Method for calculating optimal input power of rapid temperature rise of transformer

Abstract

The invention discloses a method for calculating the optimal input power of rapid temperature rise of a transformer, and belongs to the technical field of temperature rise characteristics of transformers, and the method comprises the steps: constructing a transformer thermal circuit model, carrying out the fitness calculation of parameters in the thermal circuit model, and screening out excellent thermal circuit parameter individuals; and sequentially carrying out interlace operation and uniform mutation operation on the excellent thermal circuit parameter individuals, carrying out fitness calculation on the mutated thermal circuit parameter individuals after mutation until a search termination condition is met to obtain the optimal input power of the rapid temperature rise of the transformer, and obtaining the optimal input power of the rapid temperature rise of the transformer based on a natural selection principle. According to the method, the optimal thermal circuit parameter is searched in a self-adaptive probability search mode, a large number of assumptions do not need to be carried out before calculation, the calculation efficiency of identifying the optimal thermal circuit parameter is improved, and therefore the total loss power value applied in the early stage of temperature rise can be accurately calculated, the temperature rise efficiency is improved, and the temperature rise time is shortened.

Description

Technical Field

[0001] The invention relates to the technical field of transformer temperature rise characteristics, and in particular to a method for calculating an optimal input power for rapid temperature rise of a transformer. Background Art

[0002] With the gradual acceleration of urbanization and industrialization, the demand for electricity in all walks of life has increased significantly, and the power industry has also achieved considerable development. The input distribution transformer is the basic equipment of the power system. Its working conditions and quality directly affect the operation of the power system. In the actual working process, the distribution transformer will produce corresponding losses and directly affect the temperature of the insulating medium. If the temperature exceeds a certain standard, it will directly affect the normal operation and service life of the distribution transformer. As an important means of assessing transformers, the temperature rise test plays an important role in judging the quality of transformers. During the temperature rise experiment, the calculation or identification of the thermal circuit model parameters of the distribution transformer is a very inevitable process, because it is related to the calculation of the temperature rise time constant. The thermal circuit model of the transformer is an important tool for analyzing its temperature rise characteristics, and accurate model parameters are the key to ensuring the accuracy of the model. Based on accurate thermal circuit model parameters, it ultimately leads to the determination of the rapid temperature rise input power of the distribution transformer, thereby shortening the temperature rise time without affecting the insulation performance of the transformer. However, the traditional method for identifying the thermal circuit model parameters of the transformer can only rely on experimental experience. In the initial stage of the temperature rise of the distribution transformer, a loss power within a certain range higher than the total loss is given, thereby improving the efficiency in the early stage of the temperature rise and shortening the temperature rise time. The accuracy is not high.

[0003] Chinese patent, publication number: CN114462217A, publication date: May 10, 2022, discloses a method for calculating the temperature rise of a medium-frequency transformer based on an improved thermal network model, by dividing the transformer into thermal blocks; then establishing a transformer thermal network model; calculating the thermal resistance of each part; writing the node temperature equation; and finally calculating the node temperature. The magnetic core column is divided into an air-surrounded part and a winding-surrounded part, which can more accurately obtain the transformer temperature field distribution and improve the accuracy of the temperature rise calculation, providing theoretical guidance for the transformer heat dissipation design. However, it is through assumptions and simplifications of parameters, which may lead to inconsistencies with the actual situation, resulting in inaccurate calculation results. Summary of the invention

[0004] In view of the problem that the prior art usually relies on a large number of assumptions and simplifications when calculating the optimal input power of a transformer, resulting in a result that is inconsistent with the actual situation and thus causing a low accuracy of the calculation result, the present invention provides a method for calculating the optimal input power for rapid temperature rise of a transformer. By constructing a thermal circuit model of the transformer, the fitness of the parameters in the thermal circuit model is calculated and excellent thermal circuit parameter individuals are screened out, and crossover operations and uniform mutation operations are performed on the excellent thermal circuit parameter individuals in turn, and the fitness of the mutated thermal circuit parameter individuals after mutation is calculated until the search termination condition is met to obtain the optimal input power for rapid temperature rise of the transformer. Based on the principle of natural selection, an adaptive probabilistic search method is used to find the optimal thermal circuit parameters. There is no need to make a large number of assumptions before calculation, which improves the calculation efficiency of identifying the optimal thermal circuit parameters, so that the total power loss value applied in the early stage of temperature rise can be accurately calculated, thereby improving the temperature rise efficiency and shortening the temperature rise time.

[0005] In a first aspect, a technical solution provided in an embodiment of the present invention is: a method for calculating the optimal input power for rapid temperature rise of a transformer, comprising the following steps: S1. construct a transformer thermal circuit model, and construct a thermal circuit equation group based on the transformer thermal circuit model; S2, initializing the condition parameters, and randomly generating the thermal circuit parameters in the thermal circuit equation group to obtain the initial population of the thermal circuit; S3, based on the thermal circuit equation group, the fitness of the initial thermal circuit population is calculated in turn to obtain the thermal circuit fitness of each individual in the population; based on the thermal circuit fitness of each individual in the population, the initial thermal circuit population is screened to obtain individuals with excellent thermal circuit parameters; S4, performing a crossover operation on the excellent thermal circuit parameter individuals to obtain crossover thermal circuit parameter individuals; performing a uniform mutation operation on the crossover thermal circuit parameter individuals to obtain variant thermal circuit parameter individuals; performing fitness calculation on the variant thermal circuit parameter individuals based on the thermal circuit equation group to obtain the next generation optimal fitness; S5. If the next-generation best fitness does not meet the set search termination condition, the mutated thermal circuit parameter individual is iterated. If the next-generation best fitness meets the set search termination condition, the mutated thermal circuit parameter individual is used as the best thermal circuit parameter. S6. Substitute the optimal thermal circuit parameters into the thermal circuit equation group to obtain the optimal input power for rapid temperature rise of the transformer.

[0006] In this scheme, by constructing a transformer thermal circuit model, it is helpful to extract and analyze the parameters affecting the temperature rise of the transformer, so as to facilitate the solution of the constructed thermal circuit equations to improve the calculation efficiency; in order to reduce the amount of calculation, when the thermal circuit parameters are randomly generated, the search range of some parameters can be given according to the traditional estimation method, so that when searching for the optimal thermal circuit parameters in the subsequent search, the search time will not be too long due to the search path offset, thereby affecting the calculation speed; by calculating the fitness of the initial population of the thermal circuit, the initial population of the thermal circuit is screened according to the calculation results to obtain individuals with excellent thermal circuit parameters, which helps to retain individuals with high fitness and avoid the loss of excellent genes, thereby improving the global convergence and calculation efficiency during the calculation; by performing crossover and uniform mutation operations on individuals with excellent thermal circuit parameters, it is helpful to improve the diversity of the population and thus improve the comprehensiveness of the search, prevent the optimal thermal circuit parameters from being missed, and improve the calculation accuracy; by continuously iterating the population and continuously optimizing the thermal circuit parameters, there is no need to make a lot of assumptions before the calculation. The optimal thermal circuit parameters can be found through natural selection, and the calculation accuracy is higher, and the adaptability to different types of computing environments is stronger.

[0007] Preferably, in S1, constructing a transformer thermal circuit model, and constructing a thermal circuit equation group based on the transformer thermal circuit model, comprises the following steps: A transformer thermal circuit model is constructed, and the first equation is constructed based on the winding thermal capacity, the winding-oil thermal resistance and the transformer top oil temperature in the transformer thermal circuit model; The second equation is constructed based on the winding thermal capacity, winding-oil thermal resistance, transformer top oil temperature and transformer oil tank thermal capacity in the transformer thermal circuit model; A third equation is constructed based on the thermal resistance between oil and tank in the transformer thermal circuit model, tank temperature, transformer tank heat capacity, ambient temperature, and thermal resistance between the tank and the environment; The first, second and third equations are linked to construct the thermal circuit equations.

[0008] In this scheme, in order to facilitate the analysis of thermal circuit parameters, the internal structure of the transformer is presented in the form of a transformer thermal circuit model, and a thermal circuit equation group is constructed based on the transformer thermal circuit model. At the same time, due to the complex internal structure of the transformer, in order to facilitate the solution of the thermal circuit equation group, the transformer thermal circuit model can be disassembled into a winding hot spot temperature thermal circuit model, a top oil temperature thermal circuit model and an oil tank wall thermal circuit model. According to the above thermal circuit models, the first equation, the second equation and the third equation are respectively constructed, and the above equations are correlated to obtain a thermal circuit equation group. The relationship between the thermal circuit parameters of the transformer is limited, which facilitates the subsequent search for the optimal thermal circuit parameters, can reduce the amount of calculation and improve the calculation efficiency.

[0009] Preferably, the condition parameters include at least population size, crossover probability, mutation probability and evolutionary generation; The thermal circuit parameters at least include the thermal capacity of the winding, the thermal capacity of the oil, the thermal capacity of the transformer oil tank, and the thermal resistance between the winding and the oil, the thermal resistance between the oil and the oil tank, and the thermal resistance between the oil tank and the environment.

[0010] In this scheme, through the principle of natural selection, the natural selection mechanism of biological survival of the fittest and the mechanism of gene recombination and mutation in biological reproduction and evolution are used to conduct a global adaptive probability search for thermal circuit parameters. Before the search, some parameters need to be preset, including population size, crossover probability, mutation probability and evolutionary algebra. Among them, the population size will affect the effectiveness of the search algorithm. If the population size is too small, it cannot provide enough sampling points. If the population size is too large, it will increase the amount of calculation and make the convergence time too long. The crossover probability controls the frequency of the crossover operation. If the crossover probability is too large, the structure with high fitness value will be quickly destroyed. If the probability is too small, the search will stagnate and affect the search efficiency. The mutation probability is used to increase the diversity of the population. If the mutation probability is too small, no new gene blocks will be generated. If the mutation probability is too large, the algorithm will become a random search, affecting the search accuracy. The evolutionary algebra is used to indicate a condition for the end of the algorithm. If the evolutionary algebra is too large, it will affect the operation efficiency of the algorithm. The thermal circuit parameters are used to indicate the factors that affect the temperature rise of the transformer. Extracting them is convenient for accurate search and improves the calculation accuracy.

[0011] Preferably, in S3, the fitness of the initial population of the heat circuit is calculated in sequence based on the heat circuit equation group to obtain the heat circuit fitness of each individual in the population, including the following steps: Substitute the thermal circuit parameters in the initial population of the thermal circuit into the thermal circuit equation group in turn to obtain the transformer simulation temperature set, compare the temperature in the transformer simulation temperature set with the temperature of the actual working condition of the transformer in turn to obtain the temperature difference, and obtain the thermal circuit fitness of each population individual based on the temperature difference; The higher the temperature difference, the lower the fitness, and the lower the temperature difference, the higher the fitness.

[0012] In this scheme, the thermal circuit parameters are substituted into the thermal circuit equation group for solution to obtain the simulated temperature of the transformer, and the simulated temperature is compared with the actual temperature. According to the difference, it can be judged whether the thermal circuit parameters substituted into the thermal circuit equation group are suitable for the transformer. The smaller the difference, the higher the adaptability. In this way, it can be simply and intuitively reflected whether the searched thermal circuit parameters meet the requirements, without the need for complex verification means, which is very efficient and concise.

[0013] Preferably, in S3, based on the thermal path fitness of each population individual, the initial thermal path population is iterated to obtain individuals with excellent thermal path parameters, including the following steps: The thermal path fitness threshold is set, and individuals of the thermal path initial population that exceed the thermal path fitness threshold are extracted to obtain individuals with excellent thermal path parameters.

[0014] In this scheme, in order to avoid the loss of excellent genes during the search iteration process, which affects the search efficiency, some individuals with higher fitness in the initial population are screened, and iterative search is performed based on the individuals with excellent thermal path parameters obtained after screening. This can remove the negative impact of some poor quality thermal path parameters on the computational search, thereby improving the global convergence and computational efficiency.

[0015] Preferably, in S4, performing a cross operation on the excellent thermal path parameter individuals to obtain cross thermal path parameter individuals includes the following steps: The uniform crossover operator is used to set the mask field, which is composed of a number of mask words. Based on the number of bits of the mask words, the genes of any two excellent thermal circuit parameter individuals are inherited to obtain the crossover thermal circuit parameter individuals.

[0016] In this scheme, in order to obtain individuals with high fitness, a crossover operation is performed by means of a uniform crossover operator, and new individuals are generated on this basis. The genes of the new individual are determined by setting a mask field to determine which of the two individuals the corresponding genes are inherited from. When the bit in the mask word is "0", the new individual A1 inherits the corresponding genes in the old individual A, and when the mask word bit is "1", the new individual A1 inherits the corresponding genes in the old individual B. Thus, a complete new individual A1 can be generated, and a new individual B1 can be generated similarly. In this way, the genes of the previous generation of individuals can be inherited, so that excellent genes will not disappear, thereby continuously stacking excellent genes, thereby improving individual fitness and facilitating a faster search for the optimal thermal path parameters.

[0017] Preferably, in S4, performing a uniform mutation operation on the cross-heat path parameter individuals to obtain the mutated heat path parameter individuals includes the following steps: Each locus in the coding string of each crossover thermal path parameter individual is designated as a mutation point in turn, and for each mutation point, a random number is taken from the value range of the gene corresponding to the mutation point according to the mutation probability in the conditional parameter to replace the original gene value to obtain a mutant thermal path parameter individual; The value range is the value range of the thermal circuit parameter.

[0018] In this scheme, in order to improve the diversity of the population, the gene mutation operation is performed on the cross-thermal path parameter individuals, and the gene values ​​of certain loci in the individual chromosome coding string are replaced with other alleles of the loci to form a new individual. The mutation operation is an auxiliary method for generating new individuals, which ensures the effectiveness of the algorithm, enables the algorithm to have local search capabilities, improves the search efficiency of the algorithm, and maintains the diversity of the population at the same time to prevent premature convergence.

[0019] Preferably, in S5, iterating the individual variant thermal circuit parameters comprises the following steps: The mutated thermal circuit parameter individuals are taken as excellent thermal circuit parameter individuals and crossover operations are performed to obtain crossover thermal circuit parameter individuals; uniform mutation operations are performed on crossover thermal circuit parameter individuals to obtain mutated thermal circuit parameter individuals; fitness calculations are performed on mutated thermal circuit parameter individuals based on the thermal circuit equation group to obtain the next generation optimal fitness.

[0020] In this scheme, in order to obtain the optimal thermal circuit parameters, the individual thermal circuit parameters obtained after screening, crossover operation and mutation operation are continuously iterated. Each iteration will make the new thermal circuit parameters close to the optimal thermal circuit parameters. When the new thermal circuit parameters are equal to the optimal thermal circuit parameters or meet the set iteration requirements, the iteration is stopped. In this way, each generation of thermal circuit parameters is continuously verified, which can improve the accuracy of the final structure and make the structure more reliable and effective.

[0021] Preferably, the search termination condition includes a first termination condition and a second termination condition, wherein the first termination condition is that the absolute value of the difference between the best fitness values ​​of the next generation of two consecutive generations is between a set fitness threshold value; The second termination condition is that the number of iterations is equal to the evolutionary generation in the condition parameter; The search termination condition is satisfied by satisfying any one of the first termination condition and the second termination condition.

[0022] In this scheme, when the number of iterations is equal to the set evolutionary generation, it means that the search algorithm ends, and the last search result is the optimal thermal circuit parameter. Or when the fitness obtained by substituting the thermal circuit parameters between two consecutive generations into the thermal circuit equation group differs by a small positive number, it means that the space for iterative optimization is already very small, and the thermal circuit parameters obtained in the current generation can be used as the optimal thermal circuit parameters. At this time, the difference between the effect of the theoretical optimal thermal circuit parameters on the temperature rise of the transformer can be ignored. It can be considered that the optimal thermal circuit parameters are obtained, and unlimited iteration is prevented until the set evolutionary generation is reached, thereby setting some restrictions on the algorithm to improve the search efficiency.

[0023] Preferably, the range of the thermal circuit parameters is set based on transformer parameters.

[0024] In this scheme, when the thermal circuit parameters are continuously iterated, setting the iterated thermal circuit parameters based on the transformer parameters can prevent the search direction from being offset, thereby affecting the accuracy of the calculation results.

[0025] The beneficial effects of the present invention are as follows: (1) The present invention continuously iterates and screens the heat circuit parameters by means of parameter identification, and finally obtains the optimal heat circuit parameters. Compared with the traditional calculation method, there is no need to make a large number of assumptions before calculation, which improves the calculation efficiency of identifying the optimal heat circuit parameters, so that the total power loss value applied in the early stage of temperature rise can be accurately calculated, thereby improving the temperature rise efficiency and shortening the temperature rise time; (2) The present invention sets conditional parameters by different transformer types, which can adapt to more complex parameter spaces and flexibly adapt to different types of parameter optimization problems, thus avoiding the problem of insufficient accuracy caused by simplification of complex problems encountered in the traditional method. (3) The present invention substitutes the obtained optimal thermal circuit parameters into the thermal circuit equation group to solve the accurate temperature rise input power value, thereby avoiding the traditional temperature rise test process in which researchers can only rely on experience to give a loss power within a certain range higher than the total loss to shorten the temperature rise time, so that the input power reaches the optimal value, thereby improving the temperature rise efficiency.

[0026] The above invention content is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are only for the purpose of illustrating preferred embodiments and are not to be considered as limiting the present invention. Also, the same reference symbols are used throughout the drawings to represent the same parts.

[0028] Figure 1 This is a flow chart of a method for calculating the optimal input power for rapid temperature rise of a transformer according to the present invention; Figure 2 It is a schematic diagram of the structure of the transformer thermal circuit model of the present invention; Figure 3 It is a schematic diagram of the structure of the winding hot spot temperature thermal circuit model separated from the transformer thermal circuit model in the present invention; Figure 4 It is a schematic diagram of the structure of the top oil temperature heat circuit model separated from the transformer heat circuit model in the present invention; Figure 5 It is a schematic diagram of the structure of the oil tank wall thermal circuit model separated from the transformer thermal circuit model in the present invention. DETAILED DESCRIPTION

[0029] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0030] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations (or steps) therein can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the drawings; the process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0031] Example: Figure 1 As shown, in order to solve the problem that the prior art usually relies on a large number of assumptions and simplifications when calculating the optimal input power of the transformer, resulting in the result being inconsistent with the actual situation and thus causing the calculation result to be less accurate, this embodiment provides a method for calculating the optimal input power of the transformer for rapid temperature rise, including the following steps: S1: Construct a transformer thermal circuit model and construct a thermal circuit equation group based on the transformer thermal circuit model.

[0032] In this embodiment, a transformer thermal circuit model is constructed, and a thermal circuit equation group is constructed based on the transformer thermal circuit model, including the following steps: A transformer thermal circuit model is constructed, and the first equation is constructed based on the winding thermal capacity, the winding-oil thermal resistance and the transformer top oil temperature in the transformer thermal circuit model; The second equation is constructed based on the winding thermal capacity, winding-oil thermal resistance, transformer top oil temperature and transformer oil tank thermal capacity in the transformer thermal circuit model; A third equation is constructed based on the thermal resistance between oil and tank in the transformer thermal circuit model, tank temperature, transformer tank heat capacity, ambient temperature, and thermal resistance between the tank and the environment; The first, second and third equations are linked to construct the thermal circuit equations.

[0033] In order to facilitate the analysis of the thermal circuit parameters, Figure 2 As shown in the figure, the internal structure of the transformer is presented in the form of a transformer thermal circuit model. The thermal circuit equations are constructed according to the transformer thermal circuit model. At the same time, due to the complex internal structure of the transformer, the transformer thermal circuit model can be disassembled to facilitate the solution of the thermal circuit equations, as shown in Figure 3-5As shown, it is split into a winding hot spot temperature thermal circuit model, a top oil temperature thermal circuit model and an oil tank wall thermal circuit model. According to the above thermal circuit models, the first equation, the second equation and the third equation are respectively constructed. The above equations are related to each other to obtain a thermal circuit equation group, and the relationship between the thermal circuit parameters of the transformer is limited, which is convenient for the subsequent search for the optimal thermal circuit parameters, can reduce the amount of calculation and improve the calculation efficiency.

[0034] In this embodiment, a thermal circuit equation group is constructed according to the transformer thermal circuit model, and the formula is expressed as follows: ; in, is the transformer load loss, is the no-load loss, in W; , , They represent the thermal capacity of winding, oil and transformer tank respectively, in J / K; , , They represent the thermal resistance between winding and oil, the thermal resistance between oil and oil tank, and the thermal resistance between oil tank and ring, in K / W; , , , They represent transformer winding temperature, top oil temperature, oil tank temperature and ambient temperature respectively, and the unit is K.

[0035] S2: Initialize the conditional parameters and randomly generate the thermal circuit parameters in the thermal circuit equation group to obtain the initial thermal circuit population.

[0036] In this embodiment, the condition parameters include at least population size, crossover probability, mutation probability and evolutionary generation; The thermal circuit parameters at least include the thermal capacity of the winding, the thermal capacity of the oil, the thermal capacity of the transformer oil tank, and the thermal resistance between the winding and the oil, the thermal resistance between the oil and the oil tank, and the thermal resistance between the oil tank and the environment.

[0037] This embodiment uses the principle of natural selection, draws on the natural selection mechanism of biological survival of the fittest and the mechanism of gene recombination and mutation in biological reproduction and evolution to perform global adaptive probability search on thermal circuit parameters. Before searching, some parameters need to be preset, including population size, crossover probability, mutation probability and evolutionary algebra. The population size will affect the effectiveness of the search algorithm. If the population size is too small, it cannot provide enough sampling points. If the population size is too large, it will increase the amount of calculation and make the convergence time too long. The crossover probability controls the frequency of the crossover operation. If the crossover probability is too large, the structure with high fitness value will be quickly destroyed. If the probability is too small, the search will stagnate and affect the search efficiency. The mutation probability is used to increase the diversity of the population. If the mutation probability is too small, no new gene blocks will be generated. If the mutation probability is too large, the algorithm will become a random search, affecting the search accuracy. The evolutionary algebra is used to indicate a condition for the end of the algorithm operation. If the evolutionary algebra is too large, it will affect the operation efficiency of the algorithm. The thermal circuit parameters are used to indicate the factors that affect the temperature rise of the transformer. Extracting them will facilitate accurate search and improve calculation accuracy.

[0038] In this embodiment, the population size n is more suitable between 20 and 160, and in this embodiment, n=150; the crossover probability Pc is generally 0.5-1.0, and in this embodiment, Pc=0.6; the mutation probability Pm is generally 0.001-0.1, and in this embodiment, Pm=0.1; the evolutionary generation t generally ranges from 100 to 1000, and in this embodiment, t=200. The selection of the evolutionary generation can also adopt a certain judgment criterion, and stop immediately when the criterion is met.

[0039] In this embodiment, the range of the thermal circuit parameters is set based on the transformer parameters.

[0040] In this embodiment, when the thermal circuit parameters are continuously iterated, the iterated thermal circuit parameters are set based on the transformer parameters to prevent the search direction from being offset, thereby affecting the accuracy of the calculation results. When the thermal circuit parameters are randomly generated, the search range of some parameters can be given according to the traditional estimation method, so that the search time will not be too long due to the search path offset in the subsequent search for the optimal thermal circuit parameters, thereby affecting the calculation speed. For example, for a corrugated oil tank type distribution transformer with a rated capacity of 400kVA, its initial parameter estimation is as follows: Low voltage winding: Rated turns: 25, multi-layer cylindrical oblong winding Conductor name: TU1 copper tape, interlayer insulation: 0.21mm=0.08+0.13 Wire specification: 0.75×345 GB / T 18813, 1 radial, 1 axial, Rp0.2=90-110MPa, copper busbar: 5×40 Conductor area: 1×1×258.629=258.629mm2 Average turn length: 2π×116.53 / 103=0.733m Total length of wire: 25×0.733+0.095=18.42m Wire net weight: 3×8.9×18.42×258.629 / 103=127.2kg, turn insulation weight: 0kg, wire total weight: 127.2kg; High voltage winding: Winding turns: most negative 1026, rated 1081, most positive 1134, multi-layer cylindrical oblong winding 15 layers = 14 layers × 75 turns + 1 layer × 84 turns, interlayer insulation: 0.228mm = (1 + 1 / 2) × (2 × 0.076) (check thickness: 0.374mm) Wire name: Polyester enameled copper flat wire Wire specification: QZB-2 / 130 1.65×4.15 / 1.80×4.30, 1 radial, 1 axial, Rp0.2=90-110MPa Conductor area: 1×1×6.633=6.633mm2 Average turn length: 2π×158.27 / 103=0.9944m Total length of wire: 1053×0.9944+0.397=1047.5m Wire net weight: 3×8.9×1047.5×6.633 / 103=185.5kg, turn insulation weight: 3.3kg, wire total weight: 188.8kg The total estimated value of the winding heat capacity is Cwnd = (188.8 + 127.2) × 385 = 121660 J / K; Transformer oil: Oil tank capacity: 297dm3, corrugated sheet volume: 51dm3, body oil discharge volume: 159dm3, oil volume: 193dm3 Transformer total oil weight: 170kg Then the estimated value of oil heat capacity is Coil = 170 × 2000 = 340000 J / K; Transformer tank: Weight of corrugated sheet: 117kg, weight of box cover: 22kg, weight of box wall: 15kg, weight of box bottom: 46.62kg Fuel tank weight: 190kg Then the estimated value of the heat capacity of the oil tank is Ctank=190×500=95000 J / K; The factory parameters of the transformer are as follows: Heat transfer coefficient between winding and oil = 542 W / m²·K Heat exchange area between winding and oil = 2.36m² Heat transfer coefficient between oil and tank = 212W / m²·K Heat exchange area between oil and tank = 2.81m² Heat transfer coefficient between tank and environment = 50W / m²·K Heat exchange area between tank and environment = 3.76m² Get the estimated value of thermal resistance parameter: Rwnd-oil=1 / (542×2.36)≈7.818×10-4K / W Roil-tank=1 / (212×2.81)≈1.679×10-3K / W Rtank-amb=1 / (50×3.76)≈5.319×10-3K / W; For a corrugated oil tank distribution transformer with a rated capacity of 400kVA, due to different calculation requirements under actual working conditions, the calculation of the above factory parameters will have deviations. In the parameter identification algorithm, the search range of thermal capacitance and thermal resistance parameters can be set within ±10% of the above estimated parameters to narrow the search range and speed up the convergence speed. Each of the above individuals is a possible solution and chromosome encoding is performed on it. Binary encoding and floating point encoding can be used. In this embodiment, binary encoding is used. The encoding and decoding of the binary encoding method are simple and easy, making the crossover, mutation and other operations of the algorithm easy to implement. Therefore, each individual of the initial population can be represented as a binary string.

[0041] S3: Based on the thermal circuit equations, the fitness of the initial thermal circuit population is calculated in turn to obtain the thermal circuit fitness of each individual in the population. Based on the thermal circuit fitness of each individual in the population, the initial thermal circuit population is screened to obtain individuals with excellent thermal circuit parameters.

[0042] In this embodiment, the fitness of the initial population of the heat circuit is calculated in sequence based on the heat circuit equation group to obtain the heat circuit fitness of each individual population, including the following steps: Substitute the thermal circuit parameters in the initial population of the thermal circuit into the thermal circuit equation group in turn to obtain the transformer simulation temperature set, compare the temperature in the transformer simulation temperature set with the temperature of the actual working condition of the transformer in turn to obtain the temperature difference, and obtain the thermal circuit fitness of each population individual based on the temperature difference; The higher the temperature difference, the lower the fitness, and the lower the temperature difference, the higher the fitness.

[0043] In this embodiment, the thermal circuit parameters are substituted into the thermal circuit equation group to solve and obtain the simulated temperature of the transformer, and the simulated temperature is compared with the actual temperature. According to the difference, it can be judged whether the thermal circuit parameters substituted into the thermal circuit equation group are suitable for the transformer. The smaller the difference, the higher the adaptability. In this way, it can be simply and intuitively reflected whether the searched thermal circuit parameters meet the requirements, without the need for complex verification means, which is very efficient and concise.

[0044] In this embodiment, based on the thermal path fitness of each population individual, the initial thermal path population is iterated to obtain individuals with excellent thermal path parameters, including the following steps: The thermal path fitness threshold is set, and individuals of the thermal path initial population that exceed the thermal path fitness threshold are extracted to obtain individuals with excellent thermal path parameters.

[0045] In order to avoid the loss of excellent genes during the search iteration process, thereby affecting the search efficiency, this embodiment screens some individuals with higher fitness in the initial population, and performs iterative search based on the individuals with excellent thermal path parameters obtained after screening. This can remove the negative impact of some poor quality thermal path parameters on the computational search, thereby improving global convergence and computational efficiency.

[0046] S4: Perform a crossover operation on the excellent thermal circuit parameter individuals to obtain the crossover thermal circuit parameter individuals; perform a uniform mutation operation on the crossover thermal circuit parameter individuals to obtain the mutated thermal circuit parameter individuals; perform fitness calculation on the mutated thermal circuit parameter individuals based on the thermal circuit equation group to obtain the next generation optimal fitness.

[0047] In this embodiment, performing a cross operation on excellent thermal circuit parameter individuals to obtain cross thermal circuit parameter individuals includes the following steps: The uniform crossover operator is used to set the mask field, which is composed of a number of mask words. Based on the number of bits of the mask words, the genes of any two excellent thermal circuit parameter individuals are inherited to obtain the crossover thermal circuit parameter individuals.

[0048] In order to obtain individuals with high fitness, this embodiment can use a single-point crossover operator, a two-point crossover operator or a multi-point crossover operator. The calculation amount and accuracy of each method are different, and can be selected according to actual requirements. This embodiment uses a uniform crossover operator to perform a crossover operation, and generates a new individual on this basis. By setting a mask field, it is determined which of the two individuals the genes of the new individual inherit. For example, when the bit in the mask word is "0", the new individual A1 inherits the corresponding gene in the old individual A, and when the mask word bit is "1", the new individual A1 inherits the corresponding gene in the old individual B. In this way, a complete new individual A1 can be generated, and a new individual B1 can be generated similarly. In this way, the genes of the previous generation of individuals can be inherited, so that excellent genes will not disappear, thereby continuously stacking excellent genes, thereby improving individual fitness, and facilitating and faster finding of the optimal thermal path parameters.

[0049] In this embodiment, performing a uniform mutation operation on the cross-heat path parameter individuals to obtain the mutated heat path parameter individuals includes the following steps: Each locus in the coding string of each crossover thermal path parameter individual is designated as a mutation point in turn, and for each mutation point, a random number is taken from the value range of the gene corresponding to the mutation point according to the mutation probability in the conditional parameter to replace the original gene value to obtain a mutant thermal path parameter individual; The value range is the value range of the thermal circuit parameter.

[0050] In order to improve the diversity of the population, this embodiment performs gene mutation operation on the cross-heat path parameter individuals, and replaces the gene values ​​of certain loci in the individual chromosome coding string with other alleles of the loci, thereby forming a new individual. The mutation operation is an auxiliary method for generating new individuals, which ensures the effectiveness of the algorithm, enables the algorithm to have local search capabilities, and can improve the search efficiency of the algorithm. At the same time, it maintains the diversity of the population and prevents premature convergence.

[0051] In this embodiment, uniform mutation refers to replacing the original gene value at each locus in the individual coding string with a random number that is uniformly distributed within a certain range, that is, the range of the thermal capacitance and thermal resistance parameters, with a relatively small probability, specifically including the following steps: Step 1: sequentially specify each locus in each cross-heat path parameter individual code string as a mutation point; Step 2: For each mutation point, a random number is selected from the value range of the corresponding gene with the mutation probability Pm to replace the original gene value; for example, in the gene development process, there is an individual (winding heat capacity) ,like is the variation point, and its value range is , at this point After uniform mutation operation, a new individual can be obtained: , where the new gene value of the mutation point is: ; Among them, r is a random number in the range of [0, 1] that conforms to the uniform probability distribution. The uniform mutation operation is suitable for the initial operation stage of the algorithm, so that the search point can move freely in the entire search space, thereby increasing the diversity of the group.

[0052] S5: If the next-generation best fitness does not meet the set search termination condition, the mutated thermal circuit parameter individual is iterated. If the next-generation best fitness meets the set search termination condition, the mutated thermal circuit parameter individual is used as the optimal thermal circuit parameter.

[0053] In this embodiment, the individual variant thermal circuit parameters are iterated, including the following steps: The mutated thermal circuit parameter individuals are taken as excellent thermal circuit parameter individuals and crossover operations are performed to obtain crossover thermal circuit parameter individuals; uniform mutation operations are performed on crossover thermal circuit parameter individuals to obtain mutated thermal circuit parameter individuals; fitness calculations are performed on mutated thermal circuit parameter individuals based on the thermal circuit equation group to obtain the next generation optimal fitness.

[0054] In order to obtain the optimal thermal circuit parameters, this embodiment continuously iterates the thermal circuit parameter individuals obtained after screening, crossover operation, and mutation operation. Each iteration will make the new thermal circuit parameters close to the optimal thermal circuit parameters. When the new thermal circuit parameters are equal to the optimal thermal circuit parameters or meet the set iteration requirements, the iteration is stopped. In this way, each generation of thermal circuit parameters is continuously verified, which can improve the accuracy of the final structure and make the structure more reliable and effective.

[0055] In this embodiment, the search termination condition includes a first termination condition and a second termination condition, wherein the first termination condition is that the absolute value of the difference between the best fitness values ​​of the next generation of two consecutive generations is between the set fitness thresholds; The second termination condition is that the number of iterations is equal to the evolutionary generation in the condition parameter; The search termination condition is satisfied by satisfying any one of the first termination condition and the second termination condition.

[0056] In this embodiment, when the number of iterations is equal to the set evolutionary generation, it means that the search algorithm is finished, and the last search result is the optimal thermal circuit parameter. Or when the fitness obtained by substituting the thermal circuit parameters between two consecutive generations into the thermal circuit equation group differs by a small positive number, it means that the space for iterative optimization is already very small, and the thermal circuit parameters obtained in the current generation can be used as the optimal thermal circuit parameters. At this time, the difference between the effect of the theoretical optimal thermal circuit parameters on the temperature rise of the transformer can be ignored. It can be considered that the optimal thermal circuit parameters are obtained, and unlimited iteration is prevented until the set evolutionary generation is reached, thereby setting some restrictions on the algorithm to improve the search efficiency.

[0057] In this embodiment, the formula of the first termination condition is expressed as follows: ; in is the fitness of the best individual in the newly generated population, is the fitness of the best individual in the previous generation, is some small positive number.

[0058] S6: Substitute the optimal thermal circuit parameters into the thermal circuit equation group to obtain the optimal input power for rapid temperature rise of the transformer. The formula is as follows: ; in, is the transformer load loss, is the no-load loss, in W; , , They represent the thermal capacity of winding, oil and transformer tank respectively, in J / K; , , They represent the thermal resistance between winding and oil, the thermal resistance between oil and oil tank, and the thermal resistance between oil tank and ring, in K / W; , , , They represent transformer winding temperature, top oil temperature, oil tank temperature and ambient temperature respectively, and the unit is K.

[0059] Since the winding temperature rise, oil temperature rise and oil tank temperature rise are known, that is, the final winding temperature, oil temperature and oil tank temperature are clear, according to the given temperature rise time conditions, the total power loss value applied in the early stage of temperature rise can be accurately solved according to the above formula, so that the input power reaches the optimal value, thereby improving the temperature rise efficiency and shortening the temperature rise time. The thermal capacitance and thermal resistance parameter values ​​in the thermal circuit model can be accurately identified, and the precise value of the rapid temperature rise input power can be calculated, which provides more reference value for optimizing the transformer structure and improving the temperature rise efficiency.

[0060] This embodiment, by constructing a transformer thermal circuit model, is helpful to extract and analyze the parameters affecting the temperature rise of the transformer, so as to facilitate the solution of the constructed thermal circuit equation group to improve the calculation efficiency; in order to reduce the amount of calculation, when the thermal circuit parameters are randomly generated, the search range of some parameters can be given according to the traditional estimation method, so that when searching for the optimal thermal circuit parameters in the subsequent search, the search time will not be too long due to the search path offset, thereby affecting the calculation speed; by calculating the fitness of the initial population of the thermal circuit, the initial population of the thermal circuit is screened according to the calculation results to obtain individuals with excellent thermal circuit parameters, which is helpful to retain individuals with high fitness and avoid the loss of excellent genes, thereby improving the global convergence and calculation efficiency during the calculation; by performing crossover and uniform mutation operations on individuals with excellent thermal circuit parameters, it is helpful to improve the population diversity and thus improve the comprehensiveness of the search, prevent the optimal thermal circuit parameters from being missed, and improve the calculation accuracy; by continuously iterating the population and continuously optimizing the thermal circuit parameters, there is no need to make a lot of assumptions before calculation, and the optimal thermal circuit parameters can be found only through natural selection, with higher calculation accuracy and stronger adaptability to different types of computing environments.

[0061] It can be seen from the above embodiments that at least the following substantial effects are achieved: (1) The present invention continuously iterates and screens the heat circuit parameters through parameter identification, and finally obtains the optimal heat circuit parameters. Compared with the traditional calculation method, there is no need to make a large number of assumptions before calculation, which improves the calculation efficiency of identifying the optimal heat circuit parameters, so that the total power loss value applied in the early stage of temperature rise can be accurately calculated, thereby improving the temperature rise efficiency and shortening the temperature rise time; (2) The present invention sets conditional parameters by different transformer types, which can adapt to more complex parameter spaces and flexibly adapt to different types of parameter optimization problems, thus avoiding the problem of insufficient accuracy caused by simplification of complex problems encountered in the traditional method. (3) The present invention substitutes the obtained optimal thermal circuit parameters into the thermal circuit equation group to solve the accurate temperature rise input power value, thereby avoiding the traditional temperature rise test process in which researchers can only rely on experience to give a loss power within a certain range higher than the total loss to shorten the temperature rise time, so that the input power reaches the optimal value, thereby improving the temperature rise efficiency.

[0062] The specific implementation described above is a preferred implementation of a method for calculating the optimal input power for rapid temperature rise of a transformer of the present invention, and is not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation. All equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A method for calculating the optimal input power for rapid temperature rise of a transformer, characterized in that: The following steps are involved: S1. construct a transformer thermal circuit model, and construct a thermal circuit equation group based on the transformer thermal circuit model; S2, initializing the condition parameters, and randomly generating the thermal circuit parameters in the thermal circuit equation group to obtain the initial population of the thermal circuit; S3, calculating the fitness of the initial population of the thermal circuit in turn based on the thermal circuit equation group to obtain the thermal circuit fitness of each individual in the population; Based on the thermal path fitness of each individual in the population, the initial population of thermal paths is screened to obtain individuals with excellent thermal path parameters; S4, performing a cross operation on the excellent thermal circuit parameter individuals to obtain cross thermal circuit parameter individuals; Performing uniform mutation operation on the cross-heat path parameter individuals to obtain mutated heat path parameter individuals; The fitness of the individuals with mutated thermal circuit parameters is calculated based on the thermal circuit equations to obtain the optimal fitness of the next generation; S5. If the next-generation best fitness does not meet the set search termination condition, the mutated thermal circuit parameter individual is iterated. If the next-generation best fitness meets the set search termination condition, the mutated thermal circuit parameter individual is used as the best thermal circuit parameter. S6. Substitute the optimal thermal circuit parameters into the thermal circuit equation group to obtain the optimal input power for rapid temperature rise of the transformer.

2. The method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 1, characterized in that: In S1, a transformer thermal circuit model is constructed, and a thermal circuit equation group is constructed based on the transformer thermal circuit model, including the following steps: A transformer thermal circuit model is constructed, and the first equation is constructed based on the winding thermal capacity, the winding-oil thermal resistance and the transformer top oil temperature in the transformer thermal circuit model; The second equation is constructed based on the winding thermal capacity, winding-oil thermal resistance, transformer top oil temperature and transformer oil tank thermal capacity in the transformer thermal circuit model; A third equation is constructed based on the thermal resistance between oil and tank in the transformer thermal circuit model, tank temperature, transformer tank heat capacity, ambient temperature, and thermal resistance between the tank and the environment; The first, second and third equations are linked to construct the thermal circuit equations.

3. The method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 1, characterized in that: The condition parameters include at least population size, crossover probability, mutation probability and evolutionary generations; The thermal circuit parameters at least include the thermal capacity of the winding, the thermal capacity of the oil, the thermal capacity of the transformer oil tank, and the thermal resistance between the winding and the oil, the thermal resistance between the oil and the oil tank, and the thermal resistance between the oil tank and the environment.

4. The method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 1, characterized in that: In S3, the fitness of the initial population of the heat path is calculated in turn based on the heat path equation group to obtain the heat path fitness of each individual in the population, including the following steps: Substitute the thermal circuit parameters in the initial population of the thermal circuit into the thermal circuit equation group in turn to obtain the transformer simulation temperature set, compare the temperature in the transformer simulation temperature set with the temperature of the actual working condition of the transformer in turn to obtain the temperature difference, and obtain the thermal circuit fitness of each population individual based on the temperature difference; The higher the temperature difference, the lower the fitness, and the lower the temperature difference, the higher the fitness.

5. The method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 1, characterized in that: In S3, based on the thermal path fitness of each individual in the population, the initial population of thermal paths is screened to obtain individuals with excellent thermal path parameters, including the following steps: The thermal path fitness threshold is set, and individuals of the thermal path initial population that exceed the thermal path fitness threshold are extracted to obtain individuals with excellent thermal path parameters.

6. The method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 1, characterized in that: In S4, a cross operation is performed on the excellent thermal circuit parameter individuals to obtain cross thermal circuit parameter individuals, including the following steps: The uniform crossover operator is used to set the mask field, which is composed of a number of mask words. Based on the number of bits of the mask words, the genes of any two excellent thermal circuit parameter individuals are inherited to obtain the crossover thermal circuit parameter individuals.

7. The method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 1, characterized in that: In S4, uniform mutation operation is performed on the cross-heat path parameter individuals to obtain mutated heat path parameter individuals, including the following steps: Each locus in the coding string of each crossover thermal path parameter individual is designated as a mutation point in turn, and for each mutation point, a random number is taken from the value range of the gene corresponding to the mutation point according to the mutation probability in the conditional parameter to replace the original gene value to obtain a mutant thermal path parameter individual; The value range is the value range of the thermal circuit parameter.

8. The method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 1, characterized in that: In S5, the individual variant thermal circuit parameters are iterated, including the following steps: The mutated thermal circuit parameter individuals are taken as excellent thermal circuit parameter individuals and crossover operations are performed to obtain crossover thermal circuit parameter individuals; uniform mutation operations are performed on crossover thermal circuit parameter individuals to obtain mutated thermal circuit parameter individuals; fitness calculations are performed on mutated thermal circuit parameter individuals based on the thermal circuit equation group to obtain the next generation optimal fitness.

9. The method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 1, characterized in that: The search termination condition includes a first termination condition and a second termination condition, wherein the first termination condition is that the absolute value of the difference between the best fitness values ​​of the next generation of two consecutive generations is between the set fitness thresholds; The second termination condition is that the number of iterations is equal to the evolutionary generation in the condition parameter; The search termination condition is satisfied by satisfying any one of the first termination condition and the second termination condition.

10. A method for calculating the optimal input power for rapid temperature rise of a transformer according to claim 3, 4 or 8, characterized in that: The range of the thermal circuit parameters is set based on the transformer parameters.

Images