Power transmission line icing monitoring system and method based on sunlight intensity

Abstract

The invention provides a sunshine intensity-based power transmission line icing monitoring system and method, the sunshine intensity-based power transmission line icing monitoring system is provided with an acquisition unit, a processing unit, a communication unit, an alarm unit and an automatic control system, and the processing unit has data processing, SVM algorithm analysis and data storage; the using method of the system comprises the steps that data processing and SVM algorithm analysis are conducted on data collected by the collecting unit through the processing unit, the SVM algorithm analyzes and quantifies the influence of sunlight intensity on icing of the power transmission line, the thickness of the icing of the power transmission line is monitored in real time, and the change trend of the icing is predicted. The processing unit is linked with the communication unit and the alarm unit according to the analyzed data condition, the communication unit performs data interaction with a remote management center and can perform manual control, and the alarm unit interacts with an automatic control system to perform early warning on related personnel and automatically control related equipment to process icing of the power transmission line. And safe operation of the power transmission line is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of road monitoring systems, and specifically to a transmission line icing monitoring system and method based on sunlight intensity. Background Art

[0002] In cold regions or in winter, icing on transmission lines is one of the common natural disasters in the power system. In severe cases, it can lead to accidents such as line breaks and tower collapses, affecting the safe and stable operation of the power system. Traditional icing monitoring methods mainly rely on manual inspections and simple meteorological parameter monitoring, which have problems such as low monitoring accuracy and slow response speed. In recent years, with the development of sensor technology and artificial intelligence algorithms, icing prediction methods based on meteorological data have gradually become a research hotspot. However, among many monitoring systems and solutions, most only consider temperature and humidity, and do not take into account sunlight, wind speed, etc. that affect icing, which affects the accuracy of the monitoring system. Moreover, the existing systems can only monitor the icing condition of transmission lines in real time, and cannot well distinguish the changing trend of the icing condition of the lines. They cannot well monitor and utilize some icing conditions that can be naturally resolved through environmental changes. Some require a large amount of human and material resources through artificial intervention, which is extremely uneconomical. Therefore, the applicant proposes a transmission line icing monitoring system and method based on sunlight intensity according to the monitoring requirements of transmission line icing, which can more accurately monitor the icing condition of transmission lines, predict the changing trend of the icing condition of transmission lines, automatically take relevant measures or response preparation plans according to the changing trend, enable people to more conveniently and quickly manage and monitor the safe operation of transmission lines, reduce unnecessary maintenance expenditures, and timely discover potential hazards, ensuring the safe operation of transmission lines. Summary of the Invention

[0003] To solve the above technical problems, the present invention proposes a transmission line icing monitoring system and method based on sunlight intensity, by setting an acquisition unit, a processing unit, a communication unit, an alarm unit, and an automatic control system; the processing unit has data processing, SVM algorithm analysis, and data storage; the system performs data processing and SVM algorithm analysis on the data collected by the acquisition unit through the processing unit. The SVM algorithm analysis quantifies the influence of sunlight intensity on the icing of transmission lines, monitors the thickness of the icing on transmission lines in real time, and predicts the changing trend of the icing. The processing unit links the communication unit and the alarm unit according to the analyzed data situation, gives early warnings to relevant personnel, and automatically controls relevant equipment to deal with the icing on transmission lines, ensuring the safe operation of transmission lines.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] A transmission line icing monitoring system and method based on sunlight intensity, characterized in that: the transmission line icing monitoring system based on sunlight intensity is provided with a collection unit, a processing unit, a communication unit, an alarm unit and an automatic control system; the collection unit includes a sunlight sensor, a temperature sensor, a humidity sensor and a wind speed sensor, and the processing unit is provided with a data processing module, an SVM algorithm analysis module, an icing prediction module, a data storage module and a control module; the processing unit is linked with the communication unit and the alarm unit, and the communication unit is linked with a remote management center; the alarm unit is linked with the automatic control system; the usage method of the transmission line icing monitoring system based on sunlight intensity is to collect the sunlight parameters, temperature, humidity and wind speed of the transmission line through the collection unit, and the collected data is transmitted to the processing unit. The data processing module in the processing unit preprocesses the data, performs data cleaning, normalization and feature extraction. The processed data is analyzed by the SVM algorithm analysis to obtain the icing thickness of the transmission line. The icing prediction module predicts the data after the SVM algorithm analysis. The data storage module stores the operation program of the processing unit, as well as the input data and output data. The control module controls the corresponding modules for linkage control according to the result obtained by the SVM algorithm analysis, interacts and controls with the remote management center through the communication unit, and is linked with the automatic control system through the alarm unit to give early warnings and start the automatic control device to control the affected icing transmission line according to a preset plan; when the SVM algorithm of the processing unit analyzes, it quantifies the influence of sunlight intensity on transmission line icing, and monitors the thickness of transmission line icing and predicts the change trend of icing in real time, with good monitoring effect and high accuracy.

[0006] Further, the processing formula used by the data processing module of the processing unit in the transmission line icing monitoring system based on sunlight intensity is:

[0007]

[0008] Where: I is the sunlight intensity;

[0009] T is the ambient temperature;

[0010] H is the ambient humidity;

[0011] W is the ambient wind speed;

[0012] μ is the mean value of the feature;

[0013] σ is the standard deviation of the feature;

[0014] Furthermore, the SVM algorithm analysis of the processing unit of the transmission line icing monitoring system based on sunlight intensity is specifically as follows: by finding an optimal hyperplane to separate samples of different categories and maximizing the distance between the classification boundary and the nearest sample points; for regression problems, SVM introduces an ε-insensitive loss function to find a fitting function such that the deviation between the predicted value and the true value does not exceed ε; it is divided into:

[0015] Linearly separable case:

[0016] For linearly separable data, the goal of SVM is to find a hyperplane:

[0017] w.x + b = 0

[0018] where: w is the normal vector of the hyperplane;

[0019] b is the bias term;

[0020] x is the input feature vector;

[0021] The classification decision function of the hyperplane is:

[0022] f(x) = sign(w·x + b)

[0023] Non-linearly separable case:

[0024] For non-linearly separable data, SVM maps the data to a high-dimensional space by introducing a kernel function to make it linearly separable in the high-dimensional space. The kernel functions include:

[0025] Linear kernel: K(xi, xj) = xi·xj;

[0026] Polynomial kernel: K(xi, xj) = (γxi·xj + r)d;

[0027] Radial basis function kernel: K(xi, xj) = exp(-γ||xi - xj||2);

[0028] SVM regression:

[0029] For the problem of icing thickness prediction, the goal of SVM regression is to find a fitting function:

[0030] f(x) = w·x + b

[0031] such that the deviation between the predicted value f(x) and the true value y does not exceed ε, that is:

[0032] ∣y - f(x)∣ ≤ ε

[0033] The optimization goal of SVM regression is:

[0034]

[0035] Constraints:

[0036]

[0037] Among them: C is the penalty parameter, which is used to balance the complexity of the model and the training error;

[0038] ξi and ξi* are slack variables, which are used to handle errors beyond the ε range.

[0039] Furthermore, the icing prediction module of the processing unit of the transmission line icing monitoring system based on sunlight intensity is trained using the SVM algorithm model, and its specific steps are as follows:

[0040] Step 1: Data preparation, by collecting the sunlight intensity I, ambient temperature T, ambient humidity H, ambient wind speed W, and icing thickness y of a large number of icing transmission lines;

[0041] Step 2: Model training, by normalizing the input data features to make their mean 0 and variance 1; and dividing the data set into training set and test set; then select the kernel function according to the data characteristics, and use cross-validation and grid search to optimize the hyperparameters of the SVM algorithm, including the penalty parameter C, the kernel parameter γ, and the ε value; finally, use the training set data to train the SVM to obtain the optimal hyperplane;

[0042] Step 3: Model prediction, input the real-time collected meteorological data into the trained SVM algorithm model to predict the icing thickness:

[0043] ŷ = f(x) = w·x + b

[0044] Among them: x is the input feature vector;

[0045] ŷ is the predicted icing thickness;

[0046] Step 4: Model evaluation, use the test set data to evaluate the performance of the model, and calculate the mean square error (MSE) or the mean absolute error (MAE):

[0047]

[0048] Step 5: Model prediction, input the test set data into the trained SVM algorithm model to predict the icing thickness:

[0049] ŷ = f(x) = w·x + b.

[0050] Furthermore, the usage method of the transmission line icing monitoring system based on sunlight intensity is specifically as follows:

[0051] Step 1: The collection unit collects meteorological data of the transmission line, specifically: sunshine intensity I, ambient temperature T, ambient humidity H, ambient wind speed W, and transmits the data to the processing unit via a wireless or wired network;

[0052] Step 2: The processing unit pre-processes the incoming data, cleans the data, removes outliers and noise, and normalizes the data to make it meet the input requirements of machine processing;

[0053] Step 3: The processing unit monitors the icing data of the transmission line through the SVM analysis algorithm and icing prediction, and predicts and evaluates the icing development trend;

[0054] Step 4: The control module of the processing unit performs corresponding control and early warning according to the ice monitoring data and prediction data of the transmission line, and interacts with the remote management center through the communication unit. The remote center can perform remote control through the network, send out an alarm signal through the alarm unit, and link the automatic control system to respond to the transmission line in an emergency, and take relevant measures to ensure the safe operation of the line;

[0055] Step 5: The storage module of the processing unit stores the collected data and monitoring data of the system and provides a visual interface for easy access by management personnel.

[0056] The benefits of this application are:

[0057] 1. The transmission line icing monitoring system and method based on sunshine intensity monitors the sunshine intensity, ambient temperature, humidity, wind speed and other meteorological parameters in the area where the transmission line is located in real time, combines the machine learning algorithm vector machine to build an icing prediction model, and uses the algorithm formula to quantify the impact of sunshine intensity on icing thickness, so as to monitor the icing situation of the transmission line with higher accuracy;

[0058] 2. The transmission line icing monitoring system and method based on sunshine intensity can predict the ice thickness in real time, issue early warning signals in time, and initiate corresponding deicing measures or adjust the operating parameters of the transmission line, thereby effectively preventing transmission line failures caused by icing and ensuring the safe and stable operation of the power system;

[0059] 3. The transmission line icing monitoring system based on sunshine intensity and the introduction of meteorological data such as sunshine intensity, combined with kernel functions and optimization methods, can accurately predict the ice thickness of transmission lines; its powerful nonlinear processing capabilities and generalization performance make it an ideal choice in the field of icing monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 Schematic diagram of the system service architecture of the present invention. DETAILED DESCRIPTION

[0061] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments:

[0062] As Figure 1 shown, a transmission line icing monitoring system and method based on solar radiation intensity are shown. The transmission line icing monitoring system based on solar radiation intensity is provided with a collection unit, a processing unit, a communication unit, an alarm unit and an automatic control system; the collection unit includes a solar radiation sensor, a temperature sensor, a humidity sensor and a wind speed sensor, and the processing unit is provided with a data processing module, an SVM algorithm analysis module, an icing prediction module, a data storage module and a control module; the processing unit is linked with the communication unit and the alarm unit, the communication unit is linked with a remote management center; the alarm unit is linked with the automatic control system; the usage method of the transmission line icing monitoring system based on solar radiation intensity is to collect the solar radiation parameters, temperature, humidity and wind speed of the transmission line through the collection unit, and the collected data is transmitted to the processing unit. The data processing module in the processing unit preprocesses the data, performs data cleaning, normalization and feature extraction. The processed data is analyzed through the SVM algorithm analysis to obtain the icing thickness of the transmission line. The icing prediction module predicts the data after the SVM algorithm analysis. The data storage module stores the operation program of the processing unit, as well as the input data and output data. The control module controls the corresponding modules for linkage control according to the results obtained by the SVM algorithm analysis, interacts and controls with the remote management center through the communication unit, and is linked with the automatic control system through the alarm unit to give an early warning and start the automatic control device to control the affected icing transmission line according to a preset plan. The control point of the early warning is to judge whether it exceeds the preset safety threshold according to the predicted icing thickness; if it exceeds the safety threshold, the system automatically issues an early warning signal, and the early warning signal is transmitted to the power dispatching center through a wireless communication network or a wired network; the control module then starts corresponding de-icing measures or adjusts the operation parameters of the transmission line according to the early warning signal, including starting a de-icing device, adjusting the load of the transmission line, switching to a standby line, etc.; when the SVM algorithm of the processing unit analyzes, it quantifies the influence of solar radiation intensity on the icing of the transmission line, and real-time monitors the thickness of the transmission line icing and predicts the change trend of the icing, with good monitoring effect and high accuracy.

[0063] The processing formula used by the data processing module of the processing unit in the transmission line icing monitoring system based on solar radiation intensity is:

[0064]

[0065] where: I is the solar radiation intensity;

[0066] T is the ambient temperature;

[0067] H is the ambient humidity;

[0068] W is the environmental wind speed;

[0069] μ is the mean value of the feature;

[0070] σ is the standard deviation of the feature;

[0071] The SVM algorithm analysis of the processing unit of the transmission line icing monitoring system based on solar radiation intensity is specifically as follows. By finding an optimal hyperplane, samples of different classes are separated, and the distance between the classification boundary and the nearest sample points is maximized; for regression problems, SVM finds a fitting function by introducing the ε-insensitive loss function, such that the deviation between the predicted value and the true value does not exceed ε; it is divided into:

[0072] Linearly separable case:

[0073] For linearly separable data, the goal of SVM is to find a hyperplane:

[0074] w.x + b = 0

[0075] where: w is the normal vector of the hyperplane;

[0076] b is the bias term;

[0077] x is the input feature vector;

[0078] The classification decision function of the hyperplane is:

[0079] f(x) = sign(w·x + b)

[0080] Non-linearly separable case:

[0081] For non-linearly separable data, SVM maps the data to a high-dimensional space by introducing a kernel function to make it linearly separable in the high-dimensional space. The kernel functions include:

[0082] Linear kernel: K(xi, xj) = xi·xj;

[0083] Polynomial kernel: K(xi, xj) = (γxi·xj + r)d;

[0084] Radial basis function kernel: K(xi, xj) = exp(-γ||xi - xj||2);

[0085] SVM regression:

[0086] For the problem of predicting the icing thickness, the goal of SVM regression is to find a fitting function:

[0087] f(x) = w·x + b

[0088] such that the deviation between the predicted value f(x) and the true value y does not exceed ε, that is:

[0089] |y - f(x)| ≤ ε

[0090] The optimization objective of SVM regression is:

[0091]

[0092] Constraint conditions:

[0093]

[0094] Where: C is the penalty parameter, used to balance the complexity of the model and the training error;

[0095] ξi and ξi* are slack variables, used to handle errors beyond the ε range.

[0096] The ice - covering prediction module of the processing unit of the transmission line ice - covering monitoring system based on sunshine intensity is trained using the SVM algorithm model, and its specific steps are as follows:

[0097] Step 1: Data preparation, by collecting the sunshine intensity I, ambient temperature T, ambient humidity H, ambient wind speed W and ice - covering thickness y of a large number of ice - covered transmission lines;

[0098] Step 2: Model training, by normalizing the input data features to make their mean 0 and variance 1; and dividing the data set into a training set and a test set; then selecting a kernel function according to the data characteristics, and using cross - validation and grid search to optimize the hyperparameters of the SVM algorithm, including the penalty parameter C, the kernel parameter γ and the ε value; finally, training the SVM using the training set data to obtain the optimal hyperplane;

[0099] Step 3: Model prediction, inputting the real - time collected meteorological data into the trained SVM algorithm model to predict the ice - covering thickness:

[0100] ŷ = f(x) = w·x + b

[0101] Where: x is the input feature vector;

[0102] ŷ is the predicted ice - covering thickness;

[0103] Taking the following test data as an example:

[0104] Test set data table

[0105]

[0106] Input the test set data into the trained SVM model to predict the ice - covering thickness:

[0107] ŷ = f(x) = w·x + b

[0108] Suppose the trained model parameters are as follows:

[0109] w = [0.5, -0.2, 0.3, -0.1], b = 0.1

[0110] Then the prediction results are as follows:

[0111] For the first set of test data x = [350, -7, 83, 4]x = [350, -7, 83, 4]:

[0112] ŷ1 = 0.5×350 - 0.2×(-7) + 0.3×83 - 0.1×4 + 0.1 = 175 + 1.4 + 24.9 - 0.4 + 0.1 = 200.0mm For the second set of test data x = [150, -11, 92, 6]x = [150, -11, 92, 6]:

[0113] ŷ2 = 0.5×150 - 0.2×(-11) + 0.3×92 - 0.1×6 + 0.1 = 75 + 2.2 + 27.6 - 0.6 + 0.1 = 104.3mm Predicted values: ŷ1 = 200.0mmŷ1 = 200.0mm, ŷ2 = 104.3mmŷ2 = 104.3mm;

[0114] True values: y1 = 4.0mmy1 = 4.0mm, y2 = 9.0mmy2 = 9.0mm.

[0115] By comparing the predicted values with the true values, the performance of the model can be evaluated; if the prediction error is large, the model parameters can be further optimized or the training data can be increased;

[0116] Step 4: Model evaluation, use the test set data to evaluate the performance of the model, and calculate the mean square error (MSE) or mean absolute error (MAE):

[0117]

[0118] Step 5: Model prediction, input the test set data into the trained SVM algorithm model to predict the ice coating thickness:

[0119] ŷ = f(x) = w·x + b.

[0120] The usage method of the transmission line ice coating monitoring system based on solar radiation intensity is specifically as follows:

[0121] Step 1: The acquisition unit collects the meteorological data of the transmission line, specifically: solar radiation intensity I, ambient temperature T, ambient humidity H, ambient wind speed W, and transmits the data to the processing unit through wireless or wired network;

[0122] Step 2: The processing unit preprocesses the incoming data, cleans the data, removes outliers and noise, and performs normalization to make it meet the input requirements for machine processing;

[0123] Step 3: The processing unit monitors the icing data of the transmission line through the SVM analysis algorithm and icing prediction, and predicts and evaluates the development trend of icing;

[0124] Step 4: The control module of the processing unit performs corresponding control and warning based on the icing monitoring data and prediction data of the transmission line, interacts with the remote management center through the communication unit. The remote center can perform remote control through the network, send alarm signals through the alarm unit, and link the automatic control system to perform emergency response on the transmission line, and take relevant measures to ensure the safe operation of the line;

[0125] Step 5: The storage module of the processing unit stores the collected data and monitoring data of the system, and provides a visual interface for easy access by management personnel.

[0126] The above is only a preferred embodiment of the present invention, and it is not intended to limit the present invention in any other form. Any modification or equivalent change made according to the technical essence of the present invention still belongs to the scope protected by the present invention.

Claims

1. A transmission line icing monitoring system based on sunshine intensity, characterized by: The transmission line icing monitoring system based on sunshine intensity is provided with a collection unit, a processing unit, a communication unit, an alarm unit and an automatic control system; the collection unit includes a sunshine sensor, a temperature sensor, a humidity sensor and a wind speed sensor, the processing unit is provided with a data processing module, a SVM algorithm analysis module, an icing prediction module, a data storage module and a control module; the processing unit is linked to a communication unit and an alarm unit, the communication unit is linked to a remote management center; the alarm unit is linked to an automatic control system; the method for using the transmission line icing monitoring system based on sunshine intensity is to collect sunshine parameters, temperature, humidity and wind speed of the transmission line through the collection unit, the collected data is transmitted to the processing unit, the data processing module in the processing unit pre-processes the data, and cleans the data The processed data are analyzed by SVM algorithm to obtain the ice thickness of the transmission line. The ice prediction module predicts the data after SVM algorithm analysis. The data storage module stores the operation program of the processing unit as well as the input data and the output data. The control module controls the corresponding modules for linkage control according to the results of the SVM algorithm analysis. The communication unit and the remote management center are interactively controlled. The alarm unit is linked with the automatic control system to give early warning and start the automatic control device to control the iced transmission line that is affected by the preset scheme. The SVM algorithm analysis of the processing unit quantifies the influence of sunshine intensity on the ice of the transmission line, monitors the thickness of the ice on the transmission line in real time and predicts the changing trend of the ice, with good monitoring effect and high accuracy.

2. The method of a power transmission line icing monitoring system based on sunshine intensity according to claim 1 is characterized in that: The processing formula used by the data processing module of the processing unit in the transmission line icing monitoring system based on sunshine intensity is: Where: I is the sunshine intensity; T is the ambient temperature; H is the ambient humidity; W is the ambient wind speed; μ is the mean of the feature; σ is the standard deviation of the feature.

3. The method of a power transmission line icing monitoring system based on sunshine intensity according to claim 2 is characterized in that: The SVM algorithm analysis of the processing unit of the transmission line icing monitoring system based on sunshine intensity is specifically as follows: by finding an optimal hyperplane, samples of different categories are separated and the distance between the classification boundary and the nearest sample point is maximized; for the regression problem, SVM introduces an ε-insensitive loss function to find a fitting function so that the deviation between the predicted value and the true value does not exceed ε; it is divided into: Linearly separable case: For linearly separable data, the goal of SVM is to find a hyperplane: w.x+b=0 Where: w is the normal vector of the hyperplane; B is the bias term; x is the input feature vector; The classification decision function of the hyperplane is: f(x)=sign(w·x+b) Nonlinear separable case: For nonlinearly separable data, SVM maps the data to a high-dimensional space by introducing a kernel function, making it linearly separable in the high-dimensional space. The kernel function includes: Linear kernel: K(xi,xj)=xi·xj; Polynomial kernel: K(xi,xj)=(γxi·xj+r)d; Radial basis function kernel: K(xi,xj)=exp(-γ||xi-xj||2); SVM Regression: For the ice thickness prediction problem, the goal of SVM regression is to find a fitting function: f(x)=w·x+b Make the deviation between the predicted value f(x) and the true value y not exceed ε, that is: ∣yf(x)∣≤ε The optimization goal of SVM regression is: Constraints: Where: C is the penalty parameter, which is used to balance the complexity of the model and the training error; ξi and ξi* are slack variables used to handle errors outside the range of ε.

4. The method of a power transmission line icing monitoring system based on sunshine intensity according to claim 2 is characterized in that: The icing prediction module of the processing unit of the transmission line icing monitoring system based on sunshine intensity is trained using the SVM algorithm model, and the specific steps are as follows: Step 1: Data preparation: by collecting the sunshine intensity I, ambient temperature T, ambient humidity H, ambient wind speed W and ice thickness y of a large number of ice-covered transmission lines; Step 2: Model training. Normalize the input data features to make their mean 0 and variance 1. Divide the data set into a training set and a test set. Then select the kernel function based on the data characteristics, and use cross-validation and grid search to optimize the hyperparameters of the SVM algorithm, including the penalty parameter C, the kernel parameter γ and the ε value. Finally, use the training set data to train the SVM and obtain the optimal hyperplane. Step 3: Model prediction: input the real-time collected meteorological data into the trained SVM algorithm model to predict the ice thickness: y^=f(x)=w.x+b Where: x is the input feature vector; y^ is the predicted ice thickness; Step 4: Model evaluation: Use the test set data to evaluate the performance of the model and calculate the mean square error (MSE) or mean absolute error (MAE): Step 5: Model prediction: Input the test set data into the trained SVM algorithm model to predict the ice thickness: y^=f(x)=w·x+b.

5. The method of a power transmission line icing monitoring system based on sunshine intensity according to claim 2 is characterized in that: The method for using the transmission line icing monitoring system based on sunshine intensity is specifically as follows: Step 1: The collection unit collects meteorological data of the transmission line, specifically: sunshine intensity I, ambient temperature T, ambient humidity H, ambient wind speed W, and transmits the data to the processing unit via a wireless or wired network; Step 2: The processing unit pre-processes the incoming data, cleans the data, removes outliers and noise, and normalizes the data to make it meet the input requirements of machine processing; Step 3: The processing unit monitors the icing data of the transmission line through the SVM analysis algorithm and icing prediction, and predicts and evaluates the icing development trend; Step 4: The control module of the processing unit performs corresponding control and early warning according to the ice monitoring data and prediction data of the transmission line, and interacts with the remote management center through the communication unit. The remote center can perform remote control through the network, send out an alarm signal through the alarm unit, and link the automatic control system to respond to the transmission line in an emergency, and take relevant measures to ensure the safe operation of the line; Step 5: The storage module of the processing unit stores the collected data and monitoring data of the system and provides a visual interface for easy access by management personnel.

Images