RFID-based power supply management method for tilt angle sensor of electric power tower

Abstract

The invention provides an RFID-based power supply management method for a tilt angle sensor of a power tower, and the method comprises the steps: employing a support vector machine and a decision tree, analyzing the historical operation data of the tower, building a tower fault prediction model, predicting the state of the tower through the tower fault prediction model, and obtaining the fault probability and time of the tower; according to a tower fault prediction result, a preventive maintenance plan is made, priority maintenance is arranged for the tower with the fault probability larger than a preset threshold value, and a power supply strategy is adjusted in advance for the tower predicted to have a fault, and load pressure is reduced; a tower group intelligent power supply management system is adopted, state information and a power supply strategy of a single tower are integrated into a whole power grid, the power supply load of each tower is dynamically adjusted through a load balancing algorithm, and for a tower in a dangerous state, the power supply load of the tower is reduced, and the load is transferred to a tower in a good state.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a power supply management method for a power pole tower tilt sensor based on RFID. Background Art

[0002] In the RFID-based power tower tilt sensor power supply management method, there is a key technical problem that needs to be solved. At present, the index and tilt ratio of the tower are considered separately as two independent parameters, which will lead to the failure to timely detect the abnormalities and risks of the tower, resulting in the one-sidedness of the power supply management decision, so as to fail to prevent power outages and safety accidents caused by the collapse of the tower. The tower index reflects the overall condition of the tower, including factors such as material aging and structural integrity; while the tilt ratio directly reflects the degree of tilt of the tower. However, there is a complex mutual influence relationship between the two parameters, and considering each parameter separately may lead to misjudgment of the actual condition of the tower. For example, a tower may have a high index, indicating that the overall condition is good, but at the same time it may have a large tilt ratio. Conversely, a tower with a small tilt ratio may have a low index due to long-term environmental erosion. This complex relationship makes it difficult for traditional power supply management methods to accurately assess the actual condition of the tower, thus affecting the formulation of power supply strategies. Summary of the invention

[0003] The present invention provides a power supply management method for a power pole tower tilt sensor based on RFID, which mainly includes:

[0004] The tilt angle data of the pole tower is collected by the tilt sensor with RFID tag, and the collected tilt angle data is pre-processed by data fusion algorithm to filter out abnormal values ​​and noise, so as to obtain the real-time tilt angle data of the pole tower. According to the preset tilt angle threshold, the tilt angle of the pole tower is graded into three levels: normal, warning and dangerous.

[0005] The historical operation data of the tower is obtained through the tilt sensor with RFID tag, including the age, material and load-bearing capacity of the tower. Combined with the current environmental factors including wind speed, temperature and humidity, a tower index evaluation model is established. The comprehensive index of the tower is calculated based on the tower index evaluation model to reflect the overall status of the tower.

[0006] According to the inclination level and comprehensive index of the tower, a tower status assessment matrix is ​​established. If the inclination level is normal and the comprehensive index is higher than the threshold, the tower status is judged to be good. If the inclination level is warning or the comprehensive index is lower than the threshold, the tower status is judged to be in need of attention. If the inclination level is dangerous or the comprehensive index is far below the threshold, the tower status is judged to be dangerous.

[0007] Develop corresponding power supply management strategies for different tower conditions. For towers in good condition, adopt the conventional power supply mode. For towers that need attention, increase the power supply monitoring frequency and adjust the power supply parameters. For towers in dangerous conditions, start the emergency power supply mode, limit the power supply, and issue a warning signal.

[0008] Support vector machines and decision trees are used to analyze the historical operation data of the towers, and a tower failure prediction model is established. The tower status is predicted through the tower failure prediction model to obtain the probability and time of tower failure.

[0009] According to the tower failure prediction results, a preventive maintenance plan is formulated. For towers with a failure probability greater than the preset threshold, priority maintenance is arranged. For towers predicted to fail, the power supply strategy is adjusted in advance to reduce the load pressure.

[0010] By adopting the intelligent power supply management system for tower groups, the status information and power supply strategy of individual towers are integrated into the overall power grid. Through the load balancing algorithm, the power supply load of each tower is dynamically adjusted. For towers in dangerous conditions, the power supply load of the towers is reduced and the load is transferred to towers in good conditions.

[0011] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0012] The present invention provides an RFID-based power supply management method for power pole tower tilt sensor. By real-time monitoring of the pole tower tilt, the structural changes can be identified in time to prevent large-scale power outages caused by the tilt or collapse of the pole tower. Through the fault prediction model, potential fault points can be identified in advance to carry out preventive maintenance, reduce unplanned power outages, and improve the stability and reliability of the power grid. According to the pole tower status evaluation matrix, maintenance resources can be reasonably arranged to ensure that limited maintenance resources are used first for the pole towers in poor condition, thereby improving maintenance efficiency. The probability and time of failure can be predicted to make the maintenance plan more targeted, avoid excessive or insufficient maintenance, save costs and extend the service life of the pole tower. Differentiated power supply management strategies are adopted for pole towers in different conditions to maintain power supply. The system can improve the continuity and stability of power supply and user satisfaction; limit the power supply on towers in dangerous states to avoid the expansion of the impact range due to faults, and ensure the overall power supply balance of the power grid through load balancing algorithms; provide high-quality basic data for smart grid data analysis by improving the accuracy and comprehensiveness of data collection; realize intelligent management and optimization at the power grid level by integrating the status information and power supply strategy of a single tower, and promote the digital and intelligent transformation of the power system; conduct comprehensive assessments based on environmental factors to improve the system's adaptability and response capabilities to severe weather conditions; through in-depth analysis of historical operation data, the tower index evaluation model established can better understand the performance of towers in different environments and assist risk management decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The present invention is a flow chart of a method for power supply management of a power pole tower tilt sensor based on RFID.

[0014] Figure 2 It is a schematic diagram of a power supply management method for a power pole tower tilt sensor based on RFID of the present invention. DETAILED DESCRIPTION

[0015] The technical scheme of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] like Figure 1-2 In this embodiment, a power supply management method for a power pole tower tilt sensor based on RFID may specifically include:

[0017] S101. Collect the inclination data of the pole tower through the inclination sensor with RFID tag, pre-process the collected inclination data through data fusion algorithm, filter out abnormal values ​​and noise, and obtain the real-time inclination data of the pole tower; classify the inclination of the pole tower according to the preset inclination threshold value, and divide it into three levels: normal, warning and dangerous.

[0018] The inclination data of the pole tower is obtained through an inclination sensor with an RFID tag, and the inclination data is fused through the Kalman filter algorithm to obtain the real-time inclination data of the pole tower; the real-time inclination data is divided according to a preset inclination threshold, and the inclination data is classified using the K-means clustering algorithm to determine the current inclination state of the pole tower; according to the inclination state of the pole tower, an early warning model based on the inclination change rate is established, and the inclination change trend is calculated by the exponential smoothing method to determine whether the pole tower has a continuous inclination risk; according to the inclination state of the pole tower and the inclination change trend, the C4.5 decision tree algorithm is used to generate a pole tower inclination risk assessment result, and the inclination risk is quantitatively assessed from the pole tower structure safety information and line operation factors; if the pole tower inclination risk assessment result exceeds the preset threshold, a warning message is sent to relevant personnel via SMS and email to achieve real-time monitoring and early warning of the pole tower inclination.

[0019] Specifically, the tilt sensor with RFID tag is used to collect the tilt data of the tower. The collected raw tilt data is fused by Kalman filtering algorithm to filter out outliers and random noise, and the real-time tilt data of the tower is obtained. In the process of Kalman filtering, the measurement noise covariance is determined according to the statistical analysis of historical data, and the filtering accuracy is improved by dynamically adjusting the process noise covariance. According to the preset tilt threshold, the real-time tilt data is divided into three levels: normal, warning and dangerous. The K-means clustering algorithm is used to classify the tilt data to determine the current tilt state of each tower. In K-means clustering, the maximum tilt value is selected as the initial cluster center, and the distance between the data point and the cluster center is calculated by Euclidean distance. When the change of the cluster center is less than the preset threshold, the iteration is terminated. For the tilt state of the classified tower, an early warning model based on the tilt change rate is established, and the tilt change trend is calculated by exponential smoothing method to determine whether the tower has a risk of continuous tilt. In the exponential smoothing process, an appropriate smoothing parameter α is selected, where 0<α<1, and the optimal α value is determined according to the characteristics of historical data, and the prediction period is set to 7 days. According to the tilt status and change trend of the tower, the C4.5 decision tree algorithm is used to generate the tower tilt risk assessment results. The tilt risk is quantitatively evaluated from the aspects of tower structure safety and line operation impact to obtain the tower tilt risk level. When constructing the C4.5 decision tree, the tilt change trend is used as an important feature variable, the information gain ratio is used as the feature selection criterion, and the pessimistic pruning method is used to optimize the decision tree. According to the risk assessment results, an automatic alarm mechanism is established. When the tower tilt risk level exceeds the preset threshold, warning information is sent to relevant personnel through SMS, email and other methods to realize the real-time monitoring and warning functions of tower tilt. In the tower tilt monitoring system, an inertial measurement unit composed of a three-axis acceleration sensor and a three-axis gyroscope sensor is used to collect tower tilt data, and the sampling frequency is set to 100Hz. The collected raw tilt data is fused through the Kalman filter algorithm to filter out outliers and random noise. In Kalman filtering, the measurement noise covariance is obtained through variance analysis of 1,000 sets of historical data, and the initial value is set to 0.01; the initial value of the process noise covariance is set to 0.001, and it is dynamically adjusted according to the residual sequence. The accuracy of the real-time inclination data obtained after filtering is improved to 0.01°. According to the pre-set inclination thresholds such as 0-1° for normal, 1-3° for warning, and >3° for danger, the K-means clustering algorithm is used to classify the inclination data. K-means clustering selects the maximum inclination value of 3°, the median value of 1.5°, and the minimum value of 0° as the initial cluster center, and uses the Euclidean distance to calculate the distance between the data point and the cluster center. When the cluster center changes by less than 0.01°, the iteration is terminated, and usually 3-5 iterations can converge. For the classified tower inclination state, the exponential smoothing method is used to calculate the inclination change trend.Select smoothing parameter α=0.3, and predict the inclination change in the next 7 days based on 30 days of historical data. If the predicted inclination change within 7 days exceeds 0.5°, it is determined that there is a risk of continuous inclination. The C4.5 decision tree algorithm is used to generate the tower inclination risk assessment results, and the inclination, inclination change rate, tower height, foundation type, etc. are used as feature variables. The information gain ratio is used as the feature selection criterion, the minimum sample number is set to 10, the maximum tree depth is set to 5, and the pessimistic pruning method is used to optimize the decision tree. According to the risk level output by the decision tree, including low, medium and high, when the risk level is "high", an early warning message is sent to the maintenance personnel via SMS and email, including key information such as tower number, current inclination, predicted inclination, etc., to achieve real-time monitoring and early warning of tower inclination.

[0020] S102. Obtain historical operation data of the pole tower using the inclination sensor with the RFID tag, including the age, material, and load-bearing capacity of the pole tower, and establish a pole tower index evaluation model in combination with current environmental factors including wind speed, temperature, and humidity; calculate a comprehensive index of the pole tower using the pole tower index evaluation model to reflect the overall status of the pole tower.

[0021] The historical operation data of the pole tower and the real-time environmental factor data collected by the meteorological station are obtained from the database, wherein the historical operation data of the pole tower includes the age, material and load-bearing capacity of the pole tower, and the real-time environmental factor data includes wind speed, temperature and humidity; the minimum-maximum normalization processing is performed according to the historical operation data of the pole tower and the real-time environmental factor data to obtain the normalized data, and the normalization processing adopts the formula: X normalization = (XX minimum value) / (X minimum value-X maximum value); the random forest algorithm is used to construct a pole tower index evaluation model, the number of decision trees of the random forest algorithm is a preset value, the random forest algorithm uses the Gini index as a feature selection criterion, the maximum depth of each tree of the random forest algorithm is a preset depth value, and the minimum number of leaf node samples of the random forest algorithm is a preset number of samples; according to the comprehensive index of the pole tower output by the pole tower index evaluation model, the pole tower state is divided into three levels of normal, caution and danger in combination with the preset normal threshold and danger threshold. If the comprehensive index of the pole tower is less than the danger threshold, an early warning signal is triggered, and the pole tower number, comprehensive index value and current environmental data are sent to the monitoring center.

[0022] Specifically, the historical operation data of the tower is obtained from the database, including the age, material, and load-bearing capacity of the tower. At the same time, the current environmental factors such as wind speed, temperature, and humidity data are collected in real time through the meteorological station. The historical operation data and real-time environmental data are normalized to the minimum and maximum values, and the data of different dimensions are unified to the range of 0 to 1. The calculation formula is: X normalization = (XX minimum value) / (X maximum value - X minimum value). The tower index evaluation model is constructed using the random forest algorithm, the number of decision trees is set to 100, the Gini index is used as the feature selection criterion, the maximum depth of each tree is set to 10, and the minimum number of leaf node samples is 5. The normalized data is input into the tower index evaluation model, and the comprehensive index of the tower is obtained through the combined prediction of multiple decision trees. The performance of the tower index evaluation model is evaluated using K-fold cross validation, K=5 is set, the mean square error and determination coefficient are calculated, and if the mean square error is less than 0.1 and the determination coefficient is greater than 0.8, the tower index evaluation model is judged to have good performance. According to the tower comprehensive index output by the random forest algorithm, the tower status is divided into three levels: normal, caution and dangerous, combined with the preset thresholds. The normal threshold is 0.7 and the dangerous threshold is 0.3. These thresholds are obtained through statistical analysis of historical data. For towers of dangerous level, an early warning signal is triggered, and the tower number, comprehensive index value and current environmental data are sent to the monitoring center. Combined with real-time monitoring data, the tower comprehensive index is updated every hour to realize dynamic evaluation and early warning functions. In the tower index evaluation system, the tower operation data is first extracted from the historical database, including information such as the tower age, such as 20 years, material, such as Q345 steel, and load-bearing capacity, such as 50 tons. At the same time, environmental data is collected in real time through meteorological stations installed around the towers, such as wind speed 10m / s, temperature 25℃ and humidity 60%. These data are normalized to the minimum and maximum values, for example, the 20-year tower age is normalized to 0.4, and the 50-ton load-bearing capacity is normalized to 0.8. Subsequently, the tower index evaluation model was constructed using the random forest algorithm, with 100 decision trees set, the maximum depth of each tree being 10, and the minimum number of leaf node samples being 5. The model input includes normalized tower attributes and environmental data, and outputs the tower comprehensive index. The model performance was evaluated by 5-fold cross validation, and the mean square error was calculated to be 0.08 and the determination coefficient was 0.85, proving that the model performance was good. According to the tower comprehensive index output by the model, such as 0.65, the tower status was judged by comparing the preset thresholds, which were normal >0.7 and dangerous <0.3. If a dangerous state was detected, the system automatically sent an early warning message to the monitoring center, including the tower number such as TW-001, the comprehensive index value 0.25 and the current environmental data. The system updates the tower comprehensive index once an hour. For example, when the wind speed suddenly increases to 15m / s, the recalculated comprehensive index may drop to 0.55. Although it has not reached the dangerous level, it has entered the attention state. The system will record this change and continue to monitor.This dynamic assessment method can promptly reflect changes in tower status and provide maintenance personnel with timely and accurate early warning information.

[0023] S103. A pole tower status assessment matrix is ​​established based on the pole tower inclination level and the comprehensive index. If the inclination level is normal and the comprehensive index is higher than the threshold, the pole tower status is judged to be good; if the inclination level is warning or the comprehensive index is lower than the threshold, the pole tower status is judged to be in need of attention; if the inclination level is dangerous or the comprehensive index is far below the threshold, the pole tower status is judged to be dangerous.

[0024] A two-dimensional state assessment matrix is ​​established, wherein the horizontal axis of the two-dimensional state assessment matrix represents the inclination grade, the vertical axis represents the comprehensive index interval, and the corresponding tower state judgment results are filled in the matrix cells; the comprehensive index threshold is set according to the two-dimensional state assessment matrix, the tower comprehensive index data of the most recent year is extracted from the database, the mean and standard deviation of the comprehensive index data are calculated, and the comprehensive index threshold is obtained by subtracting 1.5 times the standard deviation from the mean; the tower state judgment model is constructed using the C4.5 decision tree algorithm, wherein the input variables of the tower state judgment model are the inclination grade and the comprehensive index, and the output variable is the tower state, the maximum tree depth of the decision tree is set to 5, the minimum number of leaf node samples is set to 10, and the information gain ratio is used as the splitting criterion; the tower inclination grade and the comprehensive index monitored in real time are input into the tower state judgment model to obtain the tower current state assessment result, and the tower current state assessment result is divided into three levels: good, need attention and dangerous.

[0025] Specifically, a two-dimensional state assessment matrix is ​​established based on the inclination data and comprehensive index data of the tower. The horizontal axis of the matrix represents the inclination level, which is normal, warning and dangerous. The vertical axis represents the comprehensive index interval, and the corresponding tower state judgment results are filled in the matrix cells. The comprehensive index threshold is set, and the comprehensive index data of the tower in the most recent year is extracted from the database. The mean and standard deviation are calculated, and the mean minus 1.5 times the standard deviation is set as the threshold. If it is lower than the threshold, it is judged as an abnormal state. The C4.5 decision tree algorithm is used to construct the tower state judgment model. The input variables are the inclination level and comprehensive index, and the output variable is the tower state. The maximum tree depth is set to 5, the minimum number of leaf node samples is 10, and the information gain ratio is used as the splitting criterion. The optimal pruning parameters are selected through cross-validation. The real-time monitored tower inclination level and comprehensive index are input into the state judgment model to obtain the current state assessment results of the tower, which are divided into three levels: good, need attention and dangerous. If the state is abnormal, it is detected once an hour; if it is seriously abnormal, it is detected every 10 minutes and the emergency plan is immediately activated. In the tower condition assessment system, a 3x3 two-dimensional condition assessment matrix is ​​first constructed. The horizontal axis represents the inclination level, such as normal is 0-1°, warning is 1-3°, and danger is >3°. The vertical axis represents the comprehensive index range, such as high is >0.8, medium is 0.6-0.8, and low is <0.6. The comprehensive index data of the tower in the most recent year was extracted from the database, and the mean value was 0.75 and the standard deviation was 0.1, and the threshold was set to 0.6. The C4.5 decision tree algorithm was used to construct the condition judgment model, and the maximum tree depth was set to 5, the minimum number of leaf node samples was set to 10, and the information gain ratio was used as the splitting criterion. The optimal pruning parameters were selected through 10-fold cross validation, and the alpha value was set to 0.01. The 10,000 historical data were divided into 8,000 training sets and 2,000 test sets using the holdout method. The model performance was evaluated on the test set, and the accuracy, precision, and recall rates were 95%, 93%, and 94%, respectively. Real-time monitoring data shows that the inclination of a tower is 2.5° and the comprehensive index is 0.72, which is judged as "abnormal" after input into the model. The system automatically adjusts the detection frequency to once every 30 minutes and sends a warning message to the monitoring center, including the tower number, current inclination, comprehensive index and predicted abnormal cause. At the same time, the system activates the preset emergency plan, such as intensive monitoring of the status of surrounding towers and preparing emergency maintenance teams. This dynamic evaluation and multi-level early warning mechanism can detect potential risks in a timely manner and provide strong protection for the safe operation of the power grid.

[0026] S104. Develop corresponding power supply management strategies for different tower conditions. For towers in good condition, adopt the conventional power supply mode; for towers that need attention, increase the power supply monitoring frequency and adjust the power supply parameters; for towers in dangerous conditions, start the emergency power supply mode, limit the power supply, and send out early warning signals.

[0027] The tower status level and power supply parameters are obtained, and a power supply management strategy matrix is ​​established; according to the power supply management strategy matrix, a Mamdani fuzzy control algorithm is used to design a power supply parameter adaptive adjuster, wherein the input variable of the power supply parameter adaptive adjuster is the tower status index, and the output variables are the power supply voltage, current and power factor; according to the output result of the power supply parameter adaptive adjuster, a C4.5 decision tree is constructed as a power supply mode selector, wherein the input variables of the power supply mode selector include the tower status, load demand and power grid operation status, and the output variable is the power supply mode; the output result of the power supply mode selector is linked with the actual power supply equipment through the SCADA system to realize automated power supply management; according to the tower status and power supply mode, a multi-level warning signal release mechanism is designed, wherein the multi-level warning signal release mechanism determines the warning level based on the tower status index, power supply parameter deviation value and load change rate, and pushes the warning information in real time through SMS, email and monitoring platform.

[0028] Specifically, according to the tower status assessment results, a 3x3 power supply management strategy matrix is ​​established. The horizontal axis represents the tower status level of good, attention and danger, and the vertical axis represents the power supply parameters including voltage, current and power factor. The corresponding power supply management strategies are filled in the matrix cells. The Mamdani fuzzy control algorithm is used to design the power supply parameter adaptive regulator. The input variables are the tower status indicators, and the output variables are the power supply voltage, current and power factor. The triangle membership function is designed, 15 fuzzy rules are formulated, and the centroid method is used for defuzzification to achieve dynamic adjustment of power supply parameters. The C4.5 decision tree is constructed as the power supply mode selector. The input variables include tower status, load demand and power grid operation status. The output variable is the power supply mode including conventional, adjustment and emergency. The maximum tree depth is set to 5, the minimum number of leaf node samples is set to 10, the information gain ratio is used as the splitting criterion, and the optimal pruning parameters are selected through cross-validation. The power supply mode selection results are linked with the actual power supply equipment through the SCADA system to realize automated power supply management. A multi-level warning signal release mechanism is designed to automatically generate warning information according to the tower status and power supply mode. The warning level is divided into four levels: green (normal), yellow (slightly abnormal), orange (moderately abnormal), and red (severely abnormal). The warning level is determined based on the tower status index, power supply parameter deviation value, and load change rate. The warning information is pushed in real time through SMS, email, and monitoring platform, including the warning level, impact range, and response measures. In the tower power supply management system, a 3x3 power supply management strategy matrix is ​​first established. The horizontal axis represents the tower status level, such as good (0-0.3), attention (0.3-0.7), and danger (0.7-1). The vertical axis represents the power supply parameters, such as voltage of ±5%, current of ±10%, and power factor of 0.9-1. The Mamdani fuzzy control algorithm is used to design the power supply parameter adjuster. The input variable is the tower status index, and the output variable is the power supply parameter adjustment amount. The tower status index range is 0-1. The triangular membership function is designed to divide the input into three levels: low, medium, and high, and the output is divided into three levels: decrease, maintain, and increase. 15 fuzzy rules are formulated. For example, when the tower status is 0.6, the voltage adjustment is -2%, the current adjustment is -5%, and the power factor is adjusted to 0.95 through fuzzy reasoning. A C4.5 decision tree is constructed as a power supply mode selector. The input variables include tower status, load demand, and grid operation status, and the output variable is the power supply mode. 10,000 historical data are used for training, the maximum tree depth is set to 5, the minimum number of leaf node samples is set to 10, and the optimal pruning parameters are selected through 5-fold cross validation. The trained decision tree has an accuracy of 92% on the test set. The power supply mode selection result is sent through the SCADA system to send control instructions. For example, when switching to emergency power supply mode, the power supply power is automatically reduced to 70% of the rated value.The early warning signal release mechanism generates a red early warning signal based on a comprehensive judgment of the tower status index of 0.8, the power supply voltage deviation of -4% and the load change rate of +15% / hour, and pushes it to the operation and maintenance personnel via SMS and email, and displays the affected line map and recommended response measures on the monitoring platform.

[0029] S105. Use support vector machines and decision trees to analyze historical operation data of the pole tower and establish a pole tower failure prediction model; predict the state of the pole tower through the pole tower failure prediction model to obtain the probability and time of pole tower failure.

[0030] The historical operation data of the pole tower is obtained from the database, and the historical operation data includes the inclination angle of the pole tower, vibration frequency, stress distribution, and environmental factors; data denoising is performed through five-layer wavelet transform, and the db4 wavelet basis is selected to obtain the denoised pole tower operation data; principal component analysis is performed based on the denoised pole tower operation data, and the feature vectors whose cumulative contribution rate reaches the preset threshold are selected to determine the key feature vectors of the pole tower state; a support vector machine algorithm is used to build a pole tower fault prediction model, and a radial basis kernel function is selected. The kernel function parameters and penalty factors are optimized by the grid search method, and the optimal model parameters are selected by the cross-validation method; the CART decision tree algorithm is used to perform importance analysis on the key feature vectors, and the maximum tree depth and the minimum number of leaf node samples are set. The optimal split feature is selected by the Gini index to obtain the feature importance ranking; the historical operation data is segmented by the sliding time window method, the window size and step size are set, a time series prediction model is constructed by a multi-layer LSTM network, an activation function is selected, the batch size and the initial value of the learning rate are set, and an optimizer is used to predict the future state of the pole tower to obtain the time series prediction results of various indicators of the pole tower.

[0031] Specifically, the historical operation data of the tower is obtained from the database, including the tower inclination, vibration frequency, stress distribution, and environmental factors. The data is denoised by five-layer wavelet transform, and the db4 wavelet basis is selected. The principal component analysis is performed on the denoised data, and the eigenvectors with a cumulative contribution rate of 95% are selected to obtain the key eigenvectors of the tower state. The support vector machine algorithm is used to build a tower fault prediction model, and the radial basis kernel function is selected. The kernel function parameter γ is optimized by the grid search method, ranging from 10^-3 to 10^3, with a step size of 10, and the penalty factor C ranges from 0.1 to 1000, with a step size of 10. The 5-fold cross validation method is used to select the optimal model parameters. The CART decision tree algorithm is used to analyze the importance of the features, setting the maximum tree depth to 10 and the minimum number of leaf node samples to 5. The optimal split feature is selected by the Gini index to obtain the feature importance ranking. The sliding time window method is used to segment the historical data. The window size is set to 30 days and the step size is 1 day. The time series prediction model is constructed. A three-layer LSTM network is used. Each layer contains 64 neurons. The activation function selects ReLU. The batch size is set to 32. The initial value of the learning rate is 0.001. The Adam optimizer is used to predict the future state of the tower and obtain the time series prediction results of various indicators of the tower. Combining the output of the support vector machine and the LSTM model, the probability distribution of the tower failure is calculated by Bayesian reasoning. The prior probability is set to the historical failure rate. The posterior probability is updated by the likelihood function. When the predicted failure probability exceeds 0.8, the warning mechanism is triggered, and the time interval when the failure may occur is estimated based on the time series prediction results. In the tower failure prediction system, the tower operation data of the past five years is first extracted from the database, including the hourly recorded inclination angle 0-5°, vibration frequency 0-100Hz, stress distribution 0-500MPa and environmental factors temperature -40-60℃, humidity 0-100%, and wind speed 0-50m / s. The data were subjected to five-layer wavelet transform denoising, and the db4 wavelet basis was selected. The signal-to-noise ratio after denoising was improved by about 20%. Through principal component analysis, the first five principal components were selected as key feature vectors, and the cumulative contribution rate reached 96.8%. The fault prediction model was constructed using support vector machine. After grid search optimization, the kernel function parameter γ = 0.1 and the penalty factor C = 100 were finally determined, and the accuracy of 5-fold cross validation reached 94.3%. Subsequently, the CART decision tree was used for feature importance analysis, with the tree depth set to 8 and the minimum number of leaf node samples set to 10. The results showed that the inclination change rate, vibration frequency and wind speed were the three most important features affecting tower failures, with importance scores of 0.35, 0.28 and 0.20 respectively. The sliding window method with a window size of 30 days and a step size of 1 day was used to process the data. The three-layer LSTM network was input, with 64 neurons in each layer, ReLU activation function, batch size of 32, initial learning rate of 0.001, and 100 rounds of training using the Adam optimizer. The root mean square error on the test set was 0.15.Finally, the failure probability is calculated by combining the output of the support vector machine and LSTM model through Bayesian reasoning, and the prior probability is set to the historical annual average failure rate of 0.05. When the posterior failure probability of a tower exceeds 0.8, an early warning is triggered. For example, for the tower numbered TW-001, the system predicts that the probability of failure in the next 7-10 days is 0.85, and immediately sends an early warning message to the maintenance personnel, including the tower location, predicted failure time interval, and recommended maintenance items.

[0032] S106. Develop a preventive maintenance plan based on the tower failure prediction results. Priority maintenance is arranged for towers with a failure probability greater than a preset threshold. For towers predicted to fail, adjust the power supply strategy in advance to reduce load pressure.

[0033] A preventive maintenance priority matrix is ​​constructed according to the fault probability interval and the predicted fault occurrence time, wherein the horizontal axis of the matrix represents the fault probability interval and the vertical axis represents the predicted fault occurrence time; a fuzzy comprehensive evaluation method is used to quantitatively evaluate the maintenance priority of the tower, wherein the evaluation includes establishing an evaluation index system, determining the weight of each index through the hierarchical analysis method, constructing a fuzzy relationship matrix, and performing fuzzy synthesis operation using the weighted average method; an optimal preventive maintenance plan is formulated using a 0-1 knapsack dynamic programming algorithm, wherein the algorithm uses the maintenance priority score as the value and the maintenance time as the weight to solve the optimal maintenance combination under the constraints of human resources and time window; according to the preventive maintenance plan, an adaptive load control algorithm is used to dynamically adjust the power supply strategy for towers of different priorities, wherein the algorithm uses a fuzzy PID controller to automatically calculate the load adjustment amount by real-time monitoring of the tower load status and combining weather forecast data; if a tower is predicted to fail within a preset time range, the load reduction amount of the predicted value is additionally added to the tower, and the monitoring frequency is increased.

[0034] Specifically, according to the failure probability and predicted occurrence time output by the tower failure prediction model, a preventive maintenance priority matrix is ​​constructed. The horizontal axis represents the failure probability interval, which includes three intervals: 0-0.3, 0.3-0.6, and 0.6-1. The vertical axis represents the predicted failure occurrence time, which includes three time ranges: 0-7 days, 7-30 days, and more than 30 days. The corresponding maintenance priority scores such as 1-9 points are filled in the matrix cells. The fuzzy comprehensive evaluation method is used to quantitatively evaluate the tower maintenance priority, and an evaluation index system is established, including failure probability, predicted failure time, tower importance, and maintenance cost. The weights of each index are determined by the hierarchical analysis method, and a fuzzy relationship matrix is ​​constructed. The weighted average method is used for fuzzy synthesis operations to calculate the comprehensive score of the tower maintenance priority. The 0-1 knapsack dynamic programming algorithm is used to formulate the optimal preventive maintenance plan, with the maintenance priority score as the value and the maintenance time as the weight, to solve the optimal maintenance combination that meets the constraints of human resources and time windows. The state transition equation is f[i][j]=max(f[i-1][j],f[i-1][jw[i]]+v[i]), where i represents the tower number, j represents the available time, w[i] represents the time required to maintain tower i, and v[i] represents the maintenance priority score of tower i. According to the preventive maintenance plan, the adaptive load control algorithm is used to dynamically adjust the power supply strategy for towers of different priorities. By real-time monitoring of the tower load status and combining the weather forecast data for the next 24 hours, the fuzzy PID controller is used to automatically calculate the load adjustment amount. For towers predicted to fail within 7 days, an additional 20% load reduction is added, and the monitoring frequency is increased to once an hour. In the tower preventive maintenance system, a 3x3 maintenance priority matrix is ​​first constructed based on the output of the fault prediction model, such as a tower with a failure probability of 0.7 and a score of 9 for a tower predicted to fail within 7 days. Then, the fuzzy comprehensive evaluation method is used to quantify the priority, setting the failure probability weight to 0.4, the prediction time weight to 0.3, the tower importance weight to 0.2, and the maintenance cost weight to 0.1, and the comprehensive score is obtained through fuzzy synthesis. For example, the scores of each indicator of a tower are [0.8, 0.9, 0.7, 0.6], and the comprehensive score is calculated to be 0.79. Then, the 0-1 knapsack dynamic programming algorithm is used to formulate a maintenance plan. Assuming that there are 10 towers to be maintained, the total available time is 100 hours, and the maintenance time of each tower ranges from 8 to 15 hours. The optimal maintenance combination is solved by the state transfer equation to obtain the numbers and order of the 6 towers that can be maintained within the time limit. Finally, adaptive load control is implemented for towers of different priorities. For example, for a tower with a comprehensive score of 0.79, based on the current 70% load rate and the high temperature forecast of 35℃ in the next 24 hours, the fuzzy PID controller calculates that the load needs to be reduced by 15%.Taking into account that the tower is predicted to fail within 7 days, an additional 20% load reduction is added, and ultimately a 35% load reduction is implemented. The monitoring frequency is adjusted to once an hour to continuously track changes in the tower status.

[0035] S107. Use the intelligent power supply management system for tower groups to integrate the status information and power supply strategy of individual towers into the overall power grid. Through the load balancing algorithm, dynamically adjust the power supply load of each tower. For towers in dangerous conditions, reduce the power supply load of the towers and transfer the load to towers in good conditions.

[0036] The real-time status information of the tower is collected, and the real-time status information includes inclination, vibration, stress distribution and environmental factors; based on the real-time status information, multi-source data fusion is performed through the Dempster-Shafer evidence theory module to obtain the health index of the tower; based on the health index of the tower and the current load condition, a fuzzy control algorithm is used to formulate a power supply strategy for the tower, which includes setting a triangular membership function of the health index and the load rate, formulating a fuzzy rule base, using the Mamdani reasoning mechanism, and defuzzifying the power supply power adjustment amount through the centroid method; a power grid topology model based on graph theory is constructed, in which the tower is used as a node and the transmission line is used as a node. The minimum spanning tree is realized by using the Kruskal algorithm, and the edge weight of the minimum spanning tree is defined as the weighted sum of the line loss and the tower health index; according to the path obtained by the minimum spanning tree, the load transfer of the dangerous tower is realized, which includes: calculating the maximum transmission capacity of each path, distributing the load to be transferred in proportion, and adjusting the load distribution step by step through the intelligent switch; using the genetic algorithm to optimize the load balancing strategy of the entire power grid, the genetic algorithm includes chromosome encoding using real number encoding to represent the load proportion of each tower, generating a fitness function as the inverse of the load variance between towers, selecting roulette, selecting operations, using arithmetic crossover and uniform mutation, and solving to obtain the global optimal load distribution plan.

[0037] Specifically, the real-time status information of each tower is collected, including inclination, vibration, stress distribution and environmental factors. Multi-source data fusion is performed through the Dempster-Shafer evidence theory module, basic probability distribution is calculated, and a unified tower health index is obtained by combining evidence synthesis rules, and a tower status information database is established. According to the tower health index and current load conditions, the fuzzy control algorithm is used to formulate the power supply strategy of a single tower, the triangle membership function of the health index and load rate is set, the fuzzy rule base is formulated, and the Mamdani reasoning mechanism is used to defuzzify the appropriate power supply adjustment through the center of gravity method. A power grid topology model based on graph theory is constructed, with towers as nodes and transmission lines as edges. The minimum spanning tree is realized using the Kruskal algorithm. The edge weight is defined as the weighted sum of line loss and tower health index. The optimal load distribution scheme between different towers is determined through depth-first search. According to the path obtained by the minimum spanning tree, the load transfer of towers with poor status is realized, the maximum transmission capacity of each path is calculated, the load to be transferred is proportionally distributed, and the load distribution is adjusted step by step through intelligent switches. Genetic algorithm is used to optimize the load balancing strategy of the entire power grid. The chromosome encoding uses real number encoding to represent the load ratio of each tower. The fitness function is designed to be the inverse of the load variance between towers. Roulette is selected, the selection operation is selected, arithmetic crossover and uniform mutation are used, the population size is set to 100, and 500 generations are iterated to obtain the global optimal load distribution solution. In the intelligent power supply management system of the tower group, the real-time status information of 100 towers is first collected, including inclination 0-5°, vibration frequency 0-100Hz, stress distribution 0-500MPa and environmental factors including temperature -40-60℃, humidity 0-100%, and wind speed 0-50m / s. The Dempster-Shafer evidence theory module is used to fuse these data, and basic probabilities are assigned to each indicator, such as inclination 0.3, vibration 0.2, stress 0.3, and environment 0.2. The tower health index is obtained through orthogonal and operation. For example, the health index of a tower after fusion is 0.72. Subsequently, the fuzzy control algorithm was applied to formulate the power supply strategy, and the triangular membership function of the health index and load rate was set, such as the health index "low" is 0-0.4, "medium" is 0.3-0.7, and "high" is 0.6-1, and the load rate is "low" is 0-50%, "medium" is 40%-80%, and "high" is 70%-100%. Through Mamdani reasoning, it is concluded that the power supply power of the tower should be reduced by 5%. Then, the power grid topology diagram is constructed, which contains 100 nodes and 150 edges. The minimum spanning tree is generated using the Kruskal algorithm, and the edge weight formula is 0.6×line loss+0.4×(1-tower health index). For the tower with a health index of 0.72, three paths for transferring loads are calculated, with maximum transmission capacities of 10MW, 8MW, and 5MW respectively. The 4MW load to be transferred is allocated in a ratio of 6:5:3, and adjusted step by step through intelligent switches.Finally, the genetic algorithm was run to optimize load balancing, with a chromosome length of 100, representing the load ratio of each tower, a population size of 100, 500 iterations, a crossover probability of 0.8, and a mutation probability of 0.1. The final solution reduced the health index of the original 80% load tower to 65%, and increased the load of the surrounding towers with higher health indexes from 60% to 70%, achieving load balancing for the entire power grid.

[0038] The above embodiment is only one of the preferred implementation modes of the present invention and should not be used to limit the protection scope of the present invention. Any changes or modifications that are made to the main design concept and spirit of the present invention and have no substantive significance, and the technical problems they solve are still consistent with the present invention, should be included in the protection scope of the present invention.

Claims

1. A power supply management method for a power pole tower tilt sensor based on RFID, characterized in that: The method comprises: The tilt angle data of the pole tower is collected by the tilt sensor with RFID tag, and the collected tilt angle data is pre-processed by data fusion algorithm to filter out abnormal values ​​and noise, so as to obtain the real-time tilt angle data of the pole tower. According to the preset tilt angle threshold, the tilt angle of the pole tower is graded into three levels: normal, warning and dangerous. The historical operation data of the tower is obtained through the tilt sensor with RFID tag, including the age, material and load-bearing capacity of the tower. Combined with the current environmental factors including wind speed, temperature and humidity, a tower index evaluation model is established. The comprehensive index of the tower is calculated based on the tower index evaluation model to reflect the overall status of the tower. According to the inclination level and comprehensive index of the tower, a tower status assessment matrix is ​​established. If the inclination level is normal and the comprehensive index is higher than the threshold, the tower status is judged to be good. If the inclination level is warning or the comprehensive index is lower than the threshold, the tower status is judged to be in need of attention. If the inclination level is dangerous or the comprehensive index is far below the threshold, the tower status is judged to be dangerous. Develop corresponding power supply management strategies for different tower conditions. For towers in good condition, adopt the conventional power supply mode. For towers that need attention, increase the power supply monitoring frequency and adjust the power supply parameters. For towers in dangerous conditions, start the emergency power supply mode, limit the power supply, and issue a warning signal. Support vector machines and decision trees are used to analyze the historical operation data of the towers, and a tower failure prediction model is established. The tower status is predicted through the tower failure prediction model to obtain the probability and time of tower failure. According to the tower failure prediction results, a preventive maintenance plan is formulated. For towers with a failure probability greater than the preset threshold, priority maintenance is arranged. For towers predicted to fail, the power supply strategy is adjusted in advance to reduce the load pressure. By adopting the intelligent power supply management system for tower groups, the status information and power supply strategy of individual towers are integrated into the overall power grid. Through the load balancing algorithm, the power supply load of each tower is dynamically adjusted. For towers in dangerous conditions, the power supply load of the towers is reduced and the load is transferred to towers in good conditions.

2. The method according to claim 1, wherein: The tilt angle data of the pole tower is collected by the tilt sensor with RFID tag, and the collected tilt angle data is pre-processed by data fusion algorithm to filter out abnormal values ​​and noise, so as to obtain the real-time tilt angle data of the pole tower, and the tilt angle of the pole tower is graded according to the preset tilt angle threshold, which is divided into three levels: normal, warning and dangerous, including: Obtaining the inclination data of the tower through an inclination sensor with an RFID tag, and fusing the inclination data through a Kalman filter algorithm to obtain real-time inclination data of the tower; The real-time inclination angle data is divided according to a preset inclination angle threshold, and the inclination angle data is classified by using a K-means clustering algorithm to determine the current inclination state of the tower; According to the tilt state of the tower, an early warning model based on the tilt angle change rate is established, and the tilt angle change trend is calculated by exponential smoothing method to determine whether there is a risk of continuous tilt of the tower; According to the inclination state of the tower and the inclination angle change trend, a C4.5 decision tree algorithm is used to generate a tower inclination risk assessment result, and a quantitative assessment of the inclination risk is performed based on the tower structure safety information and line operation factors; If the tower tilt risk assessment result exceeds the preset threshold, a warning message will be sent to relevant personnel via SMS and email to achieve real-time monitoring and warning of tower tilt.

3. The method according to claim 1, wherein: The historical operation data of the tower is obtained according to the tilt sensor with RFID tag, including the age, material and load-bearing capacity of the tower, and the tower index evaluation model is established in combination with the current environmental factors including wind speed, temperature and humidity. The comprehensive index of the tower is calculated according to the tower index evaluation model to reflect the overall status of the tower, including: Obtaining historical operation data of the tower in the database and real-time environmental factor data collected by the weather station, wherein the historical operation data of the tower includes the age, material and load-bearing capacity of the tower, and the real-time environmental factor data includes wind speed, temperature and humidity; Performing minimum-maximum normalization processing on the tower historical operation data and the real-time environmental factor data to obtain normalized data, wherein the normalization processing adopts the formula: X normalization = (XX minimum value) / (X minimum value-X maximum value); A tower index evaluation model is constructed using a random forest algorithm, the number of decision trees of the random forest algorithm is a preset value, the random forest algorithm uses the Gini index as a feature selection criterion, the maximum depth of each tree of the random forest algorithm is a preset depth value, and the minimum number of leaf node samples of the random forest algorithm is a preset number of samples; According to the tower comprehensive index output by the tower index evaluation model, the tower status is divided into three levels: normal, warning and dangerous, combined with the preset normal threshold and danger threshold. If the tower comprehensive index is less than the danger threshold, an early warning signal is triggered, and the tower number, comprehensive index value and current environmental data are sent to the monitoring center.

4. The method according to claim 1, wherein: The tower state evaluation matrix is ​​established according to the tower inclination level and the comprehensive index. If the inclination level is normal and the comprehensive index is higher than the threshold, the tower state is judged to be good. If the inclination level is warning or the comprehensive index is lower than the threshold, the tower state is judged to be in need of attention. If the inclination level is dangerous or the comprehensive index is far below the threshold, the tower state is judged to be dangerous, including: Establishing a two-dimensional state assessment matrix, wherein the horizontal axis of the two-dimensional state assessment matrix represents the inclination grade, the vertical axis represents the comprehensive index interval, and the corresponding tower state determination results are filled in the matrix cells; A comprehensive index threshold is set according to the two-dimensional condition assessment matrix, comprehensive index data of the tower in the most recent year is extracted from the database, the mean and standard deviation of the comprehensive index data are calculated, and the comprehensive index threshold is obtained by subtracting 1.5 times the standard deviation from the mean; The C4.5 decision tree algorithm is used to construct a tower state determination model, wherein the input variables of the tower state determination model are the inclination grade and the comprehensive index, the output variable is the tower state, the maximum tree depth of the decision tree is set to 5, the minimum number of leaf node samples is set to 10, and the information gain ratio is used as the splitting criterion; The real-time monitored pole tower inclination level and comprehensive index are input into the pole tower state determination model to obtain the pole tower current state assessment result, which is divided into three levels: good, needing attention and dangerous.

5. The method according to claim 1, wherein: The corresponding power supply management strategy is formulated for different tower states. For towers in good condition, the conventional power supply mode is adopted. For towers that need attention, the power supply monitoring frequency is increased and the power supply parameters are adjusted. For towers in dangerous conditions, the emergency power supply mode is activated, the power supply power is limited, and an early warning signal is issued, including: Obtain tower status level and power supply parameters, and establish a power supply management strategy matrix; According to the power supply management strategy matrix, a power supply parameter adaptive regulator is designed by using the Mamdani fuzzy control algorithm, wherein the input variable of the power supply parameter adaptive regulator is the tower state index, and the output variable is the power supply voltage, current and power factor; According to the output result of the power supply parameter adaptive regulator, a C4.5 decision tree is constructed as a power supply mode selector, wherein the input variables of the power supply mode selector include tower status, load demand and power grid operation status, and the output variable is the power supply mode; The output result of the power supply mode selector is linked with the actual power supply equipment through the SCADA system to realize automatic power supply management; According to the tower status and power supply mode, a multi-level warning signal issuance mechanism is designed. The multi-level warning signal issuance mechanism determines the warning level based on the tower status index, power supply parameter deviation value and load change rate, and pushes the warning information in real time via SMS, email and monitoring platform.

6. The method according to claim 1, wherein: The support vector machine and decision tree are used to analyze the historical operation data of the tower, establish a tower failure prediction model, predict the state of the tower through the tower failure prediction model, and obtain the probability and time of tower failure, including: Acquire historical operation data of the tower from a database, wherein the historical operation data includes the tower inclination, vibration frequency, stress distribution and environmental factors; The data was denoised by five-layer wavelet transform, and the db4 wavelet basis was selected to obtain the tower operation data after denoising. Performing principal component analysis on the tower operation data after noise reduction, selecting feature vectors whose cumulative contribution rate reaches a preset threshold, and determining key feature vectors of the tower state; The support vector machine algorithm is used to build a tower fault prediction model, the radial basis kernel function is selected, the kernel function parameters and penalty factors are optimized by grid search method, and the optimal model parameters are selected by cross validation method. The CART decision tree algorithm is used to perform importance analysis on the key feature vectors, the maximum tree depth and the minimum number of leaf node samples are set, the optimal split feature is selected through the Gini index, and the feature importance ranking is obtained; The sliding time window method is used to segment the historical operation data, the window size and step size are set, a multi-layer LSTM network is used to build a time series prediction model, the activation function is selected, the batch size and the initial value of the learning rate are set, and the optimizer is used to predict the future state of the tower to obtain the time series prediction results of various indicators of the tower.

7. The method according to claim 1, wherein: The method comprises: formulating a preventive maintenance plan based on the tower failure prediction results, arranging priority maintenance for towers with a failure probability greater than a preset threshold, and adjusting the power supply strategy in advance for towers predicted to fail to reduce load pressure, including: A preventive maintenance priority matrix is ​​constructed according to the failure probability interval and the predicted failure occurrence time, wherein the horizontal axis of the matrix represents the failure probability interval and the vertical axis represents the predicted failure occurrence time; A fuzzy comprehensive evaluation method is used to quantitatively evaluate the maintenance priority of the tower, which includes establishing an evaluation index system, determining the weight of each index through a hierarchical analysis method, constructing a fuzzy relationship matrix, and using a weighted average method to perform fuzzy synthesis operations; The optimal preventive maintenance plan is formulated using a 0-1 knapsack dynamic programming algorithm, which uses the maintenance priority score as the value and the maintenance time as the weight to solve the optimal maintenance combination that meets the constraints of human resources and time window; According to the preventive maintenance plan, for towers of different priorities, an adaptive load control algorithm is used to dynamically adjust the power supply strategy. The algorithm uses a fuzzy PID controller to automatically calculate the load adjustment amount by real-time monitoring of the tower load status and combining weather forecast data; If a tower is predicted to fail within a preset time range, the load reduction of the predicted value will be additionally increased for the tower and the monitoring frequency will be increased.

8. The method according to claim 1, wherein: The tower group intelligent power supply management system is adopted to integrate the status information and power supply strategy of a single tower into the overall power grid, and dynamically adjust the power supply load of each tower through the load balancing algorithm. For towers in dangerous conditions, the power supply load of the tower is reduced and the load is transferred to towers in good conditions, including: Collecting real-time status information of the tower, wherein the real-time status information includes inclination, vibration, stress distribution and environmental factors; According to the real-time status information, multi-source data fusion is performed through the Dempster-Shafer evidence theory module to obtain the health index of the tower; According to the health index of the tower and the current load condition, a fuzzy control algorithm is used to formulate a power supply strategy for the tower, which includes setting a triangular membership function of the health index and the load rate, formulating a fuzzy rule base, using a Mamdani reasoning mechanism, and defuzzifying the power supply adjustment amount by the centroid method; A power grid topology model based on graph theory is constructed, in which towers are used as nodes and transmission lines are used as edges, and a minimum spanning tree is implemented using a Kruskal algorithm, wherein the edge weight of the minimum spanning tree is defined as a weighted sum of line loss and tower health index; According to the path obtained by the minimum spanning tree, load transfer of the dangerous tower is realized, which includes: calculating the maximum transmission capacity of each path, distributing the load to be transferred in proportion, and adjusting the load distribution step by step through the intelligent switch; A genetic algorithm is used to optimize the load balancing strategy of the entire power grid. The genetic algorithm includes chromosome encoding using real number encoding to represent the load proportion of each tower, generating a fitness function as the inverse of the load variance between towers, selecting roulette, selecting operations, using arithmetic crossover and uniform mutation, and solving to obtain a global optimal load distribution solution.

Images