A method for evaluating the health state of a rotating equipment

Abstract

The application discloses a kind of rotating equipment health state evaluation methods, belong to equipment state intelligent analysis technical field.The method is by obtaining the vibration signal in the operation process of rotating equipment, establishes health state analysis model, analyzes the health state change of rotating equipment under different rotating equipment operation length;According to the health state analysis model established, determine the size of the sliding window that is intercepted when the vibration signal analysis of rotating equipment under current state;According to the size of the sliding window that is intercepted when the vibration signal analysis of rotating equipment under current state, determine the health evaluation value of rotating equipment under current state;The health evaluation value of rotating equipment under current state is monitored, to judge whether it needs to be alarmed, management personnel are reminded according to alarm signal in advance to the rotating equipment maintenance, to reduce the safety risk of equipment failure, improve the self-adaptability of system and the accuracy of health state analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent analysis of equipment state, in particular to a rotating equipment health state evaluation method. BACKGROUND

[0002] Rotating equipment is a core component in industrial production, and its running state is directly related to production efficiency and safety. Once a fault occurs, it may cause equipment downtime, production interruption, and even safety accidents. Traditional rotating equipment health state evaluation mainly relies on periodic maintenance and manual inspection, which is not only inefficient but also difficult to grasp the running state of the equipment in real time, leading to misjudgment or missed detection. With the development of sensor technology and artificial intelligence technology, real-time monitoring of vibration signals and intelligent evaluation of rotating equipment health state have realized intelligent management of equipment and reduced safety risks caused by equipment failure.

[0003] Rotating equipment gradually degrades in health state during service. The process is relatively slow, and the signal change is not obvious in the early stage. The vibration signal characteristics remain basically unchanged in the early health state, and become more obvious monotony in the decline period. Existing fault diagnosis algorithms use a fixed size sliding window for time domain analysis, but it is difficult for the system to select vibration signal characteristics with obvious monotony over time in a limited number of tests, resulting in difficulty in extracting rotating equipment fault features and low accuracy of health state evaluation. SUMMARY

[0004] The present application aims to provide a rotating equipment health state evaluation method to solve the problems raised in the background.

[0005] To solve the above technical problems, the present application provides the following technical solution: a rotating equipment health state evaluation method, comprising the following steps:

[0006] Step S1, acquire the vibration signal in the rotating equipment operation process; establish a cloud database to store the vibration signals in the operation process of different rotating equipment as historical data; based on the historical vibration signals of the rotating equipment fault state in the cloud database, establish a health state analysis model to analyze the health state change of the rotating equipment under different rotating equipment operation time;

[0007] Step S2, determine the sliding window size change range for rotating equipment vibration signal analysis, and determine the total operation time of the rotating equipment in the current state, and determine the sliding window size for rotating equipment vibration signal analysis in the current state according to the health state analysis model established in step S1;

[0008] Step S3, the state of the rotating equipment during initial operation is taken as a full health state, vibration signals of the rotating equipment in the full health state are analyzed to obtain health factors of the rotating equipment in the full health state; vibration signals of the rotating equipment in a fault state are analyzed to obtain health factors of the rotating equipment in the fault state;

[0009] Step S4, according to the sliding window size determined in step S2, vibration signals of the rotating equipment in the current state are analyzed to obtain health factors of the rotating equipment in the current state, and according to the health factors of the rotating equipment in the full health state and the health factors of the rotating equipment in the fault state, a health evaluation value of the rotating equipment in the current state is determined.

[0010] Compared with the prior art, the present application has the following beneficial effects: by establishing a health state analysis model, the health state change of the rotating equipment under different total operation time is analyzed, so that the fault degree of the rotating equipment is determined according to the health state of the rotating equipment, and the fault characteristics of the rotating equipment are better captured; the sliding window size is adaptively adjusted according to the health state of the rotating equipment, so that the adaptive ability of the system and the accuracy of the health state analysis are improved; the abnormal rotating equipment is alarmed to remind the management personnel to repair the rotating equipment in advance, so that the safety risk of equipment failure is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 is a step schematic diagram of a rotating equipment health state evaluation method of the present application. DETAILED DESCRIPTION

[0012] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0013] The present application analyzes the health state change of the rotating equipment under different rotating equipment operation time according to the historical vibration signals, and dynamically adjusts the sliding window size intercepted during vibration signal analysis in combination with the real-time total operation time of the rotating equipment, so that the fault characteristics of the rotating equipment are better captured; according to the intercepted sliding window size, the vibration signals during the operation of the rotating equipment are analyzed to determine health factors of the rotating equipment in different states, and the health evaluation value is determined according to the health factors of the rotating equipment in different states, so that the health state of the rotating equipment is evaluated, and the adaptability of the system and the accuracy of the health state analysis are improved.

[0014] Please refer toFigure 1 The present invention provides the following technical solution:

[0015] Please see Figure 1 In this first embodiment, a method for assessing the health status of rotating equipment is provided, which includes the following steps:

[0016] Step S1: Obtain vibration signals during the operation of rotating equipment; establish a cloud database to store vibration signals from different rotating equipment operation processes as historical data; based on the historical vibration signals of rotating equipment fault states in the cloud database, establish a health status analysis model to analyze the changes in the health status of rotating equipment under different operating durations.

[0017] Specifically, the steps are as follows:

[0018] Step S11: Retrieve historical vibration signals from the cloud database indicating rotating equipment malfunctions and analyze them. Based on the retrieved historical vibration signals from different rotating equipment, determine the timestamps of the malfunctions and obtain the total operating time B1, B2, ..., B when different rotating equipment malfunctions. z Where z represents the number of rotating equipment with rotating equipment failures retrieved and analyzed from the cloud database; based on B1, B2, ..., B z The maximum value in the range is used to divide the operating time of the rotating equipment into time periods [0, b1], (b1, b2], ..., (b r-1 ,b r ], B1, B2, ..., B z Substitute these values ​​into the divided time periods to determine B1, B2, ..., B z The frequencies of occurrence in different time periods are P1, P2, ..., P. r Among them, b1, b2, ..., b r-1 b r These represent the operating times of different rotating equipment;

[0019] Step S12: Establish a health status analysis model. The frequency of the total operating time when different rotating equipment malfunctions in different time periods is used as the health status of the rotating equipment. Analyze the changes in the health status of the rotating equipment under different operating times, based on the calculation formula:

[0020]

[0021] Wherein, b represents the rotation equipment operation time length; P(b) represents the rotation equipment health status varying with the rotation equipment operation time length b; k represents the influence rate of the rotation equipment operation time length on the rotation equipment health status; m represents the influence degree of the rotation equipment operation time length on the rotation equipment health status; k>0; m>1.

[0022] Step S13, P1, P2,..., P r respectively as the training parameters of the rotation equipment health status P in step S12, b1, b2,..., b r-1 respectively as the training parameters of the rotation equipment operation time length b in step S12, are substituted into the calculation formula of step S12 to calculate the values of k and m. r respectively as the training parameters of the rotation equipment health status P in step S12, b1, b2,..., b

[0023] It should be noted that the total operation time length represents the total time length of the rotation equipment operation, including the period from the start of the operation of the rotation equipment to the end of the operation; the operation time length represents the time length of the rotation equipment operation; the above different rotation equipment are rotation equipment of the same specification and type, and each rotation equipment has a one-to-one corresponding historical vibration signal; the time period [0, b1], (b1, b2],..., (b r-1 ,b r ] is divided according to the rotation equipment operation time length, the range in each time period interval is the same, the greater the number r of the divided time periods, the higher the accuracy of the analysis of the change of the rotation equipment health status, and P u represents the frequency of B1, B2,..., B z in the u-th time period; in the early operation stage, the rotation equipment ages slowly and has few fault features, with the increase of the rotation equipment operation time length, the equipment gradually ages, the fault features change and intensify, therefore, when analyzing the change of the rotation equipment health status, k>0 and m>1; by establishing the health status analysis model, the frequency of the total operation time length of the different rotation equipment when the fault occurs in the different time periods is taken as the health status of the rotation equipment, the higher the frequency in a certain time period, the more likely the rotation equipment is to fail in the time period, by the above calculation formula, the total operation time length of the different rotation equipment when the fault occurs is taken as the analysis data of the health status of the rotation equipment, which facilitates the dynamic adjustment of the size of the sliding window and better captures the fault features of the rotation equipment; in the embodiment, the rotation equipment fault includes but is not limited to bearing fault, rotation equipment stop operation caused by mechanical structure or component loosening.

[0024] Step S2, determine the sliding window size variation range of the rotating equipment vibration signal analysis, and determine the total working time of the rotating equipment in the current state, and determine the sliding window size of the rotating equipment vibration signal analysis in the current state according to the health state analysis model established in step S1.

[0025] Specifically, the method steps are:

[0026] Step S21, determine the sliding window size variation range [T min ,T max ] of the rotating equipment vibration signal analysis; wherein, T min represents the minimum value of the sliding window length; T max represents the maximum value of the sliding window length; and determine the total working time B0 of the rotating equipment in the current state;

[0027] Step S22, according to T min , T max , B0 and the established health state analysis model, calculate the sliding window size T of the rotating equipment vibration signal analysis in the current state:

[0028]

[0029] Wherein, the value of T min should satisfy the condition: f s represents the sampling frequency of the vibration signal; f max represents the highest frequency component in the vibration signal.

[0030] It should be noted that when analyzing the vibration signal, the vibration signal needs to be divided by the sliding window, at this time, T max and T min are determined according to the maximum and minimum values of the sliding window that can be adjusted in the historical vibration signal analysis process of the cloud database, and T maxThe value of T should make the divided vibration signal contain multiple analysis periods; through the established health state analysis model, the health state change of the rotating equipment under different operating time is taken as analysis data, the total operating time B0 of the rotating equipment under the current state is substituted into the calculation formula of the health state analysis model, the health state of the rotating equipment under the current state is predicted, and the sliding window size T of the vibration signal analysis under the current state of the rotating equipment is calculated through the sliding window size change range and the health state of the rotating equipment under the current state, so that the fault characteristics of the rotating equipment are better captured; in the initial stage of the operation of the rotating equipment, the degradation of the equipment is not obvious at this time, the fault characteristics in the vibration signal may be weak and change rapidly, and the smaller sliding window can improve the time resolution at this time, so that the transient change and weak fault characteristics of the vibration signal are better captured; as the operating time of the rotating equipment increases, the equipment gradually degrades, the fault characteristics are more obvious and enter the stable stage, at this time, the sliding window is increased, the frequency resolution of the vibration signal is improved, the health state of the rotating equipment is evaluated through the analysis of the fault characteristics of the rotating equipment, and the accuracy of the health state analysis of the rotating equipment is improved.

[0031] Step S3, taking the state of the rotating equipment at the initial operation as the completely healthy state, analyzing the vibration signal of the rotating equipment under the completely healthy state to obtain the health factor under the completely healthy state of the rotating equipment; analyzing the vibration signal of the rotating equipment under the fault state to obtain the health factor under the fault state of the rotating equipment.

[0032] Step S4, according to the sliding window size of the vibration signal analysis of the rotating equipment under the current state determined in step S2, analyzing the vibration signal of the rotating equipment under the current state to obtain the health factor under the current state of the rotating equipment, and determining the health evaluation value under the current state of the rotating equipment according to the health factor under the completely healthy state of the rotating equipment and the health factor under the fault state of the rotating equipment.

[0033] Further, the method for analyzing the vibration signal of the rotating equipment to determine the health factor of the rotating equipment is as follows: through Fourier transform, the vibration signal of the rotating equipment is converted from a time domain signal to a frequency domain signal, M s characteristics are extracted from the time domain signal and the frequency domain signal, and a feature matrix is constructed, wherein T represents the sliding window size of the vibration signal analysis; M s represents the number of feature types extracted from the time domain signal and the frequency domain signal; the feature matrix is subjected to standardization processing to obtain a standardization processing result The covariance matrix of is calculated to obtain eigenvalues λ i and orthogonal eigenvectors selecting an orthogonal eigenvector pair corresponding to the maximum eigenvalue of the covariance matrix reconstructing, reducing the dimension number of the features to 1 dimension, and performing smoothing to obtain the health factor. s

[0034] It should be noted that the features extracted from the time domain signal and the frequency domain signal include mean value, peak value, standard deviation, root mean square, skewness, energy, kurtosis, sk mean value, peak factor, sk standard deviation, pulse factor, sk skewness, shape factor, sk kurtosis, and marginal factor.

[0035] Further, the method for determining the health evaluation value of the rotating equipment in the current state is as follows: determining the maximum value h max of the health factor according to the health factor in the completely healthy state of the rotating equipment; determining the minimum value h min of the health factor according to the health factor in the fault state of the rotating equipment; analyzing the vibration signal in the current state of the rotating equipment to obtain the health factors h y , h max , and h min in the current state of the rotating equipment; and calculating the health evaluation value H of the rotating equipment in the current state according to h y , h i , and h max .

[0036]

[0037] wherein y represents the number of health factors obtained by analyzing the vibration signal in the current state of the rotating equipment; h i represents the i th health factor in the current state of the rotating equipment.

[0038] It should be noted that when the vibration signal in the current state of the rotating equipment is analyzed, the vibration signal analyzed is the vibration signal generated in the period from the start of the operation to the end of the operation of the rotating equipment in the current state; a plurality of health factors are generated in the vibration signal analysis process of the rotating equipment in the current state according to the size of the sliding window; wherein the maximum value h max of the health factor is the average value of the health factor in the completely healthy state of the rotating equipment, and when the vibration signal in the completely healthy state of the rotating equipment is analyzed, the size of the sliding window is determined by the total operation time of the rotating equipment in the completely healthy state; the minimum value h min of the health factor is the average value of the health factor in the fault state of the rotating equipment, and when the vibration signal in the fault state of the rotating equipment is analyzed, the size of the sliding window is determined by the total operation time of the rotating equipment in the fault state.

[0039] Further, the health factor, health evaluation value and the size of the intercepted sliding window of the rotating equipment are digitally displayed, and the management personnel can view the digital display results through the interactive platform and adjust the size of the intercepted sliding window during the vibration signal analysis of the rotating equipment.

[0040] In the present embodiment, the health evaluation value of the rotating equipment in the current state is monitored, and whether an alarm is needed is determined according to the health evaluation alarm threshold of the rotating equipment; when the health evaluation value of the rotating equipment in the current state is less than the health evaluation alarm threshold, the rotating equipment is normal, the monitoring is continued, and steps S1-S4 are repeated to recalculate the size T of the intercepted sliding window during the vibration signal analysis of the rotating equipment in the current state, to adaptively adjust the size of the sliding window according to the health state of the rotating equipment, thereby improving the adaptive ability of the system; when the health evaluation value of the rotating equipment in the current state is greater than the health evaluation alarm threshold, the rotating equipment is abnormal, an alarm is given, and an alarm signal is sent to the management personnel; the management personnel can perform maintenance on the rotating equipment in advance according to the alarm signal, thereby reducing the safety risk of equipment failure.

[0041] In another embodiment, when the rotating equipment is in a completely healthy state, the health evaluation value of the rotating equipment is 100 points, and when the rotating equipment is in a fault state, the health evaluation value of the rotating equipment is 0 points; the state of the corresponding single machine or component of the rotating equipment is evaluated to determine the health evaluation value of the rotating equipment in the current state, and when the health evaluation value score is above 80 points, the rotating equipment is normal; when the health evaluation value score is between 40 points and 80 points, a safety warning is given; when the health evaluation value score is less than 40 points, the rotating equipment is faulty, at which time the operation of the equipment is stopped and an alarm signal is sent to the management personnel, thereby facilitating the maintenance of the management personnel.

[0042] In the second embodiment, a rotating equipment health state evaluation system is provided, which includes a signal acquisition module, a cloud database, a model analysis module, a sliding window calculation module, a health evaluation module, a monitoring module and a display module.

[0043] The signal acquisition module is used for acquiring the vibration signal in the operation process of the rotating equipment and sending the acquired vibration signal to the cloud database; the cloud database is used for storing the vibration signal in the operation process of different rotating equipment as historical data; the model analysis module is used for establishing a health state analysis model based on the historical vibration signal of the rotating equipment fault state existing in the cloud database, analyzing the health state change of the rotating equipment under different rotating equipment operation time lengths; the sliding window calculation module is used for determining the sliding window size change range intercepted in the rotating equipment vibration signal analysis and determining the total operation time length of the rotating equipment under the current state, determining the sliding window size intercepted in the rotating equipment vibration signal analysis under the current state according to the health state analysis model established in the model analysis module; the health evaluation module is used for analyzing the vibration signal under the current state of the rotating equipment, obtaining the health factor under the current state of the rotating equipment, and determining the health evaluation value under the current state of the rotating equipment according to the health factor under the complete health state of the rotating equipment and the health factor under the fault state of the rotating equipment; the monitoring module is used for monitoring the health evaluation value of the rotating equipment under the current state, judging whether alarm is needed according to the health evaluation alarm threshold of the rotating equipment, and sending an alarm signal to the manager if alarm is needed; and the display module is used for digitally displaying the health factor, the health evaluation value and the intercepted sliding window size of the rotating equipment, so that the manager can view the digital display result through the interactive platform and adjust the intercepted sliding window size in the rotating equipment vibration signal analysis.

[0044] Finally, it should be noted that: the above only for the preferred embodiments of the present application, and not for limiting the present application, although the foregoing detailed description of the present application is made with reference to the foregoing embodiments, for those skilled in the art, it still can be modified, or equivalent replacement of part of the technical features described in the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method of assessing the health of a rotating equipment, characterized by: The method comprises the following steps: Step S1, obtaining vibration signals in the operation process of the rotating equipment; establishing a cloud database, and storing the vibration signals in the operation process of different rotating equipment as historical data; based on the historical vibration signals of the rotating equipment fault state in the cloud database, a health state analysis model is established to analyze the health state change of the rotating equipment under different rotating equipment operation time; Step S2, determining the change range of the sliding window size intercepted during the analysis of the rotating equipment vibration signal, and determining the total operation time of the rotating equipment under the current state, and determining the sliding window size intercepted during the analysis of the rotating equipment vibration signal under the current state according to the health state analysis model established in step S1; the specific process is: Step S21, determine the sliding window size variation range [T min ,T max ] when analyzing the rotating equipment vibration signal; wherein T min represents the minimum value of the sliding window length; T max represents the maximum value of the sliding window length; and determine the total working time B0 of the rotating equipment under the current state; Step S22, according to T min , T max , B0 and the established health state analysis model, the size T of the sliding window intercepted in the vibration signal analysis of the rotating equipment under the current state is calculated: wherein T min The value of T f s represents the sampling frequency of the vibration signal; f max represents the highest frequency component in the vibration signal; k represents the influence rate of the operating time of the rotating equipment on the health status of the rotating equipment; and m represents the influence degree of the operating time of the rotating equipment on the health status of the rotating equipment. Step S3, taking the state of the rotating equipment in the initial operation as a completely healthy state, analyzing the vibration signal of the rotating equipment in the completely healthy state to obtain the health factor of the rotating equipment in the completely healthy state; analyzing the vibration signal of the rotating equipment in the fault state to obtain the health factor of the rotating equipment in the fault state; Step S4, according to the sliding window size intercepted during the analysis of the rotating equipment vibration signal under the current state determined in step S2, analyzing the vibration signal of the rotating equipment under the current state to obtain the health factor of the rotating equipment under the current state, and determining the health evaluation value of the rotating equipment under the current state according to the health factor of the rotating equipment in the completely healthy state and the health factor of the rotating equipment in the fault state.

2. A method of assessing the health of a rotating machine according to claim 1, characterized in that: The method step of step S1 is: Step S11, the historical vibration signals of the rotating equipment fault state are called from the cloud database for analysis, the time stamp of the rotating equipment failure is determined according to the historical vibration signals of different rotating equipment, and the total operation time B1, B2,..., B z ; wherein, z represents the data amount of the historical vibration signals of the rotating equipment failure analyzed from the cloud database, each rotating equipment has corresponding historical vibration signals; the time period of the rotating equipment operation time length [0, b1], (b1, b2],..., (b r-1 , r ] is divided, B1, B2,..., B z are substituted into the divided time period, and B1, B2,..., B z are determined. r The frequency P1, P2,..., P r-1 appears in different time periods; b1, b2,..., b r , b r respectively represent the operation time length of different rotating equipment; the size of b z is determined by the maximum value in B1, B2,..., B z . Step S12, establishing a health state analysis model, taking the frequency of the total operation time of different rotating equipment occurring faults in different time periods as the health state of the rotating equipment, analyzing the health state change of the rotating equipment under different rotating equipment operation time, according to the calculation formula: Wherein, b represents the operation time of the rotating equipment; P(b) represents the health state of the rotating equipment changing with the operation time b of the rotating equipment; k represents the influence rate of the operation time of the rotating equipment on the health state of the rotating equipment; m represents the influence degree of the operation time of the rotating equipment on the health state of the rotating equipment; k>0; m>1; Step S13, P1, P2,..., P r b1, b2,..., b r-1 , respectively, as the training parameters of the health state P of the rotating equipment in step S21, are substituted into the calculation formula of step S12 to calculate the values of k and m. r b1, b2,..., b r-1 , respectively, as the training parameters of the health state P of the rotating equipment in step S21, are substituted into the calculation formula of step S12 to calculate the values of k and m.

3. A method of assessing the health of a rotating machine according to claim 2, wherein: The method for analyzing the vibration signal of a rotating device to determine a health factor of the rotating device is: through Fourier transform, the vibration signal of the rotating device is converted from a time domain signal to a frequency domain signal, M s characteristics are extracted from the time domain signal and the frequency domain signal, and a characteristic matrix is formed wherein T represents a size of a sliding window intercepted during vibration signal analysis; M s represents a number of characteristic types extracted from the time domain signal and the frequency domain signal; the characteristic matrix is subjected to standardization processing to obtain a standardization processing result The covariance matrix of the standardization processing result is calculated to obtain eigenvalues λ i and orthogonal eigenvectors The orthogonal eigenvector corresponding to the maximum eigenvalue of the covariance matrix is selected to reconstruct a health factor , the dimension number corresponding to the M s characteristics is reduced to 1 dimension, and the health factor is obtained after smoothing processing.

4. A method of assessing the health of a rotating machine according to claim 3, wherein: The method for determining the health evaluation value under the current state of the rotating equipment comprises the following steps: determining a maximum value h of the health factor under a completely healthy state of the rotating equipment according to the health factor under the completely healthy state of the rotating equipment max ; determining a minimum value h of the health factor under a fault state of the rotating equipment according to the health factor under the fault state of the rotating equipment min ; analyzing a vibration signal under a current state of the rotating equipment to obtain health factors h1, h2,..., h y under the current state of the rotating equipment; and calculating a health evaluation value H under the current state of the rotating equipment according to h max , h min and h1, h1,..., h y . Wherein, y represents the number of health factors obtained by analyzing the vibration signal under the current state of the rotating equipment; h i represents the i th health factor under the current state of the rotating equipment.

5. A method of assessing the health of a rotating machine according to claim 4, wherein: The health factor, health evaluation value and intercepted sliding window size of the rotating equipment are digitally displayed, and the management personnel can view the digital display results through the interactive platform, and adjust the sliding window size intercepted during the analysis of the rotating equipment vibration signal.

6. The method of claim 1, wherein: The health evaluation value of the rotating equipment under the current state is monitored, and whether the alarm is needed is judged according to the health evaluation alarm threshold value of the rotating equipment; when the health evaluation value of the rotating equipment under the current state is less than the health evaluation alarm threshold value, the rotating equipment is normal, the monitoring is continued, and steps S1-S4 are repeated to recalculate the sliding window size T intercepted during the analysis of the rotating equipment vibration signal under the current state; when the health evaluation value of the rotating equipment under the current state is greater than the health evaluation alarm threshold value, the rotating equipment is abnormal, and the alarm is given, and the alarm signal is sent to the management personnel.

Images