CN116680636A - Wind driven generator ice prediction method based on time sequence analysis - Google Patents

Wind driven generator ice prediction method based on time sequence analysis Download PDF

Info

Publication number
CN116680636A
CN116680636A CN202310693921.XA CN202310693921A CN116680636A CN 116680636 A CN116680636 A CN 116680636A CN 202310693921 A CN202310693921 A CN 202310693921A CN 116680636 A CN116680636 A CN 116680636A
Authority
CN
China
Prior art keywords
ice
index
driven generator
wind driven
time sequence
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310693921.XA
Other languages
Chinese (zh)
Inventor
蔺雪峰
耿杰
文军
谭光道
苗雷
徐超
孟秀俊
蔡春辉
汪德军
胡辉
付荣方
周世银
詹彪
朱玉瑞
孟鹏飞
易彦青
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.)
Shenyang Shengshi New Energy Technology Co ltd
Huaneng Weining Wind Power Co ltd
Huaneng Clean Energy Research Institute
China Huaneng Group Co Ltd
Original Assignee
Shenyang Shengshi New Energy Technology Co ltd
Huaneng Weining Wind Power Co ltd
Huaneng Clean Energy Research Institute
China Huaneng Group 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 Shenyang Shengshi New Energy Technology Co ltd, Huaneng Weining Wind Power Co ltd, Huaneng Clean Energy Research Institute, China Huaneng Group Co Ltd filed Critical Shenyang Shengshi New Energy Technology Co ltd
Priority to CN202310693921.XA priority Critical patent/CN116680636A/en
Publication of CN116680636A publication Critical patent/CN116680636A/en
Pending legal-status Critical Current

Links

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
    • F03D80/00—Details, components or accessories not provided for in groups F03D1/00 - F03D17/00
    • F03D80/40—Ice detection; De-icing means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • 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
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2123/00—Data types
    • G06F2123/02—Data types in the time domain, e.g. time-series data

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Sustainable Development (AREA)
  • Biomedical Technology (AREA)
  • Mechanical Engineering (AREA)
  • Chemical & Material Sciences (AREA)
  • Sustainable Energy (AREA)
  • Probability & Statistics with Applications (AREA)
  • Health & Medical Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Wind Motors (AREA)

Abstract

The invention relates to the technical field of wind driven generator ice prediction, and discloses a wind driven generator ice prediction method based on time sequence analysis, which comprises the following steps: acquiring the ice performance index time sequence data of the wind driven generator, and weighting the index data based on the ice performance index weight of the wind driven generator; encoding the weighted wind driven generator ice performance index time sequence data; optimizing the constructed wind power generator ice-tracking time sequence analysis model to obtain an optimal wind power generator ice-tracking time sequence analysis model, and obtaining the ice-tracking probability of the wind power generator by utilizing the optimal wind power generator ice-tracking time sequence analysis model. The invention determines the reference index weight of the index according to the information entropy of different indexes, and corrects the reference index weight according to the correlation among the indexes, wherein the larger the index weight of the ice state can be effectively detected, the accuracy of the subsequent ice probability prediction is improved, and the ice probability calculation is carried out by combining the index data cross coding result based on the continuously updated prior probability.

Description

Wind driven generator ice prediction method based on time sequence analysis
Technical Field
The invention relates to the technical field of wind driven generator ice-caterpillar prediction, in particular to a wind driven generator ice-caterpillar prediction method based on time sequence analysis.
Background
Nowadays, wind power generation is one of the main modes of electric energy production in China, and has the characteristics of strong cleanness and no pollution, and can effectively reduce the production cost and improve the utilization rate of resources and energy sources. However, in some wind power generation areas, as time goes by, the wind power generator gradually performs ice, so that the productivity of the wind power generator is seriously reduced, and the wind power generation efficiency and the working safety are affected. Aiming at the problem, the invention provides a method for predicting the ice of the wind driven generator by time sequence analysis, which is used for predicting the ice of the wind driven generator by collecting multidimensional ice indicator data to perform time sequence analysis, performing early protection and early warning and enhancing the reliability and safety of the work of the wind driven generator.
Disclosure of Invention
In view of the above, the present invention provides a method for predicting ice on a wind driven generator based on time sequence analysis, which aims to: 1) Determining the reference index weight of the index according to the information entropy of different indexes, wherein the larger the reference index weight of the index is, the higher the accuracy of dividing the non-ice state and the ice state of the wind driven generator by the index is, further fixing the index with the maximum reference index weight, calculating to obtain the correlation between other indexes and the fixed index data, calibrating the reference index weight of the other indexes, wherein the larger the correlation is, the closer the information entropy contained by the index and the fixed index is, the smaller the calibrated reference index weight is, and weighting the acquired time sequence data of the ice index of the wind driven generator to obtain the time sequence data of the index representing the importance of different indexes, wherein the larger the weight of the index capable of effectively detecting the ice state is, and the accuracy of the subsequent ice prediction is improved; 2) The method comprises the steps of respectively carrying out coding processing on time sequence data of each index, constructing a convolution matrix by combining the most obvious coding processing results of other indexes, carrying out convolution coding operation, realizing cross coding among different indexes, constructing a wind driven generator ice time sequence analysis model by combining prior probability distribution of ice probability, carrying out iterative solution on probability distribution parameters and likelihood functions of sample data, converging to obtain probability distribution parameters, carrying out ice probability calculation by combining the current index data cross coding results based on continuously updated prior probability, and realizing wind driven generator ice prediction by combining the time sequence data.
The invention provides a method for predicting ice of a wind driven generator through time sequence analysis, which comprises the following steps:
s1: acquiring the ice performance index time sequence data of the wind driven generator, and weighting the index data based on the ice performance index weight of the wind driven generator to obtain the weighted ice performance index time sequence data of the wind driven generator;
s2: encoding the weighted wind driven generator ice performance index time sequence data to obtain encoded index vector data;
s3: constructing an ice-caterpillar-sequence analysis model of the wind driven generator, wherein the model takes index vector data as input and ice-caterpillar probability as output;
s4: optimizing the constructed wind power generator ice-tracking time sequence analysis model to obtain an optimal wind power generator ice-tracking time sequence analysis model, and obtaining the ice-tracking probability of the wind power generator by utilizing the optimal wind power generator ice-tracking time sequence analysis model.
As a further improvement of the present invention:
optionally, collecting the ice performance index time sequence data of the wind driven generator in the step S1 includes:
collecting the ice-caterpillar index time sequence data of the wind driven generator, wherein the ice-caterpillar index of the wind driven generator comprises the blade rotating speed, the active power of the wind driven generator, the blade temperature, the blade humidity and the blade image of the wind driven generator, and the collected ice-caterpillar index time sequence data x of the wind driven generator is as follows:
x＝[x 1 ,x 2 ,x 3 ,x 4 ,x 5 ] T
x i ＝(x i (t 1 ),x i (t 2 ),...,x i (t n ),...,x i (t N )),i∈[1,5]
Wherein:
x i time sequence data, x, representing ice performance index of ith wind driven generator 1 Indicating the rotating speed index of the fan blade, x 2 Indicating the active power index of the generator, x 3 Indicating the temperature index of the fan blade, x 4 Indicating the humidity index of the fan blade, x 5 Representing the image index of the fan blade;
x i (t n ) Representing t n Time sequence data of ice-tracking index of ith wind driven generator acquired at moment, t 1 Representing initial time t of acquisition of ice-caterpillar index time sequence data of wind driven generator N The method comprises the steps of representing the cut-off time of the acquisition of the ice-caterpillar index time sequence data of the wind driven generator; the time interval between adjacent data acquisition moments is deltat;
t represents the transpose.
In the embodiment of the invention, the temperature and humidity sensor is used for collecting temperature and humidity index data of the fan blades in the wind driven generator, the camera is used for collecting image data of the fan blades in the wind driven generator, the rotating speed sensor is used for collecting stroke She Zhuaisu index data of the wind driven generator, and the active power of the wind driven generator is monitored in real time.
Optionally, the determining the ice performance index weight of the wind driven generator in the step S1 includes:
determining index weights of ice-caterpillar indexes of different wind driven generators, wherein the determining process of the index weights is as follows:
s11: collecting the ice-caterpillar-index time sequence data of the wind driven generator at M moments, wherein the ice-caterpillar-index time sequence data of the wind driven generator at each moment comprises time sequence data of ice-caterpillar-indexes of five wind driven generators, and the wind driven generator is in different ice-caterpillar states in the collected M moments, including an ice-caterpillar-free state and an ice-caterpillar state;
S12: calculating information entropy H of the ice-caterpillar index time sequence data of the wind driven generator at M moments:
wherein:
M 0 representing the number of moments when the wind driven generator is in an ice-free state, M 1 The number of moments when the wind driven generator is in the ice-caterpillar state is represented;
s13: extracting image edges of the binarized fan blade images by using a Canny edge detection algorithm, calculating cosine similarity of any image edges, marking the fan blade images with cosine similarity higher than a similarity threshold as the same fan blade image, and carrying out coding representation, wherein the coding representation results of the same fan blade image are the same;
s14: calculating reference index weight w of ice-caterpillar index of any ith wind driven generator i ：
Wherein:
Ω i data set of unique item of reserved repeated data of ith wind turbine icing index in acquired M-moment wind turbine icing index time sequence data, num k (i) Representing the number of times of occurrence of data k in the time sequence data of the ice indicator of the wind driven generator at M moments for the ice indicator of the ith wind driven generator, and for the image indicator of the fan blade, the data in the data set is the encoding representation result of the image of the fan blade; in the embodiment of the invention, omega i In reserving unique data and duplicates The unique item of the data is used for reserving the coding representation result of the fan blade image in the data set for the fan blade image index;
s15: selecting the ith with highest reference index weight * The seed index calculates the corresponding data and the ith index of other indexes according to the acquired M-moment wind driven generator ice-caterpillar index time sequence data * Correlation of the data corresponding to the index of the species, wherein the data corresponding to the j-th species and the i-th species * The correlation of the seed index corresponding data is sim (i * ,j)，j,i * ∈[1,5]，j≠i * ；
S16: updating the weight of the j index, wherein the updating formula is as follows:
w j ←(1-sim(i * ,j))w j
s17: normalizing the updated weights of all indexes to obtain the weights of all indexes after normalization, wherein the normalized weight of the ith index is
Optionally, in the step S1, weighting the index data based on the ice performance index weight of the wind driven generator includes:
and carrying out index data weighting based on the ice-caterpillar index weight of the wind driven generator, wherein the index data weighting flow is as follows:
carrying out standardization processing on collected data of a fan blade rotating speed index, a generator active power index, a fan blade temperature index and a fan blade humidity index, wherein a standardization processing formula is as follows:
wherein:
represents x r (t n ) Normalized processing results of r.epsilon.1, 4]；
And carrying out weighting on the standardized data, wherein a weighting formula is as follows:
Wherein:
x ′ r (t n ) Representation ofIs a weighted result of the (a);
and carrying out graying treatment on the fan blade image to obtain a graying fan blade image, wherein t is as follows n Gray fan blade image at momentAnd is about->And (3) weighting and binarizing:
wherein:
c (·) represents binarization;
constituting the weighted wind driven generator ice indicator time sequence data x ′ ：
x ′ ＝[x ′ 1 ,x ′ 2 ,x ′ 3 ,x ′ 4 ,x ′ 5 ] T
x i ′ ＝(x i ′ (t 1 ),x i ′ (t 2 ),...,x i ′ (t n ),...,x i ′ (t N )),i∈[1,5]
Wherein:
x i ′ and the time sequence data after weighting of the ice-walking index of the ith wind driven generator is shown.
Optionally, in the step S2, encoding the weighted wind turbine ice performance index time sequence data includes:
encoding the weighted wind driven generator ice performance index time sequence data, wherein the encoding flow is as follows:
s21: inputting weighted wind driven generator ice performance index time sequence data into a coding module, and coding the weighted wind driven generator ice performance index time sequence data by the coding module, wherein a coding processing formula of time sequence data acquired by an ith index is as follows:
wherein:
representing the coding matrix for the i-th index;
c i representing the coding processing result of the time sequence data acquired by the ith index;
s22: cross coding is carried out on coding processing results of various index data, and coded index vector data are obtained:
X＝[Max(1)*c 1 ,Max(2)*c 2 ,Max(3)*c 3 ,Max(4)*c 4 ,Max(5)*c 5 ] T
Max(1)＝[max(c 2 ),max(c 3 ),max(c 4 ),max(c 5 )]
Max(2)＝[max(c 1 ),max(c 3 ),max(c 4 ),max(c 5 )]
Max(3)＝[max(c 1 ),max(c 2 ),max(c 4 ),max(c 5 )]
Max(4)＝[max(c 1 ),max(c 2 ),max(c 3 ),max(c 5 )]
Max(5)＝[max(c 1 ),max(c 2 ),max(c 3 ),max(c 4 )]
Wherein:
* Representing a convolution operation;
max (·) represents the selected maximum value;
x represents the encoded index vector data.
Optionally, constructing a wind driven generator ice-caterpillar timing analysis model in the step S3 includes:
building a wind driven generator ice-building time sequence analysis model, wherein the model takes index vector data as input and ice-building probability as output, and the wind driven generator ice-building time sequence analysis model comprises a prior probability layer, an input layer, a probability calculation layer and an output layer;
the prior probability layer is used for obtaining the prior probability of the ice-caterpillar prediction result, the input layer is used for receiving index vector data, the probability calculation layer is used for calculating the ice-caterpillar probability of the wind driven generator, the ice-caterpillar probability is used as output, and the prior probability is updated according to the output result.
Optionally, in the step S4, optimizing the constructed wind turbine ice performance timing analysis model to obtain an optimal wind turbine ice performance timing analysis model, including:
optimizing the constructed wind driven generator ice-tracking time sequence analysis model to obtain an optimal wind driven generator ice-tracking time sequence analysis model, wherein the model optimization flow is as follows:
s41: initializing probability distribution parameters theta (0) of a wind driven generator ice-caterpillar time sequence analysis model:
θ(0)＝(θ 1 (0),θ 2 (0),θ 3 (0),θ 4 (0),θ 5 (0))
Wherein:
θ i (0) The probability distribution parameters of the ith wind driven generator ice-tracking index generated by initialization are represented;
s42: q groups of independent index vector data are acquired, and the ice-carrying probability of each group of index vector data is generated in an initialized mode to form a training set data of an ice-carrying time sequence analysis model of the wind driven generator:
data＝{(X q ,p q )|q∈[1,Q]}
wherein:
X q represents the q-th set of index vector data, p q Index vector data X representing initialization generation q Ice-carrying probability of (2); the ice-caterpillar probability corresponding to the Q group index vector data obeys the prior probability distribution p; in the embodiment of the invention, the prior probability distribution p represents probability distribution of different ice-caterpillar probabilities of the wind driven generator in a real environment;
s43: setting the optimization times of the current model as d, setting the initial value of d as 0, and setting the probability distribution parameter obtained by the d-th optimization as theta (d);
s44: calculating to obtain a log likelihood function of training data in the training set:
wherein:
z represents the ice-following probability obeying the prior probability distribution p;
P(z|X q θ (d)) means X q Inputting the probability distribution parameters theta (d) into a wind driven generator ice-caterpillar timing analysis model, and outputting probability that the ice-caterpillar probability accords with the prior probability distribution p by the model;
P(X q z|theta (d)) represents a wind driven generator ice-tracking time sequence analysis model based on the probability distribution parameter theta (d), and the input and the output are respectively X q Probability of z, where z conforms to the prior probability distribution p;
s45: updating to obtain a probability distribution parameter theta (d+1) obtained by the d+1th optimization:
wherein:
θ (d+1) represents a probability distribution parameter such that L (p, θ (d)) reaches the maximum;
s46: if theta (d+1) meets the following formula, stopping training optimization, and constructing an optimal wind turbine ice-carrying time sequence analysis model based on theta (d+1):
wherein:
i & ltI & gt & lt/I & gt represents the L1 norm & ltI & gt & lt/I & gt 2 Represents an L2 norm;
epsilon represents a preset parameter variation threshold.
Optionally, in the step S4, obtaining the ice-formation probability of the wind turbine using the optimal wind turbine ice-formation timing analysis model includes:
obtaining ice-caterpillar probability of the wind driven generator by using an optimal wind driven generator ice-caterpillar time sequence analysis model, wherein the calculation flow of the ice-caterpillar probability is as follows:
the prior probability of the current ice-caterpillar prediction result is obtained:
wherein:
f represents a preset reference number of times, F 0 Representing the number of times that a preset wind driven generator is in a non-ice state, F 1 Representing the times of the preset wind driven generator in the ice-caterpillar state; in the embodiment of the invention, F is set to 100, F 0 F (F) 1 Set to 60 and 40, respectively;
alpha represents the total number of times of calculation of the current ice-carrying probability of the optimal wind driven generator ice-carrying time sequence analysis model, and alpha 0 Representing the number of times the ice-on probability is below a probability threshold, alpha 1 The times of the ice-carrying probability being more than or equal to a probability threshold value are represented;
P 0 representing probability of wind driven generator in non-ice-walking state, P 1 Indicating that the wind-driven generator is on trackProbability of ice state;
receiving index vector data x= [ Max (1) ×c 1 ,Max(2)*c 2 ,Max(3)*c 3 ,Max(4)*c 4 ,Max(5)*
c5]T；
Calculating to obtain ice-caterpillar probability P (X; theta):
wherein:
exp (·) represents an exponent based on a natural constant;
θ＝(θ 1 ,θ 2 ,θ 3 ,θ 4 ,θ 5 ) The probability distribution parameters of the optimal wind driven generator ice-caterpillar time sequence analysis model are represented;
updating the prior probability according to the ice-carrying probability P (X; theta);
if the ice-carrying probability P (X; theta) is greater than or equal to the probability threshold, the wind driven generator is in the ice-carrying state, and if the ice-carrying probability P (X; theta) is greater than or equal to the probability threshold, the wind driven generator is in the ice-carrying state.
In order to solve the above-described problems, the present invention provides an electronic apparatus including:
a memory storing at least one instruction;
the communication interface is used for realizing the communication of the electronic equipment; a kind of electronic device with high-pressure air-conditioning system
And the processor executes the instructions stored in the memory to realize the wind driven generator ice prediction method for time sequence analysis.
In order to solve the above problems, the present invention further provides a computer readable storage medium having at least one instruction stored therein, the at least one instruction being executed by a processor in an electronic device to implement the wind turbine ice prediction method for time series analysis described above.
Compared with the prior art, the invention provides a method for predicting ice of a wind driven generator through time sequence analysis, and the method has the following advantages:
firstly, the scheme provides a multi-index weight determining method, by collecting the time sequence data of the ice-caterpillar indexes of the wind driven generator at M moments, wherein the time sequence data of the ice-caterpillar indexes of the wind driven generator at each moment comprises the time sequence data of the ice-caterpillar indexes of the five wind driven generators, and the wind driven generator is in different ice-caterpillar states in the collected M moments, including a non-ice-caterpillar state and an ice-caterpillar state; calculating information entropy H of the ice-caterpillar index time sequence data of the wind driven generator at M moments:
wherein: m is M 0 Representing the number of moments when the wind driven generator is in an ice-free state, M 1 The number of moments when the wind driven generator is in the ice-caterpillar state is represented; extracting image edges of the binarized fan blade images by using a Canny edge detection algorithm, calculating cosine similarity of any image edges, marking the fan blade images with cosine similarity higher than a similarity threshold as the same fan blade image, and carrying out coding representation, wherein the coding representation results of the same fan blade image are the same; calculating reference index weight w of ice-caterpillar index of any ith wind driven generator i ：
Wherein: omega shape i Data set of unique item of reserved repeated data of ith wind turbine icing index in acquired M-moment wind turbine icing index time sequence data, num k (i) Representing the number of times of occurrence of data k in the time sequence data of the ice indicator of the wind driven generator at M moments for the ice indicator of the ith wind driven generator, and for the image indicator of the fan blade, the data in the data set is the encoding representation result of the image of the fan blade; selecting the ith with highest reference index weight * The seed index calculates the corresponding data and the ith index of other indexes according to the acquired M-moment wind driven generator ice-caterpillar index time sequence data * Correlation of seed index corresponding dataSex, wherein the j index corresponds to data and i * The correlation of the seed index corresponding data is sim (i * ,j)，j,i * ∈[1,5]，j≠i * The method comprises the steps of carrying out a first treatment on the surface of the Updating the weight of the j index, wherein the updating formula is as follows:
w j ←(1-sim(i * ,j))w j
normalizing the updated weights of all indexes to obtain the weights of all indexes after normalization, wherein the normalized weight of the ith index isAccording to the scheme, the reference index weights of the indexes are determined according to the information entropy of different indexes, wherein the larger the reference index weights of the indexes are, the higher the accuracy of dividing the non-ice state and the ice state of the wind driven generator by the indexes is, the indexes with the maximum reference index weights are further fixed, the correlation between other indexes and the fixed index data is calculated, the reference index weights of the other indexes are calibrated, the larger the correlation is, the closer the information entropy of the indexes and the fixed index is, the smaller the calibrated reference index weights are, the acquired wind driven generator ice index time sequence data are weighted, the index time sequence data representing the importance of different indexes are obtained, the larger the weight of the index of the ice state can be effectively detected, and the accuracy of the subsequent ice prediction is improved.
Meanwhile, the scheme provides an ice-caterpillar probability calculation method combining time sequence data, which utilizes an optimal wind power generator ice-caterpillar time sequence analysis model to obtain the ice-caterpillar probability of the wind power generator, wherein the calculation flow of the ice-caterpillar probability is as follows: the prior probability of the current ice-caterpillar prediction result is obtained:
wherein: f represents a preset reference number of times, F 0 Representing the number of times that a preset wind driven generator is in a non-ice state, F 1 Representing the times of the preset wind driven generator in the ice-caterpillar state; alpha represents the total number of times of calculation of the current ice-carrying probability of the optimal wind driven generator ice-carrying time sequence analysis model, and alpha 0 Representing the number of times the ice-on probability is below a probability threshold, alpha 1 The times of the ice-carrying probability being more than or equal to a probability threshold value are represented; p (P) 0 Representing probability of wind driven generator in non-ice-walking state, P 1 Representing the probability that the wind driven generator is in the ice-covered state; receiving index vector data x= [ Max (1) ×c 1 ,Max(2)*c 2 ,Max(3)*c 3 ,Max(4)*c 4 ,Max(5)*c5]T is a T; calculating to obtain ice-caterpillar probability P (X; theta):
wherein: exp (·) represents an exponent based on a natural constant; θ= (θ) 1 ,θ 2 ,θ 3 ,θ 4 ,θ 5 ) The probability distribution parameters of the optimal wind driven generator ice-caterpillar time sequence analysis model are represented; the prior probability is updated according to the ice-caterpillar probability P (X; theta). According to the scheme, the time sequence data of each index are subjected to coding processing respectively, a convolution matrix is built by combining the most obvious coding processing results of other indexes, convolution coding operation is carried out, cross coding among different indexes is achieved, a wind turbine ice-carrying time sequence analysis model is built by combining the prior probability distribution of ice-carrying probability, the probability distribution parameters and likelihood functions of sample data are subjected to iterative solution and convergence to obtain probability distribution parameters, and ice-carrying probability calculation is carried out by combining the current index data cross coding results based on continuously updated prior probability, so that wind turbine ice-carrying prediction combining the time sequence data is achieved.
Drawings
FIG. 1 is a flow chart of a method for predicting ice on a wind turbine according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of an electronic device for implementing a method for predicting ice performance of a wind turbine according to an embodiment of the present application.
The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The embodiment of the application provides a method for predicting ice of a wind driven generator through time sequence analysis. The execution main body of the wind driven generator ice prediction method for time sequence analysis comprises at least one of electronic equipment, such as a server side, a terminal and the like, which can be configured to execute the method provided by the embodiment of the application. In other words, the wind turbine ice prediction method of the time sequence analysis can be executed by software or hardware installed in a terminal device or a server device, wherein the software can be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like.
Example 1:
s1: and acquiring the ice performance index time sequence data of the wind driven generator, and weighting the index data based on the ice performance index weight of the wind driven generator to obtain the weighted ice performance index time sequence data of the wind driven generator.
The step S1 is to collect the ice performance index time sequence data of the wind driven generator, and comprises the following steps:
collecting the ice-caterpillar index time sequence data of the wind driven generator, wherein the ice-caterpillar index of the wind driven generator comprises the blade rotating speed, the active power of the wind driven generator, the blade temperature, the blade humidity and the blade image of the wind driven generator, and the collected ice-caterpillar index time sequence data x of the wind driven generator is as follows:
x＝[x 1 ,x 2 ,x 3 ,x 4 ,x 5 ] T
x i ＝(x i (t 1 ),x i (t 2 ),...,x i (t n ),...,x i (t N )),i∈[1,5]
wherein:
x i time sequence data, x, representing ice performance index of ith wind driven generator 1 Indicating the rotating speed index of the fan blade, x 2 Indicating the active power index of the generator, x 3 Indicating the temperature index of the fan blade, x 4 Indicating the humidity index of the fan blade, x 5 Representing the image index of the fan blade;
x i (t n ) Representing t n Time sequence data of ice-tracking index of ith wind driven generator acquired at moment, t 1 Representing initial time t of acquisition of ice-caterpillar index time sequence data of wind driven generator N The method comprises the steps of representing the cut-off time of the acquisition of the ice-caterpillar index time sequence data of the wind driven generator; the time interval between adjacent data acquisition moments is deltat;
t represents the transpose.
In the step S1, determining the ice performance index weight of the wind driven generator comprises the following steps:
determining index weights of ice-caterpillar indexes of different wind driven generators, wherein the determining process of the index weights is as follows:
s11: collecting the ice-caterpillar-index time sequence data of the wind driven generator at M moments, wherein the ice-caterpillar-index time sequence data of the wind driven generator at each moment comprises time sequence data of ice-caterpillar-indexes of five wind driven generators, and the wind driven generator is in different ice-caterpillar states in the collected M moments, including an ice-caterpillar-free state and an ice-caterpillar state;
s12: calculating information entropy H of the ice-caterpillar index time sequence data of the wind driven generator at M moments:
wherein:
M 0 representing the number of moments when the wind driven generator is in an ice-free state, M 1 The number of moments when the wind driven generator is in the ice-caterpillar state is represented;
s13: extracting image edges of the binarized fan blade images by using a Canny edge detection algorithm, calculating cosine similarity of any image edges, marking the fan blade images with cosine similarity higher than a similarity threshold as the same fan blade image, and carrying out coding representation, wherein the coding representation results of the same fan blade image are the same;
s14: calculating reference index weight w of ice-caterpillar index of any ith wind driven generator i ：
Wherein:
Ω i data set of unique item of reserved repeated data of ith wind turbine icing index in acquired M-moment wind turbine icing index time sequence data, num k (i) Representing the number of times of occurrence of data k in the time sequence data of the ice indicator of the wind driven generator at M moments for the ice indicator of the ith wind driven generator, and for the image indicator of the fan blade, the data in the data set is the encoding representation result of the image of the fan blade; in the embodiment of the invention, omega i The unique data and the unique item of repeated data are reserved, and for the blade image index, the encoding representation result of the blade image is reserved in the data set;
s15: selecting the ith with highest reference index weight * The seed index calculates the corresponding data and the ith index of other indexes according to the acquired M-moment wind driven generator ice-caterpillar index time sequence data * Correlation of the data corresponding to the index of the species, wherein the data corresponding to the j-th species and the i-th species * The correlation of the seed index corresponding data is sim (i * ,j)，j,i * ∈[1,5]，j≠i * ；
S16: updating the weight of the j index, wherein the updating formula is as follows:
w j ←(1-sim(i * ,j))w j
s17: normalizing the updated weights of all indexes to obtain the weights of all indexes after normalization, wherein the normalized weight of the ith index is
In the step S1, weighting index data based on the ice performance index weight of the wind driven generator comprises the following steps:
and carrying out index data weighting based on the ice-caterpillar index weight of the wind driven generator, wherein the index data weighting flow is as follows:
carrying out standardization processing on collected data of a fan blade rotating speed index, a generator active power index, a fan blade temperature index and a fan blade humidity index, wherein a standardization processing formula is as follows:
wherein:
represents x r (t n ) Normalized processing results of r.epsilon.1, 4]；
And carrying out weighting on the standardized data, wherein a weighting formula is as follows:
wherein:
x ′ r (t n ) Representation ofIs a weighted result of the (a);
and carrying out graying treatment on the fan blade image to obtain a graying fan blade image, wherein t is as follows n Gray fan blade image at momentAnd is about->And (3) weighting and binarizing:
wherein:
c (·) represents binarization;
constituting the weighted wind driven generator ice indicator time sequence data x ′ ：
x ′ ＝[x ′ 1 ,x ′ 2 ,x ′ 3 ,x ′ 4 ,x ′ 5 ] T
x i ′ ＝(x i ′ (t 1 ),x i ′ (t 2 ),...,x i ′ (t n ),...,x i ′ (t N )),i∈[1,5]
Wherein:
x i ′ and the time sequence data after weighting of the ice-walking index of the ith wind driven generator is shown.
S2: and encoding the weighted wind driven generator ice performance index time sequence data to obtain encoded index vector data.
In the step S2, encoding the weighted wind driven generator ice performance index time sequence data comprises the following steps:
Encoding the weighted wind driven generator ice performance index time sequence data, wherein the encoding flow is as follows:
s21: inputting weighted wind driven generator ice performance index time sequence data into a coding module, and coding the weighted wind driven generator ice performance index time sequence data by the coding module, wherein a coding processing formula of time sequence data acquired by an ith index is as follows:
wherein:
representing the coding matrix for the i-th index;
c i representing the coding processing result of the time sequence data acquired by the ith index;
s22: cross coding is carried out on coding processing results of various index data, and coded index vector data are obtained:
X＝[Max(1)*c 1 ,Max(2)*c 2 ,Max(3)*c 3 ,Max(4)*c 4 ,Max(5)*c 5 ] T
Max(1)＝[max(c 2 ),max(c 3 ),max(c 4 ),max(c 5 )]
Max(2)＝[max(c 1 ),max(c 3 ),max(c 4 ),max(c 5 )]
Max(3)＝[max(c 1 ),max(c 2 ),max(c 4 ),max(c 5 )]
Max(4)＝[max(c 1 ),max(c 2 ),max(c 3 ),max(c 5 )]
Max(5)＝[max(c 1 ),max(c 2 ),max(c 3 ),max(c 4 )]
wherein:
* Representing a convolution operation;
max (·) represents the selected maximum value;
x represents the encoded index vector data.
S3: and constructing an ice-caterpillar timing analysis model of the wind driven generator, wherein the model takes index vector data as input and ice-caterpillar probability as output.
In the step S3, a wind driven generator ice-tracking time sequence analysis model is constructed, and the method comprises the following steps:
building a wind driven generator ice-building time sequence analysis model, wherein the model takes index vector data as input and ice-building probability as output, and the wind driven generator ice-building time sequence analysis model comprises a prior probability layer, an input layer, a probability calculation layer and an output layer;
The prior probability layer is used for obtaining the prior probability of the ice-caterpillar prediction result, the input layer is used for receiving index vector data, the probability calculation layer is used for calculating the ice-caterpillar probability of the wind driven generator, the ice-caterpillar probability is used as output, and the prior probability is updated according to the output result.
S4: optimizing the constructed wind power generator ice-tracking time sequence analysis model to obtain an optimal wind power generator ice-tracking time sequence analysis model, and obtaining the ice-tracking probability of the wind power generator by utilizing the optimal wind power generator ice-tracking time sequence analysis model.
In the step S4, the constructed wind turbine ice-track time sequence analysis model is optimized to obtain an optimal wind turbine ice-track time sequence analysis model, and the method comprises the following steps:
optimizing the constructed wind driven generator ice-tracking time sequence analysis model to obtain an optimal wind driven generator ice-tracking time sequence analysis model, wherein the model optimization flow is as follows:
s41: initializing probability distribution parameters theta (0) of a wind driven generator ice-caterpillar time sequence analysis model:
θ(0)＝(θ 1 (0),θ 2 (0),θ 3 (0),θ 4 (0),θ 5 (0))
wherein:
θ i (0) The probability distribution parameters of the ith wind driven generator ice-tracking index generated by initialization are represented;
s42: q groups of independent index vector data are acquired, and the ice-carrying probability of each group of index vector data is generated in an initialized mode to form a training set data of an ice-carrying time sequence analysis model of the wind driven generator:
data＝{(X q ,p q )|q∈[1,Q]}
Wherein:
X q represents the q-th set of index vector data, p q Index vector data X representing initialization generation q Ice-carrying probability of (2); the ice-caterpillar probability corresponding to the Q group index vector data obeys the prior probability distribution p; in the embodiment of the invention, the prior probability distribution p represents probability distribution of different ice-caterpillar probabilities of the wind driven generator in a real environment;
s43: setting the optimization times of the current model as d, setting the initial value of d as 0, and setting the probability distribution parameter obtained by the d-th optimization as theta (d);
s44: calculating to obtain a log likelihood function of training data in the training set:
wherein:
z represents the ice-following probability obeying the prior probability distribution p;
P(z|X q θ (d)) means X q Inputting the probability distribution parameters theta (d) into a wind driven generator ice-caterpillar timing analysis model, and outputting probability that the ice-caterpillar probability accords with the prior probability distribution p by the model;
P(X q z|theta (d)) represents a wind driven generator ice-tracking time sequence analysis model based on the probability distribution parameter theta (d), and the input and the output are respectively X q Probability of z, where z conforms to the prior probability distribution p;
s45: updating to obtain a probability distribution parameter theta (d+1) obtained by the d+1th optimization:
wherein:
θ (d+1) represents a probability distribution parameter such that L (p, θ (d)) reaches the maximum;
S46: if theta (d+1) meets the following formula, stopping training optimization, and constructing an optimal wind turbine ice-carrying time sequence analysis model based on theta (d+1):
wherein:
i & ltI & gt & lt/I & gt represents the L1 norm & ltI & gt & lt/I & gt 2 Represents an L2 norm;
epsilon represents a preset parameter variation threshold.
In the step S4, the ice-caterpillar probability of the wind driven generator is obtained by using an optimal wind driven generator ice-caterpillar time sequence analysis model, and the method comprises the following steps:
obtaining ice-caterpillar probability of the wind driven generator by using an optimal wind driven generator ice-caterpillar time sequence analysis model, wherein the calculation flow of the ice-caterpillar probability is as follows:
the prior probability of the current ice-caterpillar prediction result is obtained:
wherein:
f represents a preset reference number of times, F 0 Representing the number of times that a preset wind driven generator is in a non-ice state, F 1 Representing the times of the preset wind driven generator in the ice-caterpillar state; in the embodiment of the invention, F is set to 100, F 0 F (F) 1 Set to 60 and 40, respectively;
alpha represents the total number of times of calculation of the current ice-carrying probability of the optimal wind driven generator ice-carrying time sequence analysis model, and alpha 0 Representing the number of times the ice-on probability is below a probability threshold, alpha 1 The times of the ice-carrying probability being more than or equal to a probability threshold value are represented;
P 0 representing probability of wind driven generator in non-ice-walking state, P 1 Representing the probability that the wind driven generator is in the ice-covered state;
Receiving index vector data x= [ Max (1) ×c 1 ,Max(2)*c 2 ,Max(3)*c 3 ,Max(4)*c 4 ,Max(5)*
c5]T；
Calculating to obtain ice-caterpillar probability P (X; theta):
wherein:
exp (·) represents an exponent based on a natural constant;
θ＝(θ 1 ,θ 2 ,θ 3 ,θ 4 ,θ 5 ) The probability distribution parameters of the optimal wind driven generator ice-caterpillar time sequence analysis model are represented;
updating the prior probability according to the ice-carrying probability P (X; theta);
if the ice-carrying probability P (X; theta) is greater than or equal to the probability threshold, the wind driven generator is in the ice-carrying state, and if the ice-carrying probability P (X; theta) is greater than or equal to the probability threshold, the wind driven generator is in the ice-carrying state.
Example 2:
fig. 2 is a schematic structural diagram of an electronic device for implementing a method for predicting ice on a wind turbine according to an embodiment of the present invention.
The electronic device 1 may comprise a processor 10, a memory 11, a communication interface 13 and a bus, and may further comprise a computer program, such as program 12, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, such as a removable hard disk of the electronic device 1. The memory 11 may in other embodiments also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device 1. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 11 may be used not only for storing application software installed in the electronic device 1 and various types of data, such as codes of the program 12, but also for temporarily storing data that has been output or is to be output.
The processor 10 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects the respective components of the entire electronic device using various interfaces and lines, executes or executes programs or modules (a program 12 for performing wind turbine ice prediction for time series analysis, etc.) stored in the memory 11, and invokes data stored in the memory 11 to perform various functions of the electronic device 1 and process the data.
The communication interface 13 may comprise a wired interface and/or a wireless interface (e.g. WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device 1 and other electronic devices and to enable connection communication between internal components of the electronic device.
The bus may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
Fig. 2 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 2 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device 1 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.
The electronic device 1 may optionally further comprise a user interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device 1 and for displaying a visual user interface.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The program 12 stored in the memory 11 of the electronic device 1 is a combination of instructions that, when executed in the processor 10, may implement:
acquiring the ice performance index time sequence data of the wind driven generator, and weighting the index data based on the ice performance index weight of the wind driven generator to obtain the weighted ice performance index time sequence data of the wind driven generator;
encoding the weighted wind driven generator ice performance index time sequence data to obtain encoded index vector data;
constructing an ice-caterpillar timing analysis model of the wind driven generator;
optimizing the constructed wind power generator ice-tracking time sequence analysis model to obtain an optimal wind power generator ice-tracking time sequence analysis model, and obtaining the ice-tracking probability of the wind power generator by utilizing the optimal wind power generator ice-tracking time sequence analysis model.
Specifically, the specific implementation method of the above instruction by the processor 10 may refer to descriptions of related steps in the corresponding embodiments of fig. 1 to 2, which are not repeated herein.
It should be noted that, the foregoing reference numerals of the embodiments of the present invention are merely for describing the embodiments, and do not represent the advantages and disadvantages of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article or method that comprises the element.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.

Claims (8)

1. A method for predicting ice performance of a wind driven generator through time sequence analysis is characterized by comprising the following steps:
S1: acquiring the ice performance index time sequence data of the wind driven generator, and weighting the index data based on the ice performance index weight of the wind driven generator to obtain the weighted ice performance index time sequence data of the wind driven generator;
s2: encoding the weighted wind driven generator ice performance index time sequence data to obtain encoded index vector data;
s3: constructing an ice-caterpillar-sequence analysis model of the wind driven generator, wherein the model takes index vector data as input and ice-caterpillar probability as output;
s4: optimizing the constructed wind power generator ice-tracking time sequence analysis model to obtain an optimal wind power generator ice-tracking time sequence analysis model, and obtaining the ice-tracking probability of the wind power generator by utilizing the optimal wind power generator ice-tracking time sequence analysis model.
2. The method for predicting ice performance of a wind turbine according to claim 1, wherein the step S1 of collecting the time series data of the ice performance index of the wind turbine comprises:
collecting the ice-caterpillar index time sequence data of the wind driven generator, wherein the ice-caterpillar index of the wind driven generator comprises the blade rotating speed, the active power of the wind driven generator, the blade temperature, the blade humidity and the blade image of the wind driven generator, and the collected ice-caterpillar index time sequence data x of the wind driven generator is as follows:
x＝[x 1 ,x 2 ,x 3 ,x 4 ,x 5 ] T
x i ＝(x i (t 1 ),x i (t 2 ),...,x i (t n ),...,x i (t N )),i∈[1,5]
Wherein:
x i time sequence data, x, representing ice performance index of ith wind driven generator 1 Indicating the rotating speed index of the fan blade, x 2 Indicating the active power index of the generator, x 3 Indicating the temperature index of the fan blade, x 4 Indicating the humidity index of the fan blade, x 5 Representing the image index of the fan blade;
x i (t n ) Representing t n Time sequence data of ice-tracking index of ith wind driven generator acquired at moment, t 1 The initial time of the acquisition of the ice-caterpillar index time sequence data of the wind driven generator is represented,t N the method comprises the steps of representing the cut-off time of the acquisition of the ice-caterpillar index time sequence data of the wind driven generator; the time interval between adjacent data acquisition moments is deltat;
t represents the transpose.
3. The method for predicting ice performance of a wind turbine according to claim 2, wherein determining the ice performance index weight of the wind turbine in step S1 comprises:
determining index weights of ice-caterpillar indexes of different wind driven generators, wherein the determining process of the index weights is as follows:
s11: collecting the ice-caterpillar-index time sequence data of the wind driven generator at M moments, wherein the ice-caterpillar-index time sequence data of the wind driven generator at each moment comprises time sequence data of ice-caterpillar-indexes of five wind driven generators, and the wind driven generator is in different ice-caterpillar states in the collected M moments, including an ice-caterpillar-free state and an ice-caterpillar state;
S12: calculating information entropy H of the ice-caterpillar index time sequence data of the wind driven generator at M moments:
wherein:
M 0 representing the number of moments when the wind driven generator is in an ice-free state, M 1 The number of moments when the wind driven generator is in the ice-caterpillar state is represented;
s13: extracting image edges of the binarized fan blade images by using a Canny edge detection algorithm, calculating cosine similarity of any image edges, marking the fan blade images with cosine similarity higher than a similarity threshold as the same fan blade image, and carrying out coding representation, wherein the coding representation results of the same fan blade image are the same;
s14: calculating reference index weight w of ice-caterpillar index of any ith wind driven generator i ：
Wherein:
Ω i data set of unique item of reserved repeated data of ith wind turbine icing index in acquired M-moment wind turbine icing index time sequence data, num k (i) Representing the number of times of occurrence of data k in the time sequence data of the ice indicator of the wind driven generator at M moments for the ice indicator of the ith wind driven generator, and for the image indicator of the fan blade, the data in the data set is the encoding representation result of the image of the fan blade;
s15: selecting the ith with highest reference index weight * The seed index calculates the corresponding data and the ith index of other indexes according to the acquired M-moment wind driven generator ice-caterpillar index time sequence data * Correlation of the data corresponding to the index of the species, wherein the data corresponding to the j-th species and the i-th species * The correlation of the seed index corresponding data is sim (i * ,j)，j,i * ∈[1,5]，j≠i * ；
S16: updating the weight of the j index, wherein the updating formula is as follows:
w j ←(1-sim(i * ,j))w j
s17: normalizing the updated weights of all indexes to obtain the weights of all indexes after normalization, wherein the normalized weight of the ith index is
4. The method for predicting ice performance of a wind turbine according to claim 3, wherein in step S1, the weighting of the index data is performed based on the ice performance index weight of the wind turbine, comprising:
and carrying out index data weighting based on the ice-caterpillar index weight of the wind driven generator, wherein the index data weighting flow is as follows:
carrying out standardization processing on collected data of a fan blade rotating speed index, a generator active power index, a fan blade temperature index and a fan blade humidity index, wherein a standardization processing formula is as follows:
wherein:
represents x r (t n ) Normalized processing results of r.epsilon.1, 4]；
And carrying out weighting on the standardized data, wherein a weighting formula is as follows:
wherein:
x′ r (t n ) Representation ofIs a weighted result of the (a);
and carrying out graying treatment on the fan blade image to obtain a graying fan blade image, wherein t is as follows n Gray fan blade image at momentAnd is about->And (3) weighting and binarizing:
wherein:
c (·) represents binarization;
constituting the weighted wind driven generator ice indicator time sequence data x ′ ：
x′＝[x′ 1 ,x′ 2 ,x′ 3 ,x′ 4 ,x′ 5 ] T
x′ i ＝(x′ i (t 1 ),x′ i (t 2 ),...,x′ i (t n ),...,x′ i (t N )),i∈[1,5]
Wherein:
x′ i and the time sequence data after weighting of the ice-walking index of the ith wind driven generator is shown.
5. The method for predicting ice performance of a wind turbine according to claim 4, wherein the step S2 encodes weighted wind turbine ice performance index time series data, including:
encoding the weighted wind driven generator ice performance index time sequence data, wherein the encoding flow is as follows:
s21: inputting weighted wind driven generator ice performance index time sequence data into a coding module, and coding the weighted wind driven generator ice performance index time sequence data by the coding module, wherein a coding processing formula of time sequence data acquired by an ith index is as follows:
wherein:
representing the coding matrix for the i-th index;
c i representing the coding processing result of the time sequence data acquired by the ith index;
s22: cross coding is carried out on coding processing results of various index data, and coded index vector data are obtained:
X＝[Max(1)*c 1 ,Max(2)*c 2 ,Max(3)*c 3 ,Max(4)*c 4 ,Max(5)*c 5 ] T
Max(1)＝[max(c 2 ),max(c 3 ),max(c 4 ),max(c 5 )]
Max(2)＝[max(c 1 ),max(c 3 ),max(c 4 ),max(c 5 )]
Max(3)＝[max(c 1 ),max(c 2 ),max(c 4 ),max(c 5 )]
Max(4)＝[max(c 1 ),max(c 2 ),max(c 3 ),max(c 5 )]
Max(5)＝[max(c 1 ),max(c 2 ),max(c 3 ),max(c 4 )]
wherein:
* Representing a convolution operation;
max (·) represents the selected maximum value;
x represents the encoded index vector data.
6. The method for predicting ice performance of a wind turbine according to claim 1, wherein the step S3 of constructing a model for analyzing ice performance of a wind turbine includes:
building a wind driven generator ice-building time sequence analysis model, wherein the model takes index vector data as input and ice-building probability as output, and the wind driven generator ice-building time sequence analysis model comprises a prior probability layer, an input layer, a probability calculation layer and an output layer;
the prior probability layer is used for obtaining the prior probability of the ice-caterpillar prediction result, the input layer is used for receiving index vector data, the probability calculation layer is used for calculating the ice-caterpillar probability of the wind driven generator, the ice-caterpillar probability is used as output, and the prior probability is updated according to the output result.
7. The method for predicting ice performance of a wind turbine according to claim 6, wherein in step S4, the constructed wind turbine ice performance timing analysis model is optimized to obtain an optimal wind turbine ice performance timing analysis model, comprising:
optimizing the constructed wind driven generator ice-tracking time sequence analysis model to obtain an optimal wind driven generator ice-tracking time sequence analysis model, wherein the model optimization flow is as follows:
S41: initializing probability distribution parameters theta (0) of a wind driven generator ice-caterpillar time sequence analysis model:
θ(0)＝(θ 1 (0),θ 2 (0),θ 3 (0),θ 4 (0),θ 5 (0))
wherein:
θ i (0) The probability distribution parameters of the ith wind driven generator ice-tracking index generated by initialization are represented;
s42: q groups of independent index vector data are acquired, and the ice-carrying probability of each group of index vector data is generated in an initialized mode to form a training set data of an ice-carrying time sequence analysis model of the wind driven generator:
data＝{(X q ,p q )|q∈[1,Q]}
wherein:
X q represents the q-th set of index vector data, p q Index vector data X representing initialization generation q Ice-carrying probability of (2); the ice-caterpillar probability corresponding to the Q group index vector data obeys the prior probability distribution p;
s43: setting the optimization times of the current model as d, setting the initial value of d as 0, and setting the probability distribution parameter obtained by the d-th optimization as theta (d);
s44: calculating to obtain a log likelihood function of training data in the training set:
wherein:
z represents the ice-following probability obeying the prior probability distribution p;
P(z|X q θ (d)) means X q When the wind driven generator is input into ice-caterpillar based on probability distribution parameter theta (d)In the sequence analysis model, the probability that the model output ice probability accords with the prior probability distribution p;
P(X q z|theta (d)) represents a wind driven generator ice-tracking time sequence analysis model based on the probability distribution parameter theta (d), and the input and the output are respectively X q Probability of z, where z conforms to the prior probability distribution p;
s45: updating to obtain a probability distribution parameter theta (d+1) obtained by the d+1th optimization:
wherein:
θ (d+1) represents a probability distribution parameter such that L (p, θ (d)) reaches the maximum;
s46: if theta (d+1) meets the following formula, stopping training optimization, and constructing an optimal wind turbine ice-carrying time sequence analysis model based on theta (d+1):
wherein:
i & ltI & gt & lt/I & gt represents the L1 norm & ltI & gt & lt/I & gt 2 Represents an L2 norm;
epsilon represents a preset parameter variation threshold.
8. The method for predicting ice performance of a wind turbine according to claim 7, wherein the step S4 of obtaining the ice performance probability of the wind turbine using the optimal wind turbine ice performance timing analysis model comprises:
obtaining ice-caterpillar probability of the wind driven generator by using an optimal wind driven generator ice-caterpillar time sequence analysis model, wherein the calculation flow of the ice-caterpillar probability is as follows:
the prior probability of the current ice-caterpillar prediction result is obtained:
wherein:
f represents a preset reference number of times, F 0 Representing the number of times that a preset wind driven generator is in a non-ice state, F 1 Representing the times of the preset wind driven generator in the ice-caterpillar state;
alpha represents the total number of times of calculation of the current ice-carrying probability of the optimal wind driven generator ice-carrying time sequence analysis model, and alpha 0 Representing the number of times the ice-on probability is below a probability threshold, alpha 1 The times of the ice-carrying probability being more than or equal to a probability threshold value are represented;
P 0 representing probability of wind driven generator in non-ice-walking state, P 1 Representing the probability that the wind driven generator is in the ice-covered state;
receiving index vector data x= [ Max (1) ×c 1 ,Max(2)*c 2 ,Max(3)*c 3 ,Max(4)*c 4 ,Max(5)*c5]T；
Calculating to obtain ice-caterpillar probability P (X; theta):
wherein:
exp (·) represents an exponent based on a natural constant;
θ＝(θ 1 ,θ 2 ,θ 3 ,θ 4 ,θ 5 ) The probability distribution parameters of the optimal wind driven generator ice-caterpillar time sequence analysis model are represented;
updating the prior probability according to the ice-carrying probability P (X; theta);
if the ice-carrying probability P (X; theta) is greater than or equal to the probability threshold, the wind driven generator is in the ice-carrying state, and if the ice-carrying probability P (X; theta) is greater than or equal to the probability threshold, the wind driven generator is in the ice-carrying state.
CN202310693921.XA 2023-06-13 2023-06-13 Wind driven generator ice prediction method based on time sequence analysis Pending CN116680636A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310693921.XA CN116680636A (en) 2023-06-13 2023-06-13 Wind driven generator ice prediction method based on time sequence analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310693921.XA CN116680636A (en) 2023-06-13 2023-06-13 Wind driven generator ice prediction method based on time sequence analysis

Publications (1)

Publication Number Publication Date
CN116680636A true CN116680636A (en) 2023-09-01

Family

ID=87780731

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310693921.XA Pending CN116680636A (en) 2023-06-13 2023-06-13 Wind driven generator ice prediction method based on time sequence analysis

Country Status (1)

Country Link
CN (1) CN116680636A (en)

Similar Documents

Publication Publication Date Title
CN113159147B (en) Image recognition method and device based on neural network and electronic equipment
CN113283446B (en) Method and device for identifying object in image, electronic equipment and storage medium
CN113705462B (en) Face recognition method, device, electronic equipment and computer readable storage medium
CN114998838B (en) Real-time monitoring method and device for intelligent storage cabinet of hazardous chemical substances
CN111831430A (en) Electrical equipment defect identification system based on edge calculation
CN114398557B (en) Information recommendation method and device based on double images, electronic equipment and storage medium
CN115389439B (en) River pollutant monitoring method and system based on big data
CN116662839A (en) Associated big data cluster analysis method and device based on multidimensional intelligent acquisition
CN117036732B (en) Electromechanical equipment detection system, method and equipment based on fusion model
CN117373580A (en) Performance analysis method and system for realizing titanium alloy product based on time sequence network
CN116011998B (en) Retired battery recycling and classifying treatment method and device
CN115798004A (en) Face card punching method and device based on local area, electronic equipment and medium
CN116680636A (en) Wind driven generator ice prediction method based on time sequence analysis
CN113515591B (en) Text defect information identification method and device, electronic equipment and storage medium
CN112580505B (en) Method and device for identifying network point switch door state, electronic equipment and storage medium
CN117074628B (en) Multi-sensor air quality detection equipment fault positioning method
CN111178630A (en) Load prediction method and device
CN113868093B (en) Junk file monitoring method, device, equipment and storage medium
CN116681670A (en) Method for monitoring ice-caterpillar of wind driven generator
CN117315368B (en) Intelligent operation and maintenance inspection method for large-scale data center
CN113160144B (en) Target object detection method, target object detection device, electronic equipment and storage medium
CN116883764B (en) Battery system fault identification method and device
CN115860275B (en) Method and system for land preparation benefit overall surveying and mapping acquisition
CN113191455B (en) Edge computing box election method and device, electronic equipment and medium
CN116912190A (en) Wind driven generator blade structure damage analysis method and system based on feedback network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination