CN111255674B - System and method for detecting state of rotating mechanical equipment - Google Patents

Abstract

Embodiments of the present invention provide a system and a detection method for detecting the state of a rotating mechanical device. The system includes: system hardware and system software; wherein: the system hardware includes a signal acquisition terminal, a signal transmission device and a data processing terminal; the signal acquisition terminal is used to collect the state data of the rotating mechanical equipment, and the signal transmission device uses In order to transmit the state data to the data processing terminal, the data processing terminal is used to analyze and obtain the state result of the rotating machinery equipment according to the state data; the system software includes a database, a background intelligent diagnosis model, data Analysis function, software interface and software interface module. The embodiment of the present invention can complete the modeling of the system and further fault diagnosis and early warning by relying on the collected vibration information by separating from the physical structure of the mechanical equipment.

Description

A system and detection method for detecting the state of rotating mechanical equipment

technical field

The invention relates to the technical field of fault detection, and in particular, to a system and a detection method for detecting the state of rotating machinery equipment.

Background technique

In a nuclear power plant, the pump is the main rotating mechanical equipment of the nuclear power plant and the heart of the nuclear power plant. The nuclear power plant requires the main pump to have a long life, long-term continuous and safe operation, good sealing performance, and no nuclear leakage. The strength of the noise and its main frequency components are closely related to the type, degree, location and cause of the fault, so it has become an inevitable development trend in the industry to provide effective fault diagnosis and early warning based on the vibration information of the pump.

In general, equipment status early warning based on vibration signals and identification of fault causes based on monitoring information often require technicians to have high professional level and on-site diagnosis experience, which limits the application of fault diagnosis technology, so there are the following problems: fault diagnosis Difficulty, only for the fault itself, relying on expert experience and theoretical models, insufficient early warning ability for new faults, single fault diagnosis rules, relying solely on threshold early warning.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a system and a detection method for detecting the state of a rotating mechanical equipment, which are used to solve the defects of the prior art that the diagnosis form is relatively simple and relies too much on the experience level of technicians.

In a first aspect, an embodiment of the present invention provides a system for detecting a state of a rotating mechanical device, including:

System hardware and system software; where:

The system hardware includes a signal acquisition terminal, a signal transmission device and a data processing terminal;

The signal acquisition terminal is used to collect the state data of the rotating mechanical equipment, the signal transmission device is used to transmit the state data to the data processing terminal, and the data processing terminal is used to analyze and obtain the data according to the state data. State results of the rotating machinery and equipment;

The system software includes a database, a background intelligent diagnosis model, a data analysis function, a software interface and a software interface.

Preferably, the system software adopts a three-layer architecture, which specifically includes: a data layer, a middle layer and an application layer; wherein:

The data layer includes equipment collection information, system basic information data, and system model algorithms. The equipment collection information is used to input equipment collection data into the software system and read offline data. The system basic information data includes equipment information, personnel information, parameter information and file information, the system model algorithm is used to write and encapsulate the data collected by the device using a preset programming language;

The middle layer is used to encapsulate the back-end method provided by the application server into a Web method, transmit data through the HTTP protocol, and provide a unified interface for the client;

The application layer is used to provide a human-computer interaction interface, and the administrator is responsible for providing several types of software use authorization to other users.

In a second aspect, an embodiment of the present invention provides a detection method for detecting a state of a rotating mechanical device, including:

Obtain real-time acquisition data of rotating machinery and equipment;

Calling the pre-trained parameters of the equipment state monitoring model to obtain the equipment state expectation value and the equipment state index output by the equipment state monitoring model;

The equipment state expectation value and the equipment state index are input into the equipment state early warning processing and fault diagnosis process, and the fault early warning information or fault diagnosis information of the rotating mechanical equipment is obtained.

Preferably, before acquiring the real-time collection data of the rotating mechanical equipment, the method further includes: constructing a fault diagnosis knowledge base.

Preferably, the construction of the fault diagnosis knowledge base specifically includes:

Get historical data;

Combining the fuzzy correlation analysis method and the gray correlation analysis method, the data with the preset correlation degree is mined from the historical data, and the vibration state correlation process parameters are obtained;

According to expert knowledge, based on the abnormal vibration fluctuation situation and the normal fluctuation situation, output the definition of abnormal vibration, obtain the cause of abnormal vibration fluctuation through correlation analysis, and output the relationship between the process parameters related to the vibration state and the definition of abnormal vibration;

Based on the state information of the key nodes in the historical data and the association relationship, a data-driven artificial intelligence algorithm is used to train a hierarchical model for quantifying the health state of the system;

Based on the association relationship and the hierarchical model, the fault diagnosis knowledge base is obtained.

Preferably, the equipment status early warning processing and fault diagnosis process specifically include:

Analyze the vibration state of the rotating mechanical equipment from several preset domains to obtain a primary vibration parameter set;

Based on the primary vibration parameter set, a preset optimization algorithm is used to obtain the selected feature parameter set;

According to the selected feature parameter set, adopt the qualitative trend analysis method to obtain the vibration signal of the rotating machinery and the variation trend of the process parameters;

Through the layered model, based on the vibration signal and the change trend of the process parameter, combined with the fault diagnosis knowledge base, the fault trend of the rotating machinery is predicted.

Preferably, the equipment status early warning processing and fault diagnosis process further includes:

Collect process parameters and current vibration parameters of the rotating machinery;

Through the hierarchical model, based on the process parameters and the current vibration parameters, the fault information of the rotating mechanical equipment is obtained.

Preferably, the rotating machinery includes a pump.

The system and detection method for detecting the state of rotating mechanical equipment provided by the embodiments of the present invention can complete the modeling of the system and further fault diagnosis and early warning by relying on the collected vibration information by separating from the physical structure of the mechanical equipment.

Description of drawings

In order to illustrate the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are For some embodiments of the present invention, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is an overall structural diagram of a system provided by an embodiment of the present invention;

2 is a system software architecture diagram provided by an embodiment of the present invention;

3 is a flowchart of a detection method for detecting the state of a rotating mechanical equipment provided by an embodiment of the present invention;

4 is a flow chart of a system operation provided by an embodiment of the present invention;

FIG. 5 is a flowchart of the construction of a data-driven artificial intelligence fault diagnosis knowledge base provided by an embodiment of the present invention;

6 is a flowchart of a method for early warning of a vibration state of a rotating mechanical equipment provided by an embodiment of the present invention;

FIG. 7 is a flowchart of a method for diagnosing faults of rotating mechanical equipment provided by an embodiment of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In order to solve the problems in the prior art, the embodiments of the present invention provide a vibration state information management and intelligent diagnosis system and method for rotating machinery, so as to know the running status of the rotating machinery in time and ensure its safe and reliable operation. Realize early warning, analysis and diagnosis of equipment deterioration or abnormal accidents, and improve the safety, reliability and economy of equipment operation.

FIG. 1 is an overall structural diagram of a system provided by an embodiment of the present invention, as shown in FIG. 1 , including:

System hardware and system software; where:

The system hardware includes a signal acquisition terminal, a signal transmission device and a data processing terminal;

The signal acquisition terminal is used to collect the state data of the rotating mechanical equipment, the signal transmission device is used to transmit the state data to the data processing terminal, and the data processing terminal is used to analyze and obtain the data according to the state data. State results of the rotating machinery and equipment;

The system software includes a database, a background intelligent diagnosis model, a data analysis function, a software interface and a software interface.

Specifically, the early warning and intelligent diagnosis system for diagnosing the vibration state of rotating machinery equipment includes two parts: system hardware and system software. The system hardware includes a signal acquisition terminal (sensor), a signal transmission device, and a data processing terminal. The signal acquisition terminal is used to collect the state data of the rotating mechanical equipment, and the signal transmission device is used to transmit the state data to the data processing terminal. It is used to analyze the status results of rotating machinery equipment according to the status data; the system software is mainly composed of modules such as database, background intelligent diagnosis model, data analysis function, software interface, and software interface.

The embodiment of the present invention can complete the modeling of the system and further fault diagnosis and early warning by relying on the collected vibration information by separating from the physical structure of the mechanical equipment.

Based on the above embodiment, the system software adopts a three-layer architecture, which specifically includes: a data layer, a middle layer and an application layer; wherein:

The data layer includes equipment collection information, system basic information data, and system model algorithms. The equipment collection information is used to input equipment collection data into the software system and read offline data. The system basic information data includes equipment information, personnel information, parameter information and file information, the system model algorithm is used to write and encapsulate the data collected by the device using a preset programming language;

The middle layer is used to encapsulate the back-end method provided by the application server into a Web method, transmit data through the HTTP protocol, and provide a unified interface for the client;

The application layer is used to provide a human-computer interaction interface, and the administrator is responsible for providing several types of software use authorization to other users.

Specifically, as shown in Figure 2, the software system adopts the three-layer structure of data layer, middle layer and application layer to improve the stability, security, scalability and adaptability of the system, wherein:

1) Data layer

The data layer includes equipment acquisition information, system basic information data and system model algorithm. The first is to directly collect data from the equipment and input the data into the software system, and the other is to read the offline data; the basic information data of the system includes equipment information, personnel information, parameters, files and other information, which are imported and maintained by the administrator; the software reserves the background algorithm Interface, the interface supports Python and MATLAB packages, and the specific reserved interface can be modified according to user requirements. The background intelligent diagnosis model decides whether to rewrite and encapsulate it in C++ language according to the running speed; the data can be saved to the system database server as needed, or directly provided to the application layer;

2) Middle layer

The middle layer encapsulates the back-end method provided by the application server as a Web method, transmits data through the HTTP protocol, and provides a unified interface for the client. It encapsulates the operation process in the middle, and also includes some transaction processing processes. The interface service layer dispatches the functions provided by the middle layer according to the response of the front-end interface. This layer is responsible for the realization of intermediate functions and the calculation of intermediate data, and realizes intermediate functions such as data input and output, intelligent diagnosis model algorithm, data mining, database management, and data visualization;

3) Application layer

The application layer is the interface between the user and the system for human-computer interaction. The system is equipped with an administrator, who is responsible for providing different types of software authorization to other users, and assigning different role permissions to different user groups. Hierarchical display, visualization of equipment running status, good human-computer interaction interface, simple operation.

Most of the operations provided by the embodiments of the present invention can be performed directly on the software interface, the operations are simple, the human-computer interaction is good, and the workload is small.

FIG. 3 is a flowchart of a detection method for detecting the state of a rotating mechanical equipment provided by an embodiment of the present invention, as shown in FIG. 3 , including:

S1, obtain real-time collection data of rotating mechanical equipment;

S2, call the pre-trained equipment state monitoring model parameters to obtain the equipment state expected value and equipment state index output by the equipment state monitoring model;

S3: Input the equipment status expectation value and the equipment status index into the equipment status early warning processing and fault diagnosis process to obtain fault early warning information or fault diagnosis information of the rotating mechanical equipment.

Specifically, the overall workflow of the system is shown in Figure 4. The system first performs real-time data input (collection) of equipment monitoring parameters, displays real-time vibration monitoring and operating status, and then enters the data accuracy screening process. If abnormal measurement data is detected Then perform reconstruction and alarm, and then call the parameters of the equipment state monitoring model after successful training to calculate the expected value of the equipment state and the calculation of the equipment state index, and finally enter the equipment state early warning processing process. The equipment status early warning processing process specifically includes equipment status monitoring and early warning, parameter trend analysis, calling the failure mode library to automatically generate fault diagnosis sheets and equipment status monitoring reports, and realizing closed-loop management of equipment status monitoring. Equipment diagnosis specialists, equipment inspection or equipment maintenance personnel shall complete and record the relevant information of the fault diagnosis sheet. The equipment status monitoring report is convenient for production management personnel, equipment inspection or maintenance personnel to inquire about the overall health status of the equipment, existing hidden dangers or potential failures. After an early warning occurs in the equipment, equipment maintenance personnel, specialists, etc. conduct further analysis and diagnosis based on the suspected fault information of the system, report to the leader, and generate a fault diagnosis sheet. Deviation parameters and related attributes and other information, the equipment maintenance personnel will eliminate defects or troubleshoot according to the diagnosis sheet, the equipment responsible person will check the maintenance quality, and fill in the fault handling result feedback, and the diagnostic process record can be quickly added to the fault mode. In the library, the standard diagnostic experience is formed. If there are unresolved problems, it can be included in the maintenance plan in the remarks in the diagnosis sheet, or it can be transferred to the next fault diagnosis cycle.

The embodiment of the present invention adopts a data-driven artificial intelligence fault diagnosis method, which can perform fault analysis and diagnosis only according to the data collected by sensors, overcomes the limitations of traditional fault diagnosis methods based on analytical models and qualitative models, and no longer The physical structure of the mechanical equipment needs to be considered.

Based on any of the above embodiments, FIG. 5 is a flowchart of the construction of a data-driven artificial intelligence fault diagnosis knowledge base provided by an embodiment of the present invention, as shown in FIG. 5 , including:

101. Obtain historical data;

102. Combine the fuzzy correlation analysis method and the grey correlation analysis method, excavate data with a preset correlation degree from the historical data, and obtain vibration state correlation process parameters;

103. According to expert knowledge, based on the abnormal vibration fluctuation situation and the normal fluctuation situation, output the definition of abnormal vibration, obtain the cause of abnormal vibration fluctuation through correlation analysis, and output the correlation between the process parameter associated with the vibration state and the definition of abnormal vibration;

104. Based on the state information of the key nodes in the historical data and the association relationship, adopt a data-driven artificial intelligence algorithm to train to obtain a hierarchical model for quantifying the health state of the system;

105. Obtain the fault diagnosis knowledge base based on the association relationship and the hierarchical model.

Specifically, the fault diagnosis knowledge base is the basis for the early warning and fault diagnosis of the vibration state of the rotating machinery. After the real-time data collected by the sensor is processed for anti-interference and preliminary calculation, it is compared with the data in the knowledge base. After analysis, the vibration state early warning and fault diagnosis of the headquarters can be realized.

In step 101, the first step is to obtain a large amount of historical data;

In step 102, combine the fuzzy correlation analysis method (FRA method) and the grey correlation analysis method (GRA method) to mine data with a relatively high degree of correlation from the multi-source data, and obtain process parameters closely related to the vibration state (the process parameters include equipment. voltage, current, pressure, temperature, motor current, motor coil temperature, bearing temperature, vibration value, inlet and outlet medium temperature and flow, etc.). Through the FRA method, the fuzzy information is processed, and the objective interval value is obtained from the multi-source data; through the GRA method, the whitening information is processed, and the gray correlation degree order among the multiple factors is obtained;

In step 103, according to the expert knowledge, a reasonable definition of abnormal vibration is given for abnormal vibration fluctuation and normal fluctuation, and through correlation analysis, the cause of abnormal vibration fluctuation is found, and the relationship between process parameter change and vibration state is summarized. relationship between;

In step 104, according to the state information of key nodes in the system historical data and the results of step 103, the multi-source information fusion technology and Bayesian network are integrated, and an improved dynamic Bayesian network method is proposed to solve the ambiguity of the information provided by the system. Based on the improved dynamic Bayesian network, the failure probability of different nodes in the system is obtained. Specifically, the multi-source information fusion technology is combined with the Bayesian network to form an improved dynamic Bayesian network method. Specifically, the multiple fault features of the extracted vibration signal are regarded as multi-source information from multiple different sensors, each fault feature is regarded as from a sensor, and the multiple fault features are denoised and fused by using the multi-fusion technology. Obtain relatively clear and accurate fault characteristics; then, for different nodes of the system, use dynamic Bayesian network to analyze faults, and count the probability of occurrence of fault characteristics; split the components of the pump itself and the auxiliary systems related to the pump to train the health of the system A hierarchical model of state quantification, that is, a fault diagnosis model;

In step 105, based on the results of steps 103 and 104, a fault diagnosis knowledge base is formed, and historical data is dynamically updated according to the newly stored data.

The embodiment of the present invention integrates multi-source information fusion technology and Bayesian network, proposes an improved dynamic Bayesian network method, solves the problem of ambiguity and incompleteness of information provided by the system, and effectively solves the dependence of fusion algorithm on prior information According to the prior data of the pump vibration state system, an appropriate robust and accurate algorithm can be analyzed and selected, and the data mining technology, combined with the fuzzy correlation analysis method and the gray correlation analysis method, can be used by combining the actual results of data mining with prediction. The values are systematically compared, and the existing deficiencies are found and improved to make the model more perfect.

Based on any of the above embodiments, FIG. 6 is a flowchart of a method for early warning of a vibration state of a rotating machinery equipment provided by an embodiment of the present invention. As shown in FIG. 6 , the process of early warning processing and fault diagnosis of the equipment state specifically includes:

201. Analyze the vibration state of the rotating mechanical equipment from several preset domains to obtain a primary vibration parameter set;

202. Based on the primary vibration parameter set, a preset optimization algorithm is used to obtain a selected feature parameter set;

203. According to the selected feature parameter set, adopt a qualitative trend analysis method to obtain the vibration signal and process parameter variation trend of the rotating mechanical equipment;

204. Using the hierarchical model, based on the vibration signal and the change trend of the process parameter, and in combination with the fault diagnosis knowledge base, predict the fault trend of the rotating mechanical equipment.

Specifically, in step 201, extracting the characteristic parameter set of the pump according to the vibration data of the pump is the first step in the state early warning and fault diagnosis. The performance of the pump can be confirmed by judging whether the characteristic parameter deviates from the normal value range. In this design, the vibration state of rotating machinery will be analyzed from multiple angles in time domain, frequency domain and time-frequency domain, and a primary vibration parameter set will be obtained. Through time domain analysis, the characteristics of signal mean, mean square error, kurtosis value, peak value and other characteristic parameters and related calculated values at different times are revealed. Because mechanical equipment failures are not only reflected in time domain features, but also cause frequency domain feature changes. When the vibration signal is analyzed in the time domain, the time domain characteristics of some signals are similar, but it does not mean that the signals belong to the same working condition, and it is only suitable for linear systems, and the frequency domain analysis of the vibration signal is also required. Through Fourier transform, the dynamic signal is transformed from the time domain to the frequency domain, and the complex vibration signal is decomposed into a known sine wave signal for analysis, reflecting the relationship between the signal structure and the frequency signal amplitude in the frequency domain. However, mechanical equipment vibration signals are often non-stationary signals, and their working conditions may change at any time. The object of Fourier transform is the entire signal, which leads to some local features of the vibration signal may be lost. Different vibration signal faults, the frequency domain analysis based on Fourier transform lacks adaptive ability. Therefore, it is necessary to combine the advantages of time domain analysis and frequency domain analysis to select an appropriate wavelet transform function, and extract vibration signal features from time domain and frequency domain at the same time, so as to extract signal features more comprehensively. However, wavelet analysis needs to select a suitable wavelet basis function in advance, and cannot adaptively transform wavelets according to the characteristics of vibration signals, and different wavelets have different sensitivity to noise, and some wavelet basis functions do not have good anti-noise ability. Therefore, it is necessary to extract and analyze the vibration state of rotating machinery from multiple angles of time domain, frequency domain and time-frequency domain, so as to understand the equipment state more comprehensively;

In step 202, since the sensitivity evaluation of the feature parameters is used to improve the sensitivity of the feature parameters in the feature set to be selected to the fault, the feature parameter set with high sensitivity is selected from the feature parameter set extracted in step 201, and the The genetic algorithm is an optimization algorithm, selects the characteristic parameters, and obtains the selected characteristic parameter set;

In step 203, according to the selected characteristic parameter set, referring to the change characteristic trend of the historical data of the characteristic parameter, a qualitative trend analysis is performed on the selected characteristic parameter, the vibration signal of the pump and the change trend of the process parameter are obtained, and the future operation state of the pump is obtained, Compare with the fault status in the fault diagnosis knowledge base to determine whether the pump has a fault trend;

In step 204, according to the current operating state and the future operating state of the pump, combined with the hierarchical model of the system, reverse derivation is performed to predict the failure probability of different nodes of the system.

The embodiment of the present invention can judge whether there is a potential fault by diagnosing the existing fault and comparing the actual signal with the ideal signal. Multi-level analysis and processing of the vibration information of the pump further enhances the fault diagnosis ability; it has a certain machine self-learning ability, that is, for a given new vibration abnormality and fault type, the model passes one month of self-learning of the original data, Vibration status warning and intelligent diagnosis accuracy rate is over 95%.

Based on any of the above embodiments, FIG. 7 is a flowchart of a fault diagnosis method for rotating machinery equipment provided by an embodiment of the present invention. As shown in FIG. 7 , the equipment status early warning processing and fault diagnosis process further includes:

301. Collect process parameters and current vibration parameters of the rotating mechanical equipment;

302. Obtain fault information of the rotating mechanical equipment based on the process parameter and the current vibration parameter through the layered model.

Specifically, in step 301, when judging an actual fault, what is different from the fault early warning process is that the process parameters and current vibration parameters of the pump are collected;

In step 302, based on the process parameters and current vibration parameters obtained in step 301, the fault information of the pump is obtained through the layered model.

By virtue of the machine self-learning function, the embodiments of the present invention have strong adaptability to new types of faults, provide a strong guarantee for correct fault diagnosis, and have strong adaptability to new types of faults.

The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (3)

1. a detection method for detecting the state of rotating machinery equipment, is characterized in that, comprises: Obtain real-time collection data of rotating machinery and equipment; Call the parameters of the pre-trained equipment status monitoring model to obtain the equipment status expected value and equipment status index output by the equipment status monitoring model; Input the equipment status expectation value and the equipment status index into the equipment status early warning processing and fault diagnosis process, and obtain the fault early warning information or fault diagnosis information of the rotating machinery equipment; The obtaining of the real-time collection data of the rotating mechanical equipment further includes: constructing a fault diagnosis knowledge base; The building of the fault diagnosis knowledge base specifically includes: Get historical data; Combining the fuzzy correlation analysis method and the grey correlation analysis method, the data with the preset correlation degree is mined from the historical data, and the vibration state correlation process parameters are obtained; According to expert knowledge, based on the abnormal vibration fluctuation situation and the normal fluctuation situation, output the definition of abnormal vibration, obtain the cause of abnormal vibration fluctuation through correlation analysis, and output the relationship between the process parameters related to the vibration state and the definition of abnormal vibration; Based on the state information of the key nodes in the historical data and the association relationship, a data-driven artificial intelligence algorithm is used to train a hierarchical model for quantifying the health state of the system; obtaining the fault diagnosis knowledge base based on the association relationship and the hierarchical model; The equipment status early warning processing and fault diagnosis process specifically includes: Analyze the vibration state of the rotating mechanical equipment from several preset domains to obtain a primary vibration parameter set; Based on the primary vibration parameter set, a preset optimization algorithm is used to obtain the selected feature parameter set; According to the selected feature parameter set, adopt the qualitative trend analysis method to obtain the vibration signal of the rotating machinery and the variation trend of the process parameters; Through the layered model, based on the vibration signal and the change trend of the process parameter, combined with the fault diagnosis knowledge base, the fault trend of the rotating mechanical equipment is predicted. 2. The detection method for detecting the state of rotating machinery equipment according to claim 1, wherein the equipment state early warning processing and fault diagnosis process further comprises: Collect process parameters and current vibration parameters of the rotating machinery; Through the hierarchical model, based on the process parameters and the current vibration parameters, the fault information of the rotating mechanical equipment is obtained. 3. The detection method for detecting the state of a rotating mechanical device according to any one of claims 1 to 2, wherein the rotating mechanical device comprises a pump.

Images