CN-122021318-A - GIS metal particle catcher optimal location method based on hybrid optimization algorithm

CN122021318ACN 122021318 ACN122021318 ACN 122021318ACN-122021318-A

Abstract

The invention discloses a GIS metal particle catcher optimal location method based on a hybrid optimization algorithm, which comprises the steps of firstly establishing a 500kV GIS scaling model through COMSOL and carrying out simulation to obtain a three-dimensional drop point coordinate dataset of a metal particle track in the model, importing the dataset into MATLAB, constructing an optimization model taking the minimum sum of Euclidean distances from the catcher position to all particle drop points as a target, and finally adopting the hybrid optimization algorithm consisting of WOA (whale optimization algorithm), PSO (particle swarm optimization algorithm) and GWO (gray wolf optimization algorithm), searching an optimal point in a continuous three-dimensional space under the set boundary constraint condition, so that the capturing efficiency is maximized when the particle catcher is placed at the point. The invention solves the technical problems of difficult site selection and low capturing efficiency of the metal particle catcher in the GIS, remarkably improves the capturing coverage rate of the catcher on the metal particles, and provides reliable guarantee for safe and stable operation of GIS equipment.

Inventors

ZHONG LIPENG
CHAI JUNFENG
LIU ZULONG
YI SHUANG
WANG FENG
CHEN SHE
SUN QIUQIN

Assignees

湖南大学

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (9)

1. The optimal location method of the GIS metal particle catcher based on the hybrid optimization algorithm is characterized by comprising the following steps: Firstly, establishing a scaling model of a GIS, performing simulation analysis on electric field distribution in a cavity of the scaling model of the GIS to obtain an electric field distribution result, and then simulating the motion characteristics of metal particles in the GIS to obtain a three-dimensional motion track and drop point coordinates of the metal particles; step two, deriving three-dimensional coordinate data of the metal particle drop points obtained through simulation, performing data preprocessing, and constructing a standardized three-dimensional data set of the particle drop points after removing abnormal values; Step three, defining an optimized objective function taking 'the minimum Euclidean distance sum from the geometric center point of the catcher to the falling point of all particles' as a target based on an MATLAB platform, wherein the minimum value of the optimized objective function corresponds to the maximum value of the catching efficiency of the catcher; step four, configuring core parameters of at least three optimization algorithms, wherein the core parameters comprise population scale, maximum iteration times and search space boundaries; Step five, an initialization function is called to generate an initial population of the optimization algorithm, and the position of the initial population is ensured to be in a reasonable search space through a boundary checking mechanism; Step six, starting a mixed optimization algorithm iteration process, wherein the optimization algorithms independently search for optimal solutions respectively, and meanwhile, information interaction and collaborative optimization among the algorithms are realized through fitness value comparison in the iteration process; Step seven, after iteration is terminated, extracting an optimal solution obtained by searching the optimization algorithm, and obtaining a final optimal address coordinate of the catcher through weighted fusion; And step eight, installing a metal particle catcher at a corresponding position of the GIS equipment according to the final optimal address coordinates.
2. The optimal location method for GIS metal particle catcher based on hybrid optimization algorithm as claimed in claim 1, wherein in the first step, the method for simulating the motion characteristics of metal particles in GIS is as follows: (1.1) performing simulation analysis on electric field distribution in a GIS cavity according to parameters of the scaling model to obtain an electric field distribution result; The method comprises the steps of (1.2) carrying out collision detection based on a local motion track of the metal particles, judging whether the metal particles collide with a conductor or a shell of a GIS, if not, calculating acting force, motion speed and real-time position of the metal particles, and taking the position and speed of the current track end point as initial parameters of the next time step; (1.3) when the metal particles collide with the grounding shell or the high-voltage guide rod, recalculating the electric charge quantity of the particles, determining the rebound speed of the metal particles according to the collision rebound coefficient, and obtaining the calculation of acting force, movement speed and real-time position of the final particles; And (1.4) integrating each local motion track segment into a complete three-dimensional motion track of the metal particles by means of iterative computation with multiple time steps until a set simulation period is achieved.
3. The method of claim 2, wherein the motion state of the metal particles is defined as collision if the metal particles are in contact with the conductor or the shell of the GIS, and the motion state of the metal particles is defined as non-collision if the metal particles are in a space region between the grounding shell and the high-voltage guide rod; the motion state analysis equation of the metal particles is as follows: ; Wherein, the The mass of the metal particles, (r, theta) is the coordinates of the particles, The electric field force and dielectrophoresis force respectively, the included angle formed by the metal particles and the plane is the polar angle theta, and the gas resistance Fv borne by the metal particles is decomposed into components along the direction of the electric field force Component of direction perpendicular to electric field force R is the distance between the metal particles and the center of the high-pressure guide rod; the method comprises the steps of calculating a second derivative, wherein t represents time, G is a gravity constant, and F is the combination of metal particles; the metal particles are subjected to electric field force The method comprises the following steps: ; Wherein R 1 is the radius of the high-voltage guide rod, R 2 is the inner radius of the GIS cavity, k is the polarization coefficient of the metal particles, Is the vacuum dielectric constant, ln is a logarithmic function; dielectrophoresis force exerted on the metal particles at time t The method comprises the following steps: ; Wherein, the For the relative permittivity of the insulating gas C4F7N in the cavity, Is the radius of the metal particles; Is the gradient change of the electric field; the gas resistance of the metal particles at the time t The method comprises the following steps: ; ; ; Wherein, the As a coefficient of resistance (f) of the material, In order to achieve a gas density of the gas, For the reynolds number, Is the temperature The density of the C 4 F 7 N gas at the time, For the gas flow rate, In terms of the rate of movement of the metal particles, The velocity of the foreign metal relative to the gas at time t, Is the aerodynamic viscosity coefficient of C 4 F 7 N.
4. The optimal location method for a GIS metal particle catcher based on a hybrid optimization algorithm as claimed in claim 1, wherein in the second step, the data preprocessing is performed, and the outlier removing method is as follows: Duplicate drop point coordinates are deleted, so that data redundancy is avoided; outlier rejection, namely identifying and deleting abnormal coordinates beyond the normal distribution range by adopting a3 Be criterion.
5. The optimal location method for GIS metal particle catcher based on hybrid optimization algorithm as claimed in claim 1, wherein in the third step, the optimization objective function is as follows: ; Wherein the method comprises the steps of Is the three-dimensional coordinates of the geometric center of the trap, Is the three-dimensional coordinate of the i-th metal particle drop point in the GIS, n is the total number of metal particles, Is the sum of the Euclidean distances from the geometric center of the trap to all the landing points.
6. The modeling method of electromagnetic interference model of high frequency driving circuit in consideration of near field coupling characteristics according to claim 1, wherein the boundary checking mechanism is specifically as follows: (4.1) spatial boundary constraint, namely defining an upper boundary L b and a lower boundary U b of the optimized search space based on the physical size of the GIS scaling model and the distribution range of the particle landing points: L b =min(min(data))-δ; U b =max(max(data))+δ; wherein delta is a boundary allowance, and the value is 10% of the extreme value difference of coordinates of the falling point of the metal particles, namely delta=0.1× (max (data:) -min (data:)), ensuring that a search space completely covers a particle distribution area and accords with the limit of the internal installation space of GIS equipment; (4.2) constraint of variable dimensions, namely, the optimization variable is three-dimensional coordinates (x, y and z), the dimension dim=3, and the value of each dimension needs to meet all coordinates of x, y and z represented by L b ≤pos d ≤U b ,d=x,y,z;pos d ; (4.3) constraint of algorithm parameter boundaries, namely a population scale value range is 30-100, a maximum iteration number Max_iter value range is 200-500, and balance of algorithm convergence efficiency and optimizing precision is ensured.
7. The modeling method of electromagnetic interference model of high frequency driving circuit in consideration of near field coupling characteristic according to claim 1, wherein in the fourth step, the optimization algorithm comprises whale optimization algorithm WOA, particle swarm optimization algorithm PSO and gray wolf optimization algorithm GWO; In the sixth step, the iterative process of the hybrid optimization algorithm includes: the WOA algorithm updates the population position by surrounding three behaviors of predation, bubble network attack and random search, wherein dynamic switching of a search strategy is realized by linear decrease of a parameter alpha; The PSO algorithm adjusts global searching and local searching capacity through linear decrease of inertia weight omega, and updates particle speed and position by combining an individual optimal solution PBEST and a global optimal solution GBEST, wherein the speed boundary is limited to 20% of the searching space; The GWO algorithm updates the population position by simulating the social grades of the wolves, wherein the social grades comprise the first wolves, the suboptimal wolves and the third suboptimal wolves, and the linear change of the parameter alpha is utilized to balance the exploration and development capacity.
8. The modeling method of electromagnetic interference model of high frequency driving circuit according to claim 1, wherein the weighting coefficients of the weighted fusion in the seventh step are determined based on convergence speed and iteration stability of three algorithms, the weighting coefficients of WOA, PSO, GWO are respectively set to 0.35, 0.35 and 0.3, and the final optimal coordinates are pos opt =0.35×pos WOA +0.35×pos PSO +0.3×pos GWO , wherein pos WOA 、pos PSO 、pos GWO is the optimal coordinates obtained by independent searching of the three algorithms.
9. The modeling method of the electromagnetic interference model of the high-frequency driving circuit considering near-field coupling characteristics according to claim 1, wherein the scaling model of the GIS is a COMSOL 500kV GIS scaling model.

Description

GIS metal particle catcher optimal location method based on hybrid optimization algorithm Technical Field The invention relates to the field of high-voltage transmission of power systems, in particular to a GIS metal particle catcher optimal location method based on a hybrid optimization algorithm. Background The Gas Insulated Switchgear (GIS) is core equipment of an electric power system, and is widely used for high-voltage power transmission networks of 500kV and above because of small occupied area, excellent insulation and high reliability. However, metal particles are generated during the production, assembly and operation of the high-voltage transformer, and the high-voltage transformer can migrate, suspend and even discharge under the action of an electric field, so that the insulation performance is threatened, and the failure can be caused to cause economic loss. In order to solve the potential safety hazard of metal particles, a particle catcher is often installed in a GIS in the prior art, and the metal particles are caught through physical adsorption or electric field traction. However, because the GIS internal structure is complex and the motion trail of the metal particles is random and uncertain, the installation position of the catcher determines the catching efficiency. The traditional site selection method is based on experience judgment or local test, has strong subjectivity and low capture coverage rate, and a single optimization algorithm is used for selecting a local optimal solution which is easy to collapse, so that a global optimal point position is difficult to find in a continuous three-dimensional space. Therefore, the realization of scientific site selection and maximized capture efficiency of the catcher is a technical problem to be solved urgently in the field of GIS equipment operation and maintenance. Disclosure of Invention In order to solve the problems, the invention provides a GIS metal particle catcher optimal location method based on a hybrid optimization algorithm. In order to achieve the above purpose, the technical scheme of the invention is as follows: a GIS metal particle catcher optimal location method based on a hybrid optimization algorithm comprises the following steps: Firstly, establishing a scaling model of a GIS, performing simulation analysis on electric field distribution in a cavity of the scaling model of the GIS to obtain an electric field distribution result, and then simulating the motion characteristics of metal particles in the GIS to obtain a three-dimensional motion track and drop point coordinates of the metal particles; step two, deriving three-dimensional coordinate data of the metal particle drop points obtained through simulation, performing data preprocessing, and constructing a standardized three-dimensional data set of the particle drop points after removing abnormal values; Step three, defining an optimized objective function taking 'the minimum Euclidean distance sum from the geometric center point of the catcher to the falling point of all particles' as a target based on an MATLAB platform, wherein the minimum value of the optimized objective function corresponds to the maximum value of the catching efficiency of the catcher; step four, configuring core parameters of at least three optimization algorithms, wherein the core parameters comprise population scale, maximum iteration times and search space boundaries; Step five, an initialization function is called to generate an initial population of the optimization algorithm, and the position of the initial population is ensured to be in a reasonable search space through a boundary checking mechanism; Step six, starting a mixed optimization algorithm iteration process, wherein the optimization algorithms independently search for optimal solutions respectively, and meanwhile, information interaction and collaborative optimization among the algorithms are realized through fitness value comparison in the iteration process; Step seven, after iteration is terminated, extracting an optimal solution obtained by searching the optimization algorithm, and obtaining a final optimal address coordinate of the catcher through weighted fusion; And step eight, installing a metal particle catcher at a corresponding position of the GIS equipment according to the final optimal address coordinates. In the first step, the method for simulating the motion characteristics of the metal particles in the GIS includes: (1.1) performing simulation analysis on electric field distribution in a GIS cavity according to parameters of the scaling model to obtain an electric field distribution result; The method comprises the steps of (1.2) carrying out collision detection based on a local motion track of the metal particles, judging whether the metal particles collide with a conductor or a shell of a GIS, if not, calculating acting force, motion speed and real-time position of the metal particles, and taking the position