Elevator steel wire rope tension state recognition and early warning method based on internet of things

Abstract

The application discloses an elevator steel wire rope tension state recognition and early warning method based on an internet of things, and belongs to the technical field of tension data processing, and comprises the following steps: obtaining an elevator steel wire rope tension discrete sequence uploaded by a data acquisition device through the internet of things, restoring and recognizing a missing section according to a time stamp time sequence to update the tension discrete sequence, performing sequence cutting at a serial number jump of the tension discrete sequence, marking a plurality of transmission continuous sections after the cutting is completed, performing equal-interval resampling and spike suppression on each transmission continuous section to form an analyzable sequence, calculating a tension change rate of each transmission continuous section, dividing each transmission continuous section into a plurality of disturbance segments and a plurality of stable segments based on the tension change rate, and obtaining a tension result, so that the problems of inconsistent acceptance conclusions, misguiding of adjustment directions and untraceable remote judgment caused by mistakes of mistaking normal dynamic disturbance as too large deviation or mistaking unstable readings as qualified readings in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of tension data processing technology, and in particular to a method for identifying and warning of the tension status of elevator wire ropes based on the Internet of Things. Background Technology

[0002] Elevator traction systems typically use multiple steel wire ropes as the suspension and force transmission components between the car and the counterweight. The balance of force across these ropes directly affects the contact state between the traction sheave and the wire ropes, operational comfort, and the wear level of critical components. Therefore, online identification and early warning of abnormalities in the wire rope tension is a crucial direction for the digitalization of elevator operation and maintenance. With the development of IoT technology, on-site tension acquisition devices can upload tension data from multiple steel wire ropes to a remote platform in real time, enabling centralized monitoring and alarm linkage across sites and elevator types. This provides a foundation for the transformation of maintenance from scheduled on-site visits to on-demand or preventative maintenance.

[0003] Existing elevator wire rope tension monitoring and early warning systems typically collect the tension of multiple wire ropes by placing pressure or tension sensors at the ends of the wire ropes or at specific installation locations. The data is then uploaded to a data processing platform via a data acquisition and transmission unit. The data processing unit compares the current measured value with a preset threshold to determine whether the limit has been exceeded and triggers an alarm.

[0004] For example, Chinese invention patent CN119539495A discloses a method and device for identifying rope risks by integrating multi-dimensional data. The method includes: collecting multi-dimensional operational data of the rope, wherein the multi-dimensional operational data includes visual data, tension data, displacement data, and vibration data; performing preliminary anomaly analysis based on the collected multi-dimensional operational data and combining the anomaly conditions corresponding to each dimension of operational data to provide preliminary anomaly results for each dimension of operational data; performing secondary anomaly analysis based on the preliminary anomaly results of each dimension of operational data by integrating the multi-dimensional operational data to provide rope anomaly results; and identifying rope risks by providing a risk level based on the rope anomaly results and the preliminary anomaly results of each dimension of operational data.

[0005] The above-mentioned technology has at least the following technical problems:

[0006] While existing technologies can achieve a basic closed loop from tension acquisition to uploading, threshold deviation judgment, and alarm display, in acceptance and retesting scenarios after rope replacement or rope end reinstallation, the remote center often encounters the problem of unstable windows not being automatically reproducible when outputting conclusions based on the uploaded tension time series regarding whether the replacement is successful, whether further tightening or loosening is needed, and whether it can be used as a new baseline. For the same elevator, under slight differences in personnel, stopping positions, and other scenarios, the tension series may contain unstable segments such as brake instability and short-term rebound. If readings are still taken at fixed times or the average of the entire segment is directly calculated or threshold comparisons are performed, normal dynamic disturbances may be mistakenly treated as excessive deviations, or unstable readings may be considered as meeting standards. This can lead to inconsistent acceptance conclusions, misleading adjustment directions, and untraceable remote judgments.

[0007] In the absence of on-site working condition context information, such as the elevator stop status, stability level, and operation process nodes corresponding to the time of data collection, the system can only rely on the uploaded tension time series to make judgments. If the system fails to clearly define at the algorithm level which data segments can represent the current tension state and which data segments belong to non-steady-state segments such as short-term impacts and should be removed, it is easy to generate false alarms in acceptance scenarios that should be stable, reducing the credibility of the early warning system. On the other hand, it may also mask key anomalies by excessively suppressing fluctuations, bringing safety risks. Summary of the Invention

[0008] To address the technical problem of existing technologies where tension data can easily generate false alarms in stable acceptance scenarios, thus reducing the reliability of the early warning system, this invention provides an IoT-based method for identifying and issuing early warnings of elevator wire rope tension status. The technical solution is as follows:

[0009] A method for identifying and warning of elevator wire rope tension status based on the Internet of Things (IoT) is provided. This method includes: Step 1: Acquiring the discrete sequence of elevator wire rope tension uploaded by a data acquisition device via the IoT, restoring it according to timestamps, and identifying missing segments to update the discrete tension sequence; Step 2: Segmenting the sequence at sequence number jumps, marking the segmented sequence as several continuous transmission segments, and performing equal-interval resampling and peak suppression on each continuous transmission segment to form an analyzable sequence; Step 3: Calculating the tension change rate for each continuous transmission segment, and based on the tension change rate, classifying each continuous transmission segment... The subsequent segment is divided into a disturbing segment and several stable segments. When there are insufficient stable segments, the mean value method is used to obtain the tension result of the elevator wire rope. Otherwise, the convergence limit is obtained by using an exponential stabilization model as the tension result of the elevator wire rope. Step four: Analyze the median of the tension results of each elevator wire rope as the tension reference value, analyze the tension deviation sequence, locate several abnormal elevator wire ropes and issue warnings. For several abnormal elevator wire ropes, solve the corresponding adjustment amount through the already constructed coupling influence matrix, and correct each adjustment amount in combination with the number of elevator runs.

[0010] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0011] 1. The IoT-based elevator wire rope tension status identification and early warning method provided by this invention uses the discrete sequence of wire rope tension transmitted from the IoT as the judgment basis. By restoring the discrete sequence according to the timestamp and identifying missing segments, the discrete sequence is updated, enabling the remote platform to identify false differences introduced by transmission discontinuity at the data level, avoiding inconsistencies in the conclusions of the same acceptance process due to sequence misalignment or packet loss. At the same time, the boundary of the continuous transmission segment is located by the sequence number jump, and equal-interval resampling and peak suppression are performed on each continuous transmission segment. This allows tension data generated under different scenarios such as different personnel operations or different elevator stopping positions to be compared under a unified time reference, and isolates transient impacts, brake instability fluctuations, and short-term rebound peaks from statistical calculations, thereby providing a basis for the stable reproduction of stable segments. Furthermore, the tension change rate is calculated for each continuous transmission segment, and the continuous segment is divided into disturbance segments and stable segments accordingly. This allows the platform to clearly identify the data segment representing the current tension state based on the change rate evidence and eliminate disturbance segments, avoiding misjudging normal dynamic disturbances as excessive deviations and avoiding misjudging unstable readings as meeting standards. When there are insufficient stable segments, the tension results are output using the mean method, making the conclusions conservative and interpretable when steady-state evidence is insufficient. When the stable segments meet the conditions, the convergence limit is extracted using an exponential stabilization model as the tension result, transforming the tension result from a single-moment reading into a stable value inferred from the stabilization process. This reduces the impact of minor abnormal data differences on the acceptance conclusions and enhances the traceability of the conclusions. Based on this, a tension reference value is formed by using the median of the tension results of each wire rope, and a tension deviation sequence is constructed. This can locate abnormal wire ropes and trigger early warnings while suppressing outlier interference. It covers the risk of uneven force distribution among multiple ropes caused by single ropes being too tight or too loose, and avoids individual noise points dominating the overall judgment. Furthermore, by combining the constructed coupling influence matrix, the abnormal deviation is solved into the corresponding adjustment amount, and the elevator operation number is introduced to correct the adjustment amount. This allows the remote center to provide adjustment direction and amplitude suggestions consistent with the multi-rope coupling relationship while outputting whether the standard is met. This solves the problems of difficulty in reproducing the stability window, inconsistent conclusions, misleading adjustment directions, and difficulty in tracing remote judgments in acceptance and retesting scenarios after rope replacement or rope end reinstallation. It also reduces the operation and maintenance costs and safety risks caused by false alarms and missed alarms.

[0012] 2. This invention calculates the tension change rate sequence sequentially for each continuous transmission segment and locates disturbance candidate points using the absolute value threshold of the change rate. This allows for the explicit identification of instantaneous dynamic disturbances such as brake instability, short-term impacts, and rebound fluctuations from the discrete tension sequence with quantitative evidence. This eliminates the reliance on fixed-time readings or coarse-grained processing of the overall average for remote judgment, preventing the misjudgment of normal dynamic disturbances as excessive deviations or unstable readings as meeting standards. Furthermore, by combining adjacent disturbance candidate point interval thresholds with continuous length merging, discrete disturbance points can be shaped into continuous disturbance segments, forming representative stable candidate segments in the remaining intervals. In the selection of stable candidate segments, the persistence of stable segments is constrained by the minimum allowable window length. This avoids mistaking short-lived pseudo-steady-state intervals as baselines. Simultaneously, the minimum allowable number of stable segments constrains the amount of steady-state evidence, ensuring consistent evidentiary standards for acceptance and retest conclusions. When stable segments are insufficient, the tension discrete sequences of each stable segment are averaged and the tension result is output. This allows the system to provide conservative and interpretable tension conclusions even with insufficient steady-state evidence, avoiding misleading adjustment directions due to over-extrapolation. Conversely, when stable segments meet the requirements, disturbance segments are first eliminated. Then, based on the second rate of change threshold, the rebound decay tail interval is automatically located and an exponential stabilization model is fitted. The convergence limit is inferred using the convergence characteristics of rebound decay as the tension result. This transforms the tension result from an instantaneous reading into an inference of a stable value, suppressing false alarms while avoiding the safety risks caused by masking key anomalies due to full-segment smoothing.

[0013] 3. This invention introduces a constraint relationship between the fitting error deviation value and the maximum compliance value of the fitting error deviation. This allows the system to use model interpretability as the criterion for updating tension results. In situations where there is a lack of elevator stop status and stable node context on the remote side, it avoids directly solidifying fitting distortions caused by insufficient rebound decay, residual disturbances, or inappropriate threshold settings into the tension results. When the fitting error deviation value is lower than the maximum compliance value, the tension results to be analyzed are not frequently replaced, and the tension results of the elevator wire rope are updated accordingly for tension status identification. This ensures a stable and traceable source of identification data, reducing fluctuations in acceptance conclusions caused by repeated updates. Conversely, when the fitting error deviation value is not lower than the maximum compliance value, the second rate of change threshold is lowered, and the tension results to be analyzed are updated accordingly. This is equivalent to adaptively expanding the locatable range of the rebound decay tail interval and causing the exponential stabilization fitting to fall on a data segment closer to the actual convergence process. This makes the updated tension results more representative of the current steady-state tension level of the wire rope, thereby reducing the risk of misjudging unstable readings as compliant or misjudging normal disturbances as excessive deviations. Meanwhile, the number of elevator runs between the latest update time of the coupling influence matrix and the current time is introduced as the basis for correction. This allows the adjustment suggestions to be attenuated or calibrated as the rope elongation and state drift caused by the cumulative operation. This avoids over-adjustment or under-adjustment caused by the use of outdated coupling relationships. In this way, a closed loop of abnormal early warning is achieved in the acceptance and retesting scenarios after rope replacement or rope end reinstallation, with consistent conclusions, executable adjustment amounts and traceability. This also reduces the operation and maintenance risks caused by false alarms and missed alarms. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 A flowchart of an IoT-based elevator wire rope tension state identification and early warning method provided in an embodiment of this application;

[0016] Figure 2 This is a schematic diagram of the data acquisition and preprocessing process provided in the embodiments of this application;

[0017] Figure 3 This is a schematic diagram of the tension result acquisition process provided in the embodiments of this application;

[0018] Figure 4 This is a schematic diagram of the status recognition and early warning process provided in the embodiments of this application;

[0019] Figure 5 The first set of tension timing processing comparison curves provided for embodiments of this application;

[0020] Figure 6 The second set of tension timing processing comparison curves provided for embodiments of this application. Detailed Implementation

[0021] The technical solution provided in this application will now be described with reference to the accompanying drawings.

[0022] To facilitate understanding of the embodiments of this application, the following points will be explained first:

[0023] First, in this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates an "or" relationship between the preceding and following related objects, but it does not exclude the possibility of indicating an "and" relationship. The specific meaning can be understood in conjunction with the context.

[0024] Second, the use of prefixes such as "first" and "second" in this application is solely for the purpose of distinguishing and describing different things belonging to the same category, and does not constrain the order, size, or quantity of things. For example, "first message" and "second message" are simply different messages, and there is no chronological, size, or priority relationship between them.

[0025] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0026] Example 1, as Figure 1 The diagram shown is a flowchart of an IoT-based elevator wire rope tension state identification and early warning method provided in this application embodiment. The method includes the following steps:

[0027] Taking the acceptance retest of a traction elevator after rope reinstallation as an example, to eliminate the influence of passenger load changes on tension, the acceptance process is conducted under no-load conditions, i.e., with no personnel or goods in the car, and the car stopped at a preset floor and kept stationary after the car doors are closed. No-load conditions indicate that the car load remains constant, approximately equal to the car's own weight and the weight of fixed components. This allows fluctuations in the tension sequence to better reflect the rope end adjustment state, elastic rebound, and minor environmental disturbances, rather than load disturbances caused by passengers moving up and down.

[0028] During acceptance testing, the elevator car is stopped at an intermediate floor (e.g., a floor in the middle of the shaft), maintaining the elevator in a stationary state, meaning the traction machine has no driving force output, the brake is closed, and the car and counterweight are near static equilibrium. Because a portable single-channel measuring device is used, this embodiment measures only one target wire rope at a time; after completing the measurement of that rope, the measurement is repeated for the next wire rope, thus achieving individual acceptance and archiving of multiple wire ropes. To distinguish different wire ropes of the same elevator from a remote perspective, this embodiment assigns a RopeID to each wire rope.

[0029] Tension data acquisition devices, such as tension sensors, are fixed to elevator steel cables and output the tension sample value T[k] of a single elevator steel cable at a sampling frequency of 5Hz, where k is the sampling sequence number, k=1, 2, 3, ..., y, and y is the total sampling sequence number. The data acquisition device uploads data in 5-second cycles, integrating several sampling points collected within the cycle to form a data packet, which is then uploaded to the data processing terminal. The data packet is the smallest transmission unit for a single upload and includes at least: device identifier ElevatorID, rope identifier RopeID, data packet generation timestamp t0, sampling frequency, starting sampling sequence number, and tension discrete sampling sequence. The timestamp is used for time alignment in the cloud, and the sampling sequence number is used to identify sequence number rollback caused by packet loss, out-of-order delivery, or device restart in the cloud.

[0030] For example, in one instance, the data collected by the tension data acquisition device is shown in Table 1, which presents an example table of tension values:

[0031] Table 1 Example Table of Tension Values

[0032] serial number Tension value / N Deviation value / N Deviation ratio / % 1 1946.9 -4.06 0.21 2 1933.5 9.34 0.48 3 1947.5 -4.66 0.24 4 1929.7 13.14 0.68 5 1951.5 -8.66 0.44 6 1942.2 0.64 0.03 7 1942.2 0.64 0.03 8 1960.4 -17.56 0.9 9 1925.2 17.64 0.92 10 1949.3 -6.46 0.33

[0033] Table 1 is used to display the consistency and dispersion of 10 sets of tension measurement results obtained at the same measuring point (or under the same working condition). The table numbers correspond to sampling groups 1 to 10; the tension value, in N (Newtons), represents the tension result measured in each group; the deviation value, in N (Newtons), characterizes the degree of deviation of a single set of data relative to the overall level. It is calculated by subtracting the tension value corresponding to the number from the mean of the 10 sets of tension values. Therefore, a positive deviation value indicates that the tension value of that group is lower than the mean, and a negative deviation value indicates that the tension value of that group is higher than the mean; the deviation ratio, in % (percentage), is a relative quantitative indicator of the deviation, calculated by dividing the absolute value of the deviation by the tension value. It is used to eliminate the influence of dimensions, facilitate horizontal comparison of the relative deviation of each set of data, and thus intuitively reflect the stability and fluctuation range of the tension measurement results in this batch.

[0034] Under no-load and static conditions, the actual change in tension usually manifests as slow convergence or slight fluctuations. However, IoT links may still experience out-of-order arrivals, short-term packet loss, or sudden increases in latency. To avoid misinterpreting data misalignment caused by link jitter as tension instability, the data processing end first caches and indexes data packets with the same ElevatorID and RopeID, and then performs time-series restoration by timestamp. That is, the data packets are sorted from smallest to largest by t0, and then expanded and concatenated according to the sampling sequence number to obtain the restored discrete tension sequence. The discrete tension sequence refers to the set of discrete sampling points organized by sampling sequence number and restoration time, which can be represented as {t[k], T[k]}, where t[k] is the sampling time point corresponding to the kth sampling sequence number after restoration, and T[k] is the tension value corresponding to the kth sampling sequence number.

[0035] After restoration, the data processing unit performs missing segment identification, which involves verifying the continuity of the sampling sequence differences between adjacent sampling points. If the difference between the sampling sequence number of the later sampling time point and the sampling sequence number of the earlier sampling time point is greater than 1, it is determined that a missing sampling point exists within that interval, and this interval is marked as a missing segment. A missing segment refers to a gap in the sampling points caused by packet loss or buffer failure. Its essence is a transmission anomaly rather than a true tension mutation, therefore it needs to be explicitly marked for subsequent analysis to avoid or segment it. Through time series restoration and missing segment marking, the data processing unit updates the tension discrete sequence, obtaining a traceable sequence arranged according to real time and containing missing information, reducing the impact of network out-of-order delivery or packet loss on the static acceptance conclusion.

[0036] After the tension discrete sequence is updated, the data processing end executes a segmentation strategy to divide the sequence into multiple segments at points where continuity is broken. This embodiment uses sequence number jump segmentation: when a missing segment occurs, the sequence is segmented at that point and marked as several transmission continuous segments. A transmission continuous segment refers to a continuous data interval within that segment where the sampling sequence number strictly increases and does not cross the missing segment. It represents a valid observation record that can be continuously mapped to the time axis and serves as the basic input for subsequent equal-interval resampling, peak suppression, and steady-state window identification.

[0037] To ensure that subsequent algorithms run on a unified time grid and suppress single-point noise, the data processing unit performs equally spaced resampling and peak suppression on each continuous transmission segment to form an analyzable sequence. Equal-spaced resampling involves mapping the continuous segment to a time grid with a fixed step size Δt = 0.2 s. When slight time jitter exists in the sampling, interpolation is used to fill in the gaps. Specifically, the sampling is first performed at a fixed time step size Δt = 0.2 s. Construct a standard time axis t0, t0+Δt, t0+2Δt, ..., and then project (t[k], T[k]) onto this standard time axis; if there is no corresponding sampling point near a certain standard time (due to slight early or late time jitter in the sampling time), then use the two nearest sampling points before and after it to perform linear interpolation to calculate the tension value at that time, thus obtaining an equally spaced sequence of tension values ​​at each 0.2s time; Peak suppression refers to the robust processing of occasional pulse anomalies (such as instantaneous jitter of sensor contact, single-point jumps introduced by slight slippage of rope end) under no-load static conditions. For example, using a sliding window midpoint, points that deviate from the neighborhood statistics by more than a threshold are replaced with the neighborhood representative value, thereby avoiding single-point peaks being misjudged as tension not yet stable. Specifically, it means: assuming that when the elevator is no-load and stationary, the tension of this steel wire rope fluctuates slowly at around 1000N. The neighborhood window (11 points) is: 999N, 1001N, 1000N, 1002N, 998N, 1300N, 1001N, 999N, 1000N, 1002N, 999N. Of these 11 points, except for 1300N, the rest are between 998N and 1002N. The neighborhood median m = 1000N, and the data deviation for 1300N is |1300−1000| = 300N. Assume... The technicians set the threshold to 50N. 300N clearly exceeded the threshold, so 1300N was identified as an isolated spike and replaced with the neighborhood median, which is the average of the two numbers around 1300N, resulting in 999.5N. The updated sequence is: 999N, 1001N, 1000N, 1002N, 998N, 999.5N, 1001N, 999N, 1000N, 1002N, 999N.

[0038] The analyzable sequences obtained through the above processing will serve as a unified input and evidence carrier for subsequent steps, including automatic extraction of stable windows, acceptance judgment, determination of whether standards are met, whether further tightening or loosening is needed, whether a new baseline can be solidified, and early warning display.

[0039] Figure 2This is a schematic diagram of the data acquisition and preprocessing process provided in this application embodiment. The discrete sequence of elevator wire rope tension is acquired through the Internet of Things (IoT). The sequence is then restored and updated according to timestamps. Tension data is integrated and packaged into data packets and timestamped. Based on the timestamps, the discrete tension sequence within the data packets undergoes time-series restoration processing to reduce network interference on sequence analysis. A segmentation strategy is applied to the updated discrete tension sequence, dividing it into several continuous transmission segments at sequence number transitions. Equal-interval resampling and peak suppression processing are performed on each continuous transmission segment to form an analyzable sequence. Finally, the tension change rate of each continuous transmission segment is calculated to complete the division between disturbed and stable segments.

[0040] Define the rate of change of tension as v k , To establish a reproducible boundary between disturbance and stability, a first rate of change threshold θ is predefined in the database to determine whether the change is abnormally rapid, representing the maximum permissible normal rate of change of tension. k When | exceeds θ, it indicates that the tension change near that moment is too rapid to exceed the range of natural static fluctuations, possibly corresponding to dynamic disturbances such as rebound, external contact, or instantaneous sensor jitter. Therefore, the system will satisfy |v k | Sampling points exceeding θ are marked as perturbation candidate points.

[0041] Considering that disturbance events often occur in clusters, the system further introduces an interval threshold to segment and shape the disturbance candidate points. The interval refers to the distance between two adjacent disturbance candidate points on the time axis; the interval threshold is a pre-defined engineering constraint in the database that allows for brief intervals within the same disturbance. When the interval between two disturbance candidate points does not exceed the interval threshold, the system considers them different manifestations of the same disturbance process and should be merged into the same disturbance segment. Conversely, when the interval exceeds the interval threshold, the system considers them two independent disturbance events and should belong to different disturbance segments. Based on this rule, the system merges the set of candidate points that satisfy the condition that the interval between adjacent disturbance candidate points does not exceed the interval threshold and determines the segment boundaries, thus forming several disturbance segments with start and end times.

[0042] After the disturbance segments are determined, the remaining intervals in the transmission continuum that are not covered by any disturbance segments are defined as stable candidate segments. Stable candidate segments refer to continuous intervals in which the rate of change of tension never exceeds the first rate of change threshold and does not overlap with the time range of the merged disturbance events; these intervals are more representative of the stable tension state required for static acceptance.

[0043] Since short-term rebound and minor disturbances may still occur during the static acceptance process, the data processing end needs to further screen stable candidate segments and output the final wire rope tension result in different ways when there are insufficient stable segments and sufficient stable segments.

[0044] The data processing unit calculates the window length for each stable candidate segment. The window length refers to the duration the segment continuously covers on the timeline, used to measure whether the segment is sufficient to represent a stable tension state. The system presets a minimum allowable window length, indicating that only when the window length reaches this minimum is the segment considered resistant to occasional perturbations and statistically representative. Therefore, the system filters out stable candidate segments with window lengths not less than the minimum allowable value and marks these segments as stable. Subsequently, the number of stable segments is counted and compared with the preset minimum allowable number of stable segments in the database. This minimum allowable number of stable segments constrains the lower limit of the number of valid stability evidence, preventing unrepeatable acceptance conclusions from being drawn based on only a very small number of segments.

[0045] When the comparison results show that the number of stable segments is less than the minimum allowable number of stable segments, the system determines that there are insufficient stable segments. In this case, the first analysis method is used to obtain the tension result of the wire rope. The purpose of the first analysis method is to provide a traceable and easily interpretable tension estimate even when the stability evidence is insufficient. The specific method is as follows: the tension discrete sequences corresponding to each stable segment are averaged to obtain the average tension value of each segment, and then summarized at the segment level, that is, the average tension of each segment is averaged again, and the resulting summed average is used as the tension result of the wire rope. At the same time, the start and end times and segment lengths of the stable segments involved in the calculation are retained, so that the remote center can trace which stable segments support the tension result.

[0046] Figure 3 This is a schematic diagram of the tension result acquisition process provided in this application embodiment. Based on the divided tension change rate, disturbance candidate points are marked. Adjacent disturbance candidate points are merged to form disturbance segments. The remaining intervals in the transmission continuous segment that are not marked as disturbance segments are stable candidate segments. After screening each stable candidate segment, it is determined whether there are enough stable segments. If there are not enough stable segments, the tension result of the elevator wire rope is obtained by averaging. If there are enough stable segments, the tension sequence of the rebound decay tail interval is fitted by an exponential stabilization model, and the convergence limit obtained by fitting is used as the tension result. Finally, the median of the tension result of each elevator wire rope is extracted and determined as the tension reference value.

[0047] When the comparison results show that the number of stable segments is not less than the minimum allowable number of stable segments, the system determines that there are enough stable segments. At this time, the second analysis method is used to obtain the wire rope tension result. The purpose of the second analysis method is to use a model that is more in line with the physical convergence law to obtain the final static tension that should be converged when the rebound decay process still exists but the stability evidence is sufficient, thereby improving the consistency of the acceptance baseline solidification. The specific process is as follows: First, all disturbance segments are removed to avoid strong disturbances caused by start-stop rebound or external touch that could bias the fitting. Then, based on the second rate of change threshold preset in the database, the tail interval of rebound decay is automatically located. The second rate of change threshold is a criterion boundary used to define that the rebound has entered a slow decay stage. When the absolute value of the tension change rate is lower than the duration of the second rate of change threshold and is greater than the duration threshold preset by the technician, the tension change has transitioned from strong disturbance to slow convergence. The duration threshold is a value used to define whether the tension change is slowly converging. It is suitable to be represented by a stabilization model. In the sequence after removing disturbances, a continuous interval that meets the criterion and the duration requirement is searched and marked as the tail interval.

[0048] In the tail section, the system fits the tension sequence to an exponentially stabilizing model, meaning that the tension gradually approaches a certain convergence limit over time, and the model form is as follows:

[0049] ;

[0050] Where T(k1) refers to the tension value of sampling number k1 in the tail section, and k1 is an element in the subset of sampling numbers in the tail section selected from the complete set of sampling numbers. T∞ is the convergence limit, which represents the static tension level that the wire rope should reach after the rebound is completely decayed. A is the initial deviation of the tension value of sampling number k0 in the tail section relative to the convergence limit, which is used to characterize the amount of rebound that is still remaining at the beginning of the tail section. k0 represents the starting sampling number of the tail section, which is located by the second rate of change threshold. τ is the time constant, which is used to characterize the rebound decay rate. e is the natural constant.

[0051] In a static acceptance test scenario under no-load conditions, the entire system—from the wire rope to the rope end connection, the force measuring device, and the structural components—can be considered as a damped elastic system. After a short-term operation (e.g., fine-tuning the rope end, manual touch, slight rebound), the system's tension will deviate from its final stable value. Under damping, this deviation will gradually decay over time and return to equilibrium. The underlying principle is as follows: For a first-order (or approximately first-order at the tail end) linear elastic damped system, the deviation term satisfies a similar... Its solution exhibits an exponential decay characteristic. Therefore, the tail process of tension can be written as a steady value plus an exponential decay term, where the deviation refers to the difference between the tension value after a short operation and the convergence limit.

[0052] In terms of the fitting method, the data processing end first uses the tension measurements collected in chronological order within the tail section as observation data, and uses the corresponding predicted tension value T(k1) calculated by the exponential stabilization model at the same set of sampling times as the model output. Then, for each sampling time, the deviation between the predicted value and the observed value is calculated, and the deviation is accumulated point by point within the tail section to form a total error. This total error is used to characterize the degree of fit of the current model parameters to the tail attenuation trajectory. Finally, the data processing end iteratively adjusts the model parameters to minimize the total error, thereby obtaining a set of parameter solutions that best approximate the actual attenuation process in the tail section, and determines the corresponding convergence limit as the tension result of the steel wire rope. After the fitting is completed, the obtained convergence limit is used as the tension result of the elevator steel wire rope, and the tail section range, fitting residual and convergence criterion are output simultaneously to support the interpretability, repeatability and traceability of the remote acceptance conclusion that it can be solidified as a new baseline.

[0053] Tension data was collected sequentially from multiple steel cables of the same elevator. Each cable was processed through a process involving elimination of disturbed segments until stable candidate segments were selected, leading to either a first or second analysis method to obtain the corresponding tension result. The acquisition method was recorded simultaneously with the generation of the tension result. The acquisition method refers to the calculation source path of the tension result for that steel cable: when stable segments were insufficient, the result obtained by averaging corresponded to the first analysis method; when stable segments were sufficient, the result obtained by tail-interval exponential stabilization fitting and taking the convergence limit corresponded to the second analysis method. The data processing unit aggregated the tension results and acquisition methods of each steel cable in the same acceptance batch, forming a set of tension results for each cable individually, which served as input for subsequent consistency verification and result updates.

[0054] After aggregation, the system robustly aggregates the tension results of each wire rope, taking the median as the tension reference value. This reference value characterizes the typical tension level of the elevator under no-load, static acceptance conditions. The median, rather than the mean, is used because it is less sensitive to abnormally high or low tensions in individual ropes, providing a more stable reference. Simultaneously, the system sets a threshold number of ropes as a trigger condition to determine if the sample size is sufficient for fitting consistency verification using the second analysis method. The threshold number refers to the minimum number of wire ropes required for the second analysis method to be effective. For example, it requires that the tension results of at least a certain number of wire ropes be obtained from the fitting convergence limit before the fitting quality is considered statistically representative, avoiding unstable conclusions due to standard deviation judgments on fitting errors when the sample size is too small.

[0055] When the tension results of elevator wire ropes below the defined number are determined to be obtained using the second analysis method, the tension status of the elevator wire ropes is directly identified based on the tension results of each individual elevator wire rope. When the tension results of elevator wire ropes above the defined number are determined to be obtained using the second analysis method, these wire ropes whose tension results were obtained using the second analysis method are marked as elevator wire ropes to be analyzed. The designation of an elevator wire rope as to be analyzed does not necessarily indicate an anomaly, but rather that its tension result was obtained through model fitting, and the fitting process includes evidence of error, making it suitable for further consistency checks of the fitting quality. For each elevator steel wire rope to be analyzed, the system synchronously acquires its tension result fitting error value. This fitting error value is generated during the fitting process using the second analysis method, for example, by measuring the residual between the model prediction value and the observed tension value in the tail interval (which can be the root mean square residual or the weighted residual). In this embodiment, the fitting error value is represented by the total error of the fitting process, which reflects the degree to which the tail exponential stabilization model of the steel wire rope fits the actual attenuation trajectory. A larger error usually means that there are still unremoved micro-disturbances in the tail interval, or tail positioning deviation, or heavy sensor noise, thereby reducing the reliability of the convergence limit estimation.

[0056] After obtaining the fitting error values ​​for each elevator wire rope to be analyzed, the system performs standard deviation processing on these error values ​​to obtain the fitting error deviation value. The fitting error deviation value is used to quantify whether the fitting error is consistent across different wire ropes: when the fitting errors of most wire ropes are within the same order of magnitude, the deviation value is small, indicating that the tail positioning and fitting quality are generally stable; when the fitting errors of individual wire ropes are significantly larger, the deviation value will be amplified, suggesting the existence of outlier ropes with unreliable fitting that require further processing.

[0057] The system compares the fitting error deviation value with the preset maximum compliance value of fitting error deviation in the database. The maximum compliance value of fitting error deviation is the upper limit threshold used to define whether the fitting consistency meets the acceptance and solidification requirements. It can be obtained from the statistics of historical qualified acceptance samples or configured according to the platform strategy. It is used to ensure that the tension results have consistency and repeatability in fitting quality.

[0058] Based on the above comparison results, the system determines whether the tension results of each elevator wire rope to be analyzed need to be updated: when the deviation of the fitting error does not exceed the maximum compliance value of the fitting error deviation, the system determines that the fitting quality of each wire rope to be analyzed is consistent, retains the convergence limit obtained by the second analysis method as the final tension result, and archives the tension reference value, the source of the median and the evidence of the fitting error together.

[0059] When the fitting error deviation exceeds the maximum acceptable deviation value, the system determines that the dispersion of the fitting quality between the elevator wire ropes being analyzed is too large. This indicates that the second rate of change threshold currently used for automatic positioning of the tail section is too lenient, resulting in some wire rope tail sections still containing residual rebound or micro-disturbance points, thus leading to unstable convergence limit estimation. Therefore, the system triggers a threshold self-tightening update mechanism: based on the magnitude of the fitting error deviation, the second rate of change threshold is lowered to make the tail section positioning rules more stringent. Specifically, the lowering rule involves: querying the second rate of change threshold reduction corresponding to the current fitting error deviation value from the fitting error deviation-second rate of change threshold reduction mapping table stored in the database; subtracting the second rate of change threshold reduction from the current fitting error deviation value to reduce the preset second rate of change threshold. The second rate of change threshold reduction is a positive value, representing the amount of value to be removed from the second rate of change threshold.

[0060] After the second rate of change threshold is lowered, the system re-performs tail interval positioning for each elevator wire rope to be analyzed: that is, in the tension sequence after removing disturbance segments, it re-searches for a continuous interval that satisfies the requirement that the tension rate of change is continuously lower than the updated second rate of change threshold for a duration that meets the requirement, and re-marks this interval as the tail interval. Subsequently, the system uses the updated tail interval tension value as the observation data to re-perform exponential stabilization fitting, obtains a new convergence limit, and updates the tension result of the elevator wire rope to be analyzed with this convergence limit.

[0061] After the above update is completed, the system uses the updated tension results of each elevator wire rope to be analyzed as the standard, and re-aggregates the tension results of each elevator wire rope to obtain a set of tension results for the elevator. Based on this, the system performs elevator wire rope tension status identification. Specifically, the system can calculate the tension reference value (e.g., median) and the relative deviation of each rope based on the tension results of each wire rope. It can then identify whether there are tension status categories such as uneven tension of multiple ropes, significantly high / low tension of individual ropes, or overall tightness / looseness. The identification results are used as the basis for the output of acceptance conclusions and adjustment suggestions. Since the tension results of the wire ropes to be analyzed have undergone consistency correction through threshold tightening to tail repositioning to refitting process, the final status identification conclusion has higher repeatability and traceability on the remote side.

[0062] The system analyzes the deviation between the tension result and the tension reference value of each wire rope to obtain the tension deviation value. The tension deviation value refers to the degree of deviation of the rope tension result from the tension reference value, that is, the difference between the tension result and the tension reference value, which is used to reflect whether the rope is significantly too tight or too loose. To avoid misjudging normal slight fluctuations as anomalies, the database presets a maximum tension deviation compliance value, which means: the upper limit of the maximum allowable deviation range under the conditions of elevator model, acceptance conditions, and sampling accuracy. This compliance value can be given by manufacturing / maintenance specifications, or formed by the platform based on historical qualified acceptance samples, and is permanently fixed as an anomaly judgment threshold or configured according to site strategy.

[0063] The system performs threshold judgment on the tension deviation value of each wire rope: when the absolute value of the tension deviation value of a certain wire rope is greater than the maximum tension deviation compliance value, it indicates that the tension result of the rope has deviated beyond the typical level of most wire ropes. The system locates it as an abnormal elevator wire rope. An abnormal elevator wire rope refers to a target rope with significant tension deviation under the current acceptance snapshot, which may lead to uneven force on multiple ropes or abnormal traction contact state.

[0064] With the elevator in an unloaded and stationary state, maintenance personnel conducted tension testing experiments by fine-tuning each wire rope, retesting the tension, and creating trial adjustment records. During each trial adjustment, the system acquired the tension results of each wire rope before and after the adjustment (the method for acquiring the tension results still follows the judgment and calculation process of the aforementioned first / second analysis method). The system then wrote the adjustment object, adjustment step size, tension result before adjustment, and tension result after adjustment as an experimental sample into the data processing terminal for subsequent coupling relationship modeling.

[0065] During the tension testing experiment, the system defines a coupling effect coefficient as the ratio of tension change to the trial adjustment step size. This coefficient is used to quantify the cascading effect of adjusting one rope on the tension changes of itself and other ropes. Specifically, let the s-th wire rope be selected as the adjustment object in this trial adjustment, where s = 1, 2, 3, ..., d, and d is the total number of wire ropes. The trial adjustment step size is denoted as Δu. sThe trial adjustment step length refers to the standardized fine-tuning amount applied by maintenance personnel to the rope end adjustment mechanism. To ensure that trial adjustment data from different personnel, tools, and recording methods can be compared, the system pre-establishes conversion rules for adjustment actions to equivalent displacement / equivalent step length. For example, using the pitch of the adjusting nut as a basic parameter, the rotation angle or number of turns of the nut is converted into the axial displacement of the rope end clamping point; or based on the structural parameters of the tensioning device, the action quantities such as wrench scale, ratchet click count, and stepper motor pulse count are mapped into equivalent displacement quantities. These conversion rules are fixed on the platform side in the form of mapping tables or conversion coefficients (or configured according to the equipment model), so that each trial adjustment can be uniformly converted to a standardized step length under the same dimension. This is used to calculate the coupling influence coefficient of the ratio of tension change to trial adjustment step length and to achieve comparability of cross-trial adjustment records.

[0066] The tension of the i-th wire rope before the trial adjustment is F. post i The tension result of the i-th wire rope after trial adjustment is F pre i The tension change caused by this trial adjustment is ΔF. i =F post i -F pre i Based on this, the system calculates the coupling influence coefficient, i = 1, 2, 3, ..., d, where d is the total number of wire ropes. The system calculates the coupling influence coefficient a. i,s =ΔF i / Δu s ;a i,s This represents the change in tension of the i-th wire rope caused by adjusting the unit step length of the s-th wire rope. When i=s, the self-effect is drawn (the efficiency of adjusting itself on its own tension), and when i≠s, the cross-effect is drawn (the intensity and direction of the linkage effect of adjusting a rope on other ropes).

[0067] After obtaining the coupling influence coefficients from multiple trial adjustments, the system fills in and forms a coupling influence matrix according to the mapping relationship between the regulated object and the affected object. Specifically, matrix Z is constructed with the affected object as the row and the regulated object as the column, and the matrix element is Z(i, s) = a. i,s .

[0068] After completing at least one round of trial adjustments for each wire rope as the adjustment object, a complete coupling influence matrix can be obtained. The physical meaning of this matrix is ​​that by inputting a vector composed of a certain set of adjustment step sizes, the direction and magnitude of the linkage change in the tension of each wire rope can be predicted, thus providing a calculable basis for solving the adjustment amount and anomaly early warning for multi-rope balancing.

[0069] The system calls the already constructed coupling effect matrix to inversely deduce which rope end adjustment mechanisms should be applied and how much trial adjustment should be applied to ensure that the tension deviation of each wire rope returns to the compliant range, taking into account the multi-rope linkage effect.

[0070] Specifically, the system first constructs a tension deviation vector, whose elements are the deviation of the current tension of each wire rope from the tension reference value. This deviation can be taken as the difference between the current tension and the tension reference value; a positive difference indicates that the rope is too tight, and a negative difference indicates that the rope is too loose. Subsequently, the system uses a coupling influence matrix to establish a linear prediction relationship between the adjustment amount and the tension change: each column in the coupling influence matrix corresponds to the standardized trial adjustment step size of a certain wire rope, each row corresponds to the tension change of a certain affected wire rope, and the matrix elements represent the tension change caused by a unit trial adjustment step size. Therefore, when multiple wire ropes are simultaneously subjected to trial adjustment amounts, the matrix multiplied by the trial adjustment vector yields the predicted tension change of each wire rope.

[0071] The system aims to solve for the adjustment amount by offsetting the tension deviation. Let the coupling effect matrix be Z, where the element in the i-th row and s-th column represents the change in tension of the i-th wire rope caused by adjusting the unit trial adjustment step of the s-th wire rope; let the current tension deviation vector be Q, where its i-th component is the tension deviation of the i-th wire rope; let the trial adjustment amount vector to be solved be Δu, where its s-th component is the standardized trial adjustment step size to be applied to the s-th wire rope (the positive and negative signs correspond to the directional conventions of "re-tightening" or "re-loosening", respectively).

[0072] The residual deviation after the trial adjustment is expressed as Q + Z * Δu, and Δu is solved by minimizing the weighted sum of squares of the residual deviation. The objective function can be written as: W is the weight matrix, used to assign higher weights to abnormal wire ropes, making the solution prioritize eliminating deviations in abnormal ropes. β is the regularization coefficient, used to suppress excessively large solutions that could lead to overly aggressive one-time adjustments, thereby improving the stability and feasibility of trial adjustments. The weight matrix and regularization coefficient can be adaptively determined by the platform based on acceptance compliance requirements and historical trial adjustment results. The weight matrix is ​​set primarily based on the degree of deviation exceeding limits and the risk level, assigning higher weights to abnormal wire ropes whose absolute tension deviation exceeds the maximum compliance value, with the weight increasing as the deviation approaches or exceeds the safety boundary, while weights decrease as the deviation falls within the compliance range. The wire ropes within the system are assigned a low weight to ensure that abnormal ropes are pulled back to the compliance range first during the solution process. The regularization coefficient is set based on the safe and executable upper limit of a single adjustment as a constraint benchmark, combined with the allowable step size of the rope end adjustment mechanism, the maximum empirical single fine adjustment amount, and the tolerance number of multiple rounds of closed-loop retests. This ensures that when there is uncertainty in the fitting matrix or a strong linkage effect, the solution results still tend to output a trial adjustment suggestion with controlled amplitude and step-by-step implementation, avoiding new imbalances or misjudgments of direction caused by excessive one-time adjustment. Furthermore, the coefficient can be calibrated using historical acceptance samples to balance convergence speed and adjustment stability.

[0073] The principle behind the above formula is as follows: Among all available adjustment quantities, find the set of adjustment quantities that minimizes the target. The tension linkage change caused by the adjustment quantity is linearly represented by the coupling influence matrix, and the remaining tension deviation after adjustment is taken as the target to be eliminated as much as possible. Based on the current tension deviation of each rope, the system predicts how the deviation of each rope will change after applying a certain set of trial adjustment quantities, thereby constructing a calculable error cost for the overall magnitude of the deviation after adjustment. At the same time, in order to make the solution result more in line with the maintenance priority, a weighting mechanism is adopted to make the abnormal ropes that exceed the limit contribute more to the cost, thereby prioritizing the compression of abnormal rope deviations during the solution. In addition, in order to avoid giving too large a trial adjustment quantity at one time, which would lead to excessive adjustment or introduce new imbalances, a penalty term for the adjustment quantity amplitude is further added, so that the solution achieves a compromise between eliminating deviations as much as possible and controlling the adjustment amplitude, and finally obtains a set of trial adjustment suggestions that can effectively reduce abnormal deviations and have engineering feasibility.

[0074] Considering that factors such as rope slack, friction embedding, and structural micro-deformation may cause the coupling relationship to drift over time during subsequent elevator operation, the system introduces a mechanism to correct the adjustment amount based on the number of elevator runs before using the coupling influence matrix to output adjustment amounts or warning conclusions. The system first obtains the latest update time of the coupling influence matrix and retrieves the number of elevator runs from the elevator controller count or platform operation logs to the current time. The number of elevator runs refers to the cumulative count of valid start-stop / round-trip operations completed by the elevator within that time interval, used to characterize the validity of the coupling matrix from the present. The number of elevator runs is compared with a preset maximum compliance value for elevator runs, which defines the upper limit of the number of runs that can be directly used without significant distortion of the coupling influence matrix. If the number of elevator runs is less than the maximum compliance value, no correction is needed for any adjustment amount. If the number of elevator runs is not less than the maximum compliance value, the ratio of the number of elevator runs to the maximum compliance value is processed, and the result is marked as an adjustment coefficient. Each adjustment amount is then corrected based on this adjustment coefficient, i.e., each adjustment amount is multiplied by the adjustment coefficient.

[0075] For example, if the latest update time of the coupling influence matrix for a certain elevator is the time when maintenance was completed last week, and the platform statistics show that it has been run a total of 1200 times since that update time, while the preset maximum compliance value for elevator operation is 800 times, then the calculated adjustment coefficient is 1.5. The system will conservatively correct the abnormal rope based on the original trial adjustment suggestion (standardized step size) obtained from solving the coupling influence matrix: assuming the original solution suggests tightening the second rope by 1.2 standard steps, loosening the fourth rope by 0.8 standard steps, and tightening the first rope by 0.6 standard steps, then the corrected values ​​become 1.8, 1.2, and 0.9 standard steps respectively; where the standard step size has been fixed through a mapping table as 1 standard step size = 1 / 4 turn of the nut, and the pitch of this model of adjusting nut is 2mm / turn, therefore... The final implementation can be as follows: tighten the second rope by 0.45 turns (approximately 0.90mm axial displacement), loosen the fourth rope by 0.3 turns (approximately 0.6mm axial displacement), and tighten the first rope by 0.225 turns (approximately 0.45mm axial displacement). Based on this, the platform triggers an anomaly warning output. On the one hand, it uses the reason code that the number of runs has exceeded the compliance limit, resulting in the adjustment suggestion being downgraded, and prompts the need to adopt a small-step, multi-round, retesting strategy for each round. On the other hand, it checks whether the corrected adjustment amount is still executable by comparing the maximum allowable rotation amount per run (e.g., 0.25 turns / run) with the minimum executable rotation amount (e.g., 0.05 turns). If it exceeds the limit or falls below the minimum executable amount, it directly prompts that it needs to be split into multiple runs / extended static retesting before recalculation, thereby using the corrected adjustment amount to drive traceable warnings and handling suggestions.

[0076] Figure 4This is a schematic diagram of the state recognition and early warning process provided in this application embodiment. The tension deviation value of each elevator wire rope is used to locate and mark elevator wire ropes in abnormal states, thus completing the abnormality early warning. For elevator wire ropes in abnormal states, the corresponding adjustment amount is solved using the constructed coupling influence matrix. The number of elevator runs is obtained and it is determined whether it meets the preset maximum compliance value for elevator runs. If the number of elevator runs is less than the maximum compliance value, it indicates compliance, and no adjustment amount needs to be corrected. If the number of elevator runs is not less than the maximum compliance value, it indicates non-compliance, and the adjustment amount is corrected to complete the final abnormality early warning.

[0077] In Example 2, under the same conditions as in Example 1, a weighted smoothing replacement is used to handle isolated spikes appearing in the local neighborhood. Specifically, the system first calculates a neighborhood reference level within a 5-point neighborhood window and determines whether the current point is an out-of-threshold point. An out-of-threshold point refers to a deviation of the current point from the neighborhood reference level that exceeds a preset threshold (e.g., exceeding 40N, which usually corresponds to pulse interference introduced by instantaneous sensor jitter, light touch or slight slip of the rope end). To avoid misleading the subsequent selection of stable segments due to single-point jumps, the system replaces the sampled value of the out-of-threshold point with the weighted smoothed value of the non-abnormal point in the neighborhood, thereby suppressing impulse noise while preserving the overall trend.

[0078] For example, the weighted smooth replacement is obtained as follows: within the 5-point neighborhood of the current point, sampling points that have not been judged as exceeding the threshold are selected as effective neighborhood points, and weights are assigned according to the principle that the closer to the current point, the greater the weight. Then, a weighted average is performed on the effective neighborhood points to obtain the replacement value. The weights can be preset as a symmetrical discrete weight sequence. For example, weights (0.1, 0.2, 0.4, 0.2, 0.1) are assigned to the five positions (k-2, k-1, k, k+1, k+2). When the sampling point corresponding to a certain position is judged as exceeding the threshold, it is removed from the effective neighborhood point set, and the remaining weights are renormalized before the weighted average is calculated. The final mean is then filled into the corresponding position.

[0079] Through the above processing, Example 2 can smooth out local jumps after equal-interval resampling, so that the tension sequence presents a continuous small-amplitude fluctuation pattern in the static acceptance scenario, reducing the misjudgment of rate of change exceeding the threshold and the fragmentation of stable segments caused by instantaneous noise; at the same time, since the weighted smoothing adopts the method of higher weight of nearest neighbor points, it can suppress peaks while maintaining the slow convergence or small drift trend of tension as much as possible, providing a more stable and repeatable input basis for subsequent stable candidate segment screening, tail interval positioning and tension result calculation.

[0080] Figure 5 The first set of tension timing processing comparison curves provided in the embodiments of this application are as follows: Figure 6The second set of tension time-series processing comparison curves provided in this application embodiment is shown. The horizontal axis represents time in seconds (s), and the vertical axis represents tension value in Newtons (N). The black curve in both figures represents the original discrete tension sequence directly output and uploaded by the sensor. It is affected by factors such as transmission disorder, missing segments, transient disturbances, and isolated spikes. Obvious outliers and fluctuations can be seen in the curve. The red curve represents the effective tension sequence formed after the original sequence has been preprocessed according to the disclosure process, including time-series restoration by timestamp / serial number, marking of missing segments, segmentation of continuous segments by missing segments, equal-interval resampling to unify the time base, spike suppression, and replacement / smoothing of outliers. Therefore, the red curve is calculated from the black curve through the above steps. It is smoother overall and fluctuates slightly around the stable level, which is used for subsequent steady-state identification, fitting, and deviation statistical calculation.

[0081] It should be noted that, and specifically, all data cited in this embodiment are exemplary and are only used to clearly illustrate the implementation logic and application principle of the technical solution, and do not constitute a limitation on the range of values ​​for the relevant technical parameters.

[0082] The various threshold parameters preset in this embodiment were all scientifically formulated by relevant professional and technical personnel with backgrounds in elevator operation and maintenance, wire rope mechanical property analysis, and Internet of Things monitoring technology. They were combined with the elevator's rated load, wire rope model and specifications, operating condition requirements, and historical monitoring data, and were verified through theoretical calculations, on-site calibration tests, and multiple sets of simulated operating conditions. This ensures that the parameter settings not only conform to the actual operating characteristics of the elevator equipment, but also meet the requirements of accurate identification of wire rope tension status and timely early warning response.

[0083] The embodiments described herein have been described in sufficient detail to enable those skilled in the art to practice the disclosed teachings. Other embodiments may be used and derived therefrom, such that structural and logical substitutions and changes may be made without departing from the scope of this disclosure. Therefore, the detailed description should not be construed as limiting, and the scope of the various embodiments is defined only by the appended claims and the full scope of their equivalents.

Claims

1. A method for identifying and warning of elevator wire rope tension status based on the Internet of Things, characterized in that, Includes the following steps: Step 1: Obtain the discrete sequence of elevator wire rope tension uploaded by the data acquisition device through the Internet of Things, restore it according to the time stamp sequence and identify missing segments to update the discrete sequence of tension. Step 2: Perform sequence segmentation at the sequence number jump points of the tension discrete sequence. After segmentation, mark them as several transmission continuous segments. Perform equal-interval resampling and peak suppression on each transmission continuous segment to form an analyzable sequence. Step 3: Calculate the tension change rate for each transmission continuous segment. Based on the tension change rate, divide each transmission continuous segment into a few unstable segments and several stable segments. If there are not enough stable segments, the average value method is used to obtain the tension result of the elevator wire rope. Otherwise, the convergence limit is obtained by using the exponential stabilization model as the tension result of the elevator wire rope. The division of each transmission continuous segment into several stable segments based on the tension change rate is as follows: The discrete tension sequence of each transmission continuous segment is statistically obtained, and the tension change rate sequence corresponding to each transmission continuous segment is calculated in turn to characterize the instantaneous dynamic disturbance intensity. When there is a tension change rate sequence where the absolute value of the tension change rate exceeds a preset first change rate threshold, the tension change rate is marked as a disturbance candidate point. Furthermore, a continuous length merging rule is introduced to segment and shape the disturbance candidate points. Specifically, if the interval between adjacent disturbance candidate points is less than or equal to a preset interval threshold, they are merged into the same disturbance segment. The remaining intervals in each transmission continuous segment that are not marked as disturbance segments are the stable candidate segments, thus dividing each transmission continuous segment into a few disturbance segments and several stable segments. Each stable candidate segment is statistically analyzed from each continuous transmission segment, and a screening process is performed on each stable candidate segment; The specific screening process for each stable candidate segment is as follows: Select several stable candidate segments with a window length not less than the preset minimum allowed window length, and mark them as stable segments; The number of stable segments is counted and compared with the preset minimum allowed number of stable segments; If the number of stable segments is less than the minimum allowed number of stable segments, i.e. there are not enough stable segments, the tension result of the elevator wire rope is obtained through the first analysis method. The first analysis method specifically refers to: averaging the discrete tension sequences of each stable segment, and marking the result as the tension result of the elevator wire rope. If the number of stable segments is not less than the minimum allowable number of stable segments, the tension result of the elevator wire rope is obtained through the second analysis method. The second analysis method specifically refers to: eliminating each disturbance segment, automatically locating the tail section of the rebound decay based on the preset second rate of change threshold, fitting the tension sequence of the tail section into an exponential stabilization model, and using the convergence limit obtained from the fitting as the tension result of the elevator wire rope. Step 4: Analyze the median of the tension results of each elevator wire rope as the tension reference value, analyze the tension deviation sequence, locate several abnormal elevator wire ropes and issue warnings, solve the corresponding adjustment amount for several abnormal elevator wire ropes through the already constructed coupling influence matrix, and correct each adjustment amount in combination with the number of elevator runs.

2. The method for identifying and warning of elevator wire rope tension status based on the Internet of Things as described in claim 1, characterized in that: The specific update process for the updated tension discrete sequence is as follows: The elevator wire rope tension data collected by the data acquisition device are integrated, marked as the elevator wire rope tension discrete sequence, and packaged into a data package. When the data acquisition device uploads data packets to the data processing terminal, it marks the timestamp in the data packets; The data processing end performs time-series reconstruction of the discrete tension sequence in the data packets according to the timestamp order and identifies missing segments, thereby updating the discrete tension sequence of the elevator wire rope and reducing the interference caused by network out-of-order or packet loss on the discrete tension sequence analysis. After the tension discrete sequence of the elevator wire rope is updated, a cutting strategy is executed on the tension discrete sequence of the elevator wire rope.

3. The method for identifying and warning of elevator wire rope tension status based on the Internet of Things as described in claim 2, characterized in that: The specific process of implementing the cutting strategy on the discrete sequence of tension of the elevator wire rope is as follows: The cutting strategy refers to locating the jump interval when there is a jump in the sequence number of adjacent sequences in the tension discrete sequence, determining the jump interval as a gap caused by network packet loss or delay, and cutting the tension discrete sequence into several transmission continuous segments at the gap. For each continuous transmission segment, a time reference correction process is introduced. If the timestamps of the continuous transmission segment are not of equal interval, the tension discrete sequence is resampled at equal intervals according to the timestamps to restore the unified sampling rhythm. After resampling, isolated spikes appearing in the local neighborhood are replaced.

4. The method for identifying and warning of elevator wire rope tension status based on the Internet of Things as described in claim 1, characterized in that: The median of the tension results of each elevator wire rope is used as the tension reference value. The method of obtaining the tension results of each elevator wire rope and the corresponding method of obtaining them are also included. If the tension results of elevator wire ropes below the defined number are obtained using the second analysis method, then the tension status of the elevator wire ropes will be identified directly based on the tension results of each elevator wire rope. If the tension result of elevator wire ropes with a number of ropes not less than the defined number is obtained through the second analysis method, then the elevator wire ropes whose tension result is obtained through the second analysis method will be marked as each elevator wire rope to be analyzed. Obtain the fitting error value of the tension result of each elevator wire rope to be analyzed, and process the standard deviation. The processing result is marked as the fitting error deviation value. The fitting error deviation value is compared with the preset maximum compliance value for fitting error deviation; Collect and, based on the comparison results, determine whether to update the tension results of each elevator wire rope to be analyzed.

5. The method for identifying and warning of elevator wire rope tension status based on the Internet of Things as described in claim 4, characterized in that: The specific process for determining whether to update the tension results of each elevator steel wire rope to be analyzed is as follows: If the deviation of the fitting error is lower than the maximum compliance value of the fitting error deviation, it is determined not to update the tension results of each elevator wire rope to be analyzed. Based on the tension results of each elevator wire rope to be analyzed, the tension results of each elevator wire rope are updated, and the tension status of the elevator wire rope is identified based on the tension results of each elevator wire rope. Under the condition that the deviation value of the fitting error is not lower than the maximum compliance value of the fitting error deviation, the tension results of each elevator wire rope to be analyzed are updated. Based on the deviation value of the fitting error, the preset second rate of change threshold is reduced, and the tension results of each elevator wire rope to be analyzed are updated. Based on the updated tension results of each elevator wire rope to be analyzed, the tension results of each elevator wire rope are updated again. Based on the tension results of each elevator wire rope, the tension state of the elevator wire rope is identified.

6. The method for identifying and warning of elevator wire rope tension status based on the Internet of Things as described in claim 1, characterized in that: The process of locating and issuing warnings for several abnormal elevator wire ropes is as follows: The median of the tension results for each elevator wire rope is located and marked as the tension reference value; The tension results of each elevator wire rope are compared with the tension reference value to obtain the tension deviation value of each elevator wire rope. Locate several elevator wire ropes whose absolute value of tension deviation exceeds the preset maximum tension deviation compliance value, and mark them as elevator wire ropes in abnormal condition.

7. The method for identifying and warning of elevator wire rope tension status based on the Internet of Things as described in claim 1, characterized in that: The coupling influence matrix that has been constructed specifically refers to: In the elevator wire rope tension test experiment, the coupling influence coefficient is analyzed by the ratio of the change in tension of each wire rope before and after the trial adjustment to the step size of the trial adjustment. The trial adjustment step size is the fine adjustment amount obtained by converting the adjustment action applied to the elevator wire rope into a unitary measurement value according to the preset conversion rule of the adjustment action to the equivalent displacement. The coupling influence coefficient is used to characterize the cascading effect of adjusting a certain wire rope on the tension of itself and other remaining elevator wire ropes. According to the mapping relationship between the adjusted object and the affected object, the coupling influence coefficients obtained from each trial adjustment are filled in one by one to form a coupling influence matrix.

8. The method for identifying and warning of elevator wire rope tension status based on the Internet of Things as described in claim 1, characterized in that: The adjustment parameters are corrected based on the number of elevator runs. The specific correction process is as follows: Obtain the latest update time of the coupling influence matrix, and obtain the number of elevator runs between this latest update time and the current time. The number of elevator runs is compared with the preset maximum compliant value for the number of elevator runs; If the number of elevator runs is less than the maximum compliant value for elevator runs, there is no need to correct any of the adjustment values. Under the condition that the number of elevator runs is not less than the maximum compliant value of elevator runs, the ratio of the number of elevator runs to the maximum compliant value of elevator runs is processed, and the processing result is marked as the adjustment coefficient. Based on the adjustment coefficient, each adjustment quantity is corrected, and abnormal warnings are issued through the corrected adjustment quantities.

Images