CN116943329A - Wind turbine generator gearbox filter blockage early warning diagnosis method and system - Google Patents
Wind turbine generator gearbox filter blockage early warning diagnosis method and system Download PDFInfo
- Publication number
- CN116943329A CN116943329A CN202310927092.7A CN202310927092A CN116943329A CN 116943329 A CN116943329 A CN 116943329A CN 202310927092 A CN202310927092 A CN 202310927092A CN 116943329 A CN116943329 A CN 116943329A
- Authority
- CN
- China
- Prior art keywords
- filter
- real
- time
- differential pressure
- data set
- 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
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000003745 diagnosis Methods 0.000 title claims description 10
- 230000002159 abnormal effect Effects 0.000 claims abstract description 10
- 230000005856 abnormality Effects 0.000 claims abstract description 6
- 238000001914 filtration Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 12
- 238000012417 linear regression Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 8
- 230000000903 blocking effect Effects 0.000 abstract description 6
- 239000003921 oil Substances 0.000 description 44
- 230000001419 dependent effect Effects 0.000 description 17
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 14
- 239000010687 lubricating oil Substances 0.000 description 11
- 238000001816 cooling Methods 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010977 unit operation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D35/00—Filtering devices having features not specifically covered by groups B01D24/00 - B01D33/00, or for applications not specifically covered by groups B01D24/00 - B01D33/00; Auxiliary devices for filtration; Filter housing constructions
- B01D35/14—Safety devices specially adapted for filtration; Devices for indicating clogging
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D37/00—Processes of filtration
- B01D37/04—Controlling the filtration
- B01D37/046—Controlling the filtration by pressure measuring
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D37/00—Processes of filtration
- B01D37/04—Controlling the filtration
- B01D37/048—Controlling the filtration by temperature measuring
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16N—LUBRICATING
- F16N39/00—Arrangements for conditioning of lubricants in the lubricating system
- F16N39/06—Arrangements for conditioning of lubricants in the lubricating system by filtration
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/08—Investigating permeability, pore-volume, or surface area of porous materials
- G01N15/082—Investigating permeability by forcing a fluid through a sample
- G01N15/0826—Investigating permeability by forcing a fluid through a sample and measuring fluid flow rate, i.e. permeation rate or pressure change
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/08—Investigating permeability, pore-volume, or surface area of porous materials
- G01N2015/084—Testing filters
-
- 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
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Chemical Kinetics & Catalysis (AREA)
- General Engineering & Computer Science (AREA)
- Computational Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pure & Applied Mathematics (AREA)
- Databases & Information Systems (AREA)
- Fluid Mechanics (AREA)
- Probability & Statistics with Applications (AREA)
- Software Systems (AREA)
- Operations Research (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Mechanical Engineering (AREA)
- Algebra (AREA)
- Dispersion Chemistry (AREA)
- Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides a method and a system for early warning and diagnosing the blocking of a gear box filter of a wind turbine generator, which comprise the following steps: step 1, acquiring historical data of a wind turbine to be tested; step 2, calculating a filter differential pressure threshold; step 3, acquiring real-time operation data of the wind turbine to be tested; step 4, calculating a real-time residual error mean value of the filter pressure difference; step 5, judging whether the filter of the gearbox of the wind turbine to be tested is abnormal or not according to the obtained real-time residual error average value and the obtained filter differential pressure threshold value; the invention can discover abnormality in advance, reduce the damage probability of the filter element and reduce the spare part consumption.
Description
Technical Field
The invention belongs to the technical field of wind turbine generator gear box filter blockage early warning, and particularly relates to a wind turbine generator gear box filter blockage early warning diagnosis method and system.
Background
The intelligent operation and maintenance of the wind farm becomes a development trend in the industry, and the potential fault risk is found in advance by the unit operation data in the industry, which is a current industry hotspot.
The gearbox is also very important as an important part of a transmission device of the wind turbine, the failure of a gearbox filter often leads to continuous increase of the temperature of the gearbox lubricating oil, and finally, the shutdown and the loss of the generated energy are caused, and the serious blockage of the gearbox filter can also lead to the damage of a gearbox oil pump.
At present, no reliable early warning mode exists for a gearbox filter in the industry, manual regular inspection is usually carried out, and the filter element is damaged to different degrees when abnormality is found.
Disclosure of Invention
The invention aims to provide a method for diagnosing blockage early warning of a wind turbine generator gear box filter, which solves the problem that hysteresis exists in the existing gear box fault early warning.
In order to achieve the above purpose, the invention adopts the following technical scheme:
the invention provides a wind turbine generator gearbox filter blockage early warning diagnosis method, which comprises the following steps:
step 1, acquiring historical data of a wind turbine to be tested, wherein the historical data comprise gearbox outlet oil temperature, gearbox filter inlet pressure, gearbox filter outlet pressure and filter pressure difference;
step 2, dividing the obtained historical data to obtain four data sets;
step 3, respectively modeling the four obtained data sets, and calculating to obtain a filter differential pressure threshold value corresponding to each data set according to the obtained model;
step 4, acquiring real-time operation data of the wind turbine to be tested, wherein the real-time operation data comprise real-time outlet oil temperature of a gear box, real-time inlet pressure of a gear box filter and real-time outlet pressure of the gear box filter;
step 5, dividing the obtained real-time operation data according to the step 2 to obtain four data sets;
step 6, respectively calculating the real-time residual error mean value of the filter differential pressure respectively corresponding to the four data sets of the real-time operation data by using the model corresponding to each data set obtained in the step 3;
and 7, judging whether the filter of the gearbox of the wind turbine to be tested is abnormal or not according to the real-time residual error average value obtained in the step 6 and the filter differential pressure threshold value obtained in the step 3.
In the preferred embodiment, in step 2, the obtained historical data is divided to obtain four data sets, and the specific method is as follows:
according to the fine filtration bypass pressure valve threshold value and the thermal sensitive valve threshold value of the gear box filter in the wind turbine to be tested, combining the oil temperature of the gear box outlet and the inlet pressure of the gear box filter, and dividing historical data into four data sets.
The preferred embodiment is that the historical data is divided into four data sets, and the specific method is as follows:
extracting data from the historical data that the inlet pressure of a gear box filter is smaller than a fine filtration bypass pressure valve threshold value and the outlet oil temperature of the gear box is smaller than a thermosensitive valve threshold value, and forming a first historical data set;
extracting data from the historical data that the inlet pressure of the gear box filter is larger than a fine filtration bypass pressure valve threshold value and the outlet oil temperature of the gear box is smaller than a thermosensitive valve threshold value, and forming a second historical data set;
extracting data of which the inlet pressure of a gear box filter is larger than a fine filtration bypass pressure valve threshold value and the outlet oil temperature of the gear box is larger than a thermosensitive valve threshold value from historical data to form a third historical data set;
and extracting data that the inlet pressure of the gear box filter is larger than the threshold value of the fine filtration bypass pressure valve and the oil temperature of the gear box outlet is larger than the threshold value of the thermosensitive valve from the historical data to form a fourth historical data set.
In the preferred embodiment, in the step 3, the obtained four data sets are respectively modeled, and the specific method is as follows:
performing unitary linear regression fitting on all gearbox filter inlet pressures and all filter differential pressures in the first historical data set and the second historical data set respectively to obtain a first model and a second model;
respectively carrying out normalization processing on all the oil temperatures of the outlet of the gear box, all the inlet pressures of the filter of the gear box and all the differential pressures of the filter of the gear box in the third historical data set and the fourth historical data set to obtain normalized oil temperatures of the outlet of the gear box, inlet pressures of the filter of the gear box and differential pressures of the filter of the gear box;
and performing multiple linear regression fitting on the oil temperature of the outlet of the gear box, the inlet pressure of the filter of the gear box and the pressure difference of the filter after normalization processing in the third historical data set and the fourth historical data set respectively to obtain a third model and a fourth model.
In the preferred embodiment, in step 3, a filter differential pressure threshold value corresponding to each data set is calculated according to the obtained model, and the specific method is as follows:
the calculation method of the filter differential pressure threshold corresponding to the first historical data set is as follows:
combining the inlet pressure of each gearbox filter in the first historical data set with the first model, and calculating to obtain an estimated value of the filter differential pressure;
calculating to obtain residual errors corresponding to the differential pressure of each filter according to the estimated value of the differential pressure of the filter;
calculating a first filter differential pressure threshold value corresponding to the first historical data set according to the residual error corresponding to each filter differential pressure;
the calculation method of the filter differential pressure threshold corresponding to the second historical data set is as follows:
combining the inlet pressure of each gearbox filter in the second historical data set with the second model, and calculating to obtain an estimated value of the filter differential pressure;
calculating to obtain residual errors corresponding to the differential pressure of each filter according to the estimated value of the differential pressure of the filter;
calculating a second filter differential pressure threshold value corresponding to a second historical data set according to the residual error corresponding to each filter differential pressure;
the method for calculating the filter differential pressure threshold corresponding to the third historical data set is as follows:
normalizing all the inlet pressure, the outlet oil temperature and the outlet pressure of the gear box filter in the third historical data set to obtain normalized inlet pressure, outlet oil temperature and outlet pressure of the gear box filter;
combining the inlet pressure of each normalized gearbox filter, the outlet oil temperature of each normalized gearbox and a third model, and calculating to obtain an estimated value of the filter differential pressure;
calculating to obtain residual errors corresponding to the differential pressure of each filter according to the obtained estimated value of the differential pressure of the filter and the normalized outlet pressure of the filter of the gearbox;
calculating a third filter differential pressure threshold corresponding to a third historical data set according to the residual error corresponding to each filter differential pressure;
the calculation method of the filter differential pressure threshold value corresponding to the fourth historical data set is as follows:
normalizing all the inlet pressure, the outlet oil temperature and the outlet pressure of the gear box filter in the fourth historical data set to obtain normalized inlet pressure, outlet oil temperature and outlet pressure of the gear box filter;
combining the inlet pressure of each normalized gearbox filter, the outlet oil temperature of each normalized gearbox with a fourth model, and calculating to obtain an estimated value of the filter differential pressure;
calculating to obtain residual errors corresponding to the differential pressure of each filter according to the obtained estimated value of the differential pressure of the filter and the normalized outlet pressure of the filter of the gearbox;
and calculating a fourth filter differential pressure threshold corresponding to the fourth historical data set according to the residual error corresponding to each filter differential pressure.
In the preferred embodiment, in step 6, using the model corresponding to each data set obtained in step 3, a real-time residual mean value of filter differential pressures corresponding to four data sets of real-time operation data is calculated respectively, specifically:
the four data sets are a first real-time data set, a second real-time data set, a third real-time data set and a fourth real-time data set respectively;
the method for calculating the real-time residual error mean value of the filter differential pressure corresponding to the first real-time data set comprises the following steps:
combining the real-time inlet pressure of each gearbox filter in the first real-time data set with the first model, and calculating to obtain a real-time estimated value of the filter differential pressure;
calculating to obtain a real-time residual error corresponding to each filter differential pressure according to the obtained real-time estimated value of the filter differential pressure;
calculating a real-time residual error mean value of the first filter differential pressure corresponding to the first real-time data set according to the real-time residual error corresponding to each filter differential pressure;
the method for calculating the real-time residual error mean value of the filter differential pressure corresponding to the second real-time data set comprises the following steps:
combining the real-time inlet pressure of each gearbox filter in the second real-time data set with the second model, and calculating to obtain a real-time estimated value of the filter differential pressure;
calculating to obtain a real-time residual error corresponding to each filter differential pressure according to the obtained real-time estimated value of the filter differential pressure;
calculating a real-time residual error mean value of the second filter differential pressure corresponding to the second real-time data set according to the real-time residual error corresponding to each filter differential pressure;
the method for calculating the real-time residual mean value of the filter differential pressure corresponding to the third real-time data set is as follows:
normalizing each gearbox filter real-time inlet pressure, gearbox real-time outlet oil temperature and gearbox filter real-time outlet pressure in the third real-time data set to obtain normalized gearbox filter real-time inlet pressure, gearbox real-time outlet oil temperature and gearbox filter real-time outlet pressure;
combining the real-time inlet pressure of each normalized gearbox filter, the real-time outlet oil temperature of each normalized gearbox filter and a third model, and calculating to obtain a real-time estimated value of the filter differential pressure;
calculating to obtain a real-time residual error corresponding to each filter differential pressure according to the obtained real-time estimated value of the filter differential pressure and the normalized real-time outlet pressure of the gearbox filter;
and calculating the real-time residual error mean value of the third filter differential pressure corresponding to the third real-time data set according to the obtained real-time residual error corresponding to each filter differential pressure.
The method for calculating the real-time residual error mean value of the filter differential pressure corresponding to the fourth real-time data set is as follows:
normalizing each gearbox filter real-time inlet pressure, gearbox real-time outlet oil temperature and gearbox filter real-time outlet pressure in the fourth real-time data set to obtain normalized gearbox filter real-time inlet pressure, gearbox real-time outlet oil temperature and gearbox filter real-time outlet pressure;
combining the real-time inlet pressure of each normalized gearbox filter, the real-time outlet oil temperature of each normalized gearbox filter and a fourth model, and calculating to obtain a real-time estimated value of the filter differential pressure;
calculating to obtain a real-time residual error corresponding to each filter differential pressure according to the obtained real-time estimated value of the filter differential pressure and the normalized real-time outlet pressure of the gearbox filter;
and calculating the real-time residual error mean value of the fourth filter differential pressure corresponding to the fourth real-time data set according to the obtained real-time residual error corresponding to each filter differential pressure.
In the preferred embodiment, in step 7, according to the real-time residual average value obtained in step 6 and the filter differential pressure threshold value obtained in step 3, judging whether the filter of the gearbox of the wind turbine to be tested is abnormal, and the specific method is as follows:
if it isThen the coarse filter element is deemed to be blocked;
if it isAnd->The rough filter element is considered to be blocked, and the fine filter element is also possibly blocked;
if it isAnd->The fine filter element is blocked, and the coarse filter element is not blocked;
if it isAnd->The coarse filter element is not blocked and the fine filter element is not blocked.
The preferred embodiment provides a wind turbine generator system gear box filter blockage early warning diagnosis system, which comprises:
the historical data acquisition unit is used for acquiring historical data of the wind turbine to be tested, wherein the historical data comprise gearbox outlet oil temperature, gearbox filter inlet pressure, gearbox filter outlet pressure and filter pressure difference; the method comprises the steps of acquiring real-time operation data of a wind turbine to be tested, wherein the real-time operation data comprise real-time outlet oil temperature of a gear box, real-time inlet pressure of a gear box filter and real-time outlet pressure of the gear box filter;
the data dividing unit is used for dividing the obtained historical data and the real-time operation data respectively to obtain four corresponding data sets respectively;
the threshold value calculation unit is used for respectively modeling the four obtained data sets and calculating a filter differential pressure threshold value corresponding to each data set according to the obtained model;
the residual error mean value calculation unit is used for respectively calculating the real-time residual error mean value of the filter differential pressure corresponding to each data set of the real-time operation data by using the obtained model corresponding to each data set;
the abnormality judging unit is used for judging whether the filter of the gearbox of the wind turbine to be tested is abnormal according to the obtained real-time residual average value of the filter differential pressure and the obtained filter differential pressure threshold value.
Compared with the prior art, the invention has the beneficial effects that:
according to the method for diagnosing the blocking of the wind turbine generator gear box filter, provided by the invention, mathematical models under different operation conditions are built according to the working principle of a gear box oil-water cooling system and combined with historical data, then the data to be tested of the wind turbine generator is obtained, the data to be tested of the wind turbine generator are classified according to the operation conditions, calculation comparison is carried out on the data to the mathematical models built by the historical data under the operation conditions respectively, the blocking part of the gear box filter is accurately positioned, and a prediction result is given; according to the invention, the calculation and the monitoring are carried out in real time, so that the abnormality can be found in advance, the damage probability of the filter element is reduced, and the spare part consumption is reduced.
Drawings
FIG. 1 is a schematic diagram of the oil-water cooling of a gear box;
FIG. 2 is a flow chart of the early warning method of the present invention.
Detailed Description
The invention will be described in further detail with reference to the drawings and the specific examples.
As shown in fig. 1 and 2, the method for diagnosing filter blockage of a gearbox of a wind turbine generator, provided by the invention, comprises the following steps:
step 1, acquiring historical data of a wind turbine to be tested from a SCADA system, wherein the historical data comprises gearbox outlet oil temperature T and gearbox filter inlet pressure P I And gearbox filter outlet pressure P O And calculate the filter differential pressure P D 。
Step 2, screening data according to the factory value of the filter: the fine filter bypass pressure valve threshold a and the thermal valve threshold b divide the data into four data sets, specifically:
(1) From P I <a and T<b selecting P from the data of b I And P D Will P I And P D Composing the first historical dataset D 1 And for the first historical data set D 1 The data are cleaned by using a quartile method, and the data set is that lubricating oil passes through a fine filter element, a rough filter element and an oil-water exchanger.
(2) From P I >a and T<b selecting P from the data of b I And P D Will P I And P D Composing the second historical dataset D 2 And for the second historical data set D 2 The data set is that lubricating oil passes through a rough filter element and an oil-water exchanger.
(3) From P I <a and T>b selecting T, P from the data of b I And P D Will T, P I And P D Composing the third historical dataset D 3 And for a third historical data set D 3 The data set is composed of the lubricating oil passing through the fine filter element, the coarse filter element, the oil-water exchanger and the oil-water exchange bypass.
(4) From P I >a and T>b selecting T, P from the data of b I And P D Will T, P I And form a fourth historical dataset D 4 And for the fourth historical data set D 4 The data set is that lubricating oil passes through a rough filter element, an oil-water exchanger and an oil-water exchange bypass.
Step 3, for the first historical dataset D 1 The model is built as follows:
s31, for the first historical data set D 1 All of the independent variable gearbox filter inlet pressure P I And all filter differential pressure dependent variables P D Using least square method to make unitary linear regression fit to obtain first model, i.e. linear equation f 1 (x)。
S32, the first historical data set D 1 Each P of (3) I And linear equation f 1 (x) Combining, calculating to obtain the corresponding estimated value P' of the filter differential pressure dependent variable D And calculate the residual error E corresponding to each filter differential pressure dependent variable 1 ＝P D -P` D 。
S33, calculating a residual error mean valueAnd residual variance->
S34, calculating a threshold value of the first filter differential pressure corresponding to the first historical data set
Step 4, for the second historical dataset D 2 The model is built as follows:
s41, for the second historical data set D 2 All P of (3) I And all P D Using least square method to make unitary linear regression fit to obtain second model, i.e. linear equation f 2 (x)。
S42, the second historical data set D 2 Each argument P of (1) I And linear equation f 2 (x) Combining, calculating to obtain the corresponding estimated value P' of the filter differential pressure dependent variable D And calculate the residual error E corresponding to each filter differential pressure dependent variable 2 ＝P D -P` D 。
S43, calculating residual error mean valueAnd residual variance->
S44, calculating a threshold value of the second filter differential pressure corresponding to the second historical data set
Step 5, for the third historical dataset D 3 Modeling, e.g.The following steps:
s51, for the third historical data set D 3 T, P of all of (3) I And P D Processing by min-max normalization method to obtain normalized T ', P' I "and P D And record (T) 3max |T 3min )、
Normalization formula:
s52, regarding all T', P after normalization I "and P D "multiple linear regression fitting Using least squares method, resulting in a third model, the linear equation f 3 (x)。
S53, using normalized each T' and P I ", combine f 3 (x) Calculating to obtain the corresponding estimated value P' of the filter differential pressure dependent variable D And calculate the residual error E corresponding to each filter differential pressure dependent variable 3 ＝P D ″-P` D 。
S54, calculating a residual error mean valueAnd residual variance->
S55, calculating a threshold value of the third filter differential pressure corresponding to the third historical data set
Step 6, for the fourth historical dataset D 4 The model is built as follows:
s61, for the fourth historical data set D 4 T, P of all of (3) I And P D Processing by min-max normalization method to obtain normalized T ', P' I "and P D And record (T) 4max |T 4min )、
Normalization formula:
s62, regarding all T', P after normalization I "and P D "multiple linear regression fitting Using least squares method, resulting in a fourth model, the linear equation f 4 (x)。
S63, using normalized each T' and P I ", combine f 4 (x) Calculating to obtain the corresponding estimated value P' of the filter differential pressure dependent variable D And calculate the residual error E corresponding to each filter differential pressure dependent variable 4 ＝P D ″-P` D 。
S64, calculating residual error mean valueAnd residual variance->
S65, calculating a threshold value of a fourth filter differential pressure corresponding to the fourth data set
Step 7, acquiring real-time operation data of the wind turbine to be tested from the SCADA system, wherein the real-time operation data comprise real-time oil temperature t at the outlet of the gear box and real-time inlet pressure p of a filter of the gear box i Real-time outlet pressure p of gearbox filter o And calculate the filter real-time differential pressure p d 。
Step 8, screening data according to the factory value of the filter: the fine filter bypass pressure valve threshold a and the thermal valve threshold b divide the data into four data sets:
(1) From p i <a and t<b selecting p from the data of b i And p d Will p i And p d Composing the first real-time data set d 1 。
(2) From p i >a and t<b selecting p from the data of b i And p d Will p i And p d Composing the second real-time data set d 2 。
(3) From p i <a and t>b selecting t and p from the data of b i And p d T, p i And p d Composing the third real-time data set d 3 And according to step 5 (T 3max |T 3min )、Data cleaning and normalization processing are carried out to obtain normalized t ', p' i "and p d ″。
(4) From p i >a and t>b selecting t and p from the data of b i And p d T, p i And p d Composing the fourth dataset d 4 And according to step 6 (T 4max |T 4min )、Data cleaning and normalization processing are carried out to obtain normalized t ', p' i "and p d ″。
Step 9, for the first real-time data set d 1 With a first real-time data set d 1 Each p of (3) i And f in step 3 1 (x) Combining, calculating to obtain the corresponding filter differential pressure dependent variable real-time estimated value p d And calculates the real-time residual error e corresponding to each filter differential pressure dependent variable 1 ＝p d -p` d And a real-time residual mean of the first filter differential pressure corresponding to the first real-time data set
Step 10, for a second real-time dataset d 2 With a second real-time data set d 2 Each p of (3) i And f in step 4 2 (x) Combining, calculating to obtain correspondingFilter differential pressure dependent variable real-time estimated value p d And calculates the real-time residual error e corresponding to each filter differential pressure dependent variable 2 ＝p d -p` d And a real-time residual mean of a second filter differential pressure corresponding to the second real-time data set
Step 11, for a third real-time dataset d 3 With each normalized t ", p i "and f in step 5 3 (x) Combining, calculating to obtain the corresponding filter differential pressure dependent variable real-time estimated value p d And calculates the real-time residual error e corresponding to each filter differential pressure dependent variable 3 ＝p d ″-p` d And a real-time residual mean value of the third filter differential pressure corresponding to the third real-time data set
Step 12, for a fourth real-time dataset d 4 With each normalized t ", p i "and f in step 6 4 (x) Combining, calculating to obtain the corresponding filter differential pressure dependent variable real-time estimated value p d And calculates the real-time residual error e corresponding to each filter differential pressure dependent variable 4 ＝p d ″-p` d And a fourth filter differential pressure real-time residual mean value corresponding to the fourth real-time data set
Step 13, early warning judgment:
(1) Due to d 2 、d 4 In the data set, the gearbox oil passes through the coarse filter element, so ifThen the coarse filter element is deemed to be blocked;
(2) Due to d 2 、d 4 In the data set, the gearbox oil passes through the coarse filter element ifThen the coarse filter element is deemed to be blocked; d, d 1 、d 3 In the data set, the gear box oil passes through the coarse filter element and the fine filter element, if at the same timeThe fine filter cartridge may also clog;
(3) Due to d 2 、d 4 In the data set, the gearbox oil passes through the coarse filter element ifThe coarse filter element is not blocked; d, d 1 、d 3 In the data set, the gear box oil passes through the coarse filter element and the fine filter element, if at the same timeThe fine filter element is blocked;
(4) Due to d 2 、d 4 In the data set, the gearbox oil passes through the coarse filter element ifThe coarse filter element is not blocked; d, d 1 、d 3 In the data set, the gear box oil passes through the coarse filter element and the fine filter element, if at the same timeThe fine filter cartridge is not clogged.
In the present invention, the reliable historical data is used for the local linear data set D 1 、D 2 、D 3 、D 4 Performing linear regression fitting to obtain a linear equation f 1 (x)、f 2 (x)、f 3 (x)、f 4 (x) And residual errors, wherein the residual errors of the linear regression equation obey normal distribution, and the boundary value with the probability of 99.7% according to the residual error distribution interval is [ mean-3-square, mean+3-square]The filter element blockage causes the filter differential pressure to increase, so the upper boundary value [ mean+3 ] variance is taken]Calculating residual Th 1 、Th 2 、Th 3 、Th 4 The method comprises the steps of carrying out a first treatment on the surface of the For data set d running in real time 1 、d 2 、d 3 、d 4 According to filter inlet pressure and regression equation f 1 (x)、f 2 (x)、f 3 (x)、f 4 (x) Calculating estimated values of all points, and calculating residual error mean value by using actual value-estimated value The residual was analyzed as follows:
results 1:
at d 2 、d 4 In the data set, the gear box oil only passes through the rough filter element, ifI.e. the residual error exceeds the upper boundary value with the probability of 99.7% of the residual error distribution interval, i.e. the actual filter differential pressure is too high under the same filter inlet pressure, and exceeds the reasonable range of the filter differential pressure in the historical sample data, the rough filter element is considered to be blocked.
Results 2:
at d 1 、d 3 In the data set, the gear box oil passes through a coarse filter element and a fine filter element, ifI.e. the residual error exceeds the upper boundary value with the probability of 99.7% of the residual error distribution interval, i.e. the actual filter pressure difference is too high under the same filter inlet pressure and exceeds the reasonable range of the filter pressure difference in the historical sample data, because the gear box oil passes through the rough filter element and the fine filter element simultaneously, the rough filter and the fine filter are possibly blocked, and the results are combined at the moment1, if the result 1 is true, the blocking of the rough filter element is confirmed, the blocking of the fine filter element is possible, and if the result 1 is false, the blocking of the fine filter element is confirmed.
Results 3:and->
Data d to be measured 1 、d 2 、d 3 、d 4 In the method, the residual errors are in the boundary range with the probability of 99.7% in the distribution interval, the pressure difference of the filter is considered to be in a reasonable range, and the rough filter element and the fine filter element are considered to be unblocked.
The preferred embodiment provides a wind turbine generator system gear box filter blockage early warning diagnosis system, which comprises:
the historical data acquisition unit is used for acquiring historical data of the wind turbine to be tested, wherein the historical data comprise gearbox outlet oil temperature, gearbox filter inlet pressure, gearbox filter outlet pressure and filter pressure difference; the method comprises the steps of acquiring real-time operation data of a wind turbine to be tested, wherein the real-time operation data comprise real-time outlet oil temperature of a gear box, real-time inlet pressure of a gear box filter and real-time outlet pressure of the gear box filter;
the data dividing unit is used for dividing the obtained historical data and the real-time operation data respectively to obtain four corresponding data sets respectively;
the threshold value calculation unit is used for respectively modeling the four obtained data sets and calculating a filter differential pressure threshold value corresponding to each data set according to the obtained model;
the residual error mean value calculation unit is used for respectively calculating the real-time residual error mean value of the filter differential pressure corresponding to each data set of the real-time operation data by using the obtained model corresponding to each data set;
the abnormality judging unit is used for judging whether the filter of the gearbox of the wind turbine to be tested is abnormal according to the obtained real-time residual average value of the filter differential pressure and the obtained filter differential pressure threshold value.
Because the lubricating oil cooling system of the gear box controls the lubricating oil to pass through different loops according to the oil pressure and the temperature, when in integral analysis, the pressure difference before and after the gear box filter and the oil pressure and the oil temperature before the filter have an uncertain nonlinear relation, and whether the pressure difference before and after the filter is abnormal or not cannot be effectively analyzed 1 (lubricating oil passing through the fine filtration cartridge→the coarse filtration cartridge→the oil-water exchanger), a second historical dataset D 2 (lubricating oil passing through the coarse Filter core→oil-Water exchanger), third historical dataset D 3 (lubricating oil passing through the fine filtration cartridge- & gtthe coarse filtration cartridge- & gtthe oil-water exchanger and the oil-water exchange bypass), a fourth historical dataset D 4 (lubricating oil passes through the rough filter element, the oil-water exchanger and the oil-water exchange bypass), the pressure difference between the front and the rear of the gear box filter in each data set accords with the linear relation with the oil pressure and the oil temperature before the filter, so that whether the pressure difference between the front and the rear of the filter is abnormal or not can be easily analyzed, and whether the filter element is blocked or not can be judged.
Claims (8)
1. A wind turbine generator system gear box filter blockage early warning diagnosis method is characterized by comprising the following steps:
step 1, acquiring historical data of a wind turbine to be tested, wherein the historical data comprise gearbox outlet oil temperature, gearbox filter inlet pressure, gearbox filter outlet pressure and filter pressure difference;
step 2, dividing the obtained historical data to obtain four data sets;
step 3, respectively modeling the four obtained data sets, and calculating to obtain a filter differential pressure threshold value corresponding to each data set according to the obtained model;
step 4, acquiring real-time operation data of the wind turbine to be tested, wherein the real-time operation data comprise real-time oil temperature of a gear box outlet, real-time inlet pressure of a gear box filter and real-time outlet pressure of the gear box filter;
step 5, dividing the obtained real-time operation data according to the step 2 to obtain four data sets;
step 6, respectively calculating the real-time residual error mean value of the filter differential pressure respectively corresponding to the four data sets of the real-time operation data by using the model corresponding to each data set obtained in the step 3;
and 7, judging whether the filter of the gearbox of the wind turbine to be tested is abnormal or not according to the real-time residual error average value obtained in the step 6 and the filter differential pressure threshold value obtained in the step 3.
2. The method for diagnosing blockage early warning of a wind turbine gearbox filter according to claim 1 is characterized in that in the step 2, the obtained historical data are divided to obtain four data sets, and the specific method is as follows:
according to the fine filtration bypass pressure valve threshold value and the thermal sensitive valve threshold value of the gear box filter in the wind turbine to be tested, combining the oil temperature of the gear box outlet and the inlet pressure of the gear box filter, and dividing historical data into four data sets.
3. The method for diagnosing blockage early warning of the wind turbine gearbox filter according to claim 2 is characterized in that historical data are divided into four data sets, and the method specifically comprises the following steps:
extracting data from the historical data that the inlet pressure of a gear box filter is smaller than a fine filtration bypass pressure valve threshold value and the outlet oil temperature of the gear box is smaller than a thermosensitive valve threshold value, and forming a first historical data set;
extracting data from the historical data that the inlet pressure of the gear box filter is larger than a fine filtration bypass pressure valve threshold value and the outlet oil temperature of the gear box is smaller than a thermosensitive valve threshold value, and forming a second historical data set;
extracting data of which the inlet pressure of a gear box filter is larger than a fine filtration bypass pressure valve threshold value and the outlet oil temperature of the gear box is larger than a thermosensitive valve threshold value from historical data to form a third historical data set;
and extracting data that the inlet pressure of the gear box filter is larger than the threshold value of the fine filtration bypass pressure valve and the oil temperature of the gear box outlet is larger than the threshold value of the thermosensitive valve from the historical data to form a fourth historical data set.
4. The wind turbine generator system gearbox filter blockage early warning diagnosis method according to claim 3, wherein in step 3, the obtained four data sets are respectively modeled, and the specific method is as follows:
performing unitary linear regression fitting on all gearbox filter inlet pressures and all filter differential pressures in the first historical data set and the second historical data set respectively to obtain a first model and a second model;
respectively carrying out normalization processing on all the oil temperatures of the outlet of the gear box, all the inlet pressures of the filter of the gear box and all the differential pressures of the filter of the gear box in the third historical data set and the fourth historical data set to obtain normalized oil temperatures of the outlet of the gear box, inlet pressures of the filter of the gear box and differential pressures of the filter of the gear box;
and performing multiple linear regression fitting on the oil temperature of the outlet of the gear box, the inlet pressure of the filter of the gear box and the pressure difference of the filter after normalization processing in the third historical data set and the fourth historical data set respectively to obtain a third model and a fourth model.
5. The method for diagnosing filter blockage early warning of a gearbox of a wind turbine generator according to claim 4 is characterized in that in step 3, a filter differential pressure threshold value corresponding to each data set is calculated according to an obtained model, wherein the method for calculating the filter differential pressure threshold value corresponding to the first historical data set is as follows:
combining the inlet pressure of each gearbox filter in the first historical data set with the first model, and calculating to obtain an estimated value of the filter differential pressure;
calculating to obtain residual errors corresponding to the differential pressure of each filter according to the estimated value of the differential pressure of the filter;
calculating a first filter differential pressure threshold value corresponding to the first historical data set according to the residual error corresponding to each filter differential pressure;
the calculation method of the filter differential pressure threshold corresponding to the second historical data set is as follows:
combining the inlet pressure of each gearbox filter in the second historical data set with the second model, and calculating to obtain an estimated value of the filter differential pressure;
calculating to obtain residual errors corresponding to the differential pressure of each filter according to the estimated value of the differential pressure of the filter;
calculating a second filter differential pressure threshold value corresponding to a second historical data set according to the residual error corresponding to each filter differential pressure;
the method for calculating the filter differential pressure threshold corresponding to the third historical data set is as follows:
normalizing all the inlet pressure, the outlet oil temperature and the outlet pressure of the gear box filter in the third historical data set to obtain normalized inlet pressure, outlet oil temperature and outlet pressure of the gear box filter;
combining the inlet pressure of each normalized gearbox filter, the outlet oil temperature of each normalized gearbox and a third model, and calculating to obtain an estimated value of the filter differential pressure;
calculating to obtain residual errors corresponding to the differential pressure of each filter according to the obtained estimated value of the differential pressure of the filter and the normalized outlet pressure of the filter of the gearbox;
calculating a third filter differential pressure threshold corresponding to a third historical data set according to the residual error corresponding to each filter differential pressure;
the calculation method of the filter differential pressure threshold value corresponding to the fourth historical data set is as follows:
normalizing all the inlet pressure, the outlet oil temperature and the outlet pressure of the gear box filter in the fourth historical data set to obtain normalized inlet pressure, outlet oil temperature and outlet pressure of the gear box filter;
combining the inlet pressure of each normalized gearbox filter, the outlet oil temperature of each normalized gearbox with a fourth model, and calculating to obtain an estimated value of the filter differential pressure;
calculating to obtain residual errors corresponding to the differential pressure of each filter according to the obtained estimated value of the differential pressure of the filter and the normalized outlet pressure of the filter of the gearbox;
and calculating a fourth filter differential pressure threshold corresponding to the fourth historical data set according to the residual error corresponding to each filter differential pressure.
6. The method for diagnosing filter blockage early warning of a gearbox of a wind turbine generator system according to claim 4 is characterized in that in step 6, a model corresponding to each data set obtained in step 3 is utilized to calculate real-time residual average values of filter differential pressures corresponding to four data sets of real-time operation data respectively, and specifically:
the four data sets are a first real-time data set, a second real-time data set, a third real-time data set and a fourth real-time data set respectively;
the method for calculating the real-time residual error mean value of the filter differential pressure corresponding to the first real-time data set comprises the following steps:
combining the real-time inlet pressure of each gearbox filter in the first real-time data set with the first model, and calculating to obtain a real-time estimated value of the filter differential pressure;
calculating to obtain a real-time residual error corresponding to each filter differential pressure according to the obtained real-time estimated value of the filter differential pressure;
calculating a real-time residual error mean value of the first filter differential pressure corresponding to the first real-time data set according to the real-time residual error corresponding to each filter differential pressure;
the method for calculating the real-time residual error mean value of the filter differential pressure corresponding to the second real-time data set comprises the following steps:
combining the real-time inlet pressure of each gearbox filter in the second real-time data set with the second model, and calculating to obtain a real-time estimated value of the filter differential pressure;
calculating to obtain a real-time residual error corresponding to each filter differential pressure according to the obtained real-time estimated value of the filter differential pressure;
calculating a real-time residual error mean value of the second filter differential pressure corresponding to the second real-time data set according to the real-time residual error corresponding to each filter differential pressure;
the method for calculating the real-time residual mean value of the filter differential pressure corresponding to the third real-time data set is as follows:
normalizing each gearbox filter real-time inlet pressure, gearbox outlet real-time oil temperature and gearbox filter real-time outlet pressure in the third real-time data set to obtain normalized gearbox filter real-time inlet pressure, gearbox outlet real-time oil temperature and gearbox filter real-time outlet pressure;
combining the real-time inlet pressure of each normalized gearbox filter, the real-time outlet oil temperature of each normalized gearbox filter and a third model, and calculating to obtain a real-time estimated value of the filter differential pressure;
calculating to obtain a real-time residual error corresponding to each filter differential pressure according to the obtained real-time estimated value of the filter differential pressure and the normalized real-time outlet pressure of the gearbox filter;
calculating a real-time residual error mean value of the third filter differential pressure corresponding to the third real-time data set according to the real-time residual error corresponding to each filter differential pressure;
the method for calculating the real-time residual error mean value of the filter differential pressure corresponding to the fourth real-time data set is as follows:
normalizing each gearbox filter real-time inlet pressure, gearbox real-time outlet oil temperature and gearbox filter real-time outlet pressure in the fourth real-time data set to obtain normalized gearbox filter real-time inlet pressure, gearbox real-time outlet oil temperature and gearbox filter real-time outlet pressure;
combining the real-time inlet pressure of each normalized gearbox filter, the real-time outlet oil temperature of each normalized gearbox filter and a fourth model, and calculating to obtain a real-time estimated value of the filter differential pressure;
calculating to obtain a real-time residual error corresponding to each filter differential pressure according to the obtained real-time estimated value of the filter differential pressure and the normalized real-time outlet pressure of the gearbox filter;
and calculating the real-time residual error mean value of the fourth filter differential pressure corresponding to the fourth real-time data set according to the obtained real-time residual error corresponding to each filter differential pressure.
7. The wind turbine generator system gear box filter blockage early warning diagnosis method according to claim 6 is characterized in that in step 7, according to the real-time residual error mean value obtained in step 6 and the filter pressure difference threshold value obtained in step 3, whether the wind turbine generator system gear box filter to be detected is abnormal is judged, and the specific method is as follows:
if it isThen the coarse filter element is deemed to be blocked;
if it isAnd->The rough filter element is considered to be blocked, and the fine filter element is also possibly blocked;
if it isAnd->The fine filter element is blocked, and the coarse filter element is not blocked;
if it isAnd->The coarse filter element is not blocked and the fine filter element is not blocked.
8. A wind turbine generator system gear box filter blockage early warning diagnosis system is characterized by comprising:
the historical data acquisition unit is used for acquiring historical data of the wind turbine to be tested, wherein the historical data comprise gearbox outlet oil temperature, gearbox filter inlet pressure, gearbox filter outlet pressure and filter pressure difference; the method comprises the steps of acquiring real-time operation data of a wind turbine to be tested, wherein the real-time operation data comprise real-time outlet oil temperature of a gear box, real-time inlet pressure of a gear box filter and real-time outlet pressure of the gear box filter;
the data dividing unit is used for dividing the obtained historical data and the real-time operation data respectively to obtain four corresponding data sets respectively;
the threshold value calculation unit is used for respectively modeling the four obtained data sets and calculating a filter differential pressure threshold value corresponding to each data set according to the obtained model;
the residual error mean value calculation unit is used for respectively calculating the real-time residual error mean value of the filter differential pressure corresponding to each data set of the real-time operation data by using the obtained model corresponding to each data set;
the abnormality judging unit is used for judging whether the filter of the gearbox of the wind turbine to be tested is abnormal according to the obtained real-time residual average value of the filter differential pressure and the obtained filter differential pressure threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310927092.7A CN116943329A (en) | 2023-07-26 | 2023-07-26 | Wind turbine generator gearbox filter blockage early warning diagnosis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310927092.7A CN116943329A (en) | 2023-07-26 | 2023-07-26 | Wind turbine generator gearbox filter blockage early warning diagnosis method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116943329A true CN116943329A (en) | 2023-10-27 |
Family
ID=88452472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310927092.7A Pending CN116943329A (en) | 2023-07-26 | 2023-07-26 | Wind turbine generator gearbox filter blockage early warning diagnosis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116943329A (en) |
-
2023
- 2023-07-26 CN CN202310927092.7A patent/CN116943329A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Schlechtingen et al. | Wind turbine condition monitoring based on SCADA data using normal behavior models. Part 2: Application examples | |
| US8027804B2 (en) | Statistical processing methods used in abnormal situation detection | |
| US7827006B2 (en) | Heat exchanger fouling detection | |
| CN110738274A (en) | nuclear power device fault diagnosis method based on data driving | |
| CN111537219B (en) | Fan gearbox performance detection and health assessment method based on temperature parameters | |
| CN106649755B (en) | Threshold value self-adaptive setting abnormity detection method for multi-dimensional real-time power transformation equipment data | |
| CN110378042B (en) | Wind turbine generator gearbox oil temperature anomaly detection method and system based on SCADA data | |
| CN112836941B (en) | Online health condition assessment method for high-pressure system of steam turbine of coal motor unit | |
| CN110097209A (en) | A kind of equipment deterioration analysis method based on parameter residual error | |
| CN110926809A (en) | Big data analysis-based wind turbine generator transmission chain fault early warning method | |
| CN112832999B (en) | Electric pump well working condition diagnosis system and method based on multi-sensor data fusion | |
| CN113176081B (en) | Historical data-based turbine blade wear monitoring method | |
| CN109538311B (en) | Real-time monitoring method for control performance of steam turbine in high-end power generation equipment | |
| CN111103137A (en) | Wind turbine gearbox fault diagnosis method based on deep neural network | |
| CN112861350B (en) | Temperature overheating defect early warning method for stator winding of water-cooled steam turbine generator | |
| CN112633614A (en) | Real-time fault degree diagnosis system and method based on feature extraction | |
| CN110969185A (en) | Equipment abnormal state detection method based on data reconstruction | |
| CN116383636A (en) | Coal mill fault early warning method based on PCA and LSTM fusion algorithm | |
| CN116609055A (en) | Method for diagnosing wind power gear box fault by using graph convolution neural network | |
| CN110513336B (en) | Method for determining offline water washing time of gas turbine of power station | |
| CN114326486B (en) | Process monitoring method based on probability slow feature analysis and elastic weight consolidation | |
| CN116943329A (en) | Wind turbine generator gearbox filter blockage early warning diagnosis method and system | |
| CN116628976A (en) | Comprehensive evaluation method for state change of hydraulic turbine unit equipment | |
| CN115186754A (en) | Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model | |
| CN115496188A (en) | Coal mill fault early warning method based on deep learning convolutional neural network |
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 |