Elevator brake friction performance detection method based on multi-data fusion

Abstract

The application belongs to the technical field of elevator brake detection, and specifically discloses an elevator brake friction performance detection method based on multi-data fusion. A sensing unit is integrated in the elevator brake, and the time sequence signals of three parameters, i.e., braking force, brake shoe thickness and coil temperature, are synchronously collected during braking. Meanwhile, the braking process is divided into time periods, the main parameters and response parameters are screened in each time period through response cross correlation, the period-oriented fine detection of friction performance is realized, and the physical matching degree of the detection result and the period-specificity are greatly improved. Then, the detection results of the friction performance of each period are input into a multi-factor coupling judgment model for period anomaly recognition, and combined with the physical role of the period, the abnormal period coverage and the grading diagnosis result are output, so that the specific link of performance degradation can be accurately located, thereby assisting maintenance personnel to quickly formulate a targeted maintenance strategy.

Description

Technical Field

[0001] This invention belongs to the field of elevator brake testing technology, and specifically discloses a method for testing the friction performance of elevator brakes based on multi-data fusion. Background Technology

[0002] Elevator brakes are the primary component for safe elevator operation. As the frequency of elevator operation increases, the friction materials of the brakes inevitably experience wear, thermal degradation, and other performance deterioration. If this is not detected in time, it may lead to safety accidents such as slippage or overshooting. Therefore, it is essential to test and evaluate the friction performance of elevator brakes.

[0003] The elevator braking process consists of multiple consecutive physical time periods, during which multiple physical quantities such as braking force, wear, and temperature are generated. These physical quantities form a complete braking timing chain through coupled operation.

[0004] Existing technologies already include detection schemes that utilize multi-parameter fusion. For example, Chinese invention patent CN112141843B discloses a dynamic detection system and method for detecting the braking performance of elevator brakes. By collecting dynamic parameters such as brake shoe clearance, friction pad temperature, braking deceleration, and response time during elevator operation, the host computer compares the characteristic values ​​of the dynamic parameters with the theoretical values ​​calculated through simulation using a three-dimensional dynamic model of the elevator to determine whether the brake has failed. Furthermore, a Bayesian linear model is used to perform multivariate fusion of the various dynamic parameters to calculate performance degradation parameters, and the remaining life of the brake is predicted based on a degradation state space model.

[0005] While the aforementioned scheme achieves comprehensive evaluation of multiple parameters, its core lies in fusing global parameters through a statistical model. However, braking is a process-oriented action, and the coupling relationship between different physical quantities does not remain constant throughout the entire process. Global analysis tends to mix physical quantities from different time periods together, making it difficult to highlight the contribution of the dominant physical quantities in each time period. Furthermore, performing equal analysis on less influential physical quantities introduces unrepresentative interference, resulting in a coarse analysis of friction performance that fails to accurately match the actual physical degradation process.

[0006] Furthermore, the test results of the above-mentioned scheme are only a binary judgment of whether it has failed or the remaining life value, which makes it difficult to pinpoint the specific link of performance degradation and is not conducive to maintenance personnel quickly developing targeted maintenance strategies. Summary of the Invention

[0007] In view of the shortcomings of the prior art, the present invention provides a method for detecting the friction performance of elevator brakes based on multi-data fusion, which is used to solve the problems existing in the prior art.

[0008] The objective of this invention can be achieved through the following technical solution: a method for detecting the friction performance of elevator brakes based on multi-data fusion, comprising the following steps: synchronously acquiring the timing signals of braking force, brake shoe gap, and coil temperature during the entire braking process of the elevator brake.

[0009] The timeline of a single braking action is divided into multiple time periods according to the braking process.

[0010] Within each time period, fluctuation points are extracted from the time-series signals of three parameters: braking force, brake shoe gap, and coil temperature. The master-slave relationship of the parameter pairs is determined by pairwise combinations. The response coupling degree is calculated by combining the response interval and waveform area ratio between the parameter pairs. Then, the master parameter and response parameter are selected based on the master-slave relationship and response coupling degree of the parameter pairs. A braking performance template is constructed, and the joint waveform segment of the master parameter and response parameter in each time period is extracted. The deviation of the joint waveform segment from the braking performance template is analyzed as the friction performance degradation amount in each time period.

[0011] Input the amount of friction performance degradation at each time period into the multi-factor coupled judgment model, and output the brake friction performance status diagnosis result.

[0012] Combining all the above technical solutions, the positive effects of this invention are as follows: 1. This invention simultaneously collects the timing signals of three parameters—braking force, brake shoe gap, and coil temperature—during the braking process of an elevator brake, and divides the braking process into multiple time periods. Within each time period, the main parameters and response parameters are selected through the causal response relationship between the parameters. This allows the friction performance detection to obtain targeted analysis objects for each time period. Compared with global analysis of each parameter, this effectively avoids feature submersion and interference amplification caused by the aliasing of physical quantities in different time periods, and achieves a deep match between the detection process and the braking physical mechanism, thereby improving the precision level of friction performance evaluation.

[0013] 2. This invention identifies time-period anomalies by inputting the test results of friction performance at each time period into a multi-factor coupled judgment model, and outputs the coverage of abnormal time periods and graded diagnostic results by combining the physical role of each time period. This can accurately locate the specific links of performance degradation, thereby assisting maintenance personnel in quickly formulating targeted maintenance strategies. Attached Figure Description

[0014] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0015] Figure 1 This is a diagram illustrating the implementation steps of the method of the present invention;

[0016] Figure 2This is a schematic diagram illustrating the division of the braking process into time periods in this invention;

[0017] Figure 3 This is a flowchart illustrating the implementation of step S3 in this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] See Figure 1 As shown, the present invention proposes a method for detecting the friction performance of elevator brakes based on multi-data fusion, including the following steps: S1, synchronously collecting the braking force, brake shoe gap and coil temperature timing signals during the entire braking process of the elevator brake.

[0020] When the elevator brake is engaged, the electromagnetic coil is de-energized, and the brake spring pushes the brake arm to press the brake shoes against the brake wheel, generating braking force to stop the elevator from moving. The continuous friction between the brake shoes and the brake wheel causes the brake shoe material to wear gradually, which is manifested as an increase in the gap between the brake shoes and the brake wheel when the brake is released. At the same time, the heat generated during the friction process raises the temperature of the brake coil. In the whole process, three physical quantities are generated: braking force, brake shoe gap, and coil temperature. These three physical quantities reflect the force applied by the brake, the degree of wear, and the thermal load level, and are the main indicators for assessing the degree of friction performance degradation.

[0021] S2. Divide the time axis of a single braking action into multiple time periods according to the braking process.

[0022] The braking process of an elevator brake is not a single transient event, but rather goes through multiple physical time periods in sequence. Dividing the braking process into multiple time periods according to the physical mechanism can decompose the overall braking behavior into sub-processes with clear functions, providing a time basis for subsequent time-segmented analysis of friction performance.

[0023] See Figure 2 As shown, in a preferred embodiment of the present invention, the time period division process is as follows: S21, the time when the electromagnetic coil de-energization command is issued is taken as the starting zero point of the braking action.

[0024] S22. Perform first-order differential processing on the braking force timing signal to obtain the braking force change rate signal. Starting from the initial zero point, search backward to detect the first zero-crossing point in the braking force change rate signal that changes from zero to negative. Record the time corresponding to the zero-crossing point as the braking force decrease time point.

[0025] S23. The period between the initial zero point and the time point when the braking force decreases is defined as the electromagnetic release period. This is because there is an electrical and magnetic circuit response time between the de-energization of the electromagnetic coil and the start of the decrease in braking force. This period mainly reflects the release delay of the electromagnetic system and does not involve mechanical movement. Therefore, it is defined as the electromagnetic release period.

[0026] S24. Perform first-order differential processing on the brake shoe gap timing signal to obtain the gap change rate signal. Starting from the braking force reduction time point, search backwards to detect the first peak point in the gap change rate signal that exceeds the preset jump threshold. Record the time corresponding to the peak point as the brake shoe gap elimination time point, indicating that the gap between the brake shoe and the brake wheel has been completely eliminated and contact has begun. The jump threshold is set in the following way: Perform first-order differential processing on the brake shoe gap timing signal to obtain the gap change rate signal. In a static signal with a preset duration, such as 0.5 seconds, before the zero point of the start of this braking action, calculate the amplitude standard deviation and use a multiple of the amplitude standard deviation as the jump threshold.

[0027] S25. The time period between the point when the braking force decreases and the point when the brake shoe clearance is eliminated is defined as the mechanical action period. This is because from the time when the braking force begins to decrease until the brake shoe first contacts the brake wheel, the brake arm overcomes the clearance and static friction under the action of the spring force and generates displacement. This process belongs to the pure mechanical transmission stage and does not generate effective braking force. Therefore, it is defined as the mechanical action period.

[0028] S26. Continue searching the braking force change rate signal from the time point when the brake shoe gap is eliminated. Detect the zero-crossing point where the signal first drops from a positive value to zero. Record the time corresponding to the zero-crossing point as the steady-state peak braking force time point, indicating that the braking force has reached the steady-state peak.

[0029] S27. The period between the time point when the brake shoe clearance is eliminated and the time point when the braking force reaches its steady-state peak value is defined as the contact pressure building period. This is because during this period, the brake shoe has contacted the brake wheel, and the braking force rises rapidly from zero to its peak value, reflecting the brake's pressure building capability.

[0030] S28. The time period between the steady-state peak time of the braking force and the time when the command to re-energize the electromagnetic coil is issued is defined as the braking maintenance period.

[0031] S29. The time period between the moment the command to re-energize the electromagnetic coil is issued and the moment when the braking force returns to zero and the brake shoe clearance returns to the initial value before braking is defined as the brake release recovery period. This is because after the electromagnetic coil is re-energized, the electromagnetic force overcomes the spring force to retract the brake arm, the braking force is gradually released, the brake shoe disengages from the brake wheel, and the brake shoe clearance returns to the initial clearance value before braking. This initial clearance will slowly increase with the long-term wear of the brake shoe, but the same initial value is restored during a single braking process. This period reflects the brake's reset capability.

[0032] S3. Within each time period, extract fluctuation points from the timing signals of the three parameters: braking force, brake shoe gap, and coil temperature. Determine the master-slave relationship of the parameter pairs by combining them in pairs. Calculate the response coupling degree by combining the response interval and waveform area ratio between the parameter pairs. Then, based on the master-slave relationship and response coupling degree of the parameter pairs, filter the master parameter and response parameter.

[0033] Although the braking force, brake shoe clearance, and coil temperature change dynamically throughout the braking process, the physical mechanisms involved differ at different times. For example, the electromagnetic release period is dominated by electromagnetic pressure conversion, the mechanical action period is dominated by pressure displacement conversion, and the braking maintenance period is dominated by pressure-wear-temperature coupling. This results in not all parameters being prominent at all times, and the parameters do not change independently. Since the essence of the braking process is the conversion of energy from electromagnetic energy to mechanical energy, and then to frictional heat and wear, there is a clear excitation-response relationship at each step. This makes there a causal drive between parameters. Under the causal drive, the parameters cooperate to complete the braking. Analyzing parameters independently without the causal drive can easily cause the analysis of friction performance to lose physical compatibility.

[0034] Therefore, it is necessary to select the main parameters and response parameters that are suitable for the current physical process in each time period in order to accurately characterize the braking characteristics of the time period.

[0035] See Figure 3 As shown, the specific implementation process is as follows: S31. The master-slave relationship between different parameters implies a physical causal driving relationship. By identifying the master-slave relationship of parameter pairs at each time period, erroneous analysis caused by reversing the causal relationship can be avoided. Specifically, this includes the following sub-steps: S311. Considering that after the brake operates, these three parameters have continuous waveform changes throughout the entire process, the master-slave relationship between parameters cannot be simply determined as a fixed temporal sequence. Instead, the dynamic fluctuations of the parameters must be observed. When a parameter fluctuates, if there is a causal driving relationship between this parameter and other parameters, then the fluctuation of this parameter will inevitably drive the other parameters to produce corresponding fluctuations. This temporal characteristic of fluctuation first and then response can clearly reflect the master-slave relationship.

[0036] Therefore, for each time period, the fluctuation points of each parameter are extracted from the timing signals of the braking force, brake shoe gap and coil temperature.

[0037] Specifically, the timing signals of braking force, brake shoe gap, and coil temperature are processed by first-order differentiation. The differentiated signals can amplify the transient changes of the signals and highlight the precise moment when the fluctuations occur.

[0038] Extreme points and zero-crossing points are detected in each differential signal. The extreme points of the first derivative correspond to the moment when the rate of change of the original signal is the largest, and the zero-crossing points correspond to the turning points of the original signal. These points are characteristic locations where the signal state changes. The time corresponding to the detected extreme points or zero-crossing points in the original signal is taken as the fluctuation points of each parameter.

[0039] S312. Combine the three parameters in pairs to obtain three parameter pairs: braking force-brake shoe gap, braking force-coil temperature, and brake shoe gap-coil temperature. Since the master-slave relationship needs to be analyzed later, the order of parameter pair formation is not considered here. For each parameter pair, the temporal relationship between the fluctuation points of the two is identified, and the parameter whose fluctuation point occurs first is used as the candidate master parameter, and the parameter whose fluctuation point occurs later is used as the candidate response parameter.

[0040] S32. The master-slave relationship alone can only determine the existence of a causal relationship, but it cannot quantify the quality of the response parameter's following of the master parameter, i.e., the strength of the causal drive. Therefore, by combining the response interval and waveform area ratio between parameter pairs to calculate the response coupling degree, the strength of the causal drive between parameter pairs can be effectively characterized.

[0041] Specifically, the following sub-steps are included: S321. For each parameter pair, two parameter waveform segments are extracted within the response interval between the fluctuation point of the candidate principal parameter and the fluctuation point of the candidate response parameter. The waveform segments reflect the dynamic behavior of the two within the causal interval from the excitation of the principal parameter to the reaction of the response parameter. Then, the absolute area of ​​the region enclosed by the waveform curve of the candidate principal parameter and the time axis and the absolute area of ​​the region enclosed by the waveform curve of the candidate response parameter and the time axis are statistically analyzed. These two areas respectively represent the total excitation of the principal parameter within the response interval and the total response of the response parameter within the response interval. The area statistics can be obtained by time integration of the waveform curve.

[0042] S322. Calculate the ratio of the absolute area of ​​the region enclosed by the candidate principal parameter waveform curve and the time axis to the absolute area of ​​the region enclosed by the candidate response parameter waveform curve and the time axis to obtain the energy transfer factor. This factor reflects the energy utilization efficiency or response intensity of the response parameter to the principal parameter excitation. The closer the ratio is to 1, the more fully the response parameter follows the changes of the principal parameter.

[0043] S323. The response interval represents the delay time of the response parameter in response to the excitation of the main parameter. The shorter the response interval, the faster the response parameter follows the main parameter. Therefore, the reciprocal of the response interval is taken as the time coupling factor.

[0044] S324. Multiply the time coupling factor and energy transfer factor after normalization to obtain the response coupling degree of the parameter pair. This represents the overall quality of the collaborative work between the main and secondary parameters in the current time period. The higher the value, the faster and more fully the response parameter can respond to changes in the main parameter.

[0045] Since the time coupling factor and energy transfer factor are normalized to a value between 0 and 1, multiplication calculations make it so that if either factor is too low, the coupling degree will be significantly reduced. Only when both values ​​are large can the coupling degree reach a high level.

[0046] S33. Filter the master and slave relationship and response coupling degree of the fusion parameter pairs to select master parameters and response parameters. Specifically, it includes the following sub-steps: S331. Count the total number of times each parameter is marked as a candidate master parameter in each time period. If a parameter is determined to occur first in multiple parameter pairs, it means that it is dominant in physical time sequence and is most likely to become the driving source of that time period. Thus, the parameter with the highest total number of occurrences is determined as the master parameter of the corresponding time period.

[0047] S332. The other parameter that is paired with the main parameter and has the highest response coupling degree is determined as the response parameter of the time period. This means that among the multiple candidate response parameters paired with the main parameter, the one with the best causal driving quality is selected as the representative response parameter of the time period. The main parameters and response parameters of each time period selected in this way are the most prominent in the braking participation characterization of the time period, providing a targeted analysis object for subsequent braking performance analysis.

[0048] In an example of the above operation, assuming that during the contact pressure build-up period, the three parameter pairs formed—braking force-brake shoe gap, braking force-coil temperature, and brake shoe gap-coil temperature—are judged according to the order of their fluctuation points: the braking force fluctuation point precedes the brake shoe gap fluctuation point, and the master-slave relationship of the braking force-brake shoe gap parameter pair is that the braking force is the candidate master parameter and the brake shoe gap is the candidate response parameter.

[0049] The braking force fluctuation point precedes the coil temperature fluctuation point. The master-slave relationship of the braking force-coil temperature parameter pair is that the braking force is the candidate master parameter and the coil temperature is the candidate response parameter.

[0050] The brake shoe gap fluctuation point precedes the coil temperature fluctuation point. The master-slave relationship of the brake shoe gap-coil temperature parameter pair is that the brake shoe gap is the candidate master parameter and the coil temperature is the candidate response parameter.

[0051] Because the physical process during the contact pressure build-up period involves the rapid increase of braking force driving the friction interface to generate heat, the temperature responds quickly to the braking force and has high energy transfer efficiency, thus the braking force-coil temperature pair has the highest response coupling degree. In the braking force-brake shoe gap pair, the response of wear accumulation to braking force fluctuations is relatively lagging and there is mechanical loss in energy transfer, resulting in the second highest coupling degree. In the brake shoe gap-coil temperature pair, both are response parameters and lack direct causal driving, resulting in the lowest coupling degree.

[0052] The primary parameter selected is the braking force, and the response parameter is the coil temperature. At this time, the brake is in the rapid pressure build-up phase of the braking force. The change in braking force is the excitation source, and the heat generated by friction causing the coil temperature to rise is the main response characteristic. Therefore, using the braking force-coil temperature coupling as the representative analysis object for this period can effectively reflect the thermal response characteristics and pressure build-up capability of the friction material.

[0053] Specifically, in S3 operations, when multiple fluctuation points exist between parameters within a certain time period, these fluctuation points need to be paired. Specifically: First, the fluctuation points of two parameters are matched one by one in chronological order: using a fluctuation point of one parameter as a baseline, the fluctuation point of the other parameter with the closest subsequent time is selected as the pairing object, forming a fluctuation point pair. If a fluctuation point appears and no corresponding fluctuation point of another parameter is found within the limited time window, it is considered an isolated point and ignored, not participating in subsequent calculations.

[0054] Secondly, after forming fluctuation point pairs, the master-slave relationship is determined for each pair, and the response interval and energy transfer factor are calculated to obtain the response coupling degree of that pair. Finally, the response coupling degrees of all fluctuation point pairs are combined, and the arithmetic mean or median is taken as the final response coupling degree for that time period. The above processing can effectively avoid the distortion of coupling degree calculation caused by disordered time sequence or inconsistent number of fluctuation points.

[0055] S4. Construct a braking performance template and extract the joint waveform segments of the main parameters and response parameters in each time period. Analyze the degree of deviation of the joint waveform segments from the braking performance template as the friction performance degradation amount in each time period.

[0056] After selecting the main parameters and response parameters for each time period, the performance of the main parameters and response parameters during braking can be used to evaluate the friction performance of the brake. However, the evaluation requires a reference standard. Therefore, it is necessary to first construct a braking performance template to represent the performance of the parameters of the brake in a healthy state. Then, the actual performance of the current braking action is compared with the template to accurately assess the degree of deviation of the braking state.

[0057] As a preferred embodiment of the present invention, the specific process is as follows: First stage: Constructing a braking performance template.

[0058] (11) After the brake has completed its break-in period, data from multiple consecutive braking actions are selected as a learning sample. This is because the brake components have not yet reached a stable working state during the break-in period, and the template established at this time cannot represent normal performance. After the break-in period is completed, the brake enters a stable operating stage, and only the template established in this way can serve as a reliable health benchmark.

[0059] (12) For each group of learning samples, waveform segments of the main parameter signal and the response parameter signal are extracted in each time period. The waveform is chosen as the braking performance because the waveform fully records all the details of the parameter change over time, including amplitude, shape, rise / fall rate, etc., which can comprehensively depict the performance of the parameter during the braking process.

[0060] (13) Average the principal parameter waveforms of all learning samples in the same period point by point to obtain the template principal parameter waveform of that period; average the response parameter waveforms of all learning samples in the same period point by point to obtain the template response parameter waveform of that period. This can eliminate individual differences and random interference by using statistical averaging, and construct a standard waveform that can represent the normal state of that period as a reference for subsequent comparison.

[0061] Second stage: By comparing and analyzing the deviation through templates, the difference between the actual waveform and the template in each time period during the current braking action is quantified. The specific process is as follows: (21) Extract waveform segments of the main parameter signal and the response parameter signal in each time period to form a joint waveform pair.

[0062] (22) Taking the starting zero point of the braking action as a common time reference, since the total duration of each braking action is not much different, the actual waveform of the main parameter and the template main parameter waveform are aligned point by point according to the same time point on the time axis. If necessary, the actual waveform time axis and the template waveform time axis can be linearly scaled and normalized before alignment, so that the two waveforms can achieve synchronous matching on the process time axis of the braking action. Calculate the area enclosed between the actual waveform curve of the main parameter and the template main parameter waveform curve, and record it as the comparison area of ​​the main parameter waveform.

[0063] When the waveform curves completely overlap, the enclosed area is 0. The presence of an enclosed area indicates a deviation in shape between the actual waveform and the standard waveform. This area reflects the degree of distortion of the main parameter's waveform shape relative to a healthy state during that time period; the larger the area, the more severe the deviation. The area is calculated by integrating the absolute value of the difference between the two waveforms over time.

[0064] (23) Similarly, the actual waveform of the response parameter in the joint waveform pair is aligned point by point with the template response parameter waveform of the corresponding time period, and the area enclosed between the actual waveform curve of the response parameter and the template response parameter waveform curve is calculated and recorded as the response parameter waveform comparison area.

[0065] (24) The total deviation of the waveform area is obtained by adding the waveform comparison area of ​​the main parameter and the waveform comparison area of ​​the response parameter. This deviation reflects the comprehensive deviation of the waveform shapes of the main parameter and the response parameter. The degradation of friction performance is essentially manifested as the dynamic relationship between the main parameter and the response parameter deviating from the healthy state during braking. The larger the total deviation of the waveform area, the greater the difference between the actual braking behavior and the normal state during that period. Therefore, this deviation can be used as an effective characterization of the amount of friction performance degradation during a period, so as to achieve reasonable quantification of the degradation state at different periods.

[0066] S5. Input the amount of friction performance degradation at each time period into the multi-factor coupled judgment model, and output the brake friction performance status diagnosis result.

[0067] After obtaining the friction performance degradation amount by quantifying the deviation of the main parameters and response parameters from the template at each time period, a refined diagnosis of the brake friction performance can be performed, instead of directly outputting a binary judgment of whether it has failed, thus providing a basis for refined maintenance.

[0068] Specifically, a multi-factor coupled judgment model is first constructed, and the operation is as follows: the amount of friction performance degradation calculated at each time period after the braking action is completed is compared with the abnormal judgment threshold of the corresponding time period. If the amount of friction performance degradation at a certain time period exceeds the abnormal judgment threshold, then the time period is marked as an abnormal time period.

[0069] Since the degradation of brake friction performance is related to the elevator's operating frequency, when determining the abnormal judgment threshold for each time period, for a given model and cumulative operating frequency of elevator brake, a brake of the same model, with a similar cumulative operating frequency, and whose friction performance has been confirmed to be normal through on-site inspection is selected as a reference sample.

[0070] For each reference sample brake, the amount of friction performance degradation at each time period is calculated based on the brake performance template after its break-in period.

[0071] The friction performance degradation of all reference samples under the same year was summarized in the same period, and a frequency histogram was plotted. The degradation value corresponding to the first appearance of the trough to the right of the main peak was identified as the anomaly judgment threshold under the cumulative operating frequency for the corresponding period.

[0072] The number of time periods marked as abnormal during each braking action is recorded as the number of covered time periods, which reflects the breadth of the overall performance degradation of the brake. The more covered time periods, the more extensive the physical components involved in the degradation.

[0073] Then, the brake friction performance status diagnosis results are output using the judgment results, specifically including the following: Although the brake action is divided into multiple time periods, the degree of influence of different time periods on braking varies. The electromagnetic release time period and the brake release recovery time period are classified as peripheral time periods, while the mechanical action time period, the contact pressure building time period, and the brake maintenance time period are classified as dominant time periods. This is because the peripheral time periods mainly reflect the performance of the electromagnetic response and the reset mechanism, and their abnormalities usually do not directly affect the braking capacity; while the dominant time periods directly involve the establishment of braking force and the maintenance of friction, and their abnormalities directly threaten braking safety.

[0074] When the number of covered time periods is zero, it indicates that the braking behavior in all time periods is within the normal range, with no abnormalities, and the output friction performance diagnosis result is normal.

[0075] When the abnormal period only involves the peripheral period, it indicates that the braking control mechanism is still working normally, but there may be a slight abnormality in the electromagnetic response or reset mechanism, resulting in a diagnostic result of abnormality in the peripheral components.

[0076] When the abnormal period includes the dominant period but covers only one period, it indicates that a performance degradation has occurred in a certain period of the braking dominant link, but has not yet spread to multiple links, and outputs a diagnostic result of reduced braking performance.

[0077] When the abnormal period includes the dominant period and the number of covered periods is greater than one, it indicates that multiple periods in the braking dominant link are abnormal, and the performance degradation has shown a spreading trend, outputting a diagnostic result of braking performance decline.

[0078] By diagnosing the friction performance of the elevator brake as described above, the specific points of performance degradation can be accurately located, thereby assisting maintenance personnel in quickly developing targeted maintenance strategies.

[0079] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0080] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0081] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0082] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0083] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for detecting the frictional performance of elevator brakes based on multi-data fusion, characterized in that, Includes the following steps: Synchronously acquire timing signals of braking force, brake shoe gap, and coil temperature throughout the entire braking process of the elevator brake; The timeline of a single braking action is divided into multiple time periods according to the braking process; Within each time period, fluctuation points are extracted from the time-series signals of three parameters: braking force, brake shoe gap, and coil temperature. The master-slave relationship of the parameter pairs is determined by pairwise combination. The response coupling degree is calculated by combining the response interval and waveform area ratio between the parameter pairs. Then, the master parameter and response parameter are selected based on the master-slave relationship and response coupling degree of the parameter pairs. A braking performance template is constructed, and the joint waveform segments of the principal parameter and response parameter in each time period are extracted. The deviation of the joint waveform segments from the braking performance template is analyzed as the friction performance degradation in each time period. Input the friction performance degradation at each time period into the multi-factor coupled judgment model, and output the brake friction performance status diagnosis result; The method of dividing the time axis of a single braking action into multiple time periods according to the braking process includes the following: The moment when the electromagnetic coil is de-energized is taken as the starting zero point of the braking action. The first-order differential processing of the braking force timing signal is performed to obtain the braking force change rate signal. Starting from the initial zero point, the search proceeds backward to detect the first zero-crossing point in the braking force change rate signal that changes from zero to a negative value. The time corresponding to the zero-crossing point is recorded as the braking force decrease time point. The period from the initial zero point to the point where the braking force decreases is defined as the electromagnetic release period; The brake shoe gap timing signal is processed by first-order differentiation to obtain the gap change rate signal. Starting from the time point when the braking force decreases, the first peak point in the gap change rate signal that exceeds the preset jump threshold is detected, and the time corresponding to the peak point is recorded as the brake shoe gap elimination time point. The time period between the point when the braking force decreases and the point when the brake shoe clearance is eliminated is defined as the mechanical action period. The braking force change rate signal is searched backward from the time point when the brake shoe gap is eliminated. The zero-crossing point when the signal first drops from a positive value to zero is recorded as the steady-state peak time point of the braking force. The period between the time point when the brake shoe clearance is eliminated and the time point when the braking force reaches its steady-state peak value is defined as the contact pressure build-up period. The period between the steady-state peak time of braking force and the time when the command to re-energize the electromagnetic coil is issued is defined as the braking maintenance period. The time period between the moment the command to re-energize the electromagnetic coil is issued and the later of the two events—the moment the braking force returns to zero and the moment the brake shoe clearance returns to its initial value before braking—is defined as the brake release recovery period. The process of determining the master-slave relationship of parameter pairs by pairwise combination includes the following steps: Within each time period, the fluctuation points of the three parameters—braking force, brake shoe gap, and coil temperature—are extracted from their respective timing signals. The three parameters are combined in pairs to obtain several parameter pairs. For each parameter pair, the temporal relationship between the fluctuation points of the two parameters is identified. The parameter whose fluctuation point occurs first is used as the candidate principal parameter, and the parameter whose fluctuation point occurs later is used as the candidate response parameter.

2. The elevator brake friction performance testing method based on multi-data fusion as described in claim 1, characterized in that: The fluctuation points are extracted through the following process: The timing signals of braking force, brake shoe gap and coil temperature are processed by first-order differentiation. Extreme points and zero crossings are detected from each differential signal. The time of the extreme point or zero crossing in the original signal is taken as the fluctuation point.

3. The elevator brake friction performance testing method based on multi-data fusion as described in claim 1, characterized in that: The process for calculating the response coupling degree by combining the response interval and waveform area ratio between parameter pairs is as follows: For each parameter pair, two parameter waveform segments are extracted within the response interval between the fluctuation points of the candidate principal parameter and the fluctuation points of the candidate response parameter. The ratio of the absolute area of ​​the region enclosed by the candidate principal parameter waveform curve and the time axis to the absolute area of ​​the region enclosed by the candidate response parameter waveform curve and the time axis is calculated to obtain the energy transfer factor. Take the reciprocal of the response interval as the time coupling factor; The response coupling degree of the parameter pair is obtained by multiplying the time coupling factor and the energy transfer factor after normalization.

4. The elevator brake friction performance testing method based on multi-data fusion as described in claim 3, characterized in that: The main screening parameters and response parameters include the following: Count the total number of times each parameter is marked as a candidate principal parameter within each time period, and determine the parameter with the highest total number of times as the principal parameter of the corresponding time period; The other parameter that is paired with the main parameter and has the highest response coupling is determined as the response parameter for the time period.

5. The elevator brake friction performance testing method based on multi-data fusion as described in claim 1, characterized in that: The process of constructing the braking performance template includes the following steps: After the brakes have completed their break-in period, data from multiple consecutive braking actions are selected as learning samples. For each set of learning samples, waveform segments of the principal parameter signal and the response parameter signal are extracted at each time period; The principal parameter waveforms of all learning samples within the same time period are averaged point by point to obtain the template principal parameter waveforms for the corresponding time period. The template response parameter waveforms for the corresponding time period are obtained by averaging the response parameter waveforms of all learning samples within the same time period.

6. The method for detecting the friction performance of elevator brakes based on multi-data fusion as described in claim 5, characterized in that: The amount of frictional performance degradation at each time period is described in the following analysis: Waveform segments of the main parameter signal and the response parameter signal are extracted from each time period to form a joint waveform pair; Align the actual waveform of the main parameter in the joint waveform pair with the template main parameter waveform of the corresponding time period point by point, calculate the area of ​​the region enclosed between the actual waveform curve of the main parameter and the template main parameter waveform curve, and record it as the main parameter waveform comparison area. Align the actual waveform of the response parameter in the joint waveform pair with the template response parameter waveform of the corresponding time period point by point, calculate the area of ​​the region enclosed between the actual waveform curve of the response parameter and the template response parameter waveform curve, and record it as the response parameter waveform comparison area. The total waveform area deviation is obtained by adding the waveform comparison area of ​​the main parameter and the waveform comparison area of ​​the response parameter, which is used as the amount of friction performance degradation over time.

7. The method for detecting the friction performance of elevator brakes based on multi-data fusion as described in claim 1, characterized in that: The multi-factor coupling determination model is constructed as follows: The amount of frictional performance degradation at each time period after the braking action is completed is compared with the corresponding abnormal judgment threshold. If the amount of frictional performance degradation at a certain time period exceeds the abnormal judgment threshold, the time period is marked as an abnormal time period. The number of time periods marked as abnormal periods in each braking action is counted and recorded as the number of covered time periods.

8. The elevator brake friction performance testing method based on multi-data fusion as described in claim 7, characterized in that: The diagnostic results of the output brake friction performance status include the following: The electromagnetic release period and the brake release recovery period are classified as peripheral periods, while the mechanical action period, the contact pressure building period, and the brake maintenance period are classified as dominant periods. When the number of covered time periods is zero, the output will show a normal tribological performance diagnosis result. When the abnormal period only involves the peripheral period, output the abnormal diagnosis result of the peripheral link; When the abnormal period includes the dominant period but the number of covered periods is one, output a diagnostic result for decreased braking performance. When the abnormal period includes the dominant period and the number of covered periods is greater than one, the braking performance degradation diagnosis result is output.

Images