Intelligent control method and system for electric power ice melting

Abstract

The invention relates to the technical field of electric power ice melting, in particular to an electric power ice melting intelligent control method and system.The method comprises the steps that when it is detected that the current external environment meets the grounding knife switch opening condition according to meteorological parameters of a target line of an electric power ice melting system, a grounding opening instruction is generated; based on the grounding opening instruction and the line equipment state parameters, determining an optimal opening time point of the phase selection disconnecting link; an opening time sequence rule is generated according to the optimal opening time point, and a phase selection disconnecting link is controlled to execute opening operation according to the opening time sequence rule; determining a priority coefficient of the short circuit closing instruction; in the process of controlling the short-circuit disconnecting link to execute the closing operation according to the priority coefficient, if the ground wire and the wire are successfully connected in series, verifying the actual ice melting operation sequence to obtain an ice melting operation verification result; and based on the verification result of the ice melting operation, generating an ice melting operation optimization sequence by using a decision tree pruning algorithm. According to the invention, intelligent and accurate control of the electric power ice melting process is realized, and the influence of current oscillation on the system is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power de-icing, and particularly to an intelligent control method and system for power de-icing. Background Art

[0002] During the actual operation of a power de-icing system, the accuracy and timeliness of de-icing operations are crucial for ensuring the safe and stable operation of the power system. However, existing power de-icing systems face a series of technical challenges when performing de-icing operations. The execution order and timing requirements of these operations are extremely strict. First, the earthing switch needs to be opened to achieve insulation transformation, then the corresponding phase-selection switch is selectively opened according to the de-icing requirements, and finally the short-circuit switch is closed to complete the series connection of the ground wire and the conductor. The execution order and timing requirements of this series of operations are extremely strict, and any mistake or delay in any link may have an adverse impact on the de-icing effect, and even lead to equipment damage or operation failure.

[0003] In actual operation, system response delay and operation timeout have become important factors restricting the efficiency and safety of de-icing operations. Due to uneven system resource allocation or network communication delay, key instructions may fail to be executed in a timely manner. For example, after the earthing switch is opened, the opening instruction of the phase-selection switch may be delayed due to insufficient system response priority, resulting in the ground wire not being selected when the short-circuit switch is closed, thus causing operation errors and affecting the de-icing effect. In addition, existing systems lack real-time monitoring of the execution time of key operations and cannot automatically trigger an alarm and pause the subsequent process after the operation times out. The lack of this monitoring mechanism increases the operation risk, making it difficult for the system to react in a timely manner when facing abnormal situations and further exacerbating the uncertainty of the operation.

[0004] In summary, there are many problems in existing power de-icing systems when performing de-icing operations. These problems not only affect the efficiency and safety of de-icing operations but also pose a potential threat to the stable operation of the power system. Therefore, there is an urgent need to develop a more intelligent and efficient de-icing operation method for power de-icing systems to overcome the deficiencies of the existing technology. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides an intelligent control method and system for power de-icing.

[0006] In a first aspect, the present invention provides an intelligent control method for power de-icing, and the method includes the following steps: Collect meteorological parameters of the target line of the power de-icing system in real time, and generate an earthing opening instruction when it is detected according to the meteorological parameters that the current external environment meets the preset earthing switch opening condition; Control the earthing switch to perform the opening operation according to the earthing opening instruction, and determine the optimal opening time point of the phase selection switch based on the earthing switch opening completion status and the line equipment status parameters collected in real time; According to the optimal opening time point, use the ant colony optimization algorithm to dynamically generate the opening time sequence rule, and control the phase selection switch to perform the opening operation according to the opening time sequence rule; Obtain the actual execution delay data of the phase selection switch opening operation, and determine the priority coefficient of the short-circuit closing instruction according to the actual execution delay data; During the process of controlling the short-circuit switch to perform the closing operation according to the priority coefficient, if it is monitored that the ground wire is successfully connected in series with the conductor, the actual ice melting operation sequence is verified to obtain the ice melting operation verification result; Based on the ice melting operation verification result, use the decision tree pruning algorithm to generate the optimized ice melting operation sequence.

[0007] In a further embodiment, the step of generating the earthing opening instruction when it is detected according to the meteorological parameters that the current external environment meets the preset earthing switch opening condition includes: Perform a moving average filtering process on the meteorological parameters to obtain the preprocessed meteorological parameters, and obtain the current environmental temperature data, the conductor ice coating thickness data, and the wind speed data according to the preprocessed meteorological parameters; According to the conductor ice coating thickness data, use the sliding window method to calculate the ice coating thickness growth rate at adjacent time points, and calculate the wind speed fluctuation variance feature according to the current wind speed data and the historical average wind speed; According to the difference between the current wind speed data and the historical average wind speed and the reference temperature threshold, obtain the dynamic temperature threshold by linear interpolation, and compare the current environmental temperature data with the dynamic temperature threshold to obtain the temperature comparison result; If the temperature comparison result, the ice coating thickness growth rate, and the wind speed fluctuation variance feature all meet the preset earthing switch opening condition, trigger the priority scheduling algorithm to place the earthing opening instruction that meets the earthing switch opening condition at the head of the resource allocation queue, and generate the earthing opening instruction.

[0008] In a further embodiment, the step of determining the optimal opening time point of the phase selection switch based on the earthing switch opening completion status and the line equipment status parameters collected in real time includes: When the earthing switch opening completion status is that the earthing switch has been successfully opened, collect the line equipment status parameters in real time, and perform spectrum analysis on the line equipment status parameters to obtain the line harmonic distortion rate and the equipment vibration spectrum feature; Define state variables and decision variables; among them, the state variables include the current time, the opening state of the earthing switch, the line harmonic distortion rate, and the equipment vibration spectrum characteristics, and the decision variable is the opening time point of the phase selection switch. Taking minimizing the current oscillation amplitude as the optimization objective, based on the state variables and the decision variables, establish a state transition equation through the physical characteristics and dynamic behaviors of the power system when opening at different time points. Based on the state transition equation, use the Bellman dynamic programming algorithm to iteratively calculate the current oscillation amplitude at all opening time points of the phase selection switch. In each iteration, update the state variables according to the state transition equation, and select the opening time point with the minimum current oscillation amplitude as the optimal strategy for the current step. When the change in the current oscillation amplitude is less than the preset current oscillation amplitude threshold, output the optimal opening time point of the phase selection switch.

[0009] In a further implementation, the step of dynamically generating an opening time sequence rule using the ant colony optimization algorithm according to the optimal opening time point includes: According to the line length and ice coverage distribution characteristics of the target line of the power de-icing system, segment the target line of the power de-icing system to obtain several line segments. Equivalent each of the line segments to a node, and construct a node topology graph according to the connection relationship between the nodes. Randomly initialize several ants as search agents, and assign an initial pheromone concentration to each directed edge in the node topology graph at the initial stage as the initial clue to guide the ants to search. Control each ant to start from the starting node, and after each round of iteration, evaluate the path generated by each ant to obtain the reduction value of the current oscillation amplitude of the opening path. Update the pheromone concentration on the opening path according to the reduction value of the current oscillation amplitude of the opening path to obtain the updated value of the pheromone concentration. Calculate the transfer probability of the ant transferring from the current node to the next node according to the updated value of the pheromone concentration and the node weight, and select the next node according to the transfer probability until all nodes are traversed to form a complete opening time sequence path. Repeat the above process, and after the iterative calculation ends, output the opening time sequence rule corresponding to the optimal opening time sequence path; the opening time sequence rule includes the node access order and the opening time interval.

[0010] In a further implementation, the adjacent nodes of the node topology graph are connected by directed edges, and the directed edges represent the timing dependence relationship between the opening operations of the phase selection switch.

[0011] In a further embodiment, the step of determining the priority coefficient of the short connection switch closing instruction according to the actual execution delay data includes: Obtain the oscillation amplitude of the phase selection operation current for the phase selection switch disconnection operation, and perform correlation analysis on the actual execution delay data and the oscillation amplitude of the phase selection operation current to establish a delay-oscillation amplitude mapping relationship table; Use the minimum delay value in the delay-oscillation amplitude mapping relationship table as the dynamic expected delay, calculate the deviation rate between the actual execution delay data and the dynamic expected delay, and obtain the delay deviation rate; Calculate the ratio between the oscillation amplitude of the phase selection operation current and the reference current amplitude to obtain the current volatility; Use the delay deviation rate and the current volatility as the fuzzy control input quantities, perform fuzzy processing on the fuzzy control input quantities to obtain several fuzzy sets; According to the delay-oscillation amplitude mapping relationship table, perform fuzzy reasoning on the fuzzy sets to obtain the priority coefficient of the short connection switch closing instruction.

[0012] In a further embodiment, the step of, if it is monitored that the ground wire is successfully connected in series with the conductor string, verifying the actual de-icing operation sequence to obtain the de-icing operation verification result includes: During the closing operation of the short connection switch, real-time collect the contact resistance value between the ground wire and the conductor. If the contact resistance value is lower than the preset contact resistance threshold, it is determined that the ground wire is successfully connected in series with the conductor; After the ground wire is successfully connected in series with the conductor, obtain the current oscillation characteristics during the execution process of the actual de-icing operation sequence, and extract the actual execution timestamp of the actual de-icing operation sequence; Calculate the relative deviation rate of the actual execution timestamp according to the expected time window of the preset standard operation sequence to obtain the timing deviation rate; If the timing deviation is within the timing tolerance threshold range and the current oscillation characteristics are within the oscillation characteristic threshold range, it is determined that the actual de-icing operation sequence passes the verification; otherwise, it is determined that the actual de-icing operation sequence fails the verification.

[0013] In a further embodiment, the actual execution timestamp includes the time when the disconnection instruction is issued and the time when the disconnection is completed.

[0014] In a further embodiment, the step of generating an optimized de-icing operation sequence using the decision tree pruning algorithm based on the de-icing operation verification result includes: When the de-icing operation verification result is that the actual de-icing operation sequence fails the verification, construct a decision tree model representing the current oscillation characteristics; the leaf nodes in the decision tree model represent the de-icing operation decisions; Calculate the decrease value of the Gini index before and after the split of the leaf node in the decision tree model, and calculate the current oscillation sensitivity factor based on the amplitude and oscillation frequency of the current fluctuation caused by each ice melting operation decision. Use the current oscillation sensitivity factor to weight the decrease value of the Gini index to obtain the node importance weight. Traverse all branches of the decision tree model, prune the non-critical branches according to the mean value of the node importance weights, and merge the pruned subtrees to generate a reconstructed decision tree. Map the leaf node path of the reconstructed decision tree to an optimized ice melting operation sequence.

[0015] In a second aspect, the present invention provides an intelligent control system for power ice melting, and the system includes: A data acquisition module, configured to collect meteorological parameters of a target line of the power ice melting system in real time, and generate a grounding switch opening instruction when it is detected according to the meteorological parameters that the current external environment meets a preset grounding switch opening condition. A switch opening planning module, configured to control the grounding switch to perform an opening operation according to the grounding switch opening instruction, and determine the optimal opening time point of the phase selection switch based on the grounding switch opening completion status and the line equipment status parameters collected in real time. A timing analysis module, configured to dynamically generate an opening timing rule by using an ant colony optimization algorithm according to the optimal opening time point, and control the phase selection switch to perform an opening operation according to the opening timing rule. A priority analysis module, configured to obtain the actual execution delay data of the phase selection switch opening operation, and determine the priority coefficient of the short-circuit closing instruction according to the actual execution delay data. An ice melting verification module, configured to verify the actual ice melting operation sequence to obtain an ice melting operation verification result if it is monitored that the ground wire is successfully connected in series with the wire during the process of controlling the short-circuit switch to perform a closing operation according to the priority coefficient. An ice melting optimization module, configured to generate an optimized ice melting operation sequence by using a decision tree pruning algorithm based on the ice melting operation verification result.

[0016] The present invention provides an intelligent control method and system for power de-icing. The method includes: when it is detected according to the meteorological parameters of the target line of the power de-icing system collected in real time that the current external environment meets the preset earthing switch opening condition, generating an earthing opening command; controlling the earthing switch to perform an opening operation according to the earthing opening command, and determining the optimal opening time point of the phase selection switch based on the earthing switch opening completion status and the line equipment status parameters collected in real time; according to the optimal opening time point, dynamically generating an opening time sequence rule by using the ant colony optimization algorithm, and controlling the phase selection switch to perform an opening operation according to the opening time sequence rule; obtaining the actual execution delay data of the phase selection switch opening operation, and determining the priority coefficient of the short-circuit closing command according to the actual execution delay data; during the process of controlling the short-circuit switch to perform a closing operation according to the priority coefficient, if it is monitored that the ground wire is successfully connected in series with the conductor, then verifying the actual de-icing operation sequence to obtain a de-icing operation verification result; based on the de-icing operation verification result, generating an optimized de-icing operation sequence by using the decision tree pruning algorithm. Compared with the prior art, this method realizes the intelligent and precise control of the power de-icing process by real-time meteorological monitoring and intelligent decision-making to optimize the switch operation time sequence and the de-icing operation sequence, effectively reduces the influence of current oscillation on the system, and improves the quality and efficiency of the overall de-icing operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic flow chart of the intelligent control method for power de-icing provided by an embodiment of the present invention; Figure 2 is a block diagram of the intelligent control system for power de-icing provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The following specifically illustrates the embodiments of the present invention in conjunction with the drawings. The given embodiments are only for illustrative purposes and should not be construed as limiting the present invention. The drawings are only for reference and illustration, and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0019] Refer to Figure 1 , an embodiment of the present invention provides an intelligent control method for power de-icing. As Figure 1 shown, the method includes the following steps: S1. Collect the meteorological parameters of the target line of the power de-icing system in real time, and generate an earthing opening command when it is detected according to the meteorological parameters that the current external environment meets the preset earthing switch opening condition.

[0020] In some embodiments, the step of generating an earthing opening command when it is detected according to the meteorological parameters that the current external environment meets the preset earthing switch opening condition includes: Perform a moving average filtering process on the meteorological parameters to obtain the preprocessed meteorological parameters, and obtain the current ambient temperature data, conductor ice coating thickness data, and wind speed data based on the preprocessed meteorological parameters; According to the conductor ice coating thickness data, use the sliding window method to calculate the ice coating thickness growth rate at adjacent time points, and calculate the wind speed fluctuation variance feature based on the current wind speed data and the historical average wind speed; According to the difference between the current wind speed data and the historical average wind speed and the reference temperature threshold, obtain the dynamic temperature threshold through linear interpolation, and compare the current ambient temperature data with the dynamic temperature threshold to obtain a temperature comparison result; If the temperature comparison result, the ice coating thickness growth rate, and the wind speed fluctuation variance feature all meet the preset earthing switch opening conditions, trigger the priority scheduling algorithm to place the earthing switch opening instruction that meets the earthing switch opening conditions at the head of the resource allocation queue, and generate an earthing switch opening instruction.

[0021] Specifically, in this embodiment, meteorological sensors are installed on the target line of the power de-icing system. The meteorological sensors include a temperature sensor, a humidity sensor, a wind speed sensor, and a sensor for measuring the conductor ice coating thickness. Use a data recorder to read the data of these sensors in real time, and transmit the data to the central control unit by wired or wireless means. After the central control unit receives the original meteorological data, the central control unit will use the moving average filtering algorithm to smooth the data to reduce the influence of noise and outliers. The size of the sliding window can be adjusted according to actual needs, for example, set to the average value within 5 minutes. Extract the ambient temperature, conductor ice coating thickness, and wind speed data at the current time point from the preprocessed meteorological parameter database, and use the sliding window method to calculate the ice coating thickness growth rate within adjacent time windows. Specifically, it is obtained by calculating the ratio of the difference in ice coating thickness between the current time window and the previous time window to the time interval, to obtain the ice coating thickness growth rate; at the same time, this embodiment calculates the variance between the current wind speed and the historical average wind speed as the wind speed fluctuation feature, and the historical average wind speed can be calculated from the wind speed data in the database over a period of time in the past.

[0022] In this embodiment, based on the sum of the difference between the current wind speed data and the historical average wind speed and a preset reference temperature threshold, the dynamic temperature threshold is dynamically adjusted by the linear interpolation method. The current ambient temperature is compared with the dynamic temperature threshold to obtain a temperature comparison result. Considering the temperature comparison result, the ice accretion thickness growth rate, and the wind speed fluctuation variance characteristics, it is determined whether the current external environment meets the preset earthing switch opening condition. Specifically, the temperature comparison result is that the current ambient temperature is lower than the dynamic temperature threshold, the ice accretion thickness growth rate is greater than the preset ice accretion growth critical value, and the wind speed fluctuation variance is greater than the preset variance threshold. If the earthing switch opening condition is met, the priority scheduling algorithm is triggered. This priority scheduling algorithm places the earthing switch opening instruction that meets the conditions at the head of the resource allocation queue to ensure that the instruction can be executed as soon as possible. In this embodiment, a specific earthing switch opening instruction is generated according to the priority scheduling result and sent to the control unit of the earthing switch through the communication network to perform the earthing operation.

[0023] S2. Control the earthing switch to perform the opening operation according to the earthing switch opening instruction, and determine the optimal opening time point of the phase selection switch based on the earthing switch opening completion status and the line equipment status parameters collected in real time.

[0024] In some embodiments, the step of determining the optimal opening time point of the phase selection switch based on the earthing switch opening completion status and the line equipment status parameters collected in real time includes: When the earthing switch opening completion status is that the earthing switch has been successfully opened, the line equipment status parameters are collected in real time, and the line equipment status parameters are subjected to spectral analysis to obtain the line harmonic distortion rate and the equipment vibration spectral characteristics; Define state variables and decision variables; wherein, the state variables include the current time, the earthing switch opening status, the line harmonic distortion rate, and the equipment vibration spectral characteristics, and the decision variable is the opening time point of the phase selection switch; Taking the minimization of the current oscillation amplitude as the optimization goal, based on the state variables and the decision variables, a state transition equation is established through the physical characteristics and dynamic behavior of the power system when opening at different time points; Based on the state transition equation, the Bellman dynamic programming algorithm is used to iteratively calculate the current oscillation amplitude at all opening time points of the phase selection switch; In each iteration, the state variables are updated according to the state transition equation, and the opening time point with the minimum current oscillation amplitude is selected as the optimal strategy for the current step; When the change in the current oscillation amplitude is less than the preset current oscillation amplitude threshold, the optimal opening time point of the phase selection switch is output.

[0025] Specifically, in this embodiment, a grounding opening signal is sent to the grounding switch through the power automation control system. After the driving mechanism of the grounding switch receives this grounding opening signal, it drives the switch blade to move from the closed position to the open position to complete the opening operation. During the opening process of the grounding switch, the position sensor installed on the switch blade is used to monitor the position state of the switch blade in real time. When the position sensor detects that the switch blade has reached the open position and remains stable, the system determines that the grounding switch has been successfully opened. At this time, this embodiment uses the sensors installed on the line to collect the state parameters of the line equipment in real time, and uses the Fast Fourier Transform (FFT) algorithm to perform spectral analysis on the collected state parameters of the line equipment. The line harmonic distortion rate is obtained through spectral analysis. The line harmonic distortion rate reflects the proportion of harmonic components in the line. The calculation formula for the line harmonic distortion rate is: In the formula, is the line harmonic distortion rate; are the harmonic components of each order of the line current; is the fundamental component of the line current; N is the total number of harmonic components.

[0026] At the same time, spectral analysis is performed on the vibration data to extract the vibration spectral characteristics of the equipment. The vibration spectral characteristics of the equipment include the main frequency components of the equipment vibration and their corresponding amplitudes. Then this embodiment defines the current time, the grounding switch opening state S (S = 1 indicates opened, S = 0 indicates not opened), the line harmonic distortion rate, and the vibration spectral characteristics of the equipment as state variables, and takes the opening time point of the phase selection switch as the decision variable. The opening time point of the phase selection switch is a variable that can be selected within a certain time range. Different opening time points of the phase selection switch correspond to different opening moments. With minimizing the current oscillation amplitude as the optimization goal, that is, finding the optimal opening time point so that after opening at this moment, the amplitude of the system current oscillation is the smallest, thereby reducing the impact on the stability of the power system and equipment. Then this embodiment, according to the physical characteristics and dynamic behavior of the power system, uses the power system simulation model to simulate the response of the power system when opening at different time points, and establishes a state transition equation. This state transition equation describes how the system state (including the current time, the grounding switch opening state, the line harmonic distortion rate, and the vibration spectral characteristics of the equipment) and the decision variable (opening time point) at a certain moment affect the state at the next moment and the current oscillation amplitude. At the same time, this embodiment sets the initial moment, the initial state, and the value range of the opening time point , initializes the array used to store the current oscillation amplitude at each opening time point, and uses the Bellman dynamic programming algorithm to perform iterative calculations from back to front. In each iteration, according to the state transition equation, calculate the current oscillation amplitude of the system under the current state and different opening time points, and update the state variables. In each iteration, compare different opening time points The corresponding current oscillation amplitude, and select the one with the smallest current oscillation amplitude as the optimal strategy for the current step, and record the corresponding current oscillation amplitude. When the change in the current oscillation amplitude in consecutive multiple iterations is less than the preset current oscillation amplitude threshold, it is considered that the algorithm converges. At this time, the opening time point output is the optimal opening time point of the phase selection disconnector.

[0027] S3. According to the optimal opening time point, use the ant colony optimization algorithm to dynamically generate an opening time sequence rule, and control the phase selection disconnector to perform an opening operation according to the opening time sequence rule.

[0028] In some embodiments, the step of using the ant colony optimization algorithm to dynamically generate an opening time sequence rule according to the optimal opening time point includes: According to the line length and ice coating distribution characteristics of the target line of the power de-icing system, segment the target line of the power de-icing system to obtain several line segments; Equivalent each of the line segments to a node, and construct a node topology graph according to the connection relationship between the nodes; Randomly initialize several ants as search agents, and assign an initial pheromone concentration to each directed edge in the node topology graph at the initial stage as the initial clue to guide the ants to search; Control each ant to start from the starting node, and after each round of iteration, evaluate the path generated by each ant to obtain the reduction value of the current oscillation amplitude of the opening path; Update the pheromone concentration on the opening path according to the reduction value of the current oscillation amplitude of the opening path to obtain the updated value of the pheromone concentration; Calculate the transfer probability of the ant transferring from the current node to the next node according to the updated value of the pheromone concentration and the node weight, and select the next node according to the transfer probability until all nodes are traversed to form a complete opening time sequence path; Repeat the above process, and after the iterative calculation is completed, output the opening time sequence rule corresponding to the optimal opening time sequence path; the opening time sequence rule includes the node access order and the opening time interval.

[0029] Specifically, in this embodiment, the geographic information system is used to determine key parameters such as the line length and ice coating thickness of the target line of the power de-icing system. According to these key parameters, the target line is divided into several relatively uniform line segments, each line segment represents a line section, and each line section is regarded as a node. Then, a node topology graph is constructed according to the electrical connection relationship and sequential direction relationship between the nodes. In the node topology graph, adjacent nodes are connected by directed edges, which represent the timing dependence relationship between the disconnection operations of the phase selection disconnecting switches, that is, the disconnection operation of the previous line section will affect the disconnection operation timing of the subsequent line section. Then, this embodiment sets key parameters such as the number of ants and pheromone concentration in the ant colony algorithm, and randomly generates a preset number of ants as search agents. Each ant represents a disconnection timing path searcher. In the initial stage, this embodiment assigns the same initial pheromone concentration value to each directed edge in the node topology graph as the initial clue to guide the ant search. In each round of iteration, this embodiment controls each ant to start from the starting node and select the next node according to the pheromone concentration and transition probability of the current node until all nodes are traversed to form a complete disconnection timing path. When an ant selects a node at each step, it simulates the path-finding behavior of a real ant and tends to select a path with a high pheromone concentration and a large node weight. After each round of iteration, the disconnection timing paths generated by each ant are evaluated. The power system simulation software is used to simulate the current change situation of the system under this disconnection timing to obtain the current oscillation amplitude, and it is compared with the effective value volatility of the current before disconnection, so as to obtain the current oscillation amplitude reduction value under this path as the evaluation index of the path quality. This embodiment updates the pheromone concentration on the path according to the current oscillation amplitude reduction value of the disconnection path. The specific pheromone concentration update formula is as follows: In the formula, is the updated pheromone concentration; is the pheromone evaporation coefficient; is the pheromone concentration at time t; is the pheromone concentration increment; is the pheromone intensity constant; is the current oscillation amplitude reduction value after the disconnection of the disconnection timing path; m is the number of ants; k is the ant index.

[0030] At the same time, during each step of transfer, according to the updated pheromone concentration and node weights, recalculate the transfer probability of the ant from the current node to the next node. During the iteration process, according to the calculated transfer probability, use a random selection method such as roulette to determine the next node of the ant, guiding the ant to gradually converge to the optimal path. Repeat this process until the ant traverses all nodes, forming a complete switching-off time sequence path. Repeat the above steps of ant path search, evaluation, pheromone update, and transfer probability calculation. After multiple rounds of iteration, continuously optimize the search for the switching-off time sequence path by the ant, and the pheromone concentration gradually concentrates on the path with better effect of reducing the current oscillation amplitude. After the iteration calculation is completed, select the path with the largest reduction value of the current oscillation amplitude from the switching-off time sequence paths generated by all ants as the optimal switching-off time sequence path. According to this path, output the corresponding switching-off time sequence rules. The switching-off time sequence rules include the access order of each line segment (node) and the switching-off time interval between adjacent nodes. Input the generated switching-off time sequence rules into the power automation control system, and use the timer and logic controller in the power automation control system to control the phase-selection disconnecting switch to perform the switching-off operation in sequence according to the node access order and the switching-off time interval, and monitor the current oscillation situation during the switching-off operation to ensure the safety and effectiveness of the switching-off operation.

[0031] In summary, in this embodiment, by dynamically generating the switching-off time sequence rules of the target line of the power de-icing system and controlling the phase-selection disconnecting switch to perform the switching-off operation according to the switching-off time sequence rules, the amplitude of the current oscillation during the switching-off process is effectively reduced, and the stability and safety of the power de-icing system are improved.

[0032] S4. Obtain the actual execution delay data of the phase-selection disconnecting switch switching-off operation, and determine the priority coefficient of the short-circuit closing command according to the actual execution delay data.

[0033] In some embodiments, the step of determining the priority coefficient of the short-circuit closing command according to the actual execution delay data includes: Obtain the oscillation amplitude of the phase-selection operation current of the phase-selection disconnecting switch switching-off operation, and perform correlation analysis on the actual execution delay data and the oscillation amplitude of the phase-selection operation current to establish a delay-oscillation amplitude mapping relationship table; Take the minimum delay value in the delay-oscillation amplitude mapping relationship table as the dynamic expected delay, calculate the deviation rate between the actual execution delay data and the dynamic expected delay, and obtain the delay deviation rate; Calculate the ratio between the oscillation amplitude of the phase-selection operation current and the reference current amplitude to obtain the current volatility; Take the delay deviation rate and the current volatility as fuzzy control input quantities, perform fuzzy processing on the fuzzy control input quantities, and obtain several fuzzy sets; According to the time delay - oscillation amplitude mapping relation table, perform fuzzy inference on the fuzzy set to obtain the priority coefficient of the short closing switch instruction.

[0034] Specifically, when the phase - selecting switchblade executes the opening operation, in this embodiment, the time synchronization device is used to record in real - time the time difference between the moment when the opening instruction is issued and the moment when the mechanical action of the switchblade is completed, obtain the actual execution time delay data of the phase - selecting switchblade opening operation, and monitor the current change before and after the opening operation through the current sensor, calculate the oscillation amplitude of the phase - selecting operation current caused by each phase - selecting opening operation, conduct statistical analysis on the actual execution time delay data and the oscillation amplitude of the phase - selecting operation current, find the potential relationship between the time delay and the oscillation amplitude, and establish a time delay - oscillation amplitude mapping relation table according to the statistical results. This table reflects the current oscillation amplitude generated under different time delays. Find the minimum time delay value in the mapping relation table and use it as the dynamic expected time delay. This is because a shorter time delay usually means a faster response of the opening operation, has less impact on the current oscillation of the system, and is beneficial to system stability. At the same time, calculate the deviation between each actual execution time delay data and the dynamic expected time delay, that is, the ratio of the difference between the actual execution time delay data and the dynamic expected time delay to the dynamic expected time delay, to obtain the time delay deviation rate. This time delay deviation rate index reflects the degree of deviation of the actual time delay relative to the expected time delay. In this embodiment, the reference current amplitude under the current working condition is extracted from the mapping table, compare the oscillation amplitude of the phase - selecting operation current with the reference current amplitude, calculate the ratio of the two to obtain the current volatility. The current volatility is used to measure the degree of fluctuation of the current oscillation caused by the opening operation relative to the reference level.

[0035] This embodiment defines the universe of discourse and membership function of the fuzzy set. For example, the universes of discourse of the time delay deviation rate and the current volatility can be set as [0, 1] respectively, and the membership function can be selected as a Gaussian function or a triangular function, etc. According to the actual time delay deviation rate and current volatility data, use the membership function to calculate the membership degree of each data point corresponding to the fuzzy set, and classify the data points into the fuzzy set with the highest membership degree to obtain several fuzzy sets. For example, for the time delay deviation rate, fuzzy sets such as "small", "medium", and "large" can be defined. This embodiment formulates a fuzzy rule base according to the established time delay - oscillation amplitude mapping relation table and actual operation experience and knowledge. The rule form is like "If the time delay deviation rate is small and the current volatility fluctuates slightly, then the priority coefficient is high". Then, use the Mamdani inference method to match the fuzzified time delay deviation rate and current volatility with the fuzzy rules to obtain the corresponding fuzzy output set, and perform defuzzification processing on the output set obtained by fuzzy inference to finally obtain the specific priority coefficient of the short closing switch instruction. This coefficient reflects the execution priority of the short closing switch instruction under the current time delay and current oscillation conditions.

[0036] S5. During the process of controlling the bypass switch to perform the closing operation according to the priority coefficient, if it is monitored that the ground wire is successfully connected in series with the wire, the actual ice melting operation sequence is verified to obtain the ice melting operation verification result.

[0037] In some embodiments, the step of verifying the actual ice melting operation sequence to obtain the ice melting operation verification result if it is monitored that the ground wire is successfully connected in series with the wire includes: During the closing operation of the bypass switch, the contact resistance value between the ground wire and the wire is collected in real time. If the contact resistance value is lower than the preset contact resistance threshold, it is determined that the ground wire is successfully connected in series with the wire; After the ground wire is successfully connected in series with the wire, the current oscillation characteristics during the execution of the actual ice melting operation sequence are obtained, and the actual execution timestamp of the actual ice melting operation sequence is extracted; According to the expected time window of the preset standard operation sequence, the relative deviation rate of the actual execution timestamp is calculated to obtain the timing deviation rate; If the timing deviation is within the timing tolerance threshold range and the current oscillation characteristics are within the oscillation characteristic threshold range, it is determined that the actual ice melting operation sequence passes the verification; otherwise, it is determined that the actual ice melting operation sequence fails the verification.

[0038] Specifically, during the process of controlling the bypass switch to perform the closing operation through the priority coefficient, the resistance measuring instrument in this embodiment collects the contact resistance value between the ground wire and the wire in real time. The resistance measuring instrument can be installed at the connection between the ground wire and the wire, and can quickly and accurately measure tiny resistance changes. At the same time, the collected contact resistance value is compared with the preset contact resistance threshold. If the collected contact resistance value is lower than the preset contact resistance threshold, it indicates that the contact between the ground wire and the wire is good and the electrical connection is reliable. It is determined that the ground wire is successfully connected in series with the wire, and subsequent ice melting operations can be carried out. After the ground wire is successfully connected in series with the wire, the ice melting operation is started, and the current sensor installed in the power line is used to obtain the current signal during the execution of the actual ice melting operation sequence in real time. Methods such as wavelet transform are used to analyze the oscillation characteristics of the current data and extract the current oscillation characteristics. The current oscillation characteristics include oscillation amplitude, frequency, and duration, etc., to comprehensively reflect the current changes during the ice melting operation. At the same time, during the execution of the ice melting operation sequence, the actual execution timestamps of each operation such as closing, starting ice melting, and ending ice melting in the ice melting operation sequence are recorded in this embodiment. The actual execution timestamp includes the time when the opening command is issued and the time when the opening is completed.

[0039] Then, in this embodiment, according to the expected time window of the preset standard operation sequence, the relative deviation rate between the actual execution timestamp and the expected time window is calculated. The expected time window is determined based on factors such as historical data and equipment performance, and it represents the execution time range of each step under normal circumstances. Assuming that the expected time window of the preset standard operation sequence is , the calculation formula for the relative deviation rate of the actual execution timestamp is as follows: Wherein, In the formula, is the relative deviation rate; is the breaker closing completion time; is the mean value of the expected time window; is the start time of the expected time window; is the end time of the expected time window.

[0040] In this embodiment, the relative deviation rate can quantify the degree of deviation of the actual operation in time, providing a basis for subsequent verification and judgment. If the timing deviation is within the timing tolerance threshold range and the current oscillation characteristic is within the oscillation characteristic threshold range, it is determined that the actual ice melting operation sequence passes the verification. Among them, the timing tolerance threshold represents the allowable timing deviation range, and the oscillation characteristic threshold represents the allowable current oscillation characteristic range. Otherwise, it is determined that the actual ice melting operation sequence fails the verification, and the ice melting operation needs to be adjusted and optimized to ensure the safety and effectiveness of the ice melting process.

[0041] S6. Based on the ice melting operation verification result, use the decision tree pruning algorithm to generate an optimized ice melting operation sequence.

[0042] In some embodiments, the step of using the decision tree pruning algorithm to generate an optimized ice melting operation sequence based on the ice melting operation verification result includes: When the ice melting operation verification result is that the actual ice melting operation sequence fails the verification, construct a decision tree model representing the current oscillation characteristics; the leaf nodes in the decision tree model represent ice melting operation decisions; Calculate the Gini index decrease value before and after the splitting of the leaf nodes in the decision tree model, and calculate the current oscillation sensitivity factor according to the current fluctuation amplitude and oscillation frequency caused by each ice melting operation decision; Use the current oscillation sensitivity factor to weight the Gini index decrease value to obtain the node importance weight; Traverse all branches of the decision tree model, prune the non-critical branches according to the mean value of the node importance weights, and merge the pruned subtrees to generate a reconstructed decision tree; Map the leaf node path of the reconstructed decision tree to an optimized ice melting operation sequence.

[0043] Specifically, when the verification result of the ice melting operation indicates that the actual ice melting operation sequence fails the verification, it means that there is an abnormality in the current operation sequence. In this embodiment, the collected current oscillation characteristic data is used as the characteristic variable. The current oscillation characteristic data includes the current fluctuation amplitude, oscillation frequency, etc. At the same time, the ice melting operation decision (such as adjusting the switch closing sequence, time interval, etc.) is used as the target variable. Each leaf node of the decision tree represents a specific ice melting operation decision. Then, this embodiment uses the CART algorithm to construct a decision tree model, with the current oscillation characteristics as the splitting basis, and gradually divides the data set until the leaf node represents a specific ice melting operation decision. During the construction of the decision tree, this embodiment calculates the decrease in the Gini index before and after the splitting of each leaf node. The decrease in the Gini index before and after the splitting reflects the improvement in the purity of the data set caused by the splitting of this node. The Gini index before the splitting is the Gini index of the parent node, and the Gini index after the splitting is the value obtained by weighted averaging the Gini indices of each child node according to the data volume. The decrease in the Gini index is defined as the difference between the Gini index before the splitting and the Gini index after the splitting. At the same time, this embodiment calculates the current fluctuation amplitude and oscillation frequency caused by each ice melting operation decision, and calculates the current oscillation sensitivity factor according to the changes in the current fluctuation amplitude and oscillation frequency. The current oscillation sensitivity factor reflects the influence degree of the operation decision on the current oscillation characteristics. The calculation formula of the current oscillation sensitivity factor is: In the formula, is the current oscillation sensitivity factor; is the current fluctuation amplitude of the q-th operation; is the oscillation frequency of the q-th operation.

[0044] In this embodiment, the decrease value of the Gini index is weighted by the current oscillation sensitivity factor to obtain the node importance weight, thus comprehensively considering the contribution of the node in data partitioning and the impact of the corresponding operation decision on the current oscillation. The greater the node importance weight, the more important the node is in the decision tree, and the higher the sensitivity of its corresponding operation decision to the current oscillation. Then, starting from the root node, this embodiment traverses all branches and nodes of the decision tree model, calculates the mean value of the node importance weights of all nodes in each branch (including subtrees). If the mean value of the node importance of a certain branch is lower than the preset pruning threshold, it is considered that this branch is a non-critical branch, and this branch is pruned, that is, this branch and its subtree are deleted. The pruned decision tree is merged, and adjacent leaf nodes or nodes with similar decision results are merged into new leaf nodes. After pruning and merging, the complexity of the decision tree is reduced, and at the same time, the critical operation decision path is retained to generate a reconstructed decision tree. At this time, this embodiment traverses each leaf node path of the reconstructed decision tree from the root node, maps the decision nodes in each leaf node path to specific de-icing operations, and generates an optimized de-icing operation sequence according to the order and decision results of the path, effectively reducing the impact of current oscillation on the power system and improving the success rate of de-icing operations and the stability of system operation.

[0045] An embodiment of the present invention provides a power de-icing intelligent control method. The method includes: collecting meteorological parameters of the target line of the power de-icing system in real time, and generating a grounding switch opening instruction when it is detected according to the meteorological parameters that the current external environment meets the preset grounding switch opening condition; controlling the grounding switch to perform an opening operation according to the grounding switch opening instruction, and determining the optimal opening time point of the phase selection switch based on the grounding switch opening completion state and the line equipment state parameters collected in real time; according to the optimal opening time point, dynamically generating an opening time sequence rule by using the ant colony optimization algorithm, and controlling the phase selection switch to perform an opening operation according to the opening time sequence rule; obtaining the actual execution delay data of the phase selection switch opening operation, and determining the priority coefficient of the short-circuit closing instruction according to the actual execution delay data; during the process of controlling the short-circuit switch to perform a closing operation according to the priority coefficient, if it is monitored that the ground wire is successfully connected in series with the conductor, the actual de-icing operation sequence is verified to obtain a de-icing operation verification result; based on the de-icing operation verification result, an optimized de-icing operation sequence is generated by using the decision tree pruning algorithm. Compared with the prior art, this method realizes the intelligent and precise control of the power de-icing process by real-time meteorological monitoring and intelligent decision-making to optimize the switch operation time sequence and the de-icing operation sequence, effectively reducing the impact of current oscillation on the system and improving the quality and efficiency of the overall de-icing operation.

[0046] It should be noted that the magnitudes of the serial numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0047] In one embodiment, as Figure 2 shown, an embodiment of the present invention provides an intelligent control system for power ice melting, and the system includes: A data acquisition module 101, configured to collect meteorological parameters of a target line of a power ice melting system in real time, and generate a grounding switch opening command when it is detected according to the meteorological parameters that the current external environment meets a preset grounding switch opening condition; A switch opening planning module 102, configured to control a grounding switch to perform an opening operation according to the grounding switch opening command, and determine an optimal opening time point of a phase selection switch based on the grounding switch opening completion state and the line equipment state parameters collected in real time; A timing analysis module 103, configured to dynamically generate a switch opening timing rule by using an ant colony optimization algorithm according to the optimal opening time point, and control a phase selection switch to perform an opening operation according to the switch opening timing rule; A priority analysis module 104, configured to obtain actual execution delay data of a phase selection switch opening operation, and determine a priority coefficient of a short-circuit closing command according to the actual execution delay data; An ice melting verification module 105, configured to verify an actual ice melting operation sequence to obtain an ice melting operation verification result if it is monitored that the ground wire is successfully connected in series with the conductor during the process of controlling a short-circuit switch to perform a closing operation according to the priority coefficient; An ice melting optimization module 106, configured to generate an ice melting operation optimization sequence by using a decision tree pruning algorithm based on the ice melting operation verification result.

[0048] For the specific limitations on an intelligent control system for power ice melting, reference may be made to the above limitations on an intelligent control method for power ice melting, which will not be elaborated here. Those of ordinary skill in the art can realize that each module and step described in combination with the embodiments disclosed in the present application can be implemented by hardware, software, or a combination of both. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0049] An embodiment of the present invention provides an intelligent control system for power de-icing. The system uses a data acquisition module to collect meteorological parameters of the target line of the power de-icing system in real time, and generates a grounding switch opening instruction when it detects that the current external environment meets the preset grounding switch opening condition according to the meteorological parameters. The opening planning module controls the grounding switch to perform an opening operation according to the grounding switch opening instruction, and determines the optimal opening time point of the phase selection switch based on the grounding switch opening completion status and the line equipment status parameters collected in real time. The timing analysis module dynamically generates an opening timing rule using the ant colony optimization algorithm according to the optimal opening time point, and controls the phase selection switch to perform an opening operation according to the opening timing rule. The priority analysis module obtains the actual execution delay data of the phase selection switch opening operation, and determines the priority coefficient of the short-circuit closing instruction according to the actual execution delay data. During the process of controlling the short-circuit switch to perform a closing operation according to the priority coefficient, if the ice melting verification module monitors that the ground wire and the conductor are successfully connected in series, it verifies the actual ice melting operation sequence to obtain an ice melting operation verification result. The ice melting optimization module generates an optimized ice melting operation sequence using the decision tree pruning algorithm based on the ice melting operation verification result. Compared with the prior art, the system realizes the intelligent and precise control of the power de-icing process by real-time meteorological monitoring and intelligent decision-making to optimize the switch operation timing and the ice melting operation sequence, effectively reducing the impact of current oscillation on the system and improving the quality and efficiency of the overall ice melting operation.

[0050] The above embodiments only represent several preferred embodiments of the present application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and substitutions can be made, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the protection scope of the claims.

Claims

1. An intelligent control method for electric ice melting, characterized in that: The following steps are involved: Collect meteorological parameters of the target line of the electric power ice melting system in real time, and generate a grounding switch-off instruction when it is detected according to the meteorological parameters that the current external environment meets the preset grounding switch-off conditions; Controlling the grounding switch to perform the opening operation according to the grounding opening instruction, and determining the optimal opening time point of the phase selection switch based on the grounding switch opening completion state and the line equipment state parameters collected in real time; According to the optimal opening time point, an ant colony optimization algorithm is used to dynamically generate an opening timing rule, and according to the opening timing rule, a phase-selected knife switch is controlled to perform an opening operation; Acquire actual execution delay data of the phase selection knife switch opening operation, and determine the priority coefficient of the short-circuit closing switch instruction according to the actual execution delay data; In the process of controlling the short-circuit knife switch to perform the closing operation according to the priority coefficient, if it is monitored that the ground wire and the conductor are successfully connected in series, the actual ice-melting operation sequence is verified to obtain the ice-melting operation verification result; Based on the ice-melting operation verification result, a decision tree pruning algorithm is used to generate an ice-melting operation optimization sequence.

2. The electric ice melting intelligent control method according to claim 1, characterized in that: When it is detected according to the meteorological parameters that the current external environment meets the preset grounding switch opening condition, the step of generating a grounding switch opening instruction comprises: Performing sliding average filtering on the meteorological parameters to obtain preprocessed meteorological parameters, and obtaining current ambient temperature data, conductor ice thickness data, and wind speed data according to the preprocessed meteorological parameters; According to the conductor ice thickness data, the ice thickness growth rate at adjacent time points is calculated using a sliding window method, and the wind speed fluctuation variance characteristics are calculated based on the current wind speed data and the historical average wind speed; According to the difference between the current wind speed data and the historical average wind speed and the reference temperature threshold, a dynamic temperature threshold is obtained by linear interpolation, and the current ambient temperature data is compared with the dynamic temperature threshold to obtain a temperature comparison result; If the temperature comparison result, the ice thickness growth rate and the wind speed fluctuation variance characteristics all meet the preset grounding switch opening conditions, the priority scheduling algorithm is triggered to place the grounding switch opening instruction that meets the grounding switch opening conditions at the top of the resource allocation queue, and generate a grounding switch opening instruction.

3. The electric ice melting intelligent control method according to claim 1, characterized in that: The step of determining the optimal opening time point of the phase selection switch based on the opening completion state of the grounding switch and the line equipment state parameters collected in real time comprises: When the grounding switch opening completion state is that the grounding switch has been successfully opened, the line equipment status parameters are collected in real time, and the spectrum analysis is performed on the line equipment status parameters to obtain the line harmonic distortion rate and equipment vibration spectrum characteristics; Define state variables and decision variables; wherein the state variables include the current time, the grounding switch opening state, the line harmonic distortion rate and the equipment vibration spectrum characteristics, and the decision variable is the opening time point of the phase selection switch; Taking minimizing the current oscillation amplitude as the optimization goal, based on the state variables and the decision variables, a state transfer equation is established through the physical characteristics and dynamic behaviors of the power system when the switch is opened at different time points; Based on the state transfer equation, the current oscillation amplitude at all phase-selected knife switch opening time points is iteratively calculated using the Bellman dynamic programming algorithm; In each iteration, the state variables are updated according to the state transfer equation, and the opening time point with the smallest current oscillation amplitude is selected as the optimal strategy for the current step; When the change in the current oscillation amplitude is less than a preset current oscillation amplitude threshold, the optimal opening time point of the phase-selection knife switch is output.

4. The electric ice melting intelligent control method according to claim 1, characterized in that: The step of dynamically generating a trip timing rule using an ant colony optimization algorithm according to the optimal trip time point comprises: According to the line length and ice distribution characteristics of the target line of the electric ice melting system, the target line of the electric ice melting system is segmented to obtain a number of line segments; Each of the line segments is equivalent to a node, and a node topology graph is constructed according to the connection relationship between the nodes; Randomly initialize a number of ants as search agents, and assign an initial pheromone concentration to each directed edge in the node topology graph in the initial stage as an initial clue to guide the ants' search; Control each ant to start from the starting node, and after each round of iteration, evaluate the path generated by each ant to obtain the current oscillation amplitude reduction value of the tripping path; The pheromone concentration on the gate-off path is updated according to the current oscillation amplitude reduction value of the gate-off path to obtain a pheromone concentration update value; Calculate the transfer probability of the ant from the current node to the next node according to the pheromone concentration update value and the node weight, and select the next node according to the transfer probability until all nodes are traversed to form a complete opening timing path; Repeat the above process, and after the iterative calculation is completed, output the opening timing rule corresponding to the optimal opening timing path; the opening timing rule includes the node access sequence and the opening time interval.

5. The electric ice melting intelligent control method according to claim 4, characterized in that: There are directed edges between adjacent nodes in the node topology graph, and the directed edges represent the timing dependency between the phase selection knife switch opening operations.

6. The electric ice melting intelligent control method according to claim 1, characterized in that: The step of determining the priority coefficient of the short engagement brake instruction according to the actual execution delay data comprises: Obtaining the phase selection operation current oscillation amplitude of the phase selection knife switch opening operation, and correlating and analyzing the actual execution delay data and the phase selection operation current oscillation amplitude, and establishing a time delay-oscillation amplitude mapping relationship table; Taking the minimum delay value in the delay-oscillation amplitude mapping relationship table as the dynamic expected delay, calculating the deviation rate between the actual execution delay data and the dynamic expected delay, and obtaining the delay deviation rate; Calculating the ratio between the phase selection operation current oscillation amplitude and the reference current amplitude to obtain the current fluctuation rate; Taking the time delay deviation rate and the current fluctuation rate as fuzzy control input quantities, fuzzifying the fuzzy control input quantities to obtain several fuzzy sets; According to the time delay-oscillation amplitude mapping relationship table, fuzzy reasoning is performed on the fuzzy set to obtain the priority coefficient of the short engagement brake instruction.

7. The electric ice melting intelligent control method according to claim 1, characterized in that: If it is detected that the ground wire and the conductor are successfully connected in series, the actual ice-melting operation sequence is verified to obtain the ice-melting operation verification result, including: In the closing operation of the short-circuit knife switch, the contact point resistance value of the ground wire and the conductor is collected in real time. If the contact point resistance value is lower than a preset contact resistance threshold, it is determined that the ground wire and the conductor are successfully connected in series; After the ground wire is successfully connected in series with the conductor, the current oscillation characteristics of the actual execution process of the ice-melting operation sequence are obtained, and the actual execution timestamp of the actual ice-melting operation sequence is extracted; Calculate the relative deviation rate of the actual execution timestamp according to the expected time window of the preset standard operation sequence to obtain the timing deviation rate; If the timing deviation is within the timing tolerance threshold range and the current oscillation characteristic is within the oscillation characteristic threshold range, it is determined that the actual ice-melting operation sequence verification has passed; otherwise, it is determined that the actual ice-melting operation sequence verification has failed.

8. The electric ice melting intelligent control method according to claim 7, characterized in that: The actual execution timestamp includes the time when the opening instruction is issued and the time when the opening is completed.

9. The electric ice melting intelligent control method according to claim 1, characterized in that: The step of generating an optimized sequence of ice melting operations using a decision tree pruning algorithm based on the ice melting operation verification result comprises: When the ice-melting operation verification result is that the actual ice-melting operation sequence verification fails, a decision tree model characterizing the current oscillation characteristics is constructed; a leaf node in the decision tree model represents an ice-melting operation decision; The Gini index decrease value before and after the leaf node splitting in the decision tree model is calculated, and the current oscillation sensitivity factor is calculated based on the current fluctuation amplitude and oscillation frequency caused by each ice melting operation decision; The Gini index reduction value is weighted by using the current oscillation sensitivity factor to obtain a node importance weight; Traversing all branches of the decision tree model, pruning non-critical branches according to the mean of the node importance weights, and merging the pruned subtrees to generate a reconstructed decision tree; The leaf node paths of the reconstructed decision tree are mapped to an optimized sequence of ice-melting operations.

10. An intelligent control system for electric ice melting, characterized in that: The system comprises: A data acquisition module is used to collect meteorological parameters of the target line of the electric power ice melting system in real time, and generate a grounding switch-off instruction when it is detected according to the meteorological parameters that the current external environment meets the preset grounding switch-off conditions; A trip planning module is used to control the grounding switch to perform a trip operation according to the grounding trip instruction, and determine the optimal trip time point of the phase selection switch based on the grounding switch trip completion state and the line equipment state parameters collected in real time; A timing analysis module, used to dynamically generate a switching-off timing rule by using an ant colony optimization algorithm according to the optimal switching-off time point, and control the phase-selected knife switch to perform a switching-off operation according to the switching-off timing rule; A priority analysis module, used to obtain actual execution delay data of the phase selection knife switch opening operation, and determine the priority coefficient of the short-circuit closing instruction according to the actual execution delay data; An ice-melting verification module is used to verify the actual ice-melting operation sequence and obtain the ice-melting operation verification result if it is detected that the ground wire and the conductor are successfully connected in series during the closing operation of controlling the short-circuit knife switch according to the priority coefficient; The ice melting optimization module is used to generate an ice melting operation optimization sequence based on the ice melting operation verification result by using a decision tree pruning algorithm.

Images