CN113027658B - Real-time state evaluation method for water turbine runner and application thereof - Google Patents
Real-time state evaluation method for water turbine runner and application thereof Download PDFInfo
- Publication number
- CN113027658B CN113027658B CN202110311462.5A CN202110311462A CN113027658B CN 113027658 B CN113027658 B CN 113027658B CN 202110311462 A CN202110311462 A CN 202110311462A CN 113027658 B CN113027658 B CN 113027658B
- Authority
- CN
- China
- Prior art keywords
- characteristic
- real
- turbine runner
- data
- active power
- 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.)
- Active
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03B—MACHINES OR ENGINES FOR LIQUIDS
- F03B11/00—Parts or details not provided for in, or of interest apart from, the preceding groups, e.g. wear-protection couplings, between turbine and generator
- F03B11/008—Measuring or testing arrangements
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/20—Hydro energy
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Control Of Water Turbines (AREA)
Abstract
The invention belongs to the field of monitoring and evaluating hydropower station runner equipment, and particularly relates to a real-time state evaluating method for a water turbine runner and application thereof, wherein the method comprises the following steps: acquiring a water head and active power of a water turbine unit and various sensor monitoring signals of a water turbine runner in real time, extracting various characteristic values in each sensor monitoring signal, wherein all the characteristic values jointly form a characteristic value vector; respectively obtaining standard values of various characteristics under the working conditions corresponding to the water head and the active power by adopting a nonlinear mapping relation model between the working conditions and the standard values of various characteristics, wherein all the standard values jointly form a characteristic standard value vector; and evaluating the real-time state of the turbine runner by comparing the characteristic value vector with the characteristic standard value vector. The invention adopts the nonlinear mapping relation model between the working condition and the standard value of each signal characteristic in the normal operation process to carry out real-time evaluation, eliminates the influence of the monitoring signal change caused by the working condition change on the evaluation result, and improves the rationality and the accuracy of the evaluation.
Description
Technical Field
The invention belongs to the field of monitoring and evaluation of hydropower station runner equipment, and particularly relates to a real-time state evaluation method of a water turbine runner and application thereof.
Background
With the increase of the proportion of hydroelectric power generation in the electric power system in China, the single machine capacity of the hydraulic turbine set is continuously increased, and the safe, stable and efficient operation of the hydraulic turbine set faces huge challenges. As a core component of a hydraulic turbine unit, a rotating wheel bears a large load and changes frequently in working conditions, so that fault defects are easily generated to cause major accidents. Therefore, scientific and effective monitoring and evaluation of the state of the turbine runner are important guarantees for the safety and the economy of the hydropower station.
Due to the complex working environment of the turbine runner and the limitation of the existing on-line monitoring and state evaluation technology, most power stations still adopt a 'planned maintenance' mode, and the fault defect of the runner can be discovered only when the runner is periodically maintained, so that the timeliness is lacked. At present, although some existing technologies for evaluating the real-time state of the hydraulic turbine set exist, there still exists a certain limitation in the adaptability of complex working conditions. For example, the invention patent "hydraulic turbine set fault diagnosis and state evaluation method" (application publication No. CN106988951A) and "hydraulic turbine set fault diagnosis and state of health evaluation method" (application publication No. CN110552832A) both use the difference between various types of health data and real-time monitoring data of the hydraulic turbine set during normal operation as the state evaluation basis. Although the invention can realize the real-time monitoring of the state of the unit, the influence of the working condition factors of the unit on the normal range of the monitoring value is less considered.
During the long-term stable operation of the hydropower station, even if the rotating speed is basically kept unchanged, the variation range of working condition parameters such as water head, load, flow and the like is large, so that the monitoring signal fluctuates in a large range, and great difficulty is brought to the real-time state monitoring and evaluation of the runner. At present, a method and a system for evaluating the real-time state of a turbine runner under complex working conditions are not common.
Disclosure of Invention
Aiming at the defects and the improvement requirements of the prior art, the invention provides a real-time state evaluation method of a turbine runner and application thereof, and aims to evaluate the state of the turbine runner under complex working conditions in real time and accurately.
To achieve the above object, according to one aspect of the present invention, there is provided a real-time state estimation method for a turbine runner, including:
acquiring a water head and active power of a water turbine unit and various sensor monitoring signals of a water turbine runner in real time, extracting various characteristic values in each sensor monitoring signal, and forming a characteristic value vector by all the characteristic values;
respectively adopting an established nonlinear mapping relation model between the working condition and the standard values of various characteristics to obtain the standard values of various characteristics under the working condition corresponding to the water head and the active power, wherein all the standard values jointly form a characteristic standard value vector;
and evaluating the real-time state of the turbine runner by comparing the characteristic value vector with the characteristic standard value vector.
The invention has the beneficial effects that: according to the method, the condition change and the working condition change of the unit can cause the change of the monitoring signal, wherein the monitoring signal change caused by the working condition change is normal, and the monitoring signal change caused by the unit condition needs attention, so that the health state of the unit can be obtained through the monitoring signal. Because the characteristic values of the monitoring signals of the turbine runner under different working conditions are different in the normal operation process, the invention establishes a nonlinear mapping relation model between the working conditions and the standard values of various (signal) characteristics in the normal operation process, in the actual real-time evaluation, on one hand, standard value vectors corresponding to various signal characteristics are obtained based on the current working condition parameters of the turbine runner, on the other hand, actual characteristic value vectors corresponding to various signal characteristics are obtained through the monitoring signals of the sensor, and by comparing the difference between the two vectors, the real-time state of the current water turbine runner can be estimated, and because the signal characteristic standard value in the traditional method is constant and does not change along with the working condition, the misjudgment and the missed judgment are easily caused, the invention improves the rationality and accuracy of the evaluation by eliminating the influence of the change of the monitoring signal caused by the change of the working condition on the evaluation result.
On the basis of the technical scheme, the invention can be further improved as follows.
Further, the nonlinear mapping relation model is obtained by adopting the following establishing mode:
s1, continuously acquiring discrete point data in the whole stable operation period of the water turbine, wherein each discrete point data comprises working condition parameters containing a water head and active power at a certain moment and monitoring signals of the various sensors;
s2, extracting characteristic values of various characteristics in each sensor monitoring signal in each discrete point data, and taking the characteristic values as a group of characteristic standard values corresponding to the discrete point data in the state of health of the turbine runner;
and S3, establishing a nonlinear mapping relation model between the characteristic standard value of each characteristic and the water head and active power, wherein the adopted data set is composed of data points (H, P, F) corresponding to each discrete point data, wherein H and P represent the water head and the active power in each discrete point data, and F represents the characteristic standard value of the characteristic corresponding to the discrete point data.
The invention has the further beneficial effects that: the method comprises the steps of collecting sensor monitoring signals and working condition parameters of the water turbine set in a long-term normal running state, ensuring that the working conditions recorded in the period are distributed in all possible working condition ranges of the power station as densely as possible, and enabling the established nonlinear mapping relation model to be more accurately suitable for real-time state evaluation of the water turbine runner under any working condition.
Further, in S3, a gaussian process regression is adopted to establish a nonlinear mapping relationship model between the standard value probability density distribution of the characteristic and the head and active power; simplifying the nonlinear mapping relation model between the standard value probability density distribution of the characteristics and the water head and the active power into a nonlinear mapping relation model between the mean value of the standard value probability density distribution of the characteristics and the water head and the active power, and taking the nonlinear mapping relation model as a final nonlinear mapping relation model.
The invention has the further beneficial effects that: since the monitoring signal of the sensor has volatility and randomness, the observation of a single observation value of the monitoring signal is easy to cause large errors. The method is based on Gaussian process regression, and a nonlinear mapping relation model between the probability density distribution of the health state monitoring signal characteristic standard values of the water turbine rotating wheels and the unit working condition parameters is respectively established, so that the accuracy of real-time state evaluation of the water turbine rotating wheels is ensured.
Further, the plurality of sensor monitoring signals includes: acoustic emission signals and vibration signals at four quadrant points of the top cover.
Further, the plurality of characteristic values in the vibration signal include: a root mean square value, a peak-to-peak value, a peak factor, and a kurtosis factor; the plurality of characteristic values in the acoustic emission signal include: event counting and ring counting.
The invention has the further beneficial effects that: the root mean square value reflects the average power of the signal; the peak-to-peak value reflects the size of the signal variation range; the crest factor reflects the extreme degree of the crest value in the waveform; the kurtosis factor reflects the smoothness of the signal waveform; event and ring counts reflect the amount and frequency of acoustic emission activity. The method extracts various characteristic values, is favorable for fully mining useful information contained in the signals, and improves the comprehensiveness and scientificity of state evaluation.
Further, by comparing the characteristic value vector with the characteristic standard value vector, the real-time state of the turbine runner is evaluated, specifically:
and calculating a quantitative index quantity z of the real-time state of the turbine runner, wherein the quantitative index quantity z is expressed as:
wherein F' represents the eigenvalue vector,
representing the standard value vector of the characteristic,/representing the division of the corresponding element of the vector; the larger the z value is, the more serious the current state of the turbine runner deviates from the healthy state.
The invention has the further beneficial effects that: the value range difference of each element in the signal characteristic value vector is large, the calculation mode can reduce the influence caused by the difference, and the purpose of taking the relative difference is to normalize the variation of each characteristic; the purpose of taking the two norms is to synthesize the six components to obtain a comprehensive index quantity. Therefore, the invention normalizes the difference value of each element of the signal characteristic value vector, and ensures that each signal characteristic with different value ranges has equivalent effect on the real-time state quantization index quantity of the rotating wheel.
Further, the evaluating the real-time state of the turbine runner by comparing the characteristic value vector with the characteristic standard value vector further comprises:
the state of the water turbine under the current index amount z is determined based on the established degradation levels divided according to the index amount z.
The invention also provides a real-time state evaluation system of the turbine runner, which comprises: the system comprises a sensor array, a signal acquisition unit, a data communication module and a comprehensive analysis module;
the signal acquisition unit is used for acquiring various sensor monitoring signals of the water turbine runner under the assistance of the sensor array; the data communication module is used for integrating the data acquired by the signal acquisition unit uploaded by the data communication unit and the water head and active power data of the hydraulic turbine unit acquired from the on-line monitoring system of the power station unit and transmitting the data to the comprehensive analysis module; the comprehensive analysis module is used for executing the method for evaluating the real-time state of the turbine runner to obtain an evaluation result.
Further, still include: the device comprises a storage query module and a front-end interaction module;
the data communication module is also used for respectively transmitting the integrated data to the storage query module and the front-end interaction module; the comprehensive analysis module is also used for transmitting the evaluation result to the storage query module and the front-end interaction module;
the storage query module is used for storing historical acquisition data and historical evaluation results and providing a historical query function for a user; the front-end interaction module is used for displaying the real-time data, the historical evaluation result and the historical acquisition data and providing an interaction interface for a user.
The present invention also provides a computer-readable storage medium comprising a stored computer program, wherein when the computer program is executed by a processor, the computer program controls a device on which the storage medium is located to execute a real-time state estimation method of a turbine runner as described above.
Drawings
Fig. 1 is a flow chart of a method for estimating a real-time state of a turbine runner according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a method for evaluating a real-time state of a rotating wheel of a water turbine under complex working conditions according to an embodiment of the present invention;
FIG. 3 is a schematic view of a state degradation curve of a turbine runner according to an embodiment of the present invention;
FIG. 4 is a block diagram of an overall structure of a real-time state evaluation system for a water turbine runner under complex conditions according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an arrangement of a local-level signal acquisition subsystem of the real-time state evaluation system of the water turbine runner under the complex working conditions according to the embodiment of the invention;
FIG. 6 is a schematic diagram of a plant-level comprehensive analysis subsystem arrangement of a water turbine runner real-time state evaluation system under complex conditions according to an embodiment of the present invention;
the same reference numbers will be used throughout the drawings to refer to the same or like elements or structures, wherein:
the system comprises a concrete wall 1 of a waterwheel room, a guide vane crank arm 2, a servomotor 3, vibration and acoustic emission sensors 4, 5, 6 and 7, a sensor matching cable 8, an on-site cabinet 9, a server screen cabinet 10, a display 11 and a high-performance data analysis workstation 12.
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. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Example one
A real-time status evaluation method for a turbine runner, as shown in fig. 1, includes:
acquiring a water head and active power of a water turbine unit and various sensor monitoring signals of a water turbine runner in real time, extracting various characteristic values in each sensor monitoring signal, wherein all the characteristic values jointly form a characteristic value vector;
respectively adopting an established nonlinear mapping relation model between the working condition and the standard values of the various characteristics to obtain the standard values of the various characteristics under the working condition corresponding to the water head and the active power, wherein all the standard values jointly form a characteristic standard value vector;
and evaluating the real-time state of the turbine runner by comparing the characteristic value vector with the characteristic standard value vector.
Preferably, the nonlinear mapping relationship model is obtained by the following establishing method:
s1, continuously acquiring discrete point data in the whole stable operation period of the water turbine, wherein each discrete point data comprises working condition parameters containing a water head and active power at a certain moment and monitoring signals of the various sensors;
the method comprises the steps of collecting vibration signals V (t) and acoustic emission signals A (t) at four quadrant points of a top cover of the hydraulic turbine set in a long-term normal running state, corresponding working condition parameters such as a water head H and active power P, and ensuring that the working conditions recorded in the period are distributed in all possible working condition ranges of the power station as densely as possible. The specific time length is judged according to the actual condition of the power station, and generally, six months for stable and normal operation of the unit after production can be taken.
S2, extracting various characteristic values in each sensor monitoring signal in each discrete point data, and taking the characteristic values as a group of characteristic standard values corresponding to the discrete point data in the state of health of the turbine runner;
the vibration signal characteristic values may include: root mean square value F1, peak-to-peak value F2, peak factor F3, kurtosis factor F4, etc., acoustic emission signal feature values may include: event count F5, ring count F6, etc.
Wherein, F 1 ～F 4 The calculation methods of (A) are respectively as follows:
wherein n is the number of signal sample points, x i Is the ith signal data;
F 2 ＝x max -x min
in the formula, x max And x min Maximum and minimum values in the signal samples, respectively;
in the formula (I), the compound is shown in the specification,
represents the mean of the signal samples;
F 5 and F 6 The calculation method comprises the following steps: dividing an acoustic emission original signal X (t) into N sections, and connecting maximum amplitude points in each section to obtain an upper envelope line Y (t); upper envelope Y (t)Every time the threshold value is exceeded, recording the acoustic emission event as one time, and the number of the acoustic emission events is the event count F 5 (ii) a The number of pulse peaks in the original signal corresponding to each acoustic emission event is called the ringing count in the infrasonic emission event, and the sum of the ringing counts in all acoustic emission events is the ringing count F in the acoustic emission original signal 6 。
And S3, establishing a nonlinear mapping relation model between the characteristic standard value of each characteristic and the water head and active power, wherein the adopted data set is composed of data points (H, P, F) corresponding to each discrete point data, wherein H and P represent the water head and the active power in each discrete point data, and F represents the characteristic standard value of the characteristic corresponding to the discrete point data. Preferably, in S3, a gaussian process regression is adopted to establish a nonlinear mapping relationship model between the standard value probability density distribution of the characteristic and the head and active power; simplifying the nonlinear mapping relation model between the standard value probability density distribution of the characteristics and the water head and the active power into a nonlinear mapping relation model between the mean value of the standard value probability density distribution of the characteristics and the water head and the active power, and taking the nonlinear mapping relation model as a final nonlinear mapping relation model.
Since the vibration and acoustic emission signals have volatility and randomness, the examination of single observation values thereof easily causes large errors. In the embodiment, probability density distribution N of characteristic standard values of monitoring signals of the health states of the rotating wheels of each water turbine is respectively established based on Gaussian process regression H,P (μ,σ 2 ) A nonlinear mapping relation model between the unit operating parameters (H, P) (the input of the model is the operating parameters (H, P), and the output is the mean value mu and the variance sigma of normal distribution 2 ) And taking the mean value of the probability density distribution as the characteristic standard value of the monitoring signal of the health state of the turbine runner under the working condition (H, P)
(since the mean of the probability density distribution represents the observed value with the highest probability of occurrence under this prior distribution), the mapping is written as:
where i represents the type of signal feature.
And recording the health state standard value vector of the monitoring signal characteristic value as:
for each feature, the gaussian process regression model is specifically established as follows:
3.1) the prior representation of the assumed Gaussian process is:
f(X)＝Y～N(μ f ,K ff )
in the formula, mu f Represents the mean function, K ff κ (X, X), where κ represents a covariance function;
3.2) given N sets of observation samples, record:
X*＝[(H (1) ,P (1) ),(H (2) ,P (2) ),…(H (N) ,P (N) )]
3.3)Y * obey a joint gaussian distribution with Y:
in the formula, K fy ＝κ(X,X * )，K yy ＝κ(X * ,X * )；
3.4) solving a Gaussian process regression model:
obtaining the non-linear mapping corresponding to various characteristicsAfter the relation model is established, sensor monitoring signals such as vibration signals, acoustic emission signals and the like at four quadrant points of the top cover can be collected in real time, corresponding working condition parameters such as a water head H ' and active power P ' can be collected, and a characteristic value vector F ' (F) of the real-time sensor monitoring signals is extracted 1 ′,F 2 ′,…,F 6 ') and calculating the health state standard vector of the characteristic value of the turbine runner monitoring signal under the working condition according to the working condition parameters:
(5) the real-time state quantization index quantity z of the turbine runner is defined by the following formula:
wherein, the corresponding elements of the/expression vector are divided, and the larger the z value is, the more serious the current state of the turbine runner deviates from the health standard state is.
The flow from model building to the actual evaluation application is shown in fig. 2.
Each power station can reasonably set the grade interval of the turbine runner real-time state quantization index quantity z according to the actual operation condition. And calculating a curve of the quantitative index quantity z of the state of the water turbine runner along with the change of time according to historical monitoring data accumulated during long-term operation, judging the real-time degradation level of the water turbine runner at the moment, observing the development trend of the state degradation, and reasonably formulating a targeted operation and maintenance strategy according to the trend.
As shown in FIG. 3, each power station can reasonably divide the z value interval corresponding to the deterioration grade of the runner state according to the actual operation condition of the power station. Under different degradation grade states, each power station can make corresponding operation and maintenance strategies according to the actual conditions of the power station. Because the actual conditions of each power station and the basic conditions of the unit are different, the threshold value interval of the z value is difficult to be uniformly specified, but the abstract turbine runner state is quantized into the index z in the embodiment, so that operation and maintenance personnel can quantitatively evaluate the unit runner state according to the actual conditions of the operation and maintenance personnel.
Example two
A real-time state evaluation system for a turbine runner, comprising: the system comprises a sensor array, a signal acquisition unit, a data communication module and a comprehensive analysis module;
the signal acquisition unit is used for acquiring various sensor monitoring signals of the water turbine runner under the assistance of the sensor array; the data communication module is used for integrating the data acquired by the signal acquisition unit uploaded by the data communication unit and the water head and active power data of the hydraulic turbine unit acquired from the on-line monitoring system of the power station unit and transmitting the data to the comprehensive analysis module; the comprehensive analysis module is used for executing the method for evaluating the real-time state of the turbine runner according to the first embodiment to obtain an evaluation result.
In general, the real-time state evaluation system for the turbine runner comprises an on-site level signal acquisition subsystem and a plant-station level comprehensive analysis subsystem, as shown in fig. 4.
The in-situ level signal acquisition subsystem comprises a sensor array, a signal acquisition unit and a data communication unit. As shown in fig. 5, wherein 1 is a concrete wall of a waterwheel room, 2 is a guide vane crank arm, 3 is a servomotor, 4, 5, 6 and 7 are vibration and sound emission sensor measuring point positions, 8 is a sensor matching cable, 9 is an on-site cabinet, and a signal acquisition unit and a data communication unit are installed in the on-site cabinet.
The sensor array comprises 4 vibration sensors (such as acceleration vibration sensors with frequency response range of 5-60 kHz), 4 acoustic emission sensors (such as frequency response range of 50-1300 kHz) and matched cables thereof, and is used for collecting monitoring signals at four quadrant points of the top cover; the signal acquisition unit comprises a plurality of high-speed data acquisition cards (such as 16-bit double-channel high-speed data acquisition cards with the sampling rate of 1MHz per channel), and an acquisition case with heterogeneous sensing signal synchronous acquisition performance and a plurality of (such as 12) PCI slots, and is used for realizing the high-speed synchronous acquisition of acoustic emission signals and vibration signals and facilitating the expansion of the number of acquisition channels according to actual requirements; the data communication unit comprises a gigabit network card, an RJ-45 network interface and a super-six shielding network cable, and is used for realizing data transmission between the local signal acquisition subsystem and the plant station level comprehensive analysis subsystem through a TCP/IP protocol, sending monitoring signal data and receiving control information.
The plant station level integrated analysis subsystem is integrated in the high-performance data analysis workstation, and the arrangement mode is shown in fig. 6, wherein 10 is a server screen cabinet, 11 is a display, 12 is the high-performance data analysis workstation, and the plant station level integrated analysis subsystem is configured with an i7-8700 processor, a 32G running memory and a 2TB solid state disk. The subsystem comprises a data communication module, a comprehensive analysis module, a storage query module and a front-end interaction module; the data communication module is used for integrating sensor monitoring data uploaded from the site-level signal acquisition subsystem and the water turbine unit working condition data acquired from the power station unit online monitoring system, removing error data and abnormal data in the sensor monitoring data and the water turbine unit working condition data, and respectively transmitting the data to the comprehensive analysis module, the storage query module and the front-end interaction module; the comprehensive analysis module is used for evaluating the running state of the turbine runner in real time through the method and transmitting the evaluation result to the storage query module and the front-end interaction module; the storage query module is used for storing historical data and historical evaluation results and providing a historical query function for a user; the front-end display and interaction module is used for displaying real-time data, evaluation results and historical records and providing an interaction interface for a user.
The real-time state evaluation of the turbine runner under the complex working condition is realized, the site personnel of the power station can accurately master the state of the turbine runner, the risk and operation and maintenance cost of serious faults caused by runner degradation are reduced, technical support is provided for safe, stable and efficient operation of a unit, and the digitization and intelligence level of the hydropower station is improved.
EXAMPLE III
A computer-readable storage medium comprising a stored computer program, wherein when the computer program is executed by a processor, the computer program controls an apparatus on which the storage medium is located to perform a real-time status assessment method of a turbine runner according to an embodiment.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (7)
1. A real-time state evaluation method for a turbine runner is characterized by comprising the following steps:
acquiring a water head and active power of a water turbine unit and various sensor monitoring signals of a water turbine runner in real time, extracting various characteristic values in each sensor monitoring signal, wherein all the characteristic values jointly form a characteristic value vector;
respectively adopting an established nonlinear mapping relation model between the working condition and the standard values of various characteristics to obtain the standard values of various characteristics under the working condition corresponding to the water head and the active power, wherein all the standard values jointly form a characteristic standard value vector;
evaluating the real-time state of the turbine runner by comparing the characteristic value vector with the characteristic standard value vector;
the nonlinear mapping relation model is obtained by adopting the following establishing mode:
s1, continuously acquiring discrete point data in the whole stable operation period of the water turbine, wherein each discrete point data comprises working condition parameters containing a water head and active power at a certain moment and monitoring signals of the various sensors;
s2, extracting characteristic values of various characteristics in each sensor monitoring signal in each discrete point data, and taking the characteristic values as a group of characteristic standard values corresponding to the discrete point data in the state of health of the turbine runner;
s3, adopting Gaussian process regression to establish a nonlinear mapping relation model between the standard value probability density distribution of each characteristic and the water head and the active power, and simplifying the model into a nonlinear mapping relation model between the mean value of the standard value probability density distribution of the characteristic and the water head and the active power as the characteristic standard value of each characteristic and the water head and the active power, wherein the adopted data set comprises discrete pointsForming according to corresponding data points (H, P, F), wherein H and P represent the water head and the active power in each discrete point data, and F represents a characteristic standard value of the characteristic corresponding to the discrete point data; the establishing mode specifically comprises the following steps: based on a Gaussian process regression model, the probability density distribution mean value of the working head H, the active power P and the signal characteristic value F is directly fitted
Function of the mapping relation between
i represents the type of signal feature;
the method for evaluating the real-time state of the water turbine runner specifically comprises the following steps:
and calculating a quantitative index quantity z of the real-time state of the turbine runner, wherein the quantitative index quantity z is expressed as:
wherein F' represents the eigenvalue vector,
representing the characteristic standard value vector, dividing corresponding elements of the/representing vector; the larger the z value is, the more serious the current state of the turbine runner deviates from the healthy state.
2. The method of claim 1, wherein the monitoring signals of the plurality of sensors comprise: acoustic emission signals and vibration signals at four quadrant points of the top cover.
3. The method of claim 2, wherein the plurality of characteristic values of the vibration signal comprise: a root mean square value, a peak-to-peak value, a peak factor, and a kurtosis factor; the plurality of characteristic values in the acoustic emission signal include: event counting and ring counting.
4. The method for evaluating the real-time status of a turbine runner according to any one of claims 1 to 3, wherein the evaluating the real-time status of the turbine runner by comparing the characteristic value vector with the characteristic standard value vector further comprises:
the state of the water turbine under the current index amount z is determined based on the established degradation levels divided according to the index amount z.
5. A real-time status assessment system for a turbine runner, comprising: the system comprises a sensor array, a signal acquisition unit, a data communication module and a comprehensive analysis module;
the signal acquisition unit is used for acquiring various sensor monitoring signals of the water turbine runner under the assistance of the sensor array; the data communication module is used for integrating the data acquired by the signal acquisition unit uploaded by the data communication unit and the water head and active power data of the hydraulic turbine unit acquired from the on-line monitoring system of the power station unit and transmitting the data to the comprehensive analysis module; the comprehensive analysis module is used for executing the method for evaluating the real-time state of the turbine runner as claimed in any one of claims 1 to 4 to obtain an evaluation result.
6. The system of claim 5, further comprising: the device comprises a storage query module and a front-end interaction module;
the data communication module is also used for respectively transmitting the integrated data to the storage query module and the front-end interaction module; the comprehensive analysis module is also used for transmitting the evaluation result to the storage query module and the front-end interaction module;
the storage query module is used for storing historical acquisition data and historical evaluation results and providing a historical query function for a user; the front-end interaction module is used for displaying the real-time data, the historical evaluation result and the historical acquisition data and providing an interaction interface for a user.
7. A computer-readable storage medium, comprising a stored computer program, wherein when the computer program is executed by a processor, the computer program controls a device on which the storage medium is located to perform a real-time status assessment method of a turbine runner according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110311462.5A CN113027658B (en) | 2021-03-24 | 2021-03-24 | Real-time state evaluation method for water turbine runner and application thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110311462.5A CN113027658B (en) | 2021-03-24 | 2021-03-24 | Real-time state evaluation method for water turbine runner and application thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113027658A CN113027658A (en) | 2021-06-25 |
| CN113027658B true CN113027658B (en) | 2022-08-02 |
Family
ID=76473092
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110311462.5A Active CN113027658B (en) | 2021-03-24 | 2021-03-24 | Real-time state evaluation method for water turbine runner and application thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113027658B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114397840A (en) * | 2021-12-30 | 2022-04-26 | 腾安电子科技(江苏)有限公司 | Intelligent control method and device for water turbine |
| CN115048746B (en) * | 2022-07-05 | 2024-07-26 | 西安理工大学 | Method for calculating vibration probability density curve of runner of full-through-flow turbine |
| CN116881658B (en) * | 2023-07-12 | 2024-01-26 | 南方电网调峰调频发电有限公司检修试验分公司 | Intelligent state evaluation method and system for hydroelectric generating set |
| CN117668739B (en) * | 2024-02-01 | 2024-04-26 | 中国水利水电科学研究院 | Hydroelectric generating set fault diagnosis method and system based on characteristic samples |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111077876A (en) * | 2019-12-11 | 2020-04-28 | 湖南大唐先一科技有限公司 | Power station equipment state intelligent evaluation and early warning method, device and system |
| CN111365158A (en) * | 2020-03-02 | 2020-07-03 | 东方电气集团东方电机有限公司 | Real-time state evaluation and life cycle management prediction system for water turbine runner |
| CN111522858A (en) * | 2020-03-16 | 2020-08-11 | 国家电网有限公司 | Multi-dimensional state vector-based pumping unit performance degradation early warning method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6076751B2 (en) * | 2013-01-22 | 2017-02-08 | 株式会社日立製作所 | Abnormality diagnosis method and apparatus |
| CN107341349B (en) * | 2017-06-30 | 2020-04-14 | 上海交通大学 | Method and system for evaluating health of fan, memory and controller |
| CN208950762U (en) * | 2018-05-21 | 2019-06-07 | 大唐水电科学技术研究院有限公司 | A kind of hydraulic turbine health state evaluation system |
| CN110080921B (en) * | 2019-04-11 | 2020-07-10 | 华中科技大学 | Online monitoring and evaluating method and system for main water inlet valve of pumped storage power station |
| CN110185575A (en) * | 2019-05-17 | 2019-08-30 | 天津天发总厂机电设备有限公司 | Remote monitoring method for equipment state of water turbine generator set |
| CN111486043A (en) * | 2020-04-24 | 2020-08-04 | 华能四川水电有限公司 | Lower rack fault diagnosis method based on hydro-turbo generator set runout data |
-
2021
- 2021-03-24 CN CN202110311462.5A patent/CN113027658B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111077876A (en) * | 2019-12-11 | 2020-04-28 | 湖南大唐先一科技有限公司 | Power station equipment state intelligent evaluation and early warning method, device and system |
| CN111365158A (en) * | 2020-03-02 | 2020-07-03 | 东方电气集团东方电机有限公司 | Real-time state evaluation and life cycle management prediction system for water turbine runner |
| CN111522858A (en) * | 2020-03-16 | 2020-08-11 | 国家电网有限公司 | Multi-dimensional state vector-based pumping unit performance degradation early warning method |
Non-Patent Citations (2)
| Title |
|---|
| BP神经网络在水轮机综合特性曲线中的应用;刘爱丽等;《科技信息(科学教研)》;20080501(第13期);全文 * |
| 基于三维曲面的抽水蓄能电站机组故障预警模型;安学利等;《水力发电》;20130112(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113027658A (en) | 2021-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113027658B (en) | Real-time state evaluation method for water turbine runner and application thereof | |
| CN104571079A (en) | Wireless long-distance fault diagnosis system based on multiple-sensor information fusion | |
| CN112084684B (en) | Bridge health visual monitoring system based on Internet of things | |
| CN116739384A (en) | Mining equipment operation management system based on 5G wireless communication | |
| CN114580666A (en) | Multi-mode fusion motor intelligent maintenance system | |
| CN105571638A (en) | Machinery device fault combination prediction system and method | |
| CN117454114B (en) | Subway tunnel tunneling blasting vibration safety monitoring device based on multi-point location distribution | |
| CN112267972A (en) | Intelligent judgment method for abnormity of power curve of wind turbine generator | |
| CN111639852B (en) | Real-time evaluation method and system for vibration state of hydroelectric generating set based on wavelet singular value | |
| CN118133542A (en) | Electric power safety control system based on digital twin | |
| CN107340740B (en) | Unattended intelligent Roadbed Deformation parameter automated collection systems and signal processing method | |
| CN117350447A (en) | Multisource heterogeneous power data fusion algorithm applicable to power grid | |
| CN109886538B (en) | Railway signal equipment quality evaluation method and device based on dynamic monitoring data | |
| CN116433034A (en) | Virtual power plant operation risk detection method based on machine learning | |
| CN112465404B (en) | Method and system for evaluating running state of photovoltaic power generation system for multiple time windows | |
| CN114779010A (en) | Fault traveling wave detection method based on symmetric differential energy operator and neural network | |
| CN114048950A (en) | Health degree assessment method and system for wind turbine generator | |
| CN111667020A (en) | Fault early warning method for nuclear power rotating machinery | |
| CN216595337U (en) | Real-time online monitoring device for state of rotating wheel of water turbine | |
| CN2793729Y (en) | Long-range on-line analyzing inspector for electric conductivity, pH value and oxygen dissolving value | |
| CN117596190B (en) | System and method for testing secondary signal transmission delay characteristics of transformer substation | |
| CN110766325A (en) | Financial big data intelligent analysis's model development instrument | |
| CN109102001A (en) | A kind of gene improve the rotor on-line fault diagnosis method of neural network | |
| CN118367680B (en) | Intelligent driving control system and method for monitoring power equipment | |
| CN115765172A (en) | New energy station performance online monitoring method and system based on monitoring substation |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |