CN115876507A - Fault diagnosis system based on converter valve cooling system - Google Patents
Fault diagnosis system based on converter valve cooling system Download PDFInfo
- Publication number
- CN115876507A CN115876507A CN202211429741.2A CN202211429741A CN115876507A CN 115876507 A CN115876507 A CN 115876507A CN 202211429741 A CN202211429741 A CN 202211429741A CN 115876507 A CN115876507 A CN 115876507A
- Authority
- CN
- China
- Prior art keywords
- frequency
- fault
- data
- service subsystem
- diagnosis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a fault diagnosis system based on a converter valve cooling system, which relates to the technical field of equipment diagnosis and comprises acquisition hardware, a data acquisition service subsystem, a data processing service subsystem, a data diagnosis service subsystem, a Web service subsystem and a database system.
Description
Technical Field
The invention relates to the technical field of fault diagnosis systems, in particular to a fault diagnosis system based on a converter valve cooling system.
Background
At present, direct current transmission projects in China are increasingly built, the proportion of direct current transmission capacity in the whole power grid is also increasingly large, a converter valve plays a vital role as a core device of the direct current project, and a converter valve cooling system (hereinafter referred to as a valve cooling system) is a core auxiliary device of the converter valve, so that stable and reliable operation is a necessary condition for normal operation of the converter valve. The dynamic equipment (comprising a main circulating pump, a water replenishing pump, a spray pump, an air cooler fan, a cooling tower fan, a high-pressure pump, a side filter pump, a sewage pump and the like) is used as a power source of the valve cooling system, and the fault of the dynamic equipment is the primary reason for causing the fault of the valve cooling system.
At present, the fault diagnosis for the movable equipment of the valve cooling system only stays at a state monitoring level, namely, vibration speed sensors are arranged at the bearing positions of the movable equipment to obtain vibration data of measuring points, then the vibration threshold value is set to judge whether the movable equipment has faults or not, and the severity of the faults is judged according to the vibration value of the movable equipment. According to the method, when the equipment fails, the specific part of the equipment which fails is needed to be disassembled in a shutdown state by means of manual analysis, the failure reason is given by means of manual experience, and a maintenance suggestion is proposed. Therefore, the fault diagnosis method needs to be optimized and improved according to the current condition.
Disclosure of Invention
The invention aims to provide a fault diagnosis system based on a converter valve cooling system so as to solve the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme:
a fault diagnosis system based on a converter valve cooling system; the system comprises acquisition hardware, a data acquisition service subsystem, a data processing service subsystem, a data diagnosis service subsystem, a Web service subsystem and a database system;
characterized in that the acquisition hardware: the device is used for being arranged at the position of the working component and used for acquiring corresponding working state information;
the data acquisition service subsystem: the system comprises a data acquisition service subsystem, a data processing service subsystem, a data diagnosis service subsystem and a Web service subsystem, wherein the data acquisition service subsystem is used for acquiring information acquired by acquisition hardware and transmitting the information to the data acquisition service subsystem, the data processing service subsystem, the data diagnosis service subsystem and the Web service subsystem through a communication soft bus;
the data processing service subsystem: reading the calculation requirement of the characteristic quantity according to the configuration in the database, extracting and calculating the characteristics of the acquired data, and combining the acquired data with a diagnosis model in a data diagnosis service subsystem to realize the fault diagnosis of the dynamic equipment of the valve cooling system, wherein the characteristic parameters mainly comprise: time domain characteristic parameters, frequency domain characteristic parameters and envelope signal spectrum parameters;
the data diagnosis service subsystem: reading fault model diagnosis rules according to configuration in a database, substituting various characteristic parameters obtained after processing of a data processing subsystem, and judging whether a fault model takes effect or not so as to generate a corresponding fault diagnosis result of the valve cold-working equipment, wherein the fault diagnosis result is bound with specific fault components, fault reasons, fault processing suggestions and measures, and meanwhile, the diagnosis model comprises a characteristic prediction model, and the change trend of various characteristic quantities in a future period of time can be predicted through the characteristic prediction model;
the Web service subsystem: the man-machine interaction interface is used for externally displaying various functions of the system and configuring parameters of various subsystems;
a database system: the system is used for storing original data of measuring points of the fault diagnosis system, various feature quantity data after data processing, a diagnosis model, a diagnosis result, alarm information and configuration data of each system.
On the basis of the technical scheme, the invention also provides the following optional technical scheme:
in the alternative: the time domain characteristic parameters are dimensionless index parameters and are mainly selected from the consideration of the sensitivity and the stability to fault waveforms: the waveform index W, the pulse index I, the margin index L and the kurtosis index K are subjected to feature extraction, and the state monitoring can be performed on key parts of the valve cold-working equipment from a deeper level by setting the threshold values of the corresponding indexes.
In the alternative: the time domain feature extraction process mainly comprises the following steps: firstly, according to the collected vibration signal { x } 1 ,x 2 ,…,x N (N isNumber of sampling points), acquiring data length N, and calculating average value according to the acquired data and the data length
Variance σ, mean square value>
Square root value->
Peak value X p ＝max|x i According to X |, then rms 、X mean 、X P 、X r Respectively calculate the waveform index->
Pulse index->
Margin indicator>
Kurtosis index
Outputting time domain characteristic parameters after passing index rationality check; wherein the threshold value is set by adopting Gaussian distribution data statistical calculation.
In the alternative: the frequency domain characteristic parameters are mainly fault passing frequency of the valve cold-working equipment, and the extraction process is as follows:
the method comprises the following steps: obtaining accurate FFT, wherein the accurate FFT calculation module completes FFT calculation and spectrum correction of signals, and a window function Hanning window which is the same as the window function Hanning window is used in the calculation of the frequency spectrum;
step two: the fault passing frequency is obtained, when the working surface of the valve cold-working equipment is peeled off, indented or locally corroded, periodic pulses appear in the monitoring signal, and the frequency of the pulses is called the passing frequency corresponding to the fault. The rotating frequency of the valve cold-working equipment is set as f, and considering three types of motors, rolling bearings and pumps, the fault passing frequency of each element is mainly expressed as follows: rotating frequency component f, frequency multiplication component 2f, higher harmonic 3f and blade passing frequencyRate Zf (Z is the number of blades), blade passing frequency harmonic kzf (k =2, 3), bearing inner race fault BPFI = (N/2) f r [1-b d cos(β)/d p ]And BPFO = (N/2) f for bearing outer ring fault r [1-b d cos(β)/d p ]. In the above formula f r For the motor speed, N is the number of rolling elements, bd and dp are the rolling element diameter and the bearing pitch diameter. Beta is the contact angle of the rolling elements. The rotating frequency of the valve cold-working equipment is set as f, and considering three types of motors, rolling bearings and pumps, the fault passing frequency of each element is as follows:
the theoretical frequency value can be calculated by the passing frequency calculation formula in the table, but the valve cooling device generates certain rotation speed fluctuation during actual operation, so that the passing frequency needs to be searched within a certain frequency error, and if the frequency error range during searching is +/-delta f, the frequency range during passing frequency searching is f 0 And searching the frequency corresponding to the maximum amplitude within +/-delta f, namely the actual corresponding passing frequency. The main implementation process is as follows: firstly, obtaining a corrected frequency spectrum signal according to an accurate FFT (fast Fourier transform), then calculating theoretical fault passing frequency and frequency band energy according to a calculation formula, searching actual power frequency and correcting power frequency, calculating a fault passing frequency searching range, searching fault passing frequency, and obtaining frequency band energy, amplitude and fault passing frequency.
In the alternative: envelope signal spectrum parameters, and an envelope demodulation module completes the envelope of a bearing fault signal through Hilbert transformation of the signal, so that an envelope signal of an original signal is obtained, and subsequent spectrum analysis is facilitated.
In the alternative: the envelope signal acquisition process comprises the following steps: the method comprises the steps of obtaining vibration signals, sampling frequency and analysis point number, conducting Hilbert transform on obtained data to obtain envelope signals, conducting fast Fourier transform on the signals to obtain a frequency spectrum mode function with Hanning window length normalization, then searching frequency spectrum peak values and secondary peak values and judging positions, constructing a ratio function upsilon, solving frequency errors of an inverse function of the ratio function, calculating real frequency, then calculating real amplitude, and correcting the frequency spectrum of the envelope signals.
In the alternative: the accurate FFT is obtained as follows: the method comprises the steps of obtaining a vibration signal, sampling frequency and analysis points, carrying out fast Fourier transform on the signal to obtain a frequency spectrum mode function with Hanning window length normalization, then searching a frequency spectrum peak value and a secondary peak value and judging positions, constructing a ratio function upsilon, solving a frequency error of an inverse function of the ratio function, calculating real frequency, calculating real amplitude, and correcting an envelope signal frequency spectrum.
In the alternative: the method for establishing the fault diagnosis model comprises the following steps: the method comprises the steps of establishing a motor unbalance fault model, an angle misalignment fault model, a parallel misalignment fault model, a pump blade fault model and a bearing fault model.
In the alternative: the method for reconstructing the feature prediction model of the key parameter by adopting the phase space mainly comprises the following steps: selecting characteristic parameters to be predicted and analyzed of the valve cooling equipment, and taking out a period of time sequence data of the parameters from a system database; decomposing the time series data into a matrix of m x n; regarding each row as an n-dimensional vector, and calculating K vectors and the probability closest to the last one-dimensional vector; moving Step backwards under the found K vectors; subtracting the corresponding K vectors from the moved vector, and weighting the value of the vector by multiplying the probability of the point to obtain a predicted pointing vector after Step; and pointing to the last row of the vector + matrix, namely the vector after Step. The last value of the finger vector is the predicted value. In the algorithm, n is an embedding dimension, m is obtained by intercepting time delay parameters t from time sequence data, starting from an index 0, starting from the last n-1 rows as a 0 th row, starting from an index 0+ _ t, and taking the last n-2 rows as a 1 st row, and Step is the prediction Step number. The time delay parameter, the dimension and the prediction step number need to be set on a front-end page, the time delay parameter is suggested to be set between values 1-2, the dimension is suggested to be set between 2-3, and the prediction step number 1 step represents 1 day.
Compared with the prior art, the fault diagnosis system based on the converter valve cooling system power equipment provided by the invention has the following advantages:
monitoring and protecting: the automatic monitoring of key movable equipment of the valve cooling system for 7x24 hours in all weather is realized; and (5) alarming in time when a fault is discovered, and isolating the fault.
Analysis and diagnosis: judging the nature, degree and position of the fault of the mobile equipment through a fault diagnosis model; and analyzing the fault reason.
Treatment and prevention: providing a measure for eliminating the fault through the processing measure and the suggestion of each fault model configured in the database; and the change trend of various characteristic quantities in a period of time in the future is predicted through the characteristic prediction model, so that the purpose of preventing equipment from generating faults is achieved.
Drawings
Fig. 1 is a frame diagram of the present invention.
FIG. 2 is a data processing flow of the present invention.
Fig. 3 is a flow chart of the accurate FFT computation module of the present invention.
Fig. 4 illustrates the envelope signal acquisition process of the present invention.
FIG. 5 is a feature prediction model implementation of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments.
In one embodiment, as shown in fig. 1-5, a converter valve cooling system based fault diagnosis system includes two major parts: data acquisition hardware and software systems. The software system comprises a data acquisition service subsystem, a data processing service subsystem, a data diagnosis service subsystem, a data storage subsystem, a web service subsystem and a software simulation data acquisition service subsystem. In order to be compatible with different data collectors and better expand data processing and data diagnosis services, a soft bus is adopted to connect all parts of the system. And the corresponding system functions are externally displayed by utilizing the Web server, and configuration operation inlets of the subsystems are externally provided. The data acquisition service subsystem is responsible for driving the data acquisition device to acquire corresponding state data and simultaneously controlling the action of the convertor station acquisition device. The data processing and diagnosis service subsystem is responsible for carrying out feature calculation, feature extraction, performance evaluation and diagnosis on the acquired data according to a predefined strategy. The system framework is shown in FIG. 1 below;
acquisition hardware: the acquisition hardware comprises an acquisition sensor and a digital acquisition device. The acquisition sensor mainly comprises a temperature sensor, a current sensor and a vibration sensor, and the vibration sensor adopts an acceleration vibration sensor.
The data acquisition service subsystem: the data acquisition service subsystem is mainly responsible for driving and collecting all the measuring point state data uploaded by the networked data acquisition unit, and a communication protocol adopts TCP. After receiving the data, on the one hand, the data is stored into the corresponding database in a storage period of 15 minutes according to the configuration information in the database. And simultaneously, notifying the corresponding result to the Web service subsystem in real time in a queue message eliminating mode, and displaying an interface. And on the other hand, the data processing service subsystem and the data diagnosis service subsystem are informed to process the data acquired at this time, and the state of the unit is judged. And if the data is changed into fault data, changing the data storage period to 1S until the data is restored to normal, and restoring to the original storage period.
The data processing service subsystem: the data processing service subsystem reads the characteristic quantity calculation requirements according to the configuration in the database, performs characteristic extraction and calculation on the acquired data, and realizes fault diagnosis of the dynamic equipment of the valve cooling system by combining with the diagnosis model in the data diagnosis service subsystem. The characteristic parameters used by the invention mainly comprise: time domain characteristic parameters, frequency domain characteristic parameters and envelope signal spectrum parameters.
The data diagnosis service subsystem: the monitoring and diagnosing service subsystem reads the fault model diagnosing rules according to the configuration in the database, substitutes various characteristic parameters obtained after the processing of the data processing subsystem, and judges whether the fault model takes effect or not, so that the corresponding fault diagnosing result of the valve cold-working equipment is generated, and the fault diagnosing result is bound with specific fault components, fault reasons, fault processing suggestions and measures. Meanwhile, the diagnostic model comprises a characteristic prediction model, and the change trend of various characteristic quantities in a period of time in the future can be predicted through the characteristic prediction model.
The Web service subsystem: the Web service subsystem shown in the invention is a man-machine interaction interface of the whole system, and mainly displays various functions of the system and performs parameter configuration on various subsystems.
A database system: the database system is mainly used for storing original data of a measuring point of a fault diagnosis system, various feature quantity data after data processing, a diagnosis model, a diagnosis result, alarm information, configuration data of each system and the like;
the main implementation process is shown in FIG. 2;
the system of the invention extracts typical fault characteristic parameters by processing vibration data to form a practical method set suitable for monitoring and diagnosing, and realizes the state monitoring and fault diagnosis of equipment by combining with a diagnosis model. The fault characteristic parameters of the valve cold-working equipment adopted by the invention mainly comprise: time domain characteristic parameters, frequency domain characteristic parameters and envelope signal spectrum parameters. The diagnosis system calculates the fault characteristic indexes of the valve cold-working equipment such as a waveform index, a pulse index, a kurtosis index, a margin index and the like by constructing a time domain and frequency domain characteristic extraction tool; obtaining accurate FFT spectrum characteristics and various fault passing frequency characteristics through an accurate FFT calculation module; and extracting the characteristic component frequency spectrum of the envelope signal by adopting an envelope demodulation method according to the characteristics of the formation of the impact faults.
1.1 the time domain characteristic parameters used by the system of the invention are mainly dimensionless index parameters, and are mainly selected from the consideration of sensitivity and stability to fault waveforms: the waveform index W, the pulse index I, the margin index L and the kurtosis index K are subjected to feature extraction, and the state monitoring can be performed on key parts of the valve cold-working equipment from a deeper level by setting the threshold values of the corresponding indexes.
The characteristic extraction process mainly comprises the following steps: firstly, according to the collected vibration signal { x 1 ,x 2 ,…,x N And (N is the number of sampling points), acquiring the data length N, and calculating an average value according to the acquired data and the data length
Variance σ, mean square value
Square root value>
Peak value X p ＝max|x i According to X rms 、X mean 、X P 、X r Respectively calculate the waveform index->
Pulse index->
Margin indicator->
Kurtosis index->
And outputting the time domain characteristic parameters after passing the index rationality check.
Wherein the threshold value is set by adopting Gaussian distribution data statistical calculation. Generally, alarm lines with one characteristic are arranged according to Gaussian distribution, after a system runs for a period of time, data of the period of time are counted, the average value and the deviation of the data are calculated, one positive deviation and one negative deviation are primary alarm, the area of the primary alarm accounts for 66.66%, two positive deviations and two negative deviations and three positive deviations are secondary alarms, and more positive deviations and three negative deviations are tertiary alarms.
1.2 the frequency domain characteristic parameter used by the system is mainly the fault passing frequency of the valve cold-working equipment. The extraction process is as follows:
(1) Method for obtaining accurate FFT.
The accurate FFT calculation module completes FFT calculation and spectrum correction of the signal, wherein the window function and the spectrum calculation use the same window function Hanning window. The process is shown in figure 3 below.
The fault passes through a frequency acquisition method.
When the working surface of the valve cold-working equipment is peeled off, indented or partially corroded, periodic pulses occur in the monitoring signal, and the frequency of the pulses is called the passing frequency corresponding to the fault. Rotation frequency of cold-working equipment with valveConsidering three types of motors, rolling bearings and pumps, the failure passing frequency of each element is mainly represented as follows: a frequency conversion component f, a frequency multiplication component 2f, a higher harmonic 3f, a blade passing frequency Zf (Z is the number of blades), a blade passing frequency harmonic kzf (k =2, 3), a bearing inner ring fault BPFI = (N/2) f r [1-b d cos(β)/d p ]And BPFO = (N/2) f for bearing outer ring fault r [1-b d cos(β)/d p ]. In the above formula f r For the rotational speed of the motor, N is the number of rolling elements, and bd and dp are the rolling element diameter and the bearing pitch diameter. Beta is the contact angle of the rolling elements. Setting the rotating frequency of the valve cold-working equipment as f, and considering three types of motors, rolling bearings and pumps, the fault passing frequency of each element is as follows:
the theoretical frequency value can be calculated by the passing frequency calculation formula in the table, but the valve cooling device generates certain rotation speed fluctuation during actual operation, so that the passing frequency needs to be searched within a certain frequency error, and if the frequency error range during searching is +/-delta f, the frequency range during passing frequency searching is f 0 And searching the frequency corresponding to the maximum amplitude within +/-delta f, namely the actual corresponding passing frequency. The main implementation process is as follows: firstly, obtaining a corrected frequency spectrum signal according to an accurate FFT (fast Fourier transform), then calculating theoretical fault passing frequency and frequency band energy according to a calculation formula, searching actual power frequency and correcting power frequency, calculating a fault passing frequency searching range, and searching fault passing frequency to obtain frequency band energy, amplitude and fault passing frequency.
1.3 envelope signal spectral parameters used by the system of the invention:
the envelope demodulation module completes the envelope of the bearing fault signal through the Hilbert transform of the signal, so that an envelope signal of an original signal is obtained, and subsequent spectrum analysis is facilitated. The envelope signal spectrum extraction process is shown in fig. 4 below.
The method for establishing the multiple fault diagnosis models for the valve cold-working equipment comprises the following steps:
for the cold system power equipment of the converter station valve, the main faults thereof comprise: motor imbalance faults, pump imbalance faults, corner misalignment faults, parallel misalignment faults, pump blade faults, bearing faults.
2.1 method for establishing motor unbalance fault model
Firstly, obtaining the frequency spectrum amplitudes XH, XV and XA of the output ends of the motors of the moving equipment in the horizontal direction, the vertical direction and the axial direction corresponding to the measuring points 1X, 2X and 3X through characteristic extraction, and forming a diagnosis model of the unbalanced fault of the motors of the moving equipment according to the characteristic expression of the unbalanced fault of the motors: {1XH > 0.8m/s ^2& &2XH < 0.5 x 1XH & &3XH < 0.6 x 1XH & &1XA < 0.5 x 1XH | |1XV > 0.4m/s ^2& &2XV < 0.5 x 1XV & &3XV < 0.6 x 1XV & &1XA < 0.5 x 1XV }.
2.2 method for establishing pump unbalance fault model
Firstly, obtaining spectral amplitudes XH, XV and XA of corresponding measuring points 1X, 2X and 3X in the horizontal direction, the vertical direction and the axial direction of a pump end through characteristic extraction, and forming a pump balance fault diagnosis model according to the characteristic expression of pump unbalance faults: { diameter 1X >2 × 1X (normal amplitude) } | {1X (normal amplitude) m/s ^2< diameter 1X yarn 2 × 1X (normal amplitude) } & { diameter 2X yarn 0.9 × 1X } & { diameter 3X yarn and 0.3 × 1X } & { shaft 2X < shaft 1X }.
Method for establishing 2.3-angle misalignment fault model
Firstly, obtaining the frequency spectrum amplitudes XH, XV and XA of the measuring points 1X, 2X and 3X corresponding to the horizontal direction, the vertical direction and the axial direction of the coupler end through feature extraction, and forming a coupler angle misalignment fault diagnosis model according to the coupler angle misalignment fault feature expression: 1XH >0.5 m/s ^2& &2XH < 0.5 x 1XH & &3XH < 0.6 x 1XH |2XA > 0.8 x 1XH |1XV > 0.25m/s ^2& &2XV < 0.5 x 1XV | |2XA > 0.8 x 1XH |1XV & &3XV < 0.6 x 1XV & & (1 XA >0.5 x 1XV |2XA > 0.8 x 1XV).
2.4 parallel misalignment fault model establishing method
Firstly, obtaining spectral amplitude values XH, XV and XA of measuring points 1X, 2X and 3X corresponding to the horizontal direction, the vertical direction and the axial direction of a coupler end through feature extraction to form a coupler parallel misalignment fault diagnosis model: 1XH > 0.25m/s ^2& ((2 XH/(6.28 × 100) +3 XH/(6.28 × 150) +4 XH/(6.28 × 200)) > 0.4 × 1XH/(6.28 × 50) |1XA > 1 XH) |1XV > 0.25m/s ^2& ((2 XV/(6.28 × 100) +3 XV/(6.28 × 150) +4 XV/(6.28 × 200)) > 0.4 1XV/(6.28 × 50) |1XA 1 XV).
2.5 method for establishing fault model of pump blade
Firstly, obtaining spectral amplitudes XH, XV and XA of corresponding measuring points 1X, 2X and 3X in the horizontal direction, the vertical direction and the axial direction of a pump end through characteristic extraction, and forming a pump balance fault diagnosis model according to the characteristic expression of pump unbalance faults: 1XH ^ 0.3m/s ^2& (number of blades: diameter 1X) >2.2m/s ^2& &2 times the blade passing frequency > 0.5X 1.
2.6 bearing fault model establishing method
The fault models of the bearing are mainly divided into two types: inner ring failure, outer ring failure. According to the method, the rotating speed side frequency of inner and outer ring fault passing frequencies BPFI and BPFO is obtained through feature extraction, envelope signals EBPFI1X and EBPFO1X are obtained through envelope demodulation, and a bearing fault diagnosis model is formed according to bearing fault feature expression and is divided into an outer ring fault feature model and an inner ring fault feature model.
Outer ring fault characteristic model: { EBPFO1X >0.5| | Ap >6 | arms | | BPFO >0.5}
Inner ring fault characteristic model: { EBPFI1X >0.5| | Ap >6 | arms | | BPFI >0.5}
And (3) a characteristic prediction model aiming at key parameters of the valve cold-working equipment.
The system adopts a phase space reconstruction method aiming at a characteristic prediction model of key parameters of valve cold-working equipment, and mainly comprises the following steps: selecting characteristic parameters to be predicted and analyzed of the valve cooling equipment, and taking out a period of time sequence data of the parameters from a system database; decomposing the time series data into a matrix of m x n; regarding each row as an n-dimensional vector, and calculating K vectors and the probability closest to the last one-dimensional vector; moving Step backwards under the found K vectors; subtracting the corresponding K vectors from the obtained vector after movement, and multiplying the value by the probability of the point for weighting to obtain the predicted pointing vector after Step; and pointing to the last row of the vector + matrix, namely the vector after Step. The last value of the finger vector is the predicted value. The process is shown in fig. 5, where n is the embedding dimension, m is the time delay parameter t, starting from index 0, the last n-1 is the 0 th line, the index 0+ _ t, the last n-2 is the 1 st line, and Step is the prediction Step number. The time delay parameter, the dimension and the prediction step number need to be set on a front-end page, the time delay parameter is suggested to be set between values 1-2, the dimension is suggested to be set between 2-3, and the prediction step number 1 step represents 1 day.
The above description is only for the specific embodiments of the present disclosure, but the scope of the present disclosure is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present disclosure, and all the changes or substitutions should be covered within the scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (9)
1. The fault diagnosis system based on the converter valve cooling system comprises acquisition hardware, a data acquisition service subsystem, a data processing service subsystem, a data diagnosis service subsystem, a Web service subsystem and a database system;
characterized in that the acquisition hardware: the device is used for being arranged at the position of the working component and used for acquiring corresponding working state information;
the data acquisition service subsystem: the system comprises a data acquisition service subsystem, a data processing service subsystem, a data diagnosis service subsystem and a Web service subsystem, wherein the data acquisition service subsystem is used for acquiring information acquired by acquisition hardware and transmitting the information to the data acquisition service subsystem, the data processing service subsystem, the data diagnosis service subsystem and the Web service subsystem through a communication soft bus;
the data processing service subsystem: reading the feature quantity calculation requirement according to the configuration in the database, extracting and calculating the features of the acquired data, and realizing the fault diagnosis of the valve cooling system dynamic equipment by combining with a diagnosis model in a data diagnosis service subsystem, wherein the feature parameters mainly comprise: time domain characteristic parameters, frequency domain characteristic parameters and envelope signal spectrum parameters;
the data diagnosis service subsystem: reading fault model diagnosis rules according to configuration in a database, substituting various characteristic parameters obtained after processing of a data processing subsystem, and judging whether a fault model takes effect or not so as to generate a corresponding fault diagnosis result of the valve cold-working equipment, wherein the fault diagnosis result is bound with specific fault components, fault reasons, fault processing suggestions and measures, and meanwhile, the diagnosis model comprises a characteristic prediction model, and the change trend of various characteristic quantities in a future period of time can be predicted through the characteristic prediction model;
the Web service subsystem: the man-machine interaction interface is used for externally displaying various functions of the system and configuring parameters of various subsystems;
a database system: the system is used for storing original data of measuring points of the fault diagnosis system, various feature quantity data after data processing, a diagnosis model, a diagnosis result, alarm information and configuration data of each system.
2. The converter valve cooling system based fault diagnosis system according to claim 1, wherein the time domain characteristic parameters are dimensionless index parameters, and are selected from the following aspects of sensitivity to fault waveforms and stability: the waveform index W, the pulse index I, the margin index L and the kurtosis index K are subjected to feature extraction, and the state monitoring can be performed on key parts of the valve cold-working equipment from a deeper level by setting the threshold values of the corresponding indexes.
3. The converter valve cooling system based fault diagnosis system according to claim 2, wherein the time domain feature extraction process mainly comprises the following steps: firstly, according to the collected vibration signal { x 1 ,x 2 ,…,x N And (N is the number of sampling points), acquiring the data length N, and calculating a mean value according to the acquired data and the data length
Variance sigma, mean square value->
Root of Square
Peak value X p ＝max|x i According to X rms 、X mean 、X P 、X r Respectively calculate the waveform index->
Pulse index->
Margin indicator->
Kurtosis index->
Outputting time domain characteristic parameters after passing the index rationality check;
wherein the threshold value is set by adopting Gaussian distribution data statistical calculation.
4. The fault diagnosis system based on the converter valve cooling system as claimed in claim 1, wherein the frequency domain characteristic parameter is mainly a fault passing frequency of the valve cooling equipment, and the extraction process is as follows:
the method comprises the following steps: obtaining accurate FFT, finishing FFT calculation and spectrum correction of the signal by an accurate FFT calculation module, wherein a window function and a frequency spectrum use the same window function Hanning window when calculating;
step two: obtaining the passing frequency of the fault, wherein when the working surface of the valve cold-working equipment is peeled off, indented or partially corroded, periodic pulses can appear in the monitoring signal, and the frequency of the pulses is called the passing frequency of the corresponding fault; the rotating frequency of the valve cold-working equipment is set as f, and considering three types of motors, rolling bearings and pumps, the failure passing frequency of each element is mainly represented as follows: a frequency conversion component f, a frequency multiplication component 2f, a higher harmonic 3f, a blade passing frequency Zf (Z is the number of blades), a blade passing frequency harmonic kzf (k =2, 3), a bearing inner ring fault BPFI = (N/2) f r [1-b d cos(β)/d p ]BPFO = (N/2) f for bearing outer ring fault r [1-b d cos(β)/d p ](ii) a In the above formula f r The rotating speed of the motor is N, the number of the rolling bodies is N, and bd and dp are the diameters of the rolling bodies and the pitch diameter of the bearing; β is the contact angle of the rolling element; the rotating frequency of the valve cold-working equipment is set as f, and considering three types of motors, rolling bearings and pumps, the fault passing frequency of each element is as follows:
the theoretical frequency value can be calculated by the passing frequency calculation formula in the table, but the valve cooling device generates certain rotation speed fluctuation during actual operation, so that the passing frequency needs to be searched within a certain frequency error, and if the frequency error range during searching is +/-delta f, the frequency range during passing frequency searching is f 0 Searching the frequency corresponding to the maximum amplitude within +/-delta f, namely the actual corresponding passing frequency, wherein the main implementation process is as follows: firstly, obtaining a corrected frequency spectrum signal according to an accurate FFT (fast Fourier transform), then calculating theoretical fault passing frequency and frequency band energy according to a calculation formula, searching actual power frequency and correcting power frequency, calculating a fault passing frequency searching range, and searching fault passing frequency to obtain frequency band energy, amplitude and fault passing frequency.
5. The converter valve cooling system based fault diagnosis system according to claim 1, wherein envelope signal spectrum parameters are enveloped, and the envelope demodulation module completes the envelope of the bearing fault signal through Hilbert transform of the signal, so as to obtain an envelope signal of an original signal for subsequent spectrum analysis.
6. The converter valve cooling system based fault diagnosis system according to claim 5, wherein the envelope signal acquisition process is as follows: the method comprises the steps of obtaining vibration signals, sampling frequency and analysis points, conducting Hilbert transform on obtained data to obtain envelope signals, conducting fast Fourier transform on the signals to obtain a spectrum mode function with Hanning window length normalization, then searching spectrum peak values and secondary peak values and judging positions, constructing a ratio function upsilon, solving frequency errors of an inverse function of the ratio function, calculating real frequency, calculating real amplitude, and correcting spectrum of the envelope signals.
7. The converter valve cooling system based fault diagnosis system according to claim 4, wherein the accurate FFT is obtained as follows: the method comprises the steps of obtaining a vibration signal, sampling frequency and analysis points, carrying out fast Fourier transform on the signal to obtain a frequency spectrum mode function with Hanning window length normalization, then searching a frequency spectrum peak value and a secondary peak value and judging positions, constructing a ratio function upsilon, solving a frequency error of an inverse function of the ratio function, calculating real frequency, calculating real amplitude, and correcting an envelope signal frequency spectrum.
8. The converter valve cooling system based fault diagnosis system according to claim 4, further comprising a fault diagnosis model establishing method: the method comprises the steps of establishing a motor unbalance fault model, an angle misalignment fault model, a parallel misalignment fault model, a pump blade fault model and a bearing fault model.
9. The converter valve cooling system based fault diagnosis system according to claim 4, wherein the feature prediction model of the key parameters adopts a phase space reconstruction method, and mainly comprises the following steps: selecting characteristic parameters to be predicted and analyzed of the valve cooling equipment, and taking a period of time sequence data of the parameters from a system database; decomposing the time series data into a matrix of m x n; regarding each row as an n-dimensional vector, and calculating K vectors and the probability closest to the last one-dimensional vector; moving Step backwards under the found K vectors; subtracting the corresponding K vectors from the moved vector, and weighting the value of the vector by multiplying the probability of the point to obtain a predicted pointing vector after Step; pointing to the last row of the vector + matrix, namely the vector after Step; the last value of the finger vector is a predicted value; in the algorithm, n is an embedding dimension, m is obtained by intercepting time delay parameters t from time sequence data, starting from an index 0, starting from the last n-1 lines, starting from an index 0+ _ t, starting from the last n-2 lines, and taking Step as a prediction Step number; the time delay parameter, the dimension and the prediction step number need to be set in a front-end page, the time delay parameter is suggested to be set between 1 and 2, the dimension is between 2 and 3, and the prediction step number 1 represents 1 day.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211429741.2A CN115876507A (en) | 2022-11-14 | 2022-11-14 | Fault diagnosis system based on converter valve cooling system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211429741.2A CN115876507A (en) | 2022-11-14 | 2022-11-14 | Fault diagnosis system based on converter valve cooling system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115876507A true CN115876507A (en) | 2023-03-31 |
Family
ID=85759990
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211429741.2A Pending CN115876507A (en) | 2022-11-14 | 2022-11-14 | Fault diagnosis system based on converter valve cooling system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115876507A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116298651A (en) * | 2023-05-17 | 2023-06-23 | 广东电网有限责任公司阳江供电局 | Fault monitoring method, system, equipment and medium for converter valve power module |
| CN116680630A (en) * | 2023-07-27 | 2023-09-01 | 成都雨航创科科技有限公司 | Human-vehicle motion detection method and device based on vibration and image |
| CN117851873A (en) * | 2024-03-07 | 2024-04-09 | 唐智科技湖南发展有限公司 | Bearing running state evaluation method and system based on dynamic contact angle |
-
2022
- 2022-11-14 CN CN202211429741.2A patent/CN115876507A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116298651A (en) * | 2023-05-17 | 2023-06-23 | 广东电网有限责任公司阳江供电局 | Fault monitoring method, system, equipment and medium for converter valve power module |
| CN116680630A (en) * | 2023-07-27 | 2023-09-01 | 成都雨航创科科技有限公司 | Human-vehicle motion detection method and device based on vibration and image |
| CN116680630B (en) * | 2023-07-27 | 2023-10-13 | 成都雨航创科科技有限公司 | Human-vehicle motion detection method and device based on vibration and image |
| CN117851873A (en) * | 2024-03-07 | 2024-04-09 | 唐智科技湖南发展有限公司 | Bearing running state evaluation method and system based on dynamic contact angle |
| CN117851873B (en) * | 2024-03-07 | 2024-05-28 | 唐智科技湖南发展有限公司 | Bearing running state evaluation method and system based on dynamic contact angle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115876507A (en) | Fault diagnosis system based on converter valve cooling system | |
| WO2023035869A1 (en) | Gearbox fault diagnosis model training method and gearbox fault diagnosis method | |
| CN102798529B (en) | Method and system for diagnosing bearing faults of large-size wind turbine bearing | |
| CN103411774B (en) | Wind turbines on-line early warning method under fluctuation operating mode | |
| CN105784366A (en) | Wind turbine generator bearing fault diagnosis method under variable speed | |
| CN106650122B (en) | A kind of equipment variable parameter operation methods of risk assessment | |
| CN103176128A (en) | Method and system for forcasting state of wind generating set and diagnosing intelligent fault | |
| CN105205569A (en) | Draught fan gear box state on-line evaluation model building method and on-line evaluation method | |
| CN109655266A (en) | A kind of Wind turbines Method for Bearing Fault Diagnosis based on AVMD and spectral coherence analysis | |
| CN112065668A (en) | Wind turbine generator state abnormity assessment method and system | |
| CN111504450A (en) | Equipment fault alarm method and device | |
| CN113468728B (en) | Variable pitch system fault prediction method based on neural network | |
| KR20210010123A (en) | Method for fault diagnosis based on multiple variables and apparatus using the method | |
| Zhu et al. | Adaptive combined HOEO based fault feature extraction method for rolling element bearing under variable speed condition | |
| CN112308391A (en) | Real-time monitoring and anomaly detection method for equipment state based on neural network | |
| CN104459186A (en) | Tachometer-free order-ratio analyzing method based on sparse segmentation fitting and integral approximation | |
| CN115434870B (en) | Method, device, equipment and storage medium for processing vibration data of wind power generation set | |
| CN103696913B (en) | The method of power utilization deviation during a kind of Real-Time Monitoring running of wind generating set | |
| CN114398842A (en) | Method for evaluating generated energy of operating wind power plant | |
| CN111400959B (en) | Blade fault diagnosis method and device for wind generating set | |
| CN114607571A (en) | Offshore wind power gear box fault identification method and system based on lubricating system monitoring | |
| CN113807693A (en) | Wind turbine generator generating capacity optimization effect evaluation method based on airborne radar | |
| CN113111314A (en) | Multi-unit cooperation based health state evaluation method for grid-connected wind turbine generator | |
| CN112035789A (en) | Fault diagnosis method for gear transmission system of wind driven generator | |
| CN117851873B (en) | Bearing running state evaluation method and system based on dynamic contact angle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |