Overhead ground wire uninterrupted ice melting automatic segmentation method and system and storage medium thereof

Abstract

The invention provides an overhead ground wire uninterruptible ice melting automatic segmentation method and system and a storage medium, and the method comprises the following steps: constructing an ice melting evaluation model based on the environment conditions and structure parameters of a power transmission line, and judging whether an ice melting device has enough output capability or not; if the output capability of the ice melting device is not enough to cover the whole ground wire for ice melting, constructing a plurality of segmentation configurations at the sectionable tower, generating a candidate segmentation scheme set, and performing electrical parameter calculation on each candidate segmentation scheme; taking minimization of the number of segments and minimization of ice melting energy consumption as double targets, adopting a multi-target genetic algorithm to carry out iterative evolution on the candidate segment scheme set, and screening out a Pareto optimal solution set meeting the capacity of an ice melting device and line safety constraints; and selecting the optimal segmentation configuration meeting the uninterrupted ice melting requirement from the Pareto optimal solution set, and outputting the optimal segmentation configuration. According to the invention, good lightning protection performance and economical efficiency can be taken into account under the condition of live ice melting without power failure.

Description

Technical Field

[0001] The invention belongs to the technical field of power transmission lines, and in particular relates to an automatic segmentation method, system and storage medium for non-stop ice melting of overhead ground wires. Background Art

[0002] For some UHV DC transmission lines, the line span is large, spans many areas, and the micro-meteorological conditions vary significantly. The ice-melting section may involve several high-altitude or heavily iced sections. In addition, the ice-melting method has been changed from traditional power outage short circuit to live DC ice-melting. It is necessary to design a more flexible and efficient segmentation scheme for specific lines to balance the contradiction between "the number of segment transformations" and "the actual ice-melting power consumption":

[0003] Too few segments will result in a single segment that is too long, and the ice-melting voltage or current requirement will be too high, even exceeding the rated capacity of the ice-melting device; too many segments will increase the investment in segment switches and insulators. At the same time, because the number of segments increases, more equipment needs to be connected, which may push up the energy consumption cost during the ice-melting process. Traditionally, the design of ground wire segmentation schemes often relies on manual repeated attempts, comparing the electrical parameters and energy consumption under different segment configurations, and then making subjective selections based on engineering experience. This manual calculation and trial-and-error process is inefficient and labor-intensive when facing large-scale ultra-high voltage direct current transmission lines with long distances and multiple micro-meteorological sections, and it is easy to ignore some better segment configurations. At the same time, when considering multiple factors such as line lightning protection performance, ice-melting energy consumption costs, and equipment capacity constraints, manual methods are more difficult to iterate quickly, resulting in insufficient accuracy and reliability of the scheme.

[0004] It can be seen that the existing technology mainly faces the following problems in the segmented design of "long-distance overhead ground wire without power outage and ice melting":

[0005] Large amount of manual repetitive calculations: There are many combinations of parameters such as the number of segments, segment locations, and ice area distribution, and traditional methods lack automatic optimization methods;

[0006] It is difficult to take into account both lightning protection effect and energy consumption economy: we want to reduce the investment in segmented equipment but cannot make a single segment too long, which will cause the device capacity or safety to exceed the limit;

[0007] Insufficient coverage of actual projects: It is often impossible to make detailed and global optimization of ultra-long lines crossing multiple micro-meteorological zones. Summary of the invention

[0008] The purpose of the present invention is to solve the shortcomings of the above-mentioned background technology and to provide an automatic segmentation method, system and storage medium for melting ice of overhead ground wires without power outages, which can take into account both good lightning protection performance and economy under the conditions of melting ice with power outages, avoid a large number of repeated manual calculations as much as possible and quickly recommend a reasonable and efficient ice-melting segmentation configuration.

[0009] The technical solution adopted by the present invention is: an automatic sectionalizing method for de-icing overhead ground wires without power outage, including the following steps:

[0010] Based on the environmental conditions and structural parameters of the transmission line, a de-icing evaluation model is constructed to calculate the voltage, current, and energy consumption required for de-icing the ground wire without sectionalizing, and to determine whether the de-icing device has sufficient output capacity;

[0011] If the output capacity of the de-icing device is insufficient to cover the de-icing of the entire ground wire, multiple sectionalizing configurations are constructed at the sectionalizable towers to generate a set of candidate sectionalizing schemes, and the electrical parameters of each candidate sectionalizing scheme are calculated;

[0012] With the dual objectives of minimizing the number of sections and minimizing the de-icing energy consumption, a multi-objective genetic algorithm is used to iteratively evolve the set of candidate sectionalizing schemes, and a Pareto optimal solution set that meets the capacity of the de-icing device and the line safety constraints is screened out;

[0013] Based on the preset engineering criteria, the optimal sectionalizing configuration that meets the requirements of de-icing without power outage is selected from the Pareto optimal solution set and output.

[0014] In the above technical solution, the de-icing evaluation model is built on the ATP-EMTP platform. By changing the pulse trigger angle of the DC de-icing device, the change of the de-icing voltage of the ground wire under different de-icing connection modes is simulated, and based on this, the de-icing current and the device output capacity corresponding to each candidate sectionalizing scheme are calculated; the above calculation results are used for the multi-objective genetic algorithm calculation.

[0015] In the above technical solution, the structural parameters include tower type and parameters and conductor and ground wire parameters.

[0016] In the above technical solution, the environmental condition parameters include ice thickness, wind speed, and ambient temperature.

[0017] In the above technical solution, when performing electrical evaluation and iterative evolution on the candidate sectionalizing schemes, the following constraint conditions are imposed:

[0018] The de-icing voltage and de-icing current of each section of the ground wire shall not exceed the rated output of the de-icing device;

[0019] The sectionalizing position is limited to the pre-selected sectionalizable towers;

[0020] Any candidate sectionalizing scheme shall meet the range of line lightning protection and operation safety permission.

[0021] In the above technical solution, the multi-objective genetic algorithm encodes each candidate sectionalization scheme into a chromosome individual and forms a population, calculates two objective functions of the sectionalization quantity and the ice melting energy consumption for each individual, and screens out the individuals that do not meet the device capacity and line safety requirements; obtains multiple Pareto fronts through non-dominated sorting, and continuously updates the population during iteration by combining genetic operations of crossover and mutation until a converged set of Pareto optimal solutions is obtained.

[0022] In the above technical solution, the iterative process of the multi-objective genetic algorithm includes:

[0023] Calculate the fitness of randomly generated candidate sectionalization schemes and screen out feasible solutions;

[0024] Evolve the candidate individuals for several generations with a first mutation probability, and adopt selection, crossover, mutation, non-dominated sorting and elitist retention strategies to generate corresponding offspring and merge them with the parent generation;

[0025] Select the Pareto solution with the largest crowding degree from the merged population as the parent generation for the next stage of iteration;

[0026] Continue genetic evolution for several generations with a second mutation probability less than the first mutation probability, and finally obtain a set of Pareto optimal solutions based on non-dominated sorting.

[0027] In the above technical solution, when selecting the final sectionalization scheme from the set of Pareto optimal solutions, comprehensively consider the line operation cost, equipment capacity limit, and sectionalization quantity and energy consumption; if there are multiple Pareto solutions that meet the requirements, the solution with the largest crowding degree or the solution with the objective function value closest to the ideal point can be preferentially selected as the final output.

[0028] In the above technical solution, the generation process of the candidate sectionalization scheme set includes: sequentially mapping the preset tower positions to each bit of the binary encoding. Initially, only randomly set a "1" in the encoding to represent a single section; if this initial scheme does not meet the output requirements of the ice melting device, gradually increase the number of "1"s in the encoding and distribute them randomly among each tower position to form multiple candidate sectionalization schemes with gradually increasing sectionalization numbers.

[0029] In the above technical solution, if the ice melting without sectionalization of the whole line can meet the output limit of the ice melting device and the line operation requirements, there is no need to execute the subsequent sectionalization transformation and multi-objective optimization steps, and directly output the non-sectionalization scheme.

[0030] The present invention also provides an automatic sectionalization system for live-line ice melting of overhead ground wires, including:

[0031] The ice melting assessment module is used to construct an ice melting assessment model based on the environmental conditions and structural parameters of the transmission line, calculate the voltage, current, and energy consumption required for ice melting of the ground wire without segmentation, and determine whether the ice melting device has sufficient output capacity;

[0032] The candidate segmentation scheme generation module is used to construct multiple segmentation configurations at the segmentable towers if the output capacity of the device is insufficient to cover the ice melting of the entire ground wire, generate a set of candidate segmentation schemes, and calculate the electrical parameters for each candidate segmentation scheme;

[0033] The candidate segmentation scheme screening module is used to take the minimization of the number of segments and the minimization of ice melting energy consumption as dual objectives, use a multi-objective genetic algorithm to iteratively evolve the set of candidate segmentation schemes, and screen out the Pareto optimal solution set that meets the capacity of the ice melting device and the line safety constraints;

[0034] The candidate segmentation scheme output module is used to select the optimal segmentation configuration that meets the requirements of ice melting without power outage from the Pareto optimal solution set based on preset engineering criteria and output it.

[0035] In the above technical solution, the ice melting assessment model is built on the ATP-EMTP platform. By changing the pulse trigger angle of the DC ice melting device, the change of the ice melting voltage of the ground wire under different ice melting connection modes is simulated, and based on this, the ice melting current and the device output capacity corresponding to each candidate segmentation scheme are calculated; the above calculation results are used for multi-objective genetic algorithm calculation.

[0036] In the above technical solution, the structural parameters include tower type and parameters and conductor and ground wire parameters.

[0037] In the above technical solution, the environmental condition parameters include ice thickness, wind speed, and ambient temperature.

[0038] In the above technical solution, when the candidate segmentation scheme screening module conducts electrical evaluation and iterative evolution on the candidate segmentation scheme, the following constraint conditions are imposed:

[0039] The ice melting voltage and current of each section of the ground wire shall not exceed the rated output of the ice melting device;

[0040] The segmentation position is limited to the pre-selected segmentable towers;

[0041] Any candidate segmentation scheme shall meet the range of lightning protection and operation safety of the line.

[0042] In the above technical solution, the multi-objective genetic algorithm encodes each candidate sectionalization scheme into a chromosome individual to form a population, calculates two objective functions of the number of sections and the ice melting energy consumption for each individual, and screens out the individuals that do not meet the requirements of the device capacity and line safety; obtains multiple Pareto fronts through non-dominated sorting, and continuously updates the population during iteration by combining genetic operations such as crossover and mutation until a convergent set of Pareto optimal solutions is obtained.

[0043] In the above technical solution, the iteration process of the multi-objective genetic algorithm includes:

[0044] Calculate the fitness of the randomly generated candidate sectionalization schemes and screen the feasible solutions;

[0045] Evolve the candidate individuals for several generations with the first mutation probability, and adopt selection, crossover, mutation, non-dominated sorting and elitist retention strategies to generate corresponding offspring and merge them with the parent generation;

[0046] Select the Pareto solution with the largest crowding degree from the merged population as the parent generation for the next stage of iteration;

[0047] Continue genetic evolution for several generations with the second mutation probability less than the first mutation probability, and finally obtain the set of Pareto optimal solutions based on non-dominated sorting.

[0048] In the above technical solution, when the candidate sectionalization scheme screening module selects the final sectionalization scheme from the set of Pareto optimal solutions, it comprehensively considers the line operation cost, equipment capacity limit, and the number of sections and energy consumption; if there are multiple Pareto solutions that meet the requirements, the solution with the largest crowding degree or the solution with the objective function value closest to the ideal point can be preferentially selected as the final output.

[0049] In the above technical solution, the candidate sectionalization scheme generation module maps the preset tower positions to each bit of the binary encoding. Initially, only one "1" is randomly set in the encoding to represent a single section; if the initial scheme does not meet the output requirements of the ice melting device, the number of "1"s in the encoding is gradually increased and randomly distributed among each tower position to form multiple candidate sectionalization schemes with an increasing number of sections.

[0050] In the above technical solution, the ice melting evaluation module is also used to directly output the non-sectionalization scheme without performing subsequent sectionalization transformation and multi-objective optimization steps when the ice melting of the whole line without sectionalization meets the output limit of the ice melting device and the line operation requirements.

[0051] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the overhead ground wire non-stop power ice melting automatic sectionalization method described in the above technical solution.

[0052] The present invention also provides an electronic device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the overhead ground wire live deicing automatic segmentation method as described in the above technical solution.

[0053] The beneficial effects of the present invention are as follows: The present invention can efficiently deice under live conditions while reducing manual repeated trial and error. First, it determines whether it is feasible to deice the entire line without segmentation. If not, it then performs segmented transformation and optimizes with "dual objectives", which can avoid unnecessary equipment investment or energy consumption waste caused by blind segmentation. The present invention uses an automated algorithm search to replace manual repeated calculations, effectively improving the efficiency and accuracy of engineering decision-making.

[0054] Furthermore, the present invention uses the ATP-EMTP platform to change the pulse trigger angle of the DC deicing device to simulate the electrical parameters under different deicing connection modes, and provides the simulation results to the multi-objective genetic algorithm. Using the mature electromagnetic transient software ATP-EMTP can more realistically present the transient / steady-state characteristics of long-distance lines during live deicing, accurately simulate the electrical characteristics of multi-microclimate sections and different segmented connection modes; by adjusting the pulse trigger angle, the deicing voltage, current, and power of the ground wire can be dynamically obtained, avoiding errors caused by simple theoretical formulas and reducing the uncertainty of manual estimation, thereby enabling the multi-objective optimization to obtain a more accurate and feasible segmentation scheme.

[0055] Furthermore, the present invention specifies that in the process of modeling and analysis, detailed structural parameters such as different tower types, conductor and ground wire specifications, and resistance need to be considered. The more refined structural parameters make the simulation results closer to the actual line operation conditions, ensuring the applicability of the segmented transformation scheme to complex terrains and different types of towers; it is applicable to extra-high voltage long-distance lines that "span various microclimate regions" and "multiple tower type combinations", meeting the requirements of real projects.

[0056] Furthermore, the present invention introduces environmental variables in the simulation and evaluation stage, including ice thickness, wind speed, and temperature, so as to more accurately calculate deicing energy consumption, deicing current, etc., improve the simulation accuracy of actual working conditions, and fully consider the influence of climate factors on ice distribution and melting efficiency; it adapts to the scenario of large-span lines spanning multiple weather zones, enhancing the applicability of the model to multi-microclimate sections.

[0057] Furthermore, the present invention clarifies the physical and safety limitation conditions by setting constraint conditions, avoiding the generation of unrealistic or potentially dangerous segmentation schemes; taking into account the lightning protection performance and line safety, avoiding a significant decline in the shielding function of the ground wire due to excessive segmented transformation; ensuring that the finally output segmentation strategy is feasible and reliable in actual projects.

[0058] Furthermore, the present invention introduces automated search and optimization to replace manual repeated trials; the multi-objective genetic algorithm itself can simultaneously consider multiple conflicting objectives (number of segments and energy consumption), enabling the algorithm to find a better comprehensive compromise solution; by non-dominated sorting, excellent solutions are retained and dominated solutions are discarded, significantly improving the optimization efficiency.

[0059] Furthermore, the iterative process of the multi-objective genetic algorithm executed by the present invention includes two-stage mutation rates and elitism retention. In the early stage, the search diversity is increased to avoid local optima, and in the later stage, the mutation is reduced to ensure the convergence accuracy; by elitism retention and crowding distance strategy, excellent solutions are guaranteed to be retained, maintaining the uniform distribution of the solution set in the objective space, which can search the segment space more comprehensively and improve the quality and robustness of the finally obtained segment scheme.

[0060] Furthermore, when selecting the final segment scheme, the present invention comprehensively considers the operation cost, capacity limit, number of segments and energy consumption, making the decision-making process flexible and diverse, and selecting the most suitable segment scheme according to different engineering requirements (cost-oriented or equipment-oriented); ensuring a reasonable balance among the number of segments, energy consumption and cost among multiple feasible solutions, which is suitable for the refined management of large-scale UHV lines; improving the operability and economic benefits of the scheme during actual implementation.

[0061] Furthermore, the present invention provides a progressive candidate scheme generation strategy, which prevents the initial population from being overly random and avoids generating a large number of redundant high-segment-number schemes at once; by adding "1" step by step to ensure diversity, and at the same time, when it is found that a small number of segments is not feasible, then try more segments, thereby reducing unnecessary high-segment switch inputs; better adapting to large lines and significantly improving the search efficiency and feasibility when combined with the multi-objective genetic algorithm.

[0062] Furthermore, when it is found through simulation evaluation that the output capacity of the device is sufficient to support the ice melting of the whole line, the "non-segmentation" scheme is directly adopted, omitting other steps, saving the transformation cost to the greatest extent and avoiding ground wire segmentation under non-necessary circumstances; keeping the ground wire intact and grounded has the best lightning protection performance under normal conditions, and can also reduce waste of manpower and material resources. In reality, if the whole-line DC ice melting is already sufficient, there is no need for cumbersome segmentation transformation, further improving the practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a schematic diagram of the overall method of the present invention;

[0064] Figure 2 It is a flow chart of the multi-objective genetic algorithm of the embodiment;

[0065] Figure 3 It is a schematic diagram of the encoding of the candidate segment scheme of the embodiment;

[0066] Figure 4Simulation model diagram of the ground wire segmented de-icing wiring circuit for the embodiment;

[0067] Figure 5 Schematic diagram of the initial population scheme for the embodiment;

[0068] Figure 6 Schematic diagram for solving the Pareto solution set of the initial population scheme for the embodiment;

[0069] Figure 7 Schematic diagram of the calculation results of all candidate schemes for the embodiment. Detailed implementation manners

[0070] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, which is convenient for clearly understanding the present invention, but they do not constitute a limitation to the present invention.

[0071] Embodiment 1

[0072] As Figure 1 shown, the technical solution adopted by the present invention is: an automatic segmented method for de-icing an overhead ground wire without power interruption, including the following steps:

[0073] Based on the environmental conditions and structural parameters of the transmission line, a de-icing evaluation model is constructed to calculate the voltage, current and energy consumption required for de-icing the ground wire without segmentation, and to determine whether the de-icing device has sufficient output capacity;

[0074] If the output capacity of the device is not sufficient to cover the de-icing of the entire ground wire, multiple segmented configurations are constructed at the segmentable towers to generate a candidate segmented scheme set, and the electrical parameters of each candidate segmented scheme are calculated;

[0075] Taking the minimization of the number of segments and the minimization of de-icing energy consumption as dual objectives, a multi-objective genetic algorithm is used to iteratively evolve the candidate segmented scheme set, and a Pareto optimal solution set that meets the de-icing device capacity and line safety constraints is screened out;

[0076] Based on the preset engineering criteria, the optimal segmented configuration that meets the requirements of de-icing without power interruption is selected from the Pareto optimal solution set and output.

[0077] Specifically, the structural parameters include tower types and parameters and conductor and ground wire parameters. The environmental condition parameters include ice thickness, wind speed and ambient temperature. The de-icing evaluation model is built on the ATP-EMTP platform. Based on the above structural parameters and environmental condition parameters, by changing the pulse trigger angle of the DC de-icing device, the change of the ground wire de-icing voltage under different de-icing wiring methods is simulated, and accordingly the de-icing current and device output capacity corresponding to each candidate segmented scheme are calculated; the above calculation results are used for the multi-objective genetic algorithm calculation.

[0078] Preferably, in this embodiment, a certain UHV transmission line is taken as the background, and simulation evaluations are respectively carried out for the two cases of non-segmented and segmented de-icing, and the optimal segmentation scheme is selected according to the optimization results of the multi-objective genetic algorithm, so as to balance the technical requirements and equipment capacity limitations during the live-line de-icing process of the line. The specific steps are as follows:

[0079] 1. First, the structural parameters and environmental condition parameters of the target line are collected.

[0080] The structural parameters include: tower type and main structural dimensions, conductor and ground wire materials and electrical parameters (such as cross-sectional area, resistance, inductance, capacitance, etc.), positions of strain towers or angle towers, etc.

[0081] The environmental condition parameters include: ice thickness, wind speed, ambient temperature in different micro-meteorological sections, and geographical information (altitude, terrain, etc.) of the area where the line is located.

[0082] 2. These information will be input as the basic data for modeling, and the subsequent multi-objective genetic algorithm will also determine whether the segmentation strategy meets the actual operation requirements according to the above environmental and structural data.

[0083] After completing the collection of the basic line information, the structure and operation conditions of the line are preliminarily processed using MATLAB scripts, including: formatting the line data to form an input file that matches the ATP-EMTP simulation model; generating specific electrical network topology descriptions in MATLAB according to different candidate segmentation schemes (such as the number of segments, segmentation positions), including the connection methods of each segment of the ground wire to the DC de-icing device after being separated into several segments.

[0084] 3. Next, a de-icing evaluation model is built in ATP-EMTP. Specifically, it includes:

[0085] Line equivalent modeling: Each segment of the ground wire is represented by a distributed parameter line model or a π-type equivalent model. If it is in a multi-segment parallel or series mode, appropriate nodes also need to be connected in ATP-EMTP.

[0086] DC de-icing device model: Insert the equivalent circuit of the DC de-icing device in the simulation, including a rectifier / inverter unit with adjustable pulse trigger angle, and set the maximum output voltage and current.

[0087] Change the pulse trigger angle α: Control the trigger angle of the DC de-icing device by modifying the ATP-EMTP script or interface input to simulate the dynamic change of the ground wire de-icing voltage under different wiring methods. In this way, the mutual influence relationship of "device capacity - number of segments - line conditions" can be tested in the simulation environment.

[0088] After completing the ATP-EMTP modeling, a batch script runATP_G.bat is used to automatically run the simulation: by calling this bat script in the command line or script environment, multiple ATP-EMTP input files can be batch processed; loop simulations are performed for different sectional configurations (section points corresponding to different binary encodings), and the steady-state or transient ice-melting voltage, current distribution, device output power, energy consumption and other results under each sectional scheme are output respectively. Compared with manually opening the software one by one, this batch script significantly improves the efficiency of multi-scenario and multi-scheme comparison. Since the output file format of ATP-EMTP is often its own format (such as.PLO,.PL4), in this embodiment, the pl42mat tool is used for transcoding: converting the simulation results of ATP-EMTP (such as node potential, branch current, device power) into readable MATLAB data files (.mat). Import the electrical quantity data generated after transcoding in MATLAB, and then comprehensively evaluate each candidate sectional scheme, including whether the ice-melting voltage exceeds the limit, whether the device output capacity exceeds the standard, and the calculation of energy consumption between sections.

[0089] 4. After completing the extraction of the ice-melting electrical characteristics of each scheme, it can be passed to the multi-objective genetic algorithm. The core of the algorithm is as follows: for several binary encoding schemes under random or gradually increasing number of sections strategies, run the ATP-EMTP simulation respectively to obtain the fitness (number of sections, energy consumption) and constraint check results (capacity, line safety requirements). Retain the individuals that meet the capacity limit and have relatively excellent energy consumption and number of sections, derive new schemes through gene crossover and mutation, and call runATP_G.bat for batch simulation again;: comprehensively sort the new and old schemes according to the multi-objective genetic algorithm to obtain the Pareto optimal solution. Repeat the simulation and optimization multiple times until the algorithm reaches the preset number of iterations or convergence criterion, and finally obtain a series of sectional configurations that can meet the device capacity and take into account the least number of sections and the lowest energy consumption.

[0090] 5. Finally, the algorithm will output a batch of Pareto optimal sectional schemes, and each scheme corresponds to the electrical data verified by ATP-EMTP: such as the required voltage and current peaks, and the energy consumption of the entire line during the ice-melting process. According to the project priority (for example, emphasizing fewer sections or lower energy consumption), select the most suitable sectional configuration from the Pareto solutions and use it for actual line transformation and live ice-melting operations. If it is found that the entire line can meet the capacity without sectionalization, no sectional transformation is carried out, which further simplifies the operation and saves costs.

[0091] Specifically, when performing electrical evaluation and iterative evolution on the candidate sectional schemes, the following constraint conditions are imposed: the ice-melting voltage and current of each section of the ground wire shall not exceed the rated output of the ice-melting device; the sectional position is limited to the pre-selected sectionalizable towers; any candidate sectional scheme shall meet the line lightning protection and operation safety permission range.

[0092] Preferably, the specific indicators of the constraint conditions may include:

[0093] Voltage and current limits: such as the rated output of the de-icing device, de-icing duration, etc.;

[0094] Segmentable tower positions: list of strain towers and special-function towers, accessibility for safe maintenance;

[0095] Lightning protection and operation safety permits: lightning trip rate, impulse withstand voltage, power frequency withstand voltage, thermal stability, crossing safety, etc.

[0096] During the execution of the optimization algorithm, any candidate segmentation scheme needs to first check the above conditions. Once it exceeds the equipment capacity (voltage or current) or the lightning protection, mechanical, and thermal stability indicators are non-compliant, it is regarded as infeasible. The segmentation configuration that can finally enter the Pareto optimal solution set and be selected as the actual scheme must meet all the constraints, so as to balance efficiency, economy, and overall line safety and reliability during live-line de-icing.

[0097] Typical DC de-icing devices usually have a calibrated upper limit of ±X kV voltage, such as ±15kV or ±10kV. For any segment in the candidate segmentation scheme, if the ATP-EMTP simulation or the corresponding estimation results show that the actual de-icing voltage required for this segment exceeds the maximum output voltage of the de-icing device, then this scheme is regarded as infeasible.

[0098] The highest DC current value that the de-icing device can continuously provide, such as 2000A, 3000A, etc. If the de-icing current calculated for a certain segment in the simulation is greater than the rated value of the equipment, it means that the equipment cannot safely operate the de-icing of this segment, so this candidate scheme should also be excluded or determined to not meet the constraints.

[0099] Some de-icing equipment or lines will limit the de-icing duration. For example, if a long time and high current are required for de-icing a section of the line, it may cause the equipment to overheat or the line to overheat. At this time, it is necessary to calculate the heat distribution during the de-icing process in the simulation. If it exceeds the allowable continuous operation duration of the equipment, it will also be regarded as not meeting the requirements.

[0100] In line design, it is generally only allowed to carry out disconnection transformation on strain towers, terminal towers or special-function towers. It is necessary to record the numbers of these tower positions to form a list of segmentable tower positions;

[0101] If there are m segmentable towers in a line, the coding length of the candidate segmentation scheme is m. When generating the segmentation scheme, any point where segmentation is desired must correspond to this list, otherwise it will not be considered.

[0102] When implementing an actual project, the factor of "the location for sectional transformation should be convenient for construction and maintenance" should also be considered, such as the surrounding traffic conditions or terrain conditions. If the site environment of a certain tower position is extremely complex although it is a strain tower, it can also be excluded from the optional list.

[0103] For some tower poles where lightning protection devices have been installed or there are important crossing environments, it is also necessary to conduct an actual lightning strike risk assessment to ensure that the lightning protection requirements are still met after sectionalization. Set the minimum threshold for short-term operation based on the lightning withstand level specified in national or enterprise standards. For example, it is required that each section after sectional transformation can still maintain sufficient insulation margin during high lightning current impact and will not significantly increase the failure rate. If the original design of the line has a rated annual average lightning strike tripping rate, it is necessary to ensure that the tripping rate will not increase significantly (such as the increase does not exceed a certain tolerance) after installing the sectionalization. After ice melting is completed, the grounding continuity of the overhead ground wire should be restored as soon as possible to avoid weakening the shielding effect of the line due to excessive breakpoints during the high lightning season or the next weather change.

[0104] For the insulation device of the sectionalized line or the location of the sectional switch, the safety margin of power frequency and impulse withstand voltage should be met to prevent insulation breakdown during ice melting or in the fault state. For the sectionalized conductors and overhead ground wires, it is necessary to check their short-term thermal stability, tension and span to avoid additional mechanical stress or thermal stress caused by the new sectional configuration under large current. If the sectional point is in an important crossing area such as a river or a highway, it is necessary to additionally verify the critical sag and jump risk after sectionalization to avoid unexpected safety hazards of the line under extreme working conditions such as strong wind and ice coating superposition.

[0105] In the multi-objective genetic algorithm, lightning protection and the operating safety permission range can be incorporated in the form of "constraint functions": if a certain sectionalization scheme shows that its lightning strike tripping rate exceeds the standard or the impulse withstand voltage does not meet the standard in ATP-EMTP simulation or other lightning protection calculation modules, it is directly determined to be infeasible; if over-limit occurs in mechanical load and thermal stability calculations, it is also regarded as a violation of the constraint; a penalty mechanism can also be adopted to assign a high penalty value to the candidate scheme with a slight over-limit of safety constraints, making it difficult to survive in genetic evolution until the scheme that meets the line permission stands out.

[0106] Specifically, the multi-objective genetic algorithm encodes each candidate sectionalization scheme into a chromosome individual and forms a population, calculates two objective functions of the number of sections and ice melting energy consumption for each individual, and screens out the individuals that do not meet the device capacity and line safety requirements; obtains multiple Pareto fronts through non-dominated sorting, and continuously updates the population through crossover and mutation genetic operations in iterations until a convergent set of Pareto optimal solutions is obtained. The iterative process of the multi-objective genetic algorithm includes:

[0107] Calculate the fitness of randomly generated candidate sectionalization schemes and screen out feasible solutions;

[0108] Evolve the candidate individuals for several generations with the first mutation probability, and adopt selection, crossover, mutation, non-dominated sorting, and elitist retention strategies to generate corresponding offspring and merge them with the parent generation;

[0109] Select the Pareto solution with the largest crowding degree from the merged population as the parent generation for the next stage of iteration;

[0110] Continue genetic evolution for several generations with the second mutation probability less than the first mutation probability, and finally obtain the Pareto optimal solution set based on non-dominated sorting.

[0111] Preferably, as Figure 2 shown, when the present invention automatically optimizes the candidate sectioning scheme, a multi-objective genetic algorithm is adopted and the evolution is carried out in two stages: in the first stage, the solution space is quickly explored with a higher mutation rate, and in the second stage, it is finely converged with a lower mutation rate, and finally a Pareto optimal solution set that meets the safety constraints and takes into account the minimum number of sections and the lowest energy consumption is obtained. The main process is as follows:

[0112] First, binary encoding is performed on the ground wire sectioning scheme: the encoding length of each individual is the same as the number of sectionable towers, "1" indicates insulation sectioning from this tower, and "0" remains continuous. The positions of "1" are randomly set in the encoding to form a diverse set of initial candidate sectioning schemes.

[0113] A number of randomly generated binary-encoded individuals are grouped into an "initial population". This initial population covers various combinations of the number of sections and section positions to ensure a preliminary coverage and diversity of the solution space.

[0114] Call ATP-EMTP simulation or other power system models for each individual (sectioning scheme) to calculate its corresponding key parameters such as de-icing voltage, current, and device capacity, and perform fitness evaluation on the number of sections and de-icing energy consumption in MATLAB or other programs, so as to link the simulation results with the multi-objective genetic algorithm.

[0115] After obtaining the fitness (number of sections, energy consumption) of each initial individual and constraint verification (whether the voltage and current exceed the limit), classify the initial population: if an individual violates the capacity / safety constraint, it is eliminated, and the remaining feasible solutions enter the subsequent genetic operations.

[0116] In the first stage of iteration, according to the rank and crowding degree of non-dominated sorting, "select" the individuals in the initial population, that is, preferentially select the individuals on the Pareto front or with better performance as the parent generation; then perform crossover operations (such as single-point or multi-point crossover) on the parent generation to combine their encoding parts to generate offspring.

[0117] At the beginning of the first stage, to enhance search diversity and prevent premature convergence, a relatively high mutation probability (such as 0.4) is set. The mutation operation will randomly flip several bit positions with a certain probability, enabling the offspring individuals to have the possibility of new segmentation configurations, thereby significantly expanding the search scope.

[0118] Perform non-dominated sorting on the combined population of parents and offspring:

[0119] Level 1 (Pareto front) contains solutions that are not simultaneously superior to other individuals in both objectives; Level 2 contains better solutions dominated by Level 1; and so on.

[0120] At the same time, calculate the "crowding distance" to measure the sparsity of solutions in the objective space, ensuring a uniform distribution of the Pareto front.

[0121] Continue to iterate the population that has undergone non-dominated sorting and crowding distance evaluation for a total of 40 times (i.e., repeat "selection, crossover, mutation, non-dominated sorting" for several generations), and always maintain a relatively high mutation probability (0.4) during this process to fully explore potential high-quality solutions.

[0122] After each generation of iteration is completed, combine the parents and offspring into a larger mixed population, and then select the top several individuals (elite individuals) through non-dominated sorting as the next generation. This can prevent high-quality solutions from being discarded during evolution, also known as the "elitism strategy", which can improve the stability and convergence speed of the algorithm.

[0123] After 40 iterations with a high mutation rate, the algorithm obtains an offspring population and calculates the corresponding Pareto solution set through non-dominated sorting. At this time, the Pareto solution with the largest crowding distance (or the optimal solution according to other criteria) can be selected from this Pareto solution set as the new parent, thus entering the next stage.

[0124] Solutions with a larger crowding distance tend to be in the sparse region of the objective space and can better represent different trade-off relationships. Therefore, selecting them as parents enables subsequent iterations to search the objective space more evenly; it also takes into account retaining the most potential solutions discovered so far.

[0125] In the second stage, the parents also need to be "selected and crossed", but the selected parents come from the Pareto front with a larger crowding distance to ensure a relatively high initial quality. A lower mutation probability (such as 0.2) is set in the second stage to reduce large-scale random changes after initially finding good solutions, thereby finely searching the surrounding solutions and accelerating the convergence of the algorithm in the multi-objective space.

[0126] Through the same non-dominated sorting process, combine the parents and offspring, recalculate the levels and crowding distances, and retain the excellent solutions to enter the next generation.

[0127] In the second stage, the above process is repeated 100 times. Each time, the offspring encoding is converted into a segmented configuration, and the combined simulation platform (ATP-EMTP+MATLAB) is called to evaluate the de-icing voltage, current, as well as the number of segments and de-icing energy consumption. The fitness value is updated and the survival of the fittest is carried out. As the iteration progresses, the algorithm will gradually converge to a more accurate and evenly distributed Pareto front.

[0128] Finally, when 100 iterations are completed, the algorithm outputs the final Pareto optimal solution set. This solution set contains several segmented schemes, each with a different trade-off relationship between the number of segments and energy consumption, but all of which have met the constraints such as equipment capacity and line safety requirements.

[0129] In practical applications, engineers can further make decisions on this Pareto front. If they particularly value reducing the transformation cost, they can choose the solution with a relatively small number of segments; if they pay more attention to energy conservation, they can choose the solution with lower energy consumption; at the same time, they can also consider other operating factors such as lightning protection safety margin and construction accessibility, and finally determine the most suitable segmented configuration for implementing live-line de-icing.

[0130] In this two-stage multi-objective genetic algorithm process, the switching between the high mutation rate (0.4) and the low mutation rate (0.2) effectively balances the diversity of solutions and the convergence speed during the initial and later iterations; the application of non-dominated sorting and crowding degree indicators ensures the quality of multi-objective solutions; the elite retention strategy ensures that excellent solutions will not be accidentally discarded by the next generation. Combining the combined simulation platform to conduct electrical evaluations on candidate solutions enables the finally output Pareto solution set to not only perform well in the theoretical objective function but also meet the actual engineering requirements, truly achieving an efficient trade-off between the number of segments and de-icing energy consumption.

[0131] Specifically, when selecting the final segmented scheme from the Pareto optimal solution set, the line operation cost, equipment capacity limit, as well as the number of segments and energy consumption are comprehensively considered; if there are multiple Pareto solutions that meet the requirements, the solution with the largest crowding degree or the solution with the objective function value closest to the ideal point can be preferentially selected as the final output.

[0132] Preferably, when the multi-objective genetic algorithm finishes all iterations, a Pareto optimal solution set (Pareto Front) is obtained, where each segmented scheme achieves a certain trade-off in both "minimizing the number of segments" and "minimizing de-icing energy consumption", and meets the basic constraints of "de-icing device capacity limit" and "line operation safety permission". To select the optimal single scheme from this set and put it into practice, the present invention adopts the following multi-index comprehensive evaluation process:

[0133] First, all Pareto solutions output by the algorithm are archived one by one, and each solution is recorded including but not limited to: the number of segments (N), the ice melting energy consumption (W), the utilization rate of the rated device capacity (such as the percentage of the required voltage and current of the solution relative to the output of the device in simulation or evaluation), the retrofit cost / operation cost (if the number of segments is large, the additional equipment investment and maintenance cost will increase accordingly), the additional operation safety index (which can include lightning strike tolerance, electrical thermal stability or crossing safety), and the construction feasibility (if a certain segment point is located in a remote mountainous area or special terrain, and the construction difficulty coefficient is high, it also needs to be noted).

[0134] This can connect the Pareto solutions with the actual engineering requirements, facilitating subsequent evaluation and comparison.

[0135] Some Pareto solutions are feasible in terms of theoretical electrical parameters, but there may still be major potential hazards in actual construction, economy or safety redundancy. For example: the utilization rate of the current or voltage of the solution is too close to 100%, without enough safety margin; the segment position happens to be at a tower position that is inaccessible or where an insulated sectionalizing switch cannot be installed; in a high lightning risk area, a higher lightning protection level is required, but this solution fails to meet the corresponding standards. For this part of the solutions, the present invention will eliminate or lower the priority in the local engineering evaluation.

[0136] If there are still a large number of remaining feasible solutions, the present invention can perform multi-index quantification and weighting on them:

[0137] Set the evaluation index and its corresponding weight (the smaller the better), the ice melting energy consumption (the lower the better), the retrofit / operation cost (the lower the better), and the equipment capacity utilization rate (moderate is better, not too high)

[0138] After normalizing each index, calculate a comprehensive score or a weighted objective function. If this project is most sensitive to energy consumption, increase the weight of the ice melting energy consumption; if it focuses on the one-time equipment investment, increase the weight of the retrofit / operation cost; in this way, it can flexibly reflect the actual operation and maintenance needs. Finally, the few solutions ranked at the top can be selected in ascending order (or descending order, depending on the index design) as the "optimal choices" for the next comparison.

[0139] When several preferred candidate solutions still show similar comprehensive scores, the following common multi-objective decision-making criteria can be introduced:

[0140] The largest crowding distance: On the Pareto front, the larger the crowding distance value, the sparser and more independent the solution is around the target space, which can also be understood as the solution having a more flexible target balance;

[0141] The closest distance to the ideal point: Define the ideal point in advance and calculate the Euclidean distance or weighted distance. The smaller the distance, the more balanced the consideration of each target.

[0142] If there are multiple alternative solutions with similar performance, both of these two methods can provide clear and simple means of determination.

[0143] The present invention also supports submitting a small number of the optimal ranked solutions to power design or operation and maintenance experts for manual review after obtaining the algorithm candidate solutions: If a certain solution ranks the highest in terms of the number of segments and energy consumption, but it is not recommended for implementation due to terrain, approval procedures, or subsequent operation and maintenance difficulties; the expert can exclude this solution; If there are certain special safety requirements (such as across the strait, high lightning strike area, etc.), solutions with more safety redundancy may be preferentially retained.

[0144] After a complete set of comprehensive evaluation processes of "multi-index scoring + crowding degree / ideal point comparison + expert review", select one or several segmenting solutions that best fit the local engineering conditions as the final recommended solutions. If only one unique solution is required, select one from the optimal ranked ones; If the operator hopes to have multiple options, two or three high-scoring solutions can be retained as alternatives and flexibly switched under different meteorological or load conditions.

[0145] Specifically, the generation process of the candidate segmenting solution set includes: sequentially mapping the preset tower positions to each bit of the binary code. Initially, only one "1" is randomly set in the code to represent a single segment; If this initial solution does not meet the output requirements of the de-icing device, gradually increase the number of "1"s in the code and distribute them randomly among each tower position to form multiple candidate segmenting solutions with an increasing number of segments.

[0146] Preferably, as Figure 3 shown, in the present invention, to automatically obtain a ground wire segmenting transformation solution that meets the requirements of live-line de-icing, it is necessary to encode and screen possible segmenting combinations. This embodiment represents the segmenting solution in the form of binary code, and the specific steps are as follows:

[0147] 1. Preset tower mapping and determination of coding length

[0148] In the line design or operation and maintenance data, first determine the positions of the strain towers, angle towers, or special function towers that can be subjected to insulation transformation, and number them as 1, 2, 3,..., m.

[0149] Each number corresponds to one bit in the coding string; If this bit is 1, it means to disconnect the ground wire at the corresponding tower for segmentation, and if it is 0, it means not to transform at this place and keep the original state.

[0150] For some lines, if there are a total of 10 strain towers available for transformation, the coding length is set to 10.

[0151] 2. Generation of the initial solution

[0152] Single "1" random distribution: When generating the initial segmentation scheme, only randomly select 1 bit out of 10 bits and set it to 1, and the others are all 0, so as to form the basic configuration of "only one segmentation (two sections of ground wires)". For example, 0000100000.

[0153] Randomly generate N such codes to obtain a preliminary diversified set of "two-section segmentation" schemes.

[0154] Input each scheme into the subsequent simulation and judgment module (such as ATP-EMTP simulation + multi-objective genetic algorithm) to check whether it meets the safety requirements such as the output voltage and current of the de-icing device.

[0155] 3. Scheme evaluation and iteration

[0156] If a single segmentation scheme can meet the device output capacity, it means that two-section segmentation is sufficient to complete the live-line de-icing of the entire line; in this case, this scheme can be given priority and there is no need to continue increasing the number of segments.

[0157] If all single segmentation schemes cannot meet the requirements (for example, the de-icing voltage or current still exceeds the rated value of the equipment), more segments need to be added to shorten the length of a single segment and reduce the device pressure.

[0158] 4. Gradually increase the number of segments

[0159] Based on the initial (including 1 bit "1"), gradually increase the number of "1"s in the code: 2 bits "1" represent three-section segmentation, 3 bits "1" represent four-section segmentation, and so on.

[0160] For each additional bit "1", its specific position in the code is randomly distributed among all segmentable tower positions. For example, when increasing to 3 bits "1", various combinations such as 0100100000 and 0010010000 may appear.

[0161] 5. Upper limit of the number of segments and complete candidate scheme set

[0162] In general projects, to balance lightning protection and construction economy, the number of line segments will not increase indefinitely, and the upper limit is usually set to 5 or 6 segments. In other words, at most 4 - 5 bits of "1" are allowed in the code.

[0163] When the number of "1"s reaches the established upper limit, all feasible segment combinations of this line are formed. In this embodiment, if the maximum allowable number of segments is set to 5 segments (that is, at most 4 bits "1"), various random coding schemes from 2 segments to 5 segments can be formed, constituting a complete initial candidate scheme set.

[0164] 6. Subsequent calculations and multi-objective optimization

[0165] For all the generated encodings (from 1 bit '1' to 4 bits '1'), ice melting simulation and fitness evaluation (such as the number of segments, ice melting energy consumption, and device capacity compliance) are carried out one by one, and crossover and mutation are performed during the iteration process of the multi-objective genetic algorithm.

[0166] If some schemes with a high number of segments and low energy consumption perform excellently in the evaluation, the algorithm retains them as parents for further evolution; if some schemes with a low number of segments can also meet the capacity limit, they may become the optimal solutions to reduce the transformation cost.

[0167] Through this way of segment generation that combines 'from less to more' and 'random distribution', the feasible transformation schemes of the line can be fully explored, improving the efficiency and comprehensiveness of finally finding the optimal configuration.

[0168] As Figure 3 shown, the encoding 0000100000 with only 1 bit '1' initially represents disconnecting at the 5th strain tower of the line, forming two segments. If the ice melting device still exceeds the rated voltage output range after evaluation, the number of '1's is increased to 2 bits. For example, 0010000100, which means disconnecting at the 4th and 8th towers respectively, forming three segments.

[0169] If the three-segment scheme still doesn't work, then increase it to three bits '1', randomly distributed among 10 bits, forming four segments; and so on, until reaching a maximum of 4 or 5 bits '1', or until a feasible scheme that meets the ice melting requirements is found.

[0170] Specifically, if the ice melting of the whole line without segmentation can meet the output limit of the ice melting device and the line operation requirements, there is no need to perform subsequent segmentation transformation and multi-objective optimization steps, and directly output the non-segmented scheme.

[0171] Preferably, in the automatic segmentation method for live-line ice melting of overhead ground wires described in the present invention, first, through parameters such as the ice melting simulation model and the capacity of the ice melting device, a feasibility assessment of ice melting is carried out by applying a direct current to the whole unsegmented ground wire. When it is found that this 'non-segmented' situation itself can complete ice melting within the rated output range of the device, this simplified scheme can be directly adopted without performing subsequent segmentation transformation steps. The main process is described as follows:

[0172] After obtaining the line environment parameters (such as ice thickness, temperature, wind speed, etc.) and structural parameters (such as tower distribution, ground wire type), an equivalent circuit model of the whole line without segmentation is established in the ATP-EMTP or similar electromagnetic transient / steady-state simulation platform.

[0173] Set the maximum output voltage and maximum output current of the ice melting device. Apply a direct current sufficient to melt the ice in the simulation, and monitor the actual voltage and current levels at the end of the line segment and intermediate key points.

[0174] If the overall line length is short or the performance of the de-icing device is high enough, it may be directly concluded that the entire line can be covered for de-icing within the rated output.

[0175] The de-icing voltage and current obtained through simulation should be less than the maximum output voltage and maximum output current of the de-icing device.

[0176] At the same time, check whether the electrical safety indicators of the line, such as induced voltage, operating temperature, power frequency withstand voltage, and impulse withstand voltage, are within the permitted range.

[0177] If these indicators are all within the equipment and operation safety requirements, it means that the overall line non-segmented scheme is sufficient to complete live de-icing under the current working conditions.

[0178] After confirming that non-segmented de-icing meets the capacity and safety requirements, there is no need to additionally disconnect the ground wire and perform the optimization process of the multi-objective genetic algorithm.

[0179] This can save the investment and manpower for adding sectional insulation devices, switches, and commissioning on the ground wire. At the same time, the line retains the original continuous grounding lightning protection effect; during the later operation and maintenance process, the operation is also simpler.

[0180] This embodiment takes the de-icing of the overhead ground wire of a certain UHV DC transmission line as an example. First, according to the parameters and boundary conditions of the overhead ground wire of the ±800kV transmission line, the de-icing current under different ice coating thicknesses and de-icing distances is determined. Secondly, a live de-icing process model is built in the electromagnetic transient software ATP-EMTP. The model is shown in Figure 4 。The output voltages of the device required for different de-icing distances and ice coating thicknesses are respectively simulated and calculated. The de-icing range limit under different external environments and different wiring methods can be deduced from the output voltage of the de-icing device, so as to automatically determine the sectional insulation design scheme. Finally, according to some line cases, the effectiveness of the model and conclusion is verified.

[0181] This embodiment selects a section of line XX. This section of line has a total of 100 towers, and the line length is 56.21 km. (The de-icing synchronous wind speed is 5 m / s, and the ambient temperature is -7°C). The commonly used de-icing device model for this line is: De-Icer-2000ADC / 10KVAC-M. The detailed parameters are shown in Table 1.

[0182] Table 1 DC de-icing device parameters

[0183]

[0184] By segmentally insulating the overhead ground wire, it is indeed possible to reduce the output pressure of the de-icing device to varying degrees, enabling the one-time completion of line de-icing. The more segments there are, the lower the requirements for the de-icing device. However, this will increase the investment in de-icing connection equipment and power consumption. For the above-mentioned ±800kV line, a segmented insulation scheme design is carried out. The initial population results calculated after 40 iterations are as Figure 5 shown.

[0185] When solving the Pareto solution set of the initial scheme, the calculation results are as Figure 6 shown. The Pareto solutions calculated based on the initial population are Schemes 8, 14, 12, 18, 7, 10, and 24 respectively. The number of segments and de-icing power consumption corresponding to each scheme are shown in Table 2.

[0186] Table 2 Fitness corresponding to the initial Pareto solution set

[0187]

[0188]

[0189] According to the data in Table 3, in order to ensure a reduction in the number of de-icing devices, for Schemes 7, 10, and 24, the required number of devices is too large, so they are discarded. Among all the remaining segmented schemes, Schemes 8, 14, 12, and 18 can meet the operation requirements. By calculating the crowding degree, it can be seen that the crowding degree corresponding to Scheme 18 is the largest, and the power consumption is also the minimum among the remaining schemes, which can meet the operation requirements of the actual line. Therefore, Scheme 18 with a crowding degree of 0.773357 is selected as the optimal solution. Then, taking Scheme 18 as the parent generation for iterative calculation, the calculation results after re-looping 100 times are as Figure 7 shown. The number of segments and de-icing power consumption corresponding to the optimal scheme calculated after 100 iterations are shown in Table 3.

[0190] Table 3 Fitness corresponding to the initial Pareto solution set

[0191]

[0192]

[0193] As can be seen from the data in Table 3, after 100 generations of offspring iteration, among all Pareto solution sets, the solutions on the Pareto front are selected for analysis and comparison. Among Solutions 61, 70, and 113, four-section de-icing is adopted. The de-icing power consumption of Solution 113 is the smallest, which is 7855 kW·h. The total de-icing current is 1557.15 A, not exceeding the output de-icing current of 2000 A of the de-icing device. Among Solutions 124, 131, 65, and 42, five-section de-icing of the line is adopted, and its minimum power consumption is 7551 kW·h, slightly smaller than that of the four-section de-icing. Among Solutions 106, 54, and 98, six-section de-icing is adopted, and its minimum power consumption is 7523 kW·h, slightly smaller than that of Solution 113.

[0194] In summary, adopting Solution 113 for line de-icing is the most suitable. The number of sections is the smallest among other offspring solutions on the Pareto front, and the increase in power consumption is not significant, reducing the investment in sectional equipment and ensuring economy while completing de-icing.

[0195] Embodiment 2

[0196] The present invention also provides an automatic sectionalized system for live de-icing of overhead ground wires, including:

[0197] A de-icing evaluation module, configured to construct a de-icing evaluation model based on the environmental conditions and structural parameters of the transmission line, calculate the voltage, current, and energy consumption required for ground wire de-icing without sectionalization, and determine whether the de-icing device has sufficient output capacity;

[0198] A candidate sectionalized solution generation module, configured to, if the output capacity of the device is insufficient to cover the de-icing of the entire ground wire, construct multiple sectionalized configurations at sectionalizable towers, generate a set of candidate sectionalized solutions, and calculate the electrical parameters for each candidate sectionalized solution;

[0199] A candidate sectionalized solution screening module, configured to use a multi-objective genetic algorithm to iteratively evolve the set of candidate sectionalized solutions with the dual objectives of minimizing the number of sections and minimizing the de-icing energy consumption, and screen out the Pareto optimal solution set that meets the de-icing device capacity and line safety constraints;

[0200] A candidate sectionalized solution output module, configured to select the optimal sectionalized configuration that meets the requirements of live de-icing from the Pareto optimal solution set based on a preset engineering criterion and output it.

[0201] Specifically, the de-icing evaluation model is built on the ATP-EMTP platform. By changing the pulse trigger angle of the DC de-icing device, the change in the ground wire de-icing voltage under different de-icing connection modes is simulated, and based on this, the de-icing current and device output capacity corresponding to each candidate sectionalized solution are calculated; the above calculation results are used for multi-objective genetic algorithm calculation.

[0202] Specifically, the structural parameters include tower types and parameters, and conductor and ground wire parameters.

[0203] Specifically, the environmental condition parameters include ice coating thickness, wind speed, and environmental temperature.

[0204] Specifically, when the candidate sectioning scheme screening module conducts electrical evaluation and iterative evolution on the candidate sectioning schemes, the following constraint conditions are imposed:

[0205] The ice melting voltage and current of each section of the ground wire shall not exceed the rated output of the ice melting device;

[0206] The sectioning positions are limited to the pre-selected towers that can be sectioned;

[0207] Any candidate sectioning scheme shall meet the lightning protection and operation safety permission range of the line.

[0208] Specifically, the multi-objective genetic algorithm encodes each candidate sectioning scheme into a chromosome individual and forms a population, calculates two objective functions of the number of sections and ice melting energy consumption for each individual, and eliminates the individuals that do not meet the device capacity and line safety requirements; multiple Pareto fronts are obtained through non-dominated sorting, and the population is continuously updated during iteration by combining genetic operations such as crossover and mutation until a converged set of Pareto optimal solutions is obtained.

[0209] Specifically, the iterative process of the multi-objective genetic algorithm includes:

[0210] Calculate the fitness of the randomly generated candidate sectioning schemes and screen the feasible solutions;

[0211] Evolve the candidate individuals for several generations with a first mutation probability, adopt selection, crossover, mutation, non-dominated sorting, and elitist retention strategies to generate corresponding offspring and merge them with the parent generation;

[0212] Select the Pareto solution with the largest crowding degree from the merged population as the parent generation for the next stage of iteration;

[0213] Continue genetic evolution for several generations with a second mutation probability less than the first mutation probability, and finally obtain the set of Pareto optimal solutions based on non-dominated sorting.

[0214] Specifically, when the candidate sectioning scheme screening module selects the final sectioning scheme from the set of Pareto optimal solutions, it comprehensively considers the line operation cost, equipment capacity limit, and the number of sections and energy consumption; if there are multiple Pareto solutions that meet the requirements, the solution with the largest crowding degree or the solution with the objective function value closest to the ideal point can be preferentially selected as the final output.

[0215] Specifically, the candidate segmentation scheme generation module maps the preset tower positions to each bit of the binary code. Initially, only one "1" is randomly set in the code to represent a single segmentation; if this initial scheme does not meet the output requirements of the de-icing device, the number of "1"s in the code is gradually increased and distributed randomly among each tower position to form multiple candidate segmentation schemes with an increasing number of segments.

[0216] Embodiment 3

[0217] The present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the overhead ground wire live-line de-icing automatic segmentation method described in the above technical solution.

[0218] Embodiment 4

[0219] The present invention provides an electronic device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the overhead ground wire live-line de-icing automatic segmentation method described in the above technical solution.

[0220] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0221] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the specified functions in Figure 1 one or more flows or multiple flows and / or blocks Figure 1 one or more blocks or multiple blocks.

[0222] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including instruction means, and the instruction means implements the functions in the processFigure 1 one process or multiple processes and / or blocks Figure 1 the functions specified in one block or multiple blocks.

[0223] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide for implementing the process Figure 1 one process or multiple processes and / or blocks Figure 1 the steps of the functions specified in one block or multiple blocks.

Claims

1. An automatic segmentation method for melting ice on overhead ground wires without power outage, characterized in that: The following steps are involved: Based on the environmental conditions and structural parameters of the transmission line, an ice melting assessment model is constructed to calculate the voltage, current and energy consumption required for ground line ice melting without segmentation, and to determine whether the ice melting device has sufficient output capacity; If the output capacity of the ice melting device is insufficient to cover the entire ground line to melt ice, multiple segment configurations are constructed at the segmentable tower, a set of candidate segment schemes is generated, and electrical parameters are calculated for each candidate segment scheme; With the dual objectives of minimizing the number of segments and minimizing the ice-melting energy consumption, a multi-objective genetic algorithm is used to iteratively evolve the candidate segmentation scheme set, and select the Pareto optimal solution set that meets the ice-melting device capacity and line safety constraints; Based on the preset engineering criteria, the optimal segment configuration that meets the requirement of non-stop ice melting is selected from the Pareto optimal solution set and outputted.

2. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 1 is characterized in that: The ice-melting evaluation model is built in the ATP-EMTP platform. By changing the pulse trigger angle of the DC ice-melting device to simulate the change of the ground wire ice-melting voltage under different ice-melting wiring methods, the ice-melting current and device output capacity corresponding to each candidate segmentation scheme are calculated accordingly; the above calculation results are used for multi-objective genetic algorithm calculation.

3. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 1 is characterized in that: The structural parameters include tower type and parameters and ground wire parameters.

4. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 1 is characterized in that: The environmental condition parameters include ice thickness, wind speed and ambient temperature.

5. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 1 is characterized in that: When performing electrical evaluation and iterative evolution of candidate segmentation schemes, the following constraints are imposed: The ice-melting voltage and ice-melting current of each section of ground wire shall not exceed the rated output of the ice-melting device; The segmentation locations are limited to pre-selected segmentable towers; Any candidate segmentation scheme must meet the line lightning protection and operation safety permit range.

6. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 1, characterized in that: The multi-objective genetic algorithm encodes each candidate segmentation scheme into a chromosome individual and forms a population, calculates two objective functions of segmentation quantity and ice-melting energy consumption for each individual, and screens out individuals that do not meet the device capacity and line safety requirements; obtains multiple Pareto fronts through non-dominated sorting, and continuously updates the population in iterations in combination with crossover and mutation genetic operations until a converged Pareto optimal solution set is obtained.

7. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 6, characterized in that: The iterative process of the multi-objective genetic algorithm includes: Calculate the fitness of randomly generated candidate segmentation solutions and screen feasible solutions; The candidate individuals are evolved for several generations with the first mutation probability, and the corresponding offspring are generated and merged with the parent generation using selection, crossover, mutation, non-dominated sorting and elite retention strategies; Select the Pareto solution with the largest crowding degree from the merged population as the parent of the next iteration; The genetic evolution is continued for several generations with a second mutation probability smaller than the first mutation probability, and finally a Pareto optimal solution set is obtained based on non-dominated sorting.

8. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 6 is characterized in that: When selecting the final segmentation scheme from the Pareto optimal solution set, the line operation cost, equipment capacity limitation, number of segments and energy consumption are comprehensively considered; if there are multiple Pareto solutions that meet the requirements, the solution with the largest congestion or the solution with the objective function value closest to the ideal point is preferentially selected as the final output.

9. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 1, characterized in that: The generation process of the candidate segmentation scheme set includes: mapping the preset tower positions to each bit of the binary code in sequence, and initially only randomly setting a "1" in the code to represent a single segment; if the initial scheme does not meet the output requirements of the ice melting device, then gradually increase the number of "1" in the code and distribute it randomly to each tower position to form a variety of candidate segmentation schemes with a gradually increasing number of segments.

10. The method for automatically segmenting an overhead ground wire without power outage and ice melting according to claim 1, characterized in that: If the output limit of the ice melting device and the line operation requirements can be met by melting ice without segmentation on the entire line, there is no need to perform subsequent segmentation transformation and multi-objective optimization steps, and a non-segmentation plan can be directly output.

11. An automatic segmentation system for melting ice without power outage for overhead ground wires, characterized in that: include, An ice melting assessment module is used to construct an ice melting assessment model based on the environmental conditions and structural parameters of the transmission line, calculate the voltage, current and energy consumption required for ground line ice melting without segmentation, and determine whether the ice melting device has sufficient output capacity; A candidate segmentation scheme generation module is used to construct multiple segmentation configurations at segmentable towers if the output capacity of the device is insufficient to cover the entire ground line ice melting, generate a set of candidate segmentation schemes, and calculate electrical parameters for each candidate segmentation scheme; A candidate segmentation scheme screening module is used to iteratively evolve the candidate segmentation scheme set using a multi-objective genetic algorithm with dual objectives of minimizing the number of segments and minimizing ice-melting energy consumption, and screen out a Pareto optimal solution set that meets the ice-melting device capacity and line safety constraints; The candidate segmentation scheme output module is used to select and output the optimal segmentation configuration that meets the requirements of non-power-off ice melting from the Pareto optimal solution set based on preset engineering criteria.

12. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 11, characterized in that: The ice-melting evaluation model is built in the ATP-EMTP platform. By changing the pulse trigger angle of the DC ice-melting device to simulate the change of the ground wire ice-melting voltage under different ice-melting wiring methods, the ice-melting current and device output capacity corresponding to each candidate segmentation scheme are calculated accordingly; the above calculation results are used for multi-objective genetic algorithm calculation.

13. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 11, characterized in that: The structural parameters include tower type and parameters and ground wire parameters.

14. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 1 is characterized by: The environmental condition parameters include ice thickness, wind speed and ambient temperature.

15. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 1 is characterized by: The candidate segmentation scheme screening module imposes the following constraints when performing electrical evaluation and iterative evolution on the candidate segmentation schemes: The ice-melting voltage and ice-melting current of each section of ground wire shall not exceed the rated output of the ice-melting device; The segmentation locations are limited to pre-selected segmentable towers; Any candidate segmentation scheme must meet the line lightning protection and operation safety permit range.

16. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 11, characterized in that: The multi-objective genetic algorithm encodes each candidate segmentation scheme into a chromosome individual and forms a population, calculates two objective functions of segmentation quantity and ice-melting energy consumption for each individual, and screens out individuals that do not meet the device capacity and line safety requirements; obtains multiple Pareto fronts through non-dominated sorting, and continuously updates the population in iterations in combination with crossover and mutation genetic operations until a converged Pareto optimal solution set is obtained.

17. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 16, characterized in that: The iterative process of the multi-objective genetic algorithm includes: Calculate the fitness of randomly generated candidate segmentation solutions and screen feasible solutions; The candidate individuals are evolved for several generations with the first mutation probability, and the corresponding offspring are generated and merged with the parent generation using selection, crossover, mutation, non-dominated sorting and elite retention strategies; Select the Pareto solution with the largest crowding degree from the merged population as the parent of the next iteration; The genetic evolution is continued for several generations with a second mutation probability smaller than the first mutation probability, and finally a Pareto optimal solution set is obtained based on non-dominated sorting.

18. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 16, characterized in that: When the candidate segmentation scheme screening module selects the final segmentation scheme from the Pareto optimal solution set, the line operation cost, equipment capacity limitation, number of segments and energy consumption are comprehensively considered; if there are multiple Pareto solutions that meet the requirements, the solution with the largest congestion or the solution with the objective function value closest to the ideal point can be preferentially selected as the final output.

19. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 11, characterized in that: The candidate segmentation scheme generation module maps the preset tower positions to each bit of the binary code in sequence. Initially, only one "1" is randomly set in the code to represent a single segment. If the initial scheme does not meet the output requirements of the ice melting device, the number of "1" in the code is gradually increased and randomly distributed to each tower position to form multiple candidate segmentation schemes with a gradually increasing number of segments.

20. The automatic segmentation system for melting ice without power outage of overhead ground wire according to claim 11, characterized in that: The ice melting assessment module is also used for directly outputting a non-segmented solution without executing subsequent segmented transformation and multi-objective optimization steps when the entire line can be ice-melted without segmentation, that is, the output limit of the ice melting device and the line operation requirements are met.

21. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for automatically segmenting the overhead ground wire without power outage and ice melting as described in any one of claims 1 to 10 is implemented.

22. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the automatic segmentation method for melting ice without power outage of the overhead ground wire as described in any one of claims 1-10 by executing the computer instructions.

Images