Bus plug-in type modular design method and system for high-voltage centrally installed switchgear

Abstract

The invention relates to the technical field of high-voltage electrical equipment, and discloses a high-voltage centrally installed switchgear bus plug-in modular design method and system, and the method comprises the steps: obtaining the physical parameters and environmental parameters of a bus plug-in module in real time through a sensor integrated at a bus module interface; extracting time domain features and spatial features from the collected data, and dynamically adjusting feature weights by adopting an attention mechanism to obtain target feature data; inputting target characteristic data into an XGBoost-NSGA-II model, optimizing design parameters by adopting an NSGA-II algorithm, and predicting the service life of a bus plug-in module by using an XGBoost part in the optimization process of the NSGA-II algorithm; modular design parameters output by the XGBoost-NSGA-II model are obtained, and the modular design parameters are fed back to a production and operation and maintenance link; according to the invention, intelligentization and high efficiency of the bus plug-in module design of the high-voltage centrally installed switchgear are realized.

Description

Technical Field

[0001] This invention relates to the field of high-voltage electrical equipment technology, specifically to a modular design method and system for a high-voltage intermediate switchgear busbar plug-in type. Background Technology

[0002] High-voltage switchgear is a crucial piece of equipment in power systems, with the busbar as its core component, playing a critical role in current transmission. Traditional busbar design methods often rely on experience and static calculations, making it difficult to adapt to complex and ever-changing operating environments and increasingly demanding performance requirements. In actual operation, the contact state, temperature changes, and vibration at the busbar connection points all affect the reliability and safety of the entire equipment. Simultaneously, environmental factors such as humidity and salt spray corrosion significantly impact the service life of the busbar. Traditional design methods cannot acquire these key parameters in real time and perform dynamic optimization, potentially leading to busbar modules with insufficient reliability, short service life, and high maintenance costs. Therefore, there is an urgent need for a modular design method for high-voltage switchgear busbars that can acquire relevant parameters in real time and perform dynamic optimization design to improve the performance and reliability of busbar modules and reduce production and maintenance costs. Summary of the Invention

[0003] The purpose of this invention is to solve the above-mentioned problems by designing a modular design method and system for high-voltage intermediate switchgear busbar plug-in.

[0004] The first aspect of this invention provides a modular design method for busbar plug-in of a high-voltage intermediate-voltage switchgear, the method comprising the following steps: By integrating sensors at the bus module interface, the physical and environmental parameters of the bus plug-in module can be acquired in real time. Temporal and spatial features are extracted from the collected data, and the feature weights are dynamically adjusted using an attention mechanism to obtain the target feature data. The target feature data is input into the XGBoost-NSGA-II model, and the design parameters are optimized using the NSGA-II algorithm. During the optimization process of the NSGA-II algorithm, the life of the bus plug-in module is partially predicted using XGBoost, and the PSO algorithm is introduced to adjust the hyperparameters of the XGBoost-NSGA-II model in real time. Obtain the modular design parameters output by the XGBoost-NSGA-II model and feed them back to the production and operation and maintenance processes.

[0005] Optionally, in a first implementation of the first aspect of the present invention, the step of extracting temporal and spatial features from the collected data and dynamically adjusting the feature weights using an attention mechanism to obtain target feature data includes: The missing values ​​in the collected data are first filled in using the sliding window method, then the 3σ criterion is used to identify and remove outliers that exceed the normal range, and wavelet transform is used to remove noise from the data to obtain the processed time series data. The fluctuation rate of contact resistance is calculated from the processed time series data to obtain vibration spectrum data; The vibration spectrum data is subjected to a fast Fourier transform to convert the time domain signal into a frequency domain signal. By analyzing the amplitude distribution of the frequency domain signal, the energy distribution of the vibration spectrum is obtained, so as to extract the time domain features that reflect the dynamic operating state of the module. The module installation position is detected, the deviation between the actual installation position and the preset installation position is calculated, the bending data of the busbar is collected by the displacement sensor, and the bending displacement of the busbar is analyzed to extract the spatial characteristics reflecting the spatial position of the module. Based on temporal and spatial features, an attention mechanism is used to dynamically adjust feature weights to obtain target feature data.

[0006] Optionally, in a second implementation of the first aspect of the present invention, the step of dynamically adjusting feature weights based on temporal and spatial features using an attention mechanism to obtain target feature data includes: Set an initial feature weight matrix, where the elements in the feature weight matrix represent the initial importance weights of each temporal and spatial feature; Based on different operating environments and conditions, evaluate the importance of various time-domain and spatial characteristics to module performance; Based on the importance assessment results of the features, an attention mechanism is used to dynamically adjust the feature weights. The extracted temporal and spatial features are then weighted and combined with the adjusted feature weights to obtain the target feature data.

[0007] Optionally, in a third implementation of the first aspect of the present invention, the step of using NSGA-II to optimize design parameters includes: A preset number of design parameter combinations are randomly generated as the initial population, wherein the design parameters include at least busbar thickness, silver plating uniformity, and module gap; Each individual in the population is evaluated, and based on electrical performance and structural strength constraints, the individuals are ranked according to their non-dominant relationships, thus dividing the population into different non-dominant layers. For each individual in the non-dominated layer, the crowding degree is calculated. Based on the non-dominated ranking results and the crowding degree, a tournament selection method is used to select excellent individuals from the current population as parents to enter the next generation for reproduction. Crossover and mutation operations are performed on the selected parent individuals to generate offspring individuals. The parent and offspring populations are merged, and non-dominated sorting and crowding calculations are performed again to select individuals to form a new population. Repeat the non-dominated sorting, crowding calculation, selection, crossover, and mutation operations until the preset number of iterations is reached to obtain the final design parameters.

[0008] Optionally, in the fourth implementation of the first aspect of the present invention, the electrical performance constraints are contact resistance less than or equal to 30μΩ and short-circuit withstand capability greater than or equal to 100kA / 1s; the structural strength constraints are module coaxiality deviation less than 0.5mm.

[0009] Optionally, in a fifth implementation of the first aspect of the present invention, the step of using XGBoost to partially predict the lifespan of the bus plug-in module includes: The target feature data is divided into training and validation sets. Based on decision trees, multiple decision trees are constructed using ensemble learning. When constructing each tree, a greedy algorithm is used to select the best split point to minimize the prediction error of the tree. During model training, a regularization term is introduced, and new decision trees are added iteratively. Each new tree is trained based on the prediction residuals of all previous trees. The trained XGBoost model is then used to predict the lifetime of the bus plug-in module.

[0010] Optionally, in the sixth implementation of the first aspect of the present invention, the introduction of the PSO algorithm to adjust the hyperparameters of the XGBoost-NSGA-II model in real time includes: Set a preset number of particles, each particle represents a set of model hyperparameters, and randomly assign an initial position and velocity to each particle; Calculate the fitness value of the hyperparameter combination represented by each particle in the model. Each particle updates its individual optimal position based on its own fitness value, while the swarm updates the global optimal position based on the fitness values ​​of all particles. Each particle adjusts its velocity and position based on its individual optimal position and global optimal position, and repeats the process of fitness evaluation, updating individual and global optimal positions, and adjusting velocity and position until a preset number of iterations is reached to obtain the optimal combination of hyperparameters, which is then applied to the XGBoost-NSGA-II model.

[0011] A second aspect of the present invention provides a modular design system for busbar plug-in of a high-voltage switchgear, the system comprising: The acquisition module is used to acquire the physical and environmental parameters of the bus plug-in module in real time through sensors integrated at the bus module interface; The extraction module is used to extract temporal and spatial features from the collected data, and dynamically adjusts the feature weights using an attention mechanism to obtain the target feature data. The optimization module is used to input target feature data into the XGBoost-NSGA-II model and optimize the design parameters using the NSGA-II algorithm. During the NSGA-II algorithm optimization process, XGBoost is used to partially predict the lifespan of the bus plug-in module, and the PSO algorithm is introduced to adjust the hyperparameters of the XGBoost-NSGA-II model in real time. The feedback module is used to obtain the modular design parameters output by the XGBoost-NSGA-II model and feed them back to the production and operation and maintenance processes.

[0012] A third aspect of the present invention provides a high-voltage intermediate-voltage switchgear busbar plug-in modular design device, the high-voltage intermediate-voltage switchgear busbar plug-in modular design device comprising a memory and at least one processor, the memory storing instructions; the at least one processor calling the instructions in the memory to cause the high-voltage intermediate-voltage switchgear busbar plug-in modular design device to perform the various steps of the high-voltage intermediate-voltage switchgear busbar plug-in modular design method as described in any of the preceding claims.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the high-voltage intermediate-voltage switchgear busbar plug-in modular design method as described in any of the preceding claims.

[0014] The technical solution provided by this invention acquires physical and environmental parameters of the bus plug-in module in real time by integrating sensors at the bus module interface; extracts temporal and spatial features from the collected data, dynamically adjusts feature weights using an attention mechanism to obtain target feature data; inputs the target feature data into the XGBoost-NSGA-II model, optimizes design parameters using the NSGA-II algorithm, and uses XGBoost to partially predict the lifespan of the bus plug-in module during the NSGA-II algorithm optimization process; obtains the modular design parameters output by the XGBoost-NSGA-II model, and feeds the modular design parameters back to the production and operation and maintenance stages; this invention, by integrating multiple sensors at the bus module interface to acquire physical and environmental parameters in real time, can comprehensively and timely grasp the operating status of the bus plug-in module; extracting temporal and spatial features and dynamically adjusting feature weights using an attention mechanism can highlight the impact of key features on the design under different operating environments and conditions, making the obtained target feature data more targeted and effective; inputting the target feature data into XGBoost-NSGA-II... The model enables intelligent, precise, and efficient design of the busbar plug-in module for high-voltage switchgear, improving the performance and reliability of the busbar module, reducing production and maintenance costs, and enhancing product quality and market competitiveness. Attached Figure Description

[0015] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention.

[0016] Figure 1 A flowchart illustrating the modular design method for high-voltage intermediate switchgear busbar connection provided in this embodiment of the invention; Figure 2 This is a schematic diagram of the high-voltage intermediate switchgear busbar plug-in modular design system provided in an embodiment of the present invention; Figure 3 This is a structural schematic diagram of a high-voltage intermediate switchgear busbar plug-in modular design device provided in an embodiment of the present invention. Detailed Implementation

[0017] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0018] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 The flowchart of the high-voltage intermediate switchgear busbar plug-in modular design method provided in this embodiment of the invention includes the following steps: Step 101: Real-time acquisition of physical and environmental parameters of the bus plug-in module using sensors integrated at the bus module interface; In this embodiment, by precisely integrating multiple high-performance sensors such as MEMS accelerometers, infrared temperature sensors, and contact resistance detectors at the bus module interface, comprehensive real-time monitoring of the bus plug-in module's operating status can be achieved, thereby obtaining rich and crucial physical and environmental parameters. The physical parameters include contact pressure, contact resistance, vibration spectrum, and temperature; the environmental parameters include humidity changes under IP66 protection level and salt spray corrosion intensity data obtained from 1000 hours of testing. This data is transmitted in real-time and efficiently to a cloud database via a fiber optic sensor network and wireless communication protocols such as LoRa.

[0019] Step 102: Extract temporal and spatial features from the collected data, and dynamically adjust the feature weights using an attention mechanism to obtain the target feature data; In this embodiment, when preprocessing the collected data, the sliding window method is first used to fill in missing values. That is, an appropriate window size is set, and the effective data within the window is used to calculate the estimated value of the missing position through the mean, median, or linear interpolation to ensure the continuity of the data sequence. Next, the 3σ criterion is used to identify outliers. The mean and standard deviation of the data sequence are calculated first, and data that exceed the mean ± 3 times the standard deviation are identified as outliers and removed to avoid outliers interfering with subsequent analysis. Finally, wavelet transform is used to remove noise. An appropriate wavelet basis function and decomposition level are selected to decompose the data at multiple scales, separate the wavelet coefficients corresponding to the noise and set them to zero, and then the denoised time series data is obtained through wavelet reconstruction, so that the data can more accurately reflect the operating status of the module. When extracting relevant features from the processed time series data, a certain time interval needs to be selected to calculate the fluctuation rate of the contact resistance. The fluctuation level within this time period is represented by the ratio of the difference between two adjacent contact resistances to the contact resistance of the previous time, thereby reflecting the stability of the contact state. At the same time, the vibration signal during the operation of the module is continuously collected by the sensor. After the above preprocessing, vibration spectrum data is generated, which provides a basis for subsequent frequency domain analysis. When performing a Fast Fourier Transform on the vibration spectrum data, the continuous time-domain vibration signal is decomposed into sinusoidal components of different frequencies and converted into a frequency domain signal with frequency as the abscissa and amplitude as the ordinate. By analyzing the amplitude and distribution of different frequency components in the frequency domain signal, the energy proportion in each frequency interval is calculated to obtain the energy distribution of the vibration spectrum. Thus, time-domain features that can reflect the vibration intensity and energy concentration of the module at different frequencies are extracted. These features can effectively reflect the dynamic operating status of the module. When extracting spatial features, high-precision visual inspection equipment or position sensors are used to detect the module installation position, obtain the actual installation coordinates, and compare them with the preset standard installation coordinates to calculate the installation deviation of the module in three-dimensional space, ensuring that the deviation is controlled within ±2mm. At the same time, displacement sensors are placed at key parts of the busbar to collect bending deformation data of the busbar in real time when subjected to stress or temperature changes. Through data processing and analysis, the bending displacement of the busbar is obtained, ensuring that the displacement is within ±15°, thereby reflecting the spatial position status and structural stability of the module. When dynamically adjusting feature weights based on extracted temporal and spatial features using an attention mechanism, a feature importance assessment model is first constructed. This model is then combined with the module's operating environment, such as temperature, humidity, and salt spray corrosion intensity, and working conditions, such as vibration intensity, to analyze the impact of different features on module performance, such as contact reliability, structural strength, and service life. Based on the assessment results, features with high importance are assigned higher weights, such as increasing the weight of temperature-related temporal features under high-temperature environments to 0.4, while features with low importance are de-weighted. Finally, each feature is multiplied by its corresponding adjusted weight and the results are summed to obtain target feature data that incorporates key information.

[0020] In this embodiment, when setting the initial feature weight matrix, it is necessary to comprehensively consider the operating characteristics of the high-voltage switchgear busbar plug-in module and the general influence of various features. Initial weight values ​​are assigned to each time-domain feature such as contact resistance fluctuation rate, vibration spectrum energy distribution, and spatial features such as module installation deviation and busbar bending displacement. These initial weight values ​​can be determined by referring to the design experience of similar equipment or by preliminary determination through small-scale test data, forming a matrix containing the initial importance weights of all features. Each element in the matrix corresponds one-to-one with the corresponding feature. When assessing the importance of various temporal and spatial characteristics to module performance under different operating environments and conditions, a comprehensive analysis of environmental and operating condition factors is required, including temperature variations from -15℃ to 40℃, humidity variations under IP66 protection level, 1000-hour salt spray corrosion intensity test data, and vibration intensity ±20g. For example, in high-temperature environments, the impact of temperature-related temporal characteristics on module contact and insulation performance is significantly enhanced. In conditions with severe vibration, the importance of the temporal characteristic of vibration spectrum energy distribution and the spatial characteristic of busbar bending displacement increases. By analyzing the correlation between these characteristics and performance indicators such as module operational stability and lifespan, their importance level under the current environment and operating conditions can be determined. When dynamically adjusting feature weights using an attention mechanism based on the importance assessment results, the elements in the initial feature weight matrix are modified accordingly based on the assessed importance level. For features with increased importance, such as temperature-related time-domain features under high-temperature environments, their corresponding weights are increased, for example, to 0.4. For features with decreased importance, their weights are decreased. Then, each extracted time-domain feature and spatial feature is multiplied by its adjusted corresponding weight, and all product results are summarized to form the target feature data.

[0021] Step 103: Input the target feature data into the XGBoost-NSGA-II model, and use the NSGA-II algorithm to optimize the design parameters. During the NSGA-II algorithm optimization process, XGBoost is used to partially predict the life of the bus plug-in module, and the PSO algorithm is introduced to adjust the hyperparameters of the XGBoost-NSGA-II model in real time. In this embodiment, when randomly generating a preset number of design parameter combinations as the initial population, it is necessary to clarify the reasonable value range of each parameter. Among them, the bus thickness is set within a certain range according to the current carrying capacity requirement, the uniformity of the silver plating layer is measured by the thickness deviation percentage, such as being controlled within ±5%, and the module gap needs to be set within a suitable range, such as 0.5mm~2mm, taking into account insulation performance and assembly process. By randomly sampling, a specified number of combinations are generated within each parameter range to ensure that the initial population has a certain diversity, covering different parameter combinations, and providing a broad starting point for subsequent optimization search. When evaluating each individual in the population, each design parameter combination is checked to ensure it meets the constraints based on electrical performance requirements (contact resistance ≤30μΩ, short-circuit withstand capability ≥100kA / 1s) and structural strength requirements (module coaxiality deviation <0.5mm, seismic performance meeting IEC61439-2 standard 5g vibration test). On this basis, individuals are ranked according to non-dominance relationships: if an individual is not inferior to another individual in all objectives and is superior in at least one objective, then the former dominates the latter. Individuals not dominated by other individuals are grouped into the same non-dominance layer, such as the first layer. This process is repeated for the remaining individuals to divide them into the second layer, third layer, etc., thus distinguishing the superiority or inferiority of different individuals. When calculating crowding for individuals in each non-dominated layer, the minimum and maximum values ​​of each objective function are first determined, and the values ​​of individuals on each objective are standardized. Then, the sum of the distances between an individual and its neighboring individuals on each objective is calculated. This sum is the crowding, which measures the density of individuals in the population. Based on the non-dominated ranking results, individuals at higher levels are selected first. Within the same layer, individuals with higher crowding are selected based on their crowding to maintain population diversity. When using the tournament selection method, a certain number of individuals are randomly selected from the population, their non-dominated layers and crowding are compared, and the optimal individual is selected as the parent. This process is repeated until a sufficient number of parent individuals are selected for reproduction. When performing crossover operations on the selected parent individuals, real-number crossover methods are used for parameters such as busbar thickness, silver plating uniformity, and module gaps. For example, simulated binary crossover is used to randomly generate crossover points and crossover coefficients, and the corresponding parameters of the two parent individuals are proportionally merged to generate offspring parameters. Mutation operations are achieved by adding random perturbations to the parameters of the parent individuals, such as polynomial mutation, which randomly changes some parameter values ​​within the parameter range to introduce new genetic information. After merging the parent and offspring populations, non-dominated sorting and crowding calculation are performed again. Individuals are selected in descending order of hierarchy, and within each hierarchy, individuals are selected in descending order of crowding until the preset population size is reached to form a new population. The process involves repeatedly performing non-dominated sorting, crowding calculation, selection, crossover, and mutation operations, optimizing and updating the population with each iteration. During the iteration process, the population gradually evolves towards a better combination of design parameters, the non-dominated level of individuals continuously improves, and the crowding distribution becomes more reasonable. When the number of iterations reaches a preset value, such as 100 or 200, the iteration process is terminated. At this point, the individuals in the population that are at the highest non-dominated level and have a high crowding level are the final design parameters obtained through optimization. These parameters can achieve comprehensive optimization of module performance while satisfying various constraints.

[0022] In this embodiment, when dividing the target feature data into training and validation sets, an appropriate division ratio needs to be selected based on the data volume and distribution, typically using a 7:3 or 8:2 ratio. During the division process, it is necessary to ensure that the distribution of the two sets of data is consistent to avoid the impact of data distribution differences on model training and validation. The training set is used for model parameter learning and construction, while the validation set is used to evaluate the model's performance during model training, promptly identify overfitting and other issues, and provide a basis for adjusting model parameters. When constructing multiple decision trees using ensemble learning, each decision tree is trained on different subsets of the training set. These subsets are randomly selected from the original training set using the bootstrap sampling method, and repeated sample selection is allowed. During the construction of each decision tree, for each node, a greedy algorithm is used to traverse all possible features and all possible splitting thresholds for that feature. The impurity after splitting, such as the Gini coefficient, information entropy, or squared error, is calculated. The feature that reduces the impurity the most or minimizes the squared error and the splitting threshold are selected as the optimal splitting point. In this way, decision trees are recursively constructed to ensure that the prediction error of each tree is minimized. Regularization terms are introduced during model training to control the complexity of the decision tree. Regularization terms typically include penalties for tree depth, number of leaf nodes, and leaf node weights to prevent the model from overfitting the training data. Then, new decision trees are added iteratively. Each time a new tree is added, the residual between the prediction results of all previous trees and the actual values ​​is used as the new target variable for training. That is, the new decision tree is dedicated to fitting the prediction error of the preceding model. By continuously correcting the error, the prediction accuracy of the entire ensemble model is gradually improved. When using a trained XGBoost model to predict the lifespan of bus plug-in modules, the relevant target feature data of the module to be predicted are input into the model. The model will output the corresponding lifespan prediction value based on the previously learned mapping relationship between features and lifespan. During the prediction process, the prediction accuracy can be evaluated and verified by combining the model's output results with the actual operating conditions, ensuring that the prediction results can provide a reliable reference for module design optimization, maintenance, and replacement.

[0023] In this embodiment, a preset number of particles are set, each particle represents a set of model hyperparameters, and each particle is randomly assigned an initial position and velocity; the fitness value of the hyperparameter combination represented by each particle in the model is calculated, each particle updates its individual optimal position according to its own fitness value, and the swarm updates the global optimal position according to the fitness values ​​of all particles; each particle adjusts its own velocity and position according to its individual optimal position and global optimal position, and the process of fitness evaluation, updating individual optimal and global optimal, and adjusting velocity and position is repeated until a preset number of iterations is reached to obtain the optimal hyperparameter combination, which is then applied to the XGBoost-NSGA-II model.

[0024] Step 104: Obtain the modular design parameters output by the XGBoost-NSGA-II model and feed them back to the production and operation and maintenance process.

[0025] In this embodiment, modular design parameters output by the XGBoost-NSGA-II model are acquired. These parameters cover a wide range of specific details, including module dimensions such as a 550mm narrow cabinet design, busbar layout such as a double-layer stacked design to increase current carrying capacity by 30%, interface specifications such as IP66 sealing design, and other parametric drawing information, as well as process parameters for laser welding weld strength reaching 95% of the base material and robot assembly. These modular design parameters are promptly fed back to the production stage, allowing the production department to accurately carry out manufacturing work based on parametric drawings and process parameters, ensuring that the produced busbar plug-in modules meet design standards and improving production consistency and efficiency. Simultaneously, the design parameters are fed back to the operation and maintenance stage, providing strong guidance for maintenance work. For example, a module replacement early warning mechanism is set based on a contact resistance threshold >30μΩ. Combined with blockchain traceability technology, test data such as 5000 mechanical life tests of the modules are uploaded to the blockchain, enabling quality tracking throughout the module's lifecycle and facilitating timely detection and handling of problems during operation and maintenance.

[0026] Please see Figure 2 A schematic diagram of the high-voltage intermediate switchgear busbar plug-in modular design system provided in this embodiment of the invention. The system includes: The acquisition module is used to acquire the physical and environmental parameters of the bus plug-in module in real time through sensors integrated at the bus module interface; The extraction module is used to extract temporal and spatial features from the collected data, and dynamically adjusts the feature weights using an attention mechanism to obtain the target feature data. The optimization module is used to input target feature data into the XGBoost-NSGA-II model and optimize the design parameters using the NSGA-II algorithm. During the NSGA-II algorithm optimization process, XGBoost is used to partially predict the lifespan of the bus plug-in module, and the PSO algorithm is introduced to adjust the hyperparameters of the XGBoost-NSGA-II model in real time. The feedback module is used to obtain the modular design parameters output by the XGBoost-NSGA-II model and feed them back to the production and operation and maintenance processes.

[0027] Figure 3This is a schematic diagram of a high-voltage intermediate-voltage switchgear busbar plug-in modular design device 300 provided in an embodiment of the present invention. The high-voltage intermediate-voltage switchgear busbar plug-in modular design device 300 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 310 (e.g., one or more processors) and a memory 320, and one or more storage media 330 (e.g., one or more mass storage devices) storing application programs 333 or data 332. The memory 320 and storage media 330 can be temporary or persistent storage. The program stored in the storage media 330 may include one or more modules (not shown in the diagram), each module may include a series of instruction operations on the high-voltage intermediate-voltage switchgear busbar plug-in modular design device 300. Furthermore, the processor 310 may be configured to communicate with the storage media 330 and execute a series of instruction operations in the storage media 330 on the high-voltage intermediate-voltage switchgear busbar plug-in modular design device 300 to implement the method provided in the above embodiment.

[0028] The high-voltage intermediate-voltage switchgear busbar plug-in modular design device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The high-voltage intermediate switchgear busbar plug-in modular design equipment structure shown does not constitute a limitation on the computer equipment provided by the present invention. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0029] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the various steps of the high-voltage intermediate-voltage switchgear busbar plug-in modular design method provided in the above embodiments.

[0030] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described equipment or apparatus / unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0031] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0032] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A modular design method for busbar plug-in of high-voltage intermediate-voltage switchgear, characterized in that, The method includes the following steps: By integrating sensors at the bus module interface, the physical and environmental parameters of the bus plug-in module can be acquired in real time. Temporal and spatial features are extracted from the collected data, and the feature weights are dynamically adjusted using an attention mechanism to obtain the target feature data. The target feature data is input into the XGBoost-NSGA-II model, and the design parameters are optimized using the NSGA-II algorithm. During the optimization process of the NSGA-II algorithm, the life of the bus plug-in module is partially predicted using XGBoost, and the PSO algorithm is introduced to adjust the hyperparameters of the XGBoost-NSGA-II model in real time. Obtain the modular design parameters output by the XGBoost-NSGA-II model and feed them back to the production and operation and maintenance processes.

2. The high-voltage intermediate-voltage switchgear busbar plug-in modular design method as described in claim 1, characterized in that, The process involves extracting temporal and spatial features from the collected data, dynamically adjusting feature weights using an attention mechanism, and obtaining target feature data, including: The missing values ​​in the collected data are first filled in using the sliding window method, then the 3σ criterion is used to identify and remove outliers that exceed the normal range, and wavelet transform is used to remove noise from the data to obtain the processed time series data. For the processed time series data, the fluctuation rate of contact resistance is calculated to obtain vibration spectrum data; The vibration spectrum data is subjected to a fast Fourier transform to convert the time domain signal into a frequency domain signal. By analyzing the amplitude distribution of the frequency domain signal, the energy distribution of the vibration spectrum is obtained, so as to extract the time domain features that reflect the dynamic operating state of the module. The module installation position is detected, the deviation between the actual installation position and the preset installation position is calculated, the bending data of the busbar is collected by the displacement sensor, and the bending displacement of the busbar is analyzed to extract the spatial characteristics reflecting the spatial position of the module. Based on temporal and spatial features, an attention mechanism is used to dynamically adjust feature weights to obtain target feature data.

3. The high-voltage intermediate-voltage switchgear busbar plug-in modular design method as described in claim 2, characterized in that, The method of dynamically adjusting feature weights based on temporal and spatial features using an attention mechanism to obtain target feature data includes: Set an initial feature weight matrix, where the elements in the feature weight matrix represent the initial importance weights of each temporal and spatial feature; Based on different operating environments and conditions, evaluate the importance of various temporal and spatial characteristics to module performance; Based on the importance assessment results of the features, an attention mechanism is used to dynamically adjust the feature weights. The extracted temporal and spatial features are then weighted and combined with the adjusted feature weights to obtain the target feature data.

4. The high-voltage intermediate-voltage switchgear busbar plug-in modular design method as described in claim 1, characterized in that, The optimization of design parameters using NSGA-II includes: A preset number of design parameter combinations are randomly generated as the initial population, wherein the design parameters include at least busbar thickness, silver plating uniformity, and module gap; Each individual in the population is evaluated, and based on electrical performance and structural strength constraints, the individuals are sorted according to their non-dominant relationships to divide them into different non-dominant layers. For each individual in the non-dominated layer, the crowding degree is calculated. Based on the non-dominated ranking results and the crowding degree, a tournament selection method is used to select excellent individuals from the current population as parents to enter the next generation for reproduction. Crossover and mutation operations are performed on the selected parent individuals to generate offspring individuals. The parent and offspring populations are merged, and non-dominated sorting and crowding calculations are performed again to select individuals to form a new population. Repeat the non-dominated sorting, crowding calculation, selection, crossover, and mutation operations until the preset number of iterations is reached to obtain the final design parameters.

5. The high-voltage intermediate-voltage switchgear busbar plug-in modular design method as described in claim 4, characterized in that, The electrical performance constraints are: contact resistance less than or equal to 30μΩ, and short-circuit withstand capability greater than or equal to 100kA / 1s; The structural strength constraint is that the coaxiality deviation of the modules is less than 0.5 mm.

6. The high-voltage intermediate-voltage switchgear busbar plug-in modular design method as described in claim 1, characterized in that, The method of using XGBoost to partially predict the lifespan of bus plug-in modules includes: The target feature data is divided into training and validation sets. Based on decision trees, multiple decision trees are constructed using ensemble learning. When constructing each tree, a greedy algorithm is used to select the best split point to minimize the prediction error of the tree. During model training, a regularization term is introduced, and new decision trees are added iteratively. Each new tree is trained based on the prediction residuals of all previous trees. The trained XGBoost model is then used to predict the lifetime of the bus plug-in module.

7. The high-voltage intermediate-voltage switchgear busbar plug-in modular design method as described in claim 1, characterized in that, The introduction of the PSO algorithm to adjust the hyperparameters of the XGBoost-NSGA-II model in real time includes: Set a preset number of particles, each particle represents a set of model hyperparameters, and randomly assign an initial position and velocity to each particle; Calculate the fitness value of the hyperparameter combination represented by each particle in the model. Each particle updates its individual optimal position based on its own fitness value, while the swarm updates the global optimal position based on the fitness values ​​of all particles. Each particle adjusts its velocity and position based on its individual optimal position and global optimal position, and repeats the process of fitness evaluation, updating individual and global optimal positions, and adjusting velocity and position until a preset number of iterations is reached to obtain the optimal combination of hyperparameters, which is then applied to the XGBoost-NSGA-II model.

8. A high-voltage intermediate-voltage switchgear busbar plug-in modular design system, characterized in that: The system includes: The acquisition module is used to acquire the physical and environmental parameters of the bus plug-in module in real time through sensors integrated at the bus module interface; The extraction module is used to extract temporal and spatial features from the collected data, and dynamically adjusts the feature weights using an attention mechanism to obtain the target feature data. The optimization module is used to input target feature data into the XGBoost-NSGA-II model and optimize the design parameters using the NSGA-II algorithm. During the NSGA-II algorithm optimization process, XGBoost is used to partially predict the lifespan of the bus plug-in module, and the PSO algorithm is introduced to adjust the hyperparameters of the XGBoost-NSGA-II model in real time. The feedback module is used to obtain the modular design parameters output by the XGBoost-NSGA-II model and feed them back to the production and operation and maintenance processes.

9. A high-voltage intermediate-voltage switchgear with a modular design featuring busbar plug-in interface, characterized in that: The high-voltage intermediate switchgear busbar plug-in modular design device includes a memory and at least one processor. The memory stores instructions. The at least one processor calls the instructions in the memory to cause the high-voltage intermediate switchgear busbar plug-in modular design device to perform each step of the high-voltage intermediate switchgear busbar plug-in modular design method as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement each step of the high-voltage intermediate switchgear busbar plug-in modular design method as described in any one of claims 1-7.

Images