CN115828163B - Power transmission tower health state data classification method - Google Patents
Power transmission tower health state data classification method Download PDFInfo
- Publication number
- CN115828163B CN115828163B CN202211543620.0A CN202211543620A CN115828163B CN 115828163 B CN115828163 B CN 115828163B CN 202211543620 A CN202211543620 A CN 202211543620A CN 115828163 B CN115828163 B CN 115828163B
- Authority
- CN
- China
- Prior art keywords
- hyperplane
- samples
- original
- state data
- data set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The invention discloses a method for classifying health state data of a power transmission tower, which comprises the following steps: firstly, collecting the running state data of the iron tower through a sensor array of inclination angle, vibration, temperature and humidity and wind speed; step two, according to the health state data set D of the transmission tower; third step, constructing a multi-category central segmentation hyperplane support vector machine to a power transmission tower health state data set D 1 Classifying; fourth step, using test set D tes Performing accuracy verification on a multi-category central segmentation hyperplane support vector machine model MOD; and fifthly, substituting the sample x to be classified into the center segmentation hyperplane support vector machine model MOD. The beneficial effects are that: and the balance of the data set is improved, so that the data set is suitable for machine learning. The situation that the samples simultaneously meet two hyperplanes in the learning process is avoided. In the support vector machine multi-classification problem, all samples can be covered by constructing classifiers with the same number as the number of classes.
Description
Technical Field
The invention relates to a data classification method, in particular to a power transmission tower health state data classification method.
Background
At present, extreme weather and severe environments are important reasons for causing serious tower-reversing accidents of the power transmission tower. The monitoring and early warning means aiming at the state of the iron tower are imperfect, and particularly, strong vibration and serious corrosion of the iron tower can occur in strong wind, heavy rainfall and other weather. At present, the sensor array can rapidly acquire information such as inclination angle, vibration, temperature and humidity, wind speed and the like of the tower body of the iron tower, and rainfall, snowfall and ponding depth information can be acquired through current meteorological information. The information is integrated and extracted, so that a data set of the health state of the iron tower can be obtained, and faults of the iron tower, such as ice coating of a lead wire or a ground wire or an insulator, sedimentation of a tower foundation or rust of a tower body, can be predicted by training a machine learning model through the data set of the health state of the iron tower. In the existing fault method, a multi-classification support vector machine is mainly adopted to directly classify faults of an iron tower data set. The method can generate two problems, namely, the number of samples of each fault in the iron tower data set is unbalanced, and particularly, the number of samples generated by normal operation of the iron tower is far greater than the number of samples when the iron tower breaks down. This situation is detrimental to the model training process of the support vector machine. Secondly, under the traditional multi-classification support vector machine, partial data in a large amount of data exist to simultaneously satisfy fault samples of two different hyperplanes, and the reason for this is that the classification algorithm of the traditional multi-classification support vector machine cannot completely cover all samples, so that improvement on the traditional multi-classification support vector machine algorithm is required.
Disclosure of Invention
The invention aims to solve the problems that in the prior art, various fault data in an iron tower health monitoring data set are unbalanced, and the traditional multi-classification support vector machine algorithm has defects, so that the iron tower health monitoring fault classification result is not accurate enough, and provides a power transmission iron tower health state data classification method.
The invention provides a method for classifying health state data of a power transmission tower, which comprises the following steps:
collecting the running state data of the iron tower through a sensor array of inclination angle, vibration, temperature and humidity and wind speed, acquiring rainfall and snow quantity and water accumulation quantity data at corresponding moments through real-time weather information, wherein each group consists of temperature x 1 Humidity x 2 Wind direction x 3 Wind power x 4 Rainfall and snow quantity x 5 Quantity x of accumulated water 6 Vibration signal x 7 And the magnitude of the inclination anglex 8 Form a sample x i And marks the fault type y i Wherein the fault types include four types: iron tower icing y 1 Wire galloping y 2 Settling y of tower foundation 3 Accumulated water rust y 4 From samples x obtained during a monitoring period i And corresponding fault type y i Constructing a power transmission tower health state data set D;
the second step, according to the health state data set D of the power transmission tower, adopting a synthetic minority class oversampling method to adjust the number of various samples in the data set, and improving the balance of the classified data set, and specifically comprises the following steps:
(1) Editing samples in the power transmission tower health state data set D, removing points with minimum actions on the classification process, and reducing the calculated amount of the follow-up process;
(2) Selecting a sample x by using K-nearest neighbor algorithm i The nearest K homogeneous samples x' i In K homogeneous samples x' i Randomly selecting M samples x n ，n＝[1,2,…M]；
(3) Sample x by separately summing M samples i Connecting, taking a new sample x at any point in the period N Xi is a random number, new sample x N The construction formula of (2) is as follows:
x N ＝x i +ξ(x n -x i )ξ∈(0,1)
repeating the process under the same condition to obtain any number of samples of corresponding types, so that the number of the samples is equal, and constructing a new power transmission tower health state data set D 1 ；
Third step, constructing a multi-category central segmentation hyperplane support vector machine to a power transmission tower health state data set D 1 Sorting, first from the tower-containing ice coating y 1 Wire galloping y 2 Settling y of tower foundation 3 Accumulated water rust y 4 Respectively selecting the same number of samples from the four types of samples to form a training set D tri Through training set D tri Obtaining original hyperplane OH 1 、OH 2 、OH 3 、OH 4 Calculate the original hyperplane OH 1 And OH (OH) 2 Is defined by a central dividing plane OH 12 Original hyperplane OH 2 And OH (OH) 3 Is defined by a central dividing plane OH 23 Original hyperplane OH 3 And OH (OH) 4 Is defined by a central dividing plane OH 34 Original hyperplane OH 4 And OH (OH) 1 Is defined by a central dividing plane OH 41 Dividing the center into planes OH 12 、OH 23 、OH 34 、OH 41 Inputting the model MOD to an original support vector machine model to obtain a multi-category central segmentation hyperplane support vector machine model MOD, wherein the model MOD is specifically as follows:
(1) Selecting a power transmission tower health state data set D 1 Marked y in 1 、y 2 、y 3 And y 4 Of which 70% of the samples constitute training set D tri12 The remaining 30% of the samples make up test set D tes12 Setting y 1 ＝+1，y 2 ＝y 3 ＝y 4 = -1, constraint y i (ωx i +b) is greater than or equal to 1, objective functionSolving an objective function by a Lagrangian multiplier method:
α＝{α 1 ,…,α n }
(2) And solving bias derivative of the target omega, b, converting the maximum problem into the minimum problem, wherein the Lagrange multiplier needs to meet the Karush-Kuhn-Tucker condition because of unequal relation in constraint conditions, and finally solving the equation:
meanwhile, the constraint conditions need to be satisfied:
(3) Substituting alpha obtained by finally solving the equation into an objective function to obtain a corresponding original hyperplane OH 1 ：
ω 1 T x+b 1 ＝0
(4) Repeating the above steps to obtain original hyperplane OH 2 、OH 3 、OH 4 ；
(5) Original hyperplane OH 1 、OH 2 The hyperplane normal vectors of (a) are omega respectively 1 、ω 2 Will omega 1 ，ω 2 Unitized to obtainWhen the original hyperplane OH 1 、OH 2 The included angle theta satisfies the following conditions: />When the center is divided into the plane OH 12 Normal vector omega of (2) 12 The method comprises the following steps:
when the original hyperplane OH 1 、OH 2 The included angle theta satisfies the following conditions:when the center is divided into the plane OH 12 Normal vector omega of (2) 12 The method comprises the following steps:
(6) Finding the central dividing plane OH 12 Original hyperplane OH 1 、OH 2 Is a common solution X of (2) 0 Through the original hyperplane OH 1 ，OH 2 For X 0 And (3) carrying out solving:
construction of a central dividing plane OH 12 ：
OH 12 :ω 12 T x+b 12 ＝0
X 0 Dividing plane OH for the centre 12 Is substituted into the above formula to obtain:
(7) Repeating the processes (5) and (6) to obtain a central division plane OH respectively 23 、OH 34 、OH 41 Through a central dividing plane OH 12 、OH 23 、OH 34 、OH 41 Constructing a multi-category central segmentation hyperplane support vector machine model MOD;
fourth step, using test set D tes Performing accuracy verification on a multi-category central segmentation hyperplane support vector machine model MOD, drawing a result graph, and observing classification effects;
fifthly, substituting the sample x to be classified into the center segmentation hyperplane support vector machine model MOD, and passing through the center segmentation plane OH 12 、OH 23 、OH 34 、OH 41 Judging fault type y of corresponding mark of sample x by result value i ，OH 12 、OH 23 、OH 34 、OH 41 Of the four result values, only one result value is positive, the other three values are negative, if OH 12 The result value is positive, and the category to which the sample x belongs is tower icing y 1 And so on.
The invention has the beneficial effects that:
the method for classifying the health state data of the power transmission tower is based on a multi-class central division hyperplane support vector machine, and improves and classifies the health state data set of the power transmission tower by synthesizing a minority class oversampling method and the multi-class central division hyperplane support vector machine. The method has the following advantages:
1. and the balance of the data set is improved, so that the data set is suitable for machine learning.
2. The situation that the samples simultaneously meet two hyperplanes in the learning process is avoided.
3. In the support vector machine multi-classification problem, all samples can be covered by constructing classifiers with the same number as the number of classes.
Drawings
Fig. 1 is a flow chart of a method for classifying health state data of a power transmission tower according to the present invention.
Detailed Description
Please refer to fig. 1:
the invention provides a method for classifying health state data of a power transmission tower, which comprises the following steps:
collecting the running state data of the iron tower through a sensor array of inclination angle, vibration, temperature and humidity and wind speed, acquiring rainfall and snow quantity and water accumulation quantity data at corresponding moments through real-time weather information, wherein each group consists of temperature x 1 Humidity x 2 Wind direction x 3 Wind power x 4 Rainfall and snow quantity x 5 Quantity x of accumulated water 6 Vibration signal x 7 And dip size x 8 Form a sample x i And marks the fault type y i Wherein the fault types include four types: iron tower icing y 1 Wire galloping y 2 Settling y of tower foundation 3 Accumulated water rust y 4 From samples x obtained during a monitoring period i And corresponding fault type y i Constructing a power transmission tower health state data set D;
the second step, according to the health state data set D of the power transmission tower, adopting a synthetic minority class oversampling method to adjust the number of various samples in the data set, and improving the balance of the classified data set, and specifically comprises the following steps:
(1) Editing samples in the power transmission tower health state data set D, removing points with minimum actions on the classification process, and reducing the calculated amount of the follow-up process;
(2) Utilization ofK-nearest neighbor algorithm, selecting x from a certain sample i The nearest K homogeneous samples x' i In K homogeneous samples x' i Randomly selecting M samples x n ，n＝[1,2,…M]；
(3) Sample x by separately summing M samples i Connecting, taking a new sample x at any point in the period N Xi is a random number, new sample x N The construction formula of (2) is as follows:
x N ＝x i +ξ(x n -x i )ξ∈(0,1)
repeating the process under the same condition to obtain any number of samples of corresponding types, so that the number of the samples is equal, and constructing a new power transmission tower health state data set D 1 ；
Third step, constructing a multi-category central segmentation hyperplane support vector machine to a power transmission tower health state data set D 1 Sorting, first from the tower-containing ice coating y 1 Wire galloping y 2 Settling y of tower foundation 3 Accumulated water rust y 4 Respectively selecting the same number of samples from the four types of samples to form a training set D tri Through training set D tri Obtaining original hyperplane OH 1 、OH 2 、OH 3 、OH 4 Calculate the original hyperplane OH 1 And OH (OH) 2 Is defined by a central dividing plane OH 12 Original hyperplane OH 2 And OH (OH) 3 Is defined by a central dividing plane OH 23 Original hyperplane OH 3 And OH (OH) 4 Is defined by a central dividing plane OH 34 Original hyperplane OH 4 And OH (OH) 1 Is defined by a central dividing plane OH 41 Dividing the center into planes OH 12 、OH 23 、OH 34 、OH 41 Inputting the model MOD to an original support vector machine model to obtain a multi-category central segmentation hyperplane support vector machine model MOD, wherein the model MOD is specifically as follows:
(1) Selecting a power transmission tower health state data set D 1 Marked y in 1 、y 2 、y 3 And y 4 Of which 70% of the samples constitute training set D tri12 The remaining 30% of the samples make up test set D tes12 Setting y 1 ＝+1，y 2 ＝y 3 ＝y 4 = -1, constraint y i (ωx i +b) is greater than or equal to 1, objective functionSolving an objective function by a Lagrangian multiplier method:
α＝{α 1 ,…,α n }
(2) And solving bias derivative of the target omega, b, converting the maximum problem into the minimum problem, wherein the Lagrange multiplier needs to meet the Karush-Kuhn-Tucker condition because of unequal relation in constraint conditions, and finally solving the equation:
meanwhile, the constraint conditions need to be satisfied:
(3) Substituting alpha obtained by finally solving the equation into an objective function to obtain a corresponding original hyperplane OH 1 ：
ω 1 T x+b 1 ＝0
(4) Repeating the above steps to obtain original hyperplane OH 2 、OH 3 、OH 4 ；
(5) Original hyperplane OH 1 、OH 2 The hyperplane normal vectors of (a) are omega respectively 1 、ω 2 Will omega 1 ，ω 2 Unitized to obtainWhen the original hyperplane OH 1 、OH 2 The included angle theta satisfies the following conditions: />When the center is divided into the plane OH 12 Normal vector omega of (2) 12 The method comprises the following steps:
when the original hyperplane OH 1 、OH 2 The included angle theta satisfies the following conditions:when the center is divided into the plane OH 12 Normal vector omega of (2) 12 The method comprises the following steps:
(6) Finding the central dividing plane OH 12 Original hyperplane OH 1 、OH 2 Is a common solution X of (2) 0 Through the original hyperplane OH 1 ，OH 2 For X 0 And (3) carrying out solving:
construction of a central dividing plane OH 12 ：
OH 12 :ω 12 T x+b 12 ＝0
X 0 Dividing plane OH for the centre 12 Is substituted into the above formula to obtain:
(7) Repeating the processes (5) and (6) to obtain a central division plane OH respectively 23 、OH 34 、OH 41 Through a central dividing plane OH 12 、OH 23 、OH 34 、OH 41 Constructing a multi-category central segmentation hyperplane support vector machine model MOD;
fourth step, using test set D tes Performing accuracy verification on a multi-category central segmentation hyperplane support vector machine model MOD, drawing a result graph, and observing classification effects;
fifthly, substituting the sample x to be classified into the center segmentation hyperplane support vector machine model MOD, and passing through the center segmentation plane OH 12 、OH 23 、OH 34 、OH 41 Judging fault type y of corresponding mark of sample x by result value i ，OH 12 、OH 23 、OH 34 、OH 41 Of the four result values, only one result value is positive, the other three values are negative, if OH 12 The result value is positive, and the category to which the sample x belongs is tower icing y 1 And so on.
Claims (1)
1. A power transmission tower health state data classification method is characterized in that: the method comprises the following steps:
collecting the running state data of the iron tower through a sensor array of inclination angle, vibration, temperature and humidity and wind speed, acquiring rainfall and snow quantity and water accumulation quantity data at corresponding moments through real-time weather information, wherein each group consists of temperature x 1 Humidity x 2 Wind direction x 3 Wind power x 4 Rainfall and snow quantity x 5 Quantity x of accumulated water 6 Vibration signal x 7 And dip size x 8 Form a sample x i And marks the fault type y i Wherein the fault types include four types: iron tower icing y 1 Wire galloping y 2 Settling y of tower foundation 3 Accumulated water rust y 4 From samples x obtained during a monitoring period i And corresponding fault type y i Constructing a power transmission tower health state data set D;
the second step, according to the health state data set D of the power transmission tower, adopting a synthetic minority class oversampling method to adjust the number of various samples in the data set, and improving the balance of the classified data set, and specifically comprises the following steps:
(1) Editing samples in the power transmission tower health state data set D, removing points with minimum actions on the classification process, and reducing the calculated amount of the follow-up process;
(2) Selecting a sample x by using K-nearest neighbor algorithm i The nearest K homogeneous samples x' i In K homogeneous samples x' i Randomly selecting M samples x n ，n＝[1，2…M]；
(3) Sample x by separately summing M samples i Connecting, taking a new sample x at any point in the period N Xi is a random number, new sample x N The construction formula of (2) is as follows:
x N ＝x i +ξ(x n -x i )ξ∈(0，1)
repeating the process under the same condition to obtain any number of samples of corresponding types, so that the number of the samples is equal, and constructing a new power transmission tower health state data set D 1 ；
Third step, constructing a multi-category central segmentation hyperplane support vector machine to a power transmission tower health state data set D 1 Sorting, first from the tower-containing ice coating y 1 Wire galloping y 2 Settling y of tower foundation 3 Accumulated water rust y 4 Respectively selecting the same number of samples from the four types of samples to form a training set D tri Through training set D tri Obtaining original hyperplane OH 1 、OH 2 、OH 3 、OH 4 Calculate the original hyperplane OH 1 And OH (OH) 2 Is defined by a central dividing plane OH 12 Original hyperplane OH 2 And OH (OH) 3 Is defined by a central dividing plane OH 23 Original hyperplane OH 3 And OH (OH) 4 Is defined by a central dividing plane OH 34 Original hyperplane OH 4 And OH (OH) 1 Is defined by a central dividing plane OH 41 Dividing the center into planes OH 12 、OH 23 、OH 34 、OH 41 Inputting the model MOD to an original support vector machine model to obtain a multi-category central segmentation hyperplane support vector machine model MOD, wherein the model MOD is specifically as follows:
(1) Selecting health of transmission towersState data set D 1 Marked y in 1 、y 2 、y 3 And y 4 Of which 70% of the samples constitute training set D tri12 The remaining 30% of the samples make up test set D tes12 Setting y 1 ＝+1，y 2 ＝y 3 ＝y 4 = -1, constraint y i (ωx i +b) is greater than or equal to 1, objective functionSolving an objective function by a Lagrangian multiplier method:
(2) And solving bias derivative of the target omega, b, converting the maximum problem into the minimum problem, wherein the Lagrange multiplier needs to meet the Karush-Kuhn-Tucker condition because of unequal relation in constraint conditions, and finally solving the equation:
meanwhile, the constraint conditions need to be satisfied:
(3) Substituting alpha obtained by finally solving the equation into an objective function to obtain a corresponding original hyperplane OH 1 ：
ω 1 T x+b 1 ＝0
(4) Repeating the above steps to obtain original hyperplane OH 2 、OH 3 、OH 4 ；
(5) Original hyperplane OH 1 、OH 2 The hyperplane normal vectors of (a) are omega respectively 1 、ω 2 Will omega 1 ，ω 2 Unitized to obtainTo the point ofWhen the original hyperplane OH 1 、OH 2 The included angle theta satisfies the following conditions: />When the center is divided into the plane OH 12 Normal vector omega of (2) 12 The method comprises the following steps:
when the original hyperplane OH 1 、OH 2 The included angle theta satisfies the following conditions:when the center is divided into the plane OH 12 Normal vector omega of (2) 12 The method comprises the following steps:
(6) Finding the central dividing plane OH 12 Original hyperplane OH 1 、OH 2 Is a common solution X of (2) 0 Through the original hyperplane OH 1 ，OH 2 For X 0 And (3) carrying out solving:
construction of a central dividing plane OH 12 ：
OH 12 ：ω 12 T x+b 12 ＝0
X 0 Dividing plane OH for the centre 12 Is substituted into the above formula to obtain:
(7) Repeating the processes (5) and (6) to obtain a central division plane OH respectively 23 、OH 34 、OH 41 Through a central dividing plane OH 12 、OH 23 、OH 34 、OH 41 Constructing a multi-category central segmentation hyperplane support vector machine model MOD;
fourth step, using test set D tes Performing accuracy verification on a multi-category central segmentation hyperplane support vector machine model MOD, drawing a result graph, and observing classification effects;
fifthly, substituting the sample x to be classified into the center segmentation hyperplane support vector machine model MOD, and passing through the center segmentation plane OH 12 、OH 23 、OH 34 、OH 41 Judging fault type y of corresponding mark of sample x by result value i ，OH 12 、OH 23 、OH 34 、OH 41 Of the four result values, only one result value is positive, the other three values are negative, if OH 12 The result value is positive, and the category to which the sample x belongs is tower icing y 1 And so on.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211543620.0A CN115828163B (en) | 2022-12-03 | 2022-12-03 | Power transmission tower health state data classification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211543620.0A CN115828163B (en) | 2022-12-03 | 2022-12-03 | Power transmission tower health state data classification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115828163A CN115828163A (en) | 2023-03-21 |
| CN115828163B true CN115828163B (en) | 2023-08-18 |
Family
ID=85543901
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211543620.0A Active CN115828163B (en) | 2022-12-03 | 2022-12-03 | Power transmission tower health state data classification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115828163B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239516A (en) * | 2014-09-17 | 2014-12-24 | 南京大学 | Unbalanced data classification method |
| CN106650828A (en) * | 2017-01-03 | 2017-05-10 | 电子科技大学 | Support vector machine-based intelligent terminal security level classification method |
| CN112149760A (en) * | 2020-10-28 | 2020-12-29 | 哈尔滨工业大学 | Heterogeneous inner hyperplane-based fuzzy support vector machine design method |
| CN113516160A (en) * | 2021-04-15 | 2021-10-19 | 国网上海市电力公司 | Transmission tower attitude prediction method and system based on support vector machine |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10181102B2 (en) * | 2015-01-22 | 2019-01-15 | Tata Consultancy Services Limited | Computer implemented classification system and method |
-
2022
- 2022-12-03 CN CN202211543620.0A patent/CN115828163B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239516A (en) * | 2014-09-17 | 2014-12-24 | 南京大学 | Unbalanced data classification method |
| CN106650828A (en) * | 2017-01-03 | 2017-05-10 | 电子科技大学 | Support vector machine-based intelligent terminal security level classification method |
| CN112149760A (en) * | 2020-10-28 | 2020-12-29 | 哈尔滨工业大学 | Heterogeneous inner hyperplane-based fuzzy support vector machine design method |
| CN113516160A (en) * | 2021-04-15 | 2021-10-19 | 国网上海市电力公司 | Transmission tower attitude prediction method and system based on support vector machine |
Non-Patent Citations (1)
| Title |
|---|
| Research and Implementation of Power System Data Visualization;Peiguang Chen 等;2022 5th International Conference on Robotics, Control and Automation Engineering;第169-173页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115828163A (en) | 2023-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109740175B (en) | Outlier discrimination method for power curve data of wind turbine generator | |
| CN109050535B (en) | Rapid terrain condition identification method based on vehicle attitude | |
| CN110728411A (en) | High-low altitude area combined rainfall prediction method based on convolutional neural network | |
| CN113505655B (en) | Intelligent bearing fault diagnosis method for digital twin system | |
| CN111366931B (en) | Cloud precipitation refined inversion method based on cloud radar | |
| CN111582596A (en) | Pure electric vehicle endurance mileage risk early warning method integrating traffic state information | |
| CN109255094A (en) | Commercial truck quality estimation method based on SVR_Adaboost innovatory algorithm | |
| CN112950634B (en) | Unmanned aerial vehicle inspection-based wind turbine blade damage identification method, equipment and system | |
| CN113222385B (en) | Method for constructing and evaluating driving condition of electric automobile | |
| CN109635440B (en) | Overhead transmission line icing flashover tripping probability calculation method | |
| CN105740914A (en) | Vehicle license plate identification method and system based on neighboring multi-classifier combination | |
| CN111597901A (en) | Illegal billboard monitoring method | |
| WO2024051524A1 (en) | Joint prediction method and apparatus for hydraulic, wind and photovoltaic generation power | |
| CN112287980B (en) | Power battery screening method based on typical feature vector | |
| CN115828163B (en) | Power transmission tower health state data classification method | |
| CN111709336A (en) | Highway pedestrian detection method and device and readable storage medium | |
| Molina-Garcia et al. | Vertical wind profile characterization and identification of patterns based on a shape clustering algorithm | |
| CN112364706A (en) | Small sample bearing fault diagnosis method based on class imbalance | |
| CN111367174B (en) | Linear quadratic form control improvement method based on convolutional neural network vibration identification | |
| CN112711981A (en) | High-precision transmission line defect identification method | |
| CN112860183A (en) | Multisource distillation-migration mechanical fault intelligent diagnosis method based on high-order moment matching | |
| CN106650801B (en) | A kind of polymorphic type vehicle classification method based on GPS data | |
| CN113523904A (en) | Cutter wear detection method | |
| CN109214582B (en) | Wind direction space correlation driven railway line wind speed prediction method | |
| CN113805252B (en) | System for forecasting gale in tropical cyclone landing process based on ensemble forecasting model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |