Elevator brake response performance monitoring method, system and device

Abstract

The application relates to the elevator technical field and provides an elevator brake response performance monitoring method, system and device.The method comprises the following steps: acquiring a coil current signal of an electromagnetic coil of a brake and a vibration signal of a brake body after a brake instruction is issued; acquiring a time-frequency spectrum of the current signal and a time-frequency spectrum of the vibration signal, and acquiring a complex conjugate of the time-frequency spectrum of the current signal; acquiring a cross-spectral density instantaneous value of the time-frequency spectrum of the current signal and the time-frequency spectrum of the vibration signal according to the complex conjugate and the time-frequency spectrum of the vibration signal, and performing smoothing processing on the cross-spectral density instantaneous value; acquiring a brake response coherence index according to the cross-spectral density instantaneous value after the smoothing processing, and monitoring the elevator brake response performance according to the brake response coherence index.The application uses the coherence loss of the electrical and mechanical responses as a unified degree of brake degradation, quantifies the dynamic quality of the brake response process, and is beneficial to improving the accuracy of the brake response performance monitoring.

Description

Technical Field

[0001] This invention relates to the field of elevator technology, and more specifically, to a method, system, and device for monitoring elevator braking response performance. Background Technology

[0002] An elevator brake is a core safety device in elevators, belonging to the field of mechanical engineering. Primarily used in elevator systems, it achieves power-off braking through bidirectional electromagnetic thrust. When energized, the braking mechanism separates from the motor; after power is cut off, the brake spring is compressed to generate frictional braking force. It features a compact structure, low noise, and sensitive operation, and is often used in conjunction with escalator traction machines. Elevator braking response performance refers to the dynamic response capability of the elevator brake from receiving a power-off command to generating effective braking torque; its core indicator is the braking response time.

[0003] Traditional elevator braking response performance monitoring technologies mostly rely on limit switches or single current thresholds for judgment. These methods struggle to quantify the dynamic quality of the response process, such as armature movement delay, jamming, and friction fade. Therefore, there is room for improvement in the accuracy of brake response performance monitoring. Summary of the Invention

[0004] Based on this, in order to improve the accuracy of elevator brake response performance monitoring, the present invention provides a method and system for automatic detection and early warning of home elevators, the specific technical solution of which is as follows: A method for monitoring elevator braking response performance includes the following steps: The coil current signal of the brake electromagnetic coil and the vibration signal of the brake body are obtained after the braking command is issued. Fourier transforms are performed on the coil current signal and the vibration signal respectively to obtain the time spectrum of the current signal and the time spectrum of the vibration signal, and the complex conjugate of the time spectrum of the current signal is obtained. The instantaneous cross-spectral density values ​​of the current signal time spectrum and the vibration signal time spectrum are obtained based on the complex conjugate and the vibration signal time spectrum, and the instantaneous cross-spectral density values ​​are smoothed. The braking response coherence index, which characterizes the degree of electromechanical coupling, is obtained based on the instantaneous value of the cross-spectral density after smoothing, and the braking response performance of the elevator is monitored based on the braking response coherence index.

[0005] The elevator braking response performance monitoring method acquires coil current signals and brake body vibration signals, obtains instantaneous cross-spectral density values ​​based on their time-frequency domain characteristics, and then obtains the braking response coherence index used to characterize the degree of electromechanical coupling. It uses the coherence loss of electrical and mechanical responses as a unified measure of braking degradation, quantifies the dynamic quality of the brake response process, and helps to improve the accuracy of brake response performance monitoring.

[0006] Preferably, the specific method for obtaining the braking response coherence index includes the following steps: Obtain the squared modulus of the instantaneous cross-spectral density after smoothing, and denot it as the squared modulus of the cross-power spectrum; The modulus square of the time spectrum of the current signal is smoothed to obtain the observed value of the total power spectrum of the current. At the same time, the modulus square of the time spectrum of the vibration signal is smoothed to obtain the observed value of the total power spectrum of the vibration. Obtain the first noise power spectrum estimate of the current channel, and correct the observed total current power spectrum based on the first noise power spectrum estimate to obtain the corrected current power spectrum. The second noise power spectrum estimate of the vibration channel is obtained, and the observed total vibration power spectrum is corrected based on the second noise power spectrum estimate to obtain the corrected vibration self power spectrum. The time-frequency domain amplitude squared coherence value is obtained based on the cross-power spectrum modulus square, the current self-power spectrum, and the vibration self-power spectrum. The braking response coherence index is obtained based on the time-frequency domain amplitude squared coherence value.

[0007] Preferably, obtaining the time-frequency domain amplitude square coherence value specifically includes: normalizing the square of the cross-power spectrum modulus based on the product of the current self-power spectrum and the vibration self-power spectrum to obtain the time-frequency domain amplitude square coherence value.

[0008] Preferably, obtaining the braking response coherence index specifically includes the following steps: The characteristic time-frequency domain of the braking action is obtained, and the average value of the squared coherent value of the time-frequency domain amplitude is integrated in the characteristic time-frequency domain to obtain the time-frequency domain coherent average value used to characterize the degree of electromechanical coupling. The braking response coherence index is obtained based on the time-frequency domain coherence average value.

[0009] Preferably, obtaining the braking response coherence index based on the time-frequency domain coherence average value specifically includes the following steps: Obtain the measured phase standard deviation of phase values ​​at all time-frequency points within the same characteristic time-frequency domain, and the benchmark value of the phase standard deviation within the same characteristic time-frequency domain when the brake is in a healthy state; The phase penalty value used to characterize the phase stability of the electromechanical response is obtained based on the measured phase standard deviation and the phase standard deviation benchmark value. The braking response coherence index is obtained based on the time-frequency domain coherence average value and the phase penalty value.

[0010] Preferably, obtaining the phase penalty value specifically includes the following steps: The phase fluctuation ratio is obtained based on the measured phase standard deviation and the phase standard deviation benchmark value. The phase fluctuation ratio is calculated in reverse to obtain a phase penalty value that is negatively correlated with the phase fluctuation ratio.

[0011] Preferably, obtaining the braking response coherence index specifically involves obtaining the braking response coherence index based on the weighted sum of the time-frequency domain coherence average value and the phase penalty value.

[0012] Preferably, the smoothing process specifically involves performing a sliding weighted average along a preset smoothing window length in the frequency domain.

[0013] An elevator braking response performance monitoring system, used to implement the aforementioned elevator braking response performance monitoring method, includes: The signal acquisition module is used to acquire the coil current signal of the brake electromagnetic coil and the vibration signal of the brake body after the braking command is issued. The signal processing module is used to perform Fourier transform on the coil current signal and the vibration signal respectively, obtain the time spectrum of the current signal and the time spectrum of the vibration signal, obtain the complex conjugate of the time spectrum of the current signal, obtain the instantaneous cross-spectral density value of the time spectrum of the current signal and the time spectrum of the vibration signal based on the complex conjugate and the time spectrum of the vibration signal, and perform smoothing processing on the instantaneous cross-spectral density value. The performance monitoring module is used to obtain the braking response coherence index, which characterizes the degree of electromechanical coupling, based on the smoothed instantaneous value of the cross-spectral density, and to monitor the braking response performance of the elevator based on the braking response coherence index.

[0014] An elevator braking response performance monitoring device, comprising: Controller; Memory, which stores executable instructions; The executable instructions run on the controller and implement the elevator braking response performance monitoring method. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the overall process of an elevator braking response performance monitoring method according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a specific method for obtaining the braking response coherence index in one embodiment of the present invention. Figure 3This is a schematic diagram of the process for obtaining the braking response coherence index in one embodiment of the present invention; Figure 4 This is a schematic diagram of the process for obtaining the braking response coherence index based on the squared coherence value of the time-frequency domain amplitude in one embodiment of the present invention; Figure 5 This is a schematic diagram of the process for obtaining the phase penalty value in one embodiment of the present invention; Figure 6 This is a schematic diagram of the overall structure of an elevator braking response performance monitoring system according to an embodiment of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to its embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of protection of the invention.

[0017] like Figure 1 As shown, an embodiment of the present invention provides a method for monitoring elevator braking response performance, comprising the following steps: S1, acquire the coil current signal of the brake electromagnetic coil and the vibration signal of the brake body after the braking command is issued.

[0018] Specifically, a Hall current sensor can be installed in the power supply circuit of the brake's electromagnetic coil to record the coil current signal in real time after the braking command is issued. A broadband accelerometer can also be installed on the brake body or brake arm to synchronously acquire vibration signals. The sampling frequency should be no less than 10 kHz to ensure effective capture of transient characteristics.

[0019] S2, perform Fourier transform on the coil current signal and vibration signal respectively to obtain the time spectrum of the current signal and the time spectrum of the vibration signal, and obtain the complex conjugate of the time spectrum of the current signal.

[0020] First, the coil current signal and vibration signal are subjected to DC removal and low-pass filtering to retain effective feature components. After low-pass filtering, a short-time Fourier transform is performed on the coil current signal and vibration signal to obtain the time spectrum of the current signal and the time spectrum of the vibration signal.

[0021] The time spectrum of this current signal characterizes the energy distribution and phase state of the coil current at different times and frequencies during the braking process, reflecting the transient changes of electromagnetic excitation. It includes transient information for each stage of engagement, holding, and release. Generally, the time spectrum of the current signal corresponds to a complex number at each time frequency point, including amplitude and phase information.

[0022] The time spectrum of the vibration signal can characterize the energy distribution and phase state of the mechanical structure vibration at different times and frequencies during braking, reflecting mechanical dynamics such as armature impact, friction, and spring action. This complex conjugate is used to preserve phase difference information when calculating the instantaneous cross-spectral density of the time spectrum of the current signal and the time spectrum of the vibration signal.

[0023] S3. Obtain the instantaneous cross-spectral density values ​​of the current signal time spectrum and the vibration signal time spectrum based on the complex conjugate and the vibration signal time spectrum, and smooth the instantaneous cross-spectral density values.

[0024] Smoothing the instantaneous value of the cross-spectral density is intended to reflect the true linear correlation between the two signals at that time-frequency point.

[0025] S4. Obtain the braking response coherence index, which characterizes the degree of electromechanical coupling, based on the instantaneous value of the cross-spectral density after smoothing, and monitor the braking response performance of the elevator based on the braking response coherence index.

[0026] The instantaneous cross-spectral density value after smoothing can be normalized first. Then, the normalized instantaneous cross-spectral density value can be weighted and integrated within the characteristic time-frequency domain. Finally, the weighted integral value can be normalized based on the characteristic time-frequency domain to obtain the braking response coherence index.

[0027] The braking response coherence index quantitatively describes the degree of linear correlation between the coil current signal and the vibration signal at each point in the time-frequency domain. Generally speaking, 1 indicates complete coherence, and the lower the value, the less coherent the signal.

[0028] Here, the weights in the weighted integral are prior weights based on the normal state. These prior weights can be calibrated experimentally, for example, under the condition that the brake is healthy, based on the braking response coherence index, assigning heavier weights to time-frequency regions with higher coherence and more critical mechanical response. That is to say, the prior weights are positively correlated with the braking response coherence index under the condition that the elevator is operating normally and the braking response is good.

[0029] As a preferred technical solution, the weights in the weighted integral consist of two parts: frequency weights and asymmetric time weights, and are equal to their product. The frequency weights are Gaussian frequency band envelope windows, set to... ,in, The values ​​represent, in order, the frequency independent variable, the core characteristic frequency of electromechanical coupling under healthy conditions, and the bandwidth coefficient. The bandwidth coefficient is obtained statistically from healthy samples and controls the effective frequency range. The larger the value, the wider the bandwidth and the stronger the robustness; conversely, the smaller the value, the higher the sensitivity.

[0030] Specifically, the core characteristic frequency can be understood as the center frequency with the strongest current-vibration coherence in a healthy state. It corresponds to the inherent structural frequency of the brake arm and armature assembly, and is the frequency point where energy transfer is most concentrated and electromechanical coupling is tightest when electromagnetic force drives mechanical action. It is also the frequency band richest in degradation fault characteristics. Generally, the frequency corresponding to the peak value of the time-frequency coherence function can be obtained through 1-2 healthy braking tests. In engineering, it is usually 300-800Hz, with slight differences depending on the brake structure and model.

[0031] The bandwidth coefficient, measured in Hz, is a core parameter controlling the width of the Gaussian weighted window and determines the coverage range of the effective weighted frequency band. It characterizes the dispersion of the electromechanical coupling frequency band under healthy conditions; a larger value results in a wider effective weighted bandwidth, while a smaller value results in a narrower bandwidth and higher focus. The bandwidth is typically set between 50 and 150 Hz and can be flexibly adjusted according to on-site conditions. For older elevators or scenarios with strong interference, a bandwidth of 100–150 Hz can be used to broaden the bandwidth and improve robustness; for high-speed / heavy-load elevators or scenarios requiring high sensitivity, a bandwidth of 30–80 Hz can be used to narrow the bandwidth and improve fault resolution.

[0032] The deviation between the current frequency and the center frequency is squared to ensure that the weighting function is symmetrical about the core feature frequency. The further away from the center, the greater the attenuation.

[0033] The effective energy of the brake's electromechanical coupling is highly concentrated in a narrow band near the center frequency, while out-of-band frequencies such as traction machine pulsation below 100Hz, sensor high-frequency noise above 1500Hz, and power grid harmonic interference are mostly unrelated to electromechanical coupling. After weighting with a Gaussian window, the signal in the core coupling frequency band is preserved and given a high weight, while out-of-band noise components are exponentially suppressed, which is equivalent to achieving adaptive bandpass filtering in the frequency domain, significantly improving the signal-to-noise ratio of the final health indicator. In addition, in reality, power grid voltage fluctuations, car load changes, and ambient temperature drift will cause small normal shifts in characteristic frequencies, which are not faults. The Gaussian window here uses a smooth and continuous decay curve, so the weight only decreases slowly when the frequency shifts slightly, thus avoiding a precipitous drop.

[0034] Thus, based on this frequency weighting, the weighting of the healthy core coupling frequency band is preserved, while allowing for a small frequency shift, and out-of-band noise is suppressed.

[0035] To address the causal characteristics of the braking process, the asymmetric time weighting is divided into two sets of time constants: one for the pressure build-up phase and the other for the action phase, allowing for differentiated weighting. The asymmetric time weighting is set as follows: , These are represented, in order, as the health baseline action center time, the electromagnetic voltage build-up phase time constant, and the mechanical action phase time constant. The health baseline action center time serves as the time origin; the recommended time constant for the electromagnetic voltage build-up phase is 30-50 ms, which corresponds to a wider window width, covering the gradual process of current rise; the recommended time constant for the mechanical action phase is 10-20 ms, which corresponds to a narrower window width, allowing for high focus on the transient impacts of armature engagement and brake shoe contact, thus enhancing sensitivity to mechanical faults.

[0036] Specifically, the critical moment of the healthy braking baseline is typically the moment when the armature is fully engaged and impacts the limit switch, and the brake arm generates the maximum impact vibration. It is the time origin where electromechanical coupling is most intense and fault characteristics are most concentrated during the entire braking process. It can be automatically identified and calibrated through a healthy braking test by taking the peak moment of the vibration acceleration envelope or the extreme moment of the second derivative of the current.

[0037] The time constant of the electromagnetic pressure build-up phase is used to control the decay rate of the weights before the center of the health baseline action, i.e., the braking and initiation phase. When the coil is energized, the current gradually increases, the electromagnetic force slowly overcomes the spring force, and the armature begins to move smoothly. This is a gradually changing electromagnetic voltage build-up process, which lasts longer and changes more gently. Therefore, its value is larger, the time window is wider, and it completely covers the electromagnetic excitation build-up process.

[0038] The time constant of the mechanical motion phase is used to control the decay rate of the weight at the center moment of the health baseline motion (mechanical impact phase). At that moment, the armature engages and impacts, and the brake shoe contacts the brake wheel, which is a transient impact process on the order of milliseconds. Fault information (jamming, wear, uneven wear) is highly concentrated in a very short time. Therefore, its value is smaller and the time window is narrower. By highly focusing on the core impact transient, the sensitivity to mechanical faults can be maximized.

[0039] The overall asymmetric time weighting forms an asymmetric weighting pattern that is slow at the beginning and steep at the end, perfectly matching the physical time sequence characteristics of the braking process, which is slow at the beginning and fast at the end. The larger the value, the higher the contribution of the coherent data at that moment to the final health index. The closer the value is to 0, the more the background noise and steady-state signal at that moment are suppressed and hardly participate in the index calculation.

[0040] Generally, effective fault information from braking actions is concentrated in a narrow range near the center of the health baseline action. The steady-state standby segment before the action and the holding segment after the action are basically background noise and steady-state signals, which are of no value for degradation diagnosis. Therefore, by using exponential window weighting, only the coherent data in the core action range is given high weight, and the noise and interference in the steady-state segments before and after the action are exponentially suppressed, which can significantly improve the signal-to-noise ratio of health indicators and avoid irrelevant signals diluting fault characteristics.

[0041] The first half, electromagnetic voltage buildup, is a slow, gradual process lasting tens of milliseconds. A window that is too narrow will miss electromagnetic-side fault information such as coil aging and insufficient power supply. The second half, mechanical impact, is a transient process on the order of milliseconds. A window that is too wide will introduce subsequent residual vibrations and friction noise, reducing fault resolution. A wide window with asymmetric time weighting can cover the entire voltage buildup process and retain electromagnetic-side degradation characteristics, while a narrow window can focus on the impact transient and enhance sensitivity to mechanical faults.

[0042] Furthermore, by adjusting the two time constants separately, the monitoring capability for a specific type of fault can be enhanced. For example, if the focus is on monitoring electromagnetic coil aging or abnormal power supply voltage, the time constant of the electromagnetic voltage build-up stage can be increased to broaden the weight of the voltage build-up stage and strengthen the weight of electromagnetic side characteristics. If the focus is on monitoring mechanical jamming or friction plate wear, the time constant of the mechanical action stage can be reduced to narrow the impact stage window and further improve the sensitivity to transient faults. In scenarios with strong interference and large fluctuations in operating conditions in old elevators, both time constants can be increased simultaneously to improve the overall robustness of the indicators.

[0043] Under normal elevator operation and good braking response conditions, the characteristic time-frequency domain of braking action can be determined. Within this characteristic time-frequency domain, the braking response coherence index often exhibits a consistently high value, such as greater than 0.8, corresponding to the electromechanical strong coupling stage during the armature engagement / release process. Once the braking system degrades, such as armature jamming or guide sleeve dryness, mechanical actions will be delayed or discontinuous, the time axis position and range of the coherent time-frequency region will shift, and the braking response coherence index will decrease. If the spring is weak or the friction plate is worn, it can easily lead to changes in the vibration characteristic frequency of the closed impact, and the nonlinearity of the current-vibration transmission path will be enhanced, resulting in a decrease in high-frequency coherence. Aging of the electromagnetic coil will slow down the current build-up process, causing misalignment with the timing of normal vibration, and the coherence will exhibit attenuation characteristics.

[0044] Based on the obtained braking response coherence index and the preset coherence index threshold, such as 0.6 times the normal value, if the braking response coherence index is less than the coherence index threshold, it can be determined that the braking response performance has degraded and an early warning will be triggered.

[0045] The elevator braking response performance monitoring method acquires coil current signals and brake body vibration signals, obtains instantaneous cross-spectral density values ​​based on their time-frequency domain characteristics, and then obtains the braking response coherence index used to characterize the degree of electromechanical coupling. It uses the coherence loss of electrical and mechanical responses as a unified measure of braking degradation, quantifies the dynamic quality of the brake response process, and helps to improve the accuracy of brake response performance monitoring.

[0046] In one embodiment, such as Figure 2 As shown, the specific method for obtaining the braking response coherence index includes the following steps: S41, obtain the squared modulus of the instantaneous cross-spectral density after smoothing, and denot it as the squared modulus of the cross-power spectrum.

[0047] Specifically, the smoothed instantaneous cross-spectral density value characterizes the correlation energy and relative phase of the current and vibration signals at the corresponding time-frequency points. Squaring its absolute value yields the modulus-squared value of the cross-spectral density, i.e., the modulus-squared value of the instantaneous cross-spectral density. This modulus-squared cross-power spectrum reflects the magnitude of the shared energy component between the two signals—the coil current signal and the vibration signal—at that time-frequency point. If the two signals are completely unrelated, the average value of the cross-spectral density will approach 0.

[0048] S42, the modulus square value of the time spectrum of the current signal is smoothed to obtain the observed value of the total power spectrum of the current, and the modulus square value of the time spectrum of the vibration signal is smoothed to obtain the observed value of the total power spectrum of the vibration.

[0049] The observed total current power includes both the effective braking signal power and the background noise power, while the observed total vibration power spectrum includes both the mechanical action signal power and the environmental vibration noise power. S43, obtain the first noise power spectrum estimate of the current channel, and correct the observed total current power spectrum based on the first noise power spectrum estimate to obtain the corrected current power spectrum.

[0050] S44, obtain the second noise power spectrum estimate of the vibration channel, and correct the observed total vibration power spectrum based on the second noise power spectrum estimate to obtain the corrected vibration power spectrum.

[0051] Generally, elevator scenarios involve electromagnetic interference from the power grid, pulsating vibrations of the traction machine, and impacts on the car guide rails, resulting in uncorrelated background noise in the current and vibration signals. This means that the observed total current power spectrum and total vibration power spectrum, obtained after smoothing, contain relevant noise. Therefore, noise filtering is necessary to improve anti-interference capabilities and the accuracy of braking response performance monitoring.

[0052] The first noise power spectrum estimate and the second noise power spectrum estimate can be pre-calculated through the steady-state noise reduction segment before the braking command is issued. They are used to remove noise components from the observed spectrum, eliminate the systematic underestimation of coherence values ​​caused by environmental interference, and thus obtain the true electromechanical coupling coherence.

[0053] Specifically, the first noise power spectrum estimate can be obtained by performing power spectrum statistics on the current signal during the steady-state rest period before each braking command is issued, such as 200ms before braking. During this period, the brake has not yet been activated, and the signal only contains steady-state noise such as grid ripple and sensor noise floor. The second noise power spectrum estimate can be obtained by performing vibration power spectrum statistics on the vibration signal during the steady-state rest period before each braking command is issued, such as 200ms before braking. This is used to eliminate environmental interference such as traction machine standby vibration and guide rail conducted vibration to obtain the mechanical response power of pure braking action.

[0054] For example, the corrected current self-power spectrum = observed total current power spectrum - estimated first noise power spectrum, and the corrected vibration self-power spectrum = observed total vibration power spectrum - estimated second noise power spectrum.

[0055] S45, obtain the time-frequency domain amplitude square coherence value based on the cross-power spectrum modulus square, current self-power spectrum and vibration self-power spectrum, and obtain the braking response coherence index based on the time-frequency domain amplitude square coherence value.

[0056] The aforementioned smoothing process specifically involves performing a sliding weighted average along a preset smoothing window length in the frequency domain. More specifically, neighborhood smoothing can be performed on several frequency points, such as 3 to 5 frequency points, on the frequency axis, while no smoothing is applied to the time axis. In this way, the variance of the coherent estimation can be reduced through frequency domain smoothing, thereby reducing random statistical fluctuations at a single time frequency point and ensuring the stability of the coherent estimation. At the same time, the high resolution of the time dimension can be preserved to accurately capture the millisecond-level transient moments of the armature's movement, perfectly adapting to the short-term characteristics of the braking process.

[0057] As a preferred technical solution, obtaining the time-frequency domain amplitude squared coherence value specifically includes: normalizing the squared magnitude of the cross-power spectrum based on the product of the current self-power spectrum and the vibration self-power spectrum to obtain the time-frequency domain amplitude squared coherence value. For example, the time-frequency domain amplitude squared coherence value can be defined as: cross-power spectrum squared / (current self-power spectrum × vibration self-power spectrum). When the signal-to-noise ratio is extremely low at a certain frequency, the current self-power spectrum or vibration self-power spectrum may be negative. In this case, it can be set to a minimum constant to avoid negative numbers or division by zero errors.

[0058] In the elevator braking scenario, the square coherence value of the amplitude in the time-frequency domain characterizes the linear coupling strength of the electromechanical system, with a value range of [0,1]. The closer it is to 1, the tighter the coupling. Generally speaking, the high value region corresponds to the effective coupling period of the current change driving the mechanical action, while the low value region corresponds to the degradation state of noise dominance or electromechanical decoupling.

[0059] As a preferred technical solution, such as Figure 3 As shown, obtaining the braking response coherence index specifically includes the following steps: S451, obtain the characteristic time-frequency domain of the braking action, and then integrate the squared coherent value of the time-frequency domain amplitude within the characteristic time-frequency domain and take the average value to obtain the time-frequency domain coherent average value used to characterize the degree of electromechanical coupling.

[0060] Assuming the characteristic time-frequency domain is represented by Ω, and the time-frequency domain amplitude squared coherence value is represented by r, then the time-frequency domain coherence average value is represented as... Where t and f represent the time and frequency variables, respectively. The time variable corresponds to the complete action sequence after the braking command is issued, covering the entire process of current build-up, armature movement, brake shoe engagement, and steady-state holding. The frequency variable corresponds to the frequency components of the current and vibration signals, covering the characteristic frequency band of the brake action.

[0061] Of course, in some cases, the squared amplitude coherence value in the time-frequency domain is discrete data. In this case, the time-frequency domain coherence average value can be obtained by performing an arithmetic mean on the squared amplitude coherence values ​​corresponding to multiple time points and frequency points within the characteristic time-frequency domain.

[0062] When the brake is in good condition, changes in current directly drive mechanical action, and the coupling is tight. The coherent average value in the time-frequency domain is close to 1. When degradation occurs, such as armature jamming, friction plate wear, or coil aging, the energy transfer efficiency decreases, and the coherent average value in the time-frequency domain will decrease accordingly.

[0063] S452, obtain the braking response coherence index based on the time-frequency domain coherence average value.

[0064] As a preferred technical solution, such as Figure 4 As shown, in step S452, obtaining the braking response coherence index based on the time-frequency domain coherence average value specifically includes the following steps: S4521, obtain the measured phase standard deviation of phase values ​​at all time-frequency points within the same characteristic time-frequency domain, and the reference value of the phase standard deviation within the same characteristic time-frequency domain when the brake is in a healthy state.

[0065] The measured phase standard deviation reflects the degree of fluctuation of the electromechanical phase difference. The larger the value, the worse the synchronicity of the mechanical response and the stronger the nonlinearity. The phase standard deviation benchmark value serves as a normalization benchmark for phase fluctuations, eliminating structural differences between different models of brakes.

[0066] S4522, Obtain a phase penalty value to characterize the phase stability of the electromechanical response based on the measured phase standard deviation and the phase standard deviation reference value.

[0067] S4523, obtain the braking response coherence index based on the time-frequency domain coherence average value and the phase penalty value.

[0068] Specifically, such as Figure 5As shown, obtaining the phase penalty value specifically includes the following steps: S45221, obtain the phase fluctuation ratio based on the measured phase standard deviation and the phase standard deviation reference value.

[0069] S45222, Perform reverse calculation on the phase fluctuation ratio to obtain a phase penalty value that is negatively correlated with the phase fluctuation ratio.

[0070] Under normal brake conditions or with pure noise interference, the phase ripple ratio is approximately 1. When nonlinear degradation occurs, the phase ripple ratio will be greater than 1.

[0071] For example, the phase fluctuation ratio = measured phase standard deviation / baseline phase standard deviation, and the phase penalty value = 1 - measured phase standard deviation / baseline phase standard deviation. Generally, pure noise interference only reduces amplitude coherence and does not significantly increase phase fluctuation, while nonlinear degradation such as armature jamming and wear will simultaneously reduce amplitude coherence and increase phase fluctuation. Therefore, this phase penalty value is used to distinguish between noise interference and mechanical degradation, thereby reducing the false alarm rate.

[0072] The braking response coherence index is obtained by weighting the average time-frequency domain coherence value and the phase penalty value. That is: Braking response coherence index = α × average time-frequency domain coherence value + (1-α) × phase penalty value, where α represents the weighting coefficient, which can be fine-tuned according to the brake type, and the default value is 0.7.

[0073] The braking response coherence index is essentially a weighted fusion of the amplitude and phase stability information of two-dimensional time-frequency coherence. Its core function is to map complex time-frequency plane data into an intuitive health score for on-site performance monitoring, degradation warning, and health classification of elevator brakes.

[0074] Assuming the time-frequency domain coherence average is between 0 and 1, and a phase fluctuation ratio less than 0 indicates a degradation fault scenario, a negative penalty is applied to the time-frequency domain coherence average, and the braking response coherence index is limited to the range of -0.5 to 1. Therefore, a real-time braking response coherence index greater than 0.65 can be defined as a healthy state, indicating tight electromechanical coupling, good braking response quality, and normal operation; a real-time braking response coherence index between 0.5 and 0.65 indicates a slight degradation state, suggesting potential problems such as early wear and guide sleeve dryness, requiring close monitoring; a real-time braking response coherence index between 0.35 and 0.5 indicates a moderate degradation state, indicating a significant decline in braking response quality, requiring planned maintenance; otherwise, if the real-time braking response coherence index is ≤0.35, it indicates severe brake degradation, posing a safety risk, and immediate shutdown for inspection and troubleshooting is necessary. Of course, the range of braking response coherence indices corresponding to different health levels can be adjusted appropriately based on the brake type and actual scenario.

[0075] An embodiment of the present invention also provides an elevator braking response performance monitoring system for implementing the aforementioned elevator braking response performance monitoring method, such as... Figure 6 It includes a signal acquisition module, a signal processing module, and a performance monitoring module.

[0076] The signal acquisition module is used to acquire the coil current signal of the brake electromagnetic coil and the vibration signal of the brake body after the braking command is issued; the signal processing module is used to perform Fourier transform on the coil current signal and the vibration signal respectively to obtain the time spectrum of the current signal and the time spectrum of the vibration signal, and to obtain the complex conjugate of the time spectrum of the current signal. Based on the complex conjugate and the time spectrum of the vibration signal, the instantaneous value of the cross-spectral density of the time spectrum of the current signal and the time spectrum of the vibration signal is obtained, and the instantaneous value of the cross-spectral density is smoothed.

[0077] Specifically, a Hall current sensor can be installed in the power supply circuit of the brake's electromagnetic coil to record the coil current signal in real time after the braking command is issued. A broadband accelerometer can also be installed on the brake body or brake arm to synchronously acquire vibration signals. The sampling frequency should be no less than 10 kHz to ensure effective capture of transient characteristics.

[0078] First, the coil current signal and vibration signal are subjected to DC removal and low-pass filtering to retain effective feature components. After low-pass filtering, a short-time Fourier transform is performed on the coil current signal and vibration signal to obtain the time spectrum of the current signal and the time spectrum of the vibration signal.

[0079] The time spectrum of this current signal characterizes the energy distribution and phase state of the coil current at different times and frequencies during the braking process, reflecting the transient changes of electromagnetic excitation. It includes transient information for each stage of engagement, holding, and release. Generally, the time spectrum of the current signal corresponds to a complex number at each time frequency point, including amplitude and phase information.

[0080] The performance monitoring module is used to obtain the braking response coherence index, which characterizes the degree of electromechanical coupling, based on the instantaneous value of the cross-spectral density after smoothing, and to monitor the braking response performance of the elevator based on the braking response coherence index.

[0081] The instantaneous cross-spectral density value after smoothing can be normalized first. Then, the normalized instantaneous cross-spectral density value can be weighted and integrated within the characteristic time-frequency domain. Finally, the weighted integral value can be normalized based on the characteristic time-frequency domain to obtain the braking response coherence index.

[0082] The braking response coherence index quantitatively describes the degree of linear correlation between the coil current signal and the vibration signal at each point in the time-frequency domain. Generally speaking, 1 indicates complete coherence, and the lower the value, the less coherent the signal.

[0083] Here, the weights in the weighted integral are prior weights based on the normal state. These prior weights can be calibrated experimentally, for example, under the condition that the brake is healthy, based on the braking response coherence index, assigning heavier weights to time-frequency regions with higher coherence and more critical mechanical response. That is to say, the prior weights are positively correlated with the braking response coherence index under the condition that the elevator is operating normally and the braking response is good.

[0084] Under normal elevator operation and good braking response conditions, the characteristic time-frequency domain of the braking action can be determined. Within this characteristic time-frequency domain, the braking response coherence index often exhibits a consistently high value, such as greater than 0.8, corresponding to the electromechanical strong coupling stage during the armature engagement / disengagement process. Based on the obtained braking response coherence index and a preset coherence index threshold, such as 0.6 times the normal value, if the braking response coherence index is less than the coherence index threshold, it can be determined that the braking response performance has degraded and an early warning can be triggered.

[0085] In summary, the elevator braking response performance monitoring system acquires coil current signals and brake body vibration signals, obtains the instantaneous value of cross-spectral density based on the time-frequency domain characteristics of the two, and then obtains the braking response coherence index used to characterize the degree of electromechanical coupling. It uses the coherence loss of electrical and mechanical responses as a unified measure of braking degradation, quantifies the dynamic quality of the brake response process, and helps to improve the accuracy of brake response performance monitoring.

[0086] An embodiment of the present invention also provides an elevator braking response performance monitoring device, which includes: a controller; a memory storing executable instructions; wherein the executable instructions run on the controller and implement the elevator braking response performance monitoring method.

[0087] The technical features of the embodiments described can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0088] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for monitoring the braking response performance of an elevator, characterized in that, Includes the following steps: The coil current signal of the brake electromagnetic coil and the vibration signal of the brake body are obtained after the braking command is issued. Fourier transforms are performed on the coil current signal and the vibration signal respectively to obtain the time spectrum of the current signal and the time spectrum of the vibration signal, and the complex conjugate of the time spectrum of the current signal is obtained. The instantaneous cross-spectral density values ​​of the current signal time spectrum and the vibration signal time spectrum are obtained based on the complex conjugate and the vibration signal time spectrum, and the instantaneous cross-spectral density values ​​are smoothed. The braking response coherence index, which characterizes the degree of electromechanical coupling, is obtained based on the instantaneous value of the cross-spectral density after smoothing, and the braking response performance of the elevator is monitored based on the braking response coherence index.

2. The elevator braking response performance monitoring method as described in claim 1, characterized in that, The specific method for obtaining the braking response coherence index includes the following steps: Obtain the squared modulus of the instantaneous cross-spectral density after smoothing, and denot it as the squared modulus of the cross-power spectrum; The modulus square of the time spectrum of the current signal is smoothed to obtain the observed value of the total power spectrum of the current. At the same time, the modulus square of the time spectrum of the vibration signal is smoothed to obtain the observed value of the total power spectrum of the vibration. Obtain the first noise power spectrum estimate of the current channel, and correct the observed total current power spectrum based on the first noise power spectrum estimate to obtain the corrected current power spectrum. The second noise power spectrum estimate of the vibration channel is obtained, and the observed total vibration power spectrum is corrected based on the second noise power spectrum estimate to obtain the corrected vibration self power spectrum. The time-frequency domain amplitude squared coherence value is obtained based on the cross-power spectrum modulus square, the current self-power spectrum, and the vibration self-power spectrum. The braking response coherence index is obtained based on the time-frequency domain amplitude squared coherence value.

3. The elevator braking response performance monitoring method as described in claim 2, characterized in that, Obtaining the time-frequency domain amplitude square coherence value specifically includes: normalizing the square of the cross-power spectrum modulus based on the product of the current self-power spectrum and the vibration self-power spectrum, and obtaining the time-frequency domain amplitude square coherence value.

4. The elevator braking response performance monitoring method as described in claim 3, characterized in that, Obtaining the braking response coherence index specifically includes the following steps: The characteristic time-frequency domain of the braking action is obtained, and the average value of the squared coherent value of the time-frequency domain amplitude is integrated in the characteristic time-frequency domain to obtain the time-frequency domain coherent average value used to characterize the degree of electromechanical coupling. The braking response coherence index is obtained based on the time-frequency domain coherence average value.

5. The elevator braking response performance monitoring method as described in claim 4, characterized in that, Obtaining the braking response coherence index based on the time-frequency domain coherence average value specifically includes the following steps: Obtain the measured phase standard deviation of phase values ​​at all time-frequency points within the same characteristic time-frequency domain, and the benchmark value of the phase standard deviation within the same characteristic time-frequency domain when the brake is in a healthy state; The phase penalty value used to characterize the phase stability of the electromechanical response is obtained based on the measured phase standard deviation and the phase standard deviation benchmark value. The braking response coherence index is obtained based on the time-frequency domain coherence average value and the phase penalty value.

6. The elevator braking response performance monitoring method as described in claim 5, characterized in that, Obtaining the phase penalty value involves the following steps: The phase fluctuation ratio is obtained based on the measured phase standard deviation and the phase standard deviation benchmark value. The phase fluctuation ratio is calculated in reverse to obtain a phase penalty value that is negatively correlated with the phase fluctuation ratio.

7. The elevator braking response performance monitoring method as described in claim 6, characterized in that, The braking response coherence index is obtained by weighting the average time-frequency domain coherence value and the phase penalty value.

8. The elevator braking response performance monitoring method as described in claim 7, characterized in that, The smoothing process specifically involves performing a sliding weighted average along a preset smoothing window length in the frequency domain.

9. An elevator braking response performance monitoring system, used to implement the elevator braking response performance monitoring method as described in any one of claims 1-8, characterized in that, include: The signal acquisition module is used to acquire the coil current signal of the brake electromagnetic coil and the vibration signal of the brake body after the braking command is issued. The signal processing module is used to perform Fourier transform on the coil current signal and the vibration signal respectively, obtain the time spectrum of the current signal and the time spectrum of the vibration signal, obtain the complex conjugate of the time spectrum of the current signal, obtain the instantaneous cross-spectral density value of the time spectrum of the current signal and the time spectrum of the vibration signal based on the complex conjugate and the time spectrum of the vibration signal, and perform smoothing processing on the instantaneous cross-spectral density value. The performance monitoring module is used to obtain the braking response coherence index, which characterizes the degree of electromechanical coupling, based on the instantaneous value of the cross-spectral density after smoothing, and to monitor the braking response performance of the elevator based on the braking response coherence index.

10. An elevator braking response performance monitoring device, characterized in that, The elevator braking response performance monitoring device includes: Controller; Memory, which stores executable instructions; The executable instructions run on the controller and implement the elevator braking response performance monitoring method as described in any one of claims 1 to 8.

Images