191 results about "Partial differential equation" patented technology

The invention discloses a rapid heat transfer simulation method and device based on a neural network, and relates to the technical field of physical simulation. The method comprises the steps that a hybrid neural network model is trained, the model learns operator mapping from an input function to a temperature or heat flow field, and meanwhile physical constraints such as a heat conduction partial differential equation are coded into a loss function; for a new simulation task, single forward inference is carried out by using the operator mapping, and an initial prediction result is rapidly generated; then, according to physical constraints of coding, calculating a physical residual error of initial prediction, and when the residual error exceeds a preset threshold value, executing a small amount of optimization iteration by taking the prediction as an initial value to carry out rapid local correction; the problems that a traditional numerical method is long in calculation time and an existing neural network method is insufficient in physical fidelity are solved, and high efficiency and high precision of heat transfer simulation are achieved.

The invention relates to the technical field of underground engineering safety monitoring and disaster simulation, in particular to a digital twinborn deduction platform for underground engineering disaster chain evolution simulation, which comprises a multi-disaster coupling numerical simulation module used for acquiring multi-source data including design drawing data, geological data and engineering monitoring data, and based on the multi-source data, establishing an initial stress field model by using a numerical calculation method combining a finite element and a finite difference, and carrying out leakage-settlement-structure failure multi-disaster coupling numerical simulation. According to the method, the partial differential equation for describing the evolution law of the underground engineering is used as a constraint term to be embedded into the deduction model, and online identification and dynamic extrapolation of parameters are executed by fusing real-time monitoring data, so that the accuracy of disaster evolution simulation under complex working conditions is remarkably improved; and the problem that the traditional numerical calculation method is difficult to meet the real-time requirement of digital twinning is solved.

The invention discloses a three-dimensional physical field real-time prediction method and system based on geometric deep learning and physical constraint, and relates to the technical field of computer-aided engineering and artificial intelligence. The method comprises the following steps: directly extracting native boundary representation data (B-Rep) of a three-dimensional model from a computer aided design system; constructing a heterogeneous dual graph taking a parameterized curved surface as a graph node, skipping finite element grid division, and aggregating local and global topological features by using a graph neural network; in combination with a physical information driving mechanism, a partial differential equation (PDE) residual error is introduced as a loss function for constraint training, and generalization prediction of a novel geometric structure is realized; and finally, the physical field state quantity is predicted through direct regression and is rendered in real time. An incremental reasoning mechanism based on a local topology subgraph is adopted, millisecond-level physical field real-time feedback under design modification is achieved, and the method is suitable for scheme rapid screening and trend prediction in the initial stage of design.

The invention belongs to the technical field of aero-engine performance testing, particularly relates to a physical constraint-based aero-engine state prediction model construction method and system, and aims to solve the problems of high calculation complexity and poor data quality of an existing physical information model. The method comprises the following steps: acquiring historical operation data of the aero-engine and a physical constraint rule set of engineering simplification; predicting performance parameters by adopting a deep learning model; constructing a total loss function formed by weighting a data loss item and a physical loss item to train the model; wherein the physical loss item is generated based on the deviation degree of the predicted performance parameter and the engineering simplified physical constraint rule set, and is used for replacing the complex partial differential equation constraint. According to the method, the engineering simplified physical rule is introduced, so that the calculation overhead of model training is remarkably reduced, the model is effectively guided to learn the characteristics conforming to the physical rule, and the accuracy and generalization ability of the prediction model are remarkably improved under the condition of limited data.

The invention provides an underground pipeline leakage detection method and system based on physical enhancement thermal inertia imaging, and belongs to the technical field of pipeline detection, and the method comprises the steps: obtaining an earth surface thermal field and environment meteorological parameters of a current target area; the surface thermal field and the environmental meteorological parameters are input into a double-flow physical information neural network model, a predicted background thermal field is obtained, the double-flow physical information neural network model is obtained through training based on leakage-free historical data and a total loss function, and the total loss function comprises physical loss items determined based on a one-dimensional heat conduction partial differential equation; calculating a space-time residual image of the earth surface thermal field and the predicted background thermal field, and obtaining an apparent thermal inertia distribution map of the earth surface based on the space-time residual image and environmental meteorological parameter inversion; and detecting whether the underground pipeline in the target area leaks based on the space-time residual map, the apparent thermal inertia distribution map and a preset condition. According to the invention, the accuracy of underground pipeline leakage detection can be improved.

The invention discloses a cooperative control method for a multi-domain unmanned system. The method comprises the following steps: acquiring an original sensor sequence and a late message set; according to the original sensor sequence, calculating a dominant function and anti-fact regret to obtain a value signal; generating a time delay weight according to the late message set; constructing a space-time source item based on the value signal and the time delay weight; constructing a neural pheromone field by solving partial differential equation dynamics by using space-time source items and pheromone field parameters; under the guidance of the neural pheromone field, sampling to generate a candidate path set; and selecting a cooperative path from the candidate path set for the unmanned system to execute. According to the method, the problems of value characterization and timeliness under asynchronous communication are solved, deep coupling of a decision strategy and a physical model is realized, and the robustness and the adaptive ability of collaborative decision are improved.

The invention belongs to the technical field of oil and gas field development engineering and multiphase flow numerical simulation, and relates to a rapid prediction method for gas-liquid two-phase flow key flow parameters of a horizontal gas well. Constructing the mixed momentum equation, the mass conservation equation and the drift velocity constitutive relation into a partial differential equation residual group; secondly, constructing a physical information Fourier operator network architecture, and learning integral operator mapping from working condition parameters to flow field distribution in a frequency domain by using a lifting layer, a Fourier layer and a projection layer; and finally, calculating a partial derivative of an output variable relative to a space-time coordinate by utilizing an automatic differential technology, and constructing a loss function fusing a data error and a physical PDE residual error to carry out model training. According to the method, millisecond-level accurate prediction of the pressure, the liquid holdup and the flow velocity field of the whole wellbore is achieved, and the problems that traditional numerical simulation calculation is long in time consumption, the real-time monitoring requirement is difficult to meet, and a pure data driving model lacks physical consistency are solved.

Provided are a grouting subgrade water-vapor-heat coupling simulation method and system, a device and a medium, including: constructing a subgrade water-vapor-heat coupling geometric model; acquiring a partial differential equation of a subgrade water-vapor-heat coupling process, and establishing a relationship between physical fields; setting a temperature and water boundary condition; performing mapping and free triangle mesh generation on the subgrade water-vapor-heat coupling geometric model to obtain a meshing model; selecting initial data, and performing a simulation solution on the meshing model to obtain a water-vapor-heat coupling simulation result; and analyzing the impact of a double-layer polyurethane grouting thermal insulation structure on a temperature distribution, a freeze-thaw cycle depth, and water migration of a subgrade.

The application discloses a large-scale PDE neural operator pre-training method based on a high-frequency enhancement module, a partial differential equation (PDE) data set is composed into a mixed data set, a large-scale PDE neural operator is constructed, preprocessed PDE data is mapped to a latent representation space through a space-time encoder, a frequency decomposition module is used for frequency space mapping, the frequency decomposition module includes parallel high-frequency branches and low-frequency branches, and high-frequency features and low-frequency features are obtained respectively; a multi-frequency fusion module (GFM) adaptively fuses the high-frequency features and the low-frequency features through a gating mechanism; finally, a prediction head is used for processing the fused features, and final output features, i.e., predicted physical features of a next time step, are obtained. The application firstly introduces an explicit frequency division and a high-frequency enhancement mechanism, the input field is divided into low-frequency and high-frequency parts, the low-frequency branches / high-frequency branches are used for processing respectively, and thus the model can simultaneously consider global trend modeling and local gradient detail reconstruction.

The invention provides a lithium battery cargo abnormal temperature rise identification method, and belongs to the technical field of lithium battery customs detection.The lithium battery cargo abnormal temperature rise identification method comprises the steps that a temperature sensor array is arranged on the surface of a lithium battery cargo stacking body to collect multi-point temperature time sequence data, and trend characteristics are extracted through wavelet packet decomposition denoising and singular spectrum analysis; establishing a state space model based on a heat conduction partial differential equation, estimating an internal temperature field state by using Kalman filtering, solving a heat conduction inverse problem by using a conjugate gradient regularization iterative algorithm to invert an internal three-dimensional temperature distribution field, and inputting an inversion result and statistical characteristics into a thermal anomaly identification model fused with manifold learning. Dimensionality reduction is carried out through a local linear embedding algorithm, a mahalanobis distance is calculated in a low-dimensional manifold space, an abnormal temperature rise risk score is output, when the score exceeds a preset threshold value, early warning is triggered, and the technical problem that the abnormal temperature rise in the lithium battery cargo stacking body is difficult to accurately recognize through surface temperature measurement is solved.

The invention provides an early warning method for sea surface temperature anomaly detection, which belongs to the technical field of ocean forecasting, and comprises the following steps of: constructing a multi-source sea temperature data fusion system and a multi-scale adaptive grid, generating a high-resolution temperature field by adopting ensemble Kalman filtering data assimilation, extracting an abnormal component by utilizing ensemble empirical mode decomposition, and carrying out early warning on the abnormal component. Multi-level anomaly discrimination is performed based on a sparse coding recognition model and fractal dimension mutation detection, anomaly types are distinguished in combination with an atmospheric compulsive event feature library and a random forest classifier, and partial differential equation inverse problem reverse deduction is performed on ocean endogenous anomaly to reconstruct a three-dimensional anomaly structure. And finally, the early warning level is determined through the early warning decision function and the multi-dimensional indexes, and the technical problem that the real-time performance and the accuracy of sea surface temperature anomaly detection are difficult to guarantee at the same time is solved.

The invention discloses a real-time mapping and evaluation method for the reliability of complex equipment, which is based on physical information neural networks (PINNs) and a model system engineering (MBSE) model, and is characterized in that the method comprises the following steps of: (1) carrying out real-time mapping and evaluation on the reliability of the complex equipment, and (2) carrying out real-time mapping and evaluation on the reliability of the complex equipment based on the MBSE model, and (3) carrying out real-time mapping and evaluation on the reliability of the complex equipment based on the MBSE model, and (4) carrying out real-time mapping and evaluation on the reliability of the complex equipment. The method comprises the following steps: firstly, constructing a system architecture model of equipment by utilizing MBSE, and defining reliability constraint parameters; a partial differential equation describing a component failure mechanism is extracted, a neural network agent model integrating physical information constraints is constructed, and a loss function of the neural network agent model is formed by weighting a data driving item and a physical residual item. Mapping real-time working condition parameters into PINNs input by establishing a data interface of a system architecture model and a PINNs agent model, and outputting a physical performance degradation state; and finally, calculating real-time reliability based on the probability statistical model, and dynamically feeding back the real-time reliability to the system architecture model to update a demand verification state and generate a control instruction. According to the method, the physical partial differential equation is introduced as the regularization constraint of the neural network, so that the problem of poor generalization ability of a pure data driving model under a small sample working condition is solved, and online evaluation and closed-loop control in an equipment operation stage are realized.

The application relates to the computer technical field and discloses a clothing recommendation method and system based on user preferences, which comprises the following steps: constructing a multi-dimensional utility target of user preferences, the target comprising at least one target or a plurality of combinations in clothing style matching degree, diversity index and novelty index; time evolution modeling of user recommendation preferences is carried out, a dynamic weight structure associated with user behavior feedback is established; a recommendation utility function is constructed and multi-target constraints are introduced, the nested optimization structure comprising outer layer recommendation target constraints and inner layer preference evolution constraints; a continuous evolution process of user preference weights is controlled by using partial differential equations; and user short-term behavior feedback and long-term behavior trend data are fused in the recommendation process. In the application, the multi-target recommendation utility function is introduced, and the style matching degree, diversity and novelty targets of the user are jointly modeled in the nested optimization structure, so that the clothing recommendation effect of simultaneously satisfying personalized and exploratory requirements is achieved.

The invention relates to a long straight control cable space radiation field effect calculation and evaluation method, system and device and a medium. The method comprises the following steps: constructing a partial differential equation constraint model based on a cable geometric structure and electromagnetic transient characteristics; fusing an actually measured magnetic field and an additional physical rule through a physical information neural network, and generating a complete electromagnetic data set to compensate for electric field data missing; a multi-dimensional interference factor is calculated by combining time-frequency distribution characteristics, equipment sensitive parameters and cable equivalent inductance, and the radiation field effect complexity is objectively quantified; establishing a probability mapping model of interference indexes and effect states by adopting multi-dimensional Gaussian process regression to realize risk probability distribution prediction; and finally, generating a visual probability curved surface of an electric field-magnetic field plane, and identifying a high-risk area to guide electromagnetic protection optimization. According to the method, the evaluation precision is improved in a data limited scene, and a technical closed loop from multi-dimensional feature fusion to probabilistic risk early warning is realized.

The present application relates to the technical field of smart water affairs and urban public safety, and particularly relates to a kind of artificial intelligence driven urban pipe network flood resilience evaluation method and system, comprising: constructing Riemann metric tensor by using metric matrix containing potential barrier function, and establishing the Riemann manifold embedding space of pipe network physical state;Then, based on fluid energy, a Hamilton function is constructed, and a dynamics prediction model with physical conservation is obtained by using symplectic neural network and symplectic discrete integral format training;Subsequently, a second-order Hamilton-Jacobi-Aleksandrov partial differential equation describing stochastic differential game is constructed, and a physical perception neural network is used to solve and extract the resilience safety boundary;Finally, real-time data is mapped to the manifold space, the Riemann gradient is calculated, and a quadratic programming problem is solved to generate control instructions. The problem of lack of physical constraints and safety bottom line in pipe network control under extreme random working conditions is solved, and real-time closed-loop control with physical consistency is realized.

This invention discloses a method for preparing and intelligently identifying microscopic stealth codes by integrating artificial intelligence and hyperspectral imaging, belonging to the field of anti-counterfeiting technology. The method comprises two parts: preparation and identification. During preparation, photochromic materials are doped into a substrate to form microscopic code points. A hyperspectral feature library is established, and a digital twin initial model containing partial differential equations is constructed, followed by pre-training of a physical information neural network. During identification, measured hyperspectral data of the code points to be tested are collected. Combined with environmental parameters and historical data, the expected state is predicted using the trained physical information neural network. The authenticity is determined by comparing the measured data with the expected state. This invention embeds physical laws into the identification process through a digital twin model and a physical information neural network, achieving proactive prediction of the evolution of the microscopic stealth code state, significantly improving the accuracy, robustness, and anti-counterfeiting capability of anti-counterfeiting identification.

The present application belongs to the technical field of geothermal energy development and utilization, and relates to a hydrothermal type buried pipe heat exchanger optimization design method based on PINN, comprising: 1, numerical simulation and benchmark data set generation; 2, PINN construction and training: constructing a PINN neural network, converting the mass conservation equation and the energy conservation equation in the pore medium into a partial differential equation residual, and together with the discrete data error generated by the numerical simulation software to form a composite loss function; using the nonlinear fitting capability of the PINN neural network, and embedding the physical partial differential equation of fluid flow and heat transfer in the pore medium into the loss function as a soft constraint; 3, data-driven optimization based on the proxy model; the present application not only fundamentally solves the problems of poor generalization ability and easy non-physical interpretation of traditional pure data-driven models, but also breaks through the bottleneck of convergence difficulty and high training cost when solving complex engineering problems by pure PINN.

The invention discloses a large-scale PDE neural operator pre-training method based on a high-frequency enhancement module, and the method comprises the steps: forming a mixed data set through partial differential equation PDE data, constructing a large-scale PDE neural operator, mapping the preprocessed PDE data to a potential representation space through a space-time encoder, carrying out the frequency space mapping of the data through a frequency decomposition module, and carrying out the pre-training of the large-scale PDE neural operator. The frequency decomposition module comprises a high-frequency branch and a low-frequency branch which are parallel to obtain high-frequency features and low-frequency features respectively; the multi-frequency fusion module GFM adaptively fuses the high-frequency features and the low-frequency features through a gating mechanism; and finally, the fused features are processed through a prediction head, and final output features, namely predicted physical features of the next time step, are obtained. According to the method, an explicit frequency division and high-frequency enhancement mechanism is introduced for the first time, an input field is decomposed into a low-frequency part and a high-frequency part, and a low-frequency branch and a high-frequency branch are adopted for processing, so that the model can give consideration to global trend modeling and local gradient detail reconstruction at the same time.

The invention discloses a flow field prediction method and system based on a partial differential equation operator embedded convolutional network, belongs to the technical field of flow field prediction, and solves the problems that in existing flow field prediction, the numerical calculation amount is huge, and a pure data driven model lacks physical consistency and is easy to diverge due to long-time rolling. Comprising the following steps: acquiring fine grid flow field data and physical parameters, preprocessing to obtain a fine grid tensor, and recording a physical parameter vector; performing down-sampling on the fine grid flow field, adopting a fixed convolution kernel discrete partial differential equation operator, advancing a time step, obtaining a coarse grid physical solution, performing up-sampling, and obtaining a physical priori solution; constructing a residual network, learning residual features between a physical prior solution and a flow field truth value, and superposing the residual features and the physical prior solution to form a prediction flow field at the next moment; carrying out model training; and based on the flow field at the current moment, predicting the flow field at the next moment by adopting the trained residual network. The method is suitable for flow field prediction scenes.

The application belongs to the technical field of image segmentation, and particularly relates to a complex scene target segmentation method and system based on image recognition, which comprises the following steps: performing superpixel adaptive division on an input image, fusing gray scale and texture features to determine a superpixel boundary, mapping the superpixel boundary into a graph node and calculating an edge weight, and constructing an undirected weighted graph; adaptively encoding a graph signal, utilizing a hybrid graph wavelet-Fourier joint transform to optimize and separate features, and obtaining purified graph frequency domain features; sparsely reconstructing features through adaptive super-complete dictionary learning, and obtaining target enhanced features; extracting topological parameters based on an improved persistent homology, constructing a topological constraint feature graph, mapping the topological constraint feature graph into an initial contour field, iteratively optimizing a level set and a contour through an adaptive partial differential equation, and obtaining a high-fidelity coarse segmentation result; extracting geometric features to construct a joint constraint model to repair an occluded area, and outputting a precise segmentation result. In the application, sparse topological modeling is adopted, weak features are strengthened, and target discrimination accuracy is improved.

This invention relates to the field of engineering machinery structural optimization technology, specifically to a method and system for topology optimization of excavator structures based on real-time stress field reconstruction. The method involves acquiring deformation and stress data of key parts of the excavator boom using a distributed fiber optic grating sensor array, and then modeling and simulating these data in finite element analysis software to obtain virtual data. A Kalman filter is designed to correct errors between the virtual and actual test data, and the corrected data is used to train the Kalman filter to predict the real-time stress field distribution. Furthermore, a topology optimization model of the excavator structure is established, encoding material properties into weight matrices in a neural network, and using stress constraint partial differential equations as constraints. This achieves high-precision reconstruction of the excavator boom stress field under actual working conditions, providing accurate boundary conditions for topology optimization.

This invention discloses a method for assessing the weathering of sandstone artifacts based on the fusion of multi-source heterogeneous data driven by physical-manifold collaboration, belonging to the field of cultural relic protection technology. Addressing the contradiction that surface spectral data of sandstone artifacts is dense but cannot probe the interior, while internal physical data is accurate but extremely sparse and lossy, this invention proposes a "surface-to-interior" fusion strategy. First, using a physical information deep learning model, physical partial differential equations are introduced as prior constraints to extrapolate sparse point data into a continuous deep physical tensor across the entire field, achieving a "penetrating" effect. Second, based on Riemannian manifold geometry, spectral-physical enhancement features are mapped to the tangent space to extract noise-resistant surface manifold features. Furthermore, through coupled tensor decomposition, deep mechanisms and surface properties are forcibly aligned in the latent feature space. Finally, a high-order Laplacian hypergraph model is used to achieve pixel-level classification of weathering degree. This method effectively solves the problems of spatial scale mismatch and missing physical mechanisms in multi-source data, achieving a non-destructive, full-field, and accurate quantitative assessment of the weathering status of cultural relics.

A digital twin modeling method and device for motors based on multi-field coupled PINN (Partial Differential Equations) belongs to the field of intelligent manufacturing and advanced control technology for motors. Its key features include: acquiring stator current, voltage, rotor position, and vibration signals of the motor to construct a full-domain multimodal sensing layer; constructing a multi-field coupled PINN model based on automatic differentiation; and embedding the multi-field partial differential equations (PDEs) inside the motor as prior knowledge into the network structure of the PINN model. By utilizing PINN technology, the physical laws of PDEs are integrated into the neural network, ensuring both millisecond-level inference speed and physical consistency in areas lacking sensor data, achieving both high fidelity and real-time performance. It enables precise virtual sensing of unmeasurable rotor permanent magnet temperature and local magnetic flux density saturation within the motor, achieving full-domain observability. Through edge-cloud collaboration, it can automatically evolve as the physical entities of the motor age (e.g., resistance changes, magnet demagnetization), maintaining model accuracy throughout its entire lifecycle.

This invention discloses a forest fire risk early warning system based on Internet of Things (IoT) communication, belonging to the field of data processing technology. The system comprises a data acquisition module that collects forest area data from edge devices; an environmental feature module that uses an adaptive forgetting factor recursive method to obtain environmental anomaly response vectors; a feature fusion module that extracts image feature tensors, maps these vectors to scaling matrices, combines them with a forced masking weighted feature tensor, and outputs predicted bounding box parameters; a feature inversion module that obtains a penalty matrix based on the parameters, constructs a partial differential equation for canopy drag force, and solves for the fire plume feature matrix and flow field uncertainty tensor matrix by minimizing the inverse source-finding objective function; and a communication optimization module that extracts elements as fire risk mapping weights, arranges them in descending order, and allocates time resources so that the terminal only sends fire plume elements to the decision center. This invention effectively isolates meteorological interference, improves the accuracy of locating small fire points under dense smoke, and solves the communication congestion problem during extreme fire events.

The invention relates to the technical field of physical field prediction, in particular to a physical field prediction method and device based on gradient recognition parameter tuning and a medium, and the method comprises the steps: determining a space-time computational domain based on a physical problem, and determining a control equation, an initial condition constraint and a boundary condition constraint of the physical problem; constructing a training sample data set; initializing parameters of the neural network, and pre-training the neural network; obtaining pre-trained neural network parameters, and identifying high-contribution neural blocks; on the basis of the high-contribution-degree nerve blocks, parameter fine tuning iteration is carried out on the weights of the combination points of the neural network, and a physical field prediction model is obtained; and inputting the space-time coordinates of a to-be-solved point into the physical field prediction model to obtain a physical field prediction value of the point. The adaptive algorithm framework and the staged training strategy provided by the method enable the model to adapt to different partial differential equations and gradient distribution characteristics, thereby reducing the use threshold of the physical information neural network.

The application provides a KDP crystal surface micro-nano defect DPN water-soluble repair morphology evolution simulation method, and relates to the technical field of micro-nano manufacturing, and aims to solve the problem that there is no quantitative method to simulate the evolution process of the KDP crystal surface micro-nano defect DPN water-soluble repair morphology in the prior art. The method comprises the following steps: step one, constructing a defect local growth mathematical model; step two, performing dimension conversion on the defect depth information, and performing dimension reduction processing on the model; step three, converting the defect local growth mathematical model into a standard partial differential equation form; step four, obtaining the initial value of the defect local growth mathematical model, and setting the model boundary condition; step five, performing parameterization scanning on the undetermined coefficient, and determining the undetermined coefficient value; and step six, simulating the KDP crystal surface micro-nano defect DPN water-soluble repair morphology evolution process. Through dimension conversion, the dimension reduction processing of the model is realized, and finally the simulation of the KDP crystal surface micro-nano defect DPN water-soluble repair morphology evolution process is realized.

A method for maintaining physical assets based on recommendations generated from operational data analysis and a composite model representing multiple models of the physical assets includes: constructing the composite model by a computing device using a port-reduced static condensed reduction primitive approximation with respect to at least a portion of a partial differential equation; the computing device analyzing error indices associated with at least one model within the composite model to determine if the error indices exceed a tolerance level, and accordingly increasing the number of basis functions in the port-reduced static condensed reduction primitive approximation; the computing device receiving first operational data associated with at least one region of the physical asset and updating the composite model; and the computing device providing recommendations for maintaining the physical asset based on the updated composite model.

Examples of the presently disclosed technology provide new systems and methods for real-time temperature propagation and tissue damage visualization during laser interstitial thermal therapy (LITT) procedures that do not rely on real-time MR imaging. Accordingly, examples enable performance of LITT procedures in regular operating rooms lacking MR-equipment-thereby reducing costs and improving availability for LITT procedures. Examples achieve these advantages by leveraging “discretized” patient-specific 3D brain structure representations to perform numerical methods for solving partial differential equations that estimate real-time (or close to real-time) temperature propagation within a patient's brain during a LITT procedure.

The invention provides an underground water movement substitution model construction method and system based on deep learning, and the method comprises the steps: solving a parameterized underground water movement partial differential equation in a heterogeneous aquifer based on a physical information neural network, and constructing a substitution model; constructing a loss function fusing physical information, and enabling the output of the substitution model to meet initial conditions and boundary conditions through a hard constraint method; parameters of the substitution model are optimized through an iterative training process; groundwater head distribution under different hydrogeological conditions is predicted based on the optimized substitution model, and uncertainty quantitative analysis is completed. According to the method, the prediction accuracy and reliability of the substitution model under the condition of no data are greatly improved, and the calculation time for completing uncertainty quantification is greatly shortened.