the microcontroller via an I2C interface, enabling real-time monitoring of step acceleration changes in three directions with an accuracy of ±0.06g. The LVDT step displacement sensor outputs displacement data via an analog signal, which is converted to a digital signal via an analog-to-digital converter (ADC) and then transmitted to the microcontroller. The MPU6050 handrail vibration sensor connects to the microcontroller via an I2C interface and simultaneously measures acceleration and angular velocity for monitoring handrail vibration. The handrail displacement sensor uses a photoelectric principle to output displacement data with an accuracy of ±0.1mm. The installation and data transmission methods for the drive unit vibration sensor and displacement sensor are similar. Detailed information for each sensor is provided in Table 2.

[0030] Table 2: Equipment status sensor selection and installation examples

[0031]

[0032] Step S130, while running the environmental parameter acquisition module to collect environmental data of the escalator and obtain the time-series environmental parameters, the device status monitoring module is synchronously run to collect operational data of the escalator and obtain the time-series operational parameters. Specifically, a data acquisition card (such as NIDAQ) or a microcontroller (such as Arduino, Raspberry Pi) is used as a data acquisition terminal to convert the analog signal of the sensor into a digital signal, and a timestamp is added to the collected data to ensure the time sequence of the data. The environmental parameter acquisition module and the device status monitoring module are synchronized by the internal clock or external synchronization signal (such as PPS signal) of the microcontroller to ensure that the collected data are consistent in time. For example, an Arduino microcontroller is used as a data acquisition terminal to collect the analog signal of the sensor through its built-in ADC interface and a timestamp is added to the data using its internal clock. By writing a program, it is ensured that the environmental parameter acquisition module and the device status monitoring module collect data at the same time interval, for example, once per second. The external synchronization signal (such as PPS signal) is further used to ensure that the acquisition time of the two modules is completely consistent.

[0033] In a possible implementation, the time series environmental parameters include an environmental temperature parameter, an environmental humidity parameter, an environmental wind speed parameter, and an environmental dust concentration parameter.

[0034] In a possible implementation, the timing operation parameters include step vibration data, step displacement data, handrail vibration data, handrail displacement data, drive device vibration data and drive device displacement data.

[0035] In step S200, the data transmission module receives and transmits the time sequence environment parameters and the time sequence operation parameters to the data analysis and processing module in real time based on the Internet of Things.

[0036] Specifically, wireless communication technologies such as Wi-Fi, LoRa, and NB-IoT are used to transmit the collected time series data to a cloud server or local server. The data transmission module can use an IoT development board (such as ESP8266, ESP32) or an industrial-grade IoT gateway. During the transmission process, encryption protocols such as TLS / SSL are used to ensure data security and integrity to prevent data tampering or theft. For example, using the ESP32 development board as a data transmission module, connecting to the network via Wi-Fi, the collected time series data is transmitted to a cloud platform (such as the Alibaba Cloud IoT platform) via the MQTT protocol. ESP32 supports multiple communication protocols, including Wi-Fi, Bluetooth, and LoRa. You can choose the appropriate communication method based on actual needs. During the transmission process, the TLS encryption protocol is used to ensure data security.

[0037] In step S300 , the data analysis and processing module performs data association mining on the time series environmental parameters and time series operating parameters to filter and output environmental risk association data.

[0038] Specifically, use big data processing frameworks such as Hadoop and Spark to clean, organize, and store the large amount of collected time series data. Improve data processing efficiency through distributed computing. Utilize association rule mining algorithms (such as the Apriori algorithm) and time series analysis algorithms (such as the ARIMA model) to mine the correlation between environmental parameters and equipment status, and filter out data related to environmental risks. Train machine learning models (such as decision trees, random forests, and neural networks) to identify and filter out feature data that is highly correlated with environmental risks. For example, use the Spark framework to perform distributed processing on the collected time series data, and use Spark SQL to clean and organize the data to remove noise data and outliers. Then, use the Apriori algorithm to mine the association rules between temperature, humidity, and equipment vibration, and filter out the associated data that indicates increased equipment vibration in high humidity and high temperature environments. Finally, by training the random forest model, further filter out feature data that is highly correlated with environmental risks.

[0039] like Figure 2 As shown, in a possible implementation, the data analysis and processing module performs data association mining on the time series environmental parameters and the time series operation parameters, filters and outputs environmental risk association data, and step S300 further includes step S310, aligning the time series environmental parameters and the time series operation parameters based on timestamps to obtain a multidimensional time series data set. Specifically, a time series processing library (such as Pandas) in a data processing software or programming language (such as Python, MATLAB) is used to align the timestamps of the environmental parameters and the equipment status parameters to ensure that the two are consistent in time. The aligned data is integrated into a multidimensional time series data set, where each dimension represents a parameter (such as temperature, humidity, step vibration, etc.). An example of a multidimensional time series data set is shown in Table 3.

[0040] Table 3: Example of a multidimensional time series dataset

[0041]

[0042] Step S320, using a preset segmentation scale to perform sliding segmentation on the multidimensional time series dataset, to obtain multiple environment-run two-dimensional datasets. Specifically, a sliding window function in a programming language (such as the rolling function of Pandas) is used to perform sliding segmentation on the multidimensional time series dataset according to a preset time window (such as 1 minute, 5 minutes, etc.) to obtain multiple sub-datasets. Each sub-dataset is stored as an independent data structure (such as a Pandas DataFrame) for easy subsequent processing. Each environment-run two-dimensional dataset contains all parameter data within the current time window.

[0043] Step S330, using the Pearson correlation coefficient algorithm, calculates the correlation of the multiple environment-operation two-dimensional data sets to screen out M environment-operation two-dimensional data sets whose correlation exceeds a preset threshold as M correlation parameter combinations. Specifically, a correlation coefficient calculation function in a programming language (such as the corr function in Pandas) is used to calculate the Pearson correlation coefficient between each environmental parameter and the device state parameter. Based on a preset correlation coefficient threshold (such as 0.5), the parameter combinations whose correlation coefficient exceeds the threshold are screened out. After screening, M correlation parameter combinations are obtained.

[0044] Step S340: Perform data association mining on the M association parameter combinations to filter and output the environmental risk-related data. Specifically, a data mining algorithm (such as the Apriori algorithm or the FP-Growth algorithm) is used to mine association rules on the association parameter combinations to identify combinations of environmental parameters and equipment status parameters with strong correlations. Based on the mined association rules, data related to environmental risks is filtered out.

[0045] In one possible implementation, data association mining is performed on the M association parameter combinations, and the environmental risk association data is screened and output. Step S340 further includes step S341, and based on the gradient boosting decision tree model, feature importance evaluation is performed on the M association parameter combinations to extract M key environmental risk factors. Specifically, the gradient boosting decision tree (GBDT) model in the machine learning library (such as Python's scikit-learn) is used to perform feature importance evaluation on the M association parameter combinations. The GBDT model evaluates the contribution of each feature to the model prediction through a combination of multiple decision trees. The importance score of each feature is obtained through the feature_importances_ attribute of the model, and the key environmental risk factors are extracted through threshold screening.

[0046] Step S342: Aggregate the M key environmental risk factors based on risk factor consistency to obtain N key environmental risk factors. Specifically, cluster the M key environmental risk factors using a clustering algorithm (e.g., K-Means, DBSCAN), grouping important factors with similar characteristics into one category to obtain N key environmental risk factors. Cluster analysis aggregates similar factors by calculating the similarity between factors (e.g., Euclidean distance, cosine similarity).

[0047] Step S343: Based on the N key environmental risk factors, the M association parameter combinations are reversely aggregated to obtain N groups of association parameter combinations. Specifically, based on the clustering results, the association parameter combinations corresponding to key environmental risk factors belonging to the same category are grouped together to obtain N groups of association parameter combinations. A mapping relationship is established between the key environmental risk factors and the association parameter combinations, and reverse aggregation is performed based on the mapping relationship.

[0048] Step S344, the N kinds of key environmental risk factors and N groups of associated parameter combinations are stored in the form of a structured database, and output to the risk assessment and early warning module as the environmental risk associated data. Specifically, a relational database (such as MySQL, PostgreSQL) or a NoSQL database (such as MongoDB) is used to store the N kinds of key environmental risk factors and N groups of associated parameter combinations as structured data. A data table structure is designed to store the key environmental risk factors and associated parameter combinations respectively, and establish an association relationship between them. In this way, the key environmental risk factors and associated parameter combinations are stored as structured data, which is convenient for subsequent risk assessment and early warning modules to query and process.

[0049] Step S400: After receiving the environmental risk associated data, the risk assessment and warning module performs escalator operation risk assessment by loading the environmental risk associated data into a built-in environmental risk assessment model, and outputs an operation risk assessment result, wherein the operation risk assessment result includes an operation risk level and a risk warning signal.

[0050] Specifically, an environmental risk assessment model is constructed using deep learning models (such as convolutional neural networks (CNNs) and long short-term memory (LSTMs)). These models can process time-series data and learn the complex relationships between environmental parameters and equipment status. Based on the output of the environmental risk assessment model, risks are classified into different levels (such as low, medium, and high), and corresponding thresholds are set. When the risk value exceeds the set threshold, a risk warning signal is triggered. This risk warning signal is sent to maintenance personnel in real time via SMS, email, or app push notifications. For example, an LSTM model is used to model time-series data and learn the dynamic relationship between environmental parameters and equipment status using a training dataset. The model outputs a risk score, which is then categorized into low risk (0-30), medium risk (31-70), and high risk (71-100). When the risk value exceeds 70, the system sends a high-risk warning signal to maintenance personnel via SMS and app push notifications.

[0051] In one possible implementation, step S400 further includes: interactively obtaining multiple sample-associated parameter sets of multiple key environmental risk factors; performing risk assessment backtracking on the multiple sample-associated parameter sets to obtain multiple sample risk levels and multiple sample warning signals; using the multiple sample-associated parameter sets, multiple sample risk levels and multiple sample warning signals as training data to perform parameter optimization of the standard environmental risk assessment network to obtain multiple environmental risk assessment sub-models; after labeling the multiple environmental risk assessment sub-models with the multiple key environmental risk factors, completing the construction of the environmental risk assessment model by connecting the multiple environmental risk assessment sub-models in parallel; and loading the environmental risk assessment model into the risk assessment and warning module.

[0052] Specifically, through interaction with domain experts (such as escalator maintenance engineers and data analysts), we collect sample data on a variety of key environmental risk factors. This sample data includes a combination of environmental parameters and equipment status parameters, and is annotated by experts with risk levels and warning signals. The collected sample data is stored in a structured database for subsequent processing and analysis.

[0053] Risk assessment is performed retrospectively using historical data (such as past escalator operation data and fault records). By analyzing environmental parameters and equipment status parameters in the historical data, the risk level and warning signal for each sample are determined. Based on pre-set risk assessment criteria (such as risk value range), the sample data is divided into different risk levels (such as low risk, medium risk, and high risk). Based on the risk level, a corresponding warning signal is generated (such as no warning for low risk, yellow warning for medium risk, and red warning for high risk).

[0054] Use a deep learning framework (such as TensorFlow or PyTorch) to build a standard environmental risk assessment network. This network can be a multi-layer perceptron (MLP) or a convolutional neural network (CNN) to assess environmental risks. Use sample-related parameter sets, sample risk levels, and sample warning signals as training data to train and adjust the parameters of the standard environmental risk assessment network. Use techniques such as cross-validation to optimize model parameters and improve predictive performance.

[0055] Based on different key environmental risk factors, multiple environmental risk assessment sub-models are generated. Each sub-model focuses on assessing the impact of a specific environmental risk factor. Each environmental risk assessment sub-model is labeled with the key environmental risk factor to clarify the assessment scope of each sub-model. Multiple environmental risk assessment sub-models are connected in parallel to construct a comprehensive environmental risk assessment model. By connecting in parallel, the model can simultaneously assess the impact of multiple key environmental risk factors. The output results of multiple sub-models are integrated using ensemble learning techniques (such as model fusion and voting mechanism) to obtain the final environmental risk assessment results. The constructed environmental risk assessment model is deployed in the risk assessment and early warning module so that it can receive environmental risk-related data in real time and perform risk assessment and early warning. The risk assessment and early warning module generates risk levels and early warning signals in real time based on the output results of the model, and notifies relevant personnel to take appropriate measures.

[0056] In one possible implementation, after receiving the environmental risk-related data, the risk assessment and early warning module loads the environmental risk-related data into a built-in environmental risk assessment model to perform an escalator operation risk assessment and output an operation risk assessment result. Step S400 further includes step S410, which directionally activates N environmental risk assessment sub-models in the environmental risk assessment model according to the N key environmental risk factors. Specifically, based on the input key environmental risk factors, the environmental risk assessment sub-models related to these factors are selectively activated. This can be achieved through the input interface of the model to ensure that only the sub-models related to the current risk factors are activated. A mapping relationship between the key environmental risk factors and the environmental risk assessment sub-models is established to ensure that each factor corresponds to a specific sub-model. For example, assume that the environmental risk assessment model consists of three sub-models, corresponding to the three key environmental risk factors of temperature, humidity and wind speed. When the input data contains temperature and humidity factors, the model will activate the sub-models related to temperature and humidity, and ignore the wind speed sub-model.

[0057] Step S420: Based on the mapping relationship between the N key environmental risk factors and the N groups of associated parameter combinations, the N groups of associated parameter combinations are loaded into the N environmental risk assessment sub-models, and the escalator operation risk assessment is performed simultaneously, outputting N single-dimensional risk levels and N single-dimensional warning signals. Specifically, the parameter combination associated with each key environmental risk factor is loaded into the corresponding sub-model, and the sub-model performs risk assessment based on the input data and outputs a single-dimensional risk level and a warning signal. Using multi-threading or multi-process technology, multiple sub-models are run simultaneously to improve assessment efficiency. For example, for the temperature factor, the parameter combination associated with temperature (such as temperature, step vibration, etc.) is loaded into sub-model 1, and sub-model 1 outputs a single-dimensional risk level and warning signal associated with temperature. At the same time, for the humidity factor, the parameter combination associated with humidity (such as humidity, handrail vibration, etc.) is loaded into sub-model 2, and sub-model 2 outputs a single-dimensional risk level and warning signal associated with humidity. These sub-models can run in parallel to improve the overall assessment speed.

[0058] Step S430 performs frequency statistics on the N single-dimensional risk levels and N single-dimensional warning signals to locate the primary risk level and primary warning signal. Specifically, the frequency of occurrence of each single-dimensional risk level and warning signal is counted, and the level and signal with the highest number of occurrences are determined as the primary risk level and primary warning signal. The single-dimensional risk level with the highest number of occurrences is selected as the primary risk level. If multiple risk levels have the same number of occurrences, the risk level with the highest level is selected. The single-dimensional warning signal with the highest number of occurrences is selected as the primary warning signal. If multiple warning signals have the same number of occurrences, the warning signal with the highest level is selected. For example, assume that three sub-models output the following single-dimensional risk levels and warning signals: Sub-model 1: Risk level is "medium" and warning signal is "yellow"; Sub-model 2: Risk level is "high" and warning signal is "red"; Sub-model 3: Risk level is "medium" and warning signal is "yellow". The risk level "medium" appears twice, and the risk level "high" appears once. The warning signal "yellow" appears twice, and the warning signal "red" appears once. The risk level "Medium" appears the most frequently (2 times), so "Medium" is selected as the primary risk level. The warning signal "Yellow" appears the most frequently (2 times), so "Yellow" is selected as the primary warning signal.

[0059] Step S440, if the main risk level meets the preset warning threshold, the main risk level and the main warning signal are respectively used as the operation risk level and the risk warning signal in the operation risk assessment result. Specifically, the main risk level is compared with the preset warning threshold. If the main risk level is higher than or equal to the threshold, the warning is triggered. The main risk level and the main warning signal are output as the final operation risk assessment result, and the relevant personnel are notified to take measures. For example, assume that the preset warning threshold is a "high" risk level. If the main risk level is "high", the "high" risk level and the corresponding "red" warning signal are output as the operation risk assessment result, and the maintenance personnel are notified to take emergency measures.

[0060] Step S500: The decision support module makes escalator operation and maintenance decisions based on the operation risk assessment results and outputs a real-time operation and maintenance strategy.

[0061] Specifically, based on a rule engine (such as Drools) or an expert system, corresponding operation and maintenance decisions are generated according to the risk assessment results. The rule engine can generate specific operation and maintenance strategies, such as preventive maintenance plans, environmental adaptability modification recommendations, emergency response plans, etc., based on preset rules and conditions, risk levels and warning signals. For example, when the risk assessment result is high risk, the decision support system generates the following operation and maintenance strategies according to the preset rules: immediately stop the escalator operation and conduct an emergency inspection. Arrange maintenance personnel to inspect and repair key components. Based on environmental factors (such as high humidity), it is recommended to perform moisture-proof treatment on the equipment. The embodiment of the present application adopts the collaborative collection of time-series environment and operation parameters by the environmental parameter acquisition module and the equipment status monitoring module. The data transmission module transmits these parameters to the data analysis and processing module in real time through the Internet of Things. The data analysis and processing module performs correlation mining on the parameters and screens out environmental risk-related data. The risk assessment and warning module uses a built-in model to assess the risk and output the operation risk level and warning signal. The decision support module formulates real-time operation and maintenance strategies based on the assessment results. These technical means achieve the technical effect of improving the accuracy of risk identification and the reliability of assessment results through comprehensive and real-time monitoring.

[0062] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for monitoring and risk assessment of escalator operation environment, characterized in that: The method comprises: The collaborative operation environment parameter acquisition module and the equipment status monitoring module collect monitoring data of the escalator to obtain time series environment parameters and time series operation parameters; The data transmission module receives and transmits the time sequence environment parameters and time sequence operation parameters to the data analysis and processing module in real time based on the Internet of Things; The data analysis and processing module performs data association mining on the time series environmental parameters and time series operation parameters to filter and output environmental risk association data; After receiving the environmental risk associated data, the risk assessment and warning module performs an escalator operation risk assessment by loading the environmental risk associated data into a built-in environmental risk assessment model, and outputs an operation risk assessment result, wherein the operation risk assessment result includes an operation risk level and a risk warning signal; The decision support module makes escalator operation and maintenance decisions based on the operation risk assessment results and outputs real-time operation and maintenance strategies.

2. The escalator operation environment monitoring and risk assessment method according to claim 1, characterized in that: The operating environment parameter acquisition module and the equipment status monitoring module cooperate to collect monitoring data of the escalator to obtain time sequence environmental parameters and time sequence operating parameters. The method includes: The environmental parameter acquisition module is configured on the escalator, wherein the environmental parameter acquisition module is integrated with a temperature sensor, a humidity sensor, a wind speed sensor, and a dust concentration sensor; The escalator is equipped with the equipment status monitoring module, wherein the equipment status monitoring module integrates a step vibration sensor, a step displacement sensor, a handrail vibration sensor, a handrail displacement sensor, a drive device vibration sensor, and a drive device displacement sensor; In the process of running the environmental parameter acquisition module to collect environmental data of the escalator and obtain the time sequence environmental parameters, the equipment status monitoring module is synchronously run to collect operation data of the escalator and obtain the time sequence operation parameters.

3. The escalator operation environment monitoring and risk assessment method according to claim 1, characterized in that: The data analysis and processing module performs data association mining on the time series environmental parameters and the time series operation parameters to filter and output environmental risk association data. The method includes: Aligning the timing environment parameters and timing operation parameters based on timestamps to obtain a multidimensional timing data set; Performing sliding segmentation of the multidimensional time series data set using a preset segmentation scale to obtain multiple environment-operation two-dimensional data sets; Calculating correlations among the plurality of environment-operation two-dimensional data sets using a Pearson correlation coefficient algorithm to screen out M environment-operation two-dimensional data sets having correlations exceeding a preset threshold as M correlation parameter combinations; Data association mining is performed on the M association parameter combinations to filter and output the environmental risk association data.

4. The escalator operation environment monitoring and risk assessment method according to claim 3, characterized in that: Performing data association mining on the M association parameter combinations to filter and output the environmental risk association data, the method comprising: Based on the gradient boosting decision tree model, feature importance evaluation is performed on the M association parameter combinations to extract M key environmental risk factors; Aggregating the M key environmental risk factors according to the consistency of the risk factors to obtain N key environmental risk factors; According to the N key environmental risk factors, reversely aggregate the M associated parameter combinations to obtain N groups of associated parameter combinations; The N key environmental risk factors and N groups of associated parameters are combined and stored in a structured database format, and output as the environmental risk associated data to the risk assessment and early warning module.

5. The escalator operation environment monitoring and risk assessment method according to claim 2, characterized in that: The time series environmental parameters include environmental temperature parameters, environmental humidity parameters, environmental wind speed parameters and environmental dust concentration parameters.

6. The escalator operation environment monitoring and risk assessment method according to claim 4, characterized in that: The method further comprises: Interactively obtain multiple sample association parameter sets of multiple key environmental risk factors; Perform risk assessment backtracking on the multiple sample association parameter sets to obtain multiple sample risk levels and multiple sample warning signals; Using the multiple sample association parameter sets, the multiple sample risk levels, and the multiple sample warning signals as training data, performing parameter optimization on a standard environmental risk assessment network to obtain multiple environmental risk assessment sub-models; After labeling the plurality of environmental risk assessment sub-models with the plurality of key environmental risk factors, completing the construction of the environmental risk assessment model by connecting the plurality of environmental risk assessment sub-models in parallel; The environmental risk assessment model is loaded into the risk assessment and early warning module.

7. The escalator operation environment monitoring and risk assessment method according to claim 2, characterized in that: The time sequence operation parameters include step vibration data, step displacement data, handrail vibration data, handrail displacement data, drive device vibration data and drive device displacement data.

8. The method for monitoring and risk assessment of an escalator operating environment according to claim 6, wherein: After receiving the environmental risk associated data, the risk assessment and early warning module loads the environmental risk associated data into a built-in environmental risk assessment model to perform an escalator operation risk assessment and output an operation risk assessment result. The method includes: Directively activating N environmental risk assessment sub-models in the environmental risk assessment model according to the N key environmental risk factors; According to the mapping relationship between the N key environmental risk factors and the N groups of associated parameter combinations, the N groups of associated parameter combinations are loaded into the N environmental risk assessment sub-models, and the escalator operation risk assessment is performed simultaneously to output N single-dimensional risk levels and N single-dimensional warning signals; Perform frequency statistics on the N single-dimensional risk levels and the N single-dimensional warning signals, and locate the main risk level and the main warning signal; If the main risk level meets the preset warning threshold, the main risk level and the main warning signal are used as the operation risk level and the risk warning signal in the operation risk assessment result respectively.

Images