CN108507787B - Wind power gear speed increasing box fault diagnosis test platform and method based on multi-feature fusion - Google Patents
Wind power gear speed increasing box fault diagnosis test platform and method based on multi-feature fusion Download PDFInfo
- Publication number
- CN108507787B CN108507787B CN201810689692.3A CN201810689692A CN108507787B CN 108507787 B CN108507787 B CN 108507787B CN 201810689692 A CN201810689692 A CN 201810689692A CN 108507787 B CN108507787 B CN 108507787B
- Authority
- CN
- China
- Prior art keywords
- wind power
- speed increasing
- oil
- vibration
- gear speed
- 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
- 238000012360 testing method Methods 0.000 title claims abstract description 81
- 230000004927 fusion Effects 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000003745 diagnosis Methods 0.000 title abstract description 34
- 238000001514 detection method Methods 0.000 claims abstract description 67
- 238000013210 evaluation model Methods 0.000 claims abstract description 26
- 239000008186 active pharmaceutical agent Substances 0.000 claims abstract description 22
- 230000001133 acceleration Effects 0.000 claims abstract description 19
- 230000008859 change Effects 0.000 claims abstract description 17
- 238000013135 deep learning Methods 0.000 claims abstract description 13
- 239000002245 particle Substances 0.000 claims description 55
- 238000011156 evaluation Methods 0.000 claims description 26
- 238000012549 training Methods 0.000 claims description 21
- 238000004458 analytical method Methods 0.000 claims description 16
- 238000013528 artificial neural network Methods 0.000 claims description 16
- 238000003062 neural network model Methods 0.000 claims description 15
- 238000012544 monitoring process Methods 0.000 claims description 12
- 230000009467 reduction Effects 0.000 claims description 12
- 230000008901 benefit Effects 0.000 claims description 9
- 238000012952 Resampling Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 238000005315 distribution function Methods 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000003921 particle size analysis Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 5
- 230000002093 peripheral effect Effects 0.000 abstract 2
- 239000003921 oil Substances 0.000 description 85
- 238000009434 installation Methods 0.000 description 13
- 238000005299 abrasion Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 11
- 239000010687 lubricating oil Substances 0.000 description 9
- 238000005461 lubrication Methods 0.000 description 8
- 238000000605 extraction Methods 0.000 description 6
- 230000008878 coupling Effects 0.000 description 5
- 238000010168 coupling process Methods 0.000 description 5
- 238000005859 coupling reaction Methods 0.000 description 5
- 230000006378 damage Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 5
- 239000000306 component Substances 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 239000007788 liquid Substances 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 239000003638 chemical reducing agent Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003754 machining Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000227 grinding Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000006249 magnetic particle Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/02—Gearings; Transmission mechanisms
- G01M13/021—Gearings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/257—Belief theory, e.g. Dempster-Shafer
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Wind Motors (AREA)
Abstract
The invention discloses a wind power gear speed increasing box fault diagnosis test platform and method based on multi-feature fusion, which solves the problems of single detection method, random wind load special working condition and oil change interference faced by wind power planetary gear speed increasing box fault diagnosis, and has the beneficial effects of high fault diagnosis accuracy, and the scheme is as follows: the test platform comprises: the method comprises the steps of setting a test bed base of a wind power gear speed increasing box to be tested, setting a vibration acceleration sensor and a noise signal sensor on the peripheral side of the wind power gear speed increasing box to be tested, connecting the peripheral side of the wind power gear speed increasing box to be tested with an oil information detection sensor, respectively and independently connecting each sensor with a data acquisition module, extracting vibration characteristic indexes aiming at random wind load working conditions, extracting noise characteristic indexes, extracting oil characteristic indexes aiming at oil change interference problems, and thus establishing a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory.
Description
Technical Field
The invention relates to the technical field of gearbox detection, in particular to a wind power gear speed increasing box fault diagnosis test platform and method based on multi-feature fusion.
Background
With the large-scale construction of wind power plants, wind power generation sets have entered a failure frequency period. Most wind fields are in remote zones, the distance between fans is far, the fan body is high, equipment inspection cannot be conveniently carried out, and the shutdown maintenance cost is extremely high, so that most of the current wind fields execute planned maintenance once every half year. At present, a fault early warning and diagnosis mode of a fan mainly depends on an SCADA system of a main monitoring room of a wind power plant, and all fans of the wind power plant are monitored and controlled in a centralized manner according to a large amount of collected data; however, the early failure of the wind driven generator cannot be accurately diagnosed, and a more intelligent and reliable failure diagnosis method is urgently needed.
The intelligent monitoring and health maintenance of the wind driven generator are important guarantees for wind power development, the planetary gear speed increasing box is a core component of the wind driven generator, and typical early damage such as pitting, cracking, abrasion and the like is easy to occur to gears in the component under the long-term random wind load working condition, so that the wind driven generator is potentially huge in harm. According to statistics, gear faults are most likely to occur in wind power equipment, the proportion is 60%, and the time and economic cost for overhauling the gear box are extremely high. Therefore, the early fault diagnosis research of the wind power planetary gear speed increasing box is developed, the fault diagnosis accuracy can be improved, the fault diagnosis is prevented, the time and the maintenance cost are saved, the requirements of safe and healthy development of wind power energy industry are met, and the wind power planetary gear speed increasing box has wide market requirements and industrialization prospects.
Vibration detection and oil monitoring are common fault diagnosis methods. However, in the aspect of vibration detection, because the wind driven generator is influenced by random wind load, the input rotation speed of the wind driven generator planetary gear speed increasing box is fluctuant, so that the non-stationarity characteristic of the vibration signal of the wind driven generator planetary gear speed increasing box is very obvious, and the traditional vibration signal fault characteristic extraction means cannot well analyze the non-stationary signal; in the aspect of oil monitoring, because the wind power planetary gear speed increasing box can cause abrupt change of the abrasive particle number in oil (namely, oil change interference problem) when the lubricating oil is replaced, the traditional abrasive particle number characteristic index cannot accurately reflect the actual fault state of the wind power planetary gear speed increasing box.
On the other hand, the traditional wind driven generator planetary gear speed increasing box fault diagnosis detection method is single, the fault state of the planetary gear speed increasing box cannot be accurately reflected by data obtained by only a single detection means, and uncertainty exists in the detection result; the fault characteristic information reflected by the single detection means is not comprehensive, fault extraction and analysis can be carried out only from a single layer, and multi-angle and multi-layer comprehensive evaluation of the fault state of the planetary gear speed increasing box can not be carried out. Therefore, a wind power planetary gear box fault state detection and evaluation method with multiple information feature fusion is lacking, and a corresponding wind power planetary gear speed increasing box fault diagnosis test platform is also lacking to provide convenient test data support.
In summary, the research and development of the wind power planetary gear speed increasing box fault diagnosis test platform and the comprehensive, accurate and intelligent multi-information feature fusion detection evaluation method have important significance aiming at the problems of single detection method, random wind load special working condition, oil change interference and the like faced by wind power planetary gear speed increasing box fault diagnosis.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a wind power gear speed-up box fault diagnosis test platform based on multi-feature fusion, and the test platform is used for carrying out fault information parallel acquisition by using three detection means of vibration, oil liquid and noise, so that the fault information can be analyzed and processed, the type and degree of gear damage can be fully represented from the three aspects, the accuracy of fault diagnosis is greatly improved, and the fault features are extracted and analyzed more fully from different angles.
In addition, the invention provides a wind power planetary gear box fault state detection and evaluation method based on vibration-noise-oil characteristic fusion, which can extract vibration characteristic indexes and noise characteristic indexes according to random wind load working conditions and oil characteristic indexes according to oil change interference problems, thereby establishing a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory and being beneficial to improving the comprehensiveness, intelligence and accuracy of wind power planetary gear speed increasing box fault diagnosis.
The wind power gear speed-up box fault diagnosis test platform based on multi-feature fusion comprises the following specific scheme:
wind power gear speed increasing box fault diagnosis test platform based on multi-feature fusion includes:
the surface of the test bed base can be provided with a wind power gear speed increasing box to be tested, an output shaft of the wind power gear speed increasing box to be tested is provided with a load, the load is connected with the wind power gear speed increasing box to be tested through a fixed shaft gear box, an input shaft of the wind power gear speed increasing box is connected with a servo motor through a speed reducing box, the servo motor is arranged on the test bed base through a servo motor mounting seat, and the servo motor is connected with a PLC;
the wind power planetary gear speed increasing box to be tested is respectively connected with the vibration signal detection module, the oil information detection module and the noise detection module; and the vibration signal detection module, the oil information detection module and the noise detection module are respectively connected with the PLC.
According to the test platform, by using the technical means of equal-angle sampling of the vibration signals, fault feature extraction analysis is carried out on the non-stationary vibration signals, and the influence of rotation speed fluctuation on the accuracy of the fault feature extraction analysis of the vibration signals of the wind power gear speed increasing box can be effectively reduced.
Further, the oil information detection module comprises a temperature sensor, an online dielectric constant sensor, an online viscosity sensor, an online abrasive particle monitoring sensor and a CMOS abrasive particle image sensor. And each sensor is sequentially connected with an oil pipe T-shaped three-way interface of the lubrication system of the wind power planetary gear speed increasing box through threads, and the detected temperature information of the lubricating oil, the water content information of the lubricating oil, the viscosity information of the lubricating oil, the particle size information of the abrasion abrasive particles and the type information of the abrasion abrasive particles are transmitted to a data acquisition module.
Further, one end of the wind power gear speed increasing box to be tested is sequentially connected with the fine filter, the oil pump, the coarse filter, the cooler and the oil information detection module through an oil pipe, and the oil pipe is connected to the other end of the wind power gear speed increasing box to be tested.
Further, the oil information detection sensor is arranged at the interface of the oil pipe at the end side of the wind power gear speed increasing box to be detected through threads.
Further, the vibration signal detection module comprises a pulse signal acquisition device and a plurality of vibration acceleration sensors, wherein the pulse signal acquisition device is a photoelectric encoder arranged on an input shaft of the wind power gear speed increasing box to be detected, the vibration acceleration sensors are respectively arranged on bearing seats and box bodies at two ends of the wind power gear speed increasing box, and the vibration acceleration sensors and the pulse signal acquisition device resample vibration signals acquired at equal angles and transmit the vibration signals to the data acquisition module (such as a data acquisition card).
Further, the PLC is connected with a servo driver, and the servo driver controls the servo motor to operate; the servo motor is provided with a built-in encoder, and the built-in encoder feeds back the operation parameters of the servo motor to the PLC controller, so that the closed-loop control of the rotating speed and the torque of the servo motor is realized.
The industrial personal computer is connected with the PLC, the PLC is connected with the oil pump, the start and the close of the oil pump are directly controlled, and the PLC is connected with the load. The industrial personal computer sends control parameters to the PLC, the PLC gives the control parameters after calculation and sends the control parameters to the servo driver, and the servo driver controls the servo motor to operate; the servo motor is provided with a built-in encoder, and motor operation parameters are fed back to the control unit, so that closed-loop control of the rotating speed and torque of the servo motor is realized.
The industrial personal computer integrates a servo motor speed regulation software system, a load regulation software system and a fault diagnosis software system. When the test is carried out, all the devices of the test platform can be started by the industrial personal computer.
Specifically, an input shaft at the left side of the wind power planetary gear speed increasing box to be tested is connected with an output shaft of the front reduction box through a second coupler; the input shaft at the left side of the front reduction gearbox is connected with the output shaft of the servo motor through a first coupler; an output shaft on the right side of the wind power planetary gear speed increasing box to be tested is connected with an input shaft on the left side of the fixed shaft gear box through a third coupler; the output shaft on the right side of the fixed shaft gear box is connected with a load through a fourth coupler, and the fixed shaft gear box mounting seat and the load mounting seat are of solid structures with kidney-shaped guide holes and conical positioning pin holes.
Further, the wind power gear speed increasing box to be measured is arranged on the test bed base through a wind power gear speed increasing box mounting seat, the load is arranged on the test bed base through a load mounting seat, the fixed shaft gear box is arranged on the test bed base through a fixed shaft gear box mounting seat, and the reduction gearbox is arranged on the test bed base through a reduction gearbox seat.
Further, the wind power gear speed increasing box mounting seat, the load mounting seat, the fixed shaft gear box mounting seat and the speed reducing box seat are detachably connected with the test bed base through bolts respectively. Each mount pad is the entity structure of taking waist shape guiding hole and toper location pinhole, and the mount pad is equipped with the recess and is used for fixed corresponding mechanism.
In addition, the test platform also comprises a power module, wherein the power module is respectively and independently connected with the servo motor, the vibration acceleration sensor, the noise signal sensor, the oil information detection sensor and the PLC controller to supply power.
And (3) through collecting random wind load data of the wind field, carrying out simulation modeling analysis on wind loads under different working conditions, and obtaining corresponding random wind load spectrums. Because the research object is the wind power planetary gear speed increasing box, the input end of the wind power planetary gear speed increasing box has only one rotating speed and torque signal, and in order to simplify the test process and avoid unnecessary resource waste, the obtained random wind power load spectrum only needs to be calculated into the rotating speed and torque signal of the input end of the wind power planetary gear speed increasing box in the wind power generator transmission chain. However, the rotational speed and torque signal at the input end of the wind power planetary gear speed increasing box has the characteristics of low rotational speed and large torque, the signal is not easy to directly generate and adjust through mechanical equipment, a mode of adding a speed reducer at the front end of the wind power planetary gear speed increasing box can be adopted to reversely increase speed and reduce torque, the low rotational speed and large torque signal is converted into a high rotational speed and small torque signal to control, and the signal is used as an original input control signal of a test platform, namely a random wind load signal which needs to be simulated by a driving motor of the test platform. The random wind load signal control system consists of an industrial personal computer, a PLC controller, a servo driver and a servo motor. When the system operates, signal parameters are input into the industrial personal computer according to test requirements, the industrial personal computer sends control parameters to the PLC, and the PLC gives the control parameters after calculation, so that the servo driver controls the servo motor to operate; the servo motor is provided with an encoder, and the motor operation parameters are fed back to the control unit, so that the closed-loop control of the rotating speed and the torque of the servo motor is realized.
In the process of building and installing a test platform, main assembly errors affecting the measurement precision of the platform are coaxiality errors generated by a driving motor, a wind power planetary gear speed increasing box and various input and output shafts of a load motor in the test platform during assembly. When the coaxiality error does not meet the matching standard, serious noise, vibration and flexible impact are brought in the running process of the equipment, irreversible damage is brought to each part of the rotary machine, and the service life of the whole machine is seriously influenced. From the test point of view, interference can be brought to the collection work of test data, and even the accuracy of the data is influenced. In order to reduce coaxiality errors when the rotating shafts are matched, a modularized adjustable device is added besides optimizing the structure and design parameters and improving the machining precision of the matching surfaces in the process of designing and machining the test bench mounting base. On the base installation module of all rotary mechanism, designed the location structure that can accurate centering in the horizontal direction, can effectively improve the equipment location and debugging work's in the assembly process intensity, realize accurate location installation. The bolt positioning holes on the wind power planetary gear speed increasing box mounting seat are processed into kidney-shaped guide holes, so that the kidney-shaped guide holes can be guided and finely adjusted in the horizontal direction perpendicular to the transmission shaft during mounting, then the assembly position of the wind power gear box is accurately positioned by means of positioning pin connection designed by the mounting seat, and finally all positioning bolts on the mounting seat are screwed to complete the assembly of the wind power planetary gear speed increasing box mounting seat on the test bench base.
The wind power planetary gear speed increasing box test platform integrates various fault characteristic acquisition schemes, performs fault information parallel acquisition by using three detection means of vibration, oil liquid and noise, analyzes and processes the fault information in an industrial control fault diagnosis system, and fully characterizes the fault state of the planetary gear speed increasing box from the three aspects.
Every time the input shaft of the wind power planetary gear speed increasing box rotates by one sampling angle, the photoelectric encoder acquires a rotating speed pulse, and a pulse signal is transmitted to the industrial personal computer; the industrial personal computer receives the rotating speed pulse signal, sends an acquisition instruction to the acceleration vibration signal sensor, and acquires the acceleration vibration signal of the wind power planetary gear speed increasing box once; the acceleration vibration signal acquired at this time is an equal-angle resampling vibration signal of the wind power planetary gear speed increasing box, and the acquired equal-angle resampling vibration signal is transmitted to a data acquisition module (connected with an industrial personal computer).
In order to overcome the defects of the prior art, the invention also provides a wind power planetary gear box fault state detection and evaluation method based on vibration-noise-oil characteristic fusion, which comprises the following steps:
1) Extracting vibration characteristic indexes aiming at random wind load working conditions;
2) Extracting noise characteristic indexes;
3) Extracting oil characteristic indexes aiming at the problem of oil change interference;
4) Establishing a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory;
5) And diagnosing and evaluating the fault state of the wind power planetary gear speed increasing box.
The specific steps of the step 1) are as follows:
1-1) obtaining stable equal-angle resampling vibration signal data through a vibration signal detection module of a test platform;
1-2) decomposing the equal-angle resampled vibration signal into a series of eigenmode components based on a complete set empirical mode decomposition method;
1-3) screening out an optimal IMF signal according to a kurtosis criterion, so as to achieve the purpose of filtering and denoising;
1-4) carrying out Fourier transformation on the optimal IMF signal to obtain a fault order feature spectrogram, and taking the fault order feature spectrogram as a vibration feature index for training and constructing a deep neural network evaluation model for the first time.
The specific steps of the step 2) are as follows:
2-1) obtaining noise signal data of the wind power planetary gear speed increasing box through a noise signal detection module of the test platform;
2-2) obtaining sound pressure level and octave spectrogram of the noise signal based on an acoustic calculation analysis method, and taking the sound pressure level and octave spectrogram as noise characteristic indexes;
The specific steps of the step 3) are as follows:
3-1) obtaining an oil information database through an oil information detection module of the test platform;
3-2) calculating the percentage of the number of the abrasive particles of different types to the total number of the abrasive particles based on a ferrographic analysis method;
3-3) calculating the percentage of the number of the abrasive particles with different particle diameters to the total number of the abrasive particles based on a laser particle size analysis method;
3-4) taking the proportion characteristic of the abrasive particle type distribution and the proportion characteristic of the abrasive particle size distribution, which are less affected by oil change interference, as oil characteristic indexes.
The specific steps of the step 4) are as follows:
4-1) establishing a training sample set Φ, as shown in formula (1), wherein Φ x For the x-th training sample, V x ,N x ,O x Each representing various single characteristic indices of the x-th training sample: vibration characteristic indexes, noise characteristic indexes and oil characteristic indexes;
4-2) based on remarkable advantages of deep learning in aspects of image recognition, machine learning, big data processing analysis and the like, respectively taking various single characteristic indexes in a training sample set as input quantities, and training and constructing a deep neural network evaluation model of the various single characteristic indexes; the output quantity of the model is the fault state of the wind power planetary gear speed increasing box;
4-3) introducing an identification frame in DS evidence theory into a deep neural network evaluation model, and determining a fault state identification frame theta = { F of the wind power planetary gear speed increasing box by referring to the output quantity of the deep neural network evaluation model 1 ,F 2 ,…,F n }, wherein F 1 ，F 2 ，…，F n Representing n fault states of the wind power planetary gear speed increasing box;
4-4) designing a multi-feature deep learning-DS evidence theory fusion decision rule based on the advantage of DS evidence theory in the aspect of multi-source feature information fusion, wherein the key is to describe a credibility allocation function by combining a deep neural network evaluation model of various single feature indexes, as shown in a formula (2):
m i (F 1 ，F 2 ，…，F n ，Θ)＝(p i q i1 ，p i q i2 ，…，p i q in ，1-p i ) (2)
wherein m is i The reliability distribution function of the evaluation result of the deep neural network model representing the ith single characteristic index, i=1, 2, … and k, wherein k is the total number of characteristic indexes such as vibration, noise, oil and the like; p is p i The accuracy of the evaluation result of the deep neural network model representing the ith single characteristic index; q ij The deep neural network model representing the ith single feature index evaluates the sample as the credibility of the jth fault state, j=1, 2, …, n;
identifying any fault state F in the framework Θ for the fault state j The multi-feature deep learning-DS evidence theory fusion decision rule can be represented by the following formulas (3) and (4):
in the formula, for the deep neural network model of various single characteristic indexes, the accuracy of the evaluation result of the training sample set can be taken as p i A value; q ij The value can be statistically determined according to the evaluation result of the deep neural network model.
The specific steps of the step 5) are as follows:
new vibration, noise and oil test data are continuously collected through a test platform, then vibration characteristic indexes, noise characteristic indexes and oil characteristic indexes of the test platform are respectively extracted to form a new sample to be tested, a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory established in the prior art is input, and the model can output the fault state of the planetary gear speed increasing box at the moment, so that comprehensive, accurate and intelligent diagnosis and evaluation of the fault state of the wind power planetary gear speed increasing box are realized.
Compared with the prior art, the invention has the beneficial effects that:
1) The invention provides a wind driven generator planetary gear box test platform for multi-means detection and comprehensive extraction of fault characteristics under random wind load working conditions, which integrates multiple fault characteristic acquisition schemes, performs fault information parallel acquisition by using three detection means of vibration, oil liquid and noise, analyzes and processes the fault information in an industrial control fault diagnosis system, and fully characterizes the type and degree of gear damage from the three aspects. The fault diagnosis method based on multi-information fusion not only greatly improves the accuracy of fault diagnosis, but also extracts and analyzes the fault characteristics more fully from different angles.
2) The invention provides an assembly-oriented test bed design which effectively reduces coaxiality errors between an input shaft and an output shaft of a rotating mechanism. In the process of designing and processing the mounting base of the experiment bench, the structure and design parameters are optimized, the processing precision of each matching surface is improved, and the modularized adjustable device is further increased. On the base installation module of all rotary mechanism, designed the location structure that can accurate centering in the horizontal direction, can effectively improve the equipment location and debugging work's in the assembly process intensity, realize accurate location installation.
3) The invention provides a vibration detection module of a wind driven generator planetary gear speed increasing box test platform based on random wind load, which can collect equal-angle resampling vibration signals of a wind driven generator planetary gear speed increasing box, so that the influence of rotation speed fluctuation on the accuracy of extraction and analysis of vibration signal fault characteristics of the wind driven generator planetary gear speed increasing box is effectively reduced;
4) The invention provides an oil liquid detection module which comprises a temperature sensor, an online dielectric constant sensor, an online viscosity sensor, an online abrasive particle monitoring sensor and a CMOS abrasive particle image sensor, wherein the temperature sensor, the online dielectric constant sensor, the online viscosity sensor, the online abrasive particle monitoring sensor and the CMOS abrasive particle image sensor are respectively arranged on an oil pipe T-shaped three-way interface of a gearbox lubrication system through threaded connection so as to realize online monitoring of the lubrication state and the abrasion state (including the abrasion type and the abrasion degree) of the wind power gear speed increasing box lubrication system, and meanwhile, the installation mode is beneficial to maintenance and replacement of a fault sensor so as to realize upgrading of the detection module.
5) The method is characterized in that a mode of adding a speed reducer at the front end of the wind power gear speed increasing box is adopted to carry out reverse speed increasing and torque decreasing, a low-rotation-speed high-torque signal is converted into a high-rotation-speed low-torque signal to be controlled, the low-rotation-speed high-torque signal of the wind power gear speed increasing box input end in a wind power generator transmission chain calculated through a random wind power load spectrum is well simulated, and the signal is used as an original input control signal of a test platform. The test process is simplified, and unnecessary resource waste is avoided.
6) The invention provides a wind power planetary gear box fault state detection and evaluation method based on vibration-noise-oil characteristic fusion, which can extract vibration characteristic indexes and noise characteristic indexes according to random wind load working conditions and oil characteristic indexes according to oil change interference problems, thereby establishing a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory and being beneficial to improving the comprehensiveness, intelligence and accuracy of wind power planetary gear speed increasing box fault diagnosis.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application.
FIG. 1 is a schematic diagram of the front structure of the present invention;
FIG. 2 is a schematic top view of the present invention;
FIG. 3 is a schematic diagram of a lubrication system for a wind power gear speed increasing box;
FIG. 4 is a schematic diagram of an oil information detection module;
FIG. 5 is a schematic diagram of a data acquisition system;
FIG. 6 is a schematic diagram of a servo motor control system;
FIG. 7 is a schematic design of a wind power gear speed increasing box fault diagnosis test platform.
FIG. 8 is a flow chart for detecting and evaluating the fault state of a wind power planetary gear box based on vibration-noise-oil characteristic fusion;
1, a load fixing frame; 2. a load; 3. a fourth coupling; 4. a fixed-axis gear box; 5. a third coupling; 6. a wind power gear speed increasing box to be tested; 7. a vibration acceleration sensor; 8. a second coupling; 9. a front reduction gearbox; 10. a first coupling; 11. a servo motor; 12. a servo motor fixing frame; 13. a test bed base; 14. a servo motor mounting seat; 15. a front reduction gearbox mounting seat; 16. the wind power gear speed increasing box mounting seat to be tested; 17. a fixed-axis gear box mounting seat; 18. a load mounting seat; 19. a photoelectric encoder; 20. a photoelectric encoder mounting base; 21. a fifth coupling; 22. a PLC controller; 23. an oil information sensor; 24. a noise signal sensor; 25. a data acquisition module; 26. an industrial personal computer; 27. a servo driver; 28. an overflow valve; 29. fine filter; 30. an oil pump; 31. a coarse filter; 32. a cooler; 33. a CMOS abrasive particle image sensor; 34. an on-line abrasive particle monitoring sensor; 35. an online viscosity sensor; 36. an in-line dielectric constant sensor; 37. a temperature sensor.
Detailed Description
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments in accordance with the present application. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
As introduced by the background technology, the defects in the prior art are overcome, and in order to solve the technical problems, the application provides a wind power gear speed increasing box fault diagnosis test platform based on multi-feature fusion.
In a typical embodiment of the application, as shown in fig. 1, a wind power gear speed increasing box fault diagnosis test platform based on multi-feature fusion comprises a test bed base 13, wherein a wind power gear speed increasing box installation seat 16 is arranged on the test bed base 13; the wind power gear speed increasing box mounting seat 16 is provided with a wind power gear speed increasing box 6 to be tested, and an input shaft at the left side of the wind power gear speed increasing box 6 to be tested is connected with an output shaft of the front reduction box 9 through a second coupler 8; the input shaft at the left side of the front reduction box 9 is connected with the output shaft of a servo motor 11 through a first coupler 10; an output shaft on the right side of the wind power gear speed increasing box 6 to be tested is connected with an input shaft on the left side of the fixed shaft gear box 4 through a third coupler 5; the output shaft on the right side of the fixed-shaft gear box 4 is connected with a load through a fourth coupler 3; the end of the extension shaft of the input shaft on the right side of the fixed shaft gear box 4 is connected with the photoelectric encoder 19 through a fifth coupler 21.
The power module provides power for the whole device of the wind power gear speed increasing box fault diagnosis test bed; the load 2 selects a magnetic particle brake.
The servo motor mounting seat 14, the front gearbox mounting seat 15, the wind power gear speed increasing box mounting seat 16, the fixed shaft gearbox mounting seat 17 and the load mounting seat 18 are solid structures with kidney-shaped guide holes and conical positioning pin holes.
The servo motor 11 is fixed on the servo motor fixing frame 12 in a bolt connection mode, and the servo motor fixing frame 12 is fixed on the servo motor mounting seat 14 in a bolt connection mode.
The front reduction gearbox 9 is fixed on the front reduction gearbox mounting seat 15 in a bolt connection mode. The wind power gear speed increasing box 6 is fixed on the wind power gear speed increasing box mounting seat 16 in a bolt connection mode. The fixed-axis gear box 4 is fixed on the fixed-axis gear box mounting seat 17 through a bolt connection mode. The photoelectric encoder 19 is fixed to the photoelectric encoder mounting base 20 by bolting. The load 2 is fixed on the load fixing frame 1 in a bolt connection mode, and the load fixing frame 1 is fixed on the load mounting seat 18 in a bolt connection mode.
The servo motor mounting seat 14, the front gearbox mounting seat 15, the wind power gear speed increasing box mounting seat 16, the fixed shaft gearbox mounting seat 17, the photoelectric encoder mounting seat 20 and the load mounting seat 18 are all fixed on the test bed base 13 in a bolt connection mode.
One end of the wind power gear speed increasing box 6 to be tested is connected with the overflow valve 28, the fine filter 29, the oil pump 30, the coarse filter 31, the cooler 32 and the oil information detection module in sequence through oil pipes, and finally the oil pipes are connected to the other end of the wind power gear speed increasing box 6 to be tested to form a wind power gear speed increasing box lubrication system; the oil pump 30 is connected with the industrial personal computer 26 of the Miao Hua IPC-610L.
The noise signal detection module comprises a noise signal sensor 24, wherein the noise signal sensor 24 is arranged on a box body of the wind power gear speed increasing box to be detected, and the collected noise signal is transmitted to the data acquisition module 25 and further transmitted to the research IPC-610L industrial personal computer 26.
The oil information detection module comprises a temperature sensor 37, an online dielectric constant sensor 36, an online viscosity sensor 35, an online abrasive particle monitoring sensor 34 and a CMOS abrasive particle image sensor 33, is sequentially installed on an oil pipe T-shaped three-way interface of a wind power gear speed increasing box lubrication system through threaded connection, and transmits detected lubricating oil temperature information, lubricating oil water content information, lubricating oil viscosity information, abrasive particle size information and abrasive particle type information to a data acquisition module 25 of a Siemens S7-200PLC controller 22 and to a grinding IPC-610L industrial personal computer 26.
The vibration signal detection module comprises a pulse signal acquisition device and a plurality of vibration acceleration sensors 7, wherein the pulse signal acquisition device is a photoelectric encoder 19 arranged on an input shaft of the wind power planetary gear speed increasing box to be detected, and the vibration acceleration sensors 7 are respectively arranged on bearing seats and box bodies at two ends of the wind power planetary gear speed increasing box to be detected 6. The vibration acceleration sensor and the pulse signal acquisition device transmit the acquired equal-angle resampled vibration signals to the data acquisition module 25, and further transmit the sampled equal-angle resampled vibration signals to the Miao IPC-610L industrial personal computer 26.
The industrial personal computer of the MIWA IPC-610L sends control parameters to the Siemens S7-200PLC controller, the Siemens S7-200PLC controller gives the control parameters after calculation and sends the control parameters to the servo driver 27, and the servo driver 27 controls the servo motor 11 to operate; the servo motor 11 is provided with a built-in encoder, and motor operation parameters are fed back to the control unit, so that closed-loop control of the rotating speed and torque of the servo motor 11 is realized.
When the wind power gear speed increasing box 6 to be detected is detected, firstly, a proper wind power gear speed increasing box installation seat 16 is designed according to the size of the wind power gear speed increasing box 6 to be detected, the positions of the wind power gear speed increasing box installation seat 16 to be detected, a servo motor installation seat 14, a front speed reducing box installation seat 15, a fixed shaft gear box installation seat 17 and a load installation seat 18 on a test bed base 13 are adjusted and fixed on the test bed base 13 along a waist-shaped guide hole respectively, then the wind power gear speed increasing box 6 to be detected is placed on the wind power gear speed increasing box installation seat 16 to be detected, the input shaft and the output shaft of the wind power gear speed increasing box 6 to be detected are connected through a second coupler 8 and a third coupler 5 respectively, finally, the lubricating oil inlet/outlet of the wind power gear speed increasing box 6 to be detected is connected with corresponding oil pipes in a wind power gear speed increasing box lubrication system, and the vibration acceleration sensor 7 is placed at the corresponding position of the wind power gear speed increasing box 6 to be detected. The power module provides power for the whole device of the wind power gear speed increasing box fault diagnosis test bed.
The industrial personal computer 26 of the Miao IPC-610L integrates a servo motor speed regulation software system, a load regulation software system and a fault diagnosis software system. When the detection is carried out, all the equipment of the test platform can be started by the IPC-610L industrial personal computer.
Meanwhile, the photoelectric encoder 19, the noise signal sensor 24, the temperature sensor 37, the online dielectric constant sensor 36, the online viscosity sensor 35, the online abrasive particle monitoring sensor 34, the CMOS abrasive particle image sensor 33 and the vibration acceleration sensor 7 transmit measured data to the data acquisition module 25 through data wires and transmit the data to the Miao IPC-610L industrial personal computer 26, the Miao IPC-610L industrial personal computer 26 processes and analyzes the acquired data, and a detector detects the wind power gear speed increasing box 6 according to the acceleration vibration signal, the oil information and the noise signal after processing and analysis. According to the invention, the acceleration vibration signal, the oil information and the noise signal of the wind power gear speed increasing box are comprehensively considered, so that the wind power gear speed increasing box is more accurately detected.
In addition, the invention also provides a wind power planetary gear box fault state detection and evaluation method based on vibration-noise-oil characteristic fusion, which comprises the following steps:
1) Extracting vibration characteristic indexes aiming at random wind load working conditions;
2) Extracting noise characteristic indexes;
3) Extracting oil characteristic indexes aiming at the problem of oil change interference;
4) Establishing a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory;
5) And diagnosing and evaluating the fault state of the wind power planetary gear speed increasing box.
The method comprises the following specific steps:
1) Extracting vibration characteristic indexes aiming at random wind load working conditions;
firstly, obtaining stable equal-angle resampling vibration signal data through a vibration signal detection module of a test platform; then, decomposing the equiangular resampled vibration signal into a series of eigenmode components (intrinsic mode function, IMF) based on a complete set empirical mode decomposition method (complete ensemble empirical mode decomposition with adaptive noise, CEEMDAN); then, the optimal IMF signal is screened out according to the kurtosis criterion, so as to achieve the purpose of filtering and denoising; and finally, carrying out Fourier transform on the optimal IMF signal to obtain a fault order feature spectrogram, and taking the fault order feature spectrogram as a vibration feature index for training and constructing a deep neural network evaluation model for the first time.
2) Extracting noise characteristic indexes;
Firstly, obtaining noise signal data of a wind power planetary gear speed increasing box through a noise signal detection module of a test platform; then, based on an acoustic calculation analysis method, a sound pressure level and an octave spectrogram of the noise signal are obtained and used as noise characteristic indexes.
3) Extracting oil characteristic indexes aiming at the problem of oil change interference;
when the wind power planetary gear speed increasing box is used for replacing lubricating oil, abrupt change of the abrasive particle number in the oil (namely, oil replacement interference problem) is caused, so that the traditional characteristic index of the abrasive particle number cannot accurately reflect the actual fault state of the wind power planetary gear speed increasing box; however, under the same abrasion state, abrasive particles with different abrasion types and particle diameters still enter the oil according to the proportion before oil change, so that the characteristic indexes of the proportion of the types of the abrasive particles and the proportion of the particle diameters of the abrasive particles in the extracted oil are breakthrough points for solving the problem of oil change interference.
Firstly, an oil information database is obtained through an oil information detection module of a test platform; then, calculating the percentage of the number of different types of abrasive particles (normal abrasion abrasive particles, severe sliding abrasive particles, cutting abrasive particles, fatigue abrasion abrasive particles and oxidized abrasive particles) to the total abrasive particle number based on a ferrographic analysis method; then, calculating the number of abrasive particles with different particle sizes (particle sizes of 0-10 mu m, 10-30 mu m, 30-50 mu m, 50-100 mu m and more than 100 mu m) accounting for the percentage of the total abrasive particle number based on a laser particle size analysis method; and finally, taking the proportion characteristic of the abrasive particle type distribution and the proportion characteristic of the abrasive particle size distribution, which are less influenced by oil change interference, as oil characteristic indexes.
4) Establishing a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory;
in order to improve the utilization rate of the vibration, noise and oil characteristic indexes, the sensitivity advantage of the vibration characteristic indexes in the aspect of fault location analysis, the sensitivity advantage of the noise characteristic indexes in the aspect of noise source location and the sensitivity advantage of the oil characteristic indexes in the aspect of fault quantitative analysis are integrated, the fault state of the planetary gear speed increasing box is estimated more comprehensively, accurately and intelligently, intelligent technologies such as machine learning, multi-source information fusion and the like are required to be comprehensively applied, and an effective vibration-noise-oil characteristic fusion estimation model is established. The specific scheme is as follows:
(1) Establishing a training sample set phi as shown in formula (1), wherein phi x For the x-th training sample, V x ,N x ,O x Each representing various single characteristic indices of the x-th training sample: vibration characteristic index, noise characteristic index and oil characteristic index.
(2) Based on remarkable advantages of deep learning in aspects of image recognition, machine learning, big data processing analysis and the like, respectively taking various single characteristic indexes (vibration characteristic indexes, noise characteristic indexes and oil characteristic indexes) in a training sample set as input quantities, training and constructing a deep neural network evaluation model of various single characteristic indexes; the output quantity of the model is the failure state (such as gear abrasion, gear crack, bearing abrasion and the like) of the wind power planetary gear speed increasing box.
(3) Introducing an identification frame in DS evidence theory into a deep neural network evaluation model, and determining a fault state identification frame Θ= { F of the wind power planetary gear speed increasing box by referring to the output quantity of the deep neural network evaluation model 1 ,F 2 ,…,F n }, wherein F 1 ，F 2 ，…，F n Representing n fault states of the wind power planetary gear speed increasing box.
(4) Based on the advantages of DS evidence theory in the aspect of multisource feature information fusion, a multi-feature deep learning-DS evidence theory fusion decision rule is designed, and the key is that a credibility distribution function is described by combining with a deep neural network evaluation model of various single feature indexes, as shown in a formula (2):
m i (F 1 ，F 2 ，…，F n ，Θ)＝(p i q i1 ，p i q i2 ，…，p i q in ，1-p i ) (2)
wherein m is i The reliability distribution function of the evaluation result of the deep neural network model representing the ith single characteristic index, i=1, 2, … and k, wherein k is the total number of characteristic indexes such as vibration, noise, oil and the like; p is p i The accuracy of the evaluation result of the deep neural network model representing the ith single characteristic index; q ij The deep neural network model representing the i-th single feature index evaluates the sample as the confidence level of the j-th fault state, j=1, 2, …, n.
Identifying any fault state F in the framework Θ for the fault state j The multi-feature deep learning-DS evidence theory fusion decision rule can be represented by the following formulas (3) and (4):
In the formula, for the deep neural network model of various single characteristic indexes, the accuracy of the evaluation result of the training sample set can be taken as p i A value; q ij The value can be statistically determined according to the evaluation result of the deep neural network model.
(5) Testing and correcting the deep neural network evaluation models of various single characteristic indexes through a test sample set, and perfecting a multi-characteristic deep learning-DS evidence theory fusion decision rule; thus, a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory is established.
5) Diagnosis and evaluation of fault state of wind power planetary gear speed increasing box
New vibration, noise and oil test data are continuously collected through a test platform, then vibration characteristic indexes, noise characteristic indexes and oil characteristic indexes of the test platform are respectively extracted to form a new sample to be tested, and a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory established in the prior art is input, so that the model can intelligently and autonomously output the fault state of the planetary gear speed increasing box at the moment, and comprehensive, accurate and intelligent diagnosis and evaluation of the fault state of the wind power planetary gear speed increasing box are realized.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the same, but rather, various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.
Claims (9)
1. A wind power planetary gear box fault state detection and evaluation method based on vibration-noise-oil characteristic fusion is characterized by comprising the following steps:
1) Extracting vibration characteristic indexes aiming at random wind load working conditions;
2) Extracting noise characteristic indexes;
3) Extracting oil characteristic indexes aiming at the problem of oil change interference;
4) Establishing a vibration-noise-oil characteristic fusion evaluation model based on deep learning and DS evidence theory;
5) Diagnosing and evaluating the fault state of the wind power planetary gear speed increasing box;
the specific steps of the step 4) are as follows:
4-1) establishing a training sample set Φ, as shown in formula (1), wherein Φ x For the x-th training sample, V x ,N x ,O x Each representing various single characteristic indices of the x-th training sample: vibration characteristic indexes, noise characteristic indexes and oil characteristic indexes;
4-2) based on remarkable advantages of deep learning in aspects of image recognition, machine learning and big data processing analysis, respectively taking various single characteristic indexes in a training sample set as input quantities, and training and constructing a deep neural network evaluation model of the various single characteristic indexes; the output quantity of the model is the fault state of the wind power planetary gear speed increasing box;
4-3) introducing an identification frame in DS evidence theory into a deep neural network evaluation model, and determining a fault state identification frame theta = { F of the wind power planetary gear speed increasing box by referring to the output quantity of the deep neural network evaluation model 1 ,F 2 ,…,F n }, wherein F 1 ，F 2 ，…，F n Representing n fault states of the wind power planetary gear speed increasing box;
4-4) designing a multi-feature deep learning-DS evidence theory fusion decision rule based on the advantage of DS evidence theory in the aspect of multi-source feature information fusion, wherein the key is to describe a credibility allocation function by combining a deep neural network evaluation model of various single feature indexes, as shown in a formula (2):
m i (F 1 ，F 2 ，…，F n ，Θ)＝(p i q i1 ，p i q i2 ，…，p i q in ，1-p i ) (2)
wherein m is i A reliability distribution function of the evaluation result of the deep neural network model representing the ith single characteristic index, wherein i=1, 2, … and k, and k is the total number of vibration, noise and oil characteristic indexes; p is p i The accuracy of the evaluation result of the deep neural network model representing the ith single characteristic index; q ij The deep neural network model representing the ith single feature index evaluates the sample as the credibility of the jth fault state, j=1, 2, …, n;
identifying any fault state F in the framework Θ for the fault state j Multi-feature deep learning-DSThe evidence theory fusion decision rule can be represented by the following formula (3) and formula (4):
in the formula, for the deep neural network model of various single characteristic indexes, the accuracy of the evaluation result of the training sample set can be taken as p i A value; q ij The value can be statistically determined according to the evaluation result of the deep neural network model.
2. The wind power planetary gear box fault state detection and assessment method based on vibration-noise-oil characteristic fusion according to claim 1, wherein the specific steps of the step 1) are as follows:
1-1) obtaining stable equal-angle resampling vibration signal data through a vibration signal detection module of a test platform;
1-2) decomposing the equal-angle resampled vibration signal into a series of eigenmode components based on a complete set empirical mode decomposition method;
1-3) screening out an optimal IMF signal according to a kurtosis criterion, so as to achieve the purpose of filtering and denoising;
1-4) carrying out Fourier transformation on the optimal IMF signal to obtain a fault order feature spectrogram, and taking the fault order feature spectrogram as a vibration feature index for training and constructing a deep neural network evaluation model for the first time.
3. The wind power planetary gear box fault state detection and assessment method based on vibration-noise-oil characteristic fusion according to claim 1, wherein the specific steps of the step 2) are as follows:
2-1) obtaining noise signal data of the wind power planetary gear speed increasing box through a noise signal detection module of the test platform;
2-2) obtaining sound pressure level and octave spectrogram of the noise signal based on an acoustic calculation analysis method, and taking the sound pressure level and octave spectrogram as noise characteristic indexes;
The specific steps of the step 3) are as follows:
3-1) obtaining an oil information database through an oil information detection module of the test platform;
3-2) calculating the percentage of the number of the abrasive particles of different types to the total number of the abrasive particles based on a ferrographic analysis method;
3-3) calculating the percentage of the number of the abrasive particles with different particle diameters to the total number of the abrasive particles based on a laser particle size analysis method;
3-4) taking the proportion characteristic of the abrasive particle type distribution and the proportion characteristic of the abrasive particle size distribution, which are less affected by oil change interference, as oil characteristic indexes.
4. A test platform for a wind power planetary gear box fault state detection and assessment method based on vibration-noise-oil feature fusion according to any one of claims 1-3, comprising:
the surface of the test bed base can be provided with a wind power gear speed increasing box to be tested, an output shaft of the wind power gear speed increasing box to be tested is provided with a load, the load is connected with the wind power gear speed increasing box to be tested through a fixed shaft gear box, an input shaft of the wind power gear speed increasing box to be tested is connected with a servo motor through a speed reducing box, the servo motor is arranged on the test bed base through a servo motor mounting seat, and the servo motor is connected with a PLC;
the wind power planetary gear speed increasing box to be tested is respectively connected with the vibration signal detection module, the oil information detection module and the noise detection module; and the vibration signal detection module, the oil information detection module and the noise detection module are respectively connected with the PLC.
5. The test platform for the wind power planetary gear box fault state detection and assessment method based on vibration-noise-oil characteristic fusion according to claim 4, wherein the oil information detection module comprises a temperature sensor, an online dielectric constant sensor, an online viscosity sensor, an online abrasive particle monitoring sensor and a CMOS abrasive particle image sensor.
6. The test platform of the wind power planetary gear box fault state detection and evaluation method based on vibration-noise-oil characteristic fusion according to claim 4 is characterized in that one end of the wind power gear speed increasing box to be tested is sequentially connected with a fine filter, an oil pump, a coarse filter, a cooler and the oil information detection module through an oil pipe, and the oil pipe is connected to the other end of the wind power gear speed increasing box to be tested.
7. The test platform for the wind power planetary gear box fault state detection and assessment method based on vibration-noise-oil characteristic fusion according to claim 4, wherein the wind power gear speed increasing box to be tested is arranged on the test platform base through a wind power gear speed increasing box mounting seat, a load is arranged on the test platform base through a load mounting seat, a fixed shaft gear box is arranged on the test platform base through a fixed shaft gear box mounting seat, and a reduction box is arranged on the test platform base through a reduction box seat.
8. The test platform for the wind power planetary gear box fault state detection and evaluation method based on vibration-noise-oil characteristic fusion is characterized in that the vibration signal detection module comprises a pulse signal acquisition device and a plurality of vibration acceleration sensors, wherein the pulse signal acquisition device is a photoelectric encoder arranged on an input shaft of a wind power gear speed increasing box to be tested, the vibration acceleration sensors are respectively arranged on bearing seats and box bodies at two ends of the wind power gear speed increasing box, and the vibration acceleration sensors and the pulse signal acquisition device are used for resampling vibration signals acquired at equal angles and transmitting the vibration signals to the data acquisition module.
9. The test platform for the wind power planetary gear box fault state detection and evaluation method based on vibration-noise-oil characteristic fusion according to claim 4, wherein the PLC is connected with a servo driver, and the servo driver controls the servo motor to operate; the servo motor is provided with a built-in encoder, and the built-in encoder feeds back the operation parameters of the servo motor to the PLC controller, so that the closed-loop control of the rotating speed and the torque of the servo motor is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810689692.3A CN108507787B (en) | 2018-06-28 | 2018-06-28 | Wind power gear speed increasing box fault diagnosis test platform and method based on multi-feature fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810689692.3A CN108507787B (en) | 2018-06-28 | 2018-06-28 | Wind power gear speed increasing box fault diagnosis test platform and method based on multi-feature fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108507787A CN108507787A (en) | 2018-09-07 |
| CN108507787B true CN108507787B (en) | 2024-03-15 |
Family
ID=63404076
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810689692.3A Active CN108507787B (en) | 2018-06-28 | 2018-06-28 | Wind power gear speed increasing box fault diagnosis test platform and method based on multi-feature fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108507787B (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109597315B (en) * | 2018-10-31 | 2020-08-18 | 华中科技大学 | Method, equipment and system for identifying health degradation state of mechanical equipment |
| CN109917287B (en) * | 2019-03-20 | 2021-06-08 | 华南理工大学 | Speed reduction motor quality inspection method based on empirical mode decomposition and octave spectrum analysis |
| CN110146287B (en) * | 2019-06-10 | 2021-04-27 | 东北大学 | Reliability test bed for speed increaser of offshore wind driven generator |
| CN110378042B (en) * | 2019-07-23 | 2020-12-29 | 山东大学 | Wind turbine generator gearbox oil temperature anomaly detection method and system based on SCADA data |
| CN110646200B (en) * | 2019-09-04 | 2021-10-22 | 天津工业大学 | Method for diagnosing weak local fault of gear of fixed shaft gear box |
| CN110844111B (en) * | 2019-10-11 | 2022-11-04 | 中国直升机设计研究所 | Multi-characteristic index bevel gear health state assessment method |
| CN111260640B (en) * | 2020-01-13 | 2023-03-31 | 重庆大学 | Tree generator network gear pitting image measuring method and device based on cyclean |
| CN111795817A (en) * | 2020-07-27 | 2020-10-20 | 西安交通大学 | RV reduction gear capability test device based on many sensing fuse |
| CN112362368A (en) * | 2021-01-14 | 2021-02-12 | 西门子交通技术(北京)有限公司 | Fault diagnosis method, device and system for train traction motor and readable medium |
| CN112729817B (en) * | 2021-02-04 | 2022-11-29 | 西南交通大学 | Gearbox fault diagnosis method, system and test bed based on multi-source data fusion |
| CN113340592A (en) * | 2021-06-03 | 2021-09-03 | 河北工业大学 | Involute gear abrasion and dynamics comprehensive experiment table considering external vibration excitation |
| CN113532848A (en) * | 2021-06-24 | 2021-10-22 | 南京工业大学 | Fault diagnosis system for planetary gear box |
| CN114018577B (en) * | 2021-09-28 | 2023-11-21 | 北京华控智加科技有限公司 | Equipment noise source imaging method and device, electronic equipment and storage medium |
| CN114063601A (en) * | 2021-11-12 | 2022-02-18 | 江苏核电有限公司 | Equipment state diagnosis system and method based on artificial intelligence |
| CN114136611A (en) * | 2021-12-01 | 2022-03-04 | 哈尔滨理工大学 | Gear reducer fault detection system |
| CN114383877A (en) * | 2021-12-31 | 2022-04-22 | 江苏核电有限公司 | System for be used for nuclear power station mechanical equipment failure diagnosis |
| CN115182983A (en) * | 2022-06-17 | 2022-10-14 | 同济大学 | Gear surface online monitoring imaging device based on gear box and control method |
| CN115630334B (en) * | 2022-10-26 | 2024-05-03 | 山东科技大学 | CBAM-based cross-domain mechanical fault diagnosis method based on multi-channel feature fusion and application |
| CN116091445B (en) * | 2023-01-05 | 2024-01-02 | 中国长江电力股份有限公司 | Ship lift gear and gear surface damage identification method based on deep learning |
| CN116379043B (en) * | 2023-04-11 | 2024-03-08 | 浙江勤鹏科技股份有限公司 | Fault detection method and system for spin-on oil filter |
| CN116758078B (en) * | 2023-08-18 | 2023-11-17 | 中科航迈数控软件(深圳)有限公司 | Speed change gear box observation method, system, terminal equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101566523A (en) * | 2009-05-11 | 2009-10-28 | 中能电力科技开发有限公司 | Method for monitoring state of gear case of wind generating set |
| CN103234585A (en) * | 2013-04-16 | 2013-08-07 | 内蒙古航天亿久科技发展有限责任公司 | Online monitoring and fault diagnosis system of large wind turbine units |
| CN103364189A (en) * | 2013-07-15 | 2013-10-23 | 中国水利电力物资华南公司 | Online fault diagnosis system of wind turbine generator gear case |
| CN103604601A (en) * | 2013-10-21 | 2014-02-26 | 浙江大学 | Fault diagnosis experiment platform based on wind power gearbox working condition simulation |
| CN104697787A (en) * | 2015-03-20 | 2015-06-10 | 山东大学 | Gearbox test bed based on multi-information fusion and detection method thereof |
| CN104764489A (en) * | 2015-03-27 | 2015-07-08 | 西安交通大学 | Online monitoring method for lubricating oil of wind-power transmission |
| CN106383028A (en) * | 2016-09-05 | 2017-02-08 | 北京航空航天大学 | Gear case fault diagnosis method |
| CN208254802U (en) * | 2018-06-28 | 2018-12-18 | 山东大学 | Wind power gear speed increase box fault diagnostic test platform based on multi-feature fusion |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8110936B2 (en) * | 2008-07-30 | 2012-02-07 | Hankuk Relay Co., Ltd. | Power transmission apparatus for wind power generation and wind power generator using the same |
-
2018
- 2018-06-28 CN CN201810689692.3A patent/CN108507787B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101566523A (en) * | 2009-05-11 | 2009-10-28 | 中能电力科技开发有限公司 | Method for monitoring state of gear case of wind generating set |
| CN103234585A (en) * | 2013-04-16 | 2013-08-07 | 内蒙古航天亿久科技发展有限责任公司 | Online monitoring and fault diagnosis system of large wind turbine units |
| CN103364189A (en) * | 2013-07-15 | 2013-10-23 | 中国水利电力物资华南公司 | Online fault diagnosis system of wind turbine generator gear case |
| CN103604601A (en) * | 2013-10-21 | 2014-02-26 | 浙江大学 | Fault diagnosis experiment platform based on wind power gearbox working condition simulation |
| CN104697787A (en) * | 2015-03-20 | 2015-06-10 | 山东大学 | Gearbox test bed based on multi-information fusion and detection method thereof |
| CN104764489A (en) * | 2015-03-27 | 2015-07-08 | 西安交通大学 | Online monitoring method for lubricating oil of wind-power transmission |
| CN106383028A (en) * | 2016-09-05 | 2017-02-08 | 北京航空航天大学 | Gear case fault diagnosis method |
| CN208254802U (en) * | 2018-06-28 | 2018-12-18 | 山东大学 | Wind power gear speed increase box fault diagnostic test platform based on multi-feature fusion |
Non-Patent Citations (6)
| Title |
|---|
| 变速行星齿轮系统故障诊断方法;桂勇;韩勤锴;李峥;褚福磊;;振动.测试与诊断(第02期);第220-226页 * |
| 基于DS证据理论的水下目标识别融合推理方法研究;唐政等;现代导航(第2期);第113-117页 * |
| 基于决策层的提升机减速器故障诊断方法研究;王绍进等;煤矿机械;第34卷(第05期);第282-284页 * |
| 基于深度学习的重载车辆齿轮箱的云故障诊断研究;陈康;陈良;耿臣露;王建明;;自动化应用(第12期);全文 * |
| 风机齿轮箱故障诊断与预警方法的研究进展;林丽;高建华;经昊达;王海洋;张向军;张秀丽;;仪器仪表与分析监测(第01期);全文 * |
| 风电机组齿轮箱故障诊断预警分析;陈园艺;中国优秀硕士学位论文全文数据库（工程科技Ⅱ辑）(第03期);第28-46页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108507787A (en) | 2018-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108507787B (en) | Wind power gear speed increasing box fault diagnosis test platform and method based on multi-feature fusion | |
| CN208254802U (en) | Wind power gear speed increase box fault diagnostic test platform based on multi-feature fusion | |
| CN103823409B (en) | Digit Control Machine Tool machining state multiparameter online actively monitoring system and its implementation | |
| CN106514433B (en) | A kind of magazine reliability test method and condition monitoring system | |
| CN102848266B (en) | Machine tool spindle accuracy prediction method | |
| CN111043023A (en) | Fracturing pump on-line monitoring and fault diagnosis system | |
| CN113343528B (en) | Shafting fatigue damage prediction method based on cross-point frequency response and dynamic response characteristic fusion | |
| CN113109666B (en) | Rail circuit fault diagnosis method based on deep convolutional neural network | |
| CN112943911B (en) | Wind turbine generator system gear box lubricating oil on-line monitoring device, monitoring method and system | |
| CN109542045A (en) | Numerically-controlled machine tool predictive maintenance standard testing bed and operating method | |
| CN103775832A (en) | Transient flow problem method-based oil pipeline dropping detection device | |
| CN102889988B (en) | Precision prediction method of ball screw pair | |
| CN106441843B (en) | A kind of rotating machinery fault method for waveform identification | |
| CN106228472A (en) | A kind of substation equipment intelligent patrol detection failure analysis methods | |
| CN203825002U (en) | Wind power generation gearbox lubricant online monitor system | |
| CN208456598U (en) | Shield owner's hydraulic monitoring system | |
| CN115614292A (en) | Vibration monitoring device and method for vertical water pump unit | |
| CN113155443B (en) | Lubricating oil state monitoring and fault diagnosis system and method for reducer of coal mining machine | |
| He et al. | Real-time fault diagnosis of motor bearing via improved cyclostationary analysis implemented onto edge computing system | |
| CN206351386U (en) | A kind of magazine condition monitoring system | |
| CN2771505Y (en) | Hydroelectric generating set monitoring and expert fuzzy diagnostic apparatus | |
| Zhukovytskyy et al. | Information-measuring test system of diesel locomotive hydraulic transmissions | |
| CN208431106U (en) | Wind turbines yaw system monitors mechanism on-line | |
| Liu et al. | Fault diagnosis of diesel engine information fusion based on adaptive dynamic weighted hybrid distance-taguchi method (ADWHD-T) | |
| CN1424566A (en) | Compact digital real time monitoring and controlling system for service life of large construction |
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 |