A wind power main control system fault monitoring and early warning method

Abstract

The present invention discloses a method for monitoring and warning faults of a wind power main control system, comprising: collecting historical operating status data of the wind power main control system, extracting relevant parameter information containing typical faults in the historical operating status data and preprocessing; establishing a fault warning model, dividing the preprocessed data into a training set and a test set, and using the training set and the test set to train the fault warning model; monitoring the real-time operating data of the wind power main control system and providing fault warnings according to the trained fault warning model. The method of the present invention establishes a complete and comprehensive wind power generator set status monitoring and fault diagnosis scheme, which is of great significance for correctly evaluating the operating status of the unit, preventing and reducing the occurrence of unit failures, and realizing predictive maintenance.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power main control system and fault monitoring, and in particular to a wind power main control system fault monitoring and early warning method. Background Art

[0002] As countries around the world pay more and more attention to environmental protection, energy shortage and energy conservation, wind energy, as a clean and abundant renewable energy, is increasingly valued by all countries. However, most wind turbines are built in high mountains, at sea or at the seaside where the wind is relatively strong, and their working conditions are very harsh. In addition, the wind power system has a complex structure, many controls, and high control level requirements. In addition, it operates continuously for a long time under various environmental conditions of different intensities, so it is inevitable that various faults will occur. At present, domestic condition monitoring fault diagnosis technology has been widely used in the petroleum, chemical, electric power (mainly thermal power) and metallurgical industries, and has achieved very good results, but the application of this technology in the field of wind power generation is still in the exploratory stage.

[0003] The failure mechanism of wind turbines is very complex and has many influencing factors, which are the key factors affecting whether wind turbines can operate stably in the long term. Establishing a complete and comprehensive wind turbine status monitoring and fault diagnosis program is of great significance for correctly evaluating the operating status of the unit, preventing and reducing the occurrence of unit failures, and realizing predictive maintenance. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is that the existing technical solutions are still in the exploratory stage in the field of wind power generation, and their fault monitoring and early warning accuracy is low, making it difficult to prevent and reduce the occurrence of unit failures.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: collecting historical operating status data of the wind power main control system, extracting relevant parameter information containing typical faults in the historical operating status data and preprocessing it; establishing a fault warning model, dividing the preprocessed data into a training set and a test set, and using the training set and the test set to train the fault warning model; monitoring the real-time operating data of the wind power main control system and issuing fault warnings according to the trained fault warning model.

[0008] As a preferred solution of the wind power master control system fault monitoring and early warning method described in the present invention, the historical operating status data of the wind power master control system includes power grid parameters, meteorological parameters and unit status parameters.

[0009] As a preferred solution of the wind power main control system fault monitoring and early warning method described in the present invention, the grid parameters include grid three-phase voltage, three-phase current, grid frequency, and power factor.

[0010] As a preferred solution of the wind power master control system fault monitoring and early warning method described in the present invention, the meteorological parameters include wind speed, wind direction, and ambient temperature.

[0011] As a preferred scheme of the wind power main control system fault monitoring and early warning method described in the present invention, the unit status parameters include wind wheel speed, generator speed, generator coil temperature, generator front and rear bearing temperature, gear box oil temperature, gear box front and rear bearing temperature, hydraulic system oil temperature, oil pressure, oil level, cabin vibration, cable rotation, and cabin temperature.

[0012] As an optimal solution of the wind power main control system fault monitoring and early warning method described in the present invention, the historical operating status data contains relevant parameter information of typical faults, including the power grid parameters, the meteorological parameters, and the unit status parameters, which differ from the normal operating values ​​by more than the set threshold.

[0013] As a preferred solution of the wind power main control system fault monitoring and early warning method described in the present invention, the preprocessing process of the relevant parameter information includes: data cleaning: format content cleaning, non-demand data cleaning, vacancy value cleaning, and logical error cleaning; data transformation: feature construction, data classification, and data quantification of the data; data integration: data statistics of the transformed data, and merging the data into a unified data storage; using a clustering-based outlier sample detection strategy to detect and eliminate samples that may still be abnormal in the data samples; performing signal conversion on the data obtained after detection and elimination, that is, converting the obtained data into a spectrum diagram.

[0014] As a preferred solution of the wind power main control system fault monitoring and early warning method described in the present invention, the establishment of the fault early warning model includes: based on the spectrum diagram as continuous points, a continuous point linear model is constructed using the least squares support vector machine principle; using the Chebyshev inequality, the obtained outlier parameter points are identified, and the value that deviates from the comparison point in the continuous point linear model by K standard deviations is defined as an abnormal value.

[0015] As a preferred solution of the wind power main control system fault monitoring and early warning method described in the present invention, the training set and the test set are divided in a ratio of 7:3.

[0016] Beneficial effects of the invention: The method of the invention establishes a complete and comprehensive wind turbine generator set status monitoring and fault diagnosis scheme, which is of great significance for correctly evaluating the operating status of the unit, preventing and reducing the occurrence of unit failures, and realizing predictive maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0018] Figure 1 A schematic diagram of a basic flow chart of a wind power main control system fault monitoring and early warning method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0022] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0023] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0024] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0025] Example 1

[0026] Reference Figure 1 , is an embodiment of the present invention, and provides a wind power main control system fault monitoring and early warning method, comprising:

[0027] S1: Collect historical operating status data of the wind power main control system, extract relevant parameter information containing typical faults in the historical operating status data and perform preprocessing;

[0028] It should be noted that the historical operating status data of the wind power main control system includes power grid parameters, meteorological parameters and unit status parameters.

[0029] Among them, the grid parameters include grid three-phase voltage, three-phase current, grid frequency, and power factor;

[0030] Meteorological parameters include wind speed, wind direction, and ambient temperature;

[0031] The unit status parameters include wind rotor speed, generator speed, generator coil temperature, generator front and rear bearing temperature, gearbox oil temperature, gearbox front and rear bearing temperature, hydraulic system oil temperature, oil pressure, oil level, cabin vibration, cable rotation, and cabin temperature.

[0032] Furthermore, the historical operating status data contains relevant parameter information of typical faults including power grid parameters, meteorological parameters, and unit status parameters whose difference from normal operating values ​​is greater than a set threshold.

[0033] In addition, temperature sensors, voltage sensors, current sensors, sound or imaging sensors, etc. are used to collect the required parameters.

[0034] Furthermore, the preprocessing process of relevant parameter information includes:

[0035] Data cleaning: format content cleaning, non-required data cleaning, missing value cleaning, and logical error cleaning;

[0036] Data transformation: constructing features, classifying data, and quantifying data;

[0037] Data integration: Perform data statistics on the transformed data and merge the data into a unified data storage;

[0038] A clustering-based outlier sample detection strategy is used to detect and remove samples that may still be abnormal in the data sample;

[0039] The data obtained after detection and elimination are subjected to signal conversion, that is, the obtained data are converted into a spectrum diagram.

[0040] Among them, typical faults include grid voltage flicker, overvoltage, low voltage, voltage drop, phase sequence failure, three-phase asymmetry, etc.

[0041] S2: Establish a fault warning model, divide the preprocessed data into a training set and a test set, and use the training set and the test set to train the fault warning model;

[0042] It should be noted that the establishment of the fault warning model includes:

[0043] Based on the spectrum graph as continuous points, the continuous point linear model is constructed using the principle of least squares support vector machine;

[0044] The obtained outlier parameter points are identified by using Chebyshev inequality, and the value that deviates from the comparison point in the continuous point linear model by K standard deviations is defined as an outlier.

[0045] Among them, the continuous point linear model is:

[0046] set up is a given sample set, where n represents the total number of sample sets, x k Represents the input variable, y k represents the output variable, and M represents the dimension presented by the input variable;

[0047] The least squares support vector machine is based on nonlinear functions in:

[0048]

[0049] The low-dimensional features are effectively transformed into high-dimensional features. The expression of the model is:

[0050]

[0051] Where b represents the offset, represents the mapping function, ω represents the value of the weight vector;

[0052] The final least squares support vector machine is optimized and described as follows:

[0053]

[0054]

[0055] Among them, e k represents the error, γ represents the regularization parameter, and ξ represents the error variable;

[0056] Optimization is performed based on the Lagrange method to form the optimization expression:

[0057]

[0058] Based on the optimization conditions, Eliminating ω and ξ in the above formula yields:

[0059]

[0060] The above parameters are:

[0061]

[0062] λ=[λ 1 ,λ 2 , …λ n ]

[0063] q=[y 1 ,y 2 ,…y n ] T

[0064] E = [1, 1, ... 1]

[0065] Where I is the identity matrix;

[0066] Define the formula to satisfy Mercer's symmetric kernel function, then the linear model obtained by least squares support vector machine transformation is:

[0067]

[0068] The obtained outlier parameter points are identified using Chebyshev inequality. The specific formula is as follows:

[0069] p(|x-μ|>kδ)≤1 / k 2

[0070] Assuming that the overall distribution of the variable is in an arbitrary state, the probability density function obtained after fitting analysis is f(x). When the probability value is small, it means that there are outliers in the sample. The calculation formula of the probability density function is:

[0071]

[0072] In addition, the training set and test set are divided in a ratio of 7:3.

[0073] S3: Monitor the real-time operation data of the wind power main control system and issue fault warnings based on the trained fault warning model.

[0074] The method of the present invention acquires comprehensive parameters of the wind power main control system and establishes a fault warning model for predicting faults in advance, thereby preventing and reducing the occurrence of unit failures. That is, the present invention establishes a complete and comprehensive wind turbine status monitoring and fault diagnosis solution, which is of great significance for correctly evaluating the operating status of the unit, preventing and reducing the occurrence of unit failures, and realizing predictive maintenance.

[0075] Example 2

[0076] This embodiment is another embodiment of the present invention. This embodiment is different from the first embodiment in that it provides a verification test of a wind power main control system fault monitoring and early warning method. In order to verify and illustrate the technical effects adopted in this method, this embodiment adopts a traditional technical solution and the method of the present invention for comparative testing, and compares the test results by means of scientific argumentation to verify the actual effect of this method.

[0077] Traditional technical solutions: high installation cost, limited fault warning methods in practical applications, low sampling frequency, more data and less fault data, resulting in low fault warning accuracy. To verify that this method has higher warning accuracy than traditional methods. In this embodiment, the fault warning method of traditional SCADA data (traditional method 1) and the fault warning method of ultrasonic device (traditional method 2) and this method are used to measure and compare the fault warning accuracy of the simulated wind power main control system in real time.

[0078] Test environment: The operation of the wind farm is simulated on the simulation platform. The overall range of the wind farm is 20km. 2 , the east-west length is 6km, the north-south length is about 7.1km, and the average wind speed at different heights is shown in the following table:

[0079] Table 1: Wind speed and power density at different heights of wind farms.

[0080] Height / m Wind speed / m / s <![CDATA[Wind power density W / m 2 > 13 4.9 175.8 75 5.5 310.3

[0081] 25 W2000N-99-80 units are configured. The annual average turbulence intensity and average wind speed in the area where the wind farm is located are lower than the design parameters of the W2000N-99-80 wind turbine unit. The annual minimum temperature of the wind farm is also within the designed operating temperature of the unit. The blade length is 48.38m, the diameter of the wind wheel is 99m, and the control system adopts electric variable pitch mode. The three-point support structure is selected as the drive chain of the unit. The configuration model is a dual PWM converter and a double-fed asynchronous slip-ring motor. Considering the impact of the season and wind speed on the unit, the data of 3-4 months before the abnormal state are selected as the pre-processing object by using MATLB software programming, and the simulation test of the three methods is realized. The simulation data is obtained according to the experimental results. Each method tests 200 sets of data, and the experimental results are shown in the following table.

[0082] Table 2: Experimental results.

[0083] Experimental samples Traditional method 1 Traditional method 2 Method of the present invention Prediction residuals 0.6℃ 1.3℃ 0.2℃ Fault prediction accuracy 96.254％ 90.336％ 99.012％ Latency 5min ≥10min 5s Sampling frequency 28min / time 30min / time 5min / time Failure data ratio 56％ 45％ 88％

[0084] It can be seen from the above table that the method of the present invention has better prediction accuracy and precision than the traditional fault prediction method, thereby being able to prevent and reduce the occurrence of unit failures and achieve predictive maintenance.

[0085] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A wind power main control system fault monitoring and early warning method, characterized in that: include: Collect historical operating status data of the wind power main control system, extract relevant parameter information containing typical faults in the historical operating status data and perform preprocessing; The preprocessing process of relevant parameter information includes: Data cleaning: format content cleaning, non-required data cleaning, missing value cleaning, and logical error cleaning; Data transformation: constructing features, classifying data, and quantifying data; Data integration: Perform data statistics on the transformed data and merge the data into a unified data storage; A clustering-based outlier sample detection strategy is used to detect and remove samples that may still be abnormal in the data sample; Perform signal conversion on the data obtained after detection and elimination, that is, convert the obtained data into a spectrum diagram; Establishing a fault warning model, dividing the preprocessed data into a training set and a test set, and using the training set and the test set to train the fault warning model; The training set and the test set are divided in a ratio of 7:3; Monitor the real-time operation data of the wind power main control system and provide fault warning based on the trained fault warning model; The establishment of the fault warning model includes: Based on the fact that the spectrum graph is continuous points, a continuous point linear model is constructed using the principle of least squares support vector machine; Using Chebyshev's inequality, the obtained outlier parameter points are identified, and the value that deviates from the comparison point in the continuous point linear model by K standard deviations is defined as an outlier; Among them, the continuous point linear model is: set up is a given sample set, where n represents the total number of sample sets, x k Represents the input variable, y k represents the output variable, and M represents the dimension presented by the input variable; The least squares support vector machine is based on nonlinear functions in: The low-dimensional features are effectively transformed into high-dimensional features. The expression of the model is: Where b represents the offset, represents the mapping function, ω represents the value of the weight vector; The final least squares support vector machine is optimized and described as follows: Among them, e k represents the error, γ represents the regularization parameter, and ζ represents the error variable; Optimization is performed based on the Lagrange method to form the optimization expression: Based on the optimization conditions, Eliminating ω and ξ in the above formula yields: The above parameters are: λ=[λ1,λ2,…λ n ] q=[y1,y2,…y n ] T E＝[1，1，…1] Where I is the identity matrix; Define the formula to satisfy Mercer's symmetric kernel function, then the linear model obtained by least squares support vector machine transformation is: The obtained outlier parameter points are identified using Chebyshev inequality. The specific formula is as follows: p(|x-μ|>kδ)≤1 / k 2 Assuming that the overall distribution of the variable is in an arbitrary state, the probability density function obtained after fitting analysis is f(x). When the probability value is small, it means that there are outliers in the sample. The calculation formula of the probability density function is:

2. The wind power master control system fault monitoring and early warning method according to claim 1, characterized in that: The historical operating status data of the wind power master control system includes power grid parameters, meteorological parameters and unit status parameters.

3. The wind power master control system fault monitoring and early warning method according to claim 2, characterized in that: The grid parameters include grid three-phase voltage, three-phase current, grid frequency, and power factor.

4. The wind power master control system fault monitoring and early warning method according to claim 2, characterized in that: The meteorological parameters include wind speed, wind direction, and ambient temperature.

5. The wind power master control system fault monitoring and early warning method according to claim 2, characterized in that: The unit status parameters include wind wheel speed, generator speed, generator coil temperature, generator front and rear bearing temperature, gearbox oil temperature, gearbox front and rear bearing temperature, hydraulic system oil temperature, oil pressure, oil level, cabin vibration, cable rotation, and cabin temperature.

6. The wind power master control system fault monitoring and early warning method according to any one of claims 1 to 5, characterized in that: The historical operating status data contains relevant parameter information of typical faults, including power grid parameters, meteorological parameters, and unit status parameters whose difference from normal operating values ​​is greater than a set threshold.

Images