An Adaptive Protection Method and System for Elevator Fall Prevention Based on Dynamic Load

Abstract

This application relates to the field of data processing technology and discloses an adaptive protection method and system for elevator fall prevention based on dynamic load. The method includes: collecting and filtering elevator operation data through sensors; calculating wire rope tension and load distribution deviation; segmenting the data into windows to identify abnormal states; assessing risks based on the anomaly determination results; implementing corresponding control strategies according to the risk level; comparing the control results with the expected effect, and activating mechanical safety devices when the deviation is too large. This application achieves accurate identification of elevator abnormal states, scientific quantification of risk levels, and dynamic adjustment of control strategies, thereby ensuring elevator operation safety while reducing unnecessary service interruptions and improving the accuracy and adaptability of fall prevention protection.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to an adaptive protection method and system for elevator fall prevention based on dynamic load. Background Technology

[0002] As a crucial piece of vertical transportation equipment, the safe operation of elevators is directly related to passenger safety. Traditional elevator safety systems primarily rely on passive protection mechanisms consisting of mechanical safety clamps and speed governors. When the elevator overspeeds or the steel cable breaks, the speed governor triggers the safety clamps to grab the guide rails, preventing the car from falling. With technological advancements, elevator safety systems have gradually incorporated electronic monitoring and active protection technologies, such as electronic speed governors, multiple braking systems, and broken rope protection devices. These devices monitor the elevator's operating status through sensors and trigger emergency braking in abnormal situations, improving the sensitivity and reliability of protection. However, most of these systems rely on a single or limited set of parameters for judgment, such as overspeed detection, current monitoring, or abnormal car position.

[0003] The main deficiency in existing technologies is the lack of comprehensive analysis and adaptive processing capabilities for the dynamic load characteristics of elevators. During elevator operation, parameters such as car load distribution, wire rope tension, and operating speed exhibit complex dynamic changes, making protection mechanisms triggered by a single parameter prone to misjudgment or missed detection. Furthermore, traditional protection systems typically use fixed threshold-triggered braking, failing to adjust response strategies according to different risk levels. This either results in insufficient response to dangerous situations or excessive intervention leading to operational interruptions. In addition, existing systems do not adequately consider abnormal wire rope tension changes caused by uneven load distribution, a significant contributing factor to elevator accidents. During emergency braking, the lack of dynamic adaptability to actual loads makes it difficult to achieve optimal braking performance. Summary of the Invention

[0004] This application provides an adaptive protection method and system for elevator fall prevention based on dynamic load, which can accurately identify abnormal elevator states, scientifically quantify risk levels, and dynamically adjust control strategies, thereby ensuring elevator operation safety while reducing unnecessary service interruptions and improving the accuracy and adaptability of fall prevention protection.

[0005] Firstly, this application provides an adaptive protection method for elevator fall prevention based on dynamic load. The method includes: collecting car acceleration data, wire rope tension data, car corner load data, motor speed data, and car position data using sensors installed in the car; filtering the collected data to obtain the elevator operating status; calculating static and dynamic wire rope tension values ​​based on the elevator operating status, and calculating load distribution deviation values ​​to obtain real-time elevator load characteristic data; performing window segmentation on the real-time elevator load characteristic data, extracting feature parameters, and comparing them with preset thresholds to obtain an elevator abnormal state determination result; classifying the abnormal state type, elevator operating status, and real-time elevator load characteristic data into risk level instructions based on the elevator abnormal state determination result; executing corresponding control operations according to the risk level instructions, including data monitoring, reducing operating speed, smooth deceleration, or emergency braking, to obtain a control execution result; comparing the control execution result with the expected control effect, and activating a mechanical safety device when the deviation exceeds a preset threshold.

[0006] Secondly, this application provides an elevator fall prevention adaptive protection system based on dynamic load, the elevator fall prevention adaptive protection system based on dynamic load includes:

[0007] The data acquisition module is used to collect data on car acceleration, wire rope tension, car corner loads, motor speed, and car position through sensors installed in the car. The acquired data is then filtered to obtain the elevator's operating status.

[0008] The calculation module is used to calculate the static and dynamic wire rope tension values ​​based on the elevator's operating status, and to calculate the load distribution deviation value to obtain real-time load characteristic data of the elevator.

[0009] The segmentation module is used to perform window segmentation processing on the real-time load characteristic data of the elevator, extract feature parameters and compare them with preset thresholds to obtain the elevator abnormal state determination result.

[0010] The classification module is used to classify the abnormal state type, elevator operating status and elevator real-time load characteristic data into risk level instructions based on the elevator abnormal state determination results.

[0011] The execution module is used to perform corresponding control operations according to the risk level instruction. The control operations include data monitoring, reducing the running speed, smooth deceleration, or emergency braking, and obtain the control execution result.

[0012] The comparison module is used to compare the control execution result with the expected control effect, and activate the mechanical safety device when the deviation exceeds a preset threshold.

[0013] Thirdly, a computer device is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the computer device to execute the above-described adaptive protection method for elevator fall prevention based on dynamic load.

[0014] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the aforementioned adaptive protection method for elevator fall prevention based on dynamic load.

[0015] The technical solution provided in this application achieves comprehensive monitoring and accurate judgment of elevator operating status through multi-sensor data acquisition and real-time processing, significantly improving elevator safety. Multi-sensor collaborative operation collects car acceleration data, wire rope tension data, car corner load data, motor speed data, and car position data. Combined with advanced filtering algorithms, this provides high-quality basic data for abnormal state detection, effectively reducing data noise and false positive rates. This method innovatively constructs a dynamic load model for the elevator system. By calculating static and dynamic wire rope tension values ​​and analyzing load distribution deviations, it accurately reflects the real-time load characteristics of the elevator, filling the gap in traditional fall prevention technologies' insufficient consideration of dynamic loads. In the data processing stage, window segmentation technology and multi-feature parameter extraction methods are used to improve the sensitivity and accuracy of abnormal state detection, enabling precise identification of different types of elevator abnormal states. Regarding risk assessment, this method introduces artificial intelligence algorithms to comprehensively analyze abnormal state types, elevator operating status, and load characteristics, achieving scientific quantification of risk levels. In particular, the application of artificial intelligence algorithms allows the system to learn historical data patterns, continuously optimize risk assessment parameters, and improve the accuracy and adaptability of risk identification. Another innovation of this method is its adaptive, graded control strategy based on risk level. From data monitoring and speed reduction to smooth deceleration and emergency braking, a clearly defined protection system is formed, ensuring safety in critical moments while minimizing unnecessary operational interruptions. This method also establishes a comprehensive control execution evaluation mechanism. By comparing and analyzing the control execution results with the expected effects, system parameters are continuously optimized, forming a closed-loop self-learning system that allows the fall protection capability to continuously improve with accumulated operational experience. Particularly noteworthy is the contribution of artificial intelligence algorithms in specific application areas. These algorithms can handle complex and variable load characteristic data, identify potential abnormal patterns, and adaptively adjust braking force to achieve precise braking control based on actual load conditions—a function impossible for traditional fixed-parameter fall protection systems. In summary, this invention overcomes the limitations of traditional elevator fall protection technology, establishing a comprehensive protection system based on dynamic load characteristics, realizing a shift from passive protection to active prevention, and significantly improving elevator operational safety. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1This is a schematic diagram of an embodiment of the adaptive protection method for elevator fall prevention based on dynamic load in this application.

[0018] Figure 2 This is a schematic diagram of one embodiment of the elevator fall prevention adaptive protection system based on dynamic load in this application.

[0019] Figure 3 This is a schematic block diagram of the structure of the computer device in an embodiment of the present invention. Detailed Implementation

[0020] This application provides an adaptive elevator fall protection method and system based on dynamic load. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0021] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the elevator fall prevention adaptive protection method based on dynamic load in this application includes:

[0022] Step S101: Collect car acceleration data, wire rope tension data, car corner load data, motor speed data and car position data by sensors installed in the car, and filter the collected data to obtain the elevator operating status.

[0023] Step S102: Based on the elevator's operating status, calculate the static and dynamic wire rope tension values, and calculate the load distribution deviation value to obtain the elevator's real-time load characteristic data.

[0024] Step S103: Perform window segmentation processing on the real-time load characteristic data of the elevator, extract feature parameters and compare them with preset thresholds to obtain the elevator abnormal state judgment result.

[0025] Step S104: Based on the elevator abnormal state determination results, classify the abnormal state type, elevator operating status and elevator real-time load characteristic data into risk level instructions.

[0026] Step S105: According to the risk level instruction, execute the corresponding control operation, which includes data monitoring, reducing the running speed, smooth deceleration or emergency braking, and obtain the control execution result;

[0027] Step S106: Compare the control execution result with the expected control effect. When the deviation exceeds the preset threshold, activate the mechanical safety device.

[0028] It is understood that the executing entity of this application can be an elevator fall prevention adaptive protection system based on dynamic load, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiment uses a server as an example for illustration.

[0029] Specifically, different types of sensors are installed on the elevator car: triaxial acceleration sensors are installed at the top and bottom of the car, tension sensors are installed at the wire rope connections, load sensors are installed at the four corners of the car floor, speed sensors are installed on the elevator drive system, and position sensors are installed in the hoistway. These sensors collect data at relatively high frequencies; for example, the tension sensors collect data at a frequency of 200Hz to ensure real-time information. The collected raw data is transmitted to the control unit via the industrial fieldbus CAN-Bus, and then processed by a Kalman filter to filter out high-frequency noise, remove outliers, and fill in missing values, thereby obtaining accurate elevator operating status data. Based on the elevator operating status data, static and dynamic wire rope tension values ​​are calculated. The static wire rope tension value is calculated using the principle of gravity balance, considering the car weight, rated load, counterweight weight, gravitational acceleration, and inherent parameters of the wire rope. The dynamic wire rope tension value needs to consider various dynamic factors during operation, including the effects of acceleration, speed, and load distribution. This study introduces dynamic coefficients for acceleration, velocity, and load distribution, obtained through correlation analysis and principal component analysis of historical data. Additionally, load distribution deviation values ​​were calculated, and the load values ​​at the four corners of the car were processed using four-point load variance analysis to reflect the uniformity of load distribution. These data collectively constitute the real-time load characteristic data of the elevator.

[0030] The real-time load characteristic data of the elevator is processed by window segmentation, setting a time window to extract key characteristic parameters: tension fluctuation rate, load dynamic distribution index, acceleration anomaly coefficient, speed deviation coefficient, and system response time. These characteristic parameters reflect different aspects of elevator operation. For example, the tension fluctuation rate reflects the degree of change in wire rope tension, the load dynamic distribution index reflects the change in load distribution, the acceleration anomaly coefficient reflects the degree of deviation between actual and theoretical acceleration, and the speed deviation coefficient reflects the degree of deviation between actual and theoretical speed. These characteristic parameters are compared with preset normal state thresholds to determine whether the elevator operation status is abnormal. Based on the degree of abnormality, the elevator operation status is divided into normal, slightly abnormal, moderately abnormal, and severely abnormal states. At the same time, the abnormality type is identified, such as wire rope abnormality, load distribution abnormality, mechanical jamming abnormality, and drive system abnormality. Based on the elevator abnormality status determination results, a risk level assessment is performed. The abnormality level, the hazard level of the abnormality type, the elevator operating speed, the load, and the duration of the abnormality are scored and weighted according to different weights to obtain the total risk score. Based on the total risk score, the risk levels are divided into Level 1, Level 2, Level 3, and Level 4 risks, each corresponding to different handling strategies. The risk level instructions clearly define the control measures that need to be taken.

[0031] Based on the risk level instructions, corresponding control operations are executed. The control strategy adopts a hierarchical control architecture, including a decision-making layer, a coordination layer, and an execution layer. For Level 1 risk, data monitoring is executed; for Level 2 risk, an early warning operation is executed, reducing the elevator's maximum operating speed to 85% of the rated speed; for Level 3 risk, a deceleration operation is executed, reducing the elevator speed to 50% of the rated speed and using a smooth deceleration curve; for Level 4 risk, an emergency braking operation is executed, cutting off the drive power and activating the safety devices. During the execution of control operations, changes in car speed, position, and acceleration are recorded to obtain control process status data.

[0032] The system compares the control process status data with the control command sequence, extracts the actual and expected operating status data of the elevator car, calculates the difference between the two, and determines whether the control execution meets the requirements. When the deviation exceeds a preset threshold, a mechanical safety device is activated to ensure passenger safety. Simultaneously, the elevator load characteristic calculation parameters and risk assessment parameters are adjusted based on the control execution status, forming a closed-loop self-optimizing system. Taking a high-rise office building elevator in operation as an example, when a passenger suddenly moves inside the elevator car, causing a change in load distribution, the four corner load sensors immediately detect this change. By calculating the load distribution deviation value, the system identifies uneven load at the four corners of the car, and the load dynamic distribution index exceeds a preset threshold, classifying it as a minor abnormal state. The system inputs the abnormal state type (abnormal load distribution), elevator operating status, and load characteristic data into the risk assessment stage, calculating a risk level of level two. Subsequently, the system sends a warning message to the management personnel and automatically reduces the elevator speed to 85% of its original speed for smooth operation. By monitoring the car's operating status in real time, the system ensures that the control execution results meet expectations, and passengers safely reach their destination floor.

[0033] In this embodiment, comprehensive monitoring and accurate judgment of elevator operation status are achieved through multi-sensor data acquisition and real-time processing, significantly improving elevator safety. Multi-sensor collaborative operation collects car acceleration data, wire rope tension data, car corner load data, motor speed data, and car position data. Combined with advanced filtering algorithms, this provides high-quality basic data for abnormal state detection, effectively reducing data noise and false positive rates. This method innovatively constructs a dynamic load model for the elevator system. By calculating static and dynamic wire rope tension values ​​and analyzing load distribution deviations, it accurately reflects the real-time load characteristics of the elevator, filling the gap in traditional fall prevention technologies' insufficient consideration of dynamic loads. In the data processing stage, window segmentation technology and multi-feature parameter extraction methods are used to improve the sensitivity and accuracy of abnormal state detection, enabling precise identification of different types of elevator abnormal states. Regarding risk assessment, this method introduces artificial intelligence algorithms to comprehensively analyze abnormal state types, elevator operation status, and load characteristics, achieving scientific quantification of risk levels. In particular, the application of artificial intelligence algorithms allows the system to learn historical data patterns and continuously optimize risk assessment parameters, improving the accuracy and adaptability of risk identification. Another innovation of this method is its adaptive, graded control strategy based on risk level. From data monitoring and speed reduction to smooth deceleration and emergency braking, a clearly defined protection system is formed, ensuring safety in critical moments while minimizing unnecessary operational interruptions. This method also establishes a comprehensive control execution evaluation mechanism. By comparing and analyzing the control execution results with the expected effects, system parameters are continuously optimized, forming a closed-loop self-learning system that allows the fall protection capability to continuously improve with accumulated operational experience. Particularly noteworthy is the contribution of artificial intelligence algorithms in specific application areas. These algorithms can handle complex and variable load characteristic data, identify potential abnormal patterns, and adaptively adjust braking force to achieve precise braking control based on actual load conditions—a function impossible for traditional fixed-parameter fall protection systems. In summary, this invention overcomes the limitations of traditional elevator fall protection technology, establishing a comprehensive protection system based on dynamic load characteristics, realizing a shift from passive protection to active prevention, and significantly improving elevator operational safety.

[0034] In one specific embodiment, the process of performing step S101 may specifically include the following steps:

[0035] (1) Install three-axis acceleration sensors at the top and bottom of the car to collect acceleration data of the X-axis, Y-axis and Z-axis of the car to obtain raw acceleration data;

[0036] (2) Install a tension sensor at the connection of the steel wire rope in the car and collect the tension change data of the steel wire rope at a frequency of 200Hz to obtain the original tension data;

[0037] (3) Load sensors are installed at the four corners of the car floor to collect load data at the four corners of the car and obtain the original load distribution data;

[0038] (4) Install a speed sensor on the elevator drive system to collect the speed data of the elevator drive motor and obtain the raw speed data;

[0039] (5) Install position sensors in the elevator shaft to collect real-time position data of the car and obtain raw position data;

[0040] (6) The original acceleration data, original tension data, original load distribution data, original rotational speed data and original position data are transmitted to the control unit through the industrial fieldbus CAN-Bus to obtain the control unit input data;

[0041] (7) Input the control unit input data into the Kalman filter to filter out high-frequency noise, obtain preprocessed data, and perform outlier detection and correction on the preprocessed data, remove outliers and fill missing values ​​to obtain the elevator operating status.

[0042] Specifically, triaxial accelerometers are installed at the top and bottom of the elevator car to collect acceleration data along the X-axis (horizontal lateral), Y-axis (horizontal longitudinal), and Z-axis (vertical direction) during elevator operation. These sensors are typically manufactured using MEMS (Micro-Electro-Mechanical Systems) technology, offering high sensitivity and accurately capturing minute acceleration changes during elevator operation. The raw acceleration data collected by the sensors directly reflects the smoothness of elevator operation and the presence of abnormal vibrations or impacts. Dedicated tension sensors are installed at the wire rope connections in the car, continuously monitoring changes in wire rope tension using a high sampling frequency of 200Hz. Tension sensors typically employ a strain gauge design, calculating tension values ​​by measuring the resistance changes caused by the deformation of the elastic body. High-frequency sampling ensures the capture of sudden tension fluctuations, crucial for early detection of wire rope anomalies. Tension data is directly related to core parameters for safe elevator operation and is a key monitoring indicator for fall protection.

[0043] Load sensors are installed at the four corners of the elevator car floor to form a uniformly distributed load monitoring network. These load sensors typically employ piezoelectric or resistance strain gauge designs, accurately measuring pressure changes at various points. The load data from these four corners not only allows for the calculation of the total load weight within the car but, more importantly, enables the analysis of load distribution, determining whether personnel or goods are evenly distributed within the car. This is crucial for preventing dangerous situations caused by unbalanced loads in the elevator. Speed ​​sensors are installed on the elevator drive system to monitor the elevator drive motor's rotational speed in real time. These speed sensors typically use photoelectric encoders or Hall effect sensors, determining the rotational speed by counting the number of pulses per unit time. The rotational speed data corresponds to the actual elevator speed; by comparing the theoretical speed with the actual speed, any abnormalities in the elevator drive system can be detected.

[0044] Position sensors are installed within the elevator shaft to monitor the real-time position of the elevator car. These sensors typically employ magnetic induction or photoelectric coding technology to accurately pinpoint the car's location on a floor and its deviation from the standard stopping position. Position data is crucial for determining the elevator's operational trajectory and serves as the foundation for calculating actual speed and acceleration. All raw data collected by the sensors is transmitted to the central control unit via the industrial fieldbus CAN-Bus (Controller Area Network Bus). CAN-Bus is a highly reliable serial communication protocol with strong anti-interference capabilities and a transmission rate of up to 1Mbps, making it particularly suitable for applications like elevators that require real-time performance and high reliability. The data transmitted via CAN-Bus is integrated into a unified format for control unit input, preparing it for subsequent processing.

[0045] After receiving the input data, the control unit first uses a Kalman filter to remove high-frequency noise. The Kalman filter is a recursive filtering algorithm, particularly suitable for processing dynamic system data containing random errors. This algorithm iterates through prediction and update steps, performing real-time smoothing of the sensor data. The prediction step predicts the next state based on the current system state and control input; the update step combines the predicted and actual measured values, performing a weighted average based on their respective uncertainties to obtain the optimal estimate. This process effectively filters out random noise from the sensor data while preserving its dynamic characteristics.

[0046] After filtering, outlier detection and correction are required. Outlier detection uses the 3σ (standard deviation) principle, considering data exceeding three times the standard deviation of the mean as outliers. For detected outliers, mean replacement or linear interpolation methods are used for correction. Missing data points are filled in based on the trend of previous and subsequent data to ensure data continuity. After this series of processing steps, the raw sensor data is transformed into reliable elevator operating status data, laying the foundation for subsequent dynamic load analysis and anomaly determination.

[0047] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0048] (1) Based on the car weight data and counterweight weight data in the elevator operation state, the static wire rope tension value is calculated using the gravity balance principle. The result is obtained by subtracting the counterweight weight from the sum of the car weight and the rated load, multiplying by the gravitational acceleration, and finally adding the inherent parameters of the wire rope.

[0049] (2) Extract the car acceleration, car speed and car corner load values ​​from the elevator operating status to obtain the dynamic parameter input values;

[0050] (3) Perform correlation analysis and principal component analysis on the dynamic parameter input values. By calculating the correlation coefficient matrix of the influence of each parameter on the tension and extracting the main influencing factors, the dynamic coefficient of acceleration, the dynamic coefficient of velocity and the dynamic coefficient of load distribution are obtained.

[0051] (4) Based on the load values ​​at the four corners of the car, the load distribution deviation value is calculated by using the four-point load variance analysis method. The result is obtained by summing the squares of the differences between each pair of the four corner load values ​​and the squares of the differences between the sums of the diagonal loads, and then taking the square root of the sum of the results.

[0052] (5) Calculate the adaptive correction coefficient based on historical operating data, construct the error sum of squares function between the actual tension and the calculated tension, and solve for the coefficient value that minimizes the error sum of squares function to obtain the current optimal adaptive correction coefficient value.

[0053] (6) Substitute the static wire rope tension value, acceleration dynamic coefficient, velocity dynamic coefficient, load distribution dynamic coefficient, car acceleration, car speed, load distribution deviation value and adaptive correction coefficient into the dynamic load calculation formula, and calculate by adding the static tension value and the sum of the dynamic influence terms, where the dynamic influence terms are the sum of the adaptive correction coefficient and the products of each dynamic coefficient and the corresponding parameter, to obtain the dynamic wire rope tension value.

[0054] (7) Compare and analyze the tension values ​​of dynamic wire rope and static wire rope, and obtain the tension change characteristics by calculating the difference and rate of change between the two.

[0055] (8) Combine the tension change characteristics, load distribution deviation value, dynamic wire rope tension value and static wire rope tension value into real-time load characteristic data of elevator.

[0056] Specifically, based on the weight data of the elevator car and counterweight during elevator operation, the static wire rope tension value is calculated using the principle of gravity balance. The static wire rope tension value refers to the tensile force borne by the wire rope when the elevator is stationary, and it directly relates to the safety of elevator operation. The calculation formula is:

[0057] T static =(M car +W load -M cw )×g+R rope

[0058] Among them, T static M represents the static tension value of the wire rope. car Indicates the weight of the car, W load Indicates the load weight inside the car, M cw R represents the relative weight, g represents the acceleration due to gravity, and R represents the gravitational acceleration. rope This indicates the inherent parameters of the wire rope, including its elastic deformation and inherent tension.

[0059] The car acceleration, car speed, and car corner load values ​​are extracted from elevator operation data as dynamic parameter inputs. These parameters reflect the dynamic changes of the elevator during operation and can more accurately characterize the real-time state of the elevator compared to static data. Correlation analysis and principal component analysis are performed on these dynamic parameter inputs to determine the degree of influence of each parameter on the wire rope tension. Correlation analysis determines the linear correlation between parameters by calculating the Pearson correlation coefficient matrix, while principal component analysis transforms the multidimensional parameters into several main influencing factors through eigenvalue decomposition. Through these analyses, the dynamic acceleration coefficient D is obtained. acc Speed ​​dynamic coefficient D vel and the load distribution dynamic coefficient D dist .

[0060] The load distribution deviation was calculated using a four-point load variance analysis method. This method quantifies the unevenness of load distribution by calculating the differences in load values ​​at the four corners of the car. The calculation formula is as follows:

[0061]

[0062] Where ΔW represents the load distribution deviation, and W1, W2, W3, and W4 represent the load values ​​at the four corners of the car, respectively. This formula comprehensively considers the difference in load between adjacent corners and the difference in the sum of diagonal loads, thus fully reflecting the load distribution.

[0063] To improve calculation accuracy, an adaptive correction coefficient α is introduced. This coefficient is constructed by analyzing historical operating data and comparing the difference between the actual measured tension and the calculated tension, thus establishing an error sum-of-squares function E(α):

[0064]

[0065] Among them, T measured,k T represents the measured tension value at the k-th historical data point. calculated,k (α) represents the tension value calculated using the adaptive coefficient α, and N represents the total number of historical data points. The optimal adaptive correction coefficient is found by using the least squares method to minimize E(α).

[0066] Substituting all the calculation results into the dynamic load calculation formula, we obtain the dynamic wire rope tension value:

[0067] T dynamic =T static +α·(D acc ·a+D vel ·v+D dist ·ΔW)

[0068] Among them, T dynamic This represents the dynamic wire rope tension value, 'a' represents the car acceleration, 'v' represents the car speed, 'α' represents the adaptive correction coefficient, and 'T' represents the dynamic wire rope tension value. static This represents the static tension value of the wire rope. This formula combines static tension with dynamic influencing factors to accurately reflect the actual tension borne by the wire rope during elevator operation.

[0069] Subsequently, the dynamic wire rope tension value and the static wire rope tension value were compared and analyzed, and the difference between the two, ΔT = T, was calculated. dynamic -T static and rate of change The tension variation characteristics were obtained. Finally, the tension variation characteristics, load distribution deviation value, dynamic wire rope tension value, and static wire rope tension value were combined to form real-time load characteristic data of the elevator, providing a basis for subsequent abnormal state judgment.

[0070] Taking an elevator in a high-rise building as an example, the car's self-weight is 1200kg, the counterweight weight is 1500kg, and the rated load is 800kg. When the elevator is carrying a passenger weighing 600kg, the load sensors at the four corners of the car measure 160kg at the front left, 140kg at the front right, 150kg at the rear left, and 150kg at the rear right, respectively. First, calculate the static wire rope tension value and substitute it into the formula to get (1200+600-1500)×9.8+50=2990N (assuming the inherent parameter of the wire rope is 50N).

[0071] Then calculate the load distribution deviation value and substitute it into the four-point load variance analysis formula to obtain... The dynamic coefficients obtained from historical data are respectively the acceleration dynamic coefficient D. acc =150, speed dynamic coefficient D vel =80, Load distribution dynamic coefficient Ddist =10, adaptive correction coefficient α = 0.85. When the elevator is running at a speed of 1.5 m / s, the acceleration is 0.2 m / s². 2 Substituting the values ​​into the dynamic load calculation formula, the dynamic wire rope tension value is obtained as 2990 + 0.85 × (150 × 0.2 + 80 × 1.5 + 10 × 20) = 3148.5 N. Calculating the tension change characteristics, the difference is 3148.5 - 2990 = 158.5 N, and the rate of change is (3148.5 - 2990) / 2990 × 100% = 5.3%. Finally, real-time elevator load characteristic data is generated, including static tension 2990 N, dynamic tension 3148.5 N, load distribution deviation 20.0 N, tension difference 158.5 N, and a rate of change of 5.3%.

[0072] In one specific embodiment, the process of executing step S103 may specifically include the following steps:

[0073] (1) Perform time window segmentation on the real-time load characteristic data of the elevator to obtain a data window sequence;

[0074] (2) Extract the tension fluctuation rate, load dynamic distribution index, acceleration anomaly coefficient, velocity deviation coefficient and system response time characteristic parameters from the data window sequence, and compare each parameter with its corresponding upper and lower limits of the normal state threshold, and record the number of parameters that exceed the threshold range and the degree of exceedance.

[0075] (3) Based on the comprehensive score of the number of parameters exceeding the threshold and the degree of exceeding the threshold, the elevator operating status is divided into normal status, slightly abnormal status, moderately abnormal status and severely abnormal status.

[0076] (4) By identifying the number and degree of parameters exceeding the threshold range, the abnormality type is identified and the elevator abnormality status determination result is obtained. The elevator abnormality status determination result includes one or more of the following: wire rope abnormality, load distribution abnormality, mechanical jamming abnormality and drive system abnormality.

[0077] Specifically, the real-time load characteristic data of the elevator is segmented into time windows using a sliding window technique. This typically involves setting a fixed-length time window (e.g., 2 seconds) and sliding it along the time axis in small increments (e.g., 0.5 seconds) to create a series of overlapping data segments. This processing method captures the temporal characteristics of the data, facilitating subsequent analysis of dynamic changes over a short period. In this way, continuous real-time elevator load characteristic data is converted into multiple data window sequences, each containing complete data information for a short time period. Various feature parameters are extracted from these data window sequences, each derived from different original data. The tension fluctuation rate is obtained by calculating the ratio of the standard deviation to the mean of the dynamic wire rope tension values ​​within the window, reflecting the degree of tension fluctuation over a short period. Specifically, from the time series of dynamic wire rope tension values ​​calculated in the previous steps, all tension data points within the current window are extracted, and the standard deviation and mean of these data are calculated. Dividing these by the standard deviation yields the tension fluctuation rate. This parameter is crucial for identifying wire rope anomalies, as excessive tension fluctuations usually indicate unstable wire rope operation.

[0078] The load dynamic distribution index is obtained by calculating the ratio of the maximum to the mean of the load distribution deviation values ​​within a window. This index is extracted from the time series of load distribution deviation values ​​calculated in previous steps and reflects the dynamic changes in load distribution within the elevator car. Sudden changes in load distribution (such as violent movement of passengers within the car) will cause this index value to increase, making it an important basis for judging abnormal load distribution. The acceleration anomaly coefficient is extracted from acceleration sensor data, calculated by dividing the maximum absolute value of the difference between the actual acceleration and the theoretical acceleration by the acceleration standard deviation. Theoretical acceleration is determined based on the elevator operating mode (start, constant speed, deceleration, or stop) and control commands, while actual acceleration is obtained directly from sensor data. This parameter reflects abnormal changes in acceleration during elevator operation and is a key indicator for identifying mechanical jamming anomalies.

[0079] The speed deviation coefficient is obtained by calculating the ratio of the deviation between the actual speed and the theoretical speed. It is derived from the differential of the position sensor data or directly from the speed sensor, and compared with the target speed issued by the control system. This parameter is particularly important for detecting abnormalities in the drive system, as drive system faults typically cause a significant deviation between the actual and target speeds. The system response time characteristic is obtained by calculating the time interval between the issuance of the control command and the system response, reflecting the responsiveness of the elevator control system. This parameter is calculated by comparing the control command issuance time with the corresponding elevator movement start time; an excessively long response time may indicate a problem with the control system or actuator.

[0080] After extracting these characteristic parameters, each parameter is compared with pre-set upper and lower limits of normal state thresholds. These thresholds are determined based on a large amount of historical operating data and professional expertise, reflecting the reasonable range of variation for each parameter under normal elevator conditions. The comparison process records the number of parameters exceeding the threshold range and the degree of exceeding (i.e., the numerical value exceeding the threshold), which is directly related to the assessment of the severity of the abnormal state.

[0081] Based on the number and severity of parameters exceeding thresholds, a comprehensive scoring algorithm classifies elevator operating states into different levels. The comprehensive scoring typically considers factors such as parameter importance weights, the severity of exceeding thresholds, and the duration of the anomaly. According to the scoring results, elevator operating states are classified into normal, slightly abnormal, moderately abnormal, and severely abnormal states, providing a basis for subsequent risk level assessment and control strategy selection. In addition to state level classification, it is also necessary to identify specific anomaly types. Different combinations of parameters exceeding thresholds correspond to different anomaly types. By analyzing these patterns, specific fault types such as wire rope anomalies, load distribution anomalies, mechanical jamming anomalies, and drive system anomalies can be accurately identified. For example, a significant increase in tension fluctuation rate usually indicates a wire rope anomaly; an abnormal load dynamic distribution index points to a load distribution problem; simultaneous anomalies in acceleration and speed deviation coefficients may indicate mechanical jamming; and an abnormal speed deviation coefficient accompanied by a prolonged system response time may be a manifestation of a drive system anomaly.

[0082] In one specific embodiment, the process of executing step S104 may specifically include the following steps:

[0083] (1) Assign scores to the abnormal status levels in the elevator abnormal status judgment results. Assign 1 point to normal status, 3 points to minor abnormal status, 6 points to moderate abnormal status, and 10 points to severe abnormal status to obtain the abnormal status score.

[0084] (2) The abnormality type in the elevator abnormality judgment result is scored for the degree of danger. The abnormality of load distribution is assigned a score of 2 to 5, the abnormality of mechanical jamming is assigned a score of 4 to 7, the abnormality of drive system is assigned a score of 5 to 8, and the abnormality of wire rope is assigned a score of 7 to 10. The specific score is determined according to the degree of abnormality to obtain the abnormality type score.

[0085] (3) Extract the elevator running speed value from the elevator running status, and assign a score according to the speed level. Assign 2 points for speed not exceeding 1m / s, 4 points for speed greater than 1m / s but not exceeding 2m / s, 6 points for speed greater than 2m / s but not exceeding 3m / s, 8 points for speed greater than 3m / s but not exceeding 4m / s, and 10 points for speed exceeding 4m / s to obtain the speed score value.

[0086] (4) Extract the current load value from the elevator real-time load characteristic data, and score it by dividing the current load value by the rated load value and then multiplying by 10 to obtain the load score value.

[0087] (5) Calculate the duration of the abnormal state, and calculate the duration score using an exponential function to obtain the duration score value;

[0088] (6) Input the abnormal state score, abnormal type score, speed score, load score and duration score into the multi-factor weighted scoring unit, and add them together after multiplying the abnormal state score by 0.3, the abnormal type score by 0.25, the speed score by 0.2, the load score by 0.15 and the duration score by 0.1 to obtain the total risk score;

[0089] (7) Based on the total risk score, the risk level is divided into Level 1 risk, Level 2 risk, Level 3 risk and Level 4 risk, where the total risk score is less than 3 and is Level 1 risk, the total risk score is greater than or equal to 3 and less than 5 and is Level 2 risk, the total risk score is greater than or equal to 5 and less than 7 and is Level 3 risk, and the total risk score is greater than or equal to 7 and is Level 4 risk.

[0090] (8) Assign risk levels to data monitoring strategies, early warning strategies, deceleration strategies and emergency braking strategies respectively, and generate risk level instructions.

[0091] Specifically, the abnormal state levels in the elevator abnormal state judgment results are scored and assigned values: 1 point for normal state, 3 points for minor abnormal state, 6 points for moderate abnormal state, and 10 points for severe abnormal state. This tiered scoring reflects the differences in the degree of danger of different abnormal states, and the score interval design also reflects the non-linear increasing relationship of the severity of abnormal states, enabling a more accurate quantification of the risk of elevator operation. For different types of abnormalities, different danger score ranges are set according to their threat to the safe operation of the elevator. Load distribution abnormalities are assigned 2 to 5 points, mechanical jamming abnormalities are assigned 4 to 7 points, drive system abnormalities are assigned 5 to 8 points, and wire rope abnormalities are assigned 7 to 10 points. The specific score within the same type of abnormality is determined according to the degree of abnormality; for example, a minor uneven load distribution scores close to 2 points, while a severe load imbalance scores close to 5 points. This scoring design reflects the inherent difference in danger between different types of abnormalities, especially by setting the wire rope abnormality as the highest score range, reflecting its critical impact on the overall safety of the elevator.

[0092] Elevator operating speed is another important factor in risk assessment. Scoring is based on speed levels: 2 points for speeds under 1 m / s, 4 points for speeds greater than 1 m / s but not exceeding 2 m / s, 6 points for speeds greater than 2 m / s but not exceeding 3 m / s, 8 points for speeds greater than 3 m / s but not exceeding 4 m / s, and 10 points for speeds exceeding 4 m / s. This tiered scoring reflects the positive correlation between elevator operating speed and risk; the higher the speed, the greater the damage in the event of a malfunction, requiring a higher level of safety assurance.

[0093] The load score is directly related to the ratio of the elevator's current actual load to its rated load. It is calculated by dividing the current load value by the rated load value and then multiplying by 10. This calculation method makes the load score proportional to the actual load, reflecting the additional pressure the elevator system experiences under a larger load and the potentially greater damage that could occur in the event of an accident.

[0094] The duration score for abnormal states is calculated using an exponential function:

[0095] S duration =10×(1-e -λt )

[0096] Among them, S duration The score represents the duration of the abnormal state, where t represents the duration (in seconds) and λ represents the time decay coefficient, typically set to 0.5. This exponential function model reflects the non-linear relationship between the duration of the abnormality and the risk: when the abnormality first appears, the risk increases rapidly over time; as time goes on, the risk growth rate gradually decreases, eventually approaching the maximum value of 10. This aligns with reality, as brief abnormalities may be sensor fluctuations or temporary disturbances, while persistent abnormalities are more likely to represent a real malfunction.

[0097] The total risk score is calculated by weighted summation of the above five scoring factors:

[0098] R total =0.3×S state +0.25×S type +0.2×S speed +0.15×S load +0.1×S duration

[0099] Among them, R total S represents the total risk score. state S represents the abnormal state score. type S represents the anomaly type score. speed S represents the speed score. load S represents the load score value. durationThis represents the duration score. The weighting reflects the different degrees of influence of each factor on the overall risk, with the abnormal state level and abnormal type carrying significant weight, indicating that these two factors are the main basis for risk assessment. Based on the calculated total risk score, the risk level is divided into four levels: Level 1 (low risk) with a total risk score less than 3, Level 2 (low to medium risk) with a total risk score greater than or equal to 3 and less than 5, Level 3 (medium to high risk) with a total risk score greater than or equal to 5 and less than 7, and Level 4 (high risk) with a total risk score greater than or equal to 7. Different risk levels correspond to different protection strategies: Level 1 corresponds to a data monitoring strategy, Level 2 to an early warning strategy, Level 3 to a deceleration strategy, and Level 4 to an emergency braking strategy. This correspondence between risk levels and control strategies ensures that the fall arrest system can take appropriate protective measures according to the actual risk level, avoiding normal operation interruption caused by overprotection and ensuring effective protection under high-risk conditions.

[0100] In one specific embodiment, the process of executing step S105 may specifically include the following steps:

[0101] (1) Based on the risk level instructions, the control strategy is divided into a decision-making layer, a coordination layer and an execution layer. The decision-making layer receives the risk level instructions, the coordination layer transforms the control objectives into a sequence of control instructions, and the execution layer connects the drive system and the braking system to obtain a hierarchical control architecture.

[0102] (2) For instructions with a risk level of Level 1 risk, perform data monitoring operations, continuously monitor the elevator operation status and record abnormal data to the system log to obtain the Level 1 risk response result;

[0103] (3) For instructions with a risk level of Level 2, perform early warning operations, send early warning information to the management system, and reduce the maximum operating speed of the elevator to 85% of the rated speed to obtain the Level 2 risk response result;

[0104] (4) For instructions with a risk level of three, execute a deceleration operation, forcibly reducing the elevator speed to 50% of the rated speed, and using a smooth deceleration curve to control the deceleration to not exceed 0.5 m / s. 2 Simultaneously, the safety braking system is pre-activated, resulting in a three-level risk response.

[0105] (5) For commands with a risk level of four, perform emergency braking operations, cut off the drive power supply, activate mechanical safety devices and electronic braking systems, and control the braking deceleration to 0.8-1.2 m / s². 2 Within the scope, a Level IV risk response result was obtained;

[0106] (6) During the emergency braking operation, the braking force is adjusted according to the dynamic load. The magnitude of the braking force is related to the car weight, the current load weight and the car speed, so as to obtain the adaptive braking force control effect.

[0107] (7) Real-time status monitoring of the results of Level 1 risk response, Level 2 risk response, Level 3 risk response and Level 4 risk response, recording changes in car speed, position and acceleration, and obtaining control process status data;

[0108] (8) Compare the control process status data with the control command sequence, calculate the control execution status, and obtain the control execution result.

[0109] Specifically, the control system adopts a hierarchical control architecture, dividing the control strategy into three layers: the decision-making layer, the coordination layer, and the execution layer. The decision-making layer directly receives risk level instructions from the risk assessment module, is responsible for analyzing the risk level, and determining the corresponding control objectives. The coordination layer translates the control objectives into a specific sequence of control instructions, prioritizing, timing, and allocating resources for these instructions. The execution layer is directly connected to the elevator hardware system, responsible for translating control instructions into actual electrical signals and mechanical actions, controlling the drive and braking systems to perform corresponding operations. This hierarchical architecture ensures clear division of responsibilities and efficient collaborative operation within the control system. For situations with a risk level of Level 1 (low risk), the control system performs data monitoring. Specifically, it allows the elevator to continue operating normally, but simultaneously activates an enhanced monitoring mode, increasing the sampling frequency of key parameters from 200Hz to 500Hz, expanding the range of monitored parameters to include all monitored data such as wire rope tension, car acceleration, speed, and corner loads. Simultaneously, parameter data exceeding the normal range, along with relevant timestamps and location information, are recorded in the system log, providing a basis for subsequent analysis and maintenance. This "monitoring without controlling" strategy is suitable for low-risk situations and avoids unnecessary operational disruptions.

[0110] When the risk level rises to Level 2 (low to medium risk), the control system needs to execute an early warning operation. First, it sends a standard-format warning message to the elevator management system via the industrial communication network, including key information such as the type of anomaly, the level of anomaly, the current location, and the time, triggering an alarm from the management system. Simultaneously, the control system adjusts the speed control parameters, limiting the elevator's maximum permissible operating speed to 85% of its rated speed. This speed limit is achieved by modifying the speed limit parameter in the drive system, ensuring that even if the elevator receives a full-speed operation command during subsequent operation, it will automatically limit itself to this safe speed range, allowing more reaction time to deal with potential anomalies.

[0111] When the risk level escalates to Level 3 (medium-high risk), the control system will execute a deceleration operation. Unlike the warning for Level 2 risk, deceleration at Level 3 risk is mandatory, immediately initiating the deceleration procedure regardless of the elevator's current operating state. The deceleration process employs a smooth deceleration curve, i.e., an S-shaped curve with a gradually decreasing rate of speed change, controlling the deceleration to not exceed 0.5 m / s². 2 This prevents secondary accidents such as passenger falls caused by sudden deceleration. Simultaneously, the control system pre-activates the safety braking system, switching it from standby to ready state, reducing braking system activation delay to address potential further deterioration.

[0112] When the risk level reaches Level 4 (high risk), the control system executes the most stringent emergency braking operation. First, the drive power is cut off, the inverter output is disconnected, and power supply to the drive motor is stopped. Simultaneously, mechanical safety devices and the electronic braking system, including the electromagnetic brake and mechanical safety clamp, are activated. To balance passenger safety and comfort, the control system precisely controls the braking deceleration, limiting it to 0.8-1.2 m / s². 2 Within a certain range, it is sufficient to quickly stop the elevator without causing excessive impact, thus avoiding secondary damage.

[0113] During emergency braking, the braking force is not fixed but adjusted in real time based on dynamic load conditions. The braking force calculation considers three main factors: the car's own weight, the current load weight, and the car's speed. The total mass is calculated by extracting the car's current load information from the elevator's real-time load characteristic data and combining it with the car's own weight data. Simultaneously, the car's current speed is extracted from the operating status data. Based on these parameters, the required braking force is dynamically calculated, and the brake output torque is controlled via a PWM (Pulse Width Modulation) signal to achieve adaptive adjustment of the braking force.

[0114] Throughout the control process, the control system monitors all risk response results in real time, recording changes in car speed, position, and acceleration using various sensors on the car. This data is acquired and stored at a high frequency (e.g., 1000Hz) to form complete control process status data. Subsequently, the control system compares the actual execution results with the expected control commands, calculates the deviation between the two, evaluates the accuracy and timeliness of control execution, and finally generates a control execution result report.

[0115] Taking an elevator in a high-rise building as an example, when the elevator is fully loaded (rated load 800kg) and ascends to the 15th floor, the real-time load characteristic data shows an abnormally high fluctuation rate in the wire rope tension, exceeding the preset threshold. After risk assessment, the system determines the current risk level to be level three. Upon receiving the level three risk instruction, the control system decision layer determines the control objective as "deceleration." The coordination layer then generates a series of control instructions, including speed limit instructions, deceleration curve parameters, and braking system pre-activation instructions. The execution layer receives these instructions and translates them into actual operations. The elevator, originally running at its rated speed of 2.5m / s, gradually decelerates according to an S-shaped deceleration curve, with the deceleration controlled at 0.45m / s. 2 Within 10 seconds, the speed smoothly decreased to 1.25 m / s (50% of the rated speed) and continued operation. Simultaneously, the braking system switched from standby to ready state, reducing the braking gap in preparation for intervention. Throughout the process, the control system continuously monitored the elevator's operating status, recording the entire process of speed reduction from 2.5 m / s to 1.25 m / s, comparing this data with a preset deceleration curve to confirm successful control execution and generate a control execution result. This adaptive control based on dynamic load can take appropriate measures according to different risk levels, ensuring safe elevator operation while minimizing interference with normal services.

[0116] In one specific embodiment, the process of executing step S106 may specifically include the following steps:

[0117] (1) Extract the actual speed, actual position and actual acceleration of the car from the control process state data to obtain the actual operating state data;

[0118] (2) Extract the target velocity, target position and target acceleration from the control command sequence to obtain the desired operating state data;

[0119] (3) Calculate the difference between the actual operating status data and the expected operating status data to obtain the status deviation data;

[0120] (4) Compare the state deviation data with the control accuracy requirements to determine whether the control execution meets the requirements and obtain the control accuracy evaluation results;

[0121] (5) Perform time-series analysis on the control process state data, calculate the time delay from the issuance of the control command to the response, and obtain the control response time characteristics;

[0122] (6) Generate control execution results based on state deviation data, control accuracy evaluation results and control response time characteristics.

[0123] Specifically, the actual operating parameters of the car are extracted from the control process state data. This data comes from real-time acquisition by multiple sensors. Actual speed data is obtained from the time difference of the position sensor or directly from a dedicated speed sensor, with a sampling frequency typically of 200Hz to ensure the capture of minute speed changes. Actual position data is provided by position encoders or magnetic induction position sensors installed in the hoistway, with accuracy down to the millimeter level. Actual acceleration data is obtained from three-axis accelerometers at the top and bottom of the car, and is obtained after filtering. This data undergoes time synchronization processing to eliminate sampling time differences between different sensors, forming a complete dataset of actual operating state. Target operating parameters are extracted from the control command sequence. The control command sequence is a series of commands generated by the coordination layer based on risk level instructions, containing the actions the elevator should perform at each point in time. From this, target speed commands are extracted; these are the speed values ​​the elevator should reach at each moment; target position commands indicate the specific positions the elevator should reach; and target acceleration commands specify the expected rate of acceleration or deceleration. These target values ​​constitute the desired operating state data, serving as a benchmark for evaluating the effectiveness of control execution.

[0124] Calculating the difference between actual and expected operating state data is the core step in control execution evaluation. The calculation uses a point-to-point comparison method, subtracting the actual speed from the target speed at the same moment to obtain the speed deviation; similarly, position and acceleration deviations are calculated. These deviation data are arranged in chronological order to form a time series of state deviation data, which intuitively reflects the differences between the actual and target states during control execution. The state deviation data is then compared with pre-set control accuracy requirements. These control accuracy requirements are permissible deviation ranges set based on elevator safety standards and operational comfort considerations, such as speed deviation not exceeding ±0.1 m / s, position deviation not exceeding ±10 mm, and acceleration deviation not exceeding ±0.2 m / s². 2 For each time point, the deviation value is checked to see if it exceeds the allowable range, and the number of data points exceeding the range and the degree of exceedance are counted. Based on these statistical results, control accuracy evaluation results are generated, including key indicators such as overall compliance rate, maximum deviation value and its occurrence time.

[0125] Timing analysis is a crucial method for evaluating the response performance of a control system. It calculates the time delay from control command issuance to response by analyzing the difference between the time of control command issuance and the actual response start time. This process is achieved by detecting control state change points: first, the issuance time of each control command is marked; then, the start time of the corresponding change is detected from the actual operating state data; the time difference between the two is the response delay. Typically, the response delays of multiple control points are calculated to obtain characteristic values ​​such as minimum delay, maximum delay, and average delay, forming the control response time characteristics. Based on state deviation data, control accuracy evaluation results, and control response time characteristics, a comprehensive control execution result is generated. The control execution result includes several key indicators: control accuracy, reflecting the degree of conformity between the actual state and the target state; control stability, indicating the fluctuation of the operating state; control timeliness, reflecting the system response speed; and control effectiveness, measuring the degree to which the control objective is achieved. These indicators are weighted to derive the final control execution score, which serves as the basis for subsequent safety monitoring and system parameter adjustment.

[0126] The above describes the elevator fall prevention adaptive protection method based on dynamic load in the embodiments of this application. The following describes the elevator fall prevention adaptive protection system based on dynamic load in the embodiments of this application. Please refer to [link / reference]. Figure 2 One embodiment of the elevator fall prevention adaptive protection system based on dynamic load in this application includes:

[0127] The data acquisition module is used to collect data on car acceleration, wire rope tension, car corner loads, motor speed, and car position through sensors installed in the car. The acquired data is then filtered to obtain the elevator's operating status.

[0128] The calculation module is used to calculate the static and dynamic wire rope tension values ​​based on the elevator's operating status, and to calculate the load distribution deviation value, thereby obtaining real-time load characteristic data of the elevator.

[0129] The segmentation module is used to perform window segmentation processing on the real-time load characteristic data of the elevator, extract feature parameters and compare them with preset thresholds to obtain the elevator abnormal state judgment result.

[0130] The classification module is used to classify the abnormal state type, elevator operating status and elevator real-time load characteristic data into risk level instructions based on the elevator abnormal state judgment results.

[0131] The execution module is used to perform corresponding control operations according to the risk level instructions. The control operations include data monitoring, reducing the running speed, smooth deceleration or emergency braking, and obtaining the control execution results.

[0132] The comparison module is used to compare the control execution result with the expected control effect, and activate the mechanical safety device when the deviation exceeds a preset threshold.

[0133] Through the collaborative efforts of the aforementioned components, and by acquiring and processing multi-sensor data in real time, comprehensive monitoring and accurate judgment of elevator operating status are achieved, significantly improving elevator safety. Multi-sensor collaborative operation collects car acceleration data, wire rope tension data, car corner load data, motor speed data, and car position data. Combined with advanced filtering algorithms, this provides high-quality foundational data for abnormal state detection, effectively reducing data noise and false positive rates. This method innovatively constructs a dynamic load model for the elevator system. By calculating static and dynamic wire rope tension values ​​and analyzing load distribution deviations, it accurately reflects the real-time load characteristics of the elevator, filling the gap in traditional fall prevention technologies' insufficient consideration of dynamic loads. In the data processing stage, window segmentation technology and multi-feature parameter extraction methods are employed to improve the sensitivity and accuracy of abnormal state detection, enabling precise identification of different types of elevator abnormal states. Regarding risk assessment, this method introduces artificial intelligence algorithms to comprehensively analyze abnormal state types, elevator operating status, and load characteristics, achieving scientific quantification of risk levels. In particular, the application of artificial intelligence algorithms allows the system to learn historical data patterns, continuously optimize risk assessment parameters, and improve the accuracy and adaptability of risk identification. Another innovation of this method is its adaptive, graded control strategy based on risk level. From data monitoring and speed reduction to smooth deceleration and emergency braking, a clearly defined protection system is formed, ensuring safety in critical moments while minimizing unnecessary operational interruptions. This method also establishes a comprehensive control execution evaluation mechanism. By comparing and analyzing the control execution results with the expected effects, system parameters are continuously optimized, forming a closed-loop self-learning system that allows the fall protection capability to continuously improve with accumulated operational experience. Particularly noteworthy is the contribution of artificial intelligence algorithms in specific application areas. These algorithms can handle complex and variable load characteristic data, identify potential abnormal patterns, and adaptively adjust braking force to achieve precise braking control based on actual load conditions—a function impossible for traditional fixed-parameter fall protection systems. In summary, this invention overcomes the limitations of traditional elevator fall protection technology, establishing a comprehensive protection system based on dynamic load characteristics, realizing a shift from passive protection to active prevention, and significantly improving elevator operational safety.

[0134] Reference Figure 3 This invention also provides a computer device, which can be a server, and its internal structure can be as follows: Figure 3As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0135] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.

[0136] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0137] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0138] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0139] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0140] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An adaptive protection method for elevator fall prevention based on dynamic load, characterized in that, The elevator fall prevention adaptive protection method based on dynamic load includes: Elevator operating status is obtained by collecting car acceleration data, wire rope tension data, car corner load data, motor speed data, and car position data through sensors installed in the car. The collected data is then filtered to determine the elevator's operating status. This includes: installing three-axis accelerometers at the top and bottom of the car to collect acceleration data along the X, Y, and Z axes to obtain raw acceleration data; installing tension sensors at the wire rope connections to collect wire rope tension changes at a frequency of 200Hz to obtain raw tension data; and installing load sensors at the four corners of the car floor to collect load data at the four corners to obtain raw load distribution data. A speed sensor is installed on the elevator shaft to collect the speed data of the elevator drive motor, obtaining raw speed data. A position sensor is installed in the elevator shaft to collect the real-time position data of the car, obtaining raw position data. The raw acceleration data, raw tension data, raw load distribution data, raw speed data, and raw position data are transmitted to the control unit via an industrial fieldbus (CAN-Bus) to obtain control unit input data. The control unit input data is then input into a Kalman filter for high-frequency noise filtering to obtain preprocessed data. Outlier detection and correction are performed on the preprocessed data to remove outliers and fill in missing values, thus obtaining the elevator operating status. Based on the elevator's operating status, the static and dynamic wire rope tension values ​​are calculated, and the load distribution deviation value is also calculated to obtain the elevator's real-time load characteristic data. The real-time load characteristic data of the elevator is processed by window segmentation, feature parameters are extracted and compared with preset thresholds to obtain the elevator abnormal state determination result. Based on the elevator abnormal state determination results, the abnormal state type, elevator operating status and elevator real-time load characteristic data are classified into levels to obtain risk level instructions. According to the risk level instruction, execute the corresponding control operation, which includes data monitoring, reducing the operating speed, smooth deceleration or emergency braking, and obtain the control execution result; The control execution result is compared with the expected control effect, and a mechanical safety device is activated when the deviation exceeds a preset threshold.

2. The elevator fall prevention adaptive protection method based on dynamic load according to claim 1, characterized in that, The process involves calculating the static and dynamic wire rope tension values ​​based on the elevator's operating state, and calculating the load distribution deviation value to obtain real-time load characteristic data of the elevator, including: Based on the car weight data and counterweight weight data during elevator operation, the static wire rope tension value is calculated using the principle of gravity balance. This is obtained by subtracting the counterweight weight from the sum of the car weight and the rated load, multiplying by the gravitational acceleration, and finally adding the inherent parameters of the wire rope. Extract the car acceleration, car speed, and car corner load values ​​from the elevator's operating status to obtain dynamic parameter input values; Correlation analysis and principal component analysis were performed on the dynamic parameter input values. By calculating the correlation coefficient matrix of the influence of each parameter on tension and extracting the main influencing factors, the acceleration dynamic coefficient, velocity dynamic coefficient and load distribution dynamic coefficient were obtained. Based on the load values ​​at the four corners of the car, the load distribution deviation is calculated using the four-point load variance analysis method. The result is obtained by summing the squares of the differences between each pair of the four corner load values ​​and the squares of the differences between the sums of the diagonal loads, and then taking the square root of the sum of the results. The adaptive correction coefficient is calculated based on historical operating data. By constructing the sum of squared errors between the actual tension and the calculated tension, and solving for the coefficient value that minimizes the sum of squared errors, the current optimal adaptive correction coefficient value is obtained. Substitute the static wire rope tension value, acceleration dynamic coefficient, velocity dynamic coefficient, load distribution dynamic coefficient, car acceleration, car speed, load distribution deviation value and adaptive correction coefficient into the dynamic load calculation formula, and calculate by adding the static tension value and the sum of the dynamic influence terms, where the dynamic influence terms are the sum of the adaptive correction coefficient and the products of each dynamic coefficient and the corresponding parameter, to obtain the dynamic wire rope tension value. The dynamic wire rope tension value and the static wire rope tension value are compared and analyzed. The tension change characteristics are obtained by calculating the difference and the rate of change between the two. The tension change characteristics, load distribution deviation value, dynamic wire rope tension value and static wire rope tension value are combined to form real-time load characteristic data of the elevator.

3. The elevator fall prevention adaptive protection method based on dynamic load according to claim 1, characterized in that, The step of performing window segmentation processing on the real-time load characteristic data of the elevator, extracting feature parameters and comparing them with preset thresholds to obtain the elevator abnormal state determination result includes: The real-time load characteristic data of the elevator is segmented into time windows to obtain a data window sequence; The tension fluctuation rate, load dynamic distribution index, acceleration anomaly coefficient, velocity deviation coefficient and system response time characteristic parameters are extracted from the data window sequence. Each parameter is compared with its corresponding upper and lower limits of the normal state threshold, and the number of parameters exceeding the threshold range and the degree of exceeding the threshold range are recorded. Based on a comprehensive score of the number of parameters exceeding the threshold and the degree of exceeding the threshold, the elevator operating status is divided into normal status, slightly abnormal status, moderately abnormal status, and severely abnormal status. By identifying the number of parameters exceeding the threshold range and the degree of exceeding the threshold, the abnormality type is identified, and the elevator abnormality status determination result is obtained. The elevator abnormality status determination result includes one or more of the following: wire rope abnormality, load distribution abnormality, mechanical jamming abnormality, and drive system abnormality.

4. The elevator fall prevention adaptive protection method based on dynamic load according to claim 3, characterized in that, Based on the elevator abnormality determination result, the abnormality type, elevator operating status, and real-time load characteristic data of the elevator are classified into risk level instructions, including: The abnormal state level in the elevator abnormal state determination result is scored and assigned a value: 1 point for normal state, 3 points for slight abnormal state, 6 points for moderate abnormal state, and 10 points for severe abnormal state, to obtain the abnormal state score value. The abnormality types in the elevator abnormality status determination results are scored for hazard level. Abnormal load distribution is assigned a score of 2 to 5 points, abnormal mechanical jamming is assigned a score of 4 to 7 points, abnormal drive system is assigned a score of 5 to 8 points, and abnormal wire rope is assigned a score of 7 to 10 points. The specific score is determined according to the degree of abnormality to obtain the abnormality type score. The elevator speed value is extracted from the elevator's operating status, and a score is assigned according to the speed level: 2 points for a speed not exceeding 1m / s, 4 points for a speed greater than 1m / s but not exceeding 2m / s, 6 points for a speed greater than 2m / s but not exceeding 3m / s, 8 points for a speed greater than 3m / s but not exceeding 4m / s, and 10 points for a speed exceeding 4m / s, thus obtaining the speed score. The current load value is extracted from the elevator's real-time load characteristic data, and a load score is obtained by dividing the current load value by the rated load value and then multiplying by 10. Calculate the duration of the abnormal state, and then calculate the duration score using an exponential function to obtain the duration score value; The abnormal state score, abnormal type score, speed score, load score, and duration score are input into the multi-factor weighted scoring unit. The total risk score is obtained by multiplying the abnormal state score by 0.3, the abnormal type score by 0.25, the speed score by 0.2, the load score by 0.15, and the duration score by 0.

1. Based on the total risk score, the risk levels are divided into Level 1 risk, Level 2 risk, Level 3 risk, and Level 4 risk. Specifically, a total risk score of less than 3 indicates Level 1 risk, a total risk score of 3 or more but less than 5 indicates Level 2 risk, a total risk score of 5 or more but less than 7 indicates Level 3 risk, and a total risk score of 7 or more indicates Level 4 risk. The risk levels are mapped to data monitoring strategies, early warning strategies, deceleration strategies, and emergency braking strategies, respectively, to generate risk level instructions.

5. The elevator fall prevention adaptive protection method based on dynamic load according to claim 1, characterized in that, The step involves executing corresponding control operations based on the risk level instruction. These control operations include data monitoring, reducing operating speed, smooth deceleration, or emergency braking, resulting in control execution outcomes, including: Based on the risk level instructions, the control strategy is divided into a decision-making layer, a coordination layer, and an execution layer. The decision-making layer receives the risk level instructions, the coordination layer transforms the control objective into a sequence of control instructions, and the execution layer connects the drive system and the braking system, resulting in a hierarchical control architecture. For instructions with a risk level of Level 1, data monitoring operations are performed to continuously monitor the elevator's operating status and record abnormal data to the system log, thereby obtaining a Level 1 risk response result; For instructions with a risk level of Level 2, an early warning operation is performed, an early warning message is sent to the management system, and the maximum operating speed of the elevator is reduced to 85% of the rated speed, thus obtaining a Level 2 risk response result; For instructions with a risk level of three, a deceleration operation is executed, forcibly reducing the elevator speed to 50% of the rated speed, using a smooth deceleration curve to control the deceleration to not exceed 0.5m / s. Simultaneously, the safety braking system is pre-activated, resulting in a three-level risk response. For commands with a risk level of four, an emergency braking operation is performed, cutting off the drive power and activating the mechanical safety devices and electronic braking system, controlling the braking deceleration to 0.8-1.2 m / s². Within the scope, a Level IV risk response result was obtained; During the emergency braking operation, the braking force is adjusted according to the dynamic load conditions. The magnitude of the braking force is related to the car weight, the current load weight, and the car speed, thus achieving an adaptive braking force control effect. Real-time status monitoring is performed on the results of the first-level risk response, the second-level risk response, the third-level risk response, and the fourth-level risk response, and changes in car speed, position, and acceleration are recorded to obtain control process status data. The control process status data is compared with the control command sequence to calculate the control execution status and obtain the control execution result.

6. The elevator fall prevention adaptive protection method based on dynamic load according to claim 5, characterized in that, The step of comparing the control process state data with the control command sequence, calculating the control execution status, and obtaining the control execution result includes: The actual speed, actual position, and actual acceleration of the car are extracted from the control process status data to obtain the actual operating status data. The target velocity, target position, and target acceleration are extracted from the control command sequence to obtain the desired operating state data; The difference between the actual operating status data and the expected operating status data is calculated to obtain the status deviation data; The state deviation data is compared with the control accuracy requirements to determine whether the control execution meets the requirements, and the control accuracy evaluation result is obtained. Perform timing analysis on the control process state data, calculate the time delay from the issuance of the control command to the response, and obtain the control response time characteristics; Based on the state deviation data, control accuracy evaluation results, and control response time characteristics, control execution results are generated.

7. An elevator fall prevention adaptive protection system based on dynamic load, used to implement the elevator fall prevention adaptive protection method based on dynamic load as described in any one of claims 1 to 6, characterized in that, The elevator fall prevention adaptive protection system based on dynamic load includes: The data acquisition module is used to collect data on car acceleration, wire rope tension, car corner loads, motor speed, and car position through sensors installed in the car. The acquired data is then filtered to obtain the elevator's operating status. The calculation module is used to calculate the static and dynamic wire rope tension values ​​based on the elevator's operating status, and to calculate the load distribution deviation value to obtain real-time load characteristic data of the elevator. The segmentation module is used to perform window segmentation processing on the real-time load characteristic data of the elevator, extract feature parameters and compare them with preset thresholds to obtain the elevator abnormal state judgment result; the division module is used to classify the abnormal state type, elevator operating status and elevator real-time load characteristic data into levels based on the elevator abnormal state judgment result to obtain risk level instructions. The execution module is used to perform corresponding control operations according to the risk level instruction. The control operations include data monitoring, reducing the running speed, smooth deceleration, or emergency braking, and obtain the control execution result. The comparison module is used to compare the control execution result with the expected control effect, and activate the mechanical safety device when the deviation exceeds a preset threshold.

8. A computer device, characterized in that, The system includes a memory and a processor, the memory storing a computer program that can run on the processor, characterized in that, when the processor executes the computer program, it implements the adaptive protection method for elevator fall prevention based on dynamic load as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, the computer program, when executed by a processor, causing the processor to perform the adaptive protection method for elevator fall prevention based on dynamic load as described in any one of claims 1 to 6.

Images