A light guide rod structure optimization design method, device and medium

Abstract

The present application relates to a kind of light guide rod structure optimization design method, equipment and medium, wherein the method comprises the following steps: constructing TracePro simulation model, simulating the spatial distribution of partial discharge point light source and the multi-path optical propagation process based on actual GIS equipment structure and material parameters;Define the structure parameters of light guide rod, parameterized structure modeling is carried out to light guide rod, and multi-parameter joint optimization space is constructed;Based on QAOA quantum optimization algorithm, the light flux of the exit end of light guide rod determined based on TracePro simulation model is used as objective function, and the optimal search of the structure parameter combination of light guide rod is carried out.Compared with prior art, the present application can efficiently and reliably optimize the optical detection structure of light guide rod, and significantly improve the capture ability of light guide rod to the partial discharge light signal in GIS.

Description

Technical Field

[0001] This invention relates to the field of GIS insulation defect detection, and in particular to a method, equipment and medium for optimizing the design of a light guide rod structure. Background Technology

[0002] Gas-insulated switchgear (GIS) is widely used in modern power transmission and distribution systems due to its advantages such as small size, high insulation performance, good safety and reliability, and convenient maintenance. However, during long-term operation, the insulation performance of GIS inevitably deteriorates, and one of the most significant and harmful problems is partial discharge (PD). Partial discharge is an important indicator of internal insulation defects in GIS equipment. If it is not detected and diagnosed in a timely and accurate manner, it will gradually lead to damage to the equipment's insulation materials, seriously affecting the operational safety and stability of the power system, and even causing major electrical accidents. Therefore, accurate and efficient detection of partial discharge in GIS equipment has extremely important engineering significance and economic value.

[0003] Currently, conventional methods for partial discharge detection mainly include ultra-high frequency (UHF) detection and ultrasonic detection. Although these traditional detection techniques have achieved certain results in practical applications, they are susceptible to external electromagnetic and acoustic interference in complex electromagnetic environments, leading to reduced detection sensitivity and accuracy. Furthermore, the complex structure and high airtightness of GIS equipment make it difficult for these traditional detection methods to simultaneously achieve both detection sensitivity and interference suppression performance, thus limiting their effectiveness in practical applications.

[0004] In recent years, partial discharge detection methods based on optical principles have gradually attracted widespread attention from academia and industry. Compared with traditional electromagnetic and ultrasonic detection methods, optical detection technology has significant advantages, such as strong resistance to electromagnetic and acoustic interference, low signal transmission loss, and high sensitivity. Therefore, optical detection technology is considered a promising emerging approach in the field of partial discharge detection in GIS equipment. For example, Chinese patent CN115980520A discloses a GIS partial discharge optical detection and positioning simulation method, system, medium, and terminal. It simulates actual partial discharge and performs optical detection. By establishing a functional relationship between the irradiance value at the corresponding detection location and the location of the partial discharge source, the location of the partial discharge source can be calculated by reverse calculation when the irradiance at the detector location is obtained. Currently reported optical detection methods mainly include detection methods based on fluorescent fiber optic sensors and interferometry based on fiber optic ultrasonic detection. However, although fluorescent fibers have shown certain performance advantages in laboratory research, their material is relatively soft and lacks mechanical strength, which may lead to a decrease in the sealing performance of GIS equipment during long-term operation, posing certain safety risks. While fiber optic ultrasonic testing technology has high sensitivity, it does not directly detect optical PD signals and is still easily affected by environmental acoustic vibrations, thus limiting its reliability and accuracy.

[0005] Against this backdrop, light guide rods (LGRs), characterized by high mechanical strength, excellent light transmission properties, and ease of sealed installation, are gradually becoming an important development direction for partial discharge optical detection technology in GIS. Light guide rods utilize highly transparent and high-hardness materials (such as polymethyl methacrylate, PMMA) to effectively capture and transmit the light signals generated by partial discharge through internal total internal reflection. However, existing light guide rod structural designs suffer from insufficient light signal collection efficiency, and their structural parameters (such as the microstructure design at the front end and length) have not yet been systematically optimized, failing to fully realize the potential of light guide rods in optical detection.

[0006] Therefore, developing an efficient and reliable method for optimizing the optical detection structure of light guide rods to significantly improve the ability of light guide rods to capture partial discharge optical signals in GIS is an important problem that urgently needs to be solved in the field of GIS partial discharge detection. Summary of the Invention

[0007] This invention aims to address the problems of lack of systematic optimization in the design of light guide structures, low structural response efficiency, and discrepancies between simulation evaluation and actual performance in existing optical detection of partial discharge in GIS. It proposes a method, device, and medium for optimizing the design of light guide rod structures based on quantum optimization algorithms and TracePro optical simulation. By establishing a high-fidelity TracePro optical propagation model and combining it with the Quantum Approximate Optimization Algorithm (QAOA) to jointly optimize the microstructure and geometric parameters of the light guide rod, the collection efficiency and directionality of the light guide rod for partial discharge optical signals can be significantly improved, providing a reliable design basis for the high-sensitivity deployment of optical sensors in GIS.

[0008] The objective of this invention can be achieved through the following technical solutions:

[0009] According to a first aspect of the present invention, a method for optimizing the design of a light guide rod structure is provided, the method comprising the following steps:

[0010] A TracePro simulation model was constructed to simulate the spatial distribution and multipath optical propagation process of partial discharge point sources based on the actual GIS equipment structure and material parameters.

[0011] Define the structural parameters of the light guide rod, perform parametric structural modeling of the light guide rod, and construct a multi-parameter joint optimization space;

[0012] Based on the QAOA quantum optimization algorithm, the optimal combination of structural parameters of the light guide rod is searched with the light flux at the output end of the light guide rod determined by the TracePro simulation model as the objective function.

[0013] The construction of the TracePro simulation model specifically includes the following steps:

[0014] Based on the ray tracing algorithm, a partial discharge optical signal propagation model inside a GIS device is constructed in the TracePro environment. In the partial discharge optical signal propagation model, discharge sources are set on multiple planes in the circumferential and radial directions of the shell, covering positions near the conductor, near the shell, and the central area. Multiple discharge sources with different radii are set on each plane, and each point source simulates the luminous position generated by a partial discharge defect. The light guide rod adopts an embedded design that is tightly connected to the GIS structure and is installed at the preset orifice position of the simulation model. The light guide rod has a columnar structure inside, and multiple switchable microstructures are configured at the front end. The light guide rod is flush with the inner wall of the GIS to ensure that it achieves maximum optical signal coupling without disturbing the electric field distribution of the device.

[0015] To evaluate the light signal collection capabilities of different light guide rod structures, independent simulations were performed at each discharge power source location. A receiving surface was set at the exit end of the light guide rod. The total luminous flux value was calculated and standardized by counting all photons guided to the receiving surface through internal reflection, forming a dimensionless evaluation index. Simulations of each light guide rod structure were run independently at all discharge power source locations, and the corresponding average standardized luminous flux was used as the overall performance index of the structure under complex discharge environments.

[0016] In the TracePro simulation model, the optical signal generated by partial discharge in GIS is modeled as a point light source with uniform spherical emission. Each simulated point source emits a preset number of rays, which are uniformly distributed in three-dimensional space according to the spherical coordinate system to ensure that sufficient emission angles and propagation paths are covered in the simulation process.

[0017] The structure of the light guide rod includes a front-end receiving structure and a planar end-face structure, wherein,

[0018] The front-end receiving structure is used for the initial reception and incident control of light. Its shape determines the incident angle distribution, beam focusing behavior and the initial conditions of subsequent multiple reflection paths in terms of optical characteristics. The front-end shape design is based on five typical structures, namely planar end, concave cone, concave sphere, convex cone and convex sphere structure. All structures maintain an axially symmetrical design in their overall shape to adapt to the geometric fit conditions of the GIS equipment mounting holes.

[0019] The planar end face structure serves as the basic reference model. It adopts a smooth plane with the same diameter as the rod and is set as an ideal model without incident angle adjustment capability. It is used to evaluate the performance improvement of other structures.

[0020] In the parametric modeling of planar end face structures, for concave and convex structures, key geometric features are defined through parametric methods to achieve continuous and adjustable structure in generalized parameter space; for conical structures, the vertex angle is used as the main variable, and the value range and step size are set to generate several discrete samples; for spherical structures, the radius of curvature is set as the core variable, while ensuring the continuity of the tangent at the transition between the spherical part and the rod.

[0021] The structural parameters of the light guide rod include front end shape design parameters, planar end face structure parameters, main body length of the light guide rod, and diameter of the light guide rod. The main body length of the light guide rod is set to multiple discrete values ​​according to a preset length range and step size, and the diameter of the light guide rod is set to a constant value.

[0022] The parametric structural modeling of the light guide rod specifically involves:

[0023] The structural parameters of all light guide rods are combined through the Cartesian product method to form a high-dimensional parameter space. Each set of structural parameter combinations corresponds to a unique geometric structure instance. After discretization encoding, it is converted into binary variables and uniformly represented as a bit string z = (z1, z2,..., z n ) ∈ {0, 1} n , where n is the number of structural parameters. Each set of z corresponds to a specific light guide rod structure, and its corresponding 3D model is automatically generated by the program for input into the TracePro simulation model for ray tracing simulation. At the same time, to prevent the explosion of the search space dimension, some invalid or low-correlation parameter regions are cropped, and a subspace with representativeness, obvious performance differences, and feasible processing is retained as the source of the optimization initial population.

[0024] The described QAOA quantum optimization algorithm performs the following steps to achieve the optimal search for the structural parameter combination of the light guide rod:

[0025] Construct the quantum state of QAOA:

[0026]

[0027] Among them, is the diagonal Hamiltonian corresponding to the objective function C(z), and its form is: z ∈ {0, 1} n is the bit string sample, and C(z)|z><z| is the operator projected onto the bit string z and is given its corresponding objective function value C(z); is the driving Hamiltonian, which consists of Pauli-X operators, and X i is the Pauli-X operator, and n is the number of structural parameters; are the trainable variational parameters; p is the algorithm depth, that is, the number of layers of QAOA;

[0028] Iterative optimization:

[0029] The quantum computing part of the QAOA quantum optimization algorithm generates candidate solutions in the superposition space, and the classical optimizer iteratively updates the variational parameters to maximize the expected value of the objective function corresponding to the solution samples obtained by quantum measurement:

[0030]

[0031] For each bit string sample z ∈ {0, 1} obtained by measurement of the QAOA quantum optimization algorithm n , it is decoded into the structural parameter combination of the light guide rod through a preset structural parameter dictionary;

[0032] Based on the combination of decoded light guide rod structural parameters, a three-dimensional geometric model is generated by calling the CAD modeling interface module and input into the TracePro simulation model for ray tracing simulation.

[0033] The luminous flux value output by the TracePro simulation model is returned to the QAOA quantum optimization algorithm as a feedback signal of the objective function C(z);

[0034] The QAOA quantum optimization algorithm uses a weighted average of the luminous flux values ​​from multiple samples as the current variational parameter. The performance metrics are evaluated, and the classic optimizer is called to update the variational parameters for the next iteration.

[0035] According to a second aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described thereon.

[0036] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described thereon.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. Improved optical modeling accuracy: This invention establishes a high-fidelity optical simulation model in the TracePro environment, which can fully reproduce the spatial distribution and multi-path propagation process of partial discharge optical signals inside GIS. Compared with traditional methods, it has higher realism and reliability.

[0039] 2. Systematic optimization of structural parameters: This invention adopts a parametric modeling method to incorporate key design parameters such as the microstructure morphology, radius of curvature, cone angle, and length of the light guide rod front end into a multi-dimensional optimization space, thereby enabling a systematic exploration of different structural combinations and overcoming the limitations of existing technologies that rely on experience for selection and have insufficient optimization.

[0040] 3. Introducing a quantum optimization algorithm to improve search efficiency: This invention introduces the QAOA quantum approximation optimization algorithm to conduct a global optimization search in a high-dimensional discrete combination parameter space, which has the advantages of parallel computing and fast convergence. Compared with traditional optimization methods such as genetic algorithms and particle swarm optimization, it can obtain a light guide rod structure with better performance in a shorter time.

[0041] 4. Enhanced light signal capture capability: The optimized light guide rod structure exhibits higher light flux output in both simulation and experimental verification, significantly improving the collection efficiency and directionality of partial discharge light signals in GIS, thereby effectively enhancing detection sensitivity.

[0042] 5. Improved stability and anti-interference performance: The optimized light guide rod can still maintain excellent light signal capture performance in complex electromagnetic and acoustic interference environments, and has stronger stability and reliability compared with traditional detection methods.

[0043] 6. High degree of automation and engineering feasibility: This invention establishes an automated closed-loop process of CAD modeling—TracePro simulation—QAOA optimization, which can quickly generate and verify light guide rod structure design schemes, ensuring that the results not only have theoretical feasibility, but also meet the engineering application requirements of actual processing and installation. Attached Figure Description

[0044] Figure 1 This is a flowchart of the method of the present invention;

[0045] Figure 2 This is a schematic diagram of the TracePro simulation model of a GIS device in one embodiment of the present invention;

[0046] Figure 3 This is a schematic diagram of the spectrum of a point light source in one embodiment of the present invention;

[0047] Figure 4 This is a schematic diagram of the discharge power source position in one embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram of multiple front-end receiving structures of a light guide rod in one embodiment of the present invention;

[0049] Figure 6 This is a schematic diagram of the iterative optimization process in one embodiment of the present invention;

[0050] Figure 7 This is a GIS partial discharge optical testing platform in one embodiment of the present invention;

[0051] Figure 8 This is a schematic diagram showing the relative positions of the light guide rod and the local discharge power source in one embodiment of the present invention;

[0052] Figure 9 This is a simulation comparison diagram of a light guide rod in one embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0054] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," and "third" used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0055] Example 1

[0056] This embodiment provides a method for optimizing the design of a light guide rod structure, such as... Figure 1 As shown, the method includes the following steps:

[0057] S1. Construct a TracePro simulation model to simulate the spatial distribution and multipath optical propagation process of partial discharge point sources based on the actual GIS equipment structure and material parameters.

[0058] First, based on the ray tracing algorithm, a partial discharge optical signal propagation model inside a gas-insulated switchgear (GIS) is constructed in the TracePro environment. The geometry used in the simulation is derived from the standard physical parameters of an actual GIS device, such as... Figure 2 As shown, the simulated object is a typical single-compartment GIS system, with a conductor radius of 45mm, a shell radius of 190mm, a cavity thickness of 8mm, a light guide rod insertion port diameter of 200mm, and a total length of 1250mm.

[0059] To reproduce the spatial distribution characteristics of partial discharge in actual equipment, the model sets discharge sources on multiple planes in the circumferential and radial directions of the casing, covering locations near the conductor, near the casing, and in the central region. Three discharge sources with different radii are set on each plane, with each point source simulating the luminescence location generated by a partial discharge defect.

[0060] like Figure 3 As shown, the optical signal generated by partial discharge in GIS is modeled as a point source with uniform spherical emission, ranging from 200nm to 800nm, with the main peak distributed between 300–500nm, using a white light approximation model. Each simulated point source emits 10... 5 A single ray is uniformly distributed in three-dimensional space according to a spherical coordinate system to ensure that the simulation covers sufficient emission angles and propagation paths. The internal filling medium of the GIS is sulfur hexafluoride (SF6) gas, with an optical refractive index set to 1.000783 and a negligible absorption coefficient. The surface material is alumina, and its composite reflection characteristics are characterized by a two-way reflection distribution function model, in which the specular reflection component is 0.2, the diffuse reflection component is 0.5, and the absorption component is 0.3.

[0061] The light guide rod employs an embedded design that is tightly integrated with the GIS structure, installed at the pre-set aperture position in the simulation model. It is made of polymethyl methacrylate (PMMA), which possesses excellent optical transparency and mechanical strength, with a refractive index of 1.49, effectively supporting total internal reflection for long-distance light guiding. The light guide rod has an internal columnar structure, with multiple switchable microstructure morphologies at its front end, including concave cones, concave spheres, convex cones, convex spheres, and planar ends. All front-end microstructures are defined through parametric modeling and serve as input variables for subsequent optimization algorithms. The light guide rod is 128mm long, flush with the inner wall of the GIS, to ensure maximum optical signal coupling without disturbing the equipment's electric field distribution.

[0062] The absorption and reflection characteristics of optical signals on a surface depend not only on the container material but also on the incident angle of the light, i.e., the location of the discharge source. Since partial discharge can occur anywhere within the cavity, in addition to setting material parameters, it is necessary to simulate discharge sources at different locations. This embodiment sets up four discharge source planes, denoted as I1, I2, I3, and I4. The locations of the discharge sources on each plane are as follows: Figure 4 As shown, y+45, y+115, and y+185 represent 45mm, 115mm, and 185mm in the positive y-axis direction, respectively, and the same applies to the x-axis. A total of 36 discharge points were set, covering the areas near the guide rod, near the shell, and in the middle of the cavity, making the experimental data more comprehensive and accurate.

[0063] To evaluate the light signal collection capabilities of different light guide rod structures, independent simulations were performed at each discharge point. A receiving surface was set at the exit end of the light guide rod, and the light flux guided to this port through total internal reflection was recorded. TracePro calculated the total luminous flux (Luminous Flux, LX) by counting the number of photons incident on the exit surface, and then standardized this value to form a dimensionless evaluation index. Simulations for each light guide rod structure were run independently at all discharge point locations, and the corresponding average luminous flux was used as the overall performance index of the structure under complex discharge environments. The simulation output data was used to construct the objective function, serving as input for the subsequent quantum optimization stage to guide the structural tuning of the light guide rod's geometric parameters. Through this modeling process, high-fidelity quantitative characterization of the optical performance of the light guide structure can be achieved, providing a physical basis for multi-parameter structural optimization.

[0064] S2 defines the structural parameters of the light guide rod, performs parametric structural modeling of the light guide rod, and constructs a multi-parameter joint optimization space.

[0065] After completing the TracePro simulation modeling of the partial discharge optical signal within the GIS, parametric structural modeling of the light guide rod (LGR) is required to further optimize the structure, and the boundaries of its design variables and search space must be defined. As the main acquisition and transmission channel for the partial discharge optical signal, the structural geometry of the light guide rod significantly affects the coupling efficiency, guidance path stability, and luminous flux output at the end of the signal. To enable systematic exploration of different structural configurations in subsequent optimization processes, the adjustable geometry and parameter dimensions of the light guide rod must be precisely defined first, and a physically feasible and engineering-controllable multidimensional initial design space must be constructed based on this.

[0066] The structure of the light guide rod includes a front-end receiving structure and a planar end-face structure, wherein,

[0067] The front-end receiving structure is used for the initial reception and incident control of light. Its shape, in terms of optical characteristics, determines the incident angle distribution, beam focusing behavior, and the initial conditions of subsequent multiple reflection paths. Based on experimental and simulation analysis results, the front-end shape design uses five typical structures as basic configurations: planar end, concave cone, concave sphere, convex cone, and convex sphere structures, such as... Figure 5 As shown, all structures maintain an axisymmetric design in their overall shape to accommodate the geometric fit conditions of the GIS equipment mounting holes;

[0068] The planar end-face structure serves as the basic reference model, employing a smooth plane of the same diameter as the rod. This ideal model, devoid of incident angle adjustment capabilities, is used to evaluate the performance improvements of other structures. In the parametric modeling of the planar end-face structure, key geometric features are defined parametrically for concave and convex structures to achieve continuous adjustability in the generalized parameter space. For conical structures, the apex angle is used as the main variable, ranging from 30° to 120°, with discrete samples generated in 15° increments. For spherical structures, the radius of curvature is set as the core variable, ranging from 2mm to 15mm in 1mm increments, while ensuring tangent continuity at the transition between the spherical portion and the rod to avoid abnormal interface reflections.

[0069] In the actual modeling process, all microstructures are generated into solid models through parameter-driven CAD modeling, which facilitates direct conversion with simulation software interfaces.

[0070] The main body length of the light guide rod is also included in the optimization variables. This part mainly affects the multiple reflection processes of light inside the rod and the spatial matching between it and the exit end face. In this embodiment, the length is introduced as an independent variable L into the optimization model. The length range is set from 95mm to 130mm, with a step size of 5mm, covering three typical installation states: slightly shorter than the inner wall, flush with the inner wall, and moderately extending into the GIS shell. To ensure that the electric field distribution is not disturbed, the safety of use for each length configuration is verified in the simulation with reference to the electric field simulation boundary conditions.

[0071] In this embodiment, the diameter of the light guide rod remains constant at 25mm to match the aperture size of the GIS cavity. Polymethyl methacrylate (PMMA) is used as the material, and its optical parameters are fixed in the simulation as a refractive index of 1.49 and a transmittance of 93%. Its machinability and weather resistance meet the requirements of actual engineering use. The surface finish is set to an optical polishing level to avoid light scattering loss at the interface.

[0072] To construct a complete optimization input space, all variable parameters are combined using Cartesian products to form a high-dimensional parameter space. Each parameter combination corresponds to a unique geometric structure instance, and its corresponding 3D model can be automatically generated by the program for input into TracePro for ray tracing simulation. Since the number of parameter combinations theoretically grows exponentially, to prevent the dimensionality of the search space from exploding, this method, based on previous simulation sensitivity analysis and engineering experience, prunes some invalid or low-relevance parameter regions, retaining representative subspaces with significant performance differences and feasible fabrication as the initial population source for optimization. The resulting structural design space not only possesses sufficient expressive power to support the global search characteristics of the optimization algorithm but also achieves a balance between complexity and physical constraints, ensuring that each candidate structure has good engineering applicability and simulation stability.

[0073] This design space serves as the variable input for subsequent quantum optimization stages. Its well-defined geometric boundaries and modeling parameter system provide a reliable foundation for the effective search and interpretation of results, and also ensure the consistency and comparability of the structure in multi-objective simulation evaluations. Through this systematic modeling and spatial definition approach, the structural evolution of the light guide rod can be accurately characterized and gradually approach the optimal solution.

[0074] S3, based on the QAOA quantum optimization algorithm, uses the light flux at the output end of the light guide determined by the TracePro simulation model as the objective function to perform the optimal search for the combination of structural parameters of the light guide.

[0075] After defining the structural parameter space of the light guide rod, this invention introduces the Quantum Approximate Optimization Algorithm (QAOA) to find the global optimum in a high-dimensional combinatorial parameter space in order to achieve the optimal configuration search for structural performance. QAOA is a hybrid quantum-classical variational algorithm, particularly suitable for solving combinatorial optimization problems of the maximization or minimization class. Its core is to construct a quantum Hamiltonian defined by the objective function and search for parameter combinations that maximize the objective value in the quantum state space, thereby approximating the optimum in an efficient and parallel manner.

[0076] In this embodiment, all structural parameters of the light guide rod (such as microstructure morphology, cone angle, radius of curvature, and length) are discretized and encoded into binary variables, and uniformly represented as a bit string of length n, z = (z1, z2, ..., z...). n )∈{0,1} n, where n is the number of structural parameters. Each set of z corresponds to a specific light guide rod structure, and its 3D model can be generated in the simulation platform TracePro for performance evaluation. The goal is to maximize the light flux C(z) of this structure under a specific partial discharge position, which is then transformed into a standard maximization problem.

[0077] Subsequently, the QAOA quantum optimization algorithm performs the following steps to achieve the optimal search for the combination of structural parameters of the light guide rod:

[0078] 1) Construct the quantum state of QAOA:

[0079]

[0080] where is the diagonal Hamiltonian corresponding to the objective function C(z), in the form of: z ∈ {0, 1} n is the bit string sample, and C(z)|z><z| is the operator projected onto the bit string z, and the corresponding objective function value C(z) is assigned to it; is the driving Hamiltonian, composed of Pauli-X operators, X i is the Pauli-X operator, and n is the number of structural parameters; are the trainable variational parameters; p is the depth of the algorithm, that is, the number of layers of QAOA, which determines the ability of the quantum state to approximate the optimal solution.

[0081] 2) Iterative optimization:

[0082] The optimization process of QAOA is hybrid, that is: the quantum computing part is responsible for generating candidate solutions in the superposition space, and the classical optimizer is responsible for iteratively updating the variational parameters to maximize the expected value of the objective function corresponding to the solution samples obtained by quantum measurement:

[0083]

[0084] In practical implementation, the quantum circuit in this embodiment does not rely on a real quantum processor, but runs on a high-performance quantum simulator. A quantum simulator is a software system that simulates the behavior of quantum circuits on a classical computing platform using methods such as tensor networks or matrix product states. Typical examples include IBM QiskitAer, Google Cirq Simulator, Amazon Braket Simulator, PennyLane, or QuEST. Considering the limited circuit depth (usually set to p = 1–3) and the number of optimization variables n generally controlled within 20 bits, this embodiment chooses QiskitAer Simulator as the running platform, which has good compatibility and high simulation accuracy.

[0085] like Figure 6 As shown, the optimization process includes: first, defining structural parameters; then, obtaining a bit string sample through QAOA circuit and quantum measurement, decoding it into structural parameters, generating a geometric model, and transmitting it to the TracePro platform for ray tracing; updating variational parameters through a classical optimizer and feeding them back to the QAOA circuit; and achieving structural optimization of the light guide rod through iterative processes.

[0086] Each bit string sample z∈{0,1} n After being measured by QAOA, the data needs to be decoded into specific combinations of light guide rod structural parameters. To this end, this embodiment pre-defines a parameter mapping dictionary, mapping each segment of the bit string to a specific geometric parameter. Let n = 10, where bits 1–3 are used to encode the type of the front-end microstructure (e.g., five types of structures such as concave cones and convex spheres), bits 4–6 are used to encode the cone angle or the radius of curvature of the sphere, and the remaining bits are used to encode the length of the light guide rod. This decoding process is automatically executed by a Python script, translating each bit string into a set of numerical parameters θ = {θ1, θ2, ..., θ...}. k Each parameter corresponds to a geometric dimension (e.g., a vertex angle of 60°, a length of 125mm, a radius of curvature of 5mm, etc.).

[0087] After obtaining the set of structural parameters θ, the CAD modeling interface module is called to generate a 3D geometric model. This module is built on the FreeCAD kernel and can automatically draw the light guide rod model according to parametric geometric rules, along with a microstructure front end. The specific process is as follows:

[0088] 1. The script receives the input parameter θ;

[0089] 2. Create a blank 3D space document;

[0090] 3. Call the corresponding modeling template function according to the structure type;

[0091] 4. Use Boolean operations such as revolve, cut, and fuse to generate a complete model;

[0092] 5. Define optical attribute labels on the model surface;

[0093] 6. Export as a STEP or IGES format CAD file;

[0094] Subsequently, the CAD file was transferred to the TracePro platform via API, automatically loaded and embedded into the GIS simulation scene, and used for ray tracing simulation in conjunction with PD point source settings and material property assignments. The entire geometry generation and simulation interface is automated, requiring no manual intervention. The entire path from bit string sample to luminous flux output is driven entirely by scripts, ensuring efficient linkage and consistent accuracy between the structure generation and optimization processes.

[0095] The luminous flux value output by the TracePro simulation model is returned to the QAOA optimization main program as a feedback signal to the objective function C(z). The main program then performs a weighted average of the luminous flux values ​​from multiple samples, which is used as the current parameter. The performance metrics are evaluated, and the classic optimizer Bayesian Optimization is invoked to update the variational parameters for the next iteration.

[0096] The above process achieves closed-loop automatic interaction between the quantum simulator, simulation modeler, and optimization engine, forming a hybrid quantum-classical optimization framework. The functional division of each module in the system architecture is as follows:

[0097] Quantum simulator: generating probability distributions for candidate solutions;

[0098] Geometric encoder: Parses bit strings to generate structural parameters;

[0099] CAD interface: Constructing 3D geometry;

[0100] TracePro platform: Performs ray tracing and outputs performance metrics;

[0101] Optimizer: Updates the variational angle parameters of the QAOA circuit.

[0102] Compared to traditional optimization algorithms (such as genetic algorithms, particle swarm optimization, and simulated annealing), QAOA possesses several significant advantages. First, QAOA is naturally suited for large-scale discrete combinatorial optimization problems, achieving parallel representation and updating of multiple solution states through quantum superposition, thus avoiding getting trapped in local optima. Second, QAOA's optimization process uses quantum states as the carrier for variational approximation of probability distributions, independent of heuristic search paths, making it insensitive to initial values ​​and possessing stronger global search capabilities. Third, theoretically, QAOA exhibits quantum acceleration potential for specific problems, enabling the approximate optimal solutions to complex problems with low-depth circuits, making it suitable for application in high-dimensional structural design spaces.

[0103] By coupling QAOA with a high-precision optical simulation modeling platform, this method can efficiently search for the optimal light guide rod structure configuration while maintaining the authenticity of the physical model, significantly improving the ability to collect partial discharge optical signals and providing strong support for the structural design of high-sensitivity optical detection systems.

[0104] Example 2

[0105] Based on Example 1, this embodiment builds an experimental platform, including partial discharge simulation, optical detection device and data acquisition system, to verify the optical signal detection performance of the optimal light guide rod structure obtained based on QAOA algorithm under actual working conditions. The response capabilities of the optimized structure and various comparative structures under typical partial discharge conditions are systematically evaluated.

[0106] like Figure 7 As shown, the experimental platform is constructed based on a laboratory-simulated GIS tank. The tank is 310mm high, with a shell radius of 90mm, a wall thickness of 10mm, and a conductor radius of 25mm. A sealed container is installed inside the tank to mount a needle-plate electrode. The needle is 25mm long, and the distance between the needle tip and the plate electrode is 6mm. The container is filled with 0.1MPa SF6 gas, and controlled partial discharge triggering is achieved by slowly increasing the voltage. The light guide rod is fixed to the top cover via a threaded interface. The optical signal is coupled through the rod and led out through a quartz transmission fiber, connected to a photon counter for signal detection. To avoid statistical bias in transient pulses, the average count over 60 seconds is used as the effective luminous flux indicator.

[0107] This experiment sets up multiple defect types and structural combinations for comparison. First, in terms of structural comparison, two light guide rod structures were selected for experimental verification: (1) QAOA optimized structure (concave cone, apex angle 60°, length 128mm) and (2) concave spherical structure light guide rod. Figure 8 As shown, the three structures are fixed symmetrically at the central axis of the experimental tank, and the PD source is located on the perpendicular bisector of the line connecting them. The positions of the discharge sources are switched sequentially to cover a typical spatial distribution.

[0108] like Figure 9 As shown, the experimental results are in good agreement with the simulation results, and the magnitude and trend of the normalized luminous flux are basically similar. Compared with the concave sphere, the QAOA optimized structure has a significantly better ability to collect optical signals.

[0109] There are certain differences between the experimental and simulation results. The reasons are as follows: (1) The actual manufactured tank will have burrs at the connection, that is, the connection is uneven and not smooth, and the light guide column is limited by the manufacturing process and has certain errors; (2) The partial discharge intensity is unstable and there are certain errors in the signal acquisition process.

[0110] In summary, the light guide rod structure optimized by QAOA exhibits superior optical response performance in multiple partial discharge defect scenarios. Its sensitivity, stability, and global adaptability are significantly better than traditional designs, providing a new approach and empirical support for high-performance optical partial discharge detection.

[0111] Example 3

[0112] The electronic device of this invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0113] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0114] The processing unit executes the various methods and processes described above, such as methods S1 to S3. For example, in some embodiments, methods S1 to S3 may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of methods S1 to S3 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S3 by any other suitable means (e.g., by means of firmware).

[0115] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0116] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0117] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0118] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for optimizing the design of a light guide rod structure, characterized in that, The method includes the following steps: A TracePro simulation model was constructed to simulate the spatial distribution and multipath optical propagation process of partial discharge point sources based on the actual GIS equipment structure and material parameters. Define the structural parameters of the light guide rod, perform parametric structural modeling of the light guide rod, and construct a multi-parameter joint optimization space; Based on the QAOA quantum optimization algorithm, the optimal combination of structural parameters of the light guide rod is searched with the light flux at the output end of the light guide rod determined by the TracePro simulation model as the objective function. The structure of the light guide rod includes a front-end receiving structure and a planar end-face structure, wherein, The front-end receiving structure is used for the initial reception and incident control of light. Its shape determines the incident angle distribution, beam focusing behavior and the initial conditions of subsequent multiple reflection paths in terms of optical characteristics. The front-end shape design is based on five typical structures, namely planar end, concave cone, concave sphere, convex cone and convex sphere structure. All structures maintain an axially symmetrical design in their overall shape to adapt to the geometric fit conditions of the GIS equipment mounting holes. The planar end face structure serves as a basic reference model, using a smooth plane with the same diameter as the rod body. It is set as an ideal model without incident angle adjustment capability and is used to evaluate the performance improvement of other structures. The structural parameters of the light guide rod include front end shape design parameters, planar end face structure parameters, main body length of the light guide rod, and diameter of the light guide rod. The main body length of the light guide rod is set to multiple discrete values ​​according to a preset length range and step size, and the diameter of the light guide rod is set to a constant value.

2. The method for optimizing the design of a light guide rod structure according to claim 1, characterized in that, The construction of the TracePro simulation model specifically includes the following steps: Based on the ray tracing algorithm, a partial discharge optical signal propagation model inside a GIS device is constructed in the TracePro environment. In the partial discharge optical signal propagation model, discharge sources are set on multiple planes in the circumferential and radial directions of the shell, covering positions near the conductor, near the shell, and the central area. Multiple discharge sources with different radii are set on each plane, and each point source simulates the luminous position generated by a partial discharge defect. The light guide rod adopts an embedded design that is tightly connected to the GIS structure and is installed at the preset orifice position of the simulation model. The light guide rod has a columnar structure inside, and multiple switchable microstructures are configured at the front end. The light guide rod is flush with the inner wall of the GIS to ensure that it achieves maximum optical signal coupling without disturbing the electric field distribution of the device. To evaluate the light signal collection capabilities of different light guide rod structures, independent simulations were performed at each discharge power source location. A receiving surface was set at the exit end of the light guide rod. The total luminous flux value was calculated and standardized by counting all photons guided to the receiving surface through internal reflection, forming a dimensionless evaluation index. Simulations of each light guide rod structure were run independently at all discharge power source locations, and the corresponding average standardized luminous flux was used as the overall performance index of the structure under complex discharge environments.

3. The method for optimizing the design of a light guide rod structure according to claim 1, characterized in that, In the TracePro simulation model, the optical signal generated by partial discharge in GIS is modeled as a point light source with uniform spherical emission. Each simulated point light source emits a preset number of rays, which are uniformly distributed in three-dimensional space according to the spherical coordinate system to ensure that sufficient emission angles and propagation paths are covered in the simulation process.

4. The method for optimizing the design of a light guide rod structure according to claim 1, characterized in that, In the parametric modeling of planar end face structures, for concave and convex structures, key geometric features are defined through parametric methods to achieve continuous and adjustable structure in generalized parameter space; for conical structures, the vertex angle is used as the main variable, and the value range and step size are set to generate several discrete samples; for spherical structures, the radius of curvature is set as the core variable, while ensuring the continuity of the tangent at the transition between the spherical part and the rod.

5. The method for optimizing the design of a light guide rod structure according to claim 1, characterized in that, The parametric structural modeling of the light guide rod specifically involves: All structural parameters of the light guide rods are combined using a Cartesian product to form a high-dimensional parameter space. Each combination of structural parameters corresponds to a unique geometric structure instance. After discretization and encoding, these are converted into binary variables and uniformly represented as lengths of... bit string ,in, n The number of structural parameters, for each group For a specific light guide rod structure, the program automatically generates its corresponding 3D model, which is then input into the TracePro simulation model for ray tracing simulation. At the same time, to prevent the dimensional explosion of the search space, some invalid or low-relevance parameter regions are pruned, and representative subspaces with significant performance differences and feasible processing are retained as the source of the initial population for optimization.

6. The method for optimizing the design of a light guide rod structure according to claim 1, characterized in that, The QAOA quantum optimization algorithm described above performs the following steps to achieve the optimal search for the combination of structural parameters of the light guide rod: Constructing the quantum state of QAOA: in, Is with the objective function The corresponding diagonal Hamiltonian is of the form: , It is a bit string sample. It is projected onto the bit string Operators on the target function and assign them their corresponding objective function values. ; It is the driving Hamiltonian, composed of Pauli-X operators. It is the Pauli-X operator. n It is the number of structural parameters; These are trainable variational parameters; It refers to the algorithm depth, specifically the number of layers in QAOA; Iterative optimization: The quantum computation part of the QAOA quantum optimization algorithm generates candidate solutions in the superposition space, while the classical optimizer adjusts the variational parameters. Perform iterative updates to maximize the expected value of the objective function corresponding to the solution samples obtained from quantum measurement: Each bit string sample was measured using the QAOA quantum optimization algorithm. Then, the structure parameters are decoded into a combination of light guide rod structure parameters through a preset structural parameter dictionary; Based on the combination of decoded light guide rod structural parameters, a three-dimensional geometric model is generated by calling the CAD modeling interface module and input into the TracePro simulation model for ray tracing simulation. The luminous flux value output by the TracePro simulation model is used as the objective function. The feedback signal is returned to the QAOA quantum optimization algorithm; The QAOA quantum optimization algorithm uses a weighted average of the luminous flux values ​​from multiple samples as the current variational parameter. The performance metrics are evaluated, and the classic optimizer is called to update the variational parameters for the next iteration.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 6.

Images