A fault diagnosis method, system and storage medium for analyzing equipment rotation state

Abstract

The present invention relates to a fault diagnosis method, system and storage medium for analyzing the rotation state of equipment, and relates to the field of equipment fault analysis, including: when the rotating equipment to be monitored is started, a vibration element set for calibrating the rotation control parameters of the equipment is obtained; the vibration element set is traversed to generate the calibration results of the monitoring points; the high-frequency vibration sensor of the calibration results of the monitoring points is activated, and a list of monitoring vibration state quantities is collected; based on the equipment rotation control parameters and the model of the rotating equipment to be monitored, the calibration results of the monitoring points are traversed to perform frequency statistics, and a list of reference vibration state quantities is obtained; a state deviation matrix is ​​constructed through the monitoring vibration state quantity list and the reference vibration state quantity list; a fault prediction probability is generated through a fault prediction model and sent to a fault diagnosis management end. The technical problem that the prior art cannot realize batch application in industry due to the manual configuration of vibration elements and monitoring points, resulting in low model building efficiency, poor migration ability of the fault diagnosis model, and inability to achieve industrial batch application is solved.

Description

Technical Field

[0001] The present invention relates to the field of equipment fault analysis, and in particular to a fault diagnosis method, system and storage medium for equipment rotation state analysis. Background Art

[0002] Vibration fault diagnosis of rotating equipment usually requires a model for each model. This is because the structures and transmission states of different models of equipment vary greatly, and the vibration elements and monitoring points of different equipment under different working conditions are also quite different. Therefore, when building each model, manual experience is required to configure the vibration elements and monitoring points, resulting in the model being applicable only to specific equipment.

[0003] The existing technology has low model building efficiency due to the manual configuration of vibration elements and monitoring points, and the migration ability of the fault diagnosis model is poor, which cannot be applied in batches in industry. This method limits the versatility and scalability of the fault diagnosis system and is difficult to meet the needs of industrial production for efficient and accurate diagnosis. Summary of the invention

[0004] The present invention aims to solve the technical problems in the prior art that, due to the manual configuration of vibration elements and monitoring points, the model building efficiency is low, the migration ability of the fault diagnosis model is poor, and batch application in industry cannot be achieved. A fault diagnosis method, system and storage medium for equipment rotation state analysis are provided to solve the problems.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides a fault diagnosis method for equipment rotation state analysis, comprising: when a rotating equipment to be monitored is started, obtaining equipment rotation control parameters; performing rotation state analysis according to the equipment rotation control parameters to obtain a vibration element set; traversing the vibration element set to perform monitoring point matching and generate a monitoring point calibration result; activating a high-frequency vibration sensor of the monitoring point calibration result to collect a monitoring vibration state quantity list; traversing the monitoring point calibration result based on the equipment rotation control parameters and the model of the rotating equipment to be monitored to perform frequency statistics and obtain a reference vibration state quantity list; constructing a state deviation matrix through the monitoring vibration state quantity list and the reference vibration state quantity list; fitting the state deviation matrix through a fault prediction model to generate a fault prediction probability, wherein the fault prediction probability corresponds one-to-one to the vibration element set; and sending the fault prediction probability and the vibration element set to a fault diagnosis management terminal.

[0007] In a second aspect, the present invention provides a fault diagnosis system for analyzing the rotation state of an equipment, comprising: an equipment parameter acquisition module, which is used to obtain the equipment rotation control parameters when the rotating equipment to be monitored is started; a rotation state analysis module, which is used to perform rotation state analysis based on the equipment rotation control parameters to obtain a vibration element set; a monitoring point calibration module, which is used to traverse the vibration element set to match the monitoring points and generate a monitoring point calibration result; a vibration state monitoring module, which is used to activate the high-frequency vibration sensor of the monitoring point calibration result and collect a list of monitoring vibration state quantities; a reference vibration state analysis module, which is used to traverse the monitoring point calibration results based on the equipment rotation control parameters and the model of the rotating equipment to be monitored to perform frequency statistics and obtain a list of reference vibration state quantities; a state deviation analysis module, which is used to construct a state deviation matrix through the monitoring vibration state quantity list and the reference vibration state quantity list; a fault probability prediction module, which is used to fit the state deviation matrix through a fault prediction model to generate a fault prediction probability, wherein the fault prediction probability corresponds one-to-one to the vibration element set; and an information sending module, which is used to send the fault prediction probability and the vibration element set to a fault diagnosis management terminal.

[0008] In a third aspect, the present application provides a non-transitory computer-readable storage medium, in which a computer software program is stored. When the computer software program is executed by a processor, the steps of a fault diagnosis method for analyzing the rotation state of a device described in the first aspect are implemented.

[0009] The beneficial effect of the present invention is that the vibration element set and the monitoring point calibration results of the rotating equipment to be monitored are automatically determined based on the equipment rotation control parameters, combined with the rotation state analysis and the monitoring point matching. Furthermore, based on the vibration element set and the monitoring point calibration results, the monitoring vibration state quantity list and the reference vibration state quantity list are analyzed, and then the state deviation matrix is ​​fitted through the fault prediction model to generate a technical solution for the fault prediction probability. The automatic determination of the vibration element set and the monitoring point calibration results is achieved, which is conducive to improving the efficiency of model building and is suitable for mass industrial scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic flow chart of a fault diagnosis method for analyzing a rotation state of an equipment provided by the present invention;

[0011] Figure 2 A structural schematic diagram of a fault diagnosis system for analyzing the rotation state of equipment provided by the present invention;

[0012] Figure 3 A schematic diagram of the structure of a computer-readable storage medium provided by the present invention.

[0013] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0014] Equipment parameter acquisition module 1000, rotation state analysis module 2000, monitoring point calibration module 3000, vibration state monitoring module 4000, reference vibration state analysis module 5000, state deviation analysis module 6000, fault probability prediction module 7000, information sending module 8000, computer readable storage medium 600, computer program 611. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0016] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0017] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0018] Embodiment 1:

[0019] like Figure 1 As shown, an embodiment of the present invention provides a fault diagnosis method for analyzing the rotation state of an equipment, comprising the steps of:

[0020] S10: When the rotating device to be monitored is started, the device rotation control parameters are obtained;

[0021] Specifically, when the rotating equipment to be monitored starts, the equipment rotation control parameters must be obtained first. Equipment rotation control parameters refer to parameters used to control and monitor the operating status of rotating equipment. These parameters may include the equipment's rotation speed, angle, torque, etc. This step is the basis of the fault diagnosis process, because the rotation state of the equipment directly affects the working condition of the vibration element. Accurately obtaining these parameters is helpful for subsequent state analysis and fault prediction. By obtaining equipment rotation control parameters in real time, the accuracy and timeliness of the data can be ensured, providing a reliable basis for subsequent analysis.

[0022] S20: performing rotation state analysis according to the rotation control parameters of the device to obtain a vibration element set;

[0023] Further, a rotation state analysis is performed according to the rotation control parameters of the device to obtain a vibration element set, and step S20 includes the steps of:

[0024] S21: performing self-integration to zero splitting on the rotating equipment to be monitored to obtain a tree diagram of equipment components;

[0025] S22: extracting a direct vibration element set according to the device rotation control parameter;

[0026] S23: traversing the direct vibration component set, sorting the vibration components based on the device component tree diagram, and obtaining an indirect vibration component set;

[0027] S24: Add the direct vibration element set and the indirect vibration element set into the vibration element set.

[0028] Specifically, a rotation state analysis is performed according to the rotation control parameters of the device to obtain a vibration element set. The rotation control parameters of the device include the rotation speed, angle and torque of the device. The rotation state analysis is to determine the state of the device during operation by calculating and evaluating these parameters. First, the rotating device to be monitored is self-integrated to zero split to obtain a device element tree diagram. The device element tree diagram is a tree structure diagram that describes the various components of the device and their mutual relationships. Then, according to the device rotation control parameters, a direct vibration element set is extracted. A direct vibration element set refers to a set of vibration elements directly related to the rotation of the device. Next, the direct vibration element set is traversed, and the vibration elements are sorted based on the device element tree diagram to obtain an indirect vibration element set. The indirect vibration element set is a set of vibration elements that are indirectly related to the rotation of the device through other elements. Finally, the direct vibration element set is merged with the indirect vibration element set to form a vibration element set. The vibration element set is a set of all vibration elements related to the rotation state of the device. Through the state monitoring of these elements, the operation of the device can be fully understood, thereby providing data support for fault diagnosis.

[0029] S30: traversing the vibration element set to perform monitoring point matching and generate a monitoring point calibration result;

[0030] Further, the vibration element set is traversed to perform monitoring point matching and generate monitoring point calibration results. Step S30 includes the following steps:

[0031] S31: Obtaining a first vibration element of the vibration element set;

[0032] S32: When the first vibration element is a deployable monitoring point element, adding the first vibration element to the monitoring point calibration result;

[0033] S32: when the first vibration element does not belong to a deployable monitoring point element, obtaining a set of associated vibration elements of the first vibration element, wherein the associated vibration elements are indirect vibration elements of the first vibration element;

[0034] S33: taking the device rotation control parameter as a constraint, traversing the associated vibration element set, and collecting the associated element test operation state set of the rotating device to be monitored, wherein the associated element test operation state set includes a second vibration amplitude test record value set and a second vibration frequency test record value set;

[0035] S34: taking the device rotation control parameter as a constraint, collecting a first vibration element test operation state of the rotating device to be monitored, wherein the first vibration element test operation state includes a third vibration amplitude test record value and a third vibration frequency test record value;

[0036] S35: configuring a first weight for the vibration amplitude, constructing a vibration amplitude normalized weighting function according to the first weight, configuring a second weight for the vibration frequency, constructing a vibration frequency normalized weighting function according to the second weight, wherein the sum of the first weight and the second weight is equal to 1;

[0037] S36: According to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function, traverse the second vibration amplitude test record value set and the second vibration frequency test record value set for fitting to generate a first coordinate set;

[0038] S37: fitting the third vibration amplitude test record value and the third vibration frequency test record value according to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function to generate a second coordinate;

[0039] S38: traverse the first coordinate set and calculate a Euclidean distance evaluation value set with the second coordinate;

[0040] S39: sorting the associated vibration element set from large to small according to the Euclidean distance evaluation value set to generate an associated vibration element sorting result;

[0041] S310: adding the frontmost element of the deployable monitoring point element in the sorting result of the associated vibration elements into the monitoring point calibration result.

[0042] Specifically, the monitoring point calibration result is a specific location used to monitor the operating status of the equipment.

[0043] First, a first vibration element of the vibration element set is obtained, wherein the first vibration element refers to any element of the vibration element set. If the first vibration element belongs to a deployable monitoring point element, it is added to the monitoring point calibration result, wherein the deployable monitoring point element refers to an element set of a deployable monitoring point preconfigured by a user.

[0044] If the first vibration element does not belong to a deployable monitoring point element, it is necessary to obtain a set of associated vibration elements of the vibration element, where the associated vibration element refers to an indirect vibration element associated with the first vibration element.

[0045] Next, with the equipment rotation control parameters as constraints, the associated vibration element set is traversed to collect the associated element test operation status set of the rotating equipment to be monitored, that is, with the equipment rotation control parameters as constraints, the vibration test data of the associated vibration element set of the rotating equipment to be monitored under the same control state is collected. The associated element test operation status set includes a second vibration amplitude test record value set and a second vibration frequency test record value set. The second vibration amplitude test record value set refers to the vibration amplitude mean value record information in several tests that corresponds one-to-one to the associated vibration element set; the second vibration frequency test record value set refers to the vibration frequency mean value record information in several tests that corresponds one-to-one to the associated vibration element set.

[0046] Then, with the equipment rotation control parameters as constraints, the first vibration element test operation state of the rotating equipment to be monitored is collected. The first vibration element test operation state includes a third vibration amplitude test record value and a third vibration frequency test record value. The third vibration amplitude test record value is a record value of the average value of the vibration amplitude of several tests of the first vibration element when the rotating equipment to be monitored is running with the equipment rotation control parameters; the third vibration frequency test record value is a record value of the average value of the vibration frequency of several tests of the first vibration element when the rotating equipment to be monitored is running with the equipment rotation control parameters.

[0047] A first weight is configured for the vibration amplitude, and a vibration amplitude normalization weighting function is constructed according to the first weight; a second weight is configured for the vibration frequency, and a vibration frequency normalization weighting function is constructed according to the second weight. The sum of the first weight and the second weight is equal to 1. The vibration amplitude normalization weighting function is a function used to standardize the vibration amplitude data, and the vibration frequency normalization weighting function is a function used to standardize the vibration frequency data. According to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function, the second vibration amplitude test record value set and the second vibration frequency test record value set are traversed for fitting to generate a first coordinate set. Then, the third vibration amplitude test record value and the third vibration frequency test record value are fitted to generate a second coordinate. The first coordinate set is traversed to calculate the Euclidean distance evaluation value set with the second coordinate. According to the Euclidean distance evaluation value set from large to small, the associated vibration element set is sorted to generate the associated vibration element sorting result. The Euclidean distance evaluation value set refers to the distance between the first coordinate and the second coordinate, which is used to evaluate the similarity of the two coordinates. The frontmost component of the deployable monitoring point components in the associated vibration component sorting result is added to the monitoring point calibration result.

[0048] In this way, the vibration element and the monitoring point can be accurately matched to generate the monitoring point calibration results. This step helps to ensure the accuracy and reliability of the monitoring data and provide basic data support for subsequent fault diagnosis.

[0049] S40: activating the high-frequency vibration sensor of the monitoring point calibration result, and collecting a list of monitoring vibration state quantities;

[0050] Specifically, the high-frequency vibration sensor of the monitoring point calibration result is activated, and a list of monitoring vibration state quantities is collected. The monitoring point calibration result refers to the monitoring point determined after matching and calibration. The high-frequency vibration sensor is a sensor used for high-frequency monitoring of the vibration of the equipment. Activate the high-frequency vibration sensor. These sensors are deployed at the positions determined in the monitoring point calibration results, and are specifically used to capture the high-frequency vibrations generated by the equipment during operation. Next, a list of monitoring vibration state quantities is collected. The list of monitoring vibration state quantities refers to the vibration data of each monitoring point collected by the high-frequency vibration sensor, including vibration amplitude, vibration frequency, vibration phase, vibration frequency domain characteristics and vibration time domain characteristics. These data can reflect the vibration state of each monitoring point of the equipment during operation. In this way, the operating status of the equipment can be monitored in real time and potential vibration anomalies can be captured. These data provide the necessary basis for subsequent state analysis and fault prediction, and help improve the accuracy and timeliness of equipment fault diagnosis.

[0051] S50: Based on the equipment rotation control parameters and the model of the rotating equipment to be monitored, the calibration results of the monitoring points are traversed to perform frequency statistics to obtain a list of reference vibration state quantities;

[0052] Specifically, the reference vibration state quantity list refers to the list of standard vibration data of each monitoring point when the equipment is in normal working condition. These reference data reflect the vibration characteristics of the equipment during normal operation and are used for subsequent state comparison and fault diagnosis. Frequency statistics refers to the process of statistically analyzing the vibration data of each monitoring point under the constraints of the rotation control parameters of the equipment and the model of the rotating equipment to be monitored, calculating the vibration state quantity of the frequency triggering the frequency threshold, and setting it as the reference vibration state quantity. In this way, a reference vibration state of the equipment can be established to ensure reliable reference data when performing state analysis and fault prediction. This helps to improve the accuracy and timeliness of equipment fault diagnosis.

[0053] S60: constructing a state deviation matrix through the monitoring vibration state quantity list and the reference vibration state quantity list;

[0054] Specifically, the state deviation matrix refers to a matrix that characterizes the deviation information between the monitored vibration state quantity list and the reference vibration state quantity list. Preferably, each column element of the matrix is ​​the state deviation between the monitored vibration state quantity and the reference vibration state quantity of different attribute state quantities at the same monitoring point, and each row is the state deviation between the monitored vibration state quantity and the reference vibration state quantity at different monitoring points with the same attribute.

[0055] The detailed construction process of the state deviation matrix is ​​as follows: match each data point in the list of monitored vibration state quantities with the corresponding data point in the list of reference vibration state quantities. Ensure that the real-time vibration data of each monitoring point can correspond to the reference data under its normal working state. Next, calculate the state deviation of each monitoring point. The state deviation is the difference between the monitored vibration state quantity and the reference vibration state quantity, which is usually calculated by the following formula: state deviation = monitored vibration state quantity − reference vibration state quantity state. Then, arrange the state deviation values ​​of all monitoring points into a matrix, called the state deviation matrix. In this way, you can intuitively see whether the operating status of each part of the equipment deviates from the normal value, providing basic data for subsequent fault prediction. The state deviation matrix can help identify potential abnormal areas and improve the accuracy and timeliness of equipment fault diagnosis.

[0056] S70: Fitting the state deviation matrix through a fault prediction model to generate a fault prediction probability, wherein the fault prediction probability corresponds to the vibration element set one by one;

[0057] S80: Sending the fault prediction probability and the vibration element set to a fault diagnosis management terminal.

[0058] Specifically, the fault prediction model is a model used to predict the probability of equipment failure, which may be based on machine learning or statistical analysis methods. The state deviation matrix is ​​input into the fault prediction model, which can be a machine learning model trained with historical data, capable of identifying abnormal patterns in the state deviation matrix and associating these patterns with potential faults. The model evaluates the operating state of each vibration element based on the size and distribution of the state deviation. Then, a fault prediction probability is generated. The fault prediction probability refers to the probability of failure of each vibration element, expressed as a probability value. The higher this probability value, the more likely the vibration element is to fail. In this way, the failure risk of each vibration element can be clearly pointed out, helping technicians to focus on those elements with a higher probability of failure through the fault diagnosis management end for further inspection and maintenance. In this way, the potential failure of the equipment can be predicted in advance, the initiative and pertinence of equipment maintenance can be improved, and the risk of unexpected equipment downtime and damage can be reduced, thereby improving the overall reliability and operating efficiency of the equipment.

[0059] Further, the direct vibration component set is traversed, and vibration components are sorted based on the device component tree diagram to obtain an indirect vibration component set. Step S23 includes the following steps:

[0060] S231: Obtain a first direct vibration element of the direct vibration element set;

[0061] S232: extracting the parent node element of the first direct vibration element from the device element tree diagram;

[0062] S233: performing vibration probability analysis on the parent node element based on the device rotation control parameter to obtain a first vibration identification probability;

[0063] S234: When the first vibration identification probability is greater than or equal to a vibration identification probability threshold, performing vibration element sorting based on the device element tree diagram according to the parent node element to obtain the indirect vibration element set;

[0064] S235: when the first vibration identification probability is less than the vibration identification probability threshold, extracting a brother node element of the first direct vibration element from the device element tree diagram;

[0065] S236: Performing vibration probability analysis on the brother node element based on the device rotation control parameter to obtain a second vibration identification probability;

[0066] S237: when the second vibration identification probability is greater than or equal to the vibration identification probability threshold, adding the brother node element into the first indirect vibration element set, and adding the first indirect vibration element set into the indirect vibration element set;

[0067] Specifically, the first direct vibration element refers to any direct vibration element of the direct vibration element set. The determination process of the first indirect vibration element set is described by taking the first direct vibration element as an example. The analysis methods of the indirect vibration element sets of other direct vibration elements are exactly the same. The detailed process is as follows:

[0068] The parent node element refers to the upper level element directly connected to the first direct vibration element in the device element tree diagram. The vibration probability analysis of the parent node element is performed based on the device rotation control parameters to obtain the first vibration identification probability. The device rotation control parameters include the rotation speed, angle and torque of the device. The first vibration identification probability refers to the possibility of vibration of the parent node element.

[0069] When the first vibration identification probability is greater than or equal to the vibration identification probability threshold, the vibration elements are sorted based on the parent node element and the device element tree diagram to obtain the indirect vibration element set. The vibration identification probability threshold refers to the probability threshold for judging whether the parent node element is an indirect vibration element. What is meant here is that when the first vibration identification probability is greater than or equal to the vibration identification probability threshold, the parent node element is an indirect vibration element, and because the parent node element is composed of multiple elements including the first direct vibration element, all the branch nodes from the parent node element downward are indirect vibration elements. At this time, it is also necessary to analyze the parent node of the parent node, or to analyze the brother node of the parent node, so the same analysis process as the first direct vibration element is performed on the parent node, that is, the algorithm flow of vibration element sorting is performed again. It converges until a certain node has no parent node elements and brother node elements greater than or equal to the vibration identification probability threshold, and outputs all indirect vibration elements.

[0070] When the first vibration identification probability is less than the vibration identification probability threshold, the sibling node element of the first direct vibration element is extracted from the device element tree diagram. Sibling node elements refer to other elements at the same level as the first direct vibration element in the device element tree diagram. A vibration probability analysis is performed on the sibling node element based on the device rotation control parameter to obtain a second vibration identification probability. The second vibration identification probability refers to the possibility of vibration of the sibling node element. When the second vibration identification probability is greater than or equal to the vibration identification probability threshold, the sibling node element is added to the first indirect vibration element set. The first indirect vibration element set refers to the set of indirect vibration elements of the screened first direct vibration element. The first indirect vibration element set is added to the indirect vibration element set, and then other direct vibration elements are traversed in the same way to complete the screening of indirect vibration elements.

[0071] In this way, all indirect vibration elements related to the rotational state of the equipment can be accurately screened out, which helps to comprehensively monitor the operation of the equipment and improve the accuracy and comprehensiveness of fault diagnosis.

[0072] The step of performing vibration probability analysis on the parent node element based on the device rotation control parameter to obtain a first vibration identification probability includes:

[0073] Taking the device rotation control parameter as a constraint, collecting the parent node element test operation status of the rotating device to be monitored, wherein the parent node element test operation status has a first vibration amplitude test record value and a first vibration frequency test record value;

[0074] When the first vibration amplitude test record value is greater than or equal to the vibration amplitude threshold, and the first vibration frequency test record value is greater than or equal to the vibration frequency threshold, the first vibration identification probability is set to 1;

[0075] When the first vibration amplitude test record value is smaller than the vibration amplitude threshold, or / and the first vibration frequency test record value is smaller than the vibration frequency threshold, the first vibration identification probability is set to 0.

[0076] Specifically, the detailed process of vibration probability analysis is as follows:

[0077] Taking the rotation control parameters of the equipment as constraints, collect the test operation status of the parent node element of the rotating equipment to be monitored under the same control state. The test operation status refers to the actual working status data of the parent node element during the operation of the equipment. Specifically, these status data include the first vibration amplitude test record value and the first vibration frequency test record value. The first vibration amplitude test record value: refers to the vibration amplitude test data of the parent node element. The first vibration frequency test record value: refers to the vibration frequency test data of the parent node element.

[0078] Next, conduct vibration probability analysis:

[0079] When the first vibration amplitude test record value is greater than or equal to the vibration amplitude threshold, and the first vibration frequency test record value is greater than or equal to the vibration frequency threshold, the first vibration identification probability is set to 1. The vibration amplitude threshold and the vibration frequency threshold are reference values ​​for judging whether the vibration state is abnormal. Vibration amplitude threshold: a reference value for judging whether the vibration amplitude is abnormal. Vibration frequency threshold: a reference value for judging whether the vibration frequency is abnormal.

[0080] When the first vibration amplitude test record value is less than the vibration amplitude threshold, or / and the first vibration frequency test record value is less than the vibration frequency threshold, the first vibration identification probability is set to 0. This means that under these conditions, the possibility of vibration of the parent node element is low.

[0081] In this way, the vibration probability of the parent node element can be accurately calculated, ensuring reliable data support when screening vibration elements. This helps to improve the accuracy and timeliness of fault diagnosis and ensure the reliability and effectiveness of monitoring data.

[0082] Further, based on the equipment rotation control parameters and the model of the rotating equipment to be monitored, the calibration results of the monitoring points are traversed to perform frequency statistics to obtain a list of reference vibration state quantities. Step S50 includes the following steps:

[0083] S51: setting the model of the rotating equipment to be monitored as a fixed constraint condition, and setting a fuzzy constraint condition based on the equipment rotation control parameter;

[0084] S52: According to the fixed constraint condition and the fuzzy constraint condition, a fault-free sample is collected for the first monitoring point of the monitoring point calibration result to obtain a first vibration descriptor recorded state quantity set;

[0085] S53: performing extreme value analysis on the first vibration descriptor recorded state quantity set to obtain an extreme value recorded state quantity set;

[0086] S54: deleting the extreme value record state quantity set from the first vibration descriptor record state quantity set, obtaining the mode value of the remaining record state quantity set of the first vibration descriptor, setting it as the first vibration descriptor reference state quantity, and adding it to the first monitoring point reference vibration state quantity;

[0087] S55: Adding the first monitoring point reference vibration state quantity into the reference vibration state quantity list.

[0088] Specifically, the model of the rotating equipment to be monitored is set as a fixed constraint condition. The fixed constraint condition refers to using the model of the equipment to be monitored as a screening condition to ensure that the statistical results are related to the specific equipment model. Next, fuzzy constraints are set based on the equipment rotation control parameters. Fuzzy constraints refer to setting ranges or conditions based on equipment rotation control parameters, allowing certain deviations to ensure that the statistical results can reflect the actual working status.

[0089] According to the fixed constraint condition and the fuzzy constraint condition, a fault-free sample is collected for the first monitoring point of the monitoring point calibration result to obtain a first vibration descriptor record state quantity set. Fault-free sample collection refers to collecting vibration data of the monitoring point under normal operation of the equipment. The vibration descriptor record state quantity refers to a set of parameters that describe the vibration state, including vibration amplitude and vibration frequency, etc.

[0090] The first vibration descriptor record state quantity set is subjected to extreme value analysis to obtain an extreme value record state quantity set. Extreme value analysis refers to identifying the maximum and minimum values ​​in the data to determine abnormal data. The extreme value record state quantity set refers to an extreme value data set identified in the vibration data.

[0091] Preferably, the extreme value analysis method is as follows: sort the recorded state quantities of any vibration descriptor in order from small to large to obtain the sorted result of the recorded state quantities; multiply the number of the recorded state quantities by 0.25, round up to obtain the first serial number, and extract the first eigenvalue from the sorted result of the recorded state quantities according to the first serial number; multiply the number of the recorded state quantities by 0.75, round down to obtain the second serial number, and extract the second eigenvalue from the sorted result of the recorded state quantities according to the second serial number; calculate the deviation between the second eigenvalue and the first eigenvalue, and set it as the extreme value; subtract 1.5 times the extreme value from the first eigenvalue to obtain the lower boundary of the extreme value; add 1.5 times the extreme value to the second eigenvalue to obtain the upper boundary of the extreme value. The recorded state quantities that are less than the lower boundary of the extreme value or greater than the upper boundary of the extreme value are regarded as the set of extreme recorded state quantities. The same analysis is performed on the vibration descriptor of each attribute to obtain the set of extreme recorded state quantities of the vibration descriptor of each attribute.

[0092] Delete the first vibration descriptor record state quantity set corresponding to the extreme value record state quantity set of any attribute vibration descriptor, obtain the remaining record state quantity set of the first vibration descriptor, and then calculate the mode value of the remaining record state quantity set of the first vibration descriptor, and set it as the first vibration descriptor benchmark state quantity. The mode value refers to the value with the highest frequency of occurrence in the data set, which represents the normal state of the monitoring point. The first vibration descriptor benchmark state quantity refers to the normal vibration state of the monitoring point after excluding the extreme value. Add the first monitoring point benchmark vibration state quantity to the benchmark vibration state quantity list, perform the same analysis as the first monitoring point for other monitoring points, and output after all analyses are completed. The benchmark vibration state of the equipment can be established to ensure that there is reliable reference data when performing state analysis and fault prediction. This helps to improve the accuracy and timeliness of equipment fault diagnosis.

[0093] Further, based on the device rotation control parameter, fuzzy constraint conditions are set, step 51 comprises the steps of:

[0094] S51: Construct distance evaluation function:

[0095] ;

[0096] DIS 2 = ∑ i=1 Q [ α i * x i1 - x i0 2 ] ;

[0097] in, Characterizes the first evaluation distance, Characterizes the second evaluation distance, The first eigenvalue characterizing the rotation control parameter of the i-th attribute device, The second eigenvalue characterizing the rotation control parameter of the i-th attribute device, Characterizes the rotation control parameter deviation threshold of the i-th attribute device, represents the weight adjustment parameter, Q represents the total number of device rotation control parameter attributes;

[0098] S52: configuring a first fuzzy distance threshold and a second fuzzy distance threshold, wherein the first fuzzy distance threshold and the second fuzzy distance threshold are in a logical AND relationship;

[0099] S53: constructing the fuzzy constraint condition for the device rotation control parameter according to the first fuzzy distance threshold and the second fuzzy distance threshold in combination with the distance evaluation function.

[0100] Specifically, the distance evaluation is performed according to the distance evaluation function to determine whether the first evaluation distance of the collected data can meet the first fuzzy distance threshold, and whether the second evaluation distance can meet the second fuzzy distance threshold. If both can, the fuzzy constraint condition is met, otherwise, the fuzzy constraint condition is not met.

[0101] Further, the state deviation matrix is ​​fitted by a fault prediction model to generate a fault prediction probability. Step S70 includes the following steps:

[0102] S71: traversing the monitoring point calibration results to identify the fault transfer probability, and generating a first monitoring point transfer probability matrix until an Nth monitoring point transfer probability matrix;

[0103] S72: training a first hidden Markov sub-model according to the first monitoring point transfer probability matrix, until training an Nth hidden Markov sub-model according to the Nth monitoring point transfer probability matrix;

[0104] S73: setting the first hidden Markov sub-model to the Nth hidden Markov sub-model as parallel independent nodes to generate the fault prediction model;

[0105] S74: Fitting the state deviation matrix according to the fault prediction model to generate the fault prediction probability.

[0106] Specifically, the fault transfer probability identification is performed by traversing the calibration results of the monitoring points, and the first monitoring point transfer probability matrix is ​​generated until the Nth monitoring point transfer probability matrix. The fault transfer probability identification refers to the identification of the fault transfer probability of each monitoring point. The transfer probability matrix refers to a matrix composed of the fault triggering probabilities under different state deviations. The first monitoring point transfer probability matrix is ​​used as a prediction reference standard to train the hidden Markov model. When the proportion of the deviation modulus of the predicted transfer probability and the identified transfer probability output for a preset number of consecutive times is less than or equal to the deviation modulus threshold is greater than the convergence proportion threshold, the first hidden Markov sub-model is generated. The second hidden Markov sub-model is trained in the same way until the Nth hidden Markov sub-model. The first hidden Markov sub-model to the Nth hidden Markov sub-model is set as a parallel independent node to generate the fault prediction model. The state deviation matrix is ​​fitted according to the fault prediction model to generate the fault prediction probability. The hidden Markov model can analyze the fault probability of the corresponding monitoring point based on the deviation data of a certain monitoring point. Further, since each monitoring point is associated with a corresponding component, the corresponding component is set as a fault component. The hidden Markov model is used to predict the failure probability of each vibration component. This can identify potential failure risks in advance, help technicians perform preventive maintenance, and improve the reliability and safety of equipment operation.

[0107] Further, the monitoring point calibration results are traversed to identify the fault transfer probability, and the first monitoring point transfer probability matrix is ​​generated until the Nth monitoring point transfer probability matrix. Step S71 includes the following steps:

[0108] S711: Obtain the first monitoring point of the monitoring point calibration result, and configure a state deviation vector set;

[0109] S712: traverse the state deviation vector set to collect running samples, obtain the fault sample ratio, and set it as the first monitoring point fault transfer probability set;

[0110] S713: Construct a first monitoring point transfer probability matrix according to the first monitoring point fault transfer probability set.

[0111] Specifically, the first monitoring point is any exemplary monitoring point of the monitoring point calibration result, and the state deviation vector set refers to a state deviation vector set composed of multiple groups of state deviation vectors of the pre-configured first monitoring point; the operation samples under each group of state deviation vectors of the state deviation vector set of the first monitoring point are collected, and the proportion of fault samples in each group of operation samples is analyzed, which is set as the first monitoring point fault transfer probability set; according to the first monitoring point fault transfer probability set, the first monitoring point transfer probability matrix is ​​constructed. The state deviation vector set configuration process is as follows: determine the first attribute state deviation vector interval to the Mth attribute state deviation vector interval, where M is the total number of attributes, and uniformly take values ​​of the first attribute state deviation vector interval to the Mth attribute state deviation vector interval in M-dimensional space according to the preset number to obtain the state deviation vector set. The fault transfer probability analyzed by big data is used as a reference for the evaluation of the hidden Markov model to ensure the objectivity and accuracy of the analysis results.

[0112] The fault diagnosis method for analyzing the rotation state of an equipment provided by the embodiment of the present invention has at least the following technical effects:

[0113] Based on the equipment rotation control parameters, combined with the rotation state analysis and monitoring point matching, the vibration element set and monitoring point calibration results of the rotating equipment to be monitored are automatically determined. Furthermore, based on the vibration element set and the monitoring point calibration results, the monitoring vibration state quantity list and the reference vibration state quantity list are analyzed, and then the state deviation matrix is ​​fitted through the fault prediction model to generate a technical solution for fault prediction probability. The automatic determination of the vibration element set and the monitoring point calibration results is achieved, which is conducive to improving the efficiency of model building and is suitable for mass industrial scenarios.

[0114] Embodiment 2:

[0115] like Figure 2 As shown, based on the same inventive concept of a fault diagnosis method for analyzing a rotation state of a device provided in Embodiment 1, an embodiment of the present invention further provides a fault diagnosis system for analyzing a rotation state of a device, comprising:

[0116] The equipment parameter acquisition module 1000 is used to obtain equipment rotation control parameters when the rotating equipment to be monitored is started;

[0117] A rotation state analysis module 2000, used to perform rotation state analysis according to the device rotation control parameters to obtain a vibration element set;

[0118] A monitoring point calibration module 3000, used to traverse the vibration element set to perform monitoring point matching and generate a monitoring point calibration result;

[0119] A vibration state monitoring module 4000 is used to activate the high-frequency vibration sensor of the monitoring point calibration result and collect a list of monitoring vibration state quantities;

[0120] A reference vibration state analysis module 5000 is used to perform frequency statistics on the calibration results of the monitoring points based on the equipment rotation control parameters and the model of the rotating equipment to be monitored, and obtain a list of reference vibration state quantities;

[0121] A state deviation analysis module 6000, configured to construct a state deviation matrix through the monitoring vibration state quantity list and the reference vibration state quantity list;

[0122] A fault probability prediction module 7000 is used to fit the state deviation matrix through a fault prediction model to generate a fault prediction probability, wherein the fault prediction probability corresponds to the vibration element set one by one;

[0123] The information sending module 8000 is used to send the fault prediction probability and the vibration element set to the fault diagnosis management end.

[0124] Furthermore, the rotation state analysis module 2000 executes the steps including:

[0125] Performing self-integration to zero splitting on the rotating equipment to be monitored to obtain a tree diagram of equipment components;

[0126] extracting a set of direct vibration elements according to the device rotation control parameters;

[0127] Traversing the direct vibration component set, sorting the vibration components based on the device component tree diagram, and obtaining an indirect vibration component set;

[0128] The direct vibration element set and the indirect vibration element set are added into the vibration element set.

[0129] Furthermore, the rotation state analysis module 2000 executes the steps including:

[0130] obtaining a first direct vibration element of the direct vibration element set;

[0131] Extracting the parent node element of the first direct vibration element from the device element tree diagram;

[0132] Performing a vibration probability analysis on the parent node element based on the device rotation control parameter to obtain a first vibration identification probability;

[0133] When the first vibration identification probability is greater than or equal to a vibration identification probability threshold, performing vibration element sorting based on the device element tree diagram according to the parent node element to obtain the indirect vibration element set;

[0134] When the first vibration identification probability is less than the vibration identification probability threshold, extracting a brother node element of the first direct vibration element from the device element tree diagram;

[0135] Performing a vibration probability analysis on the brother node element based on the device rotation control parameter to obtain a second vibration identification probability;

[0136] When the second vibration identification probability is greater than or equal to the vibration identification probability threshold, adding the brother node element into the first indirect vibration element set, and adding the first indirect vibration element set into the indirect vibration element set;

[0137] The step of performing vibration probability analysis on the parent node element based on the device rotation control parameter to obtain a first vibration identification probability includes:

[0138] Taking the device rotation control parameter as a constraint, collecting the parent node element test operation status of the rotating device to be monitored, wherein the parent node element test operation status has a first vibration amplitude test record value and a first vibration frequency test record value;

[0139] When the first vibration amplitude test record value is greater than or equal to the vibration amplitude threshold, and the first vibration frequency test record value is greater than or equal to the vibration frequency threshold, the first vibration identification probability is set to 1;

[0140] When the first vibration amplitude test record value is smaller than the vibration amplitude threshold, or / and the first vibration frequency test record value is smaller than the vibration frequency threshold, the first vibration identification probability is set to 0.

[0141] Furthermore, the monitoring point calibration module 3000 executes the following steps:

[0142] Obtaining a first vibration element of the vibration element set;

[0143] When the first vibration element belongs to a deployable monitoring point element, adding the first vibration element to the monitoring point calibration result;

[0144] When the first vibration element does not belong to a deployable monitoring point element, obtaining a set of associated vibration elements of the first vibration element, wherein the associated vibration elements are indirect vibration elements of the first vibration element;

[0145] Taking the device rotation control parameter as a constraint, traverse the associated vibration element set, and collect the associated element test operation state set of the rotating device to be monitored, wherein the associated element test operation state set includes a second vibration amplitude test record value set and a second vibration frequency test record value set;

[0146] Taking the device rotation control parameter as a constraint, collecting the first vibration element test operation state of the rotating device to be monitored, wherein the first vibration element test operation state includes a third vibration amplitude test record value and a third vibration frequency test record value;

[0147] Assigning a first weight to the vibration amplitude, constructing a vibration amplitude normalization weighting function according to the first weight, assigning a second weight to the vibration frequency, constructing a vibration frequency normalization weighting function according to the second weight, wherein the sum of the first weight and the second weight is equal to 1;

[0148] According to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function, traverse the second vibration amplitude test record value set and the second vibration frequency test record value set for fitting to generate a first coordinate set;

[0149] According to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function, the third vibration amplitude test record value and the third vibration frequency test record value are fitted to generate a second coordinate;

[0150] Traversing the first coordinate set, and calculating a set of Euclidean distance evaluation values ​​with the second coordinate;

[0151] Sorting the associated vibration element set from large to small according to the Euclidean distance evaluation value set to generate an associated vibration element sorting result;

[0152] The frontmost element of the deployable monitoring point element in the sorting result of the associated vibration elements is added into the monitoring point calibration result.

[0153] Furthermore, the benchmark vibration state analysis module 5000 performs the following steps:

[0154] The model of the rotating equipment to be monitored is set as a fixed constraint condition, and a fuzzy constraint condition is set based on the rotation control parameters of the equipment;

[0155] According to the fixed constraint condition and the fuzzy constraint condition, a fault-free sample is collected for a first monitoring point of the monitoring point calibration result to obtain a first vibration descriptor recorded state quantity set;

[0156] Performing extreme value analysis on the first vibration descriptor recorded state quantity set to obtain an extreme value recorded state quantity set;

[0157] The extreme value record state quantity set is deleted from the first vibration descriptor record state quantity set, and the mode value of the remaining record state quantity set of the first vibration descriptor is obtained, set as the first vibration descriptor reference state quantity, and added to the first monitoring point reference vibration state quantity;

[0158] The first monitoring point reference vibration state quantity is added to the reference vibration state quantity list.

[0159] Furthermore, the benchmark vibration state analysis module 5000 performs the following steps:

[0160] Construct a distance evaluation function:

[0161] ;

[0162] DIS 2 = ∑ i=1 Q [ α i * x i1 - x i0 2 ] ;

[0163] in, Characterizes the first evaluation distance, Characterizes the second evaluation distance, The first eigenvalue characterizing the rotation control parameter of the i-th attribute device, The second eigenvalue characterizing the rotation control parameter of the i-th attribute device, Characterizes the rotation control parameter deviation threshold of the i-th attribute device, represents the weight adjustment parameter, Q represents the total number of device rotation control parameter attributes;

[0164] Configure a first fuzzy distance threshold and a second fuzzy distance threshold, wherein the first fuzzy distance threshold and the second fuzzy distance threshold are in a logical AND relationship;

[0165] According to the first fuzzy distance threshold and the second fuzzy distance threshold, in combination with the distance evaluation function, the fuzzy constraint condition is constructed for the device rotation control parameter.

[0166] Furthermore, the failure probability prediction module 7000 performs the following steps:

[0167] Traversing the monitoring point calibration results to identify the fault transfer probability, generating a first monitoring point transfer probability matrix until an Nth monitoring point transfer probability matrix;

[0168] According to the first monitoring point transfer probability matrix, a first hidden Markov sub-model is trained, until according to the Nth monitoring point transfer probability matrix, an Nth hidden Markov sub-model is trained;

[0169] Setting the first hidden Markov sub-model to the Nth hidden Markov sub-model as parallel independent nodes to generate the fault prediction model;

[0170] The state deviation matrix is ​​fitted according to the fault prediction model to generate the fault prediction probability.

[0171] Furthermore, the failure probability prediction module 7000 performs the following steps:

[0172] Obtaining the first monitoring point of the monitoring point calibration result, and configuring a state deviation vector set;

[0173] Traversing the state deviation vector set to collect running samples, obtaining the fault sample ratio, and setting it as the first monitoring point fault transfer probability set;

[0174] A first monitoring point transfer probability matrix is ​​constructed according to the first monitoring point fault transfer probability set.

[0175] Embodiment three:

[0176] See also Figure 3 , Figure 3 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 3 As shown, this embodiment provides a computer-readable storage medium 600, on which a computer program 611 is stored. When the computer program 611 is executed by a processor, the following steps are implemented: when the rotating device to be monitored is started, the device rotation control parameters are obtained; according to the device rotation control parameters, a rotation state analysis is performed to obtain a vibration element set; the vibration element set is traversed to match the monitoring points and generate a monitoring point calibration result; the high-frequency vibration sensor of the monitoring point calibration result is activated to collect a monitoring vibration state quantity list; based on the device rotation control parameters and the model of the rotating device to be monitored, the monitoring point calibration results are traversed to perform frequency statistics to obtain a reference vibration state quantity list; a state deviation matrix is ​​constructed through the monitoring vibration state quantity list and the reference vibration state quantity list; the state deviation matrix is ​​fitted through a fault prediction model to generate a fault prediction probability, wherein the fault prediction probability corresponds to the vibration element set one by one; the fault prediction probability and the vibration element set are sent to the fault diagnosis management end.

[0177] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0178] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0179] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0180] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0181] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0182] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0183] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A fault diagnosis method based on equipment rotation state analysis, characterized in that: include: When the rotating equipment to be monitored is started, the equipment rotation control parameters are obtained; Performing rotation state analysis according to the rotation control parameters of the device to obtain a set of vibration elements; Traversing the vibration element set to perform monitoring point matching and generate monitoring point calibration results, including: obtaining a first vibration element of the vibration element set; When the first vibration element belongs to a deployable monitoring point element, adding the first vibration element to the monitoring point calibration result; When the first vibration element does not belong to a deployable monitoring point element, obtaining a set of associated vibration elements of the first vibration element, wherein the associated vibration elements are indirect vibration elements of the first vibration element; Taking the device rotation control parameter as a constraint, traverse the associated vibration element set, and collect the associated element test operation state set of the rotating device to be monitored, wherein the associated element test operation state set includes a second vibration amplitude test record value set and a second vibration frequency test record value set; Taking the device rotation control parameter as a constraint, collecting the first vibration element test operation state of the rotating device to be monitored, wherein the first vibration element test operation state includes a third vibration amplitude test record value and a third vibration frequency test record value; Assigning a first weight to the vibration amplitude, constructing a vibration amplitude normalization weighting function according to the first weight, assigning a second weight to the vibration frequency, constructing a vibration frequency normalization weighting function according to the second weight, wherein the sum of the first weight and the second weight is equal to 1; According to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function, traverse the second vibration amplitude test record value set and the second vibration frequency test record value set for fitting to generate a first coordinate set; According to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function, the third vibration amplitude test record value and the third vibration frequency test record value are fitted to generate a second coordinate; Traversing the first coordinate set, and calculating a set of Euclidean distance evaluation values ​​with the second coordinate; Sorting the associated vibration element set from large to small according to the Euclidean distance evaluation value set to generate an associated vibration element sorting result; The first element of the deployable monitoring point element in the sorting result of the associated vibration elements is added to the monitoring point calibration result, wherein the monitoring point calibration result is a specific position for monitoring the operating status of the equipment; Activate the high-frequency vibration sensor of the monitoring point calibration result and collect a list of monitoring vibration state quantities; Based on the equipment rotation control parameters and the model of the rotating equipment to be monitored, the calibration results of the monitoring points are traversed to perform frequency statistics to obtain a list of reference vibration state quantities; Constructing a state deviation matrix through the monitoring vibration state quantity list and the reference vibration state quantity list; Performing fault probability prediction on the state deviation matrix through a fault prediction model to generate a fault prediction probability, wherein the fault prediction probability corresponds to the vibration element set in one-to-one correspondence; The fault prediction probability and the vibration element set are sent to a fault diagnosis management terminal.

2. The method according to claim 1, characterized in that A rotation state analysis is performed according to the rotation control parameters of the device to obtain a vibration element set, including: Performing self-integration to zero splitting on the rotating equipment to be monitored to obtain a tree diagram of equipment components; extracting a set of direct vibration elements according to the device rotation control parameters; Traversing the direct vibration component set, sorting the vibration components based on the device component tree diagram, and obtaining an indirect vibration component set; The direct vibration element set and the indirect vibration element set are added into the vibration element set.

3. The method according to claim 2, characterized in that Traversing the direct vibration component set, sorting the vibration components based on the device component tree diagram, and obtaining the indirect vibration component set, including: obtaining a first direct vibration element of the direct vibration element set; Extracting the parent node element of the first direct vibration element from the device element tree diagram; Performing a vibration probability analysis on the parent node element based on the device rotation control parameter to obtain a first vibration identification probability; When the first vibration identification probability is greater than or equal to a vibration identification probability threshold, performing vibration element sorting based on the device element tree diagram according to the parent node element to obtain the indirect vibration element set; When the first vibration identification probability is less than the vibration identification probability threshold, extracting a brother node element of the first direct vibration element from the device element tree diagram; Performing a vibration probability analysis on the brother node element based on the device rotation control parameter to obtain a second vibration identification probability; When the second vibration identification probability is greater than or equal to the vibration identification probability threshold, adding the brother node element into the first indirect vibration element set, and adding the first indirect vibration element set into the indirect vibration element set; The step of performing vibration probability analysis on the parent node element based on the device rotation control parameter to obtain a first vibration identification probability includes: Taking the device rotation control parameter as a constraint, collecting the parent node element test operation status of the rotating device to be monitored, wherein the parent node element test operation status has a first vibration amplitude test record value and a first vibration frequency test record value; When the first vibration amplitude test record value is greater than or equal to the vibration amplitude threshold, and the first vibration frequency test record value is greater than or equal to the vibration frequency threshold, the first vibration identification probability is set to 1; When the first vibration amplitude test record value is smaller than the vibration amplitude threshold, or / and the first vibration frequency test record value is smaller than the vibration frequency threshold, the first vibration identification probability is set to 0.

4. The method according to claim 1, characterized in that Based on the equipment rotation control parameters and the model of the rotating equipment to be monitored, the monitoring point calibration results are traversed to perform frequency statistics to obtain a list of reference vibration state quantities, including: The model of the rotating equipment to be monitored is set as a fixed constraint condition, and a fuzzy constraint condition is set based on the rotation control parameters of the equipment; According to the fixed constraint condition and the fuzzy constraint condition, a fault-free sample is collected for a first monitoring point of the monitoring point calibration result to obtain a first vibration descriptor recorded state quantity set; Performing extreme value analysis on the first vibration descriptor recorded state quantity set to obtain an extreme value recorded state quantity set; The extreme value record state quantity set is deleted from the first vibration descriptor record state quantity set, and the mode value of the remaining record state quantity set of the first vibration descriptor is obtained, set as the first vibration descriptor reference state quantity, and added to the first monitoring point reference vibration state quantity; The first monitoring point reference vibration state quantity is added to the reference vibration state quantity list.

5. The method according to claim 4, characterized in that The fuzzy constraint conditions are set based on the device rotation control parameters, including: Construct a distance evaluation function: ； ； in, Characterizes the first evaluation distance, Characterizes the second evaluation distance, The first eigenvalue characterizing the rotation control parameter of the i-th attribute device, The second eigenvalue characterizing the rotation control parameter of the i-th attribute device, Characterizes the rotation control parameter deviation threshold of the i-th attribute device, represents the weight adjustment parameter, Q represents the total number of device rotation control parameter attributes; Configure a first fuzzy distance threshold and a second fuzzy distance threshold, wherein the first fuzzy distance threshold and the second fuzzy distance threshold are in a logical AND relationship; According to the first fuzzy distance threshold and the second fuzzy distance threshold, in combination with the distance evaluation function, the fuzzy constraint condition is constructed for the device rotation control parameter.

6. The method according to claim 1, characterized in that The state deviation matrix is ​​fitted by a fault prediction model to generate a fault prediction probability, including: Traversing the monitoring point calibration results to identify the fault transfer probability, generating a first monitoring point transfer probability matrix until an Nth monitoring point transfer probability matrix; According to the first monitoring point transfer probability matrix, a first hidden Markov sub-model is trained, until according to the Nth monitoring point transfer probability matrix, an Nth hidden Markov sub-model is trained; Setting the first hidden Markov sub-model to the Nth hidden Markov sub-model as parallel independent nodes to generate the fault prediction model; The fault prediction model is used to predict the fault probability of the state deviation matrix to generate the fault prediction probability.

7. The method according to claim 6, characterized in that Traversing the monitoring point calibration results to identify the fault transfer probability, generating the first monitoring point transfer probability matrix until the Nth monitoring point transfer probability matrix, including: Obtaining the first monitoring point of the monitoring point calibration result, and configuring a state deviation vector set; Traversing the state deviation vector set to collect running samples, obtaining the fault sample ratio, and setting it as the first monitoring point fault transfer probability set; A first monitoring point transfer probability matrix is ​​constructed according to the first monitoring point fault transfer probability set.

8. A fault diagnosis system based on equipment rotation state analysis, characterized in that: include: The equipment parameter acquisition module is used to obtain the equipment rotation control parameters when the rotating equipment to be monitored is started; A rotation state analysis module, used to perform rotation state analysis according to the rotation control parameters of the device to obtain a vibration element set; The monitoring point calibration module is used to traverse the vibration element set to perform monitoring point matching and generate monitoring point calibration results, including: obtaining a first vibration element of the vibration element set; When the first vibration element belongs to a deployable monitoring point element, adding the first vibration element to the monitoring point calibration result; When the first vibration element does not belong to a deployable monitoring point element, obtaining a set of associated vibration elements of the first vibration element, wherein the associated vibration elements are indirect vibration elements of the first vibration element; Taking the device rotation control parameter as a constraint, traverse the associated vibration element set, and collect the associated element test operation state set of the rotating device to be monitored, wherein the associated element test operation state set includes a second vibration amplitude test record value set and a second vibration frequency test record value set; Taking the device rotation control parameter as a constraint, collecting the first vibration element test operation state of the rotating device to be monitored, wherein the first vibration element test operation state includes a third vibration amplitude test record value and a third vibration frequency test record value; Assigning a first weight to the vibration amplitude, constructing a vibration amplitude normalization weighting function according to the first weight, assigning a second weight to the vibration frequency, constructing a vibration frequency normalization weighting function according to the second weight, wherein the sum of the first weight and the second weight is equal to 1; According to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function, traverse the second vibration amplitude test record value set and the second vibration frequency test record value set for fitting to generate a first coordinate set; According to the vibration amplitude normalization weighting function and the vibration frequency normalization weighting function, the third vibration amplitude test record value and the third vibration frequency test record value are fitted to generate a second coordinate; Traversing the first coordinate set, and calculating a set of Euclidean distance evaluation values ​​with the second coordinate; Sorting the associated vibration element set from large to small according to the Euclidean distance evaluation value set to generate an associated vibration element sorting result; The first element of the deployable monitoring point element in the sorting result of the associated vibration elements is added to the monitoring point calibration result, wherein the monitoring point calibration result is a specific position for monitoring the operating status of the equipment; A vibration state monitoring module, used to activate the high-frequency vibration sensor of the monitoring point calibration result and collect a list of monitoring vibration state quantities; A reference vibration state analysis module is used to perform frequency statistics on the calibration results of the monitoring points based on the equipment rotation control parameters and the model of the rotating equipment to be monitored, and obtain a list of reference vibration state quantities; A state deviation analysis module, used to construct a state deviation matrix through the monitoring vibration state quantity list and the reference vibration state quantity list; A fault probability prediction module, used to perform fault probability prediction on the state deviation matrix through a fault prediction model to generate a fault prediction probability, wherein the fault prediction probability corresponds to the vibration element set in a one-to-one manner; An information sending module is used to send the fault prediction probability and the vibration element set to a fault diagnosis management terminal.

9. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, it implements a fault diagnosis method based on equipment rotation state analysis as described in any one of claims 1 to 7.

Images