33results about How to "Reduce forecast bias" patented technology

The invention discloses a temperature prediction method and device for power conversion equipment, and belongs to the field of power conversion. The temperature prediction method for the power conversion equipment comprises the following steps: acquiring time sequence electrical parameters of a to-be-detected position in the power conversion equipment at a plurality of acquisition moments; converting the time sequence electrical parameter into a feature image according to a preset time sequence-to-image processing method; and based on the time sequence electrical parameters, the feature image and the environment temperature of the power conversion equipment, performing prediction through a temperature prediction model to obtain the predicted temperature of the to-be-measured position. According to the temperature prediction method of the power conversion equipment, the temperature prediction of the to-be-detected position is carried out based on the feature image obtained through conversion, the time sequence electrical parameters and the environment temperature, the incidence relation of the time sequence electrical parameters in time can be fully considered, the accuracy and authenticity of the predicted temperature obtained through prediction are improved, and the accuracy of the temperature prediction is improved. Therefore, normal operation of equipment such as power conversion equipment is effectively maintained.

The invention belongs to the technical field of specific computer models, and discloses a metal terminal cap pressing and detection integrated control method, which comprises the following steps: S1, acquiring an appearance image of a metal terminal before cap pressing, key size data of a pressing cap mounting area and initial weight data, temperature and humidity environment interference compensation and calibration of the weight data are completed synchronously through a multi-factor regression algorithm; s2, generating an adaptive pressure cap pressure threshold value and pressure holding time through the parameter prediction model; s3, applying pressure to the installation reference position of the metal terminal pressing cap according to the adaptive parameters output in the step S2, and monitoring the dynamic change of the pressure in real time; s4, newly added defect features generated in the cap pressing process are recognized through a finished product detection model after cap pressing, the overall size and the pressing cap attaching precision are verified, the actual weight is compared with a standard weight threshold value, and the screw locking state is judged in a two-way mode by combining torque data and image texture analysis; and S5, integrating the comprehensive detection results of the step S1 and the step S4, and outputting three types of judgment result rates.

The application discloses a kind of plasma air disinfection early warning and intelligent management and control method and system based on distributed sensing, comprising: respectively deploying sensing monitoring module in multiple sub-regions of open public space, each monitoring module is physically separated from disinfection device and interconnected by wireless communication network;Data preprocessing is carried out on the collected multidimensional environmental parameters;According to the number of personnel detected by the millimeter wave radar sensor, the personnel density correction factor is calculated, which is used to quantify the degree of aggravation of air quality risk by personnel activity;Based on the preprocessed multidimensional environmental parameters and personnel density correction factor, the safety factor is calculated, which is a multi-parameter weighted comprehensive index, used to quantify the comprehensive biological safety risk level under current environmental conditions;According to the numerical range of safety factor, output hierarchical control instruction, trigger corresponding level disinfection equipment linkage control strategy.

The invention relates to the technical field of machine vision and intelligent monitoring, and discloses an automatic camera tracking method and device based on machine vision, and the method comprises the steps: obtaining a real-time video frame sequence, and carrying out the preprocessing of the real-time video frame sequence, and obtaining an initial image set; identifying features of the initial image set, judging an abnormal state, and fusing the data to generate behavior change trend data; a smooth trajectory is predicted based on the data, and a parameter adjustment data set is obtained by combining camera position calculation deviation; adjusting camera parameters according to the generation instruction, detecting shielding and fusing trajectory data to obtain real-time tracking parameters; and applying parameters, complementing missing frame verification, updating a process and reextracting features, and performing time sequence analysis in case of abnormality to obtain a final tracking logic sequence. According to the method, accurate target locking can be realized, and the stability and accuracy of abnormal emotion tracking in a crowded environment are improved.

The present application relates to the technical field of artificial intelligence, and more particularly to an intelligent budget management method and system, an electronic device and a storage medium. By fusing historical data and real-time data containing external environment scenarios, business progress and resource status, a prediction model is used to dynamically generate scenario feature weights and budget target prediction values, so that the budget target can adapt to external changes such as policies and competitors, significantly reducing prediction bias. At the same time, the decomposition model calculates the initial decomposition coefficients of each unit according to the resource status data and the scenario feature weights, realizing the linkage calibration of prediction and decomposition, making the budget index allocation match the actual resource carrying capacity of each unit, improving the scientificity of the decomposition scheme and the resource utilization efficiency, and improving the intelligent level of budget management.

The application discloses a kind of virtual power plant resource optimization management method and system based on big data, it is related to clean energy field, the problem that current virtual power plant resource optimization management technology exists management efficiency is not good is solved, including steps S1: photovoltaic monitoring is carried out to target resource area, according to the monitoring result, create target resource optimization period, combine historical photovoltaic monitoring result and carry out historical photovoltaic period matching to target resource optimization period, according to the matching result, obtain photovoltaic period matching data, step S2: photovoltaic monitoring is carried out to target resource area, according to the monitoring result, carry out historical wind period matching to target resource optimization period, according to the matching result, obtain wind energy period matching data, step S3: according to wind energy period matching data and photovoltaic period matching data, the target virtual power plant in target resource optimization period is carried out power storage optimization, the application can effectively improve virtual power plant operation stability and new energy consumption capacity.

The invention relates to a new energy output prediction method and system based on an NHITS model, and the method comprises the steps: obtaining the input characteristics of a to-be-tested new energy object, forming new energy time series data, and enabling the new energy to comprise photovoltaic or wind power; the input characteristics corresponding to photovoltaic output prediction comprise time, solar irradiance, air temperature, air pressure, humidity and historical power data, and the input characteristics corresponding to wind power output prediction comprise time, meteorological variables and wind speeds and wind directions at different heights; new energy output prediction is carried out by adopting an NHITS prediction model, the NHITS prediction model carries out multi-time scale decomposition on a time sequence through a hierarchical recursive structure, the time sequence is divided into a plurality of modules stacked according to layers, and each module carries out partial interpretation on input new energy time sequence data under different time scales and outputs a prediction component; and superposing the prediction components of each layer to obtain a final prediction result. Compared with the prior art, the method can maintain high prediction precision and stability.

The invention discloses a light-storage integrated electric energy storage regulation and control system and method, and particularly relates to the technical field of electric energy storage regulation and control. Based on multi-dimensional characteristic parameters of a photovoltaic side, an energy storage side, a load side and a power grid side, a rolling prediction logic is adopted, a power generation and load trend coefficient, an adjustment coefficient and a mapping rule are combined, a photovoltaic power generation rate and a load demand power prediction result are dynamically optimized, the defects of traditional single-dimensional prediction are overcome, and the prediction efficiency is improved. Coupling influence of assembly temperature, illumination intensity and the like is fully considered, prediction deviation is greatly reduced, and reliable data support is provided for accurate regulation and control.

The application discloses a high-temperature heat accumulator design method and system based on forced convection heat exchange, and relates to the field of heat accumulator optimization design. The method comprises the following steps: a closed cycle experiment system is constructed; basic data of forced convection heat exchange and corresponding thermophysical property parameter data under different experimental conditions are obtained by using the closed cycle experiment system and a NIST database; data calculation and non-dimensional processing are performed on the above data to construct a characteristic parameter data set under each experimental condition; a plurality of candidate correlation formulas are determined by using forced convection heat exchange empirical relations and a least square method based on the characteristic parameter data set under each experimental condition; the Nusselt number prediction accuracy of the plurality of candidate correlation formulas and a plurality of reference correlation formulas is compared and analyzed based on the characteristic parameter data set under each experimental condition, and the best correlation formula is selected according to R 2 , MAPE and maximum relative deviation; and the heat accumulator shape parameters are designed based on the best correlation formula. The application improves the accuracy and reliability of high-temperature heat accumulator design.

The invention relates to the technical field of data analysis and feature engineering, in particular to a feature processing method for industry classification. The method comprises the following steps: acquiring enterprise transaction data and cross-industry cooperation information to determine industry similarity; combining the industry similarity with an industry dependency degree determined according to a pre-obtained supply chain dependency structure to construct an industry relation graph; performing propagation aggregation on adjacent information of each industry node in the industry relation graph, and performing upstream and downstream feature aggregation on a propagation aggregation result to generate enhanced industry features; the enhanced industry features are stored in a preset feature library, and each enhanced industry feature is provided with a unique index; and establishing connection between a preset classification model and the feature library, inputting each enhanced industry feature into the classification model through the unique index, and executing industry classification. According to the method, the industry classification accuracy and the dynamic adaptability can be improved through a high-dimensional enhancement industry feature and dynamic relation graph updating mechanism.

The invention discloses a traffic state prediction method and device for a large-scale road network, equipment and a medium. The traffic state prediction method of the large-scale road network comprises the following steps: taking traffic road network real-time sequence data as input of a traffic state prediction model, and performing sequence decomposition on the traffic road network real-time sequence data to generate space-time representation data; performing spatio-temporal interaction on the spatio-temporal representation data through a traffic state prediction model to generate multi-scale spatio-temporal representation data; performing future feature embedding on the multi-scale spatio-temporal representation data through the traffic state prediction model; and performing future multi-scale prediction sequence fusion on the multi-scale spatio-temporal representation data after future feature embedding through a traffic state prediction model to generate prediction sequence data. Through the above mode, traffic state prediction of a large-scale road network can be realized efficiently and precisely.

The application relates to the technical field of computer-aided design. By providing an AI all-ceramic crown bionic design and mechanical property optimization method based on deep learning, the method comprises the following steps: generating an initial crown geometric model based on crown design space constraints by using a physically embedded generative adversarial network; predicting the mechanical properties of the initial crown geometric model and outputting stress distribution data; simulating the manufacturing deformation process of the initial crown geometric model to obtain manufacturing deformation prediction data; fusing the physical field gradient of the stress distribution data and the manufacturing deformation prediction data to obtain optimization gradient data; updating the physically embedded generative adversarial network by using the optimization gradient data to obtain an optimized crown shape; and converting the optimized crown shape into an executable manufacturing instruction set, so that the technical effects of improving the anatomical shape adaptation accuracy, enhancing the mechanical property prediction accuracy and reducing the design optimization iteration number are achieved, and the clinical applicability of the crown product is improved.

The application discloses a high-resistant starch rice screening and identification and content detection method and system, and relates to the technical field of rice screening. The high-resistant starch rice screening and identification and content detection system comprises a rice screening and identification module and a rice content detection module. The application realizes non-destructive, high-throughput and high-precision comprehensive evaluation of the resistant starch of rice by constructing a multi-modal fusion screening system of starch metabolism fluctuation frequency analysis, three-dimensional seed twin modeling, virtual enzyme molecule diffusion simulation and bionic digestion dynamics school, can complete large-scale seed screening under the premise of not significantly increasing sample loss, and improves the ability to describe the dynamic behavior of starch metabolism by introducing frequency domain features and spatial topology analysis, so that the screening result not only reflects the static content difference, but also can represent the dynamic law of the formation process.

The application discloses a perovskite solar cell stability enhancement system based on multi-modal machine learning, and belongs to the technical field of machine learning. The system comprises a dynamic acquisition module, a multi-modal fusion analysis module, an instruction generation module and a control module. The dynamic acquisition module acquires real-time data and internal structure change characteristics. The multi-modal fusion analysis module performs dynamic characteristic weight distribution on each parameter in the real-time data according to the internal structure change characteristics, and obtains a real-time degradation risk probability value. The instruction generation module generates a comparison result based on the real-time degradation risk probability value, and generates a control instruction according to the comparison result. The control module converts the control instruction into a control signal, and adjusts the parameters of a preset environment cabin and a control battery manufacturing device. The multi-modal fusion analysis module and the instruction generation module are arranged, the prediction deviation is significantly reduced, and the optimization instruction delay bottleneck is broken through.

This invention discloses a real-time prediction method and system for low-altitude atmospheric waveguides for maritime communication, relating to the fields of marine atmospheric environment prediction and maritime communication technology. The method constructs a standardized dataset through multi-source data preprocessing, verifies refractive index through mesoscale optimization calculation, predicts waveguide states and interpolates them using small-scale modeling, and outputs core parameters through multi-scale prediction fusion. Finally, it corrects algorithm coefficients by comparing with measured data, forming a closed-loop prediction system to achieve efficient and accurate prediction of low-altitude atmospheric waveguide parameters at sea. This invention, through multi-source data preprocessing and dual-scale collaborative optimization combined with a closed-loop correction mechanism, outputs accurate atmospheric waveguide parameters in real time, supporting applications such as maritime beyond-line-of-sight communication and radio wave propagation assurance. It adapts to complex meteorological and oceanic conditions in different sea areas, enhances the environmental adaptability of communication systems, and provides reliable environmental data support for the stable operation of maritime communication networks.

The invention discloses a coal mine subsidence area fan site selection method, and belongs to the technical field of coal mine subsidence area wind power new energy. Aiming at the problems that the existing coal mine subsidence area lacks a scientific and applicable fan site selection technical method and the subsidence stability and the coal seam protection coal pillar influence are not comprehensively considered, the method comprises the following steps: collecting and analyzing mining history and geological data; wind resource measurement, topographic surveying and mapping and resource compliance analysis are carried out in the subsidence area where coal mining is carried out for more than three years, and candidate sites are preliminarily screened; geological exploration is carried out on the candidate sites to evaluate stability; predicting the surface residual deformation of the subsidence area, delineating a residual inclination deformation favorable area, and determining the minimum horizontal safety distance between the fan foundation and the underground coal pillar boundary; and finally determining the fan site by integrating the wind resources, the compliance, the stability, the deformation area and the safety distance. The method provides a scientific site selection basis for wind power construction in the coal mine subsidence area, ensures long-term safe and stable operation of a wind power project, and improves economic benefits.

This invention provides a method, application method, and apparatus for constructing a titanium alloy performance prediction model. The construction method includes: constructing a multimodal performance dataset based on titanium alloy samples from different heat treatment batches; constructing auxiliary image data based on microstructure images of titanium alloys acquired under different heat treatment conditions and their corresponding microstructure category labels; training a visual backbone network on the auxiliary image data to perform microstructure category classification tasks, obtaining a visual feature extractor for extracting morphological features of titanium alloys; transferring the visual feature extractor to a pre-constructed initial neural network model, and iteratively training the initial neural network model based on the multimodal performance dataset until the loss function converges or a preset number of iterations is reached, determining that the iterative training is complete, and obtaining a titanium alloy performance prediction model for outputting mechanical property prediction results. This invention can improve the accuracy and stability of titanium alloy performance prediction under small sample conditions.

This application provides a method and system for predicting urban air pollutant concentrations. By performing sensor-level standardization processing on air pollution monitoring data, the raw signals output from different types of smart sensors are uniformly mapped into standardized pollutant concentration feature vectors. Based on these feature vectors, a pollution state characterization sequence integrating temporal evolution characteristics and spatial correlation characteristics is constructed. Based on this sequence, the pollutant concentration variation patterns under different meteorological backgrounds are dynamically adjusted to form predictive input features that match real-time environmental conditions. Based on these predictive input features, pollutant concentrations are predicted, and the predicted future air pollutant concentrations for the target city within a preset timescale are output. Using this approach, collaborative modeling and adaptive prediction of the temporal evolution characteristics and spatial correlation characteristics of pollutant concentrations can be achieved in complex urban environments where multi-type sensor data and variable meteorological conditions are coupled.

The application discloses a kind of embodied large model construction methods for adaptive grasping task, it is related to embodied intelligent technical field.The method includes obtaining training dataset;Embodied large model for adaptive grasping task is constructed, including constructing visual feature extraction module, constructing language feature extraction module, constructing unified semantic representation module, constructing cross-modal transfer module, constructing dynamic modal scheduling module, constructing action output module;Training dataset is used to train the embodied large model, and the embodied large model for adaptive grasping task that training is completed is obtained.The embodied large model constructed in practical application can improve the accuracy of grasping action generation, environmental adaptability and job stability of automated grasping equipment in complex grasping scene, effectively reduce the dependence degree of artificial demonstration and repeated parameter adjustment.

The application provides obtaining a plurality of training data sets, the training data sets comprising one-to-one corresponding microfluidic chip parameter combinations and chip chamber fluid average temperature sequences, the parameter combinations comprising chip thickness, solution volume, ambient temperature, denaturation temperature, annealing temperature and extension temperature, and the chip chamber fluid average temperature sequence being a sequence of spatial average temperature of a chip chamber fluid domain changing with time; obtaining a temperature prediction optimization model, an input of the temperature prediction optimization model being the training data set, and an output being a heater set temperature sequence; and inputting the training data set into the temperature prediction optimization model to obtain the heater set temperature sequence. The method realizes reverse prediction of the heater set temperature curve, and improves prediction accuracy of the model under complex thermal cycle conditions.

The invention relates to an angle module induced effect compensation system and method based on physical-data fusion, and the system comprises a whole vehicle motion planning module which obtains a whole vehicle expected attitude and a rotation angle through calculation; the transfer effect feedforward compensation module is used for constructing a physical embedded feedforward model and estimating transfer disturbance torque in real time; the data driving prediction module is used for constructing a dual-scale space-time fusion network model, extracting long-time-sequence inertial features and short-time-sequence sudden change features of vehicle dynamics, and predicting a vehicle prediction state in a rolling manner; the multi-constraint optimization control module is used for solving the optimal residual compensation torque; the feed-forward parameter online self-adaptive calibration module is used for designing a disturbance observer based on inverse dynamics, extracting real physical disturbance in real time, and correcting feed-forward model parameters online by using a recursive least square algorithm; and an actuator actuation module. Non-linear interference of a turning effect is overcome through a feed-forward-optimization-self-adaptive closed loop mechanism, and high-precision tracking of a vehicle turning angle and coordinated stable control of a whole vehicle yaw attitude are achieved.

The application discloses a sea evaporation duct path loss adaptive prediction method based on small sample transfer learning, and belongs to the technical field of sea wireless communication. In view of the problem that the prediction accuracy of a pure simulation data driven model is reduced due to the difference between a simulation scene and a measured scene in actual sea area application, a transfer learning framework of'source domain pre-training-target domain fine-tuning' is constructed. First, a Transformer model optimized by a tree structure Parzen estimator is pre-trained by using large-scale simulation data, so as to extract general physical features of evaporation duct propagation; second, for sparse single-point measured data, the encoder parameters are frozen and the decoder is fine-tuned, and a point constraint loss function is introduced, so as to calibrate the full path prediction curve by using the measured path loss value of a link specific distance point. The application can reduce the demand for measured data, improve the path loss prediction accuracy and time sequence stability, and is suitable for sea over-the-horizon communication link modeling and performance evaluation.

This invention discloses a method and system for physical modeling and layout optimization of MOSFET saturation current. The methods include: based on effective mobility extraction technology and quasi-ballistic transport theory, impedance is decomposed into ballistic injection impedance and channel scattering impedance, thereby constructing a high-precision MOSFET saturation current physical model. The extracted physical parameters, after modifying the compact model, enable accurate prediction of the saturation current. Furthermore, an initial layout is constructed based on the modeling scheme, and the thermal failure boundary of the MOSFET device saturation current physical model is defined using thermal signature criteria. Then, based on this thermal failure boundary, geometric constraints are applied to the initial layout to generate a multi-finger parallel layout structure that suppresses self-heating effects. Overall, the solution provided by this invention significantly reduces the prediction bias of the compact model and significantly reduces the risk of large-signal linearity degradation in low-temperature drivers and low-noise amplifiers, providing reliable assurance for design.

The invention discloses a multi-source data fusion-based sintered ore FeO content prediction method and system, and belongs to the technical field of sintering process control, and the method comprises the following steps: S1, multi-source data collection and feature extraction; s2, image feature prediction; s3, temperature and process data fusion prediction; and S4, carrying out adaptive weighted fusion. According to the method, the characteristics of the tail section, the temperature distribution of the tail section and the production process data are processed in parallel, the strong time sequence characteristic extraction capability of the TCN-BiLSTM network is utilized, and the two-way prediction result is fused through adaptive weighted average, so that the FeO content of the sintered ore is predicted in real time with high precision.

The application discloses a new energy predicted power correction method and system considering turning weather influence and a storage medium, and the method comprises the following steps: acquiring early warning information, meteorological data and corresponding historical predicted power and actual output of historical turning weather; screening key factors from the early warning information and the meteorological data, and selecting mechanism factors with small coupling influence and clear mechanism from the key factors; constructing a mechanism model based on the mechanism factors, constructing an algorithm model based on the key factors, and constructing a mapping model according to the mechanism model and the algorithm model; and correcting future predicted power to be corrected by using the mapping model. The application combines physical mechanism and data driving, effectively improves the prediction accuracy of new energy under turning weather, solves the prediction deviation problem caused by the complexity of multi-factor coupling and the scarcity of historical samples, and provides a reliable working condition basis for subsequent control strategy formulation.

The invention provides an RF-GBRT-based wind wave and flow combined power generation power prediction method and system. The method comprises the following steps: S1, obtaining a historical data set and real-time collection data; s2, constructing a fusion integrated prediction model; s3, preprocessing the data acquired in real time; s4, inputting the preprocessed real-time features, and generating a power generation power prediction value; and S5, performing closed-loop optimization on the model. Through multi-algorithm deep collaboration and dynamic optimization, the problem of power prediction in a wind wave and flow coupling scene is effectively solved, and technical support is provided for efficient operation of a composite power generation system.

The application relates to the technical field of machine learning, and particularly discloses a spatial local XGBoost machine learning model based on multi-scale geographical weighting, which comprises the following modules: a feature decoupling module for constructing a double-channel input structure of geographical features and non-geographical features, realizing feature decoupling and fusion through a multi-scale spatial weight matrix; a bandwidth allocation module for dynamically determining a bandwidth scale in an XGBoost tree splitting process and establishing dynamic weights of tree levels; a constraint gain module for generating a splitting point marked with a scale; a contribution decoupling module for extracting splitting features and correlating geographical features and non-geographical features, realizing salted prediction and contribution index extraction; and a verification optimization module for optimizing model parameters through multi-scale heat maps and spatial autocorrelation analysis. The model can capture geographical spatial effects of different scales, improve spatial local prediction accuracy, and be applied to a scene with spatial heterogeneity in soil salinization analysis.

The application discloses a fusion-based adaptive pupil interaction method and belongs to the technical field of pupil positioning, and the steps of the method comprise the following steps: HOG feature extraction cooperates with an SVM model to predict key points, lock the iris position and calculate the centroid; based on the screen ratio of a display, a spatial calibration model with the same ratio as the display screen is constructed, interpolation algorithms are used to optimize pupil data, the corrected coordinate data is projected to the display, the actual position of the line-of-sight point in the screen is determined, the parameter vector of the model is iteratively updated through an L-BFGS-B optimization algorithm, the coordinate data is iteratively optimized through an HBG algorithm after mapping, and the fixation center point is determined. The application has the beneficial effects that: in view of the uncertainty of the pupil position, a calibration model is constructed by using device parameters, the data is finely calibrated by using interpolation and optimization algorithms, the error is greatly reduced, the prediction deviation is further reduced through the iterative optimization of the HBG algorithm, and the system has the advantages of simple structure, high lightweight degree, high positioning accuracy and fast response speed.

The application discloses a hydrological model differentiation calibration method based on runoff coefficient grading and coupling rainfall-melting-snow-runoff process, and comprises the following steps: collecting continuous hydrological data of a target basin and preprocessing; dividing out flood events of each flood according to the shape of the flow process line to form a flood sample set; calculating the runoff coefficient of each flood; determining a grading threshold according to the statistical distribution of all sample runoff coefficients, dividing the flood sample set into low and high runoff coefficient subsets; introducing a snow melting module into a traditional hydrological model; independently calibrating the model coupled with the snow melting module by using the two subsets respectively; obtaining the antecedent soil moisture of each flood, and determining the antecedent soil moisture threshold corresponding to the runoff coefficient grading threshold; in the flood forecasting stage, the antecedent soil moisture of the basin is obtained in real time, compared with the threshold, and the corresponding parameter set is dynamically selected and called for forecasting; the application improves the precision and stability of flood simulation and forecasting.

This invention discloses a multi-objective optimization method based on a GCN proxy model, which mainly addresses the problems of high computational complexity, low optimization efficiency, and low prediction accuracy in cascading failure simulation during critical node detection in complex networks. First, initialization and objective function construction are performed, setting relevant algorithm and model parameters, and constructing a dual objective function for attack cost and attack failure effect. A complex network dataset is generated, downloaded, and preprocessed. Second, addressing the low computational and optimization efficiency caused by multiple traversals of the entire network in each evaluation of cascading simulations during critical node detection in complex networks, this invention constructs and trains a multi-branch attention GCN proxy model. Through multi-dimensional feature extraction, multi-task learning, and group calibration, the prediction accuracy of the number of cascading failure nodes is ensured. Then, a multi-objective optimization algorithm fusing GA and PSO is used iteratively, combined with the GCN proxy model to predict the objective function, improving optimization efficiency. The convergence and diversity of solutions are balanced through GA global search and PSO local optimization. Next, the Pareto optimal solution is calibrated and verified to ensure the relative error is within a reasonable range and to verify accuracy. Finally, the experimental results are output and archived. This invention utilizes the GCN proxy model to assist in the fusion of multi-objective optimization algorithms, significantly improving optimization efficiency and prediction accuracy, and enabling precise detection of key nodes in complex networks.