1120 results about "Surrogate model" patented technology

The systems and methods provide a model deployment criterion. The model deployment criterion indicates a difference in a value against which the proxy model may be measured to determine when, if ever, the proxy model should be deployed to replace the existing rule engine. The model deployment criterion may be keyed to the proxy model (e.g., based on a difference in its size, throughput speed, number of changes, etc.), the existing rule engine (e.g., based on a difference in its age, update occurrences to its rule base, etc.), and / or comparisons between models (e.g., based on differences in results, throughput speed, efficiency, etc.).

A structural-health-monitoring system is disclosed for real-time detection and autonomous maintenance of physical structures. The system includes a sensor network comprising at least one strain gauge and one tri-axial accelerometer mounted on the structure to generate real-time sensor signals. A perception module filters and normalizes the signals and extracts numerical features such as peak amplitude and dominant frequency. A digital-twin module maintains a finite-element model updated in response to the extracted features. A data-driven surrogate model predicts sensor behavior and refines itself using machine-learning techniques. An anomaly-detection module computes an anomaly score from model residuals or classifier outputs. Upon exceeding a threshold, a maintenance module initiates a maintenance action, including generating an inspection schedule or issuing a control signal to an autonomous inspection or repair device. A learning module continuously improves system performance using reinforcement learning based on historical outcomes. The system supports predictive diagnostics, robotic repair, and automated optimization for long-term structural integrity.

The invention discloses a valve performance simulation and optimization method and system based on mixed deep learning, and the method comprises the steps: S1, carrying out the finite element simulation analysis of a valve, and solving through the thermosetting coupling of a multi-physics field, stress, deformation and temperature distribution data of the valve under different working conditions and maximum thermal stress and maximum thermal deformation displacement of key parts of the valve are obtained; s2, after data processing, taking stress, deformation and temperature distribution data of the valve under different working conditions as input parameters, taking maximum thermal stress and maximum thermal deformation displacement as output parameters, training a neural network model, and establishing a thermal-structural performance prediction model; s3, on the basis of a prediction result, a reference point-based non-dominated genetic algorithm is adopted, a reference point generation and self-adaptive agent model is combined, and multi-objective optimization is carried out to output an optimal parameter combination so as to adjust valve design; according to the invention, deep learning and multi-objective optimization are carried out on the finite element simulation result by using the neural network, and the valve performance prediction precision and optimization efficiency are improved.

The invention provides an airfoil flow field prediction method based on multi-task learning, and the method comprises the steps: firstly building an airfoil flow field prediction model based on multi-task learning, which comprises an encoder, a backbone network and a multi-head decoder; then training the established model by using an airfoil flow field data set and adopting an optimization strategy based on multi-task learning; and finally, performing geometric parameterization and grid generation processing on an airfoil which actually needs flow field prediction to obtain standardized airfoil flow field data, and inputting the airfoil flow field data into the trained airfoil flow field prediction model for prediction to obtain full flow field physical quantity distribution and lift-drag coefficient information. According to the method, a multi-task loss optimization strategy is used in the prediction model training process, so that the problem of conflict between airfoil profile surface loss and volume loss optimization can be effectively solved, and then the construction of an airfoil profile proxy model and the accurate prediction of flow field information and aerodynamic parameters are realized.

The invention provides an airfoil profile adaptive design and optimization method based on reinforcement learning and a conditional generative adversarial network, and belongs to the field of airfoil profile optimization design. The method aims at achieving the two core functions of self-adaptive optimal airfoil generation and rapid optimization. According to the first function, through interaction of a reinforcement learning agent and a conditional generative adversarial network generator, generation strategies for generating optimal airfoil profile potential variables under different design conditions are continuously learned, so that the airfoil profile with the optimal aerodynamic performance is adaptively generated, intelligent and adaptive design of the airfoil profile is achieved, and the optimization period is remarkably shortened. Secondly, a self-adaptive agent model based on meta-learning is introduced, the model not only establishes a mapping relation among design conditions, aerodynamic performance and potential variables and directly predicts the optimal potential variables of a specific airfoil profile, but also can quickly adapt to an updated database through simple adjustment; the agent model predicts the optimal potential variable corresponding to the specific airfoil profile for the self-adaptive generation framework, so that the agent can quickly optimize the specific airfoil profile under the condition of fine adjustment.

The invention relates to a thermal coupling over-reduced order prediction method and system for a multi-component assembled nonlinear system. The method comprises the following steps: collecting multi-scale physical field data; constructing an intrinsic orthogonal decomposition basis function space of a temperature field and a stress field, and establishing a double-field coupling constraint equation; constructing contact thermal resistance parameterized proxy models of a cylinder contact area, a bolt area and a free deformation area by adopting a domain discrete empirical interpolation method; constructing a parametric intrinsic mode tensor network to obtain a decline model which is used for realizing real-time reconstruction of a mode basis function through acquired tensor slices in an online stage; performing dynamic inversion based on a modal basis function reconstruction result, and outputting a predicted transient displacement field, a predicted temperature gradient field and a predicted contact stress field; and obtaining real-time parameters, and calculating a residual error with a corresponding prediction result so as to dynamically update the primary function and interpolation point distribution. Compared with the prior art, the real-time prediction of the transient thermal coupling of the multi-component contact system is realized on the premise of ensuring the precision.

The invention discloses a robust optimization method based on D optimum and considering intake and exhaust influences, a medium and a program product, and belongs to the technical field of aerodynamic optimization design of aircrafts. According to the method, a representative small sample set is constructed by introducing a D optimal criterion, optimization coverage of a design space is realized by maximizing an information matrix determinant, and an optimization objective function is formed by a weighted sum of a mean value and a standard deviation of aerodynamic performance so as to balance performance improvement and robustness optimization. An FFD method is adopted to carry out geometric parametric modeling on the aerodynamic configuration and the intake and exhaust system of the aircraft, an uncertainty quantitative agent model is established through a GPCE method, statistical characteristics and gradient information of a target function are analyzed and calculated, iterative updating is carried out through a gradient optimization algorithm, and it is ensured that design constraints and convergence criteria are met. According to the method, the uncertainty influence of geometric parameters and intake and exhaust conditions is fully considered, the aerodynamic performance of the aircraft is improved, and the adaptability and robustness under actual flight conditions are enhanced.

The invention discloses a mold and formwork design system based on virtual simulation, and particularly relates to the field of mold design, the mold and formwork design system comprises a multi-mode perception fusion module, a digital twin modeling module, a multi-physics field coupling simulation module, a hybrid intelligent optimization module, a digital twin closed loop verification module and a data center module, and each module forms a design closed loop through a data center. The multi-modal sensing fusion module adaptively collects and fuses multi-source signals to generate high-credibility data; the digital twin modeling module constructs and iteratively corrects a model based on the data; the multi-physics field coupling simulation module realizes multi-solver co-simulation under a dynamic boundary; the hybrid intelligent optimization module accelerates to generate an optimal solution through secondary optimization and an agent model; the digital twin closed-loop verification module detects defects and generates a correction instruction; the data center module is responsible for data management, cross-module scheduling and precision control; the system improves the design precision and efficiency of the mold frame, reduces the physical mold testing cost, and is suitable for a high-precision mold development scene.

The systems and methods provide a model deployment criterion. The model deployment criterion indicates a difference in a value against which the proxy model may be measured to determine when, if ever, the proxy model should be deployed to replace the existing rule engine. The model deployment criterion may be keyed to the proxy model (e.g., based on a difference in its size, throughput speed, number of changes, etc.), the existing rule engine (e.g., based on a difference in its age, update occurrences to its rule base, etc.), and / or comparisons between models (e.g., based on differences in results, throughput speed, efficiency, etc.).

The invention discloses an aircraft aerodynamic performance analysis method, which comprises the following steps: constructing a multi-dimensional parameterized model integrating aerodynamic / structure / thermal control, and outputting a global design space; based on a global design space, fusing sparse fluid dynamics and engineering data to train an agent model, and verifying the cross-speed domain reliability; decoupling the centroid / lift-drag ratio / thermal load conflict by using the proxy model, and generating a speed domain adaptive configuration library; executing fluid-structure-thermal field strong coupling simulation by utilizing a shock wave-boundary layer self-adaptive grid technology aiming at the speed domain self-adaptive configuration library; based on a fluid-structure-thermal field strong coupling mechanism, utilizing a dynamic sensing technology to predict the aerodynamic performance of the aircraft; and integrating dynamic prediction data and digital simulation, calibrating model parameters, and outputting an aerodynamic performance report and an optimization scheme. According to the method, the problems of insufficient multi-physical field integration, low cross-speed domain reliability, poor configuration self-adaption and low coupling simulation precision in aerodynamic performance analysis of a traditional aircraft are solved.

A method of optimizing a set of oil producing wells to be drilled in a reservoir includes receiving a base reservoir simulation representative of a geological oil reservoir, clustering the reservoir, and generating, in a first process, a possible pool of well positions in the reservoir, where each well position satisfies plausibility and respects operational constraints. The method further includes, in a second process, optimizing the pool of well positions, the second process maximizing a dynamic variable of the reservoir, and outputting the optimized pool of wells to a user. In embodiments, the second process further includes training a DNN surrogate model to forecast the production of the reservoir; and using the trained DNN surrogate model, optimizing the pool of wells.

The invention discloses a high-precision static aeroelastic model optimization design method based on a model correction technology, and relates to the technical field of aircraft design, and the method comprises the following steps: S1, firstly constructing an initial model, and carrying out statics pre-analysis to verify integrity; s2, executing SOL 101 statics analysis based on the initial model and outputting a physical field result; s3, carrying out consistency analysis in combination with test data and generating a correction decision; s4, screening high-priority correction parameters through local or global sensitivity analysis; s5, correcting model parameters by adopting a mixed algorithm of a gradient method and an agent model, and verifying precision and generalization ability; s6, the corrected model is output as a Nastran file and a reduced-order model in a standardized mode, and a parameter change log is recorded; s7, executing static aeroelastic coupling and flutter analysis, and feeding back a result to drive optimization iteration; s8, constructing a multidisciplinary coupling optimization model in combination with aeroelastic and flutter results to realize collaborative optimization; and S9, finally performing engineering standardization packaging on the optimization model and outputting a verification report.

The invention discloses a novel method for constructing a physical constraint oil reservoir agent model by coupling a numerical simulator, and relates to the field of intelligent oil and gas field exploitation, and the method comprises the following steps: establishing an oil reservoir model, randomly disturbing injection-production parameters, and generating an injection-production parameter data set; the injection-production parameters are input into an oil reservoir model for simulation, state parameters are obtained, and the injection-production parameters and the state parameters form label data; building an R-U-Net neural network as an agent model; coupling a numerical simulator, and calculating mixing loss; and updating network parameters of the proxy model by using mixed loss back propagation to complete proxy model training. According to the method, the quality residual generated by discrete difference and the physical loss calculated by the Jacobian matrix are fused as constraint conditions, and the agent model training process is optimized; the precise modeling capability of a numerical simulator and the efficient prediction advantage of an agent model are fused, the problem that a traditional simulator consumes long calculation time in the linear solving process is effectively solved, and the limitation that the agent model is difficult to embed into a complex physical equation is made up.

A combined hyperparameter and proxy model tuning method is described. The method involves multiple search iterations. In each search iteration, candidate hyperparameters are considered. An initial (‘seed’) hyperparameter is determined, and used to train one or more first proxy models on a target dataset. From the first proxy model(s), one or more first synthetic datasets are sampled. A first evaluation model is fitted to each first synthetic dataset, for each candidate hyperparameter, enabling each candidate hyperparameter to be scored. Based on the respective scores assigned to the candidate hyperparameters, a candidate hyperparameter is selected and used to train one or more second proxy models on the target dataset

The invention discloses an oil reservoir carbon dioxide geological sequestration process parameter multi-objective optimization method, and relates to the technical field of oil exploitation, and the method comprises the steps: employing an oil reservoir carbon dioxide geological sequestration effect prediction data set to train a variational automatic coding-deep belief neural network model, a reservoir carbon dioxide geological sequestration effect prediction agent model is obtained; constructing an oil reservoir carbon dioxide geological sequestration effect optimization model according to the safety risk coefficient, the effective sequestration coefficient and the benefit optimization coefficient; the reservoir carbon dioxide geological sequestration effect prediction agent model is combined, a non-dominated sorting genetic algorithm is utilized, and a reservoir carbon dioxide geological sequestration process parameter multi-objective optimization mathematical model is solved. The constructed reservoir carbon dioxide geological sequestration process parameter multi-objective optimization mathematical model aims at sequestration safety, potential and economy; and the optimization precision and efficiency of the oil reservoir carbon dioxide geological sequestration process parameters are improved.

The invention discloses an urban drainage system multi-target prediction method based on an SWMM and a graph convolutional neural network, and the method comprises the steps: obtaining the actual measurement data of a target urban drainage system pipeline and an inspection well, constructing an SWMM model, and outputting the data; generating a space-time diagram sequence deep learning network GraphSAGE-GRU model on the basis of a graph convolutional neural network GraphSAGE and a gated cycle unit network GRU; and taking the preprocessed data as input, inputting real-time or predicted rainfall data by utilizing the trained model, and synchronously outputting node water head and pipeline load prediction results of the target urban drainage system. According to the method, the problem that the topological structure of an urban drainage system is not considered in a traditional method can be solved, and the hydraulic attributes of the inspection well and the pipeline can be comprehensively output. The problem that a traditional agent model is too black and lacks structural information is solved, and more simulation result output is provided by setting a space-time diagram structure.

A method, computer program product, and computing system for processing a query provided to a generative AI model. A content portion retrieved by a Retrieval Augmented Generation system for the query is processed. User context information associated with a user providing the query is determined. Hyperparameters are generated for processing the prompt with the generative AI model by processing the query, the content portion, and the user context information using run-time surrogate model inversion optimization.

The systems and methods provide a model deployment criterion. The model deployment criterion indicates a difference in a value against which the proxy model may be measured to determine when, if ever, the proxy model should be deployed to replace the existing rule engine. The model deployment criterion may be keyed to the proxy model (e.g., based on a difference in its size, throughput speed, number of changes, etc.), the existing rule engine (e.g., based on a difference in its age, update occurrences to its rule base, etc.), and / or comparisons between models (e.g., based on differences in results, throughput speed, efficiency, etc.).

The invention belongs to the field of uncertainty quantification and robustness design optimization of aero-engine air-cooled turbines, and particularly discloses a one-dimensional uncertainty design optimization method and system for a reverse cooling turbine based on a one-dimensional aerodynamic analysis method for the reverse cooling turbine. Forming an augmented space by the optimization variables and the uncertainty parameters, and generating a sample set; establishing a Kriging global agent model, and generating an uncertainty parameter sample set; an ASPC model is established, and one-dimensional uncertainty quantification of the reverse cooling turbine is completed; an NSGA-II multi-objective optimization algorithm is coupled, and one-dimensional robustness design optimization of the reverse cooling turbine is completed. The one-dimensional uncertainty design optimization method and system framework of the reverse cooling turbine are provided and established, aerodynamic performance analysis of the reverse cooling turbine can be completed through simple parameters, a geometric entity is not needed, and tasks such as one-dimensional uncertainty quantification and robustness design optimization of the reverse cooling turbine can be achieved.

A combined hyperparameter and proxy model tuning method is described. The method involves multiple search iterations. In each search iteration, candidate hyperparameters are considered. An initial ('seed') hyperparameter is determined, and used to train one or more first proxy models on a target dataset. From the first proxy model(s), one or more first synthetic datasets are sampled. A first evaluation model is fitted to each first synthetic dataset, for each candidate hyperparameter, enabling each candidate hyperparameter to be scored. Based on the respective scores assigned to the candidate hyperparameters, a candidate hyperparameter is selected and used to train one or more second proxy models on the target dataset.

A computer-implemented method of generating black-box adversarial inputs to a perception component using a surrogate model of the perception component comprises receiving an initial input to the perception component and repeatedly perturbing the initial input until an adversarial input is found that satisfies an attack objective by: performing a primary attack process by perturbing the initial input based on a computed gradient of a surrogate attack loss function of the surrogate model that encodes the attack objective; wherein, if the primary attack process terminates without finding any perturbed input satisfying the promising attack condition, a backup attack process is performed to perform a randomized search of the input space of the perception component, guided by the surrogate model, until a perturbed input satisfying the promising attack condition is found; wherein the primary attack process is repeated based on the perturbed input found by the primary attack process or backup process.

The invention discloses an optimization design method for a control surface of a wide-speed-range large-attack-angle reusable carrier, and relates to optimization design of aerodynamic configuration of an aircraft. The method comprises the following steps: S1, setting a control surface aerodynamic configuration design variable range, and generating an initial sample library by adopting Latin hypercube sampling; s2, aerodynamic parameters are obtained through CFD simulation, and a Kriging proxy model is trained; s3, evaluating the precision of the proxy model by taking a U learning function as a criterion, and stopping adding points when the minimum U function value is smaller than a threshold value; s4, constructing an optimization model which takes maximization of the lift-drag ratio and the static stability margin under the hypersonic speed as a target and takes the condition that the aerodynamic parameters of the supersonic speed / subsonic speed are not lower than a reference value and the hinge moment as constraints; and S5, performing iterative optimization by adopting an improved multi-target particle swarm algorithm based on genetic algorithm crossover mutation operation, updating the proxy model after the optimal solution of each generation is subjected to CFD verification, and outputting an optimal solution set. The problem of wide-speed-range aerodynamic configuration contradictions is solved, and the comprehensive performance of the carrier is remarkably improved.

Systems, apparatuses, methods, and computer program products are disclosed for developing an optimized debt service strategy solution utilizing products across multiple product categories. An example method includes receiving, by communications hardware, a user dataset. The example method also includes processing, by surrogate modeling circuitry, the user dataset using a plurality of surrogate models. The example method also includes generating, by the surrogate modeling circuitry and based on the processing of the user dataset, a parameter estimation set. The example method also includes determining, by optimizer modeling circuitry and based on the user dataset and the parameter estimation set, at least one debt service strategy solution comprising at least one product of a first product category from a plurality of products associated with multiple product categories. The example method also includes causing presentation, by communications hardware, of the at least one debt service strategy solution.

The invention discloses an optical system multi-objective optimization method based on computer vision, and relates to the technical field of intelligent optics, and the method comprises the steps: collecting initial design parameters and an installation and adjustment tolerance range of an optical device, and obtaining an initial design vector set; based on the initial design vector set, performing simulation and image processing under different working conditions, extracting image quality and image features, and obtaining same-domain features; constructing a multi-target agent model fusing optical indexes and task performance indexes through same-domain features; obtaining a target function interface according to the multi-target agent model; on the basis of the target function interface, global optimization and local refinement are adopted to obtain a candidate optimal solution set; and through the candidate optimal solution set, re-evaluation is carried out in uncertainty evaluation and robustness screening, and a final nominal solution is obtained. According to the method, the multi-target agent model fusing the optical indexes and the task performance indexes is constructed, and prediction, constraint evaluation and risk sensitivity scoring are performed on the candidate design vectors, so that joint optimization of the optical performance and the task performance is realized.

The invention provides a centrifugal pump performance optimization method and device, computer equipment and a storage medium, and belongs to the field of centrifugal pump design. The method comprises the steps that a parameterized model of a centrifugal pump impeller is established to define the geometrical shape of the centrifugal pump impeller, a design space is determined, and a plurality of sample points are generated; for each sample point, performing computational fluid dynamics simulation, adjusting inlet pressure, monitoring a ratio of simulation lift to rated lift, determining critical cavitation pressure corresponding to the sample point, and storing flow field data to construct a data set; training an agent model by using the data set; the lift, the efficiency and the critical cavitation pressure serve as targets, design variables serve as optimization variables, performance is evaluated through an agent model, global optimization is conducted through a multi-target optimization algorithm, and a target design scheme is selected from a Pareto solution set obtained through optimization. Therefore, the centrifugal pump performance under different design variables can be quickly and accurately predicted, multiple targets are effectively balanced, and a comprehensive optimization solution is provided.

To create high-performance metasurface devices (110) in an inverse design process over a large design space (100), a deep neural network may be used as a surrogate model in lieu of a full physics simulation to more efficiently predict figures of merit for given input metasurface topologies during the iterative topology optimization. The neural network may also serve to efficiently compute, via backpropagation, gradients of the figures of merit with respect to design parameters, as are used to update the topology in each iteration.

The invention discloses a carbon dioxide oil displacement burying multi-objective optimization method based on a self-adaptive agent model, and relates to the technical field of oil reservoir injection and production optimization. The method comprises the following steps: firstly, constructing a carbon dioxide flooding embedding injection-production optimization model, then acquiring a plurality of initial samples by adopting Latin hypercube sampling to construct a database, preferably selecting each target agent model based on the database, and then generating a Pareto leading edge by utilizing a dominating class search strategy, a decomposing class search strategy and an index class search strategy; and in the optimization stage, a preferred potential solution is searched according to a hypervolume improvement maximum strategy, numerical simulation is carried out, the database is updated until a preset number of times is reached, each iteration optimization scheme, an oil reservoir net present value and a carbon dioxide burying amount are output, and multi-target optimization of carbon dioxide flooding burying is completed. According to the method, the carbon dioxide flooding multi-objective optimization efficiency is improved, meanwhile, the search direction is dynamically adjusted through the hyper-volume evaluation index, the Pareto frontier is accelerated, and accurate prediction of the carbon dioxide flooding development scheme is achieved.

The invention relates to the technical field of gas turbine performance prediction, and provides a thermodynamics-based gas turbine performance lightweight modeling method, which comprises the following steps of: 1, acquiring actual operation data of a gas turbine, and preprocessing the operation data; 2, establishing a thermodynamic model of the gas turbine; 3, establishing a gas turbine agent model; and step 4, training the gas turbine proxy model to obtain a trained gas turbine proxy model. According to the method, the influence of external environment conditions and power requirements on working condition operation conditions can be eliminated, and key gas turbine performance parameters can be quickly obtained.

The invention discloses a network flow confrontation defense method based on heuristic ant colony optimization, and belongs to the technical field of network security and privacy protection. According to the method, a self-adaptive disturbance generation strategy is designed according to spatial and temporal features of a network traffic trajectory, a disturbance injection rule meeting real-time constraint and direction consistency is established, an integrated substitution model is constructed, a feature fusion mechanism of spatial convolution, time sequence circulation and noise reduction coding is used, and a traffic fingerprint identification model with strong generalization ability is generated. An improved ant colony optimization algorithm is adopted to improve the disturbance search capability, and a batch processing method and a universal patch scoring mechanism are integrated to improve the defense universality of samples among different network flows. The method can be directly applied to a black box scene of real-time communication, gradient information or white box access permission of a target model is not needed, the attack recognition suppression rate exceeding 90% is achieved when only 20% of bandwidth overhead is increased, defense universality and deployment feasibility are remarkably improved, and meanwhile disturbance overhead is reduced.

The invention provides an automatic agent model construction method for a product CAD (computer-aided design) model, which comprises the following steps of: defining a part CAD model of a product form and a size and a dependency relationship of a logic relationship and a constraint between different parts; according to a mapping rule for converting an assembling CAD model into Sysml-based model elements based on a plurality of parts and relative positions and connection modes among the plurality of parts, realizing standardized proxy model expression of the product CAD model; and realizing automatic conversion from a native model to a Sysm l-based proxy model based on ATL, and realizing automatic construction of the whole process of the proxy model. A mechanical arm product CAD model serves as an example, an agent model of the mechanical arm product CAD model is constructed, and the abstraction and expression process of the product CAD model is displayed. According to the method, the CAD model of the product can be effectively supported to be incorporated into a digital main line, rapid transmission and feedback of information in the whole life cycle process of the product are promoted, and information cross-domain integration and collaboration are promoted.