CN116882293A - Multi-objective optimization method and device for transformer and storage medium - Google Patents
Multi-objective optimization method and device for transformer and storage medium Download PDFInfo
- Publication number
- CN116882293A CN116882293A CN202310887023.8A CN202310887023A CN116882293A CN 116882293 A CN116882293 A CN 116882293A CN 202310887023 A CN202310887023 A CN 202310887023A CN 116882293 A CN116882293 A CN 116882293A
- Authority
- CN
- China
- Prior art keywords
- parent
- population
- optimization
- target
- objective
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 162
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000006870 function Effects 0.000 claims abstract description 87
- 238000004804 winding Methods 0.000 claims description 117
- 230000035772 mutation Effects 0.000 claims description 30
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 21
- 239000000463 material Substances 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 239000003921 oil Substances 0.000 description 9
- 235000019198 oils Nutrition 0.000 description 9
- 230000002028 premature Effects 0.000 description 8
- 230000035882 stress Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000013461 design Methods 0.000 description 6
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 229910052742 iron Inorganic materials 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 108090000623 proteins and genes Proteins 0.000 description 3
- 238000005452 bending Methods 0.000 description 2
- 210000000349 chromosome Anatomy 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000001052 transient effect Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000010696 ester oil Substances 0.000 description 1
- 150000002148 esters Chemical class 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
- 235000015112 vegetable and seed oil Nutrition 0.000 description 1
- 239000008158 vegetable oil Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Computational Mathematics (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application discloses a transformer multi-objective optimization method, a device and a storage medium, wherein a multi-objective initial model is established according to objective functions, optimization parameters and optimization constraints by determining a plurality of objective functions, optimization parameters and optimization constraints; and generating an initial population, taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, obtaining a parameter optimization result according to the updated target parent population, and obtaining a multi-target optimization model according to the parameter optimization result and the multi-target initial model.
Description
Technical Field
The present application relates to the field of transformer technologies, and in particular, to a transformer multi-objective optimization method, device, and storage medium.
Background
With the increasing demand for energy, offshore wind power generation plays an important role in new energy wind power generation. The capacity of the offshore wind turbine is continuously increased, and the offshore transformer is used as an important component of wind power generation, and the existing offshore transformer adopts natural ester or mixed ester oil as an insulating heat dissipation medium to meet the related environmental protection requirements. Offshore transformers, unlike conventional transformers, require a higher level of safety. When a corresponding electrical problem occurs, such as: short circuit of the transformer, over high temperature rise of the transformer, partial discharge of the transformer, insulation aging of the transformer and the like. Due to the topography factors, staff cannot process in time, the safety requirement is higher, and the corresponding cost investment is larger. Therefore, the investment of cost is considered while the voltage level and the capacity are improved, the design requirement on the transformer is continuously improved, and the multi-objective comprehensive optimization design of the offshore transformer is a requirement for industry development.
The transformer multi-objective optimization design problem is nonlinear optimization design problem, and is a complex optimization task. For the multi-objective optimization problems of nonlinearity, strong coupling, multiple variables and complex constraint relations, the contradiction between targets is well coordinated, the selection of reasonable design parameters and the improvement of algorithm running speed are all the problems that the optimal design of the power transformer needs to be considered.
However, the conventional transformer multi-objective optimization algorithm generally adds a weighting coefficient to a plurality of objective functions, and converts the optimization into a single objective optimization problem according to the priority and the weighting of each objective function. The optimization model is simple, a conventional optimization algorithm is adopted, the initial value of the population is randomized, the running speed and the optimizing precision of the algorithm are low, and the algorithm is easy to fall into a local optimal solution.
Disclosure of Invention
The present application aims to solve at least one of the technical problems existing in the prior art. Therefore, the application provides a transformer multi-objective optimization method, a device and a storage medium, which can solve the problems of low running speed and low optimizing precision of the traditional transformer multi-objective optimization algorithm and easy sinking into a local optimal solution.
According to an embodiment of the first aspect of the application, the transformer multi-objective optimization method comprises the following steps:
determining a plurality of objective functions, optimization parameters and optimization constraints, wherein the optimization parameters comprise structural parameters of a transformer, and the optimization constraints at least comprise industry standard constraints and material characteristic constraints;
establishing a multi-objective initial model according to the objective function, the optimization parameters and the optimization constraint;
generating an initial population, wherein the initial population comprises a plurality of random solutions and artificial solutions of the multi-target initial model, and the artificial solutions are obtained according to the historical data of the transformer;
taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, and obtaining a parameter optimization result according to the updated target parent population;
obtaining a multi-objective optimization model according to the parameter optimization result and the multi-objective initial model;
the iterative strategy comprises the following steps:
non-dominant sorting is carried out on the target parent population, each artificial solution and each random solution in the target parent population are parent individuals, and the target parent population is divided into a dominant layer and a non-dominant layer;
calculating the crowdedness of each parent individual in the non-dominant layer;
screening a plurality of parent individuals according to the non-dominant relationship of the parent individuals and the crowding degree to obtain an intermediate parent population;
performing selection, crossing and mutation operations on the intermediate parent population to obtain a target offspring population;
and merging the intermediate parent population and the target offspring population to obtain a new target parent population.
The transformer multi-objective optimization method according to the embodiment of the first aspect of the application has at least the following beneficial effects:
establishing a multi-objective initial model according to the objective functions, the optimization parameters and the optimization constraints by determining a plurality of objective functions, the optimization parameters and the optimization constraints; and generating an initial population, taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, obtaining a parameter optimization result according to the updated target parent population, and obtaining a multi-target optimization model according to the parameter optimization result and the multi-target initial model. Compared with the traditional transformer multi-objective optimization algorithm, the transformer multi-objective optimization method provided by the embodiment of the application has the advantages that the artificial solution is added in the initial population, so that the convergence speed and the optimizing precision of the algorithm are improved, the premature phenomenon is overcome, and the phenomenon of sinking into the local optimal solution is avoided.
According to some embodiments of the application, the objective function includes a material cost function, a total loss function, a temperature function, and a short circuit stress function.
According to some embodiments of the application, the optimized parameters include core diameter, low voltage winding wire width, low voltage winding wire thickness, high voltage winding wire width, high voltage winding wire thickness, low voltage winding turns, high voltage winding turns, low voltage winding average radius, high voltage winding average radius, low voltage winding width wire count, high voltage winding axial wire count, high voltage winding width wire count.
According to some embodiments of the application, the selecting, crossing and mutating the intermediate parent population to obtain a target offspring population comprises:
performing selection operation on the intermediate parent population through a selection operator to obtain a plurality of parent groups, wherein each parent group comprises two parent individuals;
performing crossover operation on the two parent individuals in each parent group through crossover operators to obtain a plurality of intermediate offspring individuals;
performing mutation operation on each intermediate offspring individual through a mutation operator to obtain a plurality of target offspring individuals;
and combining all the target offspring individuals to obtain the target offspring population.
According to some embodiments of the application, the selecting operation is performed on the intermediate parent population by a selecting operator to obtain a plurality of parent groups, including:
selecting a first parent group from the intermediate parent population by a tournament selection algorithm;
and grouping the rest of the parent individuals of the intermediate parent population into a second double-parent group.
According to some embodiments of the application, the selecting the first parent group from the intermediate parent population by a tournament selection algorithm comprises:
a first parent group is selected from the intermediate parent population by a tournament selection algorithm having a tanh function as an activation function.
According to some embodiments of the application, the performing a crossover operation on the two parent individuals in each parent group by a crossover operator results in a plurality of intermediate child individuals, including:
crossing the two parent individuals in the first parent group through an arithmetic crossing operator to obtain a first intermediate offspring individual;
and crossing the two parent individuals in the second double parent group through a multi-point crossing operator to obtain a second intermediate offspring individual.
According to some embodiments of the application, the performing, by a mutation operator, a mutation operation on each of the intermediate child individuals to obtain a plurality of target child individuals includes:
and mutating each intermediate offspring individual by a single-point mutation operator to obtain a plurality of target offspring individuals.
According to a second aspect of the present application, a transformer multi-objective optimization apparatus includes:
the data input module is used for inputting a plurality of objective functions, optimization parameters and optimization constraints, wherein the optimization parameters comprise structural parameters of the transformer, and the optimization constraints at least comprise industry standard constraints and material characteristic constraints;
the model construction module is used for constructing a multi-objective initial model according to the objective function, the optimization parameters and the optimization constraint;
the strategy execution module is used for generating an initial population, wherein the initial population comprises a plurality of random solutions and artificial solutions of the multi-target initial model, and the artificial solutions are obtained according to the transformer historical data; taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, and obtaining a parameter optimization result according to the updated target parent population; the iterative strategy comprises the following steps: non-dominant sorting is carried out on the target parent population, each artificial solution and each random solution in the target parent population are parent individuals, and the target parent population is divided into a dominant layer and a non-dominant layer; calculating the crowdedness of each parent individual in the non-dominant layer; screening a plurality of parent individuals according to the non-dominant relationship of the parent individuals and the crowding degree to obtain an intermediate parent population; performing selection, crossing and mutation operations on the intermediate parent population to obtain a target offspring population; combining the intermediate parent population and the target offspring population to obtain a new target parent population;
and the result output module is used for outputting a multi-objective optimization model according to the parameter optimization result and the multi-objective initial model.
The transformer multi-objective optimization device according to the embodiment of the second aspect of the application has at least the following beneficial effects:
establishing a multi-objective initial model according to the objective functions, the optimization parameters and the optimization constraints by determining a plurality of objective functions, the optimization parameters and the optimization constraints; and generating an initial population, taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, obtaining a parameter optimization result according to the updated target parent population, and obtaining a multi-target optimization model according to the parameter optimization result and the multi-target initial model. Compared with the traditional transformer multi-objective optimization technology, the transformer multi-objective optimization device of the second aspect of the embodiment of the application has the advantages that the artificial solution is added in the initial population, so that the convergence speed and the optimizing precision of the algorithm are improved, the premature phenomenon is overcome, and the phenomenon of falling into the local optimal solution is avoided.
A computer readable storage medium according to an embodiment of the third aspect of the present application has stored therein a processor executable program for implementing a transformer multi-objective optimization method as described above when executed by a processor.
The computer-readable storage medium according to the embodiment of the third aspect of the present application has at least the following advantageous effects:
establishing a multi-objective initial model according to the objective functions, the optimization parameters and the optimization constraints by determining a plurality of objective functions, the optimization parameters and the optimization constraints; and generating an initial population, taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, obtaining a parameter optimization result according to the updated target parent population, and obtaining a multi-target optimization model according to the parameter optimization result and the multi-target initial model. Compared with the traditional transformer multi-objective optimization technology, the computer readable storage medium of the embodiment of the third aspect of the application increases the convergence speed and optimizing precision of the algorithm due to the addition of the manual solution in the initial population, overcomes the premature phenomenon and avoids sinking into the local optimal solution.
Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the application.
Drawings
The application is further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of a transformer multi-objective optimization method according to an embodiment of the application;
FIG. 2 is a flow chart of an iterative strategy in an embodiment of the application;
FIG. 3 is a flow chart of obtaining a target offspring population according to one embodiment of the present application;
FIG. 4 is a flow chart of obtaining multiple parent groups according to an embodiment of the application;
FIG. 5 is a flow chart of obtaining a plurality of intermediate offspring individuals according to one embodiment of the present application;
FIG. 6 is a schematic diagram of a transformer multi-objective optimization apparatus according to an embodiment of the application.
Reference numerals:
a transformer multi-objective optimization device 100;
a data input module 110;
a model building module 120;
a policy enforcement module 130;
and a result output module 140.
Detailed Description
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the application.
In the description of the present application, it should be understood that the direction or positional relationship indicated with respect to the description of the orientation, such as up, down, etc., is based on the direction or positional relationship shown in the drawings, is merely for convenience of describing the present application and simplifying the description, and does not indicate or imply that the apparatus or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present application.
In the description of the present application, plural means two or more. The description of the first and second is for the purpose of distinguishing between technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
In the description of the present application, unless explicitly defined otherwise, terms such as arrangement, installation, electrical connection, etc. should be construed broadly and the specific meaning of the terms in the present application can be reasonably determined by a person skilled in the art in combination with the specific contents of the technical scheme.
A transformer multi-objective optimization method, apparatus and storage medium according to an embodiment of the present application are described below with reference to fig. 1 to 6.
The transformer multi-objective optimization method according to an embodiment of the present application, as shown in fig. 1, includes, but is not limited to, step S100, step S200, step S300, step S400, and step S500.
Step S100: determining a plurality of objective functions, optimization parameters and optimization constraints, wherein the optimization parameters comprise structural parameters of the transformer, and the optimization constraints at least comprise industry standard constraints and material characteristic constraints;
in the step, a plurality of objective functions, optimization parameters and optimization constraints are determined, so that a multi-objective initial model is constructed in the subsequent step according to the objective functions, the optimization parameters and the optimization constraints.
It is understood that the objective functions include a material cost function, a total loss function, a temperature function, and a short circuit stress function, wherein the temperature function includes an average temperature function of the winding surface to oil and a hottest temperature rise function of the sudden short circuit transient winding.
The optimized parameters include core diameter, low voltage winding wire width, low voltage winding wire thickness, high voltage winding wire width, high voltage winding wire thickness, low voltage winding number of turns, high voltage winding number of turns, low voltage winding average radius, high voltage winding average radius, low voltage winding width wire number, high voltage winding axial wire number and high voltage winding width wire number.
It is understood that optimization constraints include at least industry standard constraints, material property constraints, and engineering experience constraints.
It should be noted that the material cost function is obtained by the following formula:
wherein f 1 (X) is a material cost function. G is the sum of the weight of each material, H o Is iron core window height. A is that e Is the clear area of the iron yoke. M is M o G is the center distance of the iron core column Δ Is the angular weight of the iron yoke. w (w) 1 Is the number of turns of the low-voltage coil. w (w) 2 Is the number of turns of the high-voltage coil. R is R P1 Is the average radius of the low voltage winding. R is R P2 Is the average radius of the high voltage winding. K (K) G Accessory weight coefficient. K (K) d Is an engineering factor, related to transformer capacity and voltage class. S is S N Is the transformer capacity. H k Is the axial total height of the coil.
a 0 、a 12 The air gap distance between the iron core and the low voltage side and the air gap distance between the iron core and the high voltage side are respectively related to voltage class, a 0 10mm,3 12 45mm was taken.
γ iron The density of the iron core is 7.65kg/m3. Sigma is the density of the copper wire, in particular 8.9kg/m3.b m Is the thickness of the cushion block between winding segments, and is generally 2.5mm.
M Fe 、M cu 、M oil 、M else The unit price of the iron core, the unit price of the high-low voltage winding wire and the unit price of the vegetable oil and the unit price of the accessories are respectively.
G Fe 、G cu1 、G cu2 、G oil 、G else The weight of the iron core, the weight of the low-voltage winding, the weight of the high-voltage winding, the oil discharging weight of each component and the weight of the accessories are respectively.
A c 、k c 、h 1 、h x The distance from the winding to the upper yoke, the distance from the winding to the lower yoke and the axial margin coefficient of the cross section of the iron core column are respectively set.
D、a 1 、b 1 、n 1 、a 2 、b 2 、m 2 、n 2 The diameter of the iron core, the width of the low-voltage wire, the thickness of the low-voltage wire, the number of wires of the low-voltage winding width, the width of the high-voltage wire, the thickness of the high-voltage wire, the number of wires of the high-voltage winding axial direction and the number of wires of the high-voltage winding width are respectively.
It should be noted that the total loss function is obtained by the following formula:
wherein f 2 (X) Total loss function. f (f) o 、P w The no-load loss and the load loss, respectively. K (K) po 、P o 、G Fe The loss is the additional coefficient of no-load loss, the unit loss of iron core and the weight of iron core. P (P) r 、P f 、P D The resistance loss, eddy current loss and lead loss of the high-voltage winding and the low-voltage winding are respectively. R is R L1 Is the resistance value of the low-voltage winding. R is R L2 Is the resistance value of the high-voltage winding. K (K) f 、K f1 The eddy current loss and the lead loss are, respectively, a percentage of the coil resistance loss, and are related to capacity and voltage class. r is the resistivity, typically copper at 75 degrees, 0.02135 (OMEGA mm 2/m). P (P) null And F load Indicating the values of no-load loss and load loss at that voltage level under national regulations. D. a, a 1 、b 1 、a 2 、b 2 、w 1 、w 2 、R p1 、R p2 、n 1 、m 2 、n 2 The winding wire winding structure comprises an iron core diameter, a low-voltage winding wire width, a low-voltage winding wire thickness, a high-voltage winding wire width, a high-voltage winding wire thickness, a low-voltage winding turn number, a high-voltage winding turn number, a low-voltage winding average radius, a high-voltage winding average radius, a low-voltage winding wire guiding number, a high-voltage winding axial wire guiding number and a high-voltage winding wire guiding number. L (L) 1 Is the length of the low voltage winding. L (L) 2 Is the length of the high voltage winding.
The temperature function includes an average temperature function of the winding surface to oil and a hottest temperature rise function of the sudden short-circuit instantaneous winding.
The average temperature function of the winding surface versus oil is obtained by the following formula:
wherein f 3 (X) As a function of the average temperature of the oil over the surface of the winding. q 1 、T X1 、q 2 、T X2 The temperature difference is the unit thermal load of the low-voltage winding surface, the average temperature difference of the low-voltage winding surface to oil, and the unit thermal load of the high-voltage winding surface, and the average temperature difference of the high-voltage winding surface to oil. P (P) w K is the load loss 1 、K 2 、K 3 、K 4 The coefficient related to the temperature material (copper wire is 22.1 at 85 ℃), the winding insulation correction coefficient, the total additional loss percentage of the wire and the covering coefficient of the wire cake are respectively shown. The parameter variables are: a, a 1 、b 1 、a 2 、b 2 、w 1 、R p1 、n 2 The winding wire width and the winding wire thickness of the low-voltage winding, the winding wire width and the winding wire thickness of the high-voltage winding, the number of turns of the low-voltage winding, the average radius of the low-voltage winding and the number of winding wires of the high-voltage winding are respectively. I N2 Is the rated current value of the high-voltage winding.
It should be noted that, the hottest temperature rise function of the winding is obtained by the following formula:
wherein f 4 (X) is the transient of burst short circuit, the hottest point temperature rising function of winding, theta 0 、θ 1 、θ 2 、V 1 、V 2 、I im1 、I im2 、U k 、k d The operation temperature of the transformer before short circuit, the average temperature of the hottest point of the low-voltage winding, the average temperature of the hottest point of the high-voltage winding, the short-circuit current density of the low-voltage winding, the short-circuit current density of the high-voltage winding, the short-circuit peak current of the low-voltage winding, the short-circuit peak current of the high-voltage winding, the short-circuit impedance percentage, the ratio of the maximum value to the stable value of the short-circuit current and the k of the high-capacity transformer are respectively shown d Taking 1.8.t is the duration of the short-circuit current. a, a 1 、b 1 、a 2 、b 2 The width of the low-voltage winding wire, the thickness of the low-voltage winding wire, the width of the high-voltage winding wire and the thickness of the high-voltage winding wire are respectively corresponding. I N1 Rated for low-voltage winding current, I N2 Rated for high voltage windings.
It should be noted that the short-circuit stress function is obtained by the following formula:
wherein f 5 And (X) is a short-circuit stress function, namely a high-low voltage winding amplitude-direction short-circuit force function. F (F) r1 、F r2 、σ 1 、σ 2 、σ a 、σ b 、e z Lambda is the low-voltage winding radial pressure, the high-voltage winding radial tension, the low-voltage winding radial compression stress, the high-voltage winding radial tension stress, the bending stress caused by the low-voltage winding axial force, the bending stress caused by the high-voltage winding axial force, the potential primary selection value of each turn and the total width of the winding magnetic leakage respectively.
K d 、K 1 ρ is the ratio of the maximum value and the stable value of the short-circuit current, the multiple of the stable value of the short-circuit current and the Rockwell coefficient respectively.
a 1 、b 1 、w 1 、R p1 、a 2 、b 2 、w 2 、R p2 、m 2 D corresponds to the width of the low-voltage winding wire, the thickness of the low-voltage winding wire, the number of turns of the low-voltage winding, the average radius of the low-voltage winding, the width of the high-voltage winding wire, the thickness of the high-voltage winding wire, the number of turns of the high-voltage winding, the average radius of the high-voltage winding, the axial number of the high-voltage winding wire and the diameter of the iron core respectively. U (U) 1Φ Rated for the low voltage winding. U (U) K Is the percent short circuit impedance. H K Is the axial total height of the coil.
It will be appreciated that the objective function is obtained by the following formula:
wherein F is min (X) is an objective function, X= [ D, a ] 1 ，b 1 ，n 1 ，a 2 ，b 2 ，m 2 ，n 2 ，w 1 ，w 2 ，R p1 ，R p2 ]。
The constraint conditions are as follows:
wherein B is the magnetic flux density of the core.
Step S200: establishing a multi-objective initial model according to the objective function, the optimization parameters and the optimization constraint;
in the step, a multi-target initial model is constructed according to the objective function, the optimization parameters and the optimization constraint, so that the multi-target initial model is solved in the subsequent step.
Step S300: generating an initial population, wherein the initial population comprises a plurality of random solutions and artificial solutions of a multi-target initial model, and the artificial solutions are obtained according to historical data of the transformer;
in the step, the initial population comprises a plurality of random solutions and artificial solutions of the multi-target initial model, the artificial solutions can be obtained according to the historical data and engineering experience of the transformer, and the addition of the artificial solutions to the initial population is beneficial to accelerating the convergence speed, improving the optimizing precision, overcoming the phenomena of premature and the like.
Step S400: taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, and obtaining a parameter optimization result according to the updated target parent population;
and executing an iteration strategy on the initial population by taking the initial population as a target parent population, stopping when the iteration times reach the preset iteration times, obtaining an updated target parent population through the iteration strategy, and obtaining a parameter optimization result according to the updated target parent population.
Step S500: obtaining a multi-objective optimization model according to the parameter optimization result and the multi-objective initial model;
in the step, a multi-objective optimization model is obtained by the parameter optimization result and the objective initial model.
As shown in fig. 2, the iterative strategy includes, but is not limited to, step S410, step S420, step S430, step S440, and step S450.
Step S410: non-dominant sorting is carried out on the target parent population, each artificial solution and each random solution in the target parent population are parent individuals, and the target parent population is divided into a dominant layer and a non-dominant layer;
step S420: calculating the crowdedness of each parent individual in the non-dominant layer;
in this step, the congestion level is expressed as the degree of density between a parent and an adjacent parent in the same layer, and the congestion distance is the absolute value of the difference between the objective function values. The formula is as follows:
n d for congestion distance, N ε 1 … N.Objective function value f for parent individual m Minimum value (min.)>Objective function value f for parent individual m Is the maximum value of (2); f (f) m (i+1),f m (i-1) is the objective function value of the parent individual before and after the ranking and the preceding ranking. And when the two parent individuals are in the same non-dominant level, judging whether the parent individuals are good or bad according to the crowding degree of the parent individuals, wherein the crowding degree is larger than the crowding degree. The crowding degree of the father individuals in the non-dominant layer is calculated, and the father individuals are ranked through crowding degree comparison, so that the risk of aggregation of a large number of father individuals in the evolution process can be reduced.
Step S430: screening out a plurality of parent individuals according to the non-dominant relationship and crowding degree of the parent individuals to obtain an intermediate parent population;
step S440: performing selection, crossing and mutation operations on the intermediate parent population to obtain a target offspring population;
step S450: and merging the intermediate parent population and the target offspring population to obtain a new target parent population.
In the embodiment, a multi-objective initial model is established according to a plurality of objective functions, optimization parameters and optimization constraints by determining the objective functions, the optimization parameters and the optimization constraints; and generating an initial population, taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, obtaining a parameter optimization result according to the updated target parent population, and obtaining a multi-target optimization model according to the parameter optimization result and the multi-target initial model. Compared with the traditional transformer multi-objective optimization algorithm, the transformer multi-objective optimization method provided by the embodiment of the application has the advantages that the artificial solution is added in the initial population, so that the convergence speed and the optimizing precision of the algorithm are improved, the premature phenomenon is overcome, and the phenomenon of sinking into a local optimal solution is avoided.
In accordance with an embodiment of the present application, the selection, crossover and mutation operations are performed on the intermediate parent population in step S440 to obtain the target offspring population, and as shown in fig. 3, step S440 includes, but is not limited to, step S441, step S442, step S443 and step S444.
Step S441: selecting the intermediate parent population by a selection operator to obtain a plurality of parent groups, wherein each parent group comprises two parent individuals;
step S442: performing crossover operation on the two parent individuals in each parent group through crossover operators to obtain a plurality of intermediate offspring individuals;
step S443: performing mutation operation on each intermediate offspring individual through a mutation operator to obtain a plurality of target offspring individuals;
step S444: and combining all the target offspring individuals to obtain a target offspring population.
In the step, a plurality of parent groups are selected from the intermediate parent population through a selection operator, two parent individuals in the parent groups are crossed to obtain intermediate child individuals, mutation operation is carried out on the intermediate child individuals through a mutation operator to obtain target child individuals, and the target child individuals are combined to obtain the target child population.
In accordance with some embodiments of the present application, further description is given of "performing a selection operation on the intermediate parent population by a selection operator to obtain a plurality of parent groups" in step S441. As shown in fig. 4, step S441 includes, but is not limited to, step S445 and step S446.
Step S445: selecting a first parent group from the intermediate parent population by a tournament selection algorithm;
step S446: the remaining parent individuals of the intermediate parent population are joined into a second set of parents.
In the step, the tournament algorithm adopts a sampling-back method, two different father individuals are repeatedly selected from the middle father population, the optimal father individual is selected through the tournament, and the selected optimal father individual is combined into a first double-parent group so as to conveniently execute the next cross operation. The remaining parent individuals that are not selected are joined into a second double-parent group.
In accordance with an embodiment of the present application, step S445 includes, but is not limited to, step S447, further describing "select first parent group from intermediate parent population by tournament selection algorithm" in step S445.
Step S447: the first parental group is selected from the intermediate parent population by a tournament selection algorithm having a tanh function as an activation function.
In the step, as the tournament selection algorithm adopts a put-back sampling method, selective put-back operation provides a large number of excellent parent individuals for subsequent crossover and mutation operation, but the parent individuals are easy to repeat, an individual accumulation pushing phenomenon is generated, the uniformity of solutions is poor, the solutions are trapped in local search areas, the efficiency is low, and the local convergence phenomenon is generated. Considering complex optimization constraints of the multi-objective initial model, population diversity needs to be increased to obtain solutions meeting the optimization constraints. Assuming that a parent individual A and a parent individual B in the intermediate parent population, the tournament selection algorithm judges that the parent individual A is superior to the parent individual B, and the parent individual A should be abandoned with a larger probability in the early stage of evolution, namely the probability that the parent individual A is selected is smaller, so that population diversity is facilitated, and the solution can meet complex optimization constraint. In the later period of evolution, the population gradually converges, the probability that the parent individual A is selected is increased, and the parent individual B is also left by the high probability, so that the population diversity is ensured. The tanh function is used as an activation function, namely, is used as a probability selection operator of the tournament selection algorithm, when the input value of the tanh function is smaller, the function value is smaller, the corresponding selection probability is smaller, when the input value is gradually increased, the function value is close to 1, and the corresponding probability also tends to 1. Population diversity can be increased by selecting a first parent group from the intermediate parent population using a tournament selection algorithm with the tanh function as the activation function.
the formula of the tanh function is as follows:
where P is the probability of being selected, β is the empirical coefficient of selection, and i is the algebra of evolution, i.e., the number of iterations.
In step S442, "crossover operations are performed on the two parent individuals in each parent group by crossover operators to obtain a plurality of intermediate child individuals" is further described according to an embodiment of the present application, and step S442 includes step S447 and step S448, as shown in fig. 5.
Step S447: performing crossover operation on two parent individuals in the first parent group through an arithmetic crossover operator to obtain a first intermediate offspring individual;
step S448: and performing crossover operation on the two parent individuals in the second double parent group through a multi-point crossover operator to obtain a second intermediate offspring individual.
In this step, in order to increase the selection space and accelerate the population evolution, the first parent group is crossed by using an arithmetic crossover operator, and the second parent group is crossed by using a multi-point crossover operator.
The formula of the arithmetic crossover operator is:
wherein X is I And X II Is a first intermediate offspring individual; x is X 1 And X is 2 Is a parent individual participating in the crossover; alpha is a random number between 0 and 1. Obviously, the offspring are generated by arithmetic crossover operators, and the new components thereof are still within the defined interval. And the second biparental group performs crossover operation by adopting a multi-point crossover operator to obtain a second intermediate offspring individual. The combination of the arithmetic crossover method and the multipoint crossover method can generate more excellent intermediate offspring individuals, increase the diversity of the population and improve the global searching capability.
In accordance with an embodiment of the present application, the "mutation operation is performed on each intermediate child individual by a mutation operator to obtain a plurality of target child individuals" in step S443 is further described, and step S443 includes, but is not limited to, step S449.
Step S449: and performing mutation operation on each intermediate child individual through a single-point mutation operator to obtain a plurality of target child individuals.
In the step, the principle of the single-point mutation operator is to realize the mutation of each gene value on the chromosome on the premise of not changing the sum of the total gene values of the chromosome according to a cyclic shift method for the selected genes. In order to control global and local search characteristics, a sigmoid function is adopted as a variation probability function of a single-point mutation operator, so that the initial variation probability of an algorithm is high, more local search characteristics are provided, premature and local optimal solutions are avoided, the later variation probability of the algorithm is low, more local search capacity is provided, and the convergence rate is ensured. The formula of the sigmoid function is as follows:
wherein P is 0 I is evolution algebra, N is preset iteration times and P is mutation probability max For the maximum probability of variation, α is the coefficient of variation constant.
In addition, an embodiment of the application also discloses a transformer multi-objective optimization device 100, as shown in fig. 6, including:
a data input module 110 for inputting a plurality of objective functions, optimization parameters including structural parameters of the transformer, and optimization constraints including at least industry standard constraints and material property constraints;
the model construction module 120 is configured to establish a multi-objective initial model according to the objective function, the optimization parameters and the optimization constraint;
a policy enforcement module 130 for generating an initial population, the initial population comprising a plurality of random solutions and artificial solutions of the multi-objective initial model, the artificial solutions derived from the transformer history data; taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, and obtaining a parameter optimization result according to the updated target parent population; the iterative strategy comprises the following steps: non-dominant sorting is carried out on the target parent population, each artificial solution and each random solution in the target parent population are parent individuals, and the target parent population is divided into a dominant layer and a non-dominant layer; calculating the crowdedness of each parent individual in the non-dominant layer; screening a plurality of parent individuals according to the non-dominant relationship of the parent individuals and the crowding degree to obtain an intermediate parent population; performing selection, crossing and mutation operations on the intermediate parent population to obtain a target offspring population; combining the intermediate parent population and the target offspring population to obtain a new target parent population;
and the result output module 140 is configured to output a multi-objective optimization model according to the parameter optimization result and the multi-objective initial model.
According to the transformer multi-objective optimization device 100 of the embodiment of the application, a multi-objective initial model is established according to a plurality of objective functions, optimization parameters and optimization constraints by determining the objective functions, the optimization parameters and the optimization constraints; and generating an initial population, taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, obtaining a parameter optimization result according to the updated target parent population, and obtaining a multi-target optimization model according to the parameter optimization result and the multi-target initial model. Compared with the traditional transformer multi-objective optimization technology, the transformer multi-objective optimization device 100 of the embodiment of the application increases the convergence speed and optimization precision of the algorithm due to the addition of the manual solution in the initial population, overcomes the premature phenomenon, and avoids sinking into the local optimal solution.
In addition, an embodiment of the present application also discloses a computer readable storage medium, in which a program executable by a processor is stored, where the program executable by the processor is used to implement the transformer multi-objective optimization method as described above.
According to the computer readable storage medium of the embodiment of the application, a multi-objective initial model is established according to a plurality of objective functions, optimization parameters and optimization constraints by determining the objective functions, the optimization parameters and the optimization constraints; and generating an initial population, taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, obtaining a parameter optimization result according to the updated target parent population, and obtaining a multi-target optimization model according to the parameter optimization result and the multi-target initial model. Compared with the traditional transformer multi-objective optimization technology, the computer readable storage medium of the embodiment of the application increases the convergence speed and optimizing precision of the algorithm due to the addition of the manual solution in the initial population, overcomes the premature phenomenon and avoids sinking into the local optimal solution.
Those of ordinary skill in the art will appreciate that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
The embodiments of the present application have been described in detail with reference to the accompanying drawings, but the present application is not limited to the above embodiments, and various changes can be made within the knowledge of one of ordinary skill in the art without departing from the spirit of the present application.
Claims (10)
1. The transformer multi-objective optimization method is characterized by comprising the following steps of:
determining a plurality of objective functions, optimization parameters and optimization constraints, wherein the optimization parameters comprise structural parameters of a transformer, and the optimization constraints at least comprise industry standard constraints and material characteristic constraints;
establishing a multi-objective initial model according to the objective function, the optimization parameters and the optimization constraint;
generating an initial population, wherein the initial population comprises a plurality of random solutions and artificial solutions of the multi-target initial model, and the artificial solutions are obtained according to the historical data of the transformer;
taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, and obtaining a parameter optimization result according to the updated target parent population;
obtaining a multi-objective optimization model according to the parameter optimization result and the multi-objective initial model;
the iterative strategy comprises the following steps:
non-dominant sorting is carried out on the target parent population, each artificial solution and each random solution in the target parent population are parent individuals, and the target parent population is divided into a dominant layer and a non-dominant layer;
calculating the crowdedness of each parent individual in the non-dominant layer;
screening a plurality of parent individuals according to the non-dominant relationship of the parent individuals and the crowding degree to obtain an intermediate parent population;
performing selection, crossing and mutation operations on the intermediate parent population to obtain a target offspring population;
and merging the intermediate parent population and the target offspring population to obtain a new target parent population.
2. The transformer multi-objective optimization method according to claim 1, wherein: the objective functions include a material cost function, a total loss function, a temperature function, and a short circuit stress function.
3. The transformer multi-objective optimization method according to claim 1, wherein: the optimized parameters comprise iron core diameter, low-voltage winding wire width, low-voltage winding wire thickness, high-voltage winding wire width, high-voltage winding wire thickness, low-voltage winding turns, high-voltage winding turns, low-voltage winding average radius, high-voltage winding average radius, low-voltage winding width wire number, high-voltage winding axial wire number and high-voltage winding width wire number.
4. The method of claim 1, wherein said performing the selecting, crossing and mutating operations on the intermediate parent population results in a target offspring population, comprising:
performing selection operation on the intermediate parent population through a selection operator to obtain a plurality of parent groups, wherein each parent group comprises two parent individuals;
performing crossover operation on the two parent individuals in each parent group through crossover operators to obtain a plurality of intermediate offspring individuals;
performing mutation operation on each intermediate offspring individual through a mutation operator to obtain a plurality of target offspring individuals;
and combining all the target offspring individuals to obtain the target offspring population.
5. The method of claim 4, wherein the selecting operation performed on the intermediate parent population by a selection operator to obtain a plurality of parent groups comprises:
selecting a first parent group from the intermediate parent population by a tournament selection algorithm;
and grouping the rest of the parent individuals of the intermediate parent population into a second double-parent group.
6. The method of transformer multi-objective optimization of claim 5, wherein the selecting a first parent group from the intermediate parent population by a tournament selection algorithm comprises:
a first parent group is selected from the intermediate parent population by a tournament selection algorithm having a tanh function as an activation function.
7. The transformer multi-objective optimization method according to claim 5, wherein the performing a crossover operation on the two parent individuals in each parent group by a crossover operator to obtain a plurality of intermediate child individuals comprises:
crossing the two parent individuals in the first parent group through an arithmetic crossing operator to obtain a first intermediate offspring individual;
and crossing the two parent individuals in the second double parent group through a multi-point crossing operator to obtain a second intermediate offspring individual.
8. The transformer multi-objective optimization method according to claim 5, wherein the performing a mutation operation on each of the intermediate child individuals by a mutation operator to obtain a plurality of target child individuals comprises:
and mutating each intermediate offspring individual by a single-point mutation operator to obtain a plurality of target offspring individuals.
9. The transformer multi-objective optimization device is characterized by comprising:
the data input module is used for inputting a plurality of objective functions, optimization parameters and optimization constraints, wherein the optimization parameters comprise structural parameters of the transformer, and the optimization constraints at least comprise industry standard constraints and material characteristic constraints;
the model construction module is used for constructing a multi-objective initial model according to the objective function, the optimization parameters and the optimization constraint;
the strategy execution module is used for generating an initial population, wherein the initial population comprises a plurality of random solutions and artificial solutions of the multi-target initial model, and the artificial solutions are obtained according to the transformer historical data; taking the initial population as a target parent population, executing an iteration strategy until the preset iteration times are reached, and obtaining a parameter optimization result according to the updated target parent population; the iterative strategy comprises the following steps: non-dominant sorting is carried out on the target parent population, each artificial solution and each random solution in the target parent population are parent individuals, and the target parent population is divided into a dominant layer and a non-dominant layer; calculating the crowdedness of each parent individual in the non-dominant layer; screening a plurality of parent individuals according to the non-dominant relationship of the parent individuals and the crowding degree to obtain an intermediate parent population; performing selection, crossing and mutation operations on the intermediate parent population to obtain a target offspring population; combining the intermediate parent population and the target offspring population to obtain a new target parent population;
and the result output module is used for outputting a multi-objective optimization model according to the parameter optimization result and the multi-objective initial model.
10. A computer readable storage medium, characterized in that a processor executable program is stored therein, which processor executable program is for implementing the transformer multi-objective optimization method according to any one of claims 1 to 8 when being executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310887023.8A CN116882293A (en) | 2023-07-18 | 2023-07-18 | Multi-objective optimization method and device for transformer and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310887023.8A CN116882293A (en) | 2023-07-18 | 2023-07-18 | Multi-objective optimization method and device for transformer and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116882293A true CN116882293A (en) | 2023-10-13 |
Family
ID=88254612
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310887023.8A Pending CN116882293A (en) | 2023-07-18 | 2023-07-18 | Multi-objective optimization method and device for transformer and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116882293A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104952067A (en) * | 2015-05-13 | 2015-09-30 | 西安电子科技大学 | Method for segmenting color images on basis of NSGA-II (non-dominated sorting genetic algorithm-II) evolution algorithms |
| CN109190241A (en) * | 2018-08-30 | 2019-01-11 | 哈尔滨工业大学 | Electromagnetic mechanism static characteristic optimization method |
| CN111079345A (en) * | 2019-12-20 | 2020-04-28 | 国网电力科学研究院武汉南瑞有限责任公司 | Optimal design method for double-layer shielding structure of extra-high voltage alternating current transformer sleeve |
| WO2021088436A1 (en) * | 2019-11-07 | 2021-05-14 | 苏宁云计算有限公司 | Project scheduling method and apparatus for multi-language collaborative development |
| CN114117938A (en) * | 2021-12-08 | 2022-03-01 | 吉林大学 | Adaptive multi-objective optimization algorithm based on NSGA-II framework |
| CN115310353A (en) * | 2022-07-26 | 2022-11-08 | 明珠电气股份有限公司 | Power transformer design method based on rapid multi-objective optimization |
| CN115861804A (en) * | 2022-11-28 | 2023-03-28 | 暨南大学 | SAR image countercheck sample detection system and method based on optimal feature attribution selection |
-
2023
- 2023-07-18 CN CN202310887023.8A patent/CN116882293A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104952067A (en) * | 2015-05-13 | 2015-09-30 | 西安电子科技大学 | Method for segmenting color images on basis of NSGA-II (non-dominated sorting genetic algorithm-II) evolution algorithms |
| CN109190241A (en) * | 2018-08-30 | 2019-01-11 | 哈尔滨工业大学 | Electromagnetic mechanism static characteristic optimization method |
| WO2021088436A1 (en) * | 2019-11-07 | 2021-05-14 | 苏宁云计算有限公司 | Project scheduling method and apparatus for multi-language collaborative development |
| CN111079345A (en) * | 2019-12-20 | 2020-04-28 | 国网电力科学研究院武汉南瑞有限责任公司 | Optimal design method for double-layer shielding structure of extra-high voltage alternating current transformer sleeve |
| CN114117938A (en) * | 2021-12-08 | 2022-03-01 | 吉林大学 | Adaptive multi-objective optimization algorithm based on NSGA-II framework |
| CN115310353A (en) * | 2022-07-26 | 2022-11-08 | 明珠电气股份有限公司 | Power transformer design method based on rapid multi-objective optimization |
| CN115861804A (en) * | 2022-11-28 | 2023-03-28 | 暨南大学 | SAR image countercheck sample detection system and method based on optimal feature attribution selection |
Non-Patent Citations (3)
| Title |
|---|
| 徐江雁等: "改进NSGA-Ⅱ算法及其在切削加工参数优化中的应用", 《计算机工程与应用》, vol. 53, no. 13, pages 227 - 234 * |
| 杨慧娜等: "基于改进的二代非支配排序遗传算法对电子变压器多目标优化", 《科学技术与工程》, vol. 15, no. 19, pages 139 - 145 * |
| 梁金琳等: "基于蚁群-遗传融合框架的仓储群机器人任务分配", 《计算机系统应用》, vol. 30, no. 11, pages 172 - 178 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP2006107425A (en) | Control method for photovoltaic power generation system | |
| CN107749630B (en) | Near-region power grid voltage coordination emergency control method of phase modulator | |
| CN113283073B (en) | Multi-objective optimization design method for three-phase high-frequency high-power transformer | |
| CN106446467A (en) | Optimal configuration method of fault current limiter based on adaptive particle swarm algorithm | |
| CN104484517B (en) | A kind of MMC bridge arms reactor parameter optimization method | |
| Fu et al. | Collection system topology for deep-sea offshore wind farms considering wind characteristics | |
| CN116882293A (en) | Multi-objective optimization method and device for transformer and storage medium | |
| CN106257477B (en) | A kind of intermediate frequency amorphous alloy transformer optimization method based on multi-objective genetic algorithm | |
| Ribeiro et al. | A Posteriori Strategy to Identify Robust Solutions in the Many-Objective Design Optimization of an Axial-Flux Permanent Magnet Synchronous Generator | |
| CN105743135A (en) | Extra-high voltage AC/DC optimal power receiving proportion evaluation method for receiving-end power grid | |
| Wang et al. | Site selection and capacity determination of substation based on power-weighted K-means | |
| Rao et al. | Optimal design of a furnace transformer by intelligent evolutionary methods | |
| CN109446657B (en) | Dynamic wireless power supply guide rail system of electric vehicle and parameter design method thereof | |
| Kobayashi et al. | SVR optimal placement problem in distribution network with Distributed Generators | |
| CN108988374B (en) | Optimal configuration method for position of flexible direct current interconnection device in power transmission section | |
| Zhao et al. | Optimum Design of Dry-Type Air-Gapped Iron-Core Reactor Based on Dynamic Programming and Circular Traversing Algorithm | |
| CN112560335B (en) | Multi-objective and structural parameter comprehensive optimization design method for transformer | |
| CN114420435B (en) | Method for designing cross section of mixed material wound core for transformer | |
| Watanabe et al. | Case Study of a 20 T-$\phi $400 mm Room Temperature Bore Superconducting Outsert for a 45 T Hybrid Magnet | |
| Ding et al. | Multi-objective optimal design of traction transformer using improved NSGA-II | |
| KR20070063204A (en) | Optimization designing method of transformer core | |
| CN220306090U (en) | Winding structure of short-circuit-resistant three-winding transformer | |
| CN117057291A (en) | Parameter determination method of toroidal inductor, toroidal inductor and buck-boost converter | |
| Zhao et al. | Research on application of genetic algorithm in optimization design of transformer | |
| CN115270607B (en) | Dry-type transformer optimization design method for nuclear power station by integrating expert experience |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |