CN113027696A - Fault diagnosis method and device of hydraulic variable pitch system - Google Patents

Fault diagnosis method and device of hydraulic variable pitch system Download PDF

Info

Publication number
CN113027696A
CN113027696A CN201911346697.7A CN201911346697A CN113027696A CN 113027696 A CN113027696 A CN 113027696A CN 201911346697 A CN201911346697 A CN 201911346697A CN 113027696 A CN113027696 A CN 113027696A
Authority
CN
China
Prior art keywords
fault
training
root cause
data
feature vectors
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.)
Granted
Application number
CN201911346697.7A
Other languages
Chinese (zh)
Other versions
CN113027696B (en
Inventor
刘众
肖飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jinfeng Technology Co ltd
Original Assignee
Xinjiang Goldwind Science and Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xinjiang Goldwind Science and Technology Co Ltd filed Critical Xinjiang Goldwind Science and Technology Co Ltd
Priority to CN201911346697.7A priority Critical patent/CN113027696B/en
Publication of CN113027696A publication Critical patent/CN113027696A/en
Application granted granted Critical
Publication of CN113027696B publication Critical patent/CN113027696B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D7/00—Controlling wind motors 
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24133—Distances to prototypes
    • G06F18/24143—Distances to neighbourhood prototypes, e.g. restricted Coulomb energy networks [RCEN]
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F05—INDEXING SCHEMES RELATING TO ENGINES OR PUMPS IN VARIOUS SUBCLASSES OF CLASSES F01-F04
    • F05B—INDEXING SCHEME RELATING TO WIND, SPRING, WEIGHT, INERTIA OR LIKE MOTORS, TO MACHINES OR ENGINES FOR LIQUIDS COVERED BY SUBCLASSES F03B, F03D AND F03G
    • F05B2270/00—Control
    • F05B2270/30—Control parameters, e.g. input parameters
    • F05B2270/328—Blade pitch angle
    • 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

Abstract

A fault diagnosis method and device for a hydraulic variable propeller system are provided. The fault diagnosis method comprises the following steps: acquiring fault data of a hydraulic variable pitch system; converting the fault data into fault feature vectors; calculating similarities between the fault feature vector and a plurality of training feature vectors in a training data set by using an unsupervised learning algorithm, wherein the training feature vectors are fault feature vectors with root cause labels; and determining the root cause of the training feature vector with the highest similarity as the root cause of the fault data. The fault diagnosis method can be used for quickly and accurately positioning the fault root cause of the hydraulic variable pitch system.

Description

Fault diagnosis method and device of hydraulic variable pitch system
Technical Field
The present invention relates generally to the field of fault diagnosis of wind turbine generators, and more particularly, to a method and apparatus for performing fault diagnosis on a hydraulic pitch control system of a wind turbine generator system using an unsupervised learning algorithm.
Background
Energy is the main material basis of social economy and human life, and is the power of social development. However, the reserves of non-renewable energy sources such as petroleum, coal, natural gas, etc., which are major pillars of world energy, are decreasing, wind power generation is being developed in various countries of the world, and wind power generation has become a mature scale as a new energy source.
Wind turbine generators are devices that convert wind energy into electrical energy. The variable pitch device is an electric scheme that a speed-adjustable motor drives blades through a gear or a toothed belt or a hydraulic variable pitch system that a hydraulic cylinder is controlled by an electromagnetic valve to directly act on a variable pitch bearing. At present, a hydraulic pitch control system widely applied to an offshore large megawatt unit comprises a hydraulic station (mainly comprising an oil pump, an oil tank, a heat dissipation system and related sensors) positioned in a cabin, an actuating mechanism (mainly comprising a hydraulic cylinder, an energy accumulator, a control valve group and related sensors) positioned on a hub, a rotary joint for connection, a pipeline and the like.
As shown in fig. 1, the conventional hydraulic pitch control systems are all open valve control systems, constituent components are distributed between a nacelle and an impeller and between components, tens of pipelines and cables are required to be connected, and a sensor for system state detection cannot directly detect the state of each component, each section of pipeline and each cable. The system has the advantages that high distribution and high fault points caused by a plurality of pipeline cables are achieved, and the detectability of potential faults is restrained due to economy and realizability, so that the existing system has the defects that a plurality of fault points are generated and the quick and accurate positioning is difficult, and the difficulty is brought to the prediction and accurate maintenance of the system operation state.
Disclosure of Invention
Exemplary embodiments of the present invention aim to overcome the above-mentioned disadvantages of the difficulty of fast and accurate localization of fault roots.
According to an exemplary embodiment of the present invention, there is provided a fault diagnosis method of a hydraulic variable propeller system, including: acquiring fault data of a hydraulic variable pitch system; converting the fault data into fault feature vectors; calculating similarities between the fault feature vector and a plurality of training feature vectors in a training data set by using an unsupervised learning algorithm, wherein the training feature vectors are fault feature vectors with root cause labels; selecting a root cause of the one or more training feature vectors with the highest similarity as a root cause of the fault data.
Optionally, the fault data may include a fault type and at least one of the following fault signatures: whether the machine is stopped, the pressure is high or low, the flow is high or low, the oil temperature is high or low, the vibration is high or low, and the IO state is detected.
Optionally, the step of converting the fault data into a fault feature vector may include: the fault data are numbered according to fault types, and values of the features included in the fault data are respectively converted into values of vector elements of a one-dimensional vector to generate a one-dimensional vector with a fault number as a fault feature vector.
Alternatively, the values of the respective features included in the fault data may be binary values, and the values of the vector elements of the one-dimensional vector may be decimal values.
Optionally, the unsupervised learning algorithm may include at least one of KNN and KD-tree.
Optionally, the step of calculating the similarity between the fault feature vector and a plurality of training feature vectors in a training data set using an unsupervised learning algorithm may include: when the number of training feature vectors in a training data set is smaller than a preset threshold value, calculating the similarity between the fault feature vector and a plurality of training feature vectors in the training data set by using KNN; and when the number of the training feature vectors in the training data set is equal to or larger than a preset threshold value, calculating the similarity between the fault feature vector and the plurality of training feature vectors in the training data set by using the KD-tree.
Optionally, the step of determining the root cause of the training feature vector with the highest similarity as the root cause of the fault data may include: when a plurality of training feature vectors with the highest similarity exist, determining the root cause of the training feature vectors with the highest similarity as the possible root cause of fault data of the hydraulic pitch system.
Optionally, the step of determining the root cause of the training feature vector with the highest similarity as the root cause of the fault data may include: when a plurality of training feature vectors with the highest similarity exist, determining the root with the largest number from the root of the training feature vectors with the highest similarity as the root of fault data of the hydraulic pitch system.
Optionally, the fault diagnosis method of the hydraulic pitch system may further include: and when the root cause of the fault data is determined, attaching the label of the determined root cause to the fault feature vector, and adding the fault feature vector attached with the label of the root cause as a new training feature vector into a training data set.
According to an exemplary embodiment of the present invention, there is provided a failure diagnosis device of a hydraulic variable propeller system, characterized by comprising: the data acquisition module is configured to acquire fault data of the hydraulic variable pitch system; a data conversion module configured to convert the fault data into a fault feature vector; a root cause diagnosis module configured to calculate similarities between the fault feature vector and a plurality of training feature vectors in a training data set using an unsupervised learning algorithm, and select a root cause of one or more training feature vectors with the highest similarity as a root cause of the fault data, wherein a training feature vector is a fault feature vector having a root cause label.
Optionally, the fault data may include a fault type and at least one of the following fault signatures: whether the machine is stopped, the pressure is high or low, the flow is high or low, the oil temperature is high or low, the vibration is high or low, and the IO state is detected.
Optionally, the data conversion module may be configured to: the fault data are numbered according to fault types, and values of the features included in the fault data are respectively converted into values of vector elements of a one-dimensional vector to generate a one-dimensional vector with a fault number as a fault feature vector.
Alternatively, the values of the respective features included in the fault data may be binary values, and the values of the vector elements of the one-dimensional vector may be decimal values.
Optionally, the unsupervised learning algorithm may include at least one of KNN and KD-tree.
Optionally, the root cause diagnostic module may be configured to: when the number of training feature vectors in a training data set is smaller than a preset threshold value, calculating the similarity between the fault feature vector and a plurality of training feature vectors in the training data set by using KNN; and when the number of the training feature vectors in the training data set is equal to or larger than a preset threshold value, calculating the similarity between the fault feature vector and the plurality of training feature vectors in the training data set by using the KD-tree.
Optionally, the root cause diagnostic module may be configured to: when a plurality of training feature vectors with the highest similarity exist, determining the root cause of the training feature vectors with the highest similarity as the possible root cause of fault data of the hydraulic pitch system.
Optionally, the root cause diagnostic module may be configured to: when a plurality of training feature vectors with the highest similarity exist, determining the root with the largest number from the root of the training feature vectors with the highest similarity as the root of fault data of the hydraulic pitch system.
Optionally, the fault diagnosis device of the hydraulic pitch system may further include: and the training set updating module is configured to label the fault feature vector with the determined root cause when the root cause of the fault data is determined, and add the fault feature vector labeled with the root cause as a new training feature vector into a training data set.
According to an exemplary embodiment of the invention, a system is provided comprising at least one computing device and at least one storage device storing instructions, wherein the instructions, when executed by the at least one computing device, cause the at least one computing device to perform a fault diagnosis method of a hydraulic pitch system according to the invention.
According to an exemplary embodiment of the invention, a computer-readable storage medium storing instructions is provided, wherein the instructions, when executed by at least one computing device, cause the at least one computing device to perform a method of fault diagnosis of a hydraulic pitch system according to the invention.
According to the fault diagnosis method and device of the hydraulic pitch system, the unsupervised learning algorithm is used for carrying out fault diagnosis on the hydraulic pitch system of the wind generating set, and the fault root of the hydraulic pitch system can be subjected to on-line self-learning diagnosis and positioning so as to quickly and accurately guide operation and maintenance personnel.
Drawings
These and/or other aspects and advantages of the present invention will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic view of a prior art hydraulic pitch system;
FIG. 2 is a flow chart illustrating a fault diagnosis method of a hydraulic pitch system according to an exemplary embodiment of the invention;
fig. 3 is an exemplary schematic diagram illustrating a KNN algorithm according to an exemplary embodiment of the present invention;
FIG. 4 is an exemplary schematic diagram illustrating a KD-tree algorithm in accordance with an exemplary embodiment of the present invention;
fig. 5 is a block diagram illustrating a fault diagnosis apparatus of a hydraulic pitch system according to an embodiment of the present invention.
Detailed Description
The following description with reference to the accompanying drawings is provided to assist in a comprehensive understanding of embodiments of the invention defined by the claims and their equivalents. Various specific details are included to aid understanding, but these are to be considered exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. In addition, descriptions of well-known functions and constructions are omitted for clarity and conciseness.
When the hydraulic variable pitch system has a fault, an implicit relationship exists between the fault expression, the fault characteristics and the fault root, for example, table 1 shows exemplary diagnosis contents of the hydraulic variable pitch system.
[ Table 1]
Figure BDA0002333567630000041
Figure BDA0002333567630000051
For example, when a fault manifests as a unit running at limited power or shutting down, the fault signature "system low pressure" can be obtained, the fault root of which can be oil contamination pump damage. The invention adopts an unsupervised learning algorithm to make the implicit relationship among the fault expression, the fault characteristics and the fault root of the hydraulic variable-pitch system explicit through data science and computer technology.
FIG. 2 is a flow chart illustrating a fault diagnosis method of a hydraulic pitch system according to an embodiment of the invention.
In step 201, fault data of the hydraulic pitch system may be acquired.
According to the embodiment of the invention, when the hydraulic pitch system has a fault, fault data of the hydraulic pitch system can be obtained from the fault log. Here, the fault data of the hydraulic pitch system may comprise fault types, e.g. fault location and fault behaviour. The fault data of the hydraulic pitch system may further comprise various fault characteristics, for example, including at least one of whether to shut down, pressure high and low, flow high and low, oil temperature high and low, vibration high and low, IO status, etc. Wherein the IO states may represent a set of digital quantity states of outputs and feedbacks of the plurality of solenoid valves and sensors.
According to an exemplary embodiment of the present invention, a binary value may be assigned to the fault signature. For example, if the first fault signature "down" is yes, a binary value "1" is assigned to the first fault signature "down", and if the first fault signature "down" is no, a binary value "0" is assigned to the first fault signature "down". If the second fault signature "pressure" is high, a binary value "1" is assigned to the second fault signature "pressure", and if the second fault signature "pressure" is low, a binary value "0" is assigned to the second fault signature "pressure". If the third fault signature "flow" is high, a binary value "1" is assigned to the third fault signature "flow", and if the third fault signature "flow" is low, a binary value "0" is assigned to the third fault signature "flow". If the fourth fault characteristic "oil temperature" is high, a binary value "1" is assigned to the fourth fault characteristic "oil temperature", and if the fourth fault characteristic "oil temperature" is low, a binary value "0" is assigned to the fourth fault characteristic "oil temperature". If the fifth fault signature "vibrate" is high, a binary value "1" is assigned to the fifth fault signature "vibrate", and if the fifth fault signature "vibrate" is low, a binary value "0" is assigned to the fifth fault signature "vibrate". Further, the sixth failure characteristic "IO state" may be assigned a predetermined number of binary values, for example, "0000", "0001", "0010", "0011", "0100", "0101", "0110", "0111", "1000", "1001", "1010", "1011", "1100", "1101", "1110", "1111", in accordance with the output and feedback indicated by the sixth failure characteristic "IO state". The above assignment of binary values is exemplary only, and the binary values may also be assigned to the fault signature according to other means available in the art.
In step 202, fault data of the hydraulic pitch system may be converted into fault feature vectors.
According to an exemplary embodiment of the present invention, the fault data may be numbered by fault type and values of respective features included in the fault data may be respectively converted into values of vector elements of a one-dimensional vector to generate a one-dimensional vector having a fault number as a fault feature vector. Here, the binary eigenvalues may be converted into decimal vector elements.
In particular, the fault number i may be represented as a subscript of the fault signature vector V, and its vector elements may be represented as a1，a2，…，anWherein n represents the number of failure features, and n is an integer equal to or greater than 1. Thus, the fault feature vector V may be represented as Vi＝(a1,a2,…,an). For example, when the fault data of the hydraulic pitch system is of a first fault type (i ═ 0001), and the fault data of the first fault type includes fault signatures "shutdown" 1, "pressure" 1, "IO state" 1111, the fault data may be converted into a fault signatureVector V0001＝(1,1,16)。
In step 203, a similarity between the fault feature vector and a plurality of training feature vectors in a training data set may be calculated using an unsupervised learning algorithm, wherein the training feature vectors are fault feature vectors with root cause labels.
According to an exemplary embodiment of the present invention, the training data set may include a plurality of training feature vectors, and the plurality of training feature vectors may be stored in a matrix form according to a mapping relationship between each vector element and a physical meaning in the vector according to a fault number.
For example, a plurality of training feature vectors stored in the form of a matrix are shown in tables 2 and 3 below.
[ Table 2]
Fault numbering Characteristic (shut down) Characteristic (pressure) Feature (IO State) Question classification (root cause)
0001 1 (YES) 1 (high) 1111 Leakage of
0001 0 (NO) 1 (high) 0000 Overload
0001 1 (No) 0 (Low) 0101 Clamping stagnation
0001 0 (NO) 1 (high) 1110 Interference
0001 1 (YES) 0 (Low) 1010 Pollution (b) by
0001 1 (YES) 0 (Low) 0110 Short circuit
[ Table 3]
Fault numbering Characteristic (shut down) Characteristic (pressure) Feature (IO State) Question classification (root cause)
0002 0 (NO) 1 (high) 1111 Leakage of
0002 1 (YES) 1 (high) 0010 Overload
0002 1 (No) 0 (Low) 0101 Clamping stagnation
0002 0 (NO) 1 (high) 0110 Interference
0002 1 (YES) 0 (Low) 0000 Pollution (b) by
0002 1 (YES) 0 (Low) 0111 Short circuit
According to an example embodiment of the present invention, the unsupervised learning algorithm may include at least one of KNN and KD-tree. Here, KNN is K nearest neighbor algorithm, i.e. if most of K nearest neighbor samples of a sample in the feature space belong to a certain class, the sample also belongs to this class and has the characteristics of the samples on this class. The KD-Tree is a data structure for partitioning a k-dimensional data space, is mainly applied to searching key data in the multi-dimensional space (such as range search and nearest neighbor search), and is essentially feature matching when data search is carried out in the KD-Tree.
According to an exemplary embodiment of the present invention, when the number of training feature vectors in the training data set is less than a predetermined threshold, the similarity between the fault feature vector and the plurality of training feature vectors in the training data set is calculated using KNN. When the number of training feature vectors in the training data set is equal to or greater than a predetermined threshold, in order to reduce the temporal and spatial complexity of the algorithm, the similarity between the fault feature vector and the plurality of training feature vectors in the training data set is calculated using the KD-tree.
Specifically, fig. 3 is an exemplary schematic diagram for calculating a similarity between the fault feature vector and a plurality of training feature vectors in a training data set using KNN. As shown in fig. 3, the triangle and the square respectively represent fault feature vectors with a class of root cause labels, which may be, for example, interference and leakage, respectively, and the circle represents a fault feature vector to be classified, and the coordinate of the circle on the coordinate axis is determined by the value of an element in the fault feature vector, for example, the training feature vector in the training data set matrix and the fault feature vector to be classified may be represented on the coordinate axis. Therefore, the similarity of each training feature vector and the fault feature vector to be classified can be obtained by respectively calculating the geometric distance between each training feature vector and the fault feature vector to be classified, wherein the distance is inversely proportional to the similarity.
FIG. 4 is an exemplary schematic diagram for computing the similarity between the fault feature vector and a plurality of training feature vectors in a training data set using a KD-tree. As shown in FIG. 4, the k-d tree is a binary tree with each node being a k-dimensional point. All non-leaf nodes can be viewed as partitioning the space into two half-spaces with one hyperplane. The subtree to the left of the node represents a point to the left of the hyperplane and the subtree to the right of the node represents a point to the right of the hyperplane. The method for selecting the hyperplane is as follows: each node is associated with a dimension of the k dimensions that is perpendicular to the hyperplane. Thus, if the selection is divided according to the x-axis, all nodes with x values less than a specified value will appear in the left sub-tree and all nodes with x values greater than the specified value will appear in the right sub-tree. Thus, the hyperplane can be determined using this x value, with the normal being the unit vector of the x-axis. Therefore, the similarity between the fault feature vector to be classified and each training feature vector can be obtained by respectively calculating the distance between the fault feature vector to be classified and the hyperplane where each training feature vector is located, wherein the distance is inversely proportional to the similarity.
In step 204, the root cause of the training feature vector with the highest similarity may be determined as the root cause of the fault data of the hydraulic pitch system.
According to an exemplary embodiment of the invention, when there are a plurality of training feature vectors with the highest similarity, the root cause of these training feature vectors may be determined as the possible root cause of the fault data of the hydraulic pitch system.
According to an exemplary embodiment of the invention, when there are a plurality of training feature vectors with the highest similarity, the root with the highest number among the roots of the training feature vectors may be determined as the root of the fault data of the hydraulic pitch system. For example, most of the roots of these training feature vectors are leaks, then the root of the fault data of the hydraulic pitch system may be determined to be a leak.
According to an exemplary embodiment of the invention, the fault diagnosis method of the hydraulic pitch system may further include: when the root cause of the fault data of the hydraulic pitch system is determined, the fault feature vector can be attached with a label of the determined root cause, and the fault feature vector attached with the label of the root cause is used as a new training feature vector to be added into a training data set.
According to an exemplary embodiment of the invention, the fault diagnosis method of the hydraulic pitch system may further include: when the root cause of the fault data of the hydraulic pitch system is determined, the root cause of the fault data of the hydraulic pitch system and corresponding processing opinions can be pushed to a terminal of an operation and maintenance person through an OPCUA protocol for example so as to accurately guide the operation of the operation and maintenance person.
According to the fault diagnosis method of the hydraulic variable-pitch system, not only can the fault root of the hydraulic variable-pitch system be quickly and accurately positioned, but also the operation of operation and maintenance personnel can be accurately guided.
Fig. 5 is a block diagram illustrating a fault diagnosis apparatus of a hydraulic pitch system according to an embodiment of the present invention.
As shown in fig. 5, the fault diagnosis apparatus 500 of the hydraulic pitch system according to the embodiment of the invention may include a data acquisition module 501, a data conversion module 502, and a root cause diagnosis module 503.
The data acquisition module 501 may acquire fault data of the hydraulic pitch system.
According to an exemplary embodiment of the invention, when a fault occurs in the hydraulic pitch system, the data acquisition module 501 may acquire fault data of the hydraulic pitch system from the fault log. Here, the fault data of the hydraulic pitch system may comprise fault types, e.g. fault location and fault behaviour. The fault data of the hydraulic pitch system may further comprise various fault characteristics, for example, including at least one of whether to shut down, pressure high and low, flow high and low, oil temperature high and low, vibration high and low, IO status, etc. Wherein the IO states may represent a set of digital quantity states of outputs and feedbacks of the plurality of solenoid valves and sensors.
According to an exemplary embodiment of the present invention, the fault signature may be assigned a binary value. For example, if the first fault signature "down" is yes, a binary value "1" is assigned to the first fault signature "down", and if the first fault signature "down" is no, a binary value "0" is assigned to the first fault signature "down". If the second fault signature "pressure" is high, a binary value "1" is assigned to the second fault signature "pressure", and if the second fault signature "pressure" is low, a binary value "0" is assigned to the second fault signature "pressure". If the third fault signature "flow" is high, a binary value "1" is assigned to the third fault signature "flow", and if the third fault signature "flow" is low, a binary value "0" is assigned to the third fault signature "flow". If the fourth fault characteristic "oil temperature" is high, a binary value "1" is assigned to the fourth fault characteristic "oil temperature", and if the fourth fault characteristic "oil temperature" is low, a binary value "0" is assigned to the fourth fault characteristic "oil temperature". If the fifth fault signature "vibrate" is high, a binary value "1" is assigned to the fifth fault signature "vibrate", and if the fifth fault signature "vibrate" is low, a binary value "0" is assigned to the fifth fault signature "vibrate". Further, the sixth failure characteristic "IO state" may be assigned a predetermined number of binary values, for example, "0000", "0001", "0010", "0011", "0100", "0101", "0110", "0111", "1000", "1001", "1010", "1011", "1100", "1101", "1110", "1111", in accordance with the output and feedback indicated by the sixth failure characteristic "IO state". The above assignment of binary values is exemplary only, and the binary values may also be assigned to the fault signature according to other means available in the art.
The data conversion module 502 may convert fault data of the hydraulic pitch system into fault feature vectors.
According to an exemplary embodiment of the present invention, the data conversion module 502 may number the fault data by fault type and convert values of respective features included in the fault data into values of vector elements of a one-dimensional vector, respectively, to generate a one-dimensional vector having a fault number as a fault feature vector. Here, the binary eigenvalues may be converted into decimal vector elements.
Specifically, the data conversion module 502 may represent the fault number i as an index to the fault signature vector V and its vector elements as a1，a2，…，anWherein n represents the number of failure features, and n is an integer equal to or greater than 1. Thus, the fault feature vector V may be represented as Vi＝(a1,a2,…,an). For example, when the fault data of the hydraulic pitch system is of a first fault type (i ═ 0001), and the fault data of the first fault type includes fault signatures "shutdown" 1, "pressure" 1, "IO state" 1111, the data conversion module 502 may convert the fault data into a fault signature vector V0001＝(1,1,16)。
The root cause diagnostic module 503 may calculate similarities between the fault feature vector and a plurality of training feature vectors in a training data set using an unsupervised learning algorithm, wherein the training feature vectors are fault feature vectors with root cause labels.
According to an exemplary embodiment of the present invention, the training data set may include a plurality of training feature vectors, and the plurality of training feature vectors may be stored in a matrix form according to a mapping relationship between each vector element and a physical meaning in the vector according to a fault number.
According to an example embodiment of the present invention, the unsupervised learning algorithm may include at least one of KNN and KD-tree. Here, KNN is K nearest neighbor algorithm, i.e. if most of K nearest neighbor samples of a sample in the feature space belong to a certain class, the sample also belongs to this class and has the characteristics of the samples on this class. The KD-Tree is a data structure for partitioning a k-dimensional data space, is mainly applied to searching key data in the multi-dimensional space (such as range search and nearest neighbor search), and is essentially feature matching when data search is carried out in the KD-Tree.
According to an exemplary embodiment of the present invention, when the number of training feature vectors in the training data set is less than the predetermined threshold, the root cause diagnosis module 503 may calculate the similarity between the fault feature vector and the plurality of training feature vectors in the training data set using KNN. Specifically, the root cause diagnosis module 503 may obtain the similarity between each training feature vector and the fault feature vector to be classified by respectively calculating the geometric distance between each training feature vector and the fault feature vector to be classified, wherein the distance is inversely proportional to the similarity. When the number of training feature vectors in the training data set is equal to or greater than a predetermined threshold, the root cause diagnostic module 503 calculates similarities between the fault feature vector and the plurality of training feature vectors in the training data set using the KD-tree in order to reduce the temporal and spatial complexity of the algorithm. Specifically, the root cause diagnosis module 503 may obtain the similarity between the fault feature vector to be classified and each training feature vector by calculating the distance between the fault feature vector to be classified and the hyperplane where each training feature vector is located, respectively, where the distance is inversely proportional to the similarity.
The root cause diagnostic module 503 may determine the root cause of the training feature vector with the highest similarity as the root cause of the fault data of the hydraulic pitch system.
According to an example embodiment of the invention, when there are multiple training feature vectors with the highest similarity, the root cause diagnosis module 503 may determine the root causes of these training feature vectors as possible root causes of the fault data of the hydraulic pitch system.
According to an exemplary embodiment of the invention, when there are a plurality of training feature vectors with the highest similarity, the root cause diagnosis module 503 may determine the most numerous root causes among the root causes of these training feature vectors as the root causes of the fault data of the hydraulic pitch system. For example, most of the roots of these training feature vectors are leaks, the root cause diagnostic module 503 may determine the root cause of the fault data of the hydraulic pitch system as a leak.
According to an exemplary embodiment of the invention, the fault diagnosis method of the hydraulic pitch system may further comprise a training set update module (not shown). When the root cause of the fault data of the hydraulic pitch system is determined, the training set updating module can attach the fault feature vector with the label of the determined root cause, and the fault feature vector attached with the label of the root cause is used as a new training feature vector to be added into the training data set.
According to an exemplary embodiment of the invention, the fault diagnosis method of the hydraulic pitch system may further comprise a sending module (not shown). When the root cause of the fault data of the hydraulic pitch system is determined, the sending module can push the root cause of the fault data of the hydraulic pitch system and corresponding processing opinions to a terminal of an operation and maintenance person through an OPCUA protocol to accurately guide the operation of the operation and maintenance person.
According to the fault diagnosis method and device of the hydraulic variable-pitch system, not only can the fault root of the hydraulic variable-pitch system be quickly and accurately positioned, but also the operation of operation and maintenance personnel can be accurately guided.
The fault diagnosis method and apparatus of the hydraulic pitch system according to the exemplary embodiment of the present invention have been described above with reference to fig. 2 to 5.
The systems, devices and units shown in fig. 5 may each be configured as software, hardware, firmware, or any combination thereof that performs a particular function. For example, the system, apparatus or unit may correspond to an application specific integrated circuit, may correspond to pure software code, and may correspond to a module combining software and hardware. Further, one or more functions implemented by these systems, apparatuses, or units may also be uniformly executed by components in a physical entity device (e.g., processor, client, server, or the like).
Further, the method described with reference to fig. 2 may be implemented by a program (or instructions) recorded on a computer-readable storage medium. For example, according to an exemplary embodiment of the present invention, a computer readable storage medium for color image segmentation may be provided, wherein a computer program (or instructions) for performing the steps of the fault diagnosis method of the hydraulic pitch system described with reference to fig. 2 is recorded on the computer readable storage medium. For example, the computer program (or instructions) may be for performing the following method steps: acquiring fault data of a hydraulic variable pitch system; converting the fault data into fault feature vectors; calculating similarities between the fault feature vector and a plurality of training feature vectors in a training data set by using an unsupervised learning algorithm, wherein the training feature vectors are fault feature vectors with root cause labels; and determining the root cause of the training feature vector with the highest similarity as the root cause of the fault data.
The computer program in the computer-readable storage medium may be executed in an environment deployed in a computer device such as a client, a host, a proxy device, a server, and the like, and it should be noted that the computer program may also be used to perform additional steps other than the above steps or perform more specific processing when the above steps are performed, and the content of the additional steps and the further processing is already mentioned in the description of the related method with reference to fig. 2, and therefore will not be described again here to avoid repetition.
It should be noted that the fault diagnosis device of the hydraulic pitch system according to the exemplary embodiment of the present invention may fully rely on the operation of the computer program to implement the corresponding functions, i.e. each unit corresponds to each step in the functional architecture of the computer program, so that the whole system is called by a special software package (e.g. lib library) to implement the corresponding functions.
Alternatively, the various means shown in fig. 5 may be implemented by hardware, software, firmware, middleware, microcode, or any combination thereof. When implemented in software, firmware, middleware or microcode, the program code or code segments to perform the corresponding operations may be stored in a computer-readable medium such as a storage medium, so that a processor may perform the corresponding operations by reading and executing the corresponding program code or code segments.
For example, the exemplary embodiments of the present invention may also be realized as a computing device comprising a storage component and a processor, the storage component having stored therein a set of computer-executable instructions that, when executed by the processor, perform a fault diagnosis device of a hydraulic pitch system according to an exemplary embodiment of the present invention.
In particular, the computing devices may be deployed in servers or clients, as well as on node devices in a distributed network environment. Further, the computing device may be a PC computer, tablet device, personal digital assistant, smart phone, web application, or other device capable of executing the set of instructions described above.
The computing device need not be a single computing device, but can be any device or collection of circuits capable of executing the instructions (or sets of instructions) described above, individually or in combination. The computing device may also be part of an integrated control system or system manager, or may be configured as a portable electronic device that interfaces with local or remote (e.g., via wireless transmission).
In the computing device, the processor may include a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a programmable logic device, a special purpose processor system, a microcontroller, or a microprocessor. By way of example, and not limitation, processors may also include analog processors, digital processors, microprocessors, multi-core processors, processor arrays, network processors, and the like.
Certain operations described in the method for diagnosing a fault of a hydraulic pitch system according to an exemplary embodiment of the present invention may be implemented by software, certain operations may be implemented by hardware, and further, the operations may be implemented by a combination of hardware and software.
The processor may execute instructions or code stored in one of the memory components, which may also store data. Instructions and data may also be transmitted and received over a network via a network interface device, which may employ any known transmission protocol.
The memory component may be integral to the processor, e.g., having RAM or flash memory disposed within an integrated circuit microprocessor or the like. Further, the storage component may comprise a stand-alone device, such as an external disk drive, storage array, or any other storage device usable by a database system. The storage component and the processor may be operatively coupled or may communicate with each other, such as through an I/O port, a network connection, etc., so that the processor can read files stored in the storage component.
Further, the computing device may also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, mouse, touch input device, etc.). All components of the computing device may be connected to each other via a bus and/or a network.
The operations involved in the fault diagnosis method of a hydraulic pitch system according to an exemplary embodiment of the present invention may be described as various interconnected or coupled functional blocks or functional diagrams. However, these functional blocks or functional diagrams may be equally integrated into a single logic device or operated on by non-exact boundaries.
Thus, the method described with reference to FIG. 2 may be implemented by a system comprising at least one computing device and at least one storage device storing instructions.
According to an exemplary embodiment of the invention, the at least one computing device is a computing device for fault diagnosis of a hydraulic pitch system according to an exemplary embodiment of the invention, the storage device having stored therein a set of computer-executable instructions which, when executed by the at least one computing device, perform the method steps described with reference to fig. 2. For example, when the set of computer-executable instructions is executed by the at least one computing device, the following method steps may be performed: acquiring fault data of a hydraulic variable pitch system; converting the fault data into fault feature vectors; calculating similarities between the fault feature vector and a plurality of training feature vectors in a training data set by using an unsupervised learning algorithm, wherein the training feature vectors are fault feature vectors with root cause labels; and determining the root cause of the training feature vector with the highest similarity as the root cause of the fault data.
While exemplary embodiments of the invention have been described above, it should be understood that the above description is illustrative only and not exhaustive, and that the invention is not limited to the exemplary embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the invention. Therefore, the protection scope of the present invention should be subject to the scope of the claims.

Claims (20)

1. A fault diagnosis method of a hydraulic variable propeller system is characterized by comprising the following steps:
acquiring fault data of a hydraulic variable pitch system;
converting the fault data into fault feature vectors;
calculating similarities between the fault feature vector and a plurality of training feature vectors in a training data set by using an unsupervised learning algorithm, wherein the training feature vectors are fault feature vectors with root cause labels;
and determining the root cause of the training feature vector with the highest similarity as the root cause of the fault data.
2. The fault diagnostic method of claim 1, wherein the fault data includes a fault type and at least one of the following fault signatures: whether the machine is stopped, the pressure is high or low, the flow is high or low, the oil temperature is high or low, the vibration is high or low, and the IO state is detected.
3. The fault diagnosis method according to claim 2, wherein the step of converting the fault data into a fault feature vector comprises:
the fault data are numbered according to fault types, and values of the features included in the fault data are respectively converted into values of vector elements of a one-dimensional vector to generate a one-dimensional vector with a fault number as a fault feature vector.
4. The fault diagnosing method according to claim 3, wherein the values of the respective features included in the fault data are binary values, and the values of the vector elements of the one-dimensional vector are decimal values.
5. The fault diagnosis method according to claim 1, wherein the unsupervised learning algorithm includes at least one of KNN and KD-tree.
6. The fault diagnosis method according to claim 5, wherein the step of calculating the similarity between the fault feature vector and a plurality of training feature vectors in a training data set using an unsupervised learning algorithm comprises:
when the number of training feature vectors in a training data set is smaller than a preset threshold value, calculating the similarity between the fault feature vector and a plurality of training feature vectors in the training data set by using KNN;
and when the number of the training feature vectors in the training data set is equal to or larger than a preset threshold value, calculating the similarity between the fault feature vector and the plurality of training feature vectors in the training data set by using the KD-tree.
7. The fault diagnosis method according to claim 1, wherein the step of determining the root cause of the training feature vector having the highest similarity as the root cause of the fault data comprises:
when a plurality of training feature vectors with the highest similarity exist, determining the root cause of the training feature vectors with the highest similarity as the possible root cause of fault data of the hydraulic pitch system.
8. The fault diagnosis method according to claim 1, wherein the step of determining the root cause of the training feature vector having the highest similarity as the root cause of the fault data comprises:
when a plurality of training feature vectors with the highest similarity exist, determining the root with the largest number from the root of the training feature vectors with the highest similarity as the root of fault data of the hydraulic pitch system.
9. The fault diagnosis method according to claim 1, further comprising:
and when the root cause of the fault data is determined, attaching the label of the determined root cause to the fault feature vector, and adding the fault feature vector attached with the label of the root cause as a new training feature vector into a training data set.
10. A failure diagnosis device of a hydraulic variable pitch system, comprising:
the data acquisition module is configured to acquire fault data of the hydraulic variable pitch system;
a data conversion module configured to convert the fault data into a fault feature vector;
a root cause diagnosis module configured to calculate similarities between the fault feature vector and a plurality of training feature vectors in a training data set using an unsupervised learning algorithm, and to use a root cause of a training feature vector with a highest similarity as a root cause of the fault data, wherein the training feature vector is a fault feature vector having a root cause label.
11. The fault diagnostic apparatus according to claim 10, wherein the fault data includes a fault type and at least one of the following fault signatures: whether the machine is stopped, the pressure is high or low, the flow is high or low, the oil temperature is high or low, the vibration is high or low, and the IO state is detected.
12. The fault diagnostic apparatus of claim 11, wherein the data conversion module is configured to: the fault data are numbered according to fault types, and values of the features included in the fault data are respectively converted into values of vector elements of a one-dimensional vector to generate a one-dimensional vector with a fault number as a fault feature vector.
13. The fault diagnosis device according to claim 12, characterized in that the values of the respective features included in the fault data are binary values, and the values of the vector elements of the one-dimensional vector are decimal values.
14. The fault diagnosis device according to claim 10, wherein the unsupervised learning algorithm includes at least one of KNN and KD-tree.
15. The fault diagnosis device according to claim 14, characterized in that the root cause diagnosis module is configured to:
when the number of training feature vectors in a training data set is smaller than a preset threshold value, calculating the similarity between the fault feature vector and a plurality of training feature vectors in the training data set by using KNN;
and when the number of the training feature vectors in the training data set is equal to or larger than a preset threshold value, calculating the similarity between the fault feature vector and the plurality of training feature vectors in the training data set by using the KD-tree.
16. The fault diagnosis device according to claim 10, characterized in that the root cause diagnosis module is configured to: when a plurality of training feature vectors with the highest similarity exist, determining the root cause of the training feature vectors with the highest similarity as the possible root cause of fault data of the hydraulic pitch system.
17. The fault diagnosis device according to claim 10, characterized in that the root cause diagnosis module is configured to: when a plurality of training feature vectors with the highest similarity exist, determining the root with the largest number from the root of the training feature vectors with the highest similarity as the root of fault data of the hydraulic pitch system.
18. The failure diagnosis device according to claim 10, further comprising:
and the training set updating module is configured to label the fault feature vector with the determined root cause when the root cause of the fault data is determined, and add the fault feature vector labeled with the root cause as a new training feature vector into a training data set.
19. A system comprising at least one computing device and at least one storage device storing instructions, wherein the instructions, when executed by the at least one computing device, cause the at least one computing device to perform a method of fault diagnosis of a hydraulic pitch system according to any of claims 1 to 9.
20. A computer-readable storage medium storing instructions that, when executed by at least one computing device, cause the at least one computing device to perform the method of fault diagnosis of a hydraulic pitch system of any of claims 1 to 9.
CN201911346697.7A 2019-12-24 2019-12-24 Fault diagnosis method and device of hydraulic variable pitch system Active CN113027696B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911346697.7A CN113027696B (en) 2019-12-24 2019-12-24 Fault diagnosis method and device of hydraulic variable pitch system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911346697.7A CN113027696B (en) 2019-12-24 2019-12-24 Fault diagnosis method and device of hydraulic variable pitch system

Publications (2)

Publication Number Publication Date
CN113027696A true CN113027696A (en) 2021-06-25
CN113027696B CN113027696B (en) 2022-11-15

Family

ID=76451606

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911346697.7A Active CN113027696B (en) 2019-12-24 2019-12-24 Fault diagnosis method and device of hydraulic variable pitch system

Country Status (1)

Country Link
CN (1) CN113027696B (en)

Citations (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060208493A1 (en) * 2005-03-15 2006-09-21 General Electric Company Methods and apparatus for pitch control power conversion
US20080164697A1 (en) * 2007-01-05 2008-07-10 Christian Schram Method and apparatus for controlling rotary machines
EP2128437A2 (en) * 2008-05-29 2009-12-02 General Electric Company Method for increasing energy capture in a wind turbine
CA2664264A1 (en) * 2008-06-27 2009-12-27 Fred Nitzsche Hybrid device for vibration control
WO2010022735A2 (en) * 2008-08-29 2010-03-04 Vestas Wind Systems A/S Pitch control system
CN101750210A (en) * 2009-12-24 2010-06-23 重庆大学 Fault diagnosis method based on OLPP feature reduction
CA2739137A1 (en) * 2010-08-11 2012-02-11 Sinovel Wind Group Co., Ltd. Control method of automatic lubrication system of variable-pitch bearing used for wind turbine generator system
AU2011253963A1 (en) * 2011-03-09 2012-09-27 Siemens Aktiengesellschaft Method and arrangement for detecting a blade pitch angle unbalance of a rotor blade system of a wind turbine
US20120330495A1 (en) * 2011-06-23 2012-12-27 United Technologies Corporation Mfcc and celp to detect turbine engine faults
US20130108443A1 (en) * 2006-10-23 2013-05-02 General Electric Company Methods for operating a wind turbine
CN103826299A (en) * 2014-03-05 2014-05-28 上海永畅信息科技有限公司 Wireless signal sensation based indoor augmented reality realizing method
CN103835882A (en) * 2014-03-18 2014-06-04 上海电机学院 Large wind generating set state monitoring and fault diagnosis system
CN103927394A (en) * 2014-05-04 2014-07-16 苏州大学 Multi-label active learning classification method and system based on SVM
CN104102773A (en) * 2014-07-05 2014-10-15 山东鲁能软件技术有限公司 Equipment fault warning and state monitoring method
US20150023602A1 (en) * 2013-07-19 2015-01-22 Kamil Wnuk Fast recognition algorithm processing, systems and methods
CN104931263A (en) * 2015-06-18 2015-09-23 东南大学 Bearing fault diagnosis method based on symbolic probabilistic finite state machine
CN105134510A (en) * 2015-09-18 2015-12-09 北京中恒博瑞数字电力科技有限公司 State monitoring and failure diagnosis method for wind generating set variable pitch system
CN106197886A (en) * 2016-06-29 2016-12-07 上海交通大学 Hydraulic cylinder interior leakage leakage fault diagnosis appraisal procedure
CN107689059A (en) * 2017-06-30 2018-02-13 北京金风科创风电设备有限公司 The abnormal recognition methods of wind generating set pitch control and device
CN108073158A (en) * 2017-12-05 2018-05-25 上海电机学院 Based on PCA and KNN density algorithm Wind turbines Method for Bearing Fault Diagnosis
CN108182523A (en) * 2017-12-26 2018-06-19 新疆金风科技股份有限公司 The treating method and apparatus of fault data, computer readable storage medium
KR101911061B1 (en) * 2018-03-26 2018-10-23 주식회사 대곤코퍼레이션 System of product defect discrimination using an unsupervised learning based regional convolutional auto-encoder and Method of product defect discrimination using an unsupervised learning based regional convolutional auto-encoder
CN109362049A (en) * 2018-10-09 2019-02-19 中国人民解放军海军航空大学 Consistent square root volume filtering method is weighted based on mixed information
EP3462265A1 (en) * 2017-09-30 2019-04-03 Siemens Aktiengesellschaft Method, apparatus, and device for facilitating failure diagnosis and machine-readable medium
CN109829498A (en) * 2019-01-31 2019-05-31 北京细推科技有限公司 Rough sort method, apparatus, terminal device and storage medium based on clustering
CN109947080A (en) * 2019-03-21 2019-06-28 北京明略软件系统有限公司 A kind of method, apparatus of fault diagnosis, computer storage medium and terminal
CN110118657A (en) * 2019-06-21 2019-08-13 杭州安脉盛智能技术有限公司 Based on relative entropy and K nearest neighbor algorithm Fault Diagnosis of Roller Bearings and system
CN110162013A (en) * 2019-05-29 2019-08-23 上海理工大学 A kind of breakdown of refrigeration system diagnostic method
US20190317488A1 (en) * 2018-04-12 2019-10-17 Saudi Arabian Oil Company Predicting failures in electrical submersible pumps using pattern recognition
CN110374822A (en) * 2019-08-07 2019-10-25 电子科技大学 Blade of wind-driven generator method for diagnosing faults based on depth convolution self-encoding encoder and XGBoost

Patent Citations (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060208493A1 (en) * 2005-03-15 2006-09-21 General Electric Company Methods and apparatus for pitch control power conversion
US20130108443A1 (en) * 2006-10-23 2013-05-02 General Electric Company Methods for operating a wind turbine
US20080164697A1 (en) * 2007-01-05 2008-07-10 Christian Schram Method and apparatus for controlling rotary machines
EP2128437A2 (en) * 2008-05-29 2009-12-02 General Electric Company Method for increasing energy capture in a wind turbine
CA2664264A1 (en) * 2008-06-27 2009-12-27 Fred Nitzsche Hybrid device for vibration control
WO2010022735A2 (en) * 2008-08-29 2010-03-04 Vestas Wind Systems A/S Pitch control system
CN101750210A (en) * 2009-12-24 2010-06-23 重庆大学 Fault diagnosis method based on OLPP feature reduction
CA2739137A1 (en) * 2010-08-11 2012-02-11 Sinovel Wind Group Co., Ltd. Control method of automatic lubrication system of variable-pitch bearing used for wind turbine generator system
AU2011253963A1 (en) * 2011-03-09 2012-09-27 Siemens Aktiengesellschaft Method and arrangement for detecting a blade pitch angle unbalance of a rotor blade system of a wind turbine
US20120330495A1 (en) * 2011-06-23 2012-12-27 United Technologies Corporation Mfcc and celp to detect turbine engine faults
US20150023602A1 (en) * 2013-07-19 2015-01-22 Kamil Wnuk Fast recognition algorithm processing, systems and methods
CN103826299A (en) * 2014-03-05 2014-05-28 上海永畅信息科技有限公司 Wireless signal sensation based indoor augmented reality realizing method
CN103835882A (en) * 2014-03-18 2014-06-04 上海电机学院 Large wind generating set state monitoring and fault diagnosis system
CN103927394A (en) * 2014-05-04 2014-07-16 苏州大学 Multi-label active learning classification method and system based on SVM
CN104102773A (en) * 2014-07-05 2014-10-15 山东鲁能软件技术有限公司 Equipment fault warning and state monitoring method
CN104931263A (en) * 2015-06-18 2015-09-23 东南大学 Bearing fault diagnosis method based on symbolic probabilistic finite state machine
CN105134510A (en) * 2015-09-18 2015-12-09 北京中恒博瑞数字电力科技有限公司 State monitoring and failure diagnosis method for wind generating set variable pitch system
CN106197886A (en) * 2016-06-29 2016-12-07 上海交通大学 Hydraulic cylinder interior leakage leakage fault diagnosis appraisal procedure
CN107689059A (en) * 2017-06-30 2018-02-13 北京金风科创风电设备有限公司 The abnormal recognition methods of wind generating set pitch control and device
EP3462265A1 (en) * 2017-09-30 2019-04-03 Siemens Aktiengesellschaft Method, apparatus, and device for facilitating failure diagnosis and machine-readable medium
CN108073158A (en) * 2017-12-05 2018-05-25 上海电机学院 Based on PCA and KNN density algorithm Wind turbines Method for Bearing Fault Diagnosis
CN108182523A (en) * 2017-12-26 2018-06-19 新疆金风科技股份有限公司 The treating method and apparatus of fault data, computer readable storage medium
KR101911061B1 (en) * 2018-03-26 2018-10-23 주식회사 대곤코퍼레이션 System of product defect discrimination using an unsupervised learning based regional convolutional auto-encoder and Method of product defect discrimination using an unsupervised learning based regional convolutional auto-encoder
US20190317488A1 (en) * 2018-04-12 2019-10-17 Saudi Arabian Oil Company Predicting failures in electrical submersible pumps using pattern recognition
CN109362049A (en) * 2018-10-09 2019-02-19 中国人民解放军海军航空大学 Consistent square root volume filtering method is weighted based on mixed information
CN109829498A (en) * 2019-01-31 2019-05-31 北京细推科技有限公司 Rough sort method, apparatus, terminal device and storage medium based on clustering
CN109947080A (en) * 2019-03-21 2019-06-28 北京明略软件系统有限公司 A kind of method, apparatus of fault diagnosis, computer storage medium and terminal
CN110162013A (en) * 2019-05-29 2019-08-23 上海理工大学 A kind of breakdown of refrigeration system diagnostic method
CN110118657A (en) * 2019-06-21 2019-08-13 杭州安脉盛智能技术有限公司 Based on relative entropy and K nearest neighbor algorithm Fault Diagnosis of Roller Bearings and system
CN110374822A (en) * 2019-08-07 2019-10-25 电子科技大学 Blade of wind-driven generator method for diagnosing faults based on depth convolution self-encoding encoder and XGBoost

Non-Patent Citations (10)

* Cited by examiner, † Cited by third party
Title
史铁林等: "提高大型复杂机电系统故障诊断质量的几种新方法", 《机械工程学报》 *
叶春霖等: "基于数据挖掘的风电机组叶片结冰故障诊断", 《噪声与振动控制》 *
吴斌等: "基于改进的K最近邻分类器的风机故障诊断", 《机械设计与研究》 *
唐新安等: "基于谱峭度的风力发电机组轴承故障诊断方法", 《能源与节能》 *
张军峰等: "基于聚类和支持向量机的非线性时间序列故障预报", 《控制理论与应用》 *
李东东: "基于无监督特征学习的行星齿轮箱故障特征提取和检测", 《电网技术》 *
杨伟等: "基于RBF神经网络的风力发电机组故障诊断研究", 《电气传动自动化》 *
赵晓君等: "基于PCA-KNN聚类的通用在线故障诊断算法设计", 《计算机测量与控制》 *
邬强等: "旋翼航空器传动系统诊断技术的研发和改进", 《电子产品可靠性与环境试验》 *
马小娟等: "基于Zigbee无线技术和粒子滤波的风电机组无线监测系统故障定位研究", 《可再生能源》 *

Also Published As

Publication number Publication date
CN113027696B (en) 2022-11-15

Similar Documents

Publication Publication Date Title
CN111829782B (en) Fault diagnosis method based on adaptive manifold embedding dynamic distribution alignment
Pang et al. Spatio-temporal fusion neural network for multi-class fault diagnosis of wind turbines based on SCADA data
Meng et al. Semi-supervised software defect prediction model based on tri-training.
Han et al. Intelligent fault diagnosis method for rotating machinery via dictionary learning and sparse representation-based classification
CN102903089B (en) Method for generating remote sensing image quick view under Linux environment
Jia et al. A deviation based assessment methodology for multiple machine health patterns classification and fault detection
Gao et al. Rolling bearing fault diagnosis of PSO–LSSVM based on CEEMD entropy fusion
Zhang et al. A novel wind turbine fault diagnosis method based on compressed sensing and DTL-CNN
Zhao et al. Data augmentation via randomized wavelet expansion and its application in few-shot fault diagnosis of aviation hydraulic pumps
Song et al. Short-term forecasting based on graph convolution networks and multiresolution convolution neural networks for wind power
CN107301243A (en) Switchgear fault signature extracting method based on big data platform
Si et al. Intelligent fault diagnosis for industrial big data
Lei et al. A new transferable bearing fault diagnosis method with adaptive manifold probability distribution under different working conditions
CN115660232A (en) Ultra-short-term prediction method, device and system for wind power
Xu et al. Mvse: Effort-aware heterogeneous defect prediction via multiple-view spectral embedding
Wang et al. Cross-domain fault diagnosis of rotating machinery in nuclear power plant based on improved domain adaptation method
Liu et al. A deep generative model based on CNN-CVAE for wind turbine condition monitoring
Li et al. A fault diagnosis method for rotating machinery with semi-supervised graph convolutional network and images converted from vibration signals
CN113027696B (en) Fault diagnosis method and device of hydraulic variable pitch system
Chen et al. Unsupervised Fault Detection Driven by Multivariate Time Series for Aeroengines
Xu et al. A composite quantile regression long short-term memory network with group lasso for wind turbine anomaly detection
Zhao et al. Fault diagnosis of rolling bearings with noise signal based on modified kernel principal component analysis and DC‐ResNet
Pei et al. Degradation state identification for hydraulic pumps based on multi-scale ternary dynamic analysis, NSGA-II and SVM
Li et al. Alignment subdomain-based deep convolutional transfer learning for machinery fault diagnosis under different working conditions
Xu et al. Discriminative subspace embedded dynamic geometrical and statistical alignment based on pseudo-label correction for cross-domain bearing fault diagnosis

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
CP01 Change in the name or title of a patent holder

Address after: 830026 No. 107, Shanghai Road, Urumqi economic and Technological Development Zone, the Xinjiang Uygur Autonomous Region

Patentee after: Jinfeng Technology Co.,Ltd.

Address before: 830026 No. 107, Shanghai Road, Urumqi economic and Technological Development Zone, the Xinjiang Uygur Autonomous Region

Patentee before: XINJIANG GOLDWIND SCIENCE & TECHNOLOGY Co.,Ltd.

CP01 Change in the name or title of a patent holder