45 results about "Model order reduction" patented technology

The invention relates to the field of digital twinborn technology and industrial equipment intelligent operation and maintenance, and discloses an electric power screen cabinet heat dissipation efficiency improvement and thermal management comprehensive analysis system, which comprises an offline modeling module for constructing a thermal order reduction model with efficient calculation through a model order reduction method based on high-fidelity CFD simulation data; the real-time data acquisition module is used for acquiring real-time power consumption, sparse sensor temperature and an expected load curve; the online state estimation module is used for fusing physical prediction of the thermal reduced-order model with real-time sparse measurement data and reconstructing a global three-dimensional temperature field of the power screen cabinet in real time; and the prospective scheduling analysis module is used for taking the current global temperature field as an initial condition, driving the thermal reduced-order model to perform prospective deduction by utilizing an expected load curve, and calculating and outputting a future thermal margin to early warn a thermal risk. According to the invention, comprehensive real-time sensing, prospective risk prediction and self-adaptive correction of the thermal state of the electric screen cabinet are realized, and thermal management is converted from passive response to active prevention.

The invention discloses an aerospace equipment data-real fusion performance test method and device based on reinforcement learning, and the device comprises a digital twin model building module which is used for building an aerospace equipment high-fidelity mechanism model fusing multi-physical field features, a lightweight digital twinborn body adaptive to reinforcement learning training is generated through a model order reduction technology; the digital twinborn model correction module is used for carrying out consistency calibration of multi-dimensional features and key physical parameter correction on the reduced-order model based on object reference data; the performance test module based on reinforcement learning actively explores a coupling scene of an extreme working condition and multiple faults in an interaction environment by constructing an attacker agent, and autonomously explores a performance boundary and a limit envelope of equipment; and the number-real fusion verification module is used for carrying out a hardware-in-the-loop point compensation test by taking the performance limit obtained by the digital test as guidance so as to assist in decision making and optimization of a test scheme. According to the method, the coverage, the safety and the test efficiency of the limit performance test of the aerospace equipment are improved.

The invention discloses an integrated circuit electrical characteristic simulation method and device, a terminal and a storage medium, and the method comprises the steps: obtaining a to-be-simulated integrated circuit model, carrying out the mesh generation of the to-be-simulated integrated circuit model, so as to obtain an integrated circuit mesh data set composed of a plurality of mesh units, constructing a corresponding charge basis function and a current basis function based on the grid units; step-by-step fusion updating of the charge primary function is realized by taking the charge primary function as a fusion object, and meanwhile, corresponding fusion updating is performed on the current primary function in a geometric mapping mode; and after all fusion updating is completed, obtaining an integrated circuit primary function data set, and performing electrical characteristic simulation on the to-be-simulated integrated circuit based on the integrated circuit primary function data set. According to the method, fast updating and low-order representation of a complex coupling capacitance matrix are achieved, compared with a traditional post-order reduction method, the theoretical complexity is reduced from O (n3) to O (n), and meanwhile the model order reduction efficiency is greatly improved while the model precision and the order reduction effect are guaranteed.

The invention discloses a three-dimensional model order reduction method for a special valve and electronic equipment, and belongs to the field of special valves. The method comprises the following steps: firstly, performing three-dimensional modeling according to a design file of the special valve to obtain an internal flow channel model; calculating a flow field in the special valve flow channel under different working conditions by utilizing computational fluid mechanics software; performing modal decomposition on a sample data set of the original three-dimensional model obtained by flow field calculation to obtain a spatial mode, a modal coefficient and modal energy; selecting the modal quantity according to the modal energy; fitting the modal coefficient, and establishing a relationship between the working condition parameter and the modal coefficient; and quickly predicting the internal flow field of the special valve under the unknown working condition according to the fitting model obtained by fitting and the spatial mode. According to the method, the calculation amount is effectively reduced while the calculation precision is ensured, the calculation efficiency is greatly improved compared with a numerical simulation method, and the requirement of a user for obtaining the internal flow field information of the special valve with high efficiency and high real-time performance can be met.

The invention discloses a rapid prediction method for wing flow field area decomposition. According to the method, a k-means clustering decomposition method is adopted, effective decomposition of a flow field area is achieved, area sampling is carried out based on area decomposition, main modal energy extraction is carried out on a sampled flow field data set through a model order reduction method, a basis matrix of flow field data is obtained, the basis matrix is trained through a machine learning method, and the flow field data is obtained. And finally, a wing flow field region decomposition rapid prediction framework is realized. According to the method, the calculation efficiency of a traditional numerical simulation method is remarkably improved, and the method has wide applicability.

The application discloses a semi-airborne full-time electromagnetic forward method based on an Octree grid, which performs spatial discretization on time-domain Maxwell equations, thereby effectively reducing unknowns of equations to be solved. For full-time electromagnetic field numerical simulation, the application combines the high calculation precision of the second-order backward Euler method with the high OFF-Time calculation efficiency of the model reduction algorithm, adopts the second-order BDF and the model reduction algorithm to respectively solve ON-Time and OFF-Time time-domain control equations, forms a joint algorithm to realize simulation of semi-airborne time-domain electromagnetic full-time response, and finally completes the forward calculation process. Through numerical simulation verification, in the full-time transient electromagnetic forward process, the application can keep small occupied memory and also has higher calculation efficiency, and the calculation precision is not lower than that of other methods.

The application provides a method and system for offshore wind turbine integrated dynamic feature reservation. The full state time series data of the offshore wind turbine multi-physical field coupling system is obtained, the modal parameters of the system are extracted by using a dynamic modal decomposition method, the participation factor matrix is calculated, the participation degree of each state quantity and the corresponding physical link to each mode is quantified, the dominant state quantity of the extracted mode and the corresponding physical link are located, the dominant state set is constructed, the time scale distribution of each mode is quantified, the time scale distribution characteristic atlas of different physical links of the overall system is formed, the fast time scale, analysis time scale and slow time scale boundaries in the atlas are divided according to different research needs, and the offshore wind turbine integrated reduced order model is constructed based on the singular perturbation theory. The application can effectively reserve the dynamic features of the offshore wind turbine multi-field coupling characteristics, and provides a precondition and quantitative basis for physical-based model reduction modeling.

The invention discloses a river and lake hydrodynamic model reduced-order inversion initialization method for intelligent water conservancy, and the method comprises the steps: dividing a computational domain into a plurality of sub-domains according to hydraulic characteristics, building a physical constraint equation represented by reduced-order description parameters, and determining an independent to-be-inverted parameter set through the analysis of the degree of freedom; estimating an initial state of a part of sub-domains by using observation information or a priori value at a starting moment; determining an inversion window according to the hydraulic disturbance propagation time; constructing an agent model, and performing joint inversion on the to-be-inverted parameters and the boundary process line parameters to minimize the cost function; and driving the proxy model to generate a global initial state by using the optimal parameter obtained by inversion, mapping the global initial state to a basic model grid, and taking the global initial state as an initial condition after free evolution and boundary smoothing processing. According to the method, actual measurement data are fused through order reduction-inversion, global initial field construction can be rapidly and accurately completed, and the requirement of an intelligent water conservancy system for cold start of a hydrodynamic model is met.

The invention discloses a method and a system for predicting stress response and displacement response of a jacket platform, and belongs to the technical field of health monitoring, and the method comprises the following steps: establishing a finite element model of the jacket platform, and obtaining the inherent frequency and local strain of the jacket platform through simulation; correcting the finite element model of the jacket platform by adjusting model parameters of the finite element model by respectively taking errors corresponding to the simulated and real-time acquired inherent frequency and local strain as variables; taking real wind wave load data as excitation, inputting the real wind wave load data into the finite element correction model of the jacket platform, obtaining stress and displacement response data of the jacket platform through simulation, and respectively constructing a stress response reduced-order model and a displacement response reduced-order model according to optimal modal basis vectors and modal coefficients corresponding to different data sets, and stress response and displacement response of the in-service jacket platform are monitored in real time. According to the method, finite element model correction and model order reduction technologies are combined, so that high-precision prediction of jacket platform stress and displacement response is realized.

The invention provides a global response reconstruction method and device based on an intelligent enhanced reduced-order model, and the method comprises the steps: obtaining the local structure data of a to-be-reconstructed scene, and enabling the local structure data to be obtained through the collection of the to-be-reconstructed scene based on a sensor; an orthogonal base space matrix under model order reduction is obtained, the column vector of the orthogonal base space matrix is an orthogonal base vector, and the orthogonal base vector is expanded into a low-dimensional subspace; and based on the local structure data and the orthogonal base space matrix, performing reconstruction to obtain a high-dimensional global response of the to-be-reconstructed scene matched with the local structure data. A high-dimensional and complex structure response field is projected to a low-dimensional subspace by using an orthogonal base space matrix with physical consistency, so that the high-dimensional global response of a to-be-reconstructed scene can be obtained through accurate reconstruction and deduction based on local structure data of the to-be-reconstructed scene.

This invention proposes a method and system for reducing the order of FPSO models by integrating time-recurrent neural networks and fuzzy logic, belonging to the field of marine engineering structural analysis and intelligent modeling technology. Addressing the problem of large degrees of freedom and low computational efficiency in FPSO finite element models, this invention extracts key degrees of freedom from the high-dimensional finite element model to construct low-dimensional feature vectors. It then uses a fuzzy logic system to perform regularized modeling of the dominant mechanical properties of the structure, achieving initial order reduction. Furthermore, a time-recurrent neural network is introduced to compensate for high-order residual effects that are difficult to accurately characterize using the fuzzy logic model, ultimately resulting in a low-dimensional, high-fidelity reduced-order model. The proposed method significantly reduces the degree of freedom and parameter dimension of the finite element model while maintaining the main mechanical properties of the structure, effectively reducing computational complexity and improving the model's applicability and practical engineering value in applications such as rapid finite element analysis, online structural response prediction, and engineering decision support.

This application discloses a method, device, terminal, and storage medium for simulating the electrical characteristics of integrated circuits. The method includes acquiring a model of the integrated circuit to be simulated; meshing the model to obtain an integrated circuit mesh dataset composed of multiple mesh cells; constructing corresponding charge basis functions and current basis functions based on the mesh cells; using the charge basis functions as the fusion object, progressively merging and updating the charge basis functions; simultaneously, merging and updating the current basis functions through geometric mapping; after all fusion and updates are completed, acquiring the integrated circuit basis function dataset; and performing electrical characteristic simulation on the integrated circuit to be simulated based on the integrated circuit basis function dataset. This achieves rapid updating and low-order representation of complex coupling capacitance matrices, reducing the theoretical complexity from O(n^2) to O(n^2) of traditional back-end order reduction methods. 3 The time complexity is reduced to O(n), while the model accuracy and order reduction effect are maintained, and the efficiency of model order reduction is greatly improved.

The application discloses a kind of double-ended flexible DC system stability analysis method and system, the method includes: establishing single-converter-ideal DC source system, and determining dominant mode;Based on double-ended flexible DC system, the transfer function of DC side state variable is determined;According to the dominant mode, the transfer function is simplified, and the reduced order model is obtained;Based on the reduced order model, the analytical condition of system small disturbance stability is determined, and the stability of double-ended flexible DC system is analyzed using the analytical condition.The application simplifies the model of general double-ended flexible DC system by model reduction, reveals the stability mechanism of the DC side of the system, provides the analytical condition of small disturbance stability of the DC side state variable of flexible DC transmission system, which is conducive to fully utilizing the transmission capacity of double-ended flexible DC system and improving the frequency stability of the system.

The application discloses a novel buffer layer material evaluation method for solving the buffer layer ablation problem, and belongs to the electrical engineering field.The method adopts CT scanning to obtain two-dimensional tomographic slices of the buffer layer material, and carries out AI artifact removal, filtering and three-value processing, and reconstructs a micro model reflecting real pores and fiber structures through a volume data method; the micro model is discretized into a finite element grid and optimized, and is embedded into a cable macro model to construct a macro-micro coupled three-dimensional finite element model; local grid refinement, parallel computing, model order reduction or reasonable selection of a solving method are adopted to accelerate simulation, and the electric field and current distribution are calculated under an electric-thermal-force multi-physical field coupling environment; the ablation resistance of the novel buffer material is compared with that of a polyester non-woven fiber water-blocking tape as a benchmark, and whether the ablation resistance of the novel material meets the standard is judged.The application realizes quantitative evaluation from a real micro structure to system performance, and has the advantages of high precision, good efficiency and strong engineering applicability.

The application discloses a microcircuit model area decomposition modeling method and device, and comprises the following steps: dividing a micro equivalent circuit model of a large-scale electromagnetic problem into a plurality of subareas in a CPU and obtaining corresponding data matrices; creating a reduced matrix corresponding to the subareas in a GPU, and sequentially reducing the plurality of subareas until the plurality of subareas are reduced; transmitting the data matrix of one subarea into the corresponding reduced matrix to reduce the model, updating the data matrix in the reduced matrix, and generating information data after the model of the subarea is reduced to pass the correlation between the plurality of subareas; and updating the data matrix in the corresponding reduced matrix of the next subarea combined with the information data when the model of the next subarea is reduced. The application reduces the correlation between the subareas by area decomposition of the large-scale electromagnetic problem, reduces the microcircuit model of each subarea, and solves the problem that the capacity used for solving the large-scale electromagnetic problem in industrial application is far more than the memory space of the CPU and the GPU.

The application provides a real-time trajectory tracking control method for a rope-driven soft robot arm and belongs to the technical field of soft robot motion control. First, a target function of a trajectory tracking optimal control problem is established according to a target trajectory. Second, a dynamics model of the rope-driven soft robot arm is established by introducing a strain constraint based on a position dynamics method. Third, a reduced order matrix is established by using a modal derivative to realize model reduction of the rope-driven soft robot arm, and the calculation of nonlinear terms is reduced by coefficient combination. Fourth, a calculation formula of trajectory tracking control input is established by using the target function. Finally, the deformation of the soft robot arm is solved by a numerical integration method. The application establishes a simulation framework of the rope-driven soft robot arm based on the position dynamics method to solve the simulation and control problems of the soft robot arm, and aims to provide a new strategy of a complete soft robot arm model verification and real-time control to solve the problem of interaction between the soft robot arm and the environment.

The invention discloses a power distribution network transient process accurate modeling and switching operation optimization method, system and device and a medium, and belongs to the technical field of power systems and automation, and the method comprises the steps: collecting real-time data, evaluating the operation state, constructing a high-fidelity transient simulation model, and carrying out the parameter identification based on historical data; simulating different switching operations by using the model to obtain transient process data; establishing a multi-objective optimization model aiming at minimizing transient impact and optimizing a switching sequence, solving a Pareto optimal solution set by adopting an intelligent algorithm, and giving a configuration scheme; quasi-real-time simulation is realized by adopting parallel computing and model order reduction technologies; the scheme feasibility is verified through a rehearsal system before actual operation, and an operation instruction is generated; and real-time monitoring and comparison with a simulation result are carried out, and emergency adjustment is started according to deviation. According to the method, closed-loop control from accurate modeling, decision optimization and rehearsal verification to real-time feedback is realized, and the safety and stability of switching operation of the power distribution network are improved.

The invention discloses a numerical simulation-based operation parameter optimization method for a negative pressure operation smoke exhaust device, and belongs to the technical field of medical environment control. According to the method, a CFD model of surgical smoke diffusion is established, and a training data set is generated through geometric modeling, physical modeling, grid modeling and boundary condition setting; secondly, constructing a parameterized reduced-order model to realize rapid prediction of a smoke diffusion field; thirdly, a multi-objective function considering smoke concentration control, smoke exhaust power consumption optimization and equipment adjustment smoothness is designed, and state estimation and constrained rolling optimization are achieved in combination with an extended Kalman filter; and finally, dynamically updating the optimal control strategy through sensor feedback, and outputting the optimal control strategy to the smoke exhaust device for execution. According to the method, through fusion of CFD simulation and a model order reduction technology, the bottleneck that full-order model calculation is time-consuming is solved, meanwhile, multi-target collaborative optimization is achieved by means of an MPC framework, the control precision and flexibility of operation smoke exhaust are remarkably improved, and the equipment operation energy consumption is reduced.

The invention discloses a novel buffer layer material evaluation method for solving the problem of buffer layer ablation, and belongs to the field of electrical engineering. The method comprises the following steps: acquiring a two-dimensional fault slice of a buffer layer material by adopting CT scanning, and reconstructing a microscopic model reflecting a real pore and fiber structure through a volume data method through AI artifact removal, filtering and ternary processing; discretizing the model into a finite element mesh, optimizing the finite element mesh, and embedding the finite element mesh into a cable macroscopic model to construct a macro-micro coupling three-dimensional finite element model; accelerating simulation by adopting local grid refinement, parallel calculation, model order reduction or a reasonable selection solution method, and calculating electric field and current distribution in an electricity-heat-force multi-physics field coupling environment; and by taking the polyester non-woven fiber water-blocking tape as a reference, comparing the ablation resistance of the novel buffer material, and judging whether the ablation resistance of the novel material reaches the standard or not. According to the method, quantitative evaluation from a real microstructure to system performance is realized, and the method has the advantages of high precision, good efficiency and high engineering applicability.

The embodiment of the invention provides a simulation method and device, electronic equipment and a storage medium. The method comprises the following steps: constructing a first simulation model and a second simulation model of a target simulation object; order reduction processing is conducted on the first simulation model, a first order reduction model is obtained, and the complexity of the first order reduction model is smaller than that of the first simulation model; the second simulation model is subjected to order reduction processing, a second order reduction model is obtained, and the complexity of the second order reduction model is smaller than that of the second simulation model; based on the first order reduction model and the second order reduction model, simulation testing of the target simulation object is achieved. According to the technical scheme provided by the embodiment of the invention, after the high-precision model is constructed, the calculation amount is reduced through model order reduction, so that the calculation time is shortened, real-time simulation is realized on the premise of high-precision simulation, and the simulation test efficiency and reliability are remarkably improved.

This invention discloses a nonlinear predictive function control method for a fractional-order turbine regulating system, belonging to the field of automatic control technology. First, a sudden load model of the turbine regulating system is introduced, incorporating fractional calculus and dynamic turbine transfer coefficients. Then, based on the series characteristics of the turbine regulating system, a predictive function controller based on a time-varying nonlinear state-space model is designed, effectively improving the unit's speed and stability. The fractional-order model part employs an improved Oustaloup approximation method and an optimal model order reduction algorithm, enabling efficient computer simulation calculations.

The invention discloses an intelligent state prediction method and system for power grid equipment based on digital twinning, and belongs to the technical field of intelligent operation and maintenance of power equipment. Comprising the following steps: collecting data through multiple sensors, performing edge preprocessing and hierarchical network transmission, and performing credible evidence storage by adopting an alliance chain technology; based on credible data, generating an equipment state vector by utilizing evidence theory fusion, constructing a three-layer physical model, and outputting an optimal state estimation sequence of the high-fidelity digital twin through model order reduction and data assimilation; carrying out anomaly detection and fault mode identification based on the sequence, constructing a dynamic causal network to realize fault root cause analysis, and fusing physical simulation and data driving prediction to output a state evolution result; and finally, formalizing the maintenance task into a multi-objective optimization problem, evaluating the risk by calculating the maintenance entropy, and screening and outputting a recommendation scheme based on an autophagy mechanism. According to the method, closed-loop operation and maintenance with data credibility, model high fidelity, diagnosis interpretability and decision intelligence are realized.

The invention relates to a method for efficiently calculating the temperature of a spacecraft structure in a thermal environment, and belongs to the technical field of spacecrafts. Establishing an antenna model based on finite element software, extracting a thermal parameter matrix, and establishing a traditional heat conduction equation; secondly, simulating a multi-orbit periodic temperature field through spaceflight thermal analysis software, and storing a transient temperature sample; secondly, processing a temperature sample by adopting intrinsic orthogonal decomposition (POD), and constructing a POD base mode to realize ten-thousand-dimensional model order reduction; thirdly, establishing a dimensionality reduction heat conduction equation, modeling a radiation heat flow item of the dimensionality reduction heat conduction equation by a feedforward neural network, and implicitly fitting quadruple nonlinearity through a ReLU activation function by the network by taking a temperature generalized coordinate as an input; finally, neural network expression is combined with a POD method, and a complete low-order nonlinear radiation heat conduction equation is established; the method has both physical mechanism completeness and calculation efficiency, wherein the model dimension is reduced by two orders of magnitude, and the real-time processing requirement of a satellite-borne platform is met; the neural network retains the radiation nonlinear nature, and the prediction precision is high.

A method for determining the on-orbit radiation pattern of a satellite antenna considering multiple field effects is presented. This includes: order reduction of the satellite structure finite element model, whole-satellite attitude dynamics modeling, whole-satellite on-orbit thermal analysis modeling, solar radiation moment and Earth's gravity modeling, and their coupled simulation analysis. This invention solves the problems of single-field coupling being unable to solve for continuous attitude changes, difficulties in predicting satellite electrical performance under the coupled effects of structural deformation and thermal analysis, and low computational efficiency. It provides an analytical method and implementation means for accurately predicting the on-orbit electrical performance of satellites with large flexible payloads.

The invention discloses a device thermal management-oriented simulation model order reduction method and system, and belongs to the technical field of numerical calculation and simulation analysis. The method aims at solving the problems that an existing finite element thermal simulation full-order system is large in calculation amount and difficult to process multiple heat sources efficiently, and a projection method is high in memory consumption. The method comprises the following steps: constructing a full-order heat conduction first state equation containing an input matrix representing multiple heat sources or multiple boundary conditions; performing iteration processing on the input matrix by applying a block second-order Arnoldi algorithm to generate an orthogonal basis matrix; reducing the order of the first state equation by using the orthogonal basis matrix and adopting a Galerkin projection method, and constructing a second state equation; and finally, carrying out time domain iteration solution and reduction on the reduced-order equation to obtain a full-order temperature field response result. According to the method, the dynamic characteristics under multiple heat sources are effectively captured by adopting the block second-order Arnoldi algorithm, the memory occupation is remarkably reduced by adopting Galerkin projection, and the simulation time is greatly shortened while high precision is ensured.

The invention discloses a self-adaptive model order reduction method and device, computer equipment and a storage medium. The method comprises the steps that initialization parameters are acquired; performing loop iteration until the actual moment matching order reaches the maximum total moment matching order; calculating residual errors of all moment matching points, identifying the moment matching point with the maximum residual error and recording the maximum residual error; residual errors of all the test points are calculated, the test point with the maximum residual error is recognized, and the maximum residual error is recorded; carrying out convergence check on the maximum value between the maximum residual error of the moment matching point and the maximum residual error of the test point, and if the maximum value is smaller than the tolerance, jumping out of circulation; and dynamically selecting to execute an order-increasing or point-increasing strategy based on the balance parameter: constructing an order-reducing model based on the finally obtained standard orthogonal basis matrix. The method aims at automatically solving the problem of balance between precision and speed in broadband simulation, the high-precision electromagnetic system reduced-order model can be obtained at the calculation cost as small as possible, the calculation efficiency is remarkably improved, and resource consumption is reduced.

The application provides a simulation method, device and medium for a catalytic cracking whole device, and the method comprises the following steps: performing computational fluid dynamics simulation on a reaction-regeneration system by adopting a virtual full cycle method to obtain a first catalytic cracking reaction-regeneration model; performing order reduction on the first catalytic cracking reaction-regeneration model according to the simulation result of the reaction-regeneration system to obtain a second catalytic cracking reaction-regeneration model; and coupling the second catalytic cracking reaction-regeneration model with a fractionation and absorption-stabilization system to perform joint simulation, and performing process simulation on the catalytic cracking whole device. The computational fluid dynamics simulation of the reaction-regeneration system and the process simulation of the fractionation and absorption-stabilization system are effectively coupled to realize real-time and accurate simulation of the catalytic cracking whole device.

The invention relates to a method for predicting the mechanical property of a metal additive manufacturing forming component, and belongs to the technical field of material science. Macroscopically, a finite cell method is adopted to directly carry out regularized grid discretization on computed tomography data, and a complex pretreatment process is avoided; the self-consistent clustering analysis is adopted for solving in the microcosmic process, so that the microcosmic calculation efficiency is greatly improved; and the two scales are coupled in real time for data transmission. A concurrent two-scale solving strategy is adopted to replace an existing single-scale technology, and the influence of a low scale (micro scale) on a macro scale can be naturally considered; a macroscopic finite cell method is adopted to replace a macroscopic finite element method adopted by an existing concurrent multi-scale technology, and a complex pretreatment process is avoided. Microscopic self-consistent clustering analysis is adopted to replace a microscopic crystal plasticity finite element method adopted by an existing concurrent multi-scale technology, model order reduction is carried out on a microscopic representative volume unit, and the calculation efficiency is greatly improved.

The application discloses a model order reduction method, system, terminal and computer readable storage medium of a multi-port linear circuit. The method comprises the following steps: obtaining a to-be-reduced circuit model, reducing the number of ports of the to-be-reduced circuit model according to a port reduction strategy, and obtaining a target circuit model; based on the target circuit model, a residual module and a tensor mapping structure, a target mapping equation is constructed, the tensor mapping structure is used to represent the correlation between the to-be-reduced circuit model and the target circuit model, and the residual module is used to predict the corresponding output according to the input; training data sets are generated according to the to-be-reduced circuit model, the residual module and the tensor mapping structure in the target mapping equation are trained, and a trained target mapping equation is obtained; and a reduced order result is generated according to the target circuit model and the trained target mapping equation. The application can use the trained target mapping equation on the basis of the target circuit model, so that the accuracy of the multi-port model after order reduction can be improved.

The invention provides a centrifugal machine multi-hanging-basket parameter optimization method and a hanging basket, and belongs to the technical field of centrifugal machine hanging baskets. A multi-hanging-basket coupling dynamic model is established based on lightweight rotor frame structure parameters, and a hanging basket swing angle response function and a centrifugal force distribution function are output by adopting a model order reduction technology; establishing a double-layer game optimization model taking maximization of sample processing flux as an upper-layer target and minimization of vibration energy as a lower-layer target, solving to obtain a candidate parameter set, and evaluating and correcting the candidate parameter set through a multi-hanging-basket parameter mapping model based on Tucker decomposition and convolutional neural network fusion. A self-adaptive rotating speed planning algorithm of Coriolis force compensation is adopted to dynamically adjust rotating speed parameters, and the technical problem that optimal balance between sample processing flux maximization and vibration control is difficult to achieve when design parameters of the multi-hanging-basket centrifugal machine are optimized is solved.