A method and system for detecting vibration faults in mechanical structures within building engineering projects.

Abstract

This invention relates to the field of mechanical vibration monitoring technology, specifically to a method and system for detecting vibration faults in mechanical structures within building engineering. The method involves acquiring vibration signal data and rotational speed during motor operation; determining the motor's rotational speed stability based on the rotational speed at various times during operation; constructing the frequency anomaly level of the vibration signal data at different times by combining the rotational speed stability with the power magnitude at different frequencies of the vibration signal data during motor operation; performing a difference measurement correction on the vibration signal data during motor operation based on the frequency anomaly level to obtain a correction metric; and detecting anomalies in the vibration signal data based on the correction metric. This invention improves the difference measurement correction of vibration signal data during motor operation, enabling the improved correction metric to better detect abnormal trends in the motor.

Description

Technical Field

[0001] This invention relates to the field of mechanical vibration monitoring technology, specifically to a method and system for detecting vibration faults in mechanical structures within building engineering projects. Background Technology

[0002] Common vibrating equipment in construction engineering includes rotating machinery such as electric motors, fans, and water pumps, which vibrate due to rotor imbalance or bearing defects; and reciprocating equipment such as air compressors and presses, which vibrate due to periodic impact forces. Besides centrifugal force generated by uneven mass distribution of rotating parts, leading to vibrations proportional to rotational speed, or friction and impact vibrations caused by poor bearing lubrication, wear, or improper installation, uneven magnetic field in the stator windings of motors or fluctuations in power supply frequency causing changes in electromagnetic force can also induce harmonic frequency vibrations.

[0003] In related technologies, abnormal motor vibration is often detected by using the traditional LOF anomaly detection algorithm to determine whether the motor is abnormal. However, since there are many factors that can cause anomalies in the data during motor operation, when using the traditional LOF anomaly detection algorithm to detect anomalies in the vibration data during motor operation, the traditional distance metric is only quantified based on the difference in vibration amplitude within the corresponding time neighborhood of each sample. This metric method does not take into account the changes in the motor's rotational speed and the frequency of its vibration data at different times, resulting in the distance metric being unable to accurately determine the anomaly. Summary of the Invention

[0004] To address the technical problem that traditional LOF anomaly detection algorithms, when detecting anomalies in motor vibration data, rely solely on differences in vibration amplitude, making it difficult to accurately determine whether an anomaly has occurred, this invention aims to provide a method and system for detecting vibration faults in mechanical structures within building engineering projects. The specific technical solution adopted is as follows:

[0005] In a first aspect, embodiments of the present invention provide a method for detecting vibration faults in mechanical structures within building engineering projects, the method comprising:

[0006] Acquire vibration signal data and rotational speed during motor operation;

[0007] Perform a Fourier transform on the vibration signal data during motor operation to obtain the power at the specified frequency;

[0008] The stability of the motor's speed is determined based on the motor's rotational speed at various times during operation.

[0009] By combining the stability of rotational speed and the power at different frequencies, the frequency anomaly of vibration signal data at different times can be constructed.

[0010] Based on the degree of frequency anomaly, the vibration signal data during motor operation is corrected by a difference measurement to obtain a correction measurement;

[0011] Based on the aforementioned correction metric, LOF anomaly detection is performed on the vibration signal data to determine any abnormalities in the motor vibration.

[0012] Furthermore, determining the stability of the motor's rotational speed based on the rotational speed at various moments during motor operation includes:

[0013] Obtain the speed difference value at each moment within a preset time range during motor operation;

[0014] The speed fluctuation value is determined based on the fluctuation of the speed difference value within a preset time range;

[0015] By combining the speed difference value and the speed fluctuation value, the stability of the motor speed is determined.

[0016] Furthermore, determining the motor's speed stability by combining the speed difference value and the speed fluctuation value includes:

[0017] The mean value of the speed difference over the time period is calculated, and the mean value is normalized by negative correlation to obtain the speed stability adjustment value.

[0018] The speed fluctuation value is negatively correlated and normalized to obtain the speed adjustment value;

[0019] The stability of the motor's speed is determined based on the speed stability adjustment value and the speed adjustment value; wherein, both the speed stability adjustment value and the speed adjustment value are positively correlated with the stability of the motor's speed.

[0020] Furthermore, the method of constructing the frequency anomaly degree of vibration signal data at different times by combining the rotational speed stability and the power magnitude at the frequency includes:

[0021] Based on the magnitude difference in rotational speed and the degree of rotational speed stability, the confidence weight of the frequency difference is determined;

[0022] Analyze the changes in power at various frequencies of the motor's vibration signal data at different times to determine the frequency differences of the motor at different times;

[0023] The frequency difference of the motor is corrected by the confidence weight to obtain the frequency anomaly degree of vibration signal data at different times.

[0024] Furthermore, determining the confidence weight of the frequency difference based on the magnitude difference in rotational speed and the stability of rotational speed includes:

[0025] For any two moments, calculate the difference in rotational speed between the two moments as the numerator; use the stability of rotational speed as the denominator, add an adjustment coefficient to the denominator, and use the resulting ratio as the confidence weight.

[0026] Furthermore, the step of correcting the frequency difference of the motor using the confidence weight to obtain the frequency anomaly degree of the vibration signal data at different times includes:

[0027] The frequency difference of the motor is weighted by the overall value of the confidence weights of the current time and the times within the preset neighborhood range, so as to obtain the frequency abnormality of the vibration signal data at that time.

[0028] Furthermore, the step of performing difference measurement correction on the vibration signal data during motor operation based on the degree of frequency anomaly to obtain a correction metric includes:

[0029] Based on the differences in frequency anomaly and rotational speed at corresponding moments in the two time periods, the matching correction coefficients for the two time periods are determined;

[0030] Based on the power spectrum at corresponding times in the two time periods, the distance metric between the two time periods is determined;

[0031] Based on the matching correction coefficient, the distance metric is weighted to obtain the correction metric for the two time periods.

[0032] Furthermore, determining the matching correction coefficient for the two time periods based on the difference in frequency anomaly and rotational speed at corresponding moments in the two time periods includes:

[0033] Match the time of the two time periods based on the motor speed within the time period;

[0034] The first matching correction coefficient is determined based on the frequency anomaly of the midpoint at the same time in the two time periods.

[0035] The second matching correction coefficient is determined based on the rotational speed at the same alignment time in two time periods;

[0036] Based on the first matching correction coefficient and the second matching correction coefficient, determine the single correction coefficient for the same time point pair in the two time periods;

[0037] By combining the single correction coefficients for all time pairs in the two time periods, the matching correction coefficients for the two time periods are determined.

[0038] Furthermore, the step of performing LOF anomaly detection on the vibration signal data based on the correction metric to determine the abnormality of motor vibration includes:

[0039] The obtained modified metric matrix is ​​substituted into LOF anomaly detection to obtain the anomaly score for each time period; when the anomaly score for a time period is greater than the preset anomaly score threshold, it is determined that the motor vibration is abnormal.

[0040] Secondly, a vibration fault detection system for mechanical structures within building engineering is provided, the system comprising:

[0041] The acquisition module obtains vibration signal data and rotational speed during motor operation;

[0042] The frequency domain transformation module performs Fourier transform on the vibration signal data during motor operation to obtain the power at the specified frequency.

[0043] The speed analysis module determines the stability of the motor's speed based on the speed at various moments during motor operation.

[0044] The frequency analysis module, by combining the stability of the rotational speed and the power at the frequency, constructs the frequency anomaly degree of the vibration signal data at different times;

[0045] The correction module performs difference measurement correction on the vibration signal data during motor operation based on the degree of frequency anomaly, and obtains the correction measurement.

[0046] The testing module, based on the correction metric, performs LOF anomaly detection on the vibration signal data to determine abnormal conditions of the motor vibration.

[0047] Thirdly, a server is provided, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, causing the device to perform the methods of the first aspect or any possible implementation thereof.

[0048] Fourthly, a computer program product is provided, comprising: computer program code, which, when run on a computer, causes the computer to perform the methods described in the first aspect or any possible implementation thereof.

[0049] Fifthly, a computer-readable storage medium is provided that stores computer program code, which, when executed on a computer, causes the computer to perform the methods described in the first aspect or any possible implementation thereof.

[0050] The embodiments of the present invention have at least the following beneficial effects:

[0051] This invention analyzes the changes in motor speed over time at different moments to obtain a stability index of motor speed, i.e., the degree of speed stability, thus avoiding the problem of the impact of motor bearing aging on speed. By combining Fourier transform, it quantifies the changes in data frequency in the neighborhood at different moments, obtains the degree of frequency anomaly of vibration signal data at different moments, analyzes the impact of possible component problems on vibration data at each moment, and improves the traditional difference measurement correction of vibration signal data during motor operation between time periods, so that the improved correction measurement can better detect abnormal trends in the motor. Attached Figure Description

[0052] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 A schematic flowchart of a method for detecting vibration faults in mechanical structures within a building project, provided in an embodiment of the present invention;

[0054] Figure 2 This is a schematic flowchart illustrating another implementation of a method for detecting vibration faults in mechanical structures within a building project, as provided in one embodiment of the present invention.

[0055] Figure 3 This is a schematic diagram of another method flow for a method of detecting vibration faults in mechanical structures within a building project, provided by an embodiment of the present invention.

[0056] Figure 4 This is a schematic diagram of another method flow for a method of detecting vibration faults in mechanical structures within a building project, provided by an embodiment of the present invention.

[0057] Figure 5 This is a schematic diagram of another method flow for a method of detecting vibration faults in mechanical structures within a building project, provided as an embodiment of the present invention.

[0058] Figure 6 This is a schematic diagram of the composition structure of a vibration fault detection system for mechanical structures in building engineering provided in an embodiment of this application;

[0059] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0060] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method and system for detecting vibration faults in mechanical structures within building engineering, based on the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0061] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0062] The implementation environment of the embodiments of this application is described below. The implementation environment of the mechanical structure vibration fault detection system in building engineering provided by the embodiments of this application includes a data acquisition terminal, a server, and an automated early warning terminal.

[0063] The data acquisition terminal and the automated early warning terminal are connected via a wireless network. The automated early warning terminal is connected to the server via a wireless or wired network. The data acquisition terminal can be any type of terminal, such as a data acquisition sensor. The data acquisition terminal collects vibration signal data and speed data during motor operation and uploads them to the server. The server analyzes the vibration signal data and speed data during motor operation, and then performs difference measurement correction on the vibration signal data to obtain a correction metric. Based on the correction metric, the server performs anomaly detection on the vibration signal data to determine abnormal motor vibration conditions. The server then sends the abnormal motor vibration information to the automated early warning terminal, enabling the terminal to use an early warning device to automatically issue an alert for the motor abnormality, thus achieving more accurate intelligent monitoring of motor vibration.

[0064] A server is a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0065] The following description, in conjunction with the accompanying drawings, details the specific scheme of a method and system for detecting vibration faults in mechanical structures within building engineering provided by this invention.

[0066] Please see Figure 1 The diagram illustrates a flowchart of a method for detecting vibration faults in mechanical structures within a building project, according to an embodiment of the present invention. The method includes the following steps:

[0067] Step S100: Obtain vibration signal data and rotational speed during motor operation.

[0068] Vibration sensors are used to acquire vibration signal data in real time during motor operation. In this embodiment of the invention, the vibration sensor is installed on the motor to facilitate the acquisition of vibration signal data during motor operation.

[0069] After acquiring vibration signal data from multiple different motor operation processes, the obtained vibration signal data during motor operation is divided into time periods. In this embodiment of the invention, the vibration data is sampled at a frequency of 2000Hz to capture detailed information about the vibration signal. Then, the vibration data within each time period is used as training samples at 5-second intervals for subsequent anomaly detection. That is, in this embodiment of the invention, the vibration signal data during motor operation is divided into time periods at 5-second intervals.

[0070] Changes in vibration signal data within a small time neighborhood may be due to changes in the required modes under different scenarios. These changes do not necessarily indicate anomalies in motor vibration data. Therefore, a relatively long time neighborhood is selected here, with a time interval of 5 seconds.

[0071] In this embodiment of the invention, an encoder is used to measure the motor's rotational speed. The encoder is typically mounted on the motor shaft and calculates the rotational speed by detecting the rotational position. The encoder can be a photoelectric encoder, a magnetic encoder, or other types. In other embodiments of the invention, a Hall sensor can also be used to detect changes in the magnetic field on the motor rotor, thereby measuring the rotational speed. This is typically suitable for DC motors with permanent magnets. It should be noted that there are various methods for obtaining the motor's rotational speed, which will not be elaborated here. The sampling frequency of the rotational speed is consistent with the sampling frequency of the motor vibration data, and the time interval is also consistent.

[0072] Step S200: Perform Fourier transform on the vibration signal data during motor operation to obtain the power at the specified frequency.

[0073] When performing anomaly detection and analysis on vibration signal data during motor operation, traditional distance measurements are typically considered. More specifically, the differences in vibration signal data between two different time periods are considered.

[0074] However, the amplitude of vibration signal data cannot fully represent abnormal conditions during motor operation. In the event of defects in bearings or other components in the motor, the vibration data will be affected by the quality of the components, which will affect the frequency of the vibration data and the motor speed. Therefore, Fourier transform is used to analyze the changes in vibration data and motor speed at different times, so as to improve the traditional distance measurement method and enable the obtained distance measurement to better detect abnormal trends in the motor.

[0075] Therefore, further performing a short-time Fourier transform on the vibration signal data during motor operation can also be described as using the vibration signal data during motor operation as samples and performing a short-time Fourier transform on each sample.

[0076] Based on the vibration signal data obtained during motor operation in various time periods, a short-time Fourier transform is performed on the vibration signal data in each time period to obtain the spectral information corresponding to each sample. By analyzing the frequency changes at different times, the distance measurement between subsequent samples can be improved. It should be noted that the Fourier transform converts the vibration signal data from a time domain representation to a frequency domain representation, obtaining the spectral information of the vibration signal data in the frequency domain. This spectral information can show the intensity or amplitude of each frequency component in the vibration signal data and can be used to analyze periodic vibrations, frequency components, and frequency characteristics in the vibration signal data. It can be understood that the frequency domain representation obtained after performing a Fourier transform on the vibration signal data is the power spectrum, which is the frequency domain representation obtained after performing a Fourier transform on the signal, giving the power of the signal at different frequencies.

[0077] Step S300: Determine the stability of the motor speed based on the motor speed at various times during motor operation.

[0078] The operating status of a motor can be observed not only through vibration data, but also through its rotational speed, which is a crucial factor in determining whether the motor is malfunctioning. Since vibration signal data varies significantly depending on the motor's rotational speed, the stationarity of the rotational speed data corresponding to each vibration signal should be considered before analyzing the vibration signal data.

[0079] Bearings are an important component of motors, responsible for supporting the gap between the rotor and stator. If the bearings wear out, the rotor will come into contact with the stator, and the resulting friction will cause the speed to become unstable or decrease, which will further affect the vibration data of the motor. Therefore, it is necessary to further analyze the stability of the motor speed.

[0080] In some embodiments, the stability of the motor's speed can be directly and quickly determined by analyzing the speed and speed fluctuation values ​​at various moments during motor operation. That is, step S300 described above can be achieved through... Figure 2 The steps shown are to be implemented as follows:

[0081] Step S310: Obtain the speed difference value at each moment within a preset time range during motor operation.

[0082] For any moment within a preset time range centered on a certain moment during motor operation, the difference between the rotational speed at that moment and the corresponding previous moment is taken as the rotational speed difference value at that moment. The value is taken as the rotational speed difference value within the preset time range during motor operation. For example, at this moment ;in, For the first time within the preset time range during motor operation The speed difference at any given moment; For the first time within the preset time range during motor operation Rotational speed at that moment; For the first time within the preset time range during motor operation The rotational speed at a given time; this speed difference value reflects the change in rotational speed. The smaller the change in rotational speed, the better. The smaller the value, the higher the stability of the motor at that time.

[0083] In embodiments of the present invention, the preset time range also reflects the neighboring time range associated with the current moment. The time range obtained with a certain moment as the center can, to a certain extent, reflect the stability of the motor in the local time period corresponding to that moment. The size of the preset time range can be set to 40. However, it should be noted that for the first moment of the acquisition sequence, since there is no previous moment, its speed difference value can be defined as 0.

[0084] Step S320: Determine the speed fluctuation value based on the fluctuation of the speed difference value within a preset time range.

[0085] For any given moment, a preset number of speed difference values ​​are taken forward and backward from the center of that moment. This can also be understood as taking the speed difference values ​​within a preset time range from the center of that moment. The standard deviation of the speed difference values ​​within the preset time range is used to reflect the fluctuation of the speed difference values ​​within the preset time range. Then, the standard deviation of the speed difference values ​​within the preset time range is used as the corresponding speed fluctuation value.

[0086] Taking the p-th moment in the motor's operation as an example, This represents the speed fluctuation value at the p-th moment during motor operation. For the preset quantity, This reflects the preset time range. To obtain the standard deviation function, This represents the q-th speed difference value within a preset time range corresponding to the p-th moment during motor operation.

[0087] Step S330: Combine the speed difference value and the speed fluctuation value to determine the speed stability of the motor.

[0088] The speed difference value and the speed fluctuation value are fused to obtain the motor's speed stability. In one embodiment of the invention, the mean and absolute values ​​of all speed difference values ​​within a preset time range are calculated, and the results are negatively correlated and normalized to obtain a speed stability adjustment value; the speed fluctuation value is also negatively correlated and normalized to obtain a speed adjustment value; the speed stability adjustment value and the speed adjustment value are used to determine the motor's speed stability. Both the speed stability adjustment value and the speed adjustment value are positively correlated with the motor's speed stability. It should be noted that the corresponding positive correlation can be multiplication, addition, or other positively correlated relationships.

[0089] In some possible implementations, taking the p-th moment in the motor's operation as an example, the stability of the motor's rotational speed within a preset time range centered on the p-th moment is calculated, as shown in the following formula:

[0090] ;

[0091] in, The stability of the motor's rotational speed at time p during motor operation; This represents the number of speed difference values ​​within a preset time range corresponding to the p-th moment during motor operation. This indicates the time interval corresponding to the p-th moment during motor operation within a preset time range. The speed difference at each moment, It represents the absolute value of the average value of the speed difference values ​​at all times within the preset time range corresponding to the p-th time point during motor operation; This represents the ratio of the motor's rated speed to the sampling interval, used to eliminate the influence of dimensions. This represents the speed fluctuation value at the p-th moment during motor operation; and Both are adjustable weighting coefficients, and their sum is 1. Empirical values ​​can be 0.3 and 0.7. This represents a non-zero constant, used to avoid the extreme case where the standard deviation is zero when the motor is operating at an ideal constant speed. It can be taken as an empirical value of 0.001. For the preset quantity, The preset time range and preset quantity are 20; This is a function for taking the standard deviation.

[0092] In the formula for calculating the smoothness of motor speed, the rated speed is used as the benchmark. Local speed variations are decomposed into trend and random disturbances. Through dimensionless fusion, a monotonically bounded stability measure is constructed. The smoothness of motor speed is used to characterize at least the degree to which the motor speed deviates from the ideal constant speed operation. The smaller the deviation, the greater the smoothness of motor speed. Its value ranges from 0 to 1. It can represent the trend strength within a preset time range corresponding to the p-th time, i.e., trend change, such as acceleration or deceleration; This represents the jitter intensity within a preset time range corresponding to the p-th time, i.e., random fluctuations, such as mechanical shocks and load disturbances; the trend intensity and jitter intensity together constitute the disruption to stability, so the two are combined to determine the degree of speed stability; This represents the acceleration required for the system to complete the change in rated speed within one sampling period, calculated by dividing by... This eliminates the dependency on units and sampling rates, enabling cross-device comparisons; +1 is to introduce a static benchmark, ensuring that the output result is the same under ideal constant speed conditions. .

[0093] It should be noted that in all formula calculations of this invention, to prevent calculation interruption or invalid results due to a denominator of zero, a very small positive constant is added to the denominator term that may be zero. This is a conventional technique in the art to ensure algorithm stability. Regarding the normalization function (norm), unless otherwise specified, the normalization function (norm) described in the embodiments of this invention can be the Min-Max Normalization method, which linearly maps data to... Interval.

[0094] Step S400: Combining the stability of the rotational speed and the power at the frequency, construct the frequency anomaly degree of the vibration signal data at different times.

[0095] Because motors have a relatively large number of internal components, the bearings mainly affect abnormal motor speed. When other internal components of the motor become loose or damaged, the abnormality may not be well reflected in the speed, but its vibration mode will change from normal operating vibration to irregular impact vibration. On the frequency response curve, this change is manifested as the appearance of new resonance peaks or the shift of existing resonance peaks. That is, the power in the spectral data at certain times will change significantly. Based on this characteristic, the vibration frequency anomaly index at each time point in each sample is quantified.

[0096] In some embodiments, by analyzing the stability of the rotational speed and the power changes at different frequencies, the frequency anomaly of the vibration signal data at different times can be directly and quickly determined; that is, step S400 above can be achieved through... Figure 3 The steps shown are to be implemented as follows:

[0097] Step S410: Based on the magnitude difference in rotational speed and the stability of rotational speed, determine the confidence weight of the frequency difference.

[0098] For any two moments, calculate the difference in rotational speed between the two moments as the numerator; use the stability of rotational speed as the denominator. To avoid a zero denominator leading to an invalid ratio, an adjustment coefficient is added to the denominator. In this embodiment, the adjustment coefficient is set to 0.01. In other embodiments, the adjustment coefficient can be adjusted by the implementer according to the actual situation, and the value of the adjustment coefficient should be as small as possible. For example, taking the p-th moment and the k-th moment as examples, the formula for calculating the confidence weight of the frequency difference between these two moments is: ;in, The confidence weights are the frequency differences between the p-th and k-th time points. This is the normalization function; The rotational speed of the motor at time p during operation; Let $k$ be the rotational speed of the motor at time $k$ during its operation. The stability of the motor's rotational speed at time p during motor operation; This indicates the stability of the motor's speed at the k-th moment during motor operation; 0.01 is the adjustment coefficient.

[0099] It should be noted that the confidence weight in this embodiment does not represent the confidence level of the signal measurement accuracy, but rather reflects the possibility that the frequency difference between two moments is caused by changes in the actual operating conditions or abnormal events: when the rotation speed changes significantly and the system is in a non-stationary state, the corresponding frequency difference is more likely to contain key information about faults or state transitions, and therefore is given a higher weight; conversely, small frequency fluctuations in steady state are mostly caused by measurement noise and are given a lower weight.

[0100] Step S420: Analyze the changes in power of the motor's vibration signal data at various frequencies at different times to determine the frequency differences of the motor at different times.

[0101] For any two moments, calculate the average difference in power at each frequency of the motor's vibration signal data between the two moments, and use this average as the frequency difference of the motor at the two moments. For example, taking the p-th moment and the k-th moment as examples, the formula for calculating the frequency difference of the motor at these two moments is: ;in, The maximum frequency value in the short-time Fourier transform result can be directly expressed as follows: Set to 2000; Let be the power of the short-time Fourier transform result corresponding to the vibration signal data during motor operation at time p at frequency u. Let represent the power at frequency u corresponding to the short-time Fourier transform result of the vibration signal data during motor operation at time k. The frequency difference of the motor at two time points represents the magnitude of the power change at each frequency at time p and time k. The larger the value, the more abnormal frequency components appear at the two time points.

[0102] Step S430: The frequency difference of the motor is corrected by the confidence weight to obtain the frequency anomaly degree of vibration signal data at different times.

[0103] The frequency difference of the motor is weighted by the overall value of the confidence weights of the current time and the times within the preset neighborhood range, so as to obtain the frequency abnormality of the vibration signal data at that time.

[0104] In one embodiment of the present invention, for any given time, the mean of the confidence weights between that time and any time within a preset neighborhood is calculated, and this mean is used as the adjusted confidence level. In another embodiment of the present invention, for any given time, the sum of the confidence weights between that time and any time within a preset neighborhood is calculated, and the adjusted frequency anomaly level is normalized. In this embodiment of the present invention, the preset neighborhood range is 200; in other embodiments, this value can be adjusted by the implementer according to the actual situation.

[0105] The frequency difference of the motor is weighted by the adjusted confidence level to obtain the frequency anomaly degree of the vibration signal data at that moment. Then, using the same acquisition method, the frequency anomaly degree of the vibration signal data at different moments can be determined.

[0106] In some possible implementations, taking the p-th moment in the motor's operation as an example, the frequency anomaly of the vibration signal data at the p-th moment is calculated as shown in the following formula:

[0107] ;

[0108] in, The frequency anomaly of the vibration signal data at time p is denoted as . Preset neighborhood range; The confidence weights are the frequency differences between the p-th and k-th time points. This represents the maximum frequency in the short-time Fourier transform result; Let be the power of the short-time Fourier transform result corresponding to the vibration signal data during motor operation at time p at frequency u. This represents the power of the short-time Fourier transform result of the vibration signal data during motor operation at time k at frequency u. The preset neighborhood range is defined by [-ω2, ω2], which does not include the case where k=0, and the corresponding value of ω2 is 100.

[0109] In the formula for calculating the degree of frequency anomaly, This represents the magnitude of power change at various frequencies at time p and time k. The larger the value, the more abnormal frequency components appear at these two times. This represents the ratio of the difference in rotational speed between time p and time k to the overall stationarity. When the overall stationarity is high, the confidence level of the rotational speed difference at some times is low. When the frequency is abnormal at a certain moment, its corresponding weight should be reduced; conversely, its confidence level should be increased. In this case, the vibration frequency at that moment is more likely to be abnormal.

[0110] Step S500: Based on the degree of anomaly, the vibration signal data during motor operation is corrected by a difference measurement to obtain a correction measurement;

[0111] Anomalies in the frequency of motor vibration signal data are still quantified at a single moment. Anomalies at a single moment might be due to changes in the equipment's required load or other reasons, and such anomalies are insufficient to characterize the overall anomaly. When comparing two vibration signal data points, traditional distance measurement methods only quantify the Euclidean distance between the two time periods. This method does not consider factors such as the frequency of the vibration signal data at different moments and the motor's rotational speed at those moments. Calculating based solely on vibration amplitude will result in significant deviations. Therefore, after obtaining the frequency anomaly index for each data point at each moment, it is necessary to further correct the distance measurement between the two time periods by combining the rotational speed at each moment, so that the obtained distance can better reflect the true anomaly.

[0112] In some embodiments, the correction metric is obtained directly and quickly by analyzing the degree of frequency anomaly and performing difference measurement correction on the vibration signal data during motor operation. That is, the above step S500 can be achieved by... Figure 4 The steps shown are to be implemented as follows:

[0113] Step S510: Based on the difference in frequency anomaly and rotational speed at corresponding times in the two time periods, determine the matching correction coefficient for the two time periods.

[0114] In some embodiments, the matching correction coefficient can be obtained directly and quickly by analyzing the frequency anomaly degree and rotational speed at corresponding moments in two time periods. That is, step S510 above can be achieved by... Figure 5 The steps shown are to be implemented as follows:

[0115] Step S511: Match the time of the two time periods according to the motor speed within the time period.

[0116] To prevent mismatches between the times in two time periods, the times of the two time periods are first matched based on the motor speed within the time period. In one embodiment of the invention, dynamic time warping matching is performed on the times of the two time periods to obtain multiple time pairs corresponding to the two time periods. It should be noted that the two times in each time pair belong to two different time periods. When performing dynamic time warping matching on the times of the time periods, the difference in motor speed at that time is used as the dynamic time warping distance for calculation.

[0117] Step S512: Determine the first matching correction coefficient based on the frequency anomaly degree of the midpoint at the same time in the two time periods. More specifically: Calculate the sum of the frequency anomaly degrees of the midpoint at the same time in the two time periods, and use this sum as the first matching correction coefficient.

[0118] Step S513: Determine the second matching correction coefficient based on the rotational speeds at the same alignment point in the two time periods. More specifically: Calculate the difference in rotational speeds at the same alignment point in the two time periods as the second matching correction coefficient. It should be noted that, to avoid the second matching correction coefficient being 0, a preset threshold is added to the difference in rotational speeds at the same time point in the two time periods to prevent the second matching correction coefficient from being 0. In this embodiment of the invention, the preset threshold is 0.01; in other embodiments, the implementer can adjust this value according to the actual situation.

[0119] Step S514: Determine the single correction coefficient for the same time point pair in the two time periods based on the first matching correction coefficient and the second matching correction coefficient.

[0120] Step S515: Combine the individual correction coefficients of all time pairs in the two time periods to determine the matching correction coefficients for the two time periods. More specifically: directly use the sum of the individual correction coefficients of all time pairs in the two time periods as the matching correction coefficients for the two time periods.

[0121] Taking the p-th time pair within the i-th and j-th time periods as an example, the single correction coefficient for the p-th time pair within the i-th and j-th time periods is:

[0122] ;

[0123] in, is the single correction coefficient for the pair of times p in the i-th and j-th time periods; exp represents the exponential function with the natural constant as the base. This represents the attenuation rate coefficient, which can range from 5 to 10. It is used to avoid a sharp decrease in comparability due to a large difference in rotational speed. This represents the rotational speed of the time interval belonging to the i-th time interval in the p-th time pair; Let $\mathbf{p}$ be the rotation speed of the time point belonging to the $j$ time interval in the $p$-th time pair. This indicates the rated speed, used to eliminate the influence of dimensions; This represents the anomaly reinforcement weight, used to control the influence of the anomaly degree on the correction coefficient, and its value ranges from 0.1 to 0.3. The frequency anomaly of the vibration signal data belonging to the i-th time period in the p-th time pair; The frequency anomaly of the vibration signal data belonging to the j-th time period in the p-th time pair is denoted as .

[0124] In the formula for calculating a single correction factor, This represents the speed comparability decay factor, used to screen similar operating conditions, which helps to avoid misjudging healthy samples under the same operating conditions as abnormal. This indicates an abnormal enhancement item. If an abnormal vibration pattern occurs at a certain time period, the frequency of the vibration signal data is abnormal to a large degree. Even if the rotation speed is the same, the distance should be increased to highlight the abnormality. This can avoid false alarms caused by different rotation speeds. Independently enhancing the abnormality is beneficial to improving the fault detection rate.

[0125] Step S520: Based on the power spectrum at corresponding times in the two time periods, determine the distance metric between the two time periods.

[0126] In one embodiment of the present invention, for any pair of times in two time periods, the power spectrum of the corresponding two times in the time pair is calculated as a single distance metric; in another embodiment of the present invention, for any pair of times in two time periods, the difference between different power spectral density values ​​in the power spectrum of the vibration signal data of the corresponding two times in the time pair is calculated as a single distance metric.

[0127] Taking the p-th time pair within the i-th and j-th time periods as an example, the single distance metric for the p-th time pair within the i-th and j-th time periods is: ; This represents the power spectral density value at frequency u of the time interval belonging to the i-th time interval in the p-th time pair. This represents the power spectral density value at frequency u at time point j within the p-th time-p pair. A single distance metric represents the difference in power spectra between vibration data from two time-pairs at the closest rotational speeds.

[0128] Step S530: Based on the matching correction coefficient, the distance metric is weighted to obtain the correction metric for the two time periods.

[0129] After obtaining the matching correction coefficient and the distance metric weight, the product of the matching correction coefficient and the distance metric is used as the correction metric for the two time periods. This can also be understood as the distance metric being weighted through the matching correction coefficient, thus obtaining the correction metric for the two time periods.

[0130] Taking the i-th time period and the j-th time period as an example, the correction metric for the i-th time period and the j-th time period is: ;in, For the correction measures of the i-th time period and the j-th time period; is a single correction coefficient for the p-th time pair in the i-th and j-th time periods; M is the time pair in the i-th and j-th time periods.

[0131] Step S600: Based on the correction metric, perform LOF anomaly detection on the vibration signal data to determine the abnormality of motor vibration.

[0132] Based on the aforementioned distance metric, the correction metric between vibration signal data in multiple time periods is calculated, and the resulting correction metric matrix is ​​then used in LOF anomaly detection to obtain the anomaly score for each time period.

[0133] An anomaly score threshold T is preset based on the anomaly scores obtained for each time period. In this embodiment of the invention, a suggested value of 1.1 is used for T. In other embodiments, the implementer may adjust this value according to the actual situation. When the anomaly score corresponding to a time period is greater than T, an early warning should be issued for the motor corresponding to that time period to prevent motor malfunctions.

[0134] It should be noted that the embodiments of the present invention involve several empirical parameters, such as the preset quantity ω1 (example value 20) in the calculation of rotational speed fluctuation values, the neighborhood range ω2 (example value 100) in the calculation of frequency anomaly degree, and the LOF anomaly score threshold T (example value 1.1). The values ​​of these parameters are not fixed but can be determined using standard engineering methods. Specifically, those skilled in the art can use a set of labeled (normal or abnormal) validation datasets to optimize these parameters through experiments. For example, the window sizes of ω1 and ω2 can be determined by testing the impact of different values ​​on the final detection accuracy; the threshold T can be selected by analyzing the ROC curve of the anomaly scores on the validation set to achieve a balance between false positive and false negative rates. The example values ​​given in the specification are only one possible implementation and do not constitute a limitation of the present invention.

[0135] This application provides a vibration fault detection system for mechanical structures within building engineering, such as... Figure 6 As shown, system 600 includes:

[0136] The acquisition module 610 is used to acquire vibration signal data and rotational speed during motor operation;

[0137] The frequency domain transformation module 620 is used to perform Fourier transform on the vibration signal data during motor operation to obtain the power at the frequency.

[0138] The speed analysis module 630 is used to determine the stability of the motor speed based on the speed at various times during motor operation.

[0139] The frequency analysis module 640 is used to combine the rotational speed stability and the power at the frequency to construct the frequency anomaly degree of vibration signal data at different times.

[0140] The correction module 650 is used to perform difference measurement correction on the vibration signal data during motor operation based on the degree of frequency anomaly, and obtain the correction measurement.

[0141] The test module 660 is used to perform LOF anomaly detection on the vibration signal data based on the correction metric to determine the abnormality of motor vibration.

[0142] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the vibration fault detection system for mechanical structures in building engineering and the vibration fault detection method for mechanical structures in building engineering provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiments and will not be repeated here.

[0143] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. For example, as shown... Figure 7 As shown, the computer device 700 includes: a memory 710, a processor 720, and a computer program 730 stored in the memory 710 and running on the processor 720, wherein when the processor 720 executes the computer program 730, the computer device can execute any of the aforementioned methods for detecting vibration faults in mechanical structures within building engineering.

[0144] Furthermore, this application also protects an apparatus that may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to perform a method for detecting vibration faults in mechanical structures within building engineering provided in this application.

[0145] This embodiment can divide the device into functional modules based on the above method example. For example, each module can correspond to a separate function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0146] When each module is divided according to its function, the device may also include a signal uploading module, a determination module, and an adjustment module. It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here.

[0147] It should be understood that the device provided in this embodiment is used to perform the above-described method for detecting vibration faults in mechanical structures within building engineering, and therefore can achieve the same effect as the above-described implementation method.

[0148] When using integrated units, the device may include a processing module and a storage module. When applied to a workpiece, the processing module can be used to control and manage the workpiece's operations. The storage module can be used to support the execution of program code by the workpiece.

[0149] The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits as disclosed in this application. The processor may also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and microprocessors, etc., and the storage module may be a memory.

[0150] In addition, the device provided in the embodiments of this application may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute the vibration fault detection method for mechanical structures in building engineering provided in the above embodiments.

[0151] This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to implement the vibration fault detection method for mechanical structures in building engineering provided in the above embodiment.

[0152] This embodiment also provides a computer program product. When the computer program product is run on a computer, it causes the computer to perform the above-mentioned related steps to realize the vibration fault detection method for mechanical structures in building engineering provided in the above embodiment.

[0153] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0154] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0155] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0156] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for detecting vibration faults in mechanical structures within building engineering, characterized in that, The method includes the following steps: Acquire vibration signal data and rotational speed during motor operation; Perform a Fourier transform on the vibration signal data during motor operation to obtain the power at the specified frequency; The stability of the motor's speed is determined based on the motor's rotational speed at various times during operation. By combining the stability of rotational speed and the power at different frequencies, the frequency anomaly of vibration signal data at different times can be constructed. Based on the degree of frequency anomaly, the vibration signal data during motor operation is corrected by a difference measurement to obtain a correction measurement; Based on the aforementioned correction metric, LOF anomaly detection is performed on the vibration signal data to determine any abnormalities in the motor vibration. The method for obtaining the frequency anomaly level is as follows: Match the two time periods based on the motor speed within the time period; determine a first matching correction coefficient based on the frequency anomaly level at the midpoint of the same time point in the two time periods; determine a second matching correction coefficient based on the motor speed at the midpoint of the same time point in the two time periods; determine a single correction coefficient for the same time point pair in the two time periods based on the first and second matching correction coefficients; combine the single correction coefficients for all time point pairs in the two time periods to determine the matching correction coefficient for the two time periods; determine the distance metric for the two time periods based on the power spectrum at corresponding times in the two time periods; and weight the distance metric based on the matching correction coefficient to obtain the correction metric for the two time periods. The method for obtaining the first matching correction coefficient is as follows: calculate the sum of the frequency anomalies of the midpoint time at the same time in two time periods, and use it as the first matching correction coefficient; The second matching correction coefficient is obtained by calculating the difference in rotational speed at the same alignment time in two time periods, which is used as the second matching correction coefficient. The matching correction coefficient is obtained by directly using the sum of the individual correction coefficients of all time pairs in the two time periods as the matching correction coefficient for the two time periods.

2. The method for detecting vibration faults in mechanical structures within building engineering according to claim 1, characterized in that, The determination of the motor's speed stability based on the motor's speed at various moments during operation includes: Obtain the speed difference value at each moment within a preset time range during motor operation; The speed fluctuation value is determined based on the fluctuation of the speed difference value within a preset time range; By combining the speed difference value and the speed fluctuation value, the stability of the motor speed is determined.

3. The method for detecting vibration faults in mechanical structures within building engineering according to claim 2, characterized in that, The determination of the motor's speed stability by combining the speed difference value and the speed fluctuation value includes: The mean value of the speed difference over the time period is calculated, and the mean value is normalized by negative correlation to obtain the speed stability adjustment value. The speed fluctuation value is negatively correlated and normalized to obtain the speed adjustment value; The stability of the motor's speed is determined based on the speed stability adjustment value and the speed adjustment value; wherein, both the speed stability adjustment value and the speed adjustment value are positively correlated with the stability of the motor's speed.

4. The method for detecting vibration faults in mechanical structures within building engineering according to claim 1, characterized in that, The method of combining the stability of rotational speed and the power at different frequencies to construct the frequency anomaly degree of vibration signal data at different times includes: Based on the magnitude difference in rotational speed and the degree of rotational speed stability, the confidence weight of the frequency difference is determined; Analyze the changes in power at various frequencies of the motor's vibration signal data at different times to determine the frequency differences of the motor at different times; The frequency difference of the motor is corrected by the confidence weight to obtain the frequency anomaly degree of vibration signal data at different times.

5. The method for detecting vibration faults in mechanical structures within a building project according to claim 4, characterized in that, The determination of the confidence weight of the frequency difference based on the magnitude difference and the stability of the rotational speed includes: For any two moments, calculate the difference in rotational speed between the two moments as the numerator; use the stability of rotational speed as the denominator, add an adjustment coefficient to the denominator, and use the resulting ratio as the confidence weight.

6. The method for detecting vibration faults in mechanical structures within building engineering according to claim 4, characterized in that, The step of correcting the frequency difference of the motor using the confidence weight to obtain the frequency anomaly degree of vibration signal data at different times includes: The frequency difference of the motor is weighted by the overall value of the confidence weights of the current time and the times within the preset neighborhood range, so as to obtain the frequency abnormality of the vibration signal data at that time.

7. The method for detecting vibration faults in mechanical structures within building engineering according to claim 1, characterized in that, The step of performing LOF anomaly detection on the vibration signal data based on the correction metric to determine the abnormality of motor vibration includes: The obtained modified metric matrix is ​​substituted into LOF anomaly detection to obtain the anomaly score for each time period; when the anomaly score for a time period is greater than the preset anomaly score threshold, it is determined that the motor vibration is abnormal.

8. A vibration fault detection system for mechanical structures in building engineering, characterized in that, The system includes: The acquisition module obtains vibration signal data and rotational speed during motor operation; The frequency domain transformation module performs Fourier transform on the vibration signal data during motor operation to obtain the power at the specified frequency. The speed analysis module determines the stability of the motor's speed based on the speed at various moments during motor operation. The frequency analysis module, by combining the stability of the rotational speed and the power at the frequency, constructs the frequency anomaly degree of the vibration signal data at different times; The method for obtaining the frequency anomaly level is as follows: Match the two time periods based on the motor speed within the time period; determine a first matching correction coefficient based on the frequency anomaly level at the midpoint of the same time point in the two time periods; determine a second matching correction coefficient based on the motor speed at the midpoint of the same time point in the two time periods; determine a single correction coefficient for the same time point pair in the two time periods based on the first and second matching correction coefficients; combine the single correction coefficients for all time point pairs in the two time periods to determine the matching correction coefficient for the two time periods; determine the distance metric for the two time periods based on the power spectrum at corresponding times in the two time periods; and weight the distance metric based on the matching correction coefficient to obtain the correction metric for the two time periods. The method for obtaining the first matching correction coefficient is as follows: calculate the sum of the frequency anomalies of the midpoint time at the same time in two time periods, and use it as the first matching correction coefficient; The second matching correction coefficient is obtained by calculating the difference in rotational speed at the same alignment time in two time periods, which is used as the second matching correction coefficient. The matching correction coefficient is obtained by directly using the sum of the single correction coefficients of all time pairs in the two time periods as the matching correction coefficient for the two time periods. The correction module performs difference measurement correction on the vibration signal data during motor operation based on the degree of frequency anomaly, and obtains the correction measurement. The testing module, based on the correction metric, performs LOF anomaly detection on the vibration signal data to determine abnormal conditions of the motor vibration.

Images