Equipment fault prediction method and system for digital workshop

Abstract

The invention relates to the technical field of intelligent manufacturing, particularly discloses an equipment fault prediction method and system for a digital workshop, and particularly aims at analyzing vibration signals of equipment so as to effectively monitor the operation state of the equipment. Compared with the prior art, the fault diagnosis method has the advantages that the accuracy and the real-time performance of fault diagnosis are remarkably improved, the adaptive capacity of the system is enhanced, and the fault diagnosis method is suitable for various industrial application scenes.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a method and system for predicting equipment failure in a digital workshop. Background Art

[0002] With the development of Industry 4.0 and smart manufacturing, the importance of digital workshops as a key component for realizing the intelligence of production processes has become increasingly prominent. In digital workshops, the stable operation of production equipment is crucial to ensuring product quality and improving production efficiency. However, since production equipment is in a high-intensity working state for a long time, equipment failures occur from time to time, which not only leads to production interruptions, but also may cause huge economic losses. Therefore, effective fault prediction of production equipment has become an urgent problem to be solved.

[0003] At present, there are many equipment status monitoring technologies on the market, mainly including simple alarm systems based on thresholds, fault diagnosis based on machine learning, and fault diagnosis based on expert systems. These technologies have their own advantages, but also have certain limitations, such as lack of adaptability to different equipment types and application scenarios, poor real-time performance, and high implementation difficulty. In addition, the existing systems still need to be improved in terms of feature extraction accuracy and fault type identification accuracy.

[0004] In order to address the shortcomings of the prior art, the present invention proposes an improved equipment status monitoring method based on vibration signals. The method extracts the key time domain features of the vibration signals and calculates the vibration score of the equipment in combination with the weight of the features, so as to judge the status of the equipment and identify the specific fault type. This method not only improves the accuracy and real-time performance of fault diagnosis, but also enhances the scalability and flexibility of the system, so that it can better adapt to various industrial application scenarios. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for predicting equipment failures in a digital workshop in order to solve the problems of lack of adaptability to different equipment types and application scenarios, poor real-time performance and high implementation difficulty.

[0006] The present invention is implemented through the following technical solutions: a method and system for predicting equipment failure in a digital workshop, obtaining vibration signal data of the equipment, extracting vibration signal features, and obtaining a vibration score of the equipment;

[0007] The vibration signal characteristics include the vibration peak value, mean value, standard deviation and kurtosis of the vibration signal;

[0008] Generate equipment abnormality signal and equipment normal signal based on the vibration score of the equipment;

[0009] Based on the abnormal signal of the equipment, the vibration signal data is analyzed for abnormality. If the vibration signal data is abnormal, a corresponding early warning signal is generated;

[0010] Based on the abnormal signals of the equipment, all vibration signals within the monitoring period are integrated to construct a spectrum diagram;

[0011] Based on the spectrum diagram, identify the fundamental frequency of the equipment and its harmonic frequencies, and obtain abnormal peaks;

[0012] Based on the early warning signals, each early warning signal is analyzed separately to obtain the specific abnormal cause of the equipment and send an alarm notification.

[0013] As a further solution of the present invention: the specific process of extracting the vibration signal characteristics is:

[0014] Divide the monitoring period into several time sub-units of equal length, divide the time sub-unit into several time small units of equal length, and collect the vibration signal value of the equipment when the time small unit is in the middle moment;

[0015] All vibration signal values ​​in the time subunit are integrated into a vibration sequence, and the vibration signal value with the largest absolute value in the vibration sequence is extracted and recorded as the vibration peak value Fz;

[0016] Based on the vibration peak value Fz, the vibration peak value ratio Fzp of the vibration signal in the time subunit is obtained;

[0017] Based on the vibration sequence, the average data Jz of the vibration signal in the time subunit is obtained by mean calculation;

[0018] Based on the average data Jz, the mean ratio Jzp and standard deviation Bz of the vibration signal in the time subunit are obtained;

[0019] Based on the standard deviation Bz, the standard deviation ratio Bzp of the vibration signal in the time subunit and the kurtosis Qd of the vibration signal in the time subunit are obtained;

[0020] The kurtosis ratio Qdp of the vibration signal in the time subunit is obtained based on the kurtosis Qd.

[0021] As a further solution of the present invention: the specific process of obtaining the vibration score of the equipment is:

[0022] Based on the vibration peak ratio Fzp, mean ratio Jzp, standard deviation ratio Bzp and kurtosis ratio Qdp of the vibration signal in the time subunit, a weight w1 is assigned to the vibration peak ratio Fzp, a weight w2 is assigned to the mean ratio Jzp, a weight w3 is assigned to the standard deviation ratio Bzp, and a weight w4 is assigned to the kurtosis ratio Qdp;

[0023] The vibration score ZD of the equipment is calculated by the formula ZD=w1×Fzp+w2×Jzp+w3×Bzp+w4×Qdp.

[0024] As a further solution of the present invention: the process of generating the device abnormal signal and the device normal signal is:

[0025] Based on the vibration score ZD of the device, comparing the vibration score ZD of the device with a vibration score threshold of the device;

[0026] If the vibration score ZD of the device is greater than or equal to the vibration score threshold of the device, it means that there is an abnormality in the device, and a device abnormality signal is generated;

[0027] If the vibration score ZD of the device is less than the vibration score threshold of the device, it means that there is no abnormal problem with the device, and a normal signal of the device is generated.

[0028] As a further solution of the present invention: the specific process of abnormal analysis of vibration signal data is as follows:

[0029] Based on the vibration peak value Fz, average data Jz, standard deviation Bz and kurtosis Qd of the vibration signal in the time subunit, the statistical value ZF of the vibration peak value, the statistical value ZJ of the average data, the statistical value ZB of the standard deviation and the statistical value ZQ of the kurtosis are obtained.

[0030] As a further solution of the present invention: the specific process of generating the corresponding warning signal:

[0031] Based on the statistic ZF of the vibration peak, the statistic ZJ of the average data, the statistic ZB of the standard deviation, and the statistic ZQ of the kurtosis, the absolute value of the statistic ZF of the vibration peak is compared with the statistic threshold of the vibration peak;

[0032] If the absolute value of the statistical value ZF of the vibration peak value is greater than or equal to the statistical value threshold of the vibration peak value, an early warning signal of equipment bearing damage and gear meshing is generated;

[0033] Compare the statistical value ZJ of the average data with the statistical value threshold of the average data;

[0034] If the statistical value ZJ of the average data is greater than or equal to the statistical value threshold of the average data, an equipment imbalance and misalignment warning signal is generated;

[0035] Compare the statistical value ZB of the standard deviation of the vibration signal in the time subunit with the statistical value threshold of the standard deviation;

[0036] If the statistical value ZB of the standard deviation is greater than or equal to the statistical value threshold of the standard deviation, an internal wear warning signal of the equipment is generated;

[0037] Compare the kurtosis statistic ZQ of the vibration signal in the time subunit with the kurtosis statistic threshold;

[0038] If the kurtosis statistical value ZQ is greater than or equal to the kurtosis statistical value threshold, it means that the kurtosis is abnormal, and a local equipment failure warning signal is generated.

[0039] As a further solution of the present invention: the process of constructing the spectrum graph is:

[0040] Integrate all vibration signal data within the monitoring period, use the fast Fourier transform algorithm to convert the vibration signal data into the frequency domain, and obtain the spectrum data of the vibration signal within the monitoring period;

[0041] The converted spectrum data is displayed as a graph, with the horizontal axis representing frequency and the vertical axis representing amplitude.

[0042] As a further solution of the present invention: the process of identifying the fundamental frequency and harmonic frequency of the equipment and obtaining the abnormal peak value is as follows:

[0043] Get the motor speed Zs of the device through the formula Get the fundamental frequency f0 of the device, and traverse the spectrum graph to get the frequency peak closest to the fundamental frequency f0 in the spectrum graph, which is recorded as the fundamental frequency peak;

[0044] Identify the frequency peak that is an integer multiple of the fundamental frequency f0 in the spectrum graph and record it as the harmonic frequency nf0 of the device;

[0045] Wherein, the harmonic frequency is an integer multiple of the fundamental frequency, n is the multiple, and n is a positive integer greater than 1, n=2, 3, ..., n, when the fundamental frequency is f0, then 2f0, 3f0, ..., nf0 are the harmonic frequencies;

[0046] Obtain peaks in the spectrum that are inconsistent with the fundamental frequency and fundamental frequency harmonic frequencies, calculate the difference between these peaks and the fundamental frequency peak, and if the difference is greater than the peak difference threshold, mark the peak as an abnormal peak.

[0047] As a further solution of the present invention: the process of analyzing each warning signal is as follows:

[0048] Based on the equipment imbalance and misalignment warning signals, the fundamental frequency peak value is compared with the fundamental frequency peak value threshold value. If the fundamental frequency peak value is greater than or equal to the fundamental frequency peak value threshold value, an equipment imbalance alarm signal is generated;

[0049] The peak values ​​at the base frequency and twice the base frequency are compared with the base frequency peak value threshold at the same time. If the peak values ​​at the base frequency and twice the base frequency are both greater than or equal to the base frequency peak value threshold, an equipment misalignment alarm signal is generated.

[0050] Based on the equipment bearing damage and gear meshing warning signals, the number and diameter of the rolling elements in the equipment bearing, the diameter of the raceway in the bearing, the number of teeth and the speed of the driving gear in the equipment gearbox are obtained to obtain the bearing characteristic frequency and the gear meshing frequency;

[0051] Based on the bearing characteristic frequency and the gear meshing frequency, if an abnormal peak corresponding to the bearing characteristic frequency or the gear meshing frequency appears in the spectrum graph, a bearing damage alarm signal or a gear damage alarm signal is generated;

[0052] Based on the internal wear warning signal of the equipment, the gear rotation frequency is obtained through the gear speed. Based on the gear rotation frequency, if an abnormal peak corresponding to the gear rotation frequency appears in the spectrum graph, a gear wear alarm signal is generated;

[0053] Based on the number and diameter of the rolling elements in the equipment bearing and the diameter of the raceway in the bearing, the cage rotation frequency is obtained;

[0054] Based on the cage rotation frequency, if an abnormal peak corresponding to the cage rotation frequency appears in the spectrum graph, a bearing wear alarm signal is generated;

[0055] Based on the equipment local fault warning signal, the equipment power frequency is obtained. If an abnormal peak corresponding to the equipment power frequency appears in the spectrum graph, a motor fault alarm signal is generated;

[0056] Obtain the operating frequency of the pump or compressor of the equipment, and if an abnormal peak corresponding to the operating frequency of the pump or compressor appears in the spectrum graph, generate a pump or compressor fault signal;

[0057] If the device has a valve for controlling the flow of fluid, the operating frequency of the valve is obtained. If an abnormal peak corresponding to the operating frequency of the valve appears in the spectrum graph, a valve fault signal is generated.

[0058] As a further solution of the present invention: an equipment failure prediction system for a digital workshop, characterized by comprising a data acquisition and processing module, an equipment abnormality judgment module, an equipment abnormality analysis module and an equipment analysis and alarm module;

[0059] The data acquisition and processing module is used to obtain the vibration signal data of the equipment, extract the vibration signal characteristics, and obtain the vibration score of the equipment;

[0060] The vibration signal characteristics include the vibration peak value, mean value, standard deviation and kurtosis of the vibration signal;

[0061] The equipment abnormality judgment module generates equipment abnormality signals and equipment normal signals based on the vibration scores of the equipment;

[0062] The equipment abnormality analysis module performs abnormal analysis on the vibration signal data based on the equipment abnormal signal. If the vibration signal data is abnormal, a corresponding warning signal is generated;

[0063] The equipment analysis and alarm module integrates all vibration signals within the monitoring period and constructs a spectrum diagram based on equipment abnormality signals;

[0064] Based on the spectrum diagram, identify the fundamental frequency of the equipment and its harmonic frequencies, and obtain abnormal peaks;

[0065] Based on the early warning signals, each early warning signal is analyzed separately to obtain the specific abnormal cause of the equipment and send an alarm notification.

[0066] Beneficial effects of the present invention:

[0067] (1) The present invention can more accurately capture the subtle changes in the operation of the equipment through the optimized time domain feature extraction algorithm, especially greatly improve the sensitivity to early fault signals. Compared with the traditional method based on frequency domain analysis, the present invention can effectively reduce the misjudgment phenomenon caused by signal frequency overlap. In addition, by introducing the feature weight calculation mechanism, the recognition accuracy of key features is further improved, so that the system can accurately judge the equipment status in a complex working environment, thereby effectively preventing potential equipment failures and reducing the risk of unplanned downtime;

[0068] (2) By using data processing technology, the present invention realizes the real-time collection and analysis of equipment vibration signals. This not only means that the equipment status information can be fed back instantly, but also the monitoring strategy can be quickly adjusted according to the changes in actual working conditions. For example, when an abnormal signal is detected, the system can automatically trigger the early warning mechanism and take timely response measures. At the same time, due to the use of a modular design concept, the monitoring parameters can be flexibly configured according to different equipment types and usage scenarios, which greatly enhances the scope of application and practicality of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The present invention will be further described below in conjunction with the accompanying drawings.

[0070] Figure 1 It is a flow chart of a method for predicting equipment failure in a digital workshop of the present invention;

[0071] Figure 2 It is a flow chart of generating early warning signals in a method for predicting equipment failure in a digital workshop in the present invention;

[0072] Figure 3 It is a flow chart of an equipment failure prediction system for a digital workshop in the present invention; DETAILED DESCRIPTION

[0073] The present invention is further described below through specific embodiments in conjunction with the accompanying drawings, but the present invention is not limited thereto. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0074] Example 1

[0075] See also Figure 1As shown, the present invention is a method for predicting equipment failure in a digital workshop, comprising:

[0076] Obtain the vibration signal value of the device and extract the time domain characteristics of the vibration signal, including the vibration peak value, average data, standard deviation and kurtosis of the vibration signal;

[0077] Based on the vibration peak value, average data, standard deviation and kurtosis of the vibration signal, the vibration peak value ratio Fzp, mean value ratio Jzp, standard deviation ratio Bzp and kurtosis ratio Qdp are obtained;

[0078] Assign weight w1 to the vibration peak ratio, w2 to the mean ratio, w3 to the standard deviation ratio, and w4 to the kurtosis ratio;

[0079] The vibration score ZD of the equipment is calculated by the formula ZD = w1×Fzp+w2×Jzp+w3×Bzp+w4×Qdp;

[0080] Generate equipment abnormality signal and equipment normal signal based on the vibration score ZD of the equipment;

[0081] Based on the abnormal signal of the equipment, the vibration peak is analyzed abnormally. If the vibration peak is abnormal, the equipment bearing damage and gear meshing warning signal is generated;

[0082] Perform abnormal analysis on average data. If the average data is abnormal, generate equipment imbalance and misalignment warning signals;

[0083] Perform abnormal analysis on the standard deviation. If the standard deviation is abnormal, generate an early warning signal for internal wear of the equipment.

[0084] Perform abnormal analysis on the standard deviation. If the standard deviation is abnormal, generate an early warning signal for internal wear of the equipment.

[0085] Based on the abnormal signals of the equipment, all vibration signals within the monitoring period are integrated to construct a spectrum diagram;

[0086] Based on the spectrum diagram, identify the fundamental frequency of the equipment and its harmonic frequencies, and obtain abnormal peaks;

[0087] Based on the equipment imbalance and misalignment warning signals, bearing damage and gear meshing warning signals, equipment internal wear warning signals and equipment local failure warning signals, each warning signal is analyzed separately to obtain the specific abnormal cause of the equipment and send an alarm notification.

[0088] Example 2

[0089] Divide the monitoring period into several time sub-units of equal length, divide the time sub-unit into several time small units of equal length, and collect the vibration signal value of the equipment when the time small unit is in the middle moment;

[0090] All vibration signal values ​​are processed to extract the time domain characteristics of the vibration signal. Specifically:

[0091] All vibration signal values ​​in the time subunit are integrated into a vibration sequence, and the vibration signal value with the largest absolute value in the vibration sequence is extracted and recorded as the vibration peak value Fz;

[0092] Integrate all vibration peaks within the monitoring period and extract the maximum vibration peak Fz max and minimum vibration peak Fz min ;

[0093] By formula The vibration peak value ratio Fzp of the vibration signal in the time subunit is calculated;

[0094] Based on the vibration sequence, the average data Jz of the vibration signal in the time subunit is obtained by mean calculation;

[0095] Integrate all the average data within the monitoring period and extract the maximum average data Jz max and the minimum average data Jz min ;

[0096] By formula The mean value ratio Jzp of the vibration signal in the time subunit is calculated;

[0097] Based on the average data Jz of the vibration signal, the formula Calculate the standard deviation Bz of the vibration signal within the time subunit;

[0098] Integrate all standard deviations within the monitoring period and extract the maximum standard deviation Bz max and the minimum standard deviation Bz min ;

[0099] By formula The standard deviation ratio Bzp of the vibration signal within the time subunit is calculated;

[0100] Based on the standard deviation Bz of the vibration signal within the time subunit, the formula The kurtosis Qd of the vibration signal in the time subunit is calculated;

[0101] Integrate all the kurtosis within the monitoring period and extract the maximum kurtosis Qd max and minimum kurtosis Qd min ;

[0102] By formula The kurtosis ratio Qdp of the vibration signal in the time subunit is calculated;

[0103] Based on the vibration peak ratio Fzp, mean ratio Jzp, standard deviation ratio Bzp and kurtosis ratio Qdp of the vibration signal in the time subunit, a weight w1 is assigned to the vibration peak ratio Fzp, a weight w2 is assigned to the mean ratio Jzp, a weight w3 is assigned to the standard deviation ratio Bzp, and a weight w4 is assigned to the kurtosis ratio Qdp;

[0104] The vibration score ZD of the equipment is calculated by the formula ZD = w1×Fzp+w2×Jzp+w3×Bzp+w4×Qdp;

[0105] Based on the vibration score ZD of the device, comparing the vibration score ZD of the device with a vibration score threshold of the device;

[0106] If the vibration score ZD of the device is greater than or equal to the vibration score threshold of the device, it means that there is an abnormality in the device, and a device abnormality signal is generated;

[0107] If the vibration score ZD of the device is less than the vibration score threshold of the device, it means that there is no abnormality in the device, and a normal signal of the device is generated.

[0108] Example 3

[0109] refer to Figure 2 As shown, based on the abnormal signal of the equipment generated above, since the time domain characteristics of the vibration signal are a series of statistical indicators extracted from the original vibration signal of the equipment, the operating status of the equipment can be understood through these indicators, and the abnormal factors of the equipment can be determined by analyzing whether these indicators are abnormal or not;

[0110] Based on the vibration peak value Fz of the vibration signal in the time subunit, all vibration peak values ​​in the monitoring period are integrated, and the mean value Fu of the vibration peak value is obtained through the mean calculation formula;

[0111] Based on the mean value uF of the vibration peak, the standard deviation σ of the vibration peak is obtained through the standard deviation calculation formula F ;

[0112] Based on the mean value Fu of the vibration peak and the standard deviation σ of the vibration peak F , through the formula Calculate and obtain the statistical value ZF of the vibration peak value;

[0113] Compare the absolute value of the statistical value ZF of the vibration peak value of the vibration signal in the time subunit with the statistical value threshold of the vibration peak value;

[0114] If the absolute value of the statistical value ZF is greater than or equal to the statistical value threshold of the vibration peak value, it means that the vibration peak value is abnormal, and an early warning signal of equipment bearing damage and gear meshing is generated;

[0115] It should be noted that the vibration peak of the equipment can reflect the presence of transient events or impacts in the vibration signal. Peak data with abnormally high absolute values ​​indicates that there is a fault or impact event in the equipment, such as bearing damage or gear meshing abnormality.

[0116] Based on the average data Jz of the vibration signal in the time subunit, all the average data in the monitoring period are integrated, and the mean value Ju of the average data is obtained through the mean calculation formula;

[0117] Based on the mean value Ju of the average data, the standard deviation σ of the average data is obtained through the standard deviation calculation formula J ;

[0118] The mean Ju and standard deviation σ based on the average data J , through the formula Calculate the statistical value ZJ of the average data;

[0119] Compare the statistical value ZJ of the average data of the vibration signal in the time subunit with the statistical value threshold of the average data;

[0120] If the statistical value ZJ of the average data is greater than or equal to the statistical value threshold of the average data, it means that the average data is abnormal, and an early warning signal of equipment imbalance and misalignment is generated;

[0121] It should be noted that the average data of the vibration signal is usually used to determine whether the vibration signal has an obvious DC component. Ideally, the average data of the vibration signal should be zero, indicating that the amplitude of the vibration signal in the positive and negative directions is roughly equal. If the average data deviates significantly from zero, it indicates that the equipment has misalignment or imbalance problems.

[0122] Based on the standard deviation Bz of the vibration signal in the time subunit, all standard deviations in the monitoring period are integrated, and the mean value Bu of the standard deviation is obtained through the mean calculation formula;

[0123] Based on the mean value Bu of the standard deviation, the standard deviation σ of the standard deviation is obtained through the standard deviation calculation formula B ;

[0124] Mean Bu and standard deviation σ based on standard deviation B , through the formula Calculate the statistical value ZB of the standard deviation;

[0125] Compare the statistical value ZB of the standard deviation of the vibration signal in the time subunit with the statistical value threshold of the standard deviation;

[0126] If the statistical value ZB of the standard deviation is greater than or equal to the statistical value threshold of the standard deviation, it means that the standard deviation is abnormal, and an early warning signal of internal wear of the equipment is generated;

[0127] It should be noted that the larger the standard deviation, the greater the change in the amplitude of the vibration, indicating that the equipment has a large vibration or instability. When the equipment is worn inside, the standard deviation will increase.

[0128] Based on the kurtosis Qd of the vibration signal in the time subunit, all the kurtosis in the monitoring period are integrated, and the mean value Qu of the kurtosis is obtained through the mean calculation formula;

[0129] Based on the mean value Qu of the kurtosis, the standard deviation σ of the kurtosis is obtained through the standard deviation calculation formula Q ;

[0130] The mean Qu and standard deviation σ based on kurtosis Q , through the formula Calculate the statistical value ZQ of kurtosis;

[0131] Compare the kurtosis statistic ZQ of the vibration signal in the time subunit with the kurtosis statistic threshold;

[0132] If the kurtosis statistical value ZQ is greater than or equal to the kurtosis statistical value threshold, it means that the kurtosis is abnormal, and a local equipment failure warning signal is generated;

[0133] It should be noted that the higher the kurtosis of the vibration signal, the higher the frequency of occurrence of sharp peaks in the signal. The increase in kurtosis is related to local faults. When a local fault occurs in the equipment, sharp peaks will be generated in the vibration signal. These peaks appear as an increase in kurtosis in the time domain.

[0134] Example 4

[0135] Based on the abnormal signal of the equipment, the vibration signal of the equipment is obtained when the time unit in the time subunit is in the middle moment, all the vibration signals in the monitoring period are integrated, and the fast Fourier transform algorithm is used to convert the vibration signal into the frequency domain to obtain the spectrum data of the vibration signal in the monitoring period;

[0136] The converted spectrum data is displayed as a graph, with the horizontal axis representing frequency and the vertical axis representing amplitude;

[0137] It should be noted that the fast Fourier transform algorithm can convert the vibration signal into a frequency domain signal, thereby displaying the amplitude of different frequency components. The spectrum diagram constructed by the spectrum can display the intensity of each frequency component in the device;

[0138] Based on the spectrum diagram, analyze and identify the fundamental frequency and harmonic frequencies of the device, specifically:

[0139] Get the motor speed Zs of the device through the formula Get the fundamental frequency f0 of the device, and traverse the spectrum graph to get the frequency peak closest to the fundamental frequency f0 in the spectrum graph, which is recorded as the fundamental frequency peak;

[0140] Identify the frequency peak that is an integer multiple of the fundamental frequency f0 in the spectrum graph and record it as the harmonic frequency nf0 of the device;

[0141] Wherein, the harmonic frequency is an integer multiple of the fundamental frequency, n is the multiple, and n is a positive integer greater than 1, n=2, 3, ..., n, when the fundamental frequency is f0, then 2f0, 3f0, ..., nf0 are the harmonic frequencies;

[0142] Based on the equipment imbalance and misalignment warning signals, the fundamental frequency peak value is compared with the fundamental frequency peak value threshold. If the fundamental frequency peak value is greater than or equal to the fundamental frequency peak value threshold, it means that the peak value at the fundamental frequency is an abnormal peak value, and an equipment imbalance alarm signal is generated;

[0143] The peak values ​​at the base frequency and twice the base frequency are compared with the base frequency peak value threshold at the same time. If the peak values ​​at the base frequency and twice the base frequency are both greater than or equal to the base frequency peak value threshold, it means that the peak values ​​at the base frequency and twice the base frequency are both abnormal peak values, and an equipment misalignment alarm signal is generated.

[0144] Obtain peaks in the spectrum that are inconsistent with the fundamental frequency and the fundamental frequency harmonic frequency, calculate the difference between these peaks and the fundamental frequency peak, and if the difference is greater than the peak difference threshold, mark the peak as an abnormal peak;

[0145] Based on the equipment bearing damage and gear meshing warning signals, the number N and diameter d of the rolling elements in the equipment bearing and the diameter D of the raceway in the bearing are obtained;

[0146] By formula The bearing characteristic frequency Gd is calculated;

[0147] Based on the bearing characteristic frequency Gd, if an abnormal peak corresponding to the bearing characteristic frequency Gd appears in the spectrum diagram, it means that the equipment is bearing damaged, and a bearing damage alarm signal is generated;

[0148] Obtain the number of teeth Nc and the speed Zsc of the driving gear in the equipment gearbox;

[0149] By formula The gear meshing frequency Cl is calculated;

[0150] Based on the gear meshing frequency Cl, if an abnormal peak corresponding to the gear meshing frequency Cl appears in the spectrum diagram, it means that the equipment is gear damaged, and a gear damage alarm signal is generated;

[0151] Based on the internal wear warning signal of the equipment, the formula Get the gear rotation frequency fz;

[0152] Based on the gear rotation frequency fz, if an abnormal peak corresponding to the gear rotation frequency fz appears in the spectrum diagram, it means that the equipment is gear worn, and a gear wear alarm signal is generated;

[0153] Based on the number N and diameter d of the rolling elements in the equipment bearing and the diameter D of the bearing raceway, the formula The cage rotation frequency Bc is calculated;

[0154] Based on the cage rotation frequency Bc, if an abnormal peak corresponding to the cage rotation frequency Bc appears in the spectrum diagram, it means that the equipment is abraded by the bearing and the parts connected to the bearing, and a bearing wear alarm signal is generated;

[0155] Based on the equipment local fault warning signal, the equipment power frequency is obtained. If an abnormal peak corresponding to the equipment power frequency appears in the spectrum diagram, it means that the equipment motor has a fault, and a motor fault alarm signal is generated;

[0156] Based on the fact that the device is a hydraulic or pneumatic system, the operating frequency of the pump or compressor is obtained. If an abnormal peak corresponding to the operating frequency of the pump or compressor appears in the spectrum diagram, it means that the pump or compressor of the device is faulty, and a pump or compressor fault signal is generated;

[0157] Obtain the operating frequency of the valve that controls the flow of fluid. If an abnormal peak corresponding to the operating frequency of the valve appears in the spectrum, it means that the equipment valve is faulty and a valve fault signal is generated;

[0158] Based on the above types of problems, corresponding alarm signals are generated according to the actual situation, and targeted alarm reminders are sent to management personnel.

[0159] Example 5

[0160] refer to Figure 3 As shown, the present invention is a method for predicting equipment failure in a digital workshop, comprising: a data acquisition and processing module, an equipment abnormality judgment module, an equipment abnormality analysis module and an equipment analysis and alarm module;

[0161] The data acquisition and processing module is used to obtain the vibration signal value of the equipment and extract the time domain characteristics of the vibration signal, including the vibration peak value, average data, standard deviation and kurtosis of the vibration signal;

[0162] Based on the vibration peak value, average data, standard deviation and kurtosis of the vibration signal, the vibration peak value ratio Fzp, mean value ratio Jzp, standard deviation Bzp and kurtosis ratio Qdp are obtained;

[0163] The equipment abnormality judgment module is used to assign a weight w1 to the vibration peak ratio, a weight w2 to the mean ratio, a weight w3 to the standard deviation, and a weight w4 to the kurtosis ratio;

[0164] The vibration score ZD of the equipment is calculated by the formula ZD = w1×Fzp+w2×Jzp+w3×Bzp+w4×Qdp;

[0165] Generate equipment abnormality signal and equipment normal signal based on the vibration score ZD of the equipment;

[0166] The equipment abnormality analysis module performs abnormal analysis on the vibration peak value based on the equipment abnormal signal. If the vibration peak value is abnormal, it generates an early warning signal of equipment bearing damage and gear meshing.

[0167] Perform abnormal analysis on average data. If the average data is abnormal, generate equipment imbalance and misalignment warning signals;

[0168] Perform abnormal analysis on the standard deviation. If the standard deviation is abnormal, generate an early warning signal for internal wear of the equipment.

[0169] Perform abnormal analysis on the kurtosis. If the kurtosis is abnormal, generate a warning signal of local equipment failure.

[0170] The equipment analysis and alarm module integrates all vibration signals within the monitoring period and constructs a spectrum diagram based on equipment abnormality signals;

[0171] Based on the spectrum diagram, identify the fundamental frequency of the equipment and its harmonic frequencies, and obtain abnormal peaks;

[0172] Based on the equipment imbalance and misalignment warning signals, bearing damage and gear meshing warning signals, equipment internal wear warning signals and equipment local failure warning signals, each warning signal is analyzed separately to obtain the specific abnormal cause of the equipment and send an alarm notification.

[0173] Working principle of the present invention:

[0174] The present invention can more accurately capture the slight changes in the operation of the equipment through the optimized time domain feature extraction algorithm, especially greatly improve the sensitivity to early fault signals. Compared with the traditional method based on frequency domain analysis, the present invention can effectively reduce the misjudgment phenomenon caused by signal frequency overlap. In addition, by introducing the feature weight calculation mechanism, the recognition accuracy of key features is further improved, so that the system can accurately judge the equipment status in a complex working environment, thereby effectively preventing potential equipment failures and reducing the risk of unplanned downtime;

[0175] By using data processing technology, the present invention realizes the real-time collection and analysis of equipment vibration signals, which not only means that the equipment status information can be fed back instantly, but also the monitoring strategy can be quickly adjusted according to the changes in actual working conditions. For example, when an abnormal signal is detected, the system can automatically trigger the early warning mechanism and take timely countermeasures. At the same time, due to the use of a modular design concept, the monitoring parameters can be flexibly configured according to different equipment types and usage scenarios, which greatly enhances the scope of application and practicality of the system.

[0176] The above are only preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for predicting equipment failure in a digital workshop, characterized in that: The following steps are involved: Obtain vibration signal data of the device, extract vibration signal features, and obtain the vibration score of the device; Among them, the vibration signal characteristics include the vibration peak value, mean value, standard deviation and kurtosis of the vibration signal; Generate equipment abnormality signal and equipment normal signal based on the vibration score of the equipment; Based on the abnormal signal of the equipment, the vibration signal data is analyzed for abnormality. If the vibration signal data is abnormal, a corresponding early warning signal is generated; Based on the abnormal signals of the equipment, all vibration signals within the monitoring period are integrated to construct a spectrum diagram; Based on the spectrum diagram, identify the fundamental frequency of the equipment and its harmonic frequencies, and obtain abnormal peaks; Based on the early warning signals, each early warning signal is analyzed separately to obtain the specific abnormal cause of the equipment and send an alarm notification.

2. The method for predicting equipment failure in a digital workshop according to claim 1, characterized in that: The specific process of extracting vibration signal features is as follows: Divide the monitoring period into several time sub-units of equal length, divide the time sub-unit into several time small units of equal length, and collect the vibration signal value of the equipment when the time small unit is in the middle moment; All vibration signal values ​​in the time subunit are integrated into a vibration sequence, and the vibration signal value with the largest absolute value in the vibration sequence is extracted and recorded as the vibration peak value Fz; Based on the vibration peak value Fz, the vibration peak value ratio Fzp of the vibration signal in the time subunit is obtained; Based on the vibration sequence, the average data Jz of the vibration signal in the time subunit is obtained by mean calculation; Based on the average data Jz, the mean ratio Jzp and standard deviation Bz of the vibration signal in the time subunit are obtained; Based on the standard deviation Bz, the standard deviation ratio Bzp of the vibration signal in the time subunit and the kurtosis Qd of the vibration signal in the time subunit are obtained; The kurtosis ratio Qdp of the vibration signal in the time subunit is obtained based on the kurtosis Qd.

3. The method for predicting equipment failure in a digital workshop according to claim 2, characterized in that: The specific process of obtaining the vibration score of the equipment is: Based on the vibration peak ratio Fzp, mean ratio Jzp, standard deviation ratio Bzp and kurtosis ratio Qdp of the vibration signal in the time subunit, a weight w1 is assigned to the vibration peak ratio Fzp, a weight w2 is assigned to the mean ratio Jzp, a weight w3 is assigned to the standard deviation ratio Bzp, and a weight w4 is assigned to the kurtosis ratio Qdp; The vibration score ZD of the equipment is calculated by the formula ZD=w1×Fzp+w2×Jzp+w3×Bzp+w4×Qdp.

4. The method for predicting equipment failure in a digital workshop according to claim 3 is characterized in that: The process of generating device abnormal signal and device normal signal is: Based on the vibration score ZD of the device, comparing the vibration score ZD of the device with a vibration score threshold of the device; If the vibration score ZD of the device is greater than or equal to the vibration score threshold of the device, it means that there is an abnormality in the device, and a device abnormality signal is generated; If the vibration score ZD of the device is less than the vibration score threshold of the device, it means that there is no abnormal problem with the device, and a normal signal of the device is generated.

5. The method for predicting equipment failure in a digital workshop according to claim 2, characterized in that: The specific process of abnormal analysis of vibration signal data is as follows: Based on the vibration peak value Fz of the vibration signal in the time subunit, the mean value uF and standard deviation σ of the vibration peak value are obtained F ; By formula Calculate and obtain the statistical value ZF of the vibration peak value; Based on the average data Jz of the vibration signal in the time subunit, the mean Ju and standard deviation σ of the average data are obtained J ; By formula Calculate the statistical value ZJ of the average data; Based on the standard deviation Bz of the vibration signal in the time subunit, the mean Bu of the standard deviation and the standard deviation σ of the standard deviation Bz are obtained. F ; By formula Calculate and obtain the statistical value ZF of the vibration peak value; Based on the kurtosis Qd of the vibration signal in the time subunit, the mean Qu and standard deviation σ of the kurtosis are obtained Q ; By formula The kurtosis statistic ZQ is calculated.

6. The method for predicting equipment failure in a digital workshop according to claim 5, characterized in that: The specific process of generating corresponding warning signals: Based on the statistic ZF of the vibration peak, the statistic ZJ of the average data, the statistic ZB of the standard deviation, and the statistic ZQ of the kurtosis, the absolute value of the statistic ZF of the vibration peak is compared with the statistic threshold of the vibration peak; If the absolute value of the statistical value ZF of the vibration peak value is greater than or equal to the statistical value threshold of the vibration peak value, an early warning signal of equipment bearing damage and gear meshing is generated; Compare the statistical value ZJ of the average data with the statistical value threshold of the average data; If the statistical value ZJ of the average data is greater than or equal to the statistical value threshold of the average data, an equipment imbalance and misalignment warning signal is generated; Compare the statistical value ZB of the standard deviation of the vibration signal in the time subunit with the statistical value threshold of the standard deviation; If the statistical value ZB of the standard deviation is greater than or equal to the statistical value threshold of the standard deviation, an internal wear warning signal of the equipment is generated; Compare the kurtosis statistic ZQ of the vibration signal in the time subunit with the kurtosis statistic threshold; If the kurtosis statistical value ZQ is greater than or equal to the kurtosis statistical value threshold, it means that the kurtosis is abnormal, and a local equipment failure warning signal is generated.

7. The method for predicting equipment failure in a digital workshop according to claim 1, characterized in that: The process of constructing a spectrogram is: Integrate all vibration signal data within the monitoring period, use the fast Fourier transform algorithm to convert the vibration signal data into the frequency domain, and obtain the spectrum data of the vibration signal within the monitoring period; The converted spectrum data is displayed as a graph, with the horizontal axis representing frequency and the vertical axis representing amplitude.

8. The method for predicting equipment failure in a digital workshop according to claim 7, characterized in that: The process of identifying the fundamental frequency and harmonic frequencies of the equipment and obtaining the abnormal peaks is as follows: Get the motor speed Zs of the device through the formula Get the fundamental frequency f0 of the device, and traverse the spectrum graph to get the frequency peak closest to the fundamental frequency f0 in the spectrum graph, which is recorded as the fundamental frequency peak; Identify the frequency peak that is an integer multiple of the fundamental frequency f0 in the spectrum graph and record it as the harmonic frequency nf0 of the device; Wherein, the harmonic frequency is an integer multiple of the fundamental frequency, n is the multiple, and n is a positive integer greater than 1, n=2, 3, ..., n, when the fundamental frequency is f0, then 2f0, 3f0, ..., nf0 are the harmonic frequencies; Obtain peaks in the spectrum that are inconsistent with the fundamental frequency and fundamental frequency harmonic frequencies, calculate the difference between these peaks and the fundamental frequency peak, and if the difference is greater than the peak difference threshold, mark the peak as an abnormal peak.

9. The method for predicting equipment failure in a digital workshop according to claim 8, characterized in that: The process of analyzing each warning signal separately is as follows: Based on the equipment imbalance and misalignment warning signals, the fundamental frequency peak value is compared with the fundamental frequency peak value threshold value. If the fundamental frequency peak value is greater than or equal to the fundamental frequency peak value threshold value, an equipment imbalance alarm signal is generated; The peak values ​​at the base frequency and twice the base frequency are compared with the base frequency peak value threshold at the same time. If the peak values ​​at the base frequency and twice the base frequency are both greater than or equal to the base frequency peak value threshold, an equipment misalignment alarm signal is generated. Based on the equipment bearing damage and gear meshing warning signals, the number and diameter of the rolling elements in the equipment bearing, the diameter of the raceway in the bearing, the number of teeth and the speed of the driving gear in the equipment gearbox are obtained to obtain the bearing characteristic frequency and the gear meshing frequency; Based on the bearing characteristic frequency and the gear meshing frequency, if an abnormal peak corresponding to the bearing characteristic frequency or the gear meshing frequency appears in the spectrum graph, a bearing damage alarm signal or a gear damage alarm signal is generated; Based on the internal wear warning signal of the equipment, the gear rotation frequency is obtained through the gear speed. Based on the gear rotation frequency, if an abnormal peak corresponding to the gear rotation frequency appears in the spectrum graph, a gear wear alarm signal is generated; Based on the number and diameter of the rolling elements in the equipment bearing and the diameter of the raceway in the bearing, the cage rotation frequency is obtained; Based on the cage rotation frequency, if an abnormal peak corresponding to the cage rotation frequency appears in the spectrum graph, a bearing wear alarm signal is generated; Based on the equipment local fault warning signal, the equipment power frequency is obtained. If an abnormal peak corresponding to the equipment power frequency appears in the spectrum graph, a motor fault alarm signal is generated; Obtain the operating frequency of the pump or compressor of the equipment, and if an abnormal peak corresponding to the operating frequency of the pump or compressor appears in the spectrum graph, generate a pump or compressor fault signal; If the device has a valve for controlling the flow of fluid, the operating frequency of the valve is obtained. If an abnormal peak corresponding to the operating frequency of the valve appears in the spectrum graph, a valve fault signal is generated.

10. An equipment failure prediction system for a digital workshop, characterized in that: Including data acquisition and processing module, equipment abnormality judgment module, equipment abnormality analysis module and equipment analysis and alarm module; The data acquisition and processing module is used to obtain the vibration signal data of the equipment, extract the vibration signal characteristics, and obtain the vibration score of the equipment; The vibration signal characteristics include the vibration peak value, mean value, standard deviation and kurtosis of the vibration signal; The equipment abnormality judgment module generates equipment abnormality signals and equipment normal signals based on the vibration scores of the equipment; The equipment abnormality analysis module performs abnormal analysis on the vibration signal data based on the equipment abnormal signal. If the vibration signal data is abnormal, a corresponding warning signal is generated; The equipment analysis and alarm module integrates all vibration signals within the monitoring period and constructs a spectrum diagram based on equipment abnormality signals; Based on the spectrum diagram, identify the fundamental frequency of the equipment and its harmonic frequencies, and obtain abnormal peaks; Based on the early warning signals, each early warning signal is analyzed separately to obtain the specific abnormal cause of the equipment and send an alarm notification.

Images