Elevator multi-caliper disc brake diagnostic methods, systems, apparatus, media, and products

Abstract

The invention provides an elevator multi-caliper disc brake diagnosis method, system and device, a medium and a product, and relates to the field of fault diagnosis of elevator multi-caliper disc brakes. The diagnosis method comprises the following steps: acquiring a state transition matrix among preset states in a Markov jump system state transition model; first operation data of the multi-caliper disc brake are obtained and preprocessed; according to the preprocessed first operation data, a current braking system state estimation value and a corresponding operation data range are obtained; second operation data of the multi-caliper disc brake are collected in real time and preprocessed; and when the preprocessed second operation data exceed the operation data range, the fault type of the multi-caliper disc brake is determined according to the current braking system state estimation value, the Markov jump system state transition model and the state transition matrix. According to the invention, fault diagnosis of the multi-caliper disc brake of the elevator can be realized, and support is provided for daily overhaul and maintenance of an elevator brake system.

Description

Technical Field

[0001] The present invention relates to the field of fault diagnosis of elevator multi-caliper disc brakes, and in particular to a diagnosis method, system, equipment, medium and product of elevator multi-caliper disc brakes. Background Art

[0002] Elevator braking systems are essential for safe operation, from stopping to emergency stops. The braking system is equipped with numerous safety devices to ensure stable operation. Failure or malfunction of any of these devices can pose a serious threat to passenger safety. Multi-caliper disc brakes, the actuators for braking in most high-speed elevators, are a crucial component for elevator safety. Their braking performance is directly linked to the reliable stopping of the traction motor and the safe landing of the car. Multi-caliper disc brakes typically consist of multiple calipers and brake discs, capable of applying braking torque simultaneously at multiple contact points, generating a high braking torque in a short period of time and enhancing emergency braking effectiveness. Compared to conventional low-speed elevators, high-speed elevators operate under high-speed, variable load conditions for extended periods, resulting in higher speeds, longer braking distances during emergency braking, higher friction component temperatures, and severe wear. Consequently, the risk of failure of key brake components over time is significantly higher than in conventional elevators. Therefore, based on the study of the performance characteristics of high-speed elevator multi-caliper disc brakes, it is necessary to analyze the impact of the failure of a brake caliper on the comprehensive braking capacity under high-speed and variable load conditions, so as to effectively detect and warn the faulty elevator brake.

[0003] Numerous researchers have proposed numerous methods for elevator brake fault detection, such as a "host computer + embedded controller" intelligent elevator brake monitoring system that monitors elevator brake operating data in real time to determine the brake's operating status, enabling real-time monitoring and feedback of brake faults; and a support vector machine-based abnormal feature detection strategy that implements elevator brake fault detection based on deep clustering. However, these methods are all targeted at common drum, disc, and belt brakes, lacking research specific to multi-caliper disc brakes, making them difficult to universally apply to multi-caliper disc brake fault diagnosis. Summary of the Invention

[0004] The embodiments of the present invention provide a diagnostic method, system, device, storage medium, and computer program product for an elevator multi-caliper disc brake. Based on the comprehensive braking capacity evaluation results during the actual braking process of the elevator, and in accordance with the reliability index of the multi-caliper disc brake and the single-caliper fault-brake performance relationship model, the method analyzes whether one or more brake caliper faults have caused an unsatisfactory elevator braking effect, thereby providing support for the routine inspection and maintenance of elevator braking systems using multi-caliper disc brakes.

[0005] According to a first aspect of the present invention, an embodiment of the present invention provides a method for diagnosing a multi-caliper brake of an elevator, wherein the braking system of the elevator includes a multi-caliper brake, and the multi-caliper brake includes multiple brake calipers. The diagnostic method includes: obtaining a state transfer matrix between each preset state in a Markov jump system state transfer model; obtaining first operating data of the multi-caliper brake and preprocessing it; obtaining a current braking system state estimate value based on the preprocessed first operating data, and obtaining an operating data range corresponding to the current braking system state estimate value; collecting the second operating data of the multi-caliper brake in real time and preprocessing it; when the preprocessed second operating data exceeds the operating data range, determining the fault type of the multi-caliper brake based on the current braking system state estimate value, the Markov jump system state transfer model and the state transfer matrix.

[0006] The above-mentioned embodiment of the present invention is based on the Markov jump system state transition model, obtains the brake system state estimation value through the operating data of the multi-disc brake, and accurately detects the fault type of the multi-disc brake when there is a difference between the operating data of the multi-disc brake and the state estimation value, providing support for the daily inspection and maintenance of the elevator braking system using the multi-disc brake.

[0007] In some embodiments of the present invention, the diagnostic method further includes: establishing a physical model of the multi-caliper brake based on the physical structure of the multi-caliper brake and the geometric relationship between the calipers of the multi-caliper brake; and obtaining a comprehensive reliability index of the multi-caliper brake according to the physical model of the multi-caliper brake.

[0008] In some embodiments of the present invention, the diagnostic method further includes: establishing the Markov jump system state transition model that describes the characteristic mapping between a single brake caliper failure and the braking performance of the multi-caliper brake based on the comprehensive reliability index of the multi-caliper brake, the brake caliper failure and the comprehensive braking capacity evaluation results of the high-speed elevator.

[0009] In some embodiments of the present invention, the diagnostic method further comprises: updating the state transition matrix according to the first operating data and the second operating data of the multi-caliper brake and maintenance information of the elevator.

[0010] According to the above embodiments of the present invention, the robustness of the diagnosis method can be improved by continuously optimizing the state transition matrix.

[0011] In some embodiments of the present invention, the first operating data and the second operating data include: a temperature of the multi-caliper brake, a vibration signal of the multi-caliper brake, and a pressure signal of the multi-caliper brake.

[0012] In some embodiments of the present invention, an observer technique is used to obtain an estimated value of the current braking system state based on the preprocessed first operating data and a dynamic equation, wherein the dynamic equation is a linear dynamic equation for describing the dynamic characteristics of the braking system.

[0013] In some embodiments of the present invention, the linear dynamic equation is:

[0014]

[0015] in, is a state vector, the state vector including state information of the braking system; is the input vector; is an output vector, which includes readings of various acquisition sensors; 、 、 is a Markov jump system dynamic characteristic matrix, wherein the Markov jump system dynamic characteristic matrix is ​​related to a preset state in a Markov jump system state transition model; is the process noise; is the measurement noise.

[0016] In some embodiments of the present invention, the state transition probability in the state transition matrix is ​​updated by the input vector.

[0017] In some embodiments of the present invention, a Kalman filter or an extended Kalman filter technique is used to calculate the current brake system state estimate.

[0018] In some embodiments of the present invention, using Kalman filter technology to calculate the current braking system state estimate includes: using Kalman filter technology to calculate the braking clearance and braking time based on the preprocessed first operating data; generating fusion data based on the braking clearance and braking time and the temperature of the multi-caliper brake; and calculating the current braking system state estimate based on the fusion data and a Markov jump system state transition model.

[0019] In some embodiments of the present invention, determining the fault type of the multi-caliper brake based on the current braking system state estimate, the Markov jump system state transition model and the state transition matrix includes: comparing the current braking system state estimate with each preset state in the Markov jump system state transition model, and calculating the probability value of each Markov state in the Markov jump system state transition model in combination with the state transition matrix, and determining the fault type of the multi-caliper brake based on the probability value.

[0020] In some embodiments of the present invention, the diagnostic method further includes: after determining the fault type, locating the faulty brake caliper according to the following steps: extracting the fault characteristics of the vibration signal of each brake caliper in the preprocessed first operating data and the preprocessed second operating data, the fault characteristics including time domain fault characteristics and frequency domain fault characteristics; inputting the time domain fault characteristics and frequency domain fault characteristics of the vibration signal into a fault mode recognition model, and the fault mode recognition model outputs the fault mode of the brake caliper; and locating one or more faulty brake calipers according to the abnormality degree value of the fault characteristics of the vibration signal of each brake caliper.

[0021] In some embodiments of the present invention, the time domain fault characteristics of the vibration signal include: peak value, mean value, variance, and kurtosis of the vibration signal; the frequency domain fault characteristics of the vibration signal include: main frequency and spectrum energy distribution.

[0022] In some embodiments of the present invention, the diagnostic method further comprises: after locating the faulty brake caliper, generating an operational suggestion based on the faulty one or more brake calipers.

[0023] In some embodiments of the present invention, the diagnostic method further comprises: displaying the status of the brake system, the faulty brake caliper, and operation suggestions via a graphical interface.

[0024] According to the above-mentioned embodiment of the present invention, by visually displaying the real-time status, fault details and operation suggestions of the brake system, it is convenient for operation and maintenance personnel to inspect and maintain the elevator brake system.

[0025] According to the second aspect of the present invention, an embodiment of the present invention provides an elevator multi-caliper brake diagnostic system, which is used to implement the aforementioned elevator multi-caliper brake diagnostic method, and the diagnostic system includes: a model parameter acquisition module, which is used to obtain the state transfer matrix between each preset state in the Markov jump system state transfer model; a data acquisition module, which is used to obtain the first operating data of the multi-caliper brake, and to collect the second operating data of the multi-caliper brake in real time, and preprocess the first operating data and the second operating data; a state estimation module, which is used to obtain the current brake system state estimation value based on the preprocessed first operating data, and obtain the operating data range corresponding to the current brake system state estimation value; a fault diagnosis module, which is used to determine the fault type of the multi-caliper brake according to the current brake system state estimation value, the Markov jump system state transfer model and the state transfer matrix when the preprocessed second operating data exceeds the operating data range.

[0026] The above-mentioned embodiment of the present invention is based on the Markov jump system state transition model, obtains the brake system state estimation value through the operating data of the multi-disc brake, and accurately detects the fault type of the multi-disc brake when there is a difference between the operating data of the multi-disc brake and the state estimation value, providing support for the daily inspection and maintenance of the elevator braking system using the multi-disc brake.

[0027] In some embodiments of the present invention, the model parameter acquisition module establishes a physical model of the multi-caliper brake based on the physical structure of the multi-caliper brake and the geometric relationship between the calipers of the multi-caliper brake; and obtains a comprehensive reliability index of the multi-caliper brake according to the physical model of the multi-caliper brake.

[0028] In some embodiments of the present invention, the model parameter acquisition module establishes the Markov jump system state transition model that describes the characteristic mapping between a single brake caliper failure and the braking performance of the multi-caliper brake based on the comprehensive reliability index of the multi-caliper brake, the brake caliper failure and the comprehensive braking capacity evaluation results of the high-speed elevator.

[0029] In some embodiments of the present invention, the model parameter acquisition module is further configured to update the state transition matrix according to the first operating data and the second operating data of the multi-caliper brake and maintenance information of the elevator.

[0030] According to the above embodiments of the present invention, the robustness of the diagnosis system can be improved by continuously optimizing the state transfer matrix.

[0031] In some embodiments of the present invention, the first operating data and the second operating data include: a temperature of the multi-caliper brake, a vibration signal of the multi-caliper brake, and a pressure signal of the multi-caliper brake.

[0032] In some embodiments of the present invention, the state estimation module adopts observer technology to obtain the current braking system state estimation value based on the preprocessed first operating data and the dynamic equation, wherein the dynamic equation is a linear dynamic equation used to describe the dynamic characteristics of the braking system.

[0033] In some embodiments of the present invention, the linear dynamic equation is:

[0034]

[0035] in, is a state vector, the state vector including state information of the braking system; is the input vector; is an output vector, which includes readings of various acquisition sensors; 、 、 is a Markov jump system dynamic characteristic matrix, wherein the Markov jump system dynamic characteristic matrix is ​​related to a preset state in a Markov jump system state transition model; is the process noise; is the measurement noise.

[0036] In some embodiments of the present invention, the model parameter acquisition module updates the state transition probability in the state transition matrix through the input vector.

[0037] In some embodiments of the present invention, the state estimation module uses a Kalman filter or an extended Kalman filter technology to calculate the current braking system state estimate.

[0038] In some embodiments of the present invention, the state estimation module uses Kalman filter technology to calculate the current braking system state estimation value, including: using Kalman filter technology to calculate the braking clearance and braking time based on the preprocessed first operating data; generating fusion data based on the braking clearance and braking time and the temperature of the multi-caliper brake; calculating the current braking system state estimation value based on the fusion data and the Markov jump system state transfer model.

[0039] In some embodiments of the present invention, determining the fault type of the multi-caliper brake based on the current braking system state estimate, the Markov jump system state transition model and the state transition matrix includes: comparing the current braking system state estimate with each preset state in the Markov jump system state transition model, and calculating the probability value of each Markov state in the Markov jump system state transition model in combination with the state transition matrix, and determining the fault type of the multi-caliper brake based on the probability value.

[0040] In some embodiments of the present invention, the diagnostic system further includes a fault location module for locating the faulty brake caliper after determining the fault type according to the following steps: extracting fault characteristics of the vibration signal of each brake caliper in the preprocessed first operating data and the preprocessed second operating data, the fault characteristics including time domain fault characteristics and frequency domain fault characteristics; inputting the time domain fault characteristics and frequency domain fault characteristics of the vibration signal into a fault mode recognition model, and the fault mode recognition model outputs the fault mode of the brake caliper; and locating one or more faulty brake calipers according to the abnormality degree value of the fault characteristics of the vibration signal of each brake caliper.

[0041] In some embodiments of the present invention, the time domain fault characteristics of the vibration signal include: peak value, mean value, variance, and kurtosis of the vibration signal; the frequency domain fault characteristics of the vibration signal include: main frequency and spectrum energy distribution.

[0042] In some embodiments of the present invention, the diagnostic system further comprises a suggestion generating module for generating an operation suggestion based on the faulty one or more brake calipers.

[0043] In some embodiments of the present invention, the diagnostic system further comprises a display module for displaying the status of the brake system, the faulty brake caliper and operation suggestions through a graphical interface.

[0044] According to the above-mentioned embodiment of the present invention, by visually displaying the real-time status, fault details and operation suggestions of the brake system, it is convenient for operation and maintenance personnel to inspect and maintain the elevator brake system.

[0045] According to a third aspect of the present invention, an embodiment of the present invention provides a computer-readable storage medium having computer-readable instructions stored thereon. When the computer-readable instructions are executed by a processor, the computer performs the following operations: the operations include the steps included in the elevator multi-caliper brake diagnostic method based on the Markov jump system model as described in any of the above embodiments.

[0046] According to a fourth aspect of the present invention, an embodiment of the present invention provides a computer device including a memory and a processor, wherein the memory is used to store one or more computer-readable instructions, wherein the one or more computer-readable instructions, when executed by the processor, can implement the elevator multi-caliper brake diagnosis method based on the Markov jump system model as described in any of the above embodiments.

[0047] According to a fifth aspect of the present invention, an embodiment of the present invention provides a computer program product comprising a computer program, which, when executed by a processor, implements the elevator multi-caliper brake diagnostic method based on the Markov jump system model as described in any of the above embodiments.

[0048] As can be seen from the above, the elevator multi-caliper brake diagnostic method, system, equipment, medium and computer program product provided in the embodiments of the present invention are based on the comprehensive braking capacity evaluation results during the actual braking process of the elevator. According to the multi-caliper disc brake reliability index and the single-caliper fault-brake performance relationship model, they analyze whether there is one or more brake caliper failures that lead to unsatisfactory elevator braking effect, and provide support for the daily inspection and maintenance of elevator braking systems using multi-caliper disc brakes. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 11 is a flow chart of a diagnostic method for an elevator multi-caliper brake based on a Markov jump system model according to embodiment 1 of the present invention; Figure 2 1 is a flow chart of a method for diagnosing and positioning an elevator multi-caliper brake based on a Markov jump system model according to embodiment 3 of the present invention; Figure 3 1 is a flow chart of a method for diagnosing and positioning an elevator multi-caliper brake based on a Markov jump system model according to a fourth embodiment of the present invention; Figure 4 2 is a schematic diagram of the architecture of an elevator multi-caliper brake diagnostic system based on a Markov jump system model according to a fifth embodiment of the present invention; Figure 5 1 is a schematic diagram of a layer framework of an elevator multi-caliper brake fault diagnosis system based on a Markov jump system model according to a sixth embodiment of the present invention; Figure 6 yes Figure 5 A schematic diagram of the processing operations performed by the components of each level of the fault diagnosis system shown; Figure 7 is a schematic diagram of a disc brake measurement structure according to embodiment 7 of the present invention; Figure 8 yes Figure 7 Schematic diagram of the structure of the touch terminal.

[0050] The reference numerals are explained as follows: 10-vibration sensor, 20-touch terminal, 30-temperature sensor, 40-coil voltage detection module, 1-shell, 2-acquisition module circuit board, 3-lithium battery, 4-rubber sleeve, 5-touch screen fixing plate, 6-touch screen, 7-upper shell, 8-panel. DETAILED DESCRIPTION

[0051] The various aspects of the present invention are described in detail below in conjunction with the accompanying drawings and specific embodiments. Among them, well-known modules, units and their connections, links, communications or operations are not shown or described in detail. In addition, the described features, architectures or functions can be combined in any manner in one or more embodiments. It should be understood by those skilled in the art that the various embodiments described below are only for illustration and are not intended to limit the scope of protection of the present invention. It can also be easily understood that the modules or units or processing methods in the various embodiments described herein and shown in the drawings can be combined and designed in various different configurations.

[0052] The following is a brief explanation of the terms used in the following text.

[0053] MJS: Markov Jump System, Markov jump system.

[0054] FMEA: Failure Mode and Effects Analysis, a systematic risk assessment tool that aims to identify potential failure modes in a system, design, process or service through structured analysis, evaluate their impact on system performance, and determine improvement priorities.

[0055] EKF: Extended Kalman Filter, which is an extended version of the Kalman filter (KF) in nonlinear systems.

[0056] ABS: Acrylonitrile Butadiene Styrene, a thermoplastic polymer material, its Chinese name is acrylonitrile-butadiene-styrene copolymer.

[0057] PBT: Polybutylene Terephthalate, a thermoplastic engineering plastic, its Chinese name is polybutylene terephthalate.

[0058] Effective contact area of ​​caliper disc: the sum of the areas where the surface of the caliper disc friction material and the friction surface of the brake disc actually make microscopic contact and transmit friction force during the braking process.

[0059] [Example 1] Figure 1 1 is a flow chart of a diagnostic method for an elevator multi-caliper brake based on a Markov jump system model according to Embodiment 1 of the present invention, wherein the elevator braking system includes the multi-caliper brake, which includes a plurality of brake calipers.

[0060] like Figure 1 As shown, in embodiment 1 of the present invention, the elevator multi-caliper brake diagnostic method based on the Markov jump system model may include at least the following steps S11, S12, S13, S14 and S15, which are described in detail below.

[0061] In step S11 , a state transfer matrix between various preset states in a Markov jump system state transfer model is obtained.

[0062] In some embodiments, a Markov jump system state transition model is established based on the comprehensive reliability index of the multi-caliper brake, brake caliper failures, and comprehensive high-speed elevator braking capacity assessment results, describing the characteristic mapping between a single brake caliper failure and the braking performance of the multi-caliper brake. The multi-caliper brake physical model is established based on the physical structure of the multi-caliper brake and the geometric relationship between the calipers of the multi-caliper brake; and the comprehensive reliability index of the multi-caliper brake is obtained based on the multi-caliper brake physical model.

[0063] In further embodiments, the transition probability matrix and other model parameters are continuously updated using newly collected data, and the validity of the model is periodically verified and necessary corrections are made.

[0064] The reliability indicators include operational reliability and inherent reliability. The inherent reliability includes, but is not limited to, one or more of the following: brake spring pressure reliability, braking torque reliability, friction pad wear reliability, braking action time reliability, brake disc clearance reliability, and hinge structural performance reliability. Specifically, operational reliability is reflected in on-site inspections of special equipment and regular maintenance by maintenance companies, while inherent reliability is determined by relevant factors in the design and production of the brake model to be diagnosed. Operational reliability and inherent reliability are consistent to a certain extent, and their mathematical relationship is expressed as: ,in, It is the overall reliability of the brake; It is inherent reliability; It is reliability in use.

[0065] For example, for a single-caliper disc brake, The indicators are determined by the input indicators of each component in the design phase, including but not limited to the braking force, the clearance and time between braking and releasing, the braking torque, etc., among which the braking torque is the most important indicator. The braking force and the friction coefficient value are the key factors affecting the braking torque, and are also the key factors determining the reliability of the disc brake. The use of high-quality spring materials and various measures to ensure the stability of the braking friction factor can output a continuous and stable braking torque, which is the guarantee for accurate and safe braking of the elevator on each floor. It can be concluded from a large number of failure sites that another major indicator affecting the safety and reliability of the brake is the hinge failure rate, because the severity of its damage directly determines the opening and closing action of the brake and the braking torque for hovering. Therefore, the inherent reliability mathematical relationship of the disc brake is expressed as: ,in, The reliability of the spring in closing and releasing the brake; is the friction disc braking torque reliability; It is the reliability of the hinge structure.

[0066] in, 、 The calculation method is as follows: (1) Calculation: Elevators generally use normally closed brakes. In the lifting system, butterfly spring groups are used to shorten the space field. For ordinary hydraulic disc brakes, the life is approximately distributed in an exponential curve. The failure rate is obtained through a large amount of statistical data, and then the reliability of the brake spring is calculated. ; (2) Calculation: Friction reliability, i.e., the stability and reliability of the braking torque, can be determined through a single-caliper brake test. During the test, the friction coefficient and torque distribution on the brake disc are determined, and the final braking torque and friction coefficient mean and standard deviation data are obtained to calculate the friction coefficient reliability index. ; (3) Calculation: The hinge structure belongs to the serial system logic control. When each hinge can complete the specified action normally, the reliability of the disc brake can be guaranteed. By assuming the average failure rate of the mechanical hinge, the reliability index of the hinge is obtained. Then, according to the specific number of hinges of the assumed brake, the reliability of the serial system is obtained. Finally, based on the reliability index analysis of the single-caliper disc brake, the comprehensive reliability index of the multi-caliper disc brake is obtained by considering the specific structure of the multi-caliper disc brake and the geometric relationship between different calipers.

[0067] Specifically, the reliability of each caliper unit of a multi-caliper brake is calculated using the following formula: , in, Indicates the A clamp disc; Calculated using the Weibull life distribution model for disc spring groups: , Where t is a time variable, which represents the time from the start of operation to failure or reaching a specific state (such as wear threshold) of the multi-caliper brake. is the characteristic lifespan, is the shape parameter; Based on the brake bench test data, the tolerance range probability of the normal distribution of the friction coefficient is calculated and derived: , in, is the standard normal cumulative function, z is the standardized friction coefficient, which is used in the standard normal cumulative function Φ(z) to quantitatively evaluate the amplitude distribution characteristics of the friction coefficient; Calculation based on the series system model: , in, is the number of hinge points of a single clamp disc, is the average reliability of the hinge point.

[0068] Also, consider The redundant structure of the parallel clamps and the actual load imbalance introduce the load factor Corrected comprehensive reliability index of multi-caliper disc brake : , in, Indicates the The actual load borne by each clamp unit, Indicates the total load borne by all unit clamps of all clamps in the system.

[0069] Finally, the correctness of the braking performance reliability of the multi-caliper disc brake can be verified through experiments.

[0070] In step S12 , first operating data of the multi-caliper brake is acquired and pre-processed.

[0071] In some embodiments, the first operating data and the second operating data described later include but are not limited to one or more of the following: temperature of the multi-caliper brake, vibration signal of the multi-caliper brake, pressure signal of the multi-caliper brake, and displacement signal of the brake caliper.

[0072] In this embodiment, sensors are installed to monitor the operating parameters of the brake caliper, such as pressure, temperature, and displacement, and continuously record data during the operation of the brake system. Alternatively, a vibration sensor, temperature sensor, coil voltage detection module, and the like can be installed using the disc brake measurement structure shown in Example 7 below to obtain vibration, temperature, and pressure signals from a multi-caliper disc brake.

[0073] In step S13, a current brake system state estimation value is obtained according to the pre-processed first operating data, and an operating data range corresponding to the current brake system state estimation value is obtained.

[0074] In some embodiments, an observer technique is used to obtain an estimated value of the current braking system state based on the preprocessed first operating data and a dynamic equation, wherein the dynamic equation is a linear dynamic equation for describing the dynamic characteristics of the braking system.

[0075] The linear dynamic equation is shown in the following formula (1):

[0076] (1) in, is a state vector, the state vector including state information of the braking system; is the input vector; is an output vector, which includes readings of various acquisition sensors; 、 、 is a Markov jump system dynamic characteristic matrix, wherein the Markov jump system dynamic characteristic matrix is ​​related to a preset state in a Markov jump system state transition model; is the process noise; is the measurement noise.

[0077] In a further embodiment, the state transition probabilities in the state transition matrix are updated by the input vector.

[0078] In an optional embodiment, a Kalman filter or an extended Kalman filter technique is used to calculate the current brake system state estimate. Exemplarily, using the Kalman filter technique to calculate the current brake system state estimate includes: using the Kalman filter technique to calculate brake clearance and braking time based on the preprocessed first operating data; generating fused data based on the brake clearance and braking time and the temperature of the multi-caliper brake; and calculating the current brake system state estimate based on the fused data and a Markov jump system state transition model.

[0079] In step S14 , second operating data of the multi-caliper brake is collected in real time and pre-processed.

[0080] In a further embodiment, the diagnostic method further comprises: updating the state transition matrix according to the first operating data and the second operating data of the multi-caliper brake and the maintenance information of the elevator.

[0081] In step S15 , when the preprocessed second operating data exceeds the operating data range, the fault type of the multi-caliper brake is determined according to the current brake system state estimation value, the Markov jump system state transition model and the state transition matrix.

[0082] In some embodiments, determining the fault type of the multi-caliper brake based on the current brake system state estimate, the Markov jump system state transition model, and the state transition matrix may include: comparing the current brake system state estimate with each preset state in the Markov jump system state transition model, and calculating the probability value of each Markov state in the Markov jump system state transition model in combination with the state transition matrix, and determining the fault type of the multi-caliper brake based on the probability value.

[0083] The diagnostic method of Example 1 of the present invention is used to obtain a brake system state estimate through the operating data of the multi-caliper brake on the basis of the Markov jump system state transition model. When there is a difference between the operating data of the multi-caliper brake and the state estimate, the fault type / potential fault of the multi-caliper brake is accurately detected, thereby providing support for the routine inspection and maintenance of the elevator brake system using the multi-caliper brake.

[0084] In a further embodiment, the diagnostic method further includes, after determining the fault type, locating the faulty brake caliper according to the following steps: extracting fault features from the vibration signal of each brake caliper in the preprocessed first operating data and the preprocessed second operating data, the fault features including time-domain fault features and frequency-domain fault features; inputting the time-domain fault features and frequency-domain fault features of the vibration signals into a fault pattern recognition model, which outputs the fault mode of the brake caliper; and locating one or more faulty brake calipers based on the abnormality level of the fault features of the vibration signals of each brake caliper. The time-domain fault features of the vibration signals include, but are not limited to, one or more of the following: peak value, mean, variance, and kurtosis; and the frequency-domain fault features of the vibration signals include, but are not limited to, dominant frequency and spectral energy distribution. This allows for precise location of the fault, resolving the problem that conventional fault detection methods focus solely on the overall operating status of the brake, providing only information such as whether an elevator brake fault has occurred and the type of fault, but failing to accurately locate the fault.

[0085] In a further embodiment, the diagnostic method further comprises: after locating the faulty brake caliper, generating an operational suggestion based on the faulty one or more brake calipers. Furthermore, the diagnostic method further comprises: displaying the brake system status, the faulty brake caliper, and the operational suggestion via a graphical interface.

[0086] Using the diagnostic method described above, the reliability of the spring application and release, as well as the hinge structure reliability, of a single-caliper disc brake are analyzed. A comprehensive reliability index for the multi-caliper disc brake is established based on the specific structure of the multi-caliper disc brake and the geometric relationships between the different calipers. Secondly, through experiments and testing, comprehensive braking capacity evaluation results for high-speed elevators under different caliper failure and performance conditions are collected. Using feature selection and regression analysis, a multi-state transition model is developed to describe the qualitative relationship between individual caliper failures and braking performance. Finally, based on the comprehensive braking capacity evaluation results during actual elevator braking, the multi-caliper disc brake reliability index and the single-caliper failure-brake performance relationship model are used to analyze whether one or more caliper failures are causing unsatisfactory elevator braking performance. Furthermore, if an abnormality occurs during braking or exceeds a set maintenance threshold, the system automatically issues an alarm and performs fault analysis, accurately indicating the specific fault location, providing support for routine inspection and maintenance of elevator braking systems using multi-caliper disc brakes.

[0087] [Example 2] Embodiment 2 of the present invention provides a method for diagnosing and locating faults of an elevator multi-caliper brake based on a Markov jump system model, wherein the elevator braking system includes the multi-caliper brake, which includes multiple brake calipers.

[0088] In a second embodiment of the present invention, the method may include at least: Step 1: Establishing a Markov Jump System (MJS) model for different brake caliper faults; and Step 2: Fault diagnosis and location. These steps are described in detail below.

[0089] Step 1: Establishing the Markov jump system model includes the following steps: Step S211: define the possible states of different brake calipers in the brake system, that is, define the state space. For example, through a finite set Indicates the possible states of different brake calipers in the brake system. It is in normal operating state; A minor fault in a single brake caliper (for example, minor wear of the friction pad); Several brake calipers had minor faults; A single brake caliper has a moderate fault (for example, moderate wear of the friction pad); It is a moderate failure of multiple brake calipers; A single brake caliper has a serious fault (for example, the brake caliper is stuck); There was a serious failure of multiple brake calipers.

[0090] In the above example, all possible states of the defined brake system fully cover the normal working state of the brake as well as various fault states, such as the following states: .

[0091] Among them, normal operating state ( ) indicates that all brake components are working normally, no faults have occurred, and the system is in optimal working condition.

[0092] A single brake caliper has a minor fault condition ( ) indicates that a brake caliper in the braking system is slightly worn or has slightly degraded performance, but the overall braking performance is still within a safe range and does not affect the normal operation of the elevator.

[0093] Multiple brake caliper minor fault conditions ( ) indicates that there are minor faults in multiple brake calipers in the brake system, which may have a certain cumulative impact on braking performance and require close attention.

[0094] Moderate fault condition of a single brake caliper ( ) indicates that a brake caliper in the braking system has moderate wear or significant performance degradation, which has a significant impact on the braking performance and may cause abnormal operation of the elevator.

[0095] Multiple brake caliper moderate fault conditions ( ) indicates that there are moderate faults in multiple brake calipers in the braking system, the braking performance is seriously degraded, and there are safety hazards in the operation of the elevator.

[0096] Single brake caliper serious fault state ( ) indicates that a brake caliper in the braking system is severely worn or completely fails, the braking performance is basically lost, and the elevator cannot brake safely.

[0097] Multiple brake caliper critical fault conditions ( ) indicates that there are serious faults in multiple brake calipers in the braking system, the braking system has completely failed, and the elevator is in an extremely dangerous state.

[0098] Step S212: Construct a state transfer matrix. Describe the system from the state Transfer to state The probability of , which can be estimated based on historical data, expert knowledge or fault tree analysis. For example, based on the state set S in step S211, assume that the state transition matrix is ​​as follows:

[0099] in, Indicates that the system is in state Transfer to state The probability, for example, Indicates that the system is in normal operation. Transferred to a minor fault condition of a single brake caliper The probability of . At the same time, And the sum of the elements in each row is equal to 1, that is: .

[0100] For example, It may indicate the probability of a minor failure of a single brake caliper under normal operating conditions due to some reason (such as normal wear of the friction pad). It may indicate the probability that when a single brake caliper has a serious fault, due to failure to repair it in time or the fault spreads, multiple brake calipers will suffer serious faults.

[0101] By collecting a large amount of historical operation data and combining it with the knowledge of industry experts, we can scientifically estimate the probability of transitioning from one state to another. Based on these probability values, we construct a state transition matrix. Each element P( | ) indicates that the system is in state Transfer to state The probability of failure provides a quantitative basis for state prediction and fault mode identification.

[0102] In actual situations, the state transfer matrix The transition probabilities between different states in the state matrix may be affected by various factors, such as the brake's operating environment, maintenance status, and the driver's driving habits. For example, regular maintenance of an elevator's multi-disc brake may increase the probability of recovering from a faulty state to a normal state; prolonged high-load operation may also increase the probability of transitioning from a normal state to a faulty state. Alternatively, a detailed system analysis and failure mode and effects analysis (FMEA) can be performed to obtain an accurate state transition matrix.

[0103] Step S213: Define the system dynamic equation. For each state , define a dynamic equation to describe the operating behavior of the system in this state.

[0104] In an exemplary embodiment, the dynamic characteristics of the braking system are described using the linear dynamic equation shown in the following formula (2):

[0105] (2) in, is a state vector, the state vector containing state information of the braking system; is the input vector; is an output vector, which includes readings of various acquisition sensors; 、 、 is the dynamic characteristic matrix of the Markov jump system, and the dynamic characteristic matrix of the Markov jump system is consistent with the state in the state transfer model of the Markov jump system. Related; is the process noise; is the measurement noise.

[0106] Among them, the input vector This could be an external force / event that actuates the brake or other external input. For example, scheduled maintenance may increase (from failure to normal) value, and long-term high-load operation may increase (from normal to fault).

[0107] In this embodiment, the application of the dynamic equation in fault diagnosis and location includes: (1) State estimation: During the fault diagnosis process, the dynamic equation is used to estimate the system state. By comparing the difference between the estimated state and the actual measured value, potential faults are detected. (2) Fault isolation: When a fault is detected, the dynamic equation and the state transfer matrix work together in the fault isolation algorithm. By analyzing the path and probability of the system state transition, it can help determine which brake caliper has a fault. (3) Model update: By continuously updating the model parameters (such as the state transfer matrix and the system matrix), the model can adapt to changes in the system over time, improve the efficiency and accuracy of fault diagnosis and location, and thus achieve real-time analysis, rapid warning, and risk avoidance of elevator safety risks.

[0108] In another exemplary embodiment, the dynamic equation is shown as the following formula (3): (3) in, is the state vector at time k; is the input vector; is the state vector at time k+1; 、 It depends on the state in the state transition model of the Markov jump system Related system matrices and input matrices; is the process noise. Characterize the state transition relationship (for example, brake caliper failure will change the attenuation law of the braking torque, making Changes in wear-related parameters); Reflects the effect of input on the state (e.g. when the brake caliper fails, the same control input The generated braking torque may be reduced, resulting in matrix elements change). In other words, 、 According to the status ( Therefore, combined with the state transfer matrix, the model can describe the dynamic behavior of the brake under different brake caliper fault states and provide a basis for fault diagnosis and location.

[0109] Step 2, fault diagnosis and location, specifically includes the following steps S221 to S224: Step S221: Data collection: High-precision sensors are used to collect real-time brake operation data, such as braking time, braking force, brake temperature, vibration signals, etc., and the collected data is pre-processed to improve data quality.

[0110] In this embodiment, vibration sensors are used to accurately locate faults and assess their severity. Specifically, a triaxial vibration accelerometer is placed on the brake shoe corresponding to each brake caliper. It continuously collects vibration signals during braking at a high sampling frequency, monitoring and recording the operating status of each brake caliper in real time. The collected data is then pre-processed by denoising and normalizing it to eliminate environmental noise and other interfering factors, improve data quality, and unify data of different dimensions to a common scale for subsequent analysis. Furthermore, by deeply extracting the time and frequency domain features of the vibration signal and using machine learning algorithms for intelligent pattern recognition, the fault mode of the brake caliper can be accurately determined.

[0111] Step S222: State estimation. Using Kalman filtering and other observer techniques, the system state is accurately estimated based on the data collected in step S221. , get the estimated value of the current system state. Specifically including the following steps 2.1~2.3: Step 2.1: Calculate the brake clearance.

[0112] Based on the principle of Newtonian mechanics, displacement can be obtained by the quadratic integration of acceleration over time. That is, the vibration acceleration signal during the operation of the brake can be collected in real time by the vibration acceleration sensor. In theory, Performing an integration can get the speed signal , the integral formula is: . Then Perform secondary integration to obtain the displacement signal, i.e. the brake clearance data. , the integral formula is: .

[0113] However, direct integration will introduce error accumulation effect. Therefore, in this embodiment, the extended Kalman filter (EKF) is introduced to combine the double integration process with state estimation to avoid the error accumulation caused by direct integration. Specifically, the EKF converts the acceleration signal As input, through the nonlinear state transfer function Predict the state at the next moment and use the Kalman gain Correct the predicted values.

[0114] The state prediction equation is shown in the following formula (4): (4) The covariance prediction equation is shown in the following formula (5): (5) The calculation formula of Kalman gain is shown in the following formula (6): (6) In the above formulas (4) to (6), represents the prior state estimate at time k, express The posterior state estimate at time t, express System input at any time, represents the nonlinear state transfer function of the system, express The prior covariance estimate of time, express The posterior covariance estimate at time , express exist The Jacobian matrix at , express The process noise covariance matrix at time , represents the observation matrix, represents the measurement noise covariance matrix. is the observation matrix of the kth time step, and its mathematical expression is:

[0115] in, is the observation vector of the kth time step, which contains the data collected by the sensor in real time (such as temperature, vibration, and pressure signals). is the system state vector at the kth time step, representing the key state parameters of the braking system (such as brake clearance, braking time, and friction coefficient). is the observation matrix, which defines the linear mapping relationship from the state vector to the observation vector, that is, Represents the predicted sensor reading based on the current state. is the measurement noise at the kth time step, which represents the random error in the sensor measurement process.

[0116] In this embodiment, the EKF uses the acceleration signal as the observation value and continuously optimizes the state estimate through iterative calculations. By adjusting the process noise covariance matrix Q and the measurement noise covariance matrix R, the filter can be adapted to the noise characteristics of the actual system, effectively suppressing the integration error and ultimately outputting an accurate brake gap estimate.

[0117] Step 2.2: Calculate the brake time.

[0118] The braking time is calculated based on the smoother and more accurate speed data obtained through Kalman filtering. The braking time is defined as the time required for the speed to increase from 0 to the maximum value and then decrease from the maximum value to 0. The specific calculation steps include: (1) Determine the speed threshold: Set the speed threshold to 10% of the maximum speed, that is, ,in is the maximum speed during braking.

[0119] (2) Identify the speed change point: Identify the point where the speed first reaches Time point , the speed reaches Time point , and the final drop in speed Time point .

[0120] (3) Calculate braking time: braking time Calculated by the following formula (7): =( (7) That is, braking time is based on the duration of the acceleration phase ( ) and the deceleration phase duration ( ) and then divided by the coefficient b. The value of the coefficient b is obtained based on a large amount of actual data analysis, for example, the value of the coefficient b is 0.8.

[0121] Through the above method, the present invention can accurately calculate the braking time of the brake, providing an important basis for brake performance evaluation and fault diagnosis.

[0122] Step 2.3: Brake state estimation based on data fusion In this embodiment, a data fusion method is used to combine the brake gap value, braking time, and synchronously measured temperature values, and a Markov jump system (MJS) model is introduced for comprehensive judgment, thereby comprehensively evaluating the brake status. The specific steps include: (1) Markov jump system (MJS) model construction: The different states of the brake (such as normal, stuck, insufficient braking force, overheating failure, etc.) are modeled as different states of the Markov chain. The state transition probability matrix is ​​defined to describe the possibility of random jumps between different modes of the brake state.

[0123] (2) Data fusion and state estimation: In the data fusion process, not only the fusion of sensor data is considered, but also the probability of each Markov state is calculated in combination with the MJS model. Multi-sensor data is fused using methods such as weighted averaging, Kalman filtering, or neural networks, while considering the possibility of brake state jumps. The fused data contains comprehensive information on brake clearance, braking time, and temperature values, as well as the possibility of brake state jumps, which is used to quantitatively evaluate the overall state of the brake.

[0124] (3) Calculation of brake state probability: Based on the MJS model, the probability of the brake being in each state is calculated. Specifically, the sensor data and the state transition probability matrix are combined to update the state probability using Bayesian theorem or similar methods.

[0125] Among them, the possibility of the brake state jump is based on the prior probability of historical data and physical laws (such as brake failure mechanism, historical fault statistics), describing the randomness of state transition (such as ), which can predict the evolution path of potential faults and build a risk warning baseline; the brake state possibility is the posterior probability of integrating real-time sensor data, quantifying the probability of being in each state. It can locate the current fault state in real time and provide an accurate diagnosis basis.

[0126] During fault diagnosis, the brake state transition probability provides the basis for calculating the brake state probability, reflecting the probability of the brake transitioning between different states. Brake state probability, based on the transition probability and combined with real-time sensor data, quantitatively assesses the probability of the brake currently being in each state. The two factors interact and contribute to the fault diagnosis process: transition probability is used to predict potential fault risks, while state probability is used to accurately assess the current state, providing support for fault diagnosis and decision-making.

[0127] (4) Brake state estimation and decision-making: Based on the fused data and state probability, the brake state is estimated. A threshold or decision rule is set. When the probability of a certain state exceeds the threshold, the brake is determined to be in that state. Based on the brake state, appropriate maintenance measures or alarms are taken.

[0128] (5) Judgment of possible brake problems: In an exemplary embodiment, a dual-indicator fault judgment criterion is constructed by analyzing the characteristic combination of the brake gap and the braking time. For example: a. When the brake gap is too small and the braking time is short, and the possibility of a stuck state is high, it is judged that the brake has a stuck phenomenon; b. When the brake gap is too large and the braking time is long, and the possibility of an insufficient braking force state is high, it is judged that the brake has insufficient braking force.

[0129] In another exemplary embodiment, possible problems with the brake can be analyzed based on a combination of temperature and braking time. For example, when the temperature is too high and the braking time is prolonged during braking, and the possibility of an overheating failure state is high, it is determined that the brake is at risk of overheating failure.

[0130] In some embodiments, a comprehensive analysis of possible brake problems can be performed by combining multiple indicators, for example, combining multiple indicators such as brake clearance, braking time, and temperature values, as well as the possibility of the brake state, to determine other possible brake problems.

[0131] Step S223: Fault mode identification: Based on the brake state estimated in step S222, the probability of each Markov state is calculated to identify the current fault mode of the system.

[0132] In this embodiment, the specific pattern recognition method includes carefully comparing the estimated system state with each preset state in the Markov jump system model and, in conjunction with the state transition matrix, calculating the probability of each Markov state. By deeply analyzing the differences between the estimated state and the actual measured value, the current system fault mode can be accurately identified.

[0133] The estimated value of the current system state obtained in step S222 provides the basis for calculating the probability of a Markov state. The estimated value of the current system state is a preliminary estimate / approximation of the system's current state based on sensor data and techniques such as Kalman filtering. The Markov state probability, based on this estimate, combines the Markov jump system model and the state transition matrix to further refine and quantify the probability of the system being in each preset state. The two are interdependent, completing the process from preliminary state estimation to accurate fault mode identification.

[0134] In an exemplary embodiment, assuming that the system is in a normal operating state ( ) The corresponding brake temperature should be within the normal range. In the data acquisition step, the sensor collects the brake temperature data in real time as the actual measurement value. Based on the collected data, the Kalman filter and other observer technologies are used to estimate the current system state. Assume that (Normal operating state). However, if the actual measured value shows that the brake temperature is abnormally high and exceeds the normal range, the estimated state ( ) and the actual measured value. By deeply analyzing this difference and combining the state transfer matrix and Markov jump system model, it can be determined that the system may have shifted from a normal operating state to a fault state, such as a minor fault state of a single brake caliper ( Furthermore, according to the identified failure mode, corresponding maintenance measures are taken, such as inspecting and replacing the faulty brake caliper.

[0135] Step S224: Fault location: Based on the identified fault mode, determine which brake caliper has failed and the severity of the fault.

[0136] In some implementations, fault features are extracted from pre-processed collected data (e.g., vibration signals) to comprehensively capture vibration information during braking, laying the foundation for subsequent intelligent identification and precise location of fault modes. Fault feature extraction includes deep mining of time-domain features, precise analysis of frequency-domain features, and the fusion of time-frequency analysis.

[0137] Specifically, deep mining of time-domain features involves extracting key parameters of the vibration signal, such as peak value, mean, variance, and kurtosis (i.e., vibration fault signatures). These parameters comprehensively reflect the intensity and variation patterns of the vibration signal during the braking process. In-depth analysis of these time-domain features provides a crucial basis for fault identification, helping to accurately determine the operating status of the brake caliper.

[0138] Furthermore, to more accurately analyze fault characteristics, this embodiment performs precise frequency domain analysis: Time-domain signals are converted to frequency-domain signals through a Fast Fourier Transform (FFT), extracting frequency-domain features such as the dominant frequency and spectral energy distribution. These features help accurately identify the presence of abnormal frequency components during braking, and thus accurately determine whether the brake caliper is faulty, providing strong support for fault location.

[0139] Furthermore, intelligent fault pattern recognition and precise location are achieved based on deeply extracted fault features. Specifically, advanced machine learning algorithms, such as support vector machines and neural networks, are used to deeply train and learn the extracted fault features, establishing a highly accurate fault pattern recognition model. The vibration signal characteristics of each brake caliper are input into this fault pattern recognition model, which automatically and accurately identifies the fault mode of the brake caliper. After identifying the fault mode, the model combines the vibration sensor data corresponding to each brake caliper and compares and analyzes the abnormality of each brake caliper's vibration signal (i.e., the degree of deviation of the fault characteristics from the normal state) to further determine which brake caliper has failed.

[0140] For example, during an elevator's braking process, the system detects a sudden and significant increase in the energy of a specific frequency band in the vibration signal of a particular brake caliper, while the vibration signals of the other brake calipers remain normal. This abnormal vibration signal characteristic is then fed into the fault pattern recognition model, which analyzes it and determines that it matches the fault pattern of a "stuck brake caliper." Therefore, the system determines that the brake caliper exhibiting the abnormal vibration signal has experienced a stuck fault, immediately triggering the appropriate alarm device and prompting maintenance personnel to inspect and maintain the brake caliper. Furthermore, the controller can be adjusted to compensate for the fault's effects.

[0141] In this embodiment, the constructed model is applied to the monitoring of the brake system, and the occurrence of faults can be predicted by real-time detection of the system status, and corresponding measures can be taken for prevention and maintenance.

[0142] The method for diagnosing and locating elevator multi-caliper brake faults according to Example 2 of the present invention constructs a Markov jump system model to accurately estimate the different brake states and identify the current system fault mode. Furthermore, by combining the vibration sensor data corresponding to each brake caliper and comparing and analyzing the abnormality of each caliper's vibration signal, the specific brake caliper at fault can be determined, achieving accurate fault location.

[0143] [Example 3] Figure 2 1 is a flow chart of a method for diagnosing and positioning an elevator multi-caliper brake based on a Markov jump system model according to embodiment 3 of the present invention.

[0144] like Figure 2 As shown, in embodiment 3 of the present invention, the method for diagnosing and positioning an elevator multi-caliper brake may include at least the following steps: Firstly, a physical model is established based on the physical structure of the multi-caliper disc brake and the geometric relationship between different calipers to evaluate the comprehensive reliability index of the brake.

[0145] Secondly, information such as brake pad friction, brake disc torque, and brake disc clearance is obtained for each caliper in a multi-disc brake (caliper C1, C2, ..., CN) as individual caliper data. This data is then integrated. Through fault experiments and tests, comprehensive high-speed elevator braking capacity evaluation results are collected under different caliper fault and actual conditions. A Markov jump system state transition model is used to explore the qualitative relationship between individual caliper faults and braking performance, achieving a "fault-performance" feature mapping. The overall performance of the multi-disc brake (such as braking force, stability, heat dissipation, durability, friction coefficient, wear rate, and noise) is determined based on parameters such as caliper shape, area, distribution, effective contact area, number of calipers, location, and material.

[0146] Finally, by designing a relationship mining and decision-making model based on causal reasoning, self-attention mechanisms, and decision trees, we were able to mine the reliability indicators of multi-caliper disc brakes and the relationship between single-caliper faults and brake performance. This allowed us to analyze the correlation between evaluation results and faults, determine whether a fault has occurred, and assess the fault risk level. In particular, the model simultaneously considers both fault type and location, providing effective data support for efficient inspection and maintenance of elevator brake systems.

[0147] The method for diagnosing and locating a multi-caliper brake for an elevator, as described in Example 3 of the present invention, uses comprehensive braking capacity assessment results from actual elevator braking, along with multi-caliper brake reliability indicators and a single-caliper fault-brake performance relationship model, to analyze whether one or more brake caliper faults are causing unsatisfactory braking performance. When an abnormality occurs during braking or exceeds a set maintenance threshold, an automatic alarm is generated and a fault analysis is performed, accurately indicating the specific fault location. This provides support for routine inspection and maintenance of elevator braking systems using multi-caliper brakes.

[0148] [Example 4] Figure 3 1 is a flow chart of a method for diagnosing and positioning an elevator multi-caliper brake based on a Markov jump system model according to embodiment 4 of the present invention.

[0149] like Figure 3 As shown, in embodiment 4 of the present invention, the method for implementing the diagnosis and positioning of the elevator multi-caliper brake may at least include the following steps S31, S32, S33, S34, S35, S36 and S37, and these steps are described in detail below.

[0150] In step S31, the fault diagnosis system state and parameters are initialized. The core of the fault diagnosis system is a Markov jump model. In this embodiment, the Markov jump model is established based on the methods described in Examples 1 to 3 to describe the characteristic mapping between a single brake caliper fault and the braking performance of a multi-disc brake. This allows for the performance evaluation of the multi-disc brake to be determined based on collected sensor data. The correlation between the performance evaluation results and the fault is analyzed to determine whether a fault has occurred, assess the fault risk level, and ultimately determine the fault type and location, providing effective data support for efficient inspection and maintenance of elevator braking systems.

[0151] In step S32, real-time sensor data is collected and pre-processed. The sensor collects operating parameters of the multi-caliper brake, including but not limited to one or more of the following: braking torque, friction, clearance, and response time.

[0152] In step S33 , the observer is used to perform state estimation based on the preprocessed sensor data.

[0153] In step S34, it is determined whether there is a significant difference between the estimated state and the actual measured value. If there is no significant difference between the estimated state and the actual measured value, the process returns to step S32; if there is a significant difference between the estimated state and the actual measured value, the process proceeds to preliminary fault detection in step S35.

[0154] In step S35, preliminary fault detection is performed to determine the type of fault.

[0155] In step S36, a fault isolation algorithm is executed to obtain a specific fault diagnosis result. In this embodiment, when a fault is detected, the dynamic equations and the state transition matrix are applied to the fault isolation algorithm. By analyzing the paths and probabilities of system state transitions, the faulty brake caliper is determined. The fault isolation algorithm precisely locates the faulty brake caliper by analyzing system state transition probabilities and sensor characteristics. Its core steps include: 1. Extracting time-domain and frequency-domain features of temperature, vibration, and pressure signals; 2. Inputting them into a fault pattern recognition model; and 3. Identifying the faulty component based on outliers. This fault isolation algorithm relies on a Markov state transition matrix and Kalman filter state estimation.

[0156] In step S37, appropriate measures are taken based on the fault diagnosis and location results. In this embodiment, a pattern recognition algorithm is used, combined with time-domain and frequency-domain feature extraction, to accurately locate the fault location. Furthermore, a decision recommendation is generated based on the fault location results, and appropriate control measures are implemented, such as activating the alarm system, adjusting elevator operating parameters, or performing maintenance.

[0157] In further embodiments, model parameters (such as the state transition matrix and system matrix) are updated based on the fault diagnosis and location results, and the process returns to step S32 to continue fault diagnosis and location. By continuously updating model parameters, the model can adapt to system changes over time, improving the efficiency and accuracy of fault diagnosis and location, thereby enabling real-time analysis of elevator safety risks, rapid early warning, and risk avoidance.

[0158] [Example 5] Figure 4 1 is a schematic diagram of the architecture of an elevator multi-caliper brake diagnostic system based on a Markov jump system model according to embodiment 5 of the present invention. The elevator braking system includes the multi-caliper brake, which includes a plurality of brake calipers.

[0159] like Figure 4 As shown, the diagnostic system includes: a model parameter acquisition module 410 , a data acquisition module 420 , a state estimation module 430 , a fault diagnosis module 440 , a fault location module 450 , a suggestion generation module 460 , and a display module 470 .

[0160] The model parameter acquisition module 410 is used to obtain a state transition matrix between various preset states in the Markov jump system state transition model.

[0161] In some embodiments, a Markov jump system state transition model is established based on the comprehensive reliability index of the multi-caliper brake, brake caliper failures, and comprehensive high-speed elevator braking capacity assessment results, describing the characteristic mapping between a single brake caliper failure and the braking performance of the multi-caliper brake. The multi-caliper brake physical model is established based on the physical structure of the multi-caliper brake and the geometric relationship between the calipers of the multi-caliper brake; and the comprehensive reliability index of the multi-caliper brake is obtained based on the multi-caliper brake physical model.

[0162] In a further embodiment, the state transition matrix is ​​updated according to the first operating data and the second operating data of the multi-caliper brake and maintenance information of the elevator.

[0163] The data acquisition module 420 is used to acquire first operating data of the multi-caliper brake, acquire second operating data of the multi-caliper brake in real time, and pre-process the first operating data and the second operating data.

[0164] In some embodiments, the first operating data and the second operating data described later include but are not limited to one or more of the following: temperature of the multi-caliper brake, vibration signal of the multi-caliper brake, pressure signal of the multi-caliper brake, and displacement signal of the brake caliper.

[0165] In this embodiment, sensors are installed to monitor the working parameters of the brake caliper, such as pressure, temperature, displacement, etc., and continuously record data during the operation of the brake system.

[0166] The state estimation module 430 is configured to obtain a current brake system state estimation value based on the preprocessed first operating data, and obtain an operating data range corresponding to the current brake system state estimation value.

[0167] In some embodiments, an observer technique is used to obtain an estimated value of the current braking system state based on the preprocessed first operating data and a dynamic equation, wherein the dynamic equation is a linear dynamic equation for describing the dynamic characteristics of the braking system.

[0168] In an optional embodiment, a Kalman filter or an extended Kalman filter technique is used to calculate the current brake system state estimate. Exemplarily, using the Kalman filter technique to calculate the current brake system state estimate includes: using the Kalman filter technique to calculate brake clearance and braking time based on the preprocessed first operating data; generating fused data based on the brake clearance and braking time and the temperature of the multi-caliper brake; and calculating the current brake system state estimate based on the fused data and a Markov jump system state transition model.

[0169] The fault diagnosis module 440 is configured to determine a fault type of the multi-caliper brake according to the current brake system state estimation value, the Markov jump system state transition model, and the state transition matrix when the preprocessed second operating data exceeds the operating data range.

[0170] In some embodiments, determining the fault type of the multi-caliper brake based on the current brake system state estimate, the Markov jump system state transition model, and the state transition matrix may include: comparing the current brake system state estimate with each preset state in the Markov jump system state transition model, and calculating the probability value of each Markov state in the Markov jump system state transition model in combination with the state transition matrix, and determining the fault type of the multi-caliper brake based on the probability value.

[0171] The fault location module 450 is used to locate the faulty brake caliper according to the following steps: extracting the fault characteristics of the vibration signal of each brake caliper in the preprocessed first operating data and the preprocessed second operating data, the fault characteristics including time domain fault characteristics and frequency domain fault characteristics; inputting the time domain fault characteristics and frequency domain fault characteristics of the vibration signal into a fault mode recognition model, and the fault mode recognition model outputs the fault mode of the brake caliper; and locating one or more faulty brake calipers according to the abnormality degree value of the fault characteristics of the vibration signal of each brake caliper.

[0172] The suggestion generating module 460 is configured to generate an operation suggestion based on the faulty one or more brake calipers.

[0173] The display module 470 is used to display the status of the brake system, faulty brake calipers, and operation suggestions through a graphical interface. By visually displaying the real-time status of the brake system, fault details, and operation suggestions, it is convenient for operation and maintenance personnel to inspect and maintain the elevator brake system.

[0174] The diagnostic system of Example 5 of the present invention, based on a Markov jump system state transition model, uses the operating data of the multi-caliper brake to obtain an estimated brake system state. When there is a discrepancy between the operating data and the estimated state, the system accurately detects the fault type or potential fault of the multi-caliper brake, providing support for the routine inspection and maintenance of elevator brake systems using multi-caliper brakes. Furthermore, the faulty brake caliper can be precisely located.

[0175] [Example 6] Figure 5 1 is a schematic diagram of a layer framework of an elevator multi-caliper brake fault diagnosis system based on a Markov jump system model according to a sixth embodiment of the present invention; Figure 6 yes Figure 5 Schematic diagram of the processing operations performed by each hierarchical architectural component in the fault diagnosis system shown.

[0176] like Figure 5 As shown, the fault diagnosis system includes: a sensor layer 510, a data acquisition layer 520, a state estimation layer 530, a fault diagnosis and positioning layer 540, a decision and control layer 550, a user interface layer 560 and a model update and optimization module 570.

[0177] The sensor layer 510 deploys multiple types of sensors (such as vibration, temperature, and pressure) to collect real-time physical parameters of the brake caliper. This sensor layout is optimized using a multi-point layout strategy, as well as sensor calibration and maintenance strategies.

[0178] The data acquisition layer 520 is used to pre-process the collected original signals (denoising, normalization), extract time domain / frequency domain features (peak value, spectrum energy, etc.) and store them.

[0179] The state estimation layer 530 uses an extended Kalman filter (EKF) to calculate key state parameters (brake clearance, braking time); defines the Markov discrete state space (normal, minor fault, moderate fault, major fault, etc.) and the state transition probability matrix, and converts the EKF parameter estimation results into the likelihood of each discrete state through feature mapping; based on the Bayesian recursive framework, it integrates the Markov state transition prior and the EKF likelihood information to calculate the posterior probability of the brake being in each discrete state, and finally uses the state corresponding to the maximum posterior probability as the system state estimate.

[0180] The fault diagnosis and location layer 540 uses advanced machine learning algorithms, such as support vector machines and neural networks, to conduct in-depth training and learning on the extracted fault features and establish a high-precision fault pattern recognition model; the vibration signal characteristics of each brake caliper are input into the fault pattern recognition model, and the model can automatically and accurately identify the fault mode of the brake caliper; after identifying the fault mode, combined with the vibration sensor data corresponding to each brake caliper, the abnormality of each brake caliper vibration signal (that is, the degree of deviation between the fault characteristics and the normal state) is compared and analyzed to determine the brake caliper with a specific fault.

[0181] Furthermore, based on the MJS model, the state space is defined ( ~ ) and the transition probability matrix, and dynamically adjust the transition probability matrix in combination with historical data.

[0182] The decision-making and control layer 550 triggers alarms, adjusts braking force distribution, and provides maintenance recommendations based on the diagnostic results. The alarm module in the decision-making and control layer uses a threshold-triggered mechanism to generate alarms and associates corresponding control strategies based on the alarm level (fault risk level). These control strategies include, but are not limited to, adaptive force distribution (for example, adjusting the braking force of other calipers in the event of a jam) and fault-tolerant control (for example, switching to a degraded control mode in the event of a fault to ensure basic braking). Maintenance recommendations include fault location and severity (for example, "Brake caliper C2 is moderately worn; replacement recommended within 72 hours").

[0183] The user interface layer 560 is used to visually display real-time brake status, fault details (fault type and location, risk assessment level), and operation logs, and supports human-computer interaction. The operation logs can be used for historical data query and maintenance record management.

[0184] The model update and optimization module 570 continuously optimizes the state transfer matrix and algorithm parameters through online learning to improve the robustness of the system.

[0185] like Figure 6 As shown, Figure 5The components of each hierarchical structure in the fault diagnosis system shown implement fault diagnosis and location of elevator multi-caliper brake by performing the following processing operations:

[0186] Operation S61: The sensor layer monitors the working parameters of the brake caliper in real time, such as pressure, temperature, displacement, vibration, etc., to provide basic data for subsequent fault diagnosis.

[0187] Operation S62: The data acquisition layer collects the data transmitted by the sensor layer and performs preprocessing, such as denoising and normalization, to improve data quality.

[0188] Operation S63: The state estimation layer uses an observer technology such as a Kalman filter to perform state estimation based on the data transmitted by the data acquisition layer to obtain an estimated value of the current system state.

[0189] Operation S64: The fault diagnosis and localization layer detects potential faults based on a Markov jump system model by comparing the difference between the estimated state and the actual measured value. Once a fault is detected, further analysis is performed to determine which brake caliper is faulty (fault isolation). The fault location is accurately located (fault localization) through a pattern recognition algorithm combined with time and frequency domain feature extraction.

[0190] Operation S65: The decision and control layer generates decision suggestions based on the results transmitted by the fault diagnosis and positioning layer, such as alarm, adjustment of operating parameters or maintenance, and implements corresponding control measures, such as activating the alarm system, adjusting elevator operating parameters, etc.

[0191] Operation S66: The user interface layer provides a graphical interface to display system status, fault information, and operation suggestions, allowing operators to view system status, receive alarm information, and perform corresponding operations based on operation suggestions.

[0192] [Example 7] Figure 7 2 is a schematic diagram of a disc brake measurement structure according to embodiment 7 of the present invention.

[0193] like Figure 7As shown, the vibration sensor 10 is magnetically fixed to the brake. Block brakes use eight sensors mounted on the four corners of each brake block. Disc brakes use a corresponding number of sensors attached directly to the brake. Drum brakes use four sensors mounted on the brake arm and brake spring. Each sensor transmits vibration signals to the touch terminal 20 via a wired connection. The temperature sensor 30 uses non-contact measurement, directly aligned with the brake shoe, and transmits temperature data wirelessly. The coil voltage detection module 40 is connected in parallel to the voltage lines at both ends of the brake coil to obtain voltage and transmit data via a wired connection. During testing, it is necessary to check whether each sensor is securely fixed and whether the signal transmission is normal. By simulating vibration, heating the brake shoe, and powering the coil, the touch terminal is verified to verify that it can accurately receive and display vibration, temperature, and voltage data. Those skilled in the art will also understand that the specific measurement and installation procedures require the selection and fixing of sensors based on the brake type.

[0194] In some implementations, vibration sensors, temperature sensors, and coil voltage detection modules serve as the sensor layer to monitor and collect brake operating parameters in real time. These parameters are then uploaded to a host computer in the data collection layer for collection and preprocessing. Furthermore, the preprocessed data is used to implement the diagnostic and positioning methods described in Examples 1 through 4, and the processing operations performed by the various hierarchical components in Example 6 are also performed based on the preprocessed data.

[0195] The structure of the touch terminal 20 is as follows: Figure 8 As shown, the touch terminal 20 can be mainly composed of a lower shell 1, an acquisition module circuit board 2, a lithium battery 3, a rubber sleeve 4, a touch screen fixing plate 5, a touch screen 6, an upper shell 7 and a panel 8.

[0196] Among them, the lower shell 1 serves as the basic supporting structure of the equipment, carrying all internal components. The material of the lower shell 1 is ABS. ABS material is a thermoplastic engineering plastic with excellent wear resistance, impact resistance and chemical corrosion resistance, making the lower shell both light and durable.

[0197] The acquisition module circuit board 2 is the functional core of the device, which is used to collect and process vibration data, temperature data and coil voltage data, and transmit the data and results to the touch screen for user operation.

[0198] The lithium battery 3 provides the necessary power for the device, ensuring its independent operation. For example, a 12V lithium battery is used. This battery has advantages such as high energy density, long life, fast charging, and environmental friendliness. It can stably power the acquisition module circuit board and touch screen.

[0199] The material of the rubber sleeve 4 is PBT, which can play a buffering and protective role to prevent internal components from being impacted.

[0200] The material of the touch screen fixing plate 5 is aluminum-6061, and is used to fix the touch screen.

[0201] The touch screen 6 is used to provide a user interface for operation and information display. The user can input instructions through the touch screen, and the device can also display data or status information through the touch screen.

[0202] The upper shell 7 cooperates with the lower shell 1 to form a complete device shell to protect the internal components. Like the lower shell 1, the upper shell 7 is also made of ABS material, which has the characteristics of lightness, beauty and durability.

[0203] Panel 8 is used to cover the touch screen, protect the screen and provide a clear display effect. In addition, the material of panel 8 is transparent acrylic. The acrylic panel has the characteristics of high transparency, wear resistance, and easy processing and molding. It can not only protect the touch screen from scratches, but also ensure that the user can clearly see the information on the screen.

[0204] Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented by combining software with a hardware platform. Based on this understanding, all or part of the technical solution of the present invention that contributes to the background art can be embodied in the form of a computer software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for enabling a computer device (such as a personal computer, server, or network device) to execute the methods described in various embodiments of the present invention, or portions thereof.

[0205] Correspondingly, embodiments of the present invention further provide a computer-readable storage medium having computer-readable instructions or a program stored thereon. When executed by a processor, the computer-readable instructions or program causes the computer to perform the following operations, which include the steps included in the diagnostic method described in any of the above embodiments, which are not further described here. The storage medium may include, for example, an optical disk, a hard disk, a floppy disk, a flash memory, a magnetic tape, and the like.

[0206] In addition, embodiments of the present invention further provide a computer device comprising a memory and a processor, wherein the memory is configured to store one or more computer-readable instructions or programs, wherein the one or more computer-readable instructions or programs, when executed by the processor, can implement the diagnostic method described in any of the above embodiments. The computer device can be, for example, a server, a desktop computer, a laptop computer, a tablet computer, or the like.

[0207] The present invention also provides a computer program product comprising a computer program, wherein the computer program includes program code for executing the diagnostic method shown in the flowchart. When the computer program product is executed in a computer system, the program code is used to enable the computer system to implement the diagnostic method provided by the embodiment of the present disclosure.

[0208] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect via the Internet).

[0209] Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may be modified or some of the technical features thereof may be replaced with equivalents. Such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A diagnostic method for an elevator multi-caliper brake, wherein the elevator's braking system includes a multi-caliper brake, which includes a plurality of brake calipers, and wherein: The diagnostic method comprises: Obtaining a state transfer matrix between various preset states in a Markov jump system state transfer model; Acquiring first operating data of the multi-caliper brake and performing preprocessing; Obtaining a current brake system state estimate value based on the preprocessed first operating data, and obtaining an operating data range corresponding to the current brake system state estimate value; collecting second operating data of the multi-caliper brake in real time and performing preprocessing; When the preprocessed second operating data exceeds the operating data range, the fault type of the multi-caliper brake is determined according to the current brake system state estimation value, the Markov jump system state transition model and the state transition matrix.

2. The diagnostic method according to claim 1, wherein The diagnostic method further comprises: Establishing a physical model of the multi-caliper brake based on the physical structure of the multi-caliper brake and the geometric relationship between the calipers of the multi-caliper brake; A comprehensive reliability index of the multi-caliper brake is obtained according to the multi-caliper brake physical model.

3. The diagnostic method according to claim 2, wherein The diagnostic method further comprises: Based on the comprehensive reliability index of the multi-caliper brake, brake caliper failure and high-speed elevator comprehensive braking capacity evaluation results, the Markov jump system state transition model is established to describe the characteristic mapping between a single brake caliper failure and the braking performance of the multi-caliper brake.

4. The diagnostic method according to claim 1, wherein The diagnostic method further includes updating the state transition matrix according to the first operating data and the second operating data of the multi-caliper brake and the maintenance information of the elevator.

5. The diagnostic method according to claim 1, wherein The first operating data and the second operating data include: a temperature of the multi-caliper brake, a vibration signal of the multi-caliper brake, and a pressure signal of the multi-caliper brake.

6. The diagnostic method according to claim 1, wherein An observer technique is used to obtain a current brake system state estimate value based on the preprocessed first operating data and a dynamic equation, wherein the dynamic equation is a linear dynamic equation for describing the dynamic characteristics of the brake system.

7. The diagnostic method according to claim 6, wherein The linear dynamic equation is: in, is a state vector, the state vector including state information of the braking system; is the input vector; is an output vector, which includes readings of various acquisition sensors; 、 、 is a Markov jump system dynamic characteristic matrix, wherein the Markov jump system dynamic characteristic matrix is ​​related to a preset state in a Markov jump system state transition model; is the process noise; is the measurement noise.

8. The diagnostic method according to claim 7, wherein The state transition probability in the state transition matrix is ​​updated according to the input vector.

9. The diagnostic method according to claim 6, wherein The current braking system state estimation value is calculated using a Kalman filter or an extended Kalman filter technology.

10. The diagnostic method according to claim 9, wherein Calculating the current brake system state estimate using the Kalman filter technique includes: Using Kalman filter technology, calculating the brake clearance and braking time according to the preprocessed first operating data; generating fusion data according to the brake clearance and the braking time and the temperature of the multi-caliper brake; The current braking system state estimation value is calculated based on the fused data and a Markov jump system state transition model.

11. The diagnostic method according to claim 1, wherein Determining the fault type of the multi-caliper brake according to the current brake system state estimation value, the Markov jump system state transition model and the state transition matrix includes: The current brake system state estimate is compared with each preset state in the Markov jump system state transition model, and the probability value of each Markov state in the Markov jump system state transition model is calculated in combination with the state transition matrix, and the fault type of the multi-caliper brake is determined based on the probability value.

12. The diagnostic method according to claim 1, wherein The diagnostic method further includes: after determining the fault type, locating the faulty brake caliper according to the following steps: Extracting fault features of the vibration signal of each brake caliper in the preprocessed first operating data and the preprocessed second operating data, the fault features including time domain fault features and frequency domain fault features; Inputting the time domain fault features and frequency domain fault features of the vibration signal into a fault mode recognition model, the fault mode recognition model outputting a fault mode of the brake caliper; and One or more faulty brake calipers are located according to the abnormality level of the fault characteristic of the vibration signal of each brake caliper.

13. The diagnostic method according to claim 12, wherein The time domain fault characteristics of the vibration signal include: peak value, mean value, variance and kurtosis of the vibration signal; The frequency domain fault characteristics of the vibration signal include: main frequency and spectrum energy distribution.

14. The diagnostic method according to claim 12, wherein The diagnostic method further includes generating an operational recommendation based on the faulty brake caliper or calipers after locating the faulty brake caliper.

15. The diagnostic method according to claim 14, wherein The diagnostic method further includes: displaying the status of the brake system, the faulty brake caliper, and operation suggestions through a graphical interface.

16. An elevator multi-caliper brake diagnostic system, characterized in that: The diagnostic system is used to implement the diagnostic method for an elevator multi-caliper brake according to any one of claims 1 to 15, and the diagnostic system includes: A model parameter acquisition module is used to obtain the state transfer matrix between each preset state in the Markov jump system state transfer model; a data acquisition module, configured to acquire first operating data of the multi-caliper brake, acquire second operating data of the multi-caliper brake in real time, and pre-process the first operating data and the second operating data; a state estimation module, configured to obtain a current brake system state estimation value based on the preprocessed first operating data, and obtain an operating data range corresponding to the current brake system state estimation value; A fault diagnosis module is used to determine the fault type of the multi-caliper brake according to the current brake system state estimation value, the Markov jump system state transition model and the state transition matrix when the preprocessed second operating data exceeds the operating data range.

17. The diagnostic system according to claim 16, wherein: The diagnostic system further includes a fault location module configured to locate a faulty brake caliper according to the following steps: Extracting fault features of the vibration signal of each brake caliper in the preprocessed first operating data and the preprocessed second operating data, the fault features including time domain fault features and frequency domain fault features; Inputting the time domain fault features and frequency domain fault features of the vibration signal into a fault mode recognition model, the fault mode recognition model outputting a fault mode of the brake caliper; and One or more faulty brake calipers are located according to the abnormality level of the fault characteristic of the vibration signal of each brake caliper.

18. The diagnostic system according to claim 16, wherein: The diagnostic system further includes a recommendation generating module for generating an operating recommendation based on the faulty one or more brake calipers.

19. The diagnostic system according to claim 18, wherein The diagnostic system further includes a display module for displaying the status of the brake system, faulty brake calipers, and operation suggestions through a graphical interface.

20. A computer-readable storage medium storing computer-readable instructions, characterized in that: The computer-readable instructions are executed by a processor to implement the diagnostic method according to any one of claims 1 to 15.

21. A computer device comprising a memory and a processor, The memory stores computer-readable instructions, characterized in that: The processor executes the computer-readable instructions to implement the diagnostic method according to any one of claims 1 to 15.

22. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the diagnostic method according to any one of claims 1 to 15 is implemented.

Images