62 results about "Nonlinear dynamic systems" patented technology

The invention relates to the technical field of BIM, in particular to a BIM-based construction progress intelligent management and control method and system, and the method comprises the steps: constructing a BIM digital twin cloud model, fusing a panoramic image and LiDAR point cloud data, dynamically selecting a data source update model through an algorithm, extracting engineering topology, building a construction network diagram, and generating a state evolution trajectory in combination with environment and resource data. Predicting and visualizing milestone time, comparing field data with a BIM model to generate progress deviation information, performing stability analysis, triggering resource allocation when a threshold value is exceeded, synchronizing a material supplier and a field manager, dynamically adjusting personnel, equipment and materials, dynamically adjusting a milestone plan based on multiple data, and presenting and guiding allocation through the BIM model. The construction progress is ensured to be consistent with the plan, and the complex interaction relationship and dynamic characteristics in the construction process are captured through nonlinear dynamic system modeling in combination with LiDAR point cloud and other high-precision data.

The invention relates to the technical field of fatigue evaluation, and discloses an offshore wind turbine blade frequency domain fatigue evaluation method and system based on an MIMO nonlinear dynamic system, and the method comprises the steps: collecting the load data of an offshore wind turbine blade; preprocessing the load data; an MIMO nonlinear dynamic system is established; establishing a dynamic mapping relation between the preprocessed load data and the frequency domain response by using an MIMO nonlinear dynamic system; decomposing the time domain dynamic response data by using a frequency domain decomposition method, and quantifying the influence weight of each frequency band load on the blade stress based on a frequency domain energy distribution matrix to obtain a frequency domain energy distribution function of the total stress; and calculating the fatigue damage of the blade through a frequency domain fitting optimization method based on the frequency domain energy distribution function of the total stress. The method solves the problems that fatigue damage evaluation confidence is insufficient under the multi-load coupling effect and actual engineering requirements are difficult to meet in the prior art, and has the characteristic that the coupling effect of the wind load and the wave load can be accurately processed.

The invention discloses a permanent magnet synchronous motor full-speed-domain position-sensorless hybrid control method and a permanent magnet synchronous motor full-speed-domain position-sensorless hybrid control system, and the permanent magnet synchronous motor full-speed-domain position-sensorless hybrid control method is constructed by taking a permanent magnet synchronous motor nonlinear dynamic system model as a starting point. A hybrid control method combining high-frequency pulsating square wave signal injection under a linear dynamic system model, a rotor polarity judgment and rotor initial position estimation algorithm and an unscented Kalman filtering extension algorithm in an additive noise form under a nonlinear discrete dynamic system model is adopted; and obtaining the optimal estimation observation value of the system state variable in the static, low-speed, medium-speed and high-speed full-speed domain range under the condition that the permanent magnet synchronous motor is not provided with a mechanical position sensor. The optimal estimation observation value is used for compensating and feeding back a permanent magnet synchronous motor drive control system adjusting strategy, disturbance of load torque is improved, the noise amount and disturbance amount introduced by three-phase current measurement are eliminated, and efficient and accurate control operation of the permanent magnet synchronous motor is achieved.

The invention discloses an attitude control method, system, equipment and medium for an aircraft model full-aircraft wind tunnel test based on incremental nonlinear dynamic inverse, and the control method comprises the steps: building a nonlinear dynamic system according to an aircraft physical model of a wind tunnel test platform, model parameters and CFD pneumatic simulation data, the aircraft height, the vertical speed, the pitch angle and the pitch angle speed are in a system state, and the aircraft height is output of the system; a pitch angle in the system is used as an intermediate variable, the structure of the nonlinear dynamic system is simplified by ignoring a secondary item, and the processed nonlinear dynamic system is obtained; designing an INDI control algorithm, enabling the incremental system to be closed-loop, converting the incremental system into an error tracking system, configuring a pole of the error tracking system, and realizing trajectory tracking of the flight height; the system, the equipment and the medium are used for implementing the method. According to the invention, the control precision and robustness of the aircraft in a complex environment are effectively improved.

The invention discloses a friction torque prediction method based on deep learning, and belongs to the technical field of intelligent manufacturing and machining, and the method comprises the following steps: analyzing the kinematic characteristics of the overall structure of a numerical control machine tool machining center, and describing the dynamic response characteristics of a servo motor by combining the milling process of a numerical control machine tool and referring to a classical Stribeck friction model; the method comprises the following steps: respectively constructing system dynamic models related to X and Y feed axes, determining a dynamic relationship between excitation and response under the action of time-varying disturbance, solving the problems of nonlinear dynamic system modeling and parameter identification under a time-lag condition, and taking a servo system mechanism as a model architecture, introducing a deep neural network as a function approximation model, so as to realize the nonlinear dynamic system modeling and parameter identification. And the problem of quantitative characterization of friction parameters is effectively solved. Prediction conditions of different driving shaft friction torques are systematically evaluated from a time domain perspective, and results show that the method has better prediction precision and stability.

This application discloses a multivariate time series prediction method and apparatus, relating to the field of data prediction technology. The method includes: constructing a network model based on a graph convolutional neural network according to the topology of a target nonlinear dynamic system; the network model includes an adaptive graph module, an encoder, a spatial neural differential equation module, a temporal neural differential equation module, and a decoder connected in sequence; training the network model using a historical dataset of the target nonlinear dynamic system to obtain a multivariate time series prediction model; introducing state feedback through the spatial neural differential equation module to reveal the evolution pattern of the spatiotemporal time series in the spatial dimension, and introducing neural differential equations based on nonlinear state transition theory to simulate the state evolution at the temporal level, thereby improving the effectiveness of data prediction by fusing spatial features while suppressing feature oversmoothing.

The present invention relates to the fields of member management and data analysis, and discloses a member health assessment method based on member churn prediction, comprising the following steps: multi-sensor data acquisition and fusion: using multiple sensors to collect member behavior data, environmental data, social data, and device data, and fusing the data to ensure data accuracy and comprehensiveness; behavior pattern modeling: based on the collected member behavior data, a nonlinear dynamic system modeling method is used to model member behavior to capture long-term trends and short-term fluctuations in member behavior; member churn prediction: based on member behavior data and a churn prediction model, the risk of future member churn is predicted and dynamically adjusted. A nonlinear dynamic system model and a fractal analysis model are used to model member behavior patterns, and the ability to capture behavioral complexity is improved through phase space reconstruction and fractal dimension analysis.

This invention discloses a monitoring method and system for the preparation process of ternary cathode materials based on RVAE. It constructs a nonlinear dynamic system model of the sintering process based on a variational autoencoder; assigns different weights to samples at different times in the constructed nonlinear dynamic system model of the sintering process, derives the loss function of the nonlinear dynamic system model of the sintering process, and trains the model parameters through backpropagation; defines the statistics of the nonlinear dynamic system model of the sintering process based on the cyclic variational autoencoder, and obtains the control threshold of the nonlinear dynamic system model of the sintering process through kernel density estimation; collects online data as a test set for the nonlinear dynamic system model, calculates the monitoring statistics online and compares them with the control limits to determine whether a fault has occurred. This invention can significantly improve the fault detection rate and false alarm rate, providing a strong guarantee for the stable operation of the sintering process.

The application relates to a configuration design method of a three-dimensional orbiting sail with posture stability, and belongs to the field of spacecraft attitude dynamics and structure design.The application is realized by the following method: considering the environmental perturbation of atmospheric resistance, atmospheric resistance moment and gravity gradient moment, an orbit attitude coupling dynamics model of a three-dimensional orbiting sail system is established based on a position vector and Euler angles; the three-dimensional orbiting sail and a spacecraft body are regarded as rigid bodies, and the windward area is considered to be blocked by airflow so as to more accurately describe the orbit movement and attitude movement of the system in the orbiting process; the attitude movement is simplified for the attitude dynamics of the system; the attitude stability of the system in the orbiting process is analyzed by using a phase plane method, a Lyapunov energy function method and other stability analysis methods of nonlinear dynamics systems; the relationship between important structure parameters such as a cone angle and a support rod length of the three-dimensional orbiting sail and the attitude stability is further constructed; and the three-dimensional orbiting sail satisfying the preset attitude stability requirement is obtained based on the relationship, so that the configuration design is realized.

The invention discloses a pathological speech feature processing method and system based on event driving, and the method comprises the steps: obtaining a pathological modal signal, carrying out the phase-space reconstruction through the pathological modal signal, obtaining a phase-space trajectory, and building a nonlinear dynamic system model based on the pathological modal signal and the phase-space trajectory; calculating a Lyapunov index of the system model to determine system characteristics; calculating a phase locking value between the pathological modal signal and an acoustic reference feature of a preset voice regulation and control system so as to couple the intensity; determining a target coding mode of each pathological modal signal based on the system characteristics and the coupling strength, and coding the pathological modal signals to generate a corresponding pulse event sequence; and all pulse event sequences are fused based on the phase locking value to generate a time event stream reflecting a pathological regulation mechanism, so that nonlinear pathological components in the pathological speech are effectively extracted, the pathological modality is coded into an event stream structure, and accurate and explainable pathological speech evaluation is supported.

The invention relates to an adaptive adjustment system for a nonlinear dynamic system. The invention relates to a method, in particular an at least partially computer-implemented method, for operating a control system (1) for a technical device (2) for controlling a technical system (3) using a proportional component and at least one of a differential component and an integral component as a control component, comprising the steps of: providing a control parameter (Kp, Kd, Kp) of a parameter set; ki) for calculating a control component as a function of an operating range determined by an operating point (B) and / or an operating state (Z) of the technical system (3); performing an adjustment on the technical device on the basis of the adjustment deviation (e) and the adjustment parameter; adapting (S7) the adjustment parameters for the one or more operating ranges on the basis of the adjustment behavior within an optimization range (H) assigned to the adaptation condition as a function of the presence of the adaptation condition, wherein the optimization range (H) determines the duration within which a change process using the adjustment deviation (e) is used for adapting the adjustment parameters.

The application discloses a kind of self-adapting state noise compensation's maneuverable spacecraft orbit determination method, belong to aerospace technology field.The application implementation method is: establishing the general dynamic model and only angle observation model of unknown mutation for random nonlinear dynamic system;System state and covariance matrix are predicted based on dynamic model;System measurement residual and covariance matrix are calculated at t k+1 Instant, the matching degree of the prediction result of system based on dynamic model and inorganic model is represented by index function Λ k+1 ;Index function at t k+1 Instant is compared with the average of index function from t1 to t k When the difference between evaluation index and average is greater than threshold value, it is determined that system state mutation occurs, and state noise compensation is made to system prediction result;Orbit determination algorithm gain is calculated based on the result of system state noise compensation, and system prediction result noise compensation result and measurement result z k+1 are weighted to output t k+1 System state and covariance estimate value, and state estimate value at each time is iteratively solved, so that high-precision orbit determination of maneuverable spacecraft is realized.

The present invention proposes a coal preparation calorific value prediction method and system based on edge-cloud collaboration. The method includes: selecting relevant variables affecting calorific value as input and the clean coal calorific value as output, constructing a calorific value digital twin model including a linear model and nonlinear terms; performing parameter identification to obtain identification errors; constructing a nonlinear dynamic system v(k) based on the nonlinear terms and identification errors; and using a long short-term memory (LSTM) multi-layer neural network to construct an offline deep learning model for v(k), an edge-side online deep learning model, and a cloud-based deep learning correction model. The correction is corrected according to all variables updated in real time by a cloud database using a preset self-correction mechanism. The system constructs a calorific value endpoint prediction model based on the nonlinear dynamic system, inputs a density setpoint and related variable data, and outputs a predicted calorific value data. The beneficial effect is to achieve real-time calorific value prediction, assist in density decision-making and calorific value control in the coal preparation process, and improve the quality of clean coal.

The invention discloses a Hamiltonian theory-based perturbation multi-symplectic numerical solution containing a disturbance system, and belongs to the technical field of numerical calculation of a nonlinear dynamic system. The method comprises the following steps: firstly, establishing a mathematical model containing a disturbance mechanical structure, converting the mathematical model into a Dosin Hamiltonian system, introducing small parameters to unfold the system into a power series form, and decomposing an original system into a series of linear Dosin Hamiltonian equations through a perturbation theory; and carrying out numerical solution on the perturbation equation of each order by adopting a Preissmann Box Dosin difference format, and finally synthesizing a dynamic response solution of the system. According to the method, the influence of small disturbance on long-term dynamic behaviors is effectively captured while the sympathetic structure of the system is maintained, the method has the characteristics of high precision and strong structure maintenance, the method is suitable for dynamic response analysis of disturbance-containing mechanical systems and generalized nonlinear systems, and an efficient and reliable numerical tool is provided for design and optimization of engineering structures such as aircrafts.

The application discloses a kind of light reserve pool computing performance promotion method and system based on deviation physical model, including the error data of deviation physical model of initial data as input data import input layer, mismatch between deviation physical model and reserve pool state is analyzed in reserve pool layer to train output weight, the data between based on deviation physical model and reserve pool is weighted, and output weight is obtained by ridge regression, and prediction data is output by output layer based on output weight.The application combines the scheme of time delay semiconductor laser reserve pool based on deviation physical model and light injection, considers the data-assisted prediction effect of predicting chaotic nonlinear dynamic system, breaks through the instability of single prediction based on deviation physical model and the mutual exclusion relationship between virtual node interval and the number of virtual nodes in single time delay optical reserve pool system of light injection, maintains the prerequisite of high-speed information processing rate of system, and improves the prediction performance of system.

The application discloses a nonlinear multi-scale dynamic system control mechanism inference method based on Koopman operator theory, which is based on the Koopman operator theory and establishes a reduced-order mechanism model of a dynamic system from observed time series data. First, the observed time series data is dynamically sampled based on the adaptive sampling strategy designed by the application, the sampling rate is adjusted according to the data change, and the information of different time scales is effectively separated. Then, cubic B-spline interpolation is applied to locally smooth and numerically differentiate the discrete sampling points. On this basis, the nonlinear modeling capability of the Koopman operator theory is used to innovatively construct a Block-Hankel matrix to capture the multi-element dynamic characteristics of the system; and the kernel singular value decomposition technology is used to extract the main dynamic mode, so that the system identification is extended to a nonlinear dynamic system. Finally, the SINDy algorithm is used to identify the reduced-order model, so as to realize accurate modeling of a complex nonlinear multi-scale system. On the basis of the adaptive sampling and multi-scale data processing capability, the application combines the advantages of Koopman embedding and sparse regression, has strong robustness and universality, and has theoretical and practical significance for mechanism modeling of nonlinear and multi-scale systems.

The invention provides a mechanical arm motion system identification method based on an improved RLS algorithm, and relates to the technical field of industrial control process system identification, and the method comprises the following steps: S1, constructing a mechanical arm motion system model; and S2, constructing an identification process of a hierarchical least square algorithm based on unscented Kalman filtering. According to the mechanical arm motion system fractional order modeling and interaction estimation method disclosed by the invention, the problem of low mechanical arm parameter identification precision is solved. According to the method, firstly, a mechanical arm fractional order discrete state space model is built, then an identification process combining unscented Kalman filtering and an improved RLS algorithm is built, the system state and parameters are estimated through interactive iteration of the unscented Kalman filtering and the improved RLS algorithm, and an error monotone decreasing strategy and a self-adaptive forgetting factor optimization algorithm are added. The method is high in convergence speed, high in identification precision, capable of processing noise interference, moderate in calculated amount, capable of achieving online real-time identification, adaptive to the strong nonlinear characteristic of the mechanical arm and capable of being popularized to other fractional order nonlinear dynamic systems.

The present invention discloses a vehicle stability analysis method based on central manifold dimensionality reduction, comprising: establishing a vehicle dynamics model, a wheel dynamics model, and a magic formula tire model; establishing a vehicle lateral and longitudinal coupled five-degree-of-freedom nonlinear dynamics system model; performing dimensionality reduction processing on the nonlinear dynamics system based on central manifold theory to obtain a nonlinear dynamics system model after dimensionality reduction; drawing a phase plane trajectory diagram of the nonlinear dynamics system to determine the equilibrium state in the vehicle's steering and braking integration; and controlling vehicle stability based on unstable regions, stable regions, and instability boundaries. The method establishes a connection between vehicle braking and steering stability, revealing the nonlinear dynamics mechanism by which braking torque, front wheel angle, and road adhesion coefficient affect handling stability; shortening the nonlinear dynamics stability analysis cycle, improving real-time performance, and increasing analysis accuracy; reflecting the system's equilibrium state, and being able to evaluate the impact of initial conditions and parameter changes on system motion.

The invention discloses a depth Koopman modeling method and system fused with differential quadratic programming, and the method comprises the steps: collecting and processing the historical operation data, state data and input data of a nonlinear dynamic system, and dividing the data into a training set and a verification set; a depth Koopman modeling framework integrated with a differentiable quadratic programming layer is constructed; and training the modeling framework by using a historical operation data set, and optimizing to-be-trained parameters through training to finally obtain a dimension raising mapping function of the modeled nonlinear dynamic system and an optimal high-dimensional global linear dynamic model corresponding to the function. According to the method, the dimension raising mapping function can be automatically learned and optimized, and the current optimal high-dimensional global linear dynamic model is calculated in real time by utilizing the differentiable programming layer in the learning process, so that the manual selection process of the dimension raising mapping function with subjectivity and blindness is avoided; and the optimality of the obtained high-dimensional global linear dynamic model can be effectively ensured. Therefore, the modeling precision of the method is effectively improved.

The invention discloses a social network inference method and a security monitoring method, and relates to the technical field of network identification. A social network inference method comprises the following steps: collecting income time sequence characteristic data of all nodes in a game dynamics system; calculating a reference degree sequence formed by all node reference degrees; inhibiting the behavior of the node i, and obtaining new income time sequence characteristic data of all nodes after the node i is inhibited; obtaining a new reference degree sequence based on new income time sequence feature data of all nodes obtained after the suppression of the node i; and comparing a new reference degree sequence obtained after the node i is inhibited with the reference degree sequence of the nodes, finding out the nodes with changed reference degrees so as to obtain network nodes connected with the node i, judging the number of hidden nodes connected with the node i, and finally inferring the network structure of the whole social network. A complex network structure in a nonlinear dynamic system is captured, connection information between nodes is deeply mined, and accuracy and comprehensiveness are improved.

Robotic polishing and assisted rehabilitation therapy applications still face problems such as low motion accuracy, poor stability, and weak compliance. This paper discloses a robot compliant control method that integrates multiple adaptive control mechanisms. This method combines theoretical approaches such as globally stable nonlinear dynamic system learning, Lyapunov stability constraints, vector-valued sparse Gaussian process models, and passive power systems to innovatively construct a robot compliant control method that integrates multiple adaptive control mechanisms, enabling the robot to efficiently adapt to uncertain contact surfaces such as those with easily deformable and weak rigidity.

The invention provides a two-part synchronization method and system of a hybrid coupling fuzzy cooperative competition quaternion neural network, and belongs to the field of nonlinear power systems. The method is used for realizing that a cooperative competition hybrid coupling fuzzy quaternion neural network and a leader model can reach a two-part synchronous state under the action of an event triggering sampling machine controller under the non-periodic DoS attack, and comprises the following steps: S1, designing a two-part synchronous controller; s2, introducing the two synchronous controllers into a cooperative competition hybrid coupling fuzzy quaternion neural network model; and S3, designing algorithm iteration, and obtaining the maximum DoS attack rate allowed by the model after introduction of the two controllers. According to the invention, an event triggering sampler controller under the aperiodic DoS attack is designed, so that a cooperative competition hybrid coupling fuzzy quaternion neural network and a leader can reach a two-part synchronization state, and a two-part synchronization error system can reach a stable state under the aperiodic DoS attack; and the method is of great significance to research on dynamics of a nonlinear system.

The application discloses a piezoelectric driving optimal trajectory tracking control method based on an adaptive step length, and comprises the following steps: a piezoelectric ceramic driver is mathematically modeled by using an electromechanical coupling nonlinear dynamic system model to obtain a piezoelectric ceramic driver model; the piezoelectric ceramic driver model is converted into a linear controllable canonical subsystem; an expected displacement x * (0) at time t=0 is taken as an initial actual tracking displacement x(0); a feedback control u is obtained by using the linear controllable canonical subsystem; the initial actual tracking displacement x(0) is substituted into the linear controllable canonical subsystem, and an actual tracking displacement is obtained by using a Runge-Kutta-Felberg method with an adaptive step length; a performance index function of the linear controllable canonical subsystem is determined, the feedback control u and the actual tracking displacement are substituted into the performance index function, and an optimal performance index of the piezoelectric ceramic driver model is obtained. In the calculation process, the piezoelectric ceramic driver can be improved in calculation efficiency while maintaining high precision.

The invention discloses a piezoelectric ceramic driver control method based on a multi-grid method, and belongs to the technical field of precise driving control. Aiming at the problems of low positioning precision caused by inherent nonlinear characteristics of hysteresis, creep and the like of a piezoelectric ceramic driver, many parameters of a traditional complex model and large calculation amount, the method comprises the following steps: firstly, establishing an electromechanical coupling nonlinear power system model, and accurately converting an original system into a linear system through a feedback linearization technology; then constructing an optimal control problem, deducing a regular equation set by using a Pontryagin minimum principle, and performing time discretization by using an implicit Euler format to ensure numerical stability; and finally, introducing a multi-grid algorithm to efficiently solve the large-scale discrete system. According to the method, the calculation complexity is remarkably reduced while the model precision is reserved, the convergence speed is increased by more than 8 times compared with a traditional iteration method, the robustness to parameter changes is high, a precondition device does not need to be adjusted, the positioning error can be controlled within the range, and the method is suitable for the high-end manufacturing fields such as precise positioning and optical focusing.

This invention relates to the field of engineering machinery safety technology, specifically disclosing a load stability prediction system for telescopic boom forklifts based on the force exerted by the boom. The system collects motion sensing data and stress wave sensing data at the boom hinge; calculates a cross-spectral coherence function based on the motion data to generate a hinge coherence disorder index; performs time-frequency transformation on the stress wave data and automatically identifies energy impact patches, extracting their peak energy, center frequency, and duration; and generates a transient collision intensity spectrum index based on sensitive frequency band mapping and aggregation; constructs a time-series comprehensive feature vector from the two indices, inputs it into a pre-trained multi-layer gated recurrent unit structure for nonlinear dynamic system identification, and outputs a fused stability assessment state quantity; maps this state quantity to a virtual potential energy surface for multi-step forward extrapolation, calculates the dynamic stability margin value, and issues an early warning when the margin is lower than an adaptive threshold; this invention achieves early and accurate prediction of nonlinear jump instability.

The invention relates to a dynamic nonlinear system having a plurality of degrees of freedom. The system has at least one potential element, and eigenmodes of the system are produced by means of the potential element. A potential is produced by means of the at least one potential element, and the potential causes an acceleration tangential to the basic trajectories of the system, a basic trajectory being a trajectory of the potential-free system.

The application discloses a traction system fault detection method based on neighborhood restriction generalized autoencoder and belongs to the technical field of fault diagnosis. In view of defects such as great information loss, insufficient interpretability of potential variables and poor adaptability to nonlinear dynamic systems in existing high-speed train traction system fault detection methods, the method realizes high-precision fault detection through two stages of offline learning and online detection. In the offline learning stage, the Mahalanobis distance is used to determine a sample neighborhood set and weights, a neighborhood restriction generalized autoencoder loss function is constructed by fusing local linear reconstruction error and mutual information regularization terms, and optimal encoders and decoders are trained. Finally, normal state residuals are calculated, and a fault detection threshold is determined based on statistics. In the online detection stage, data are collected in real time and stacked data are constructed, real-time residuals are calculated by using the trained neighborhood restriction generalized autoencoder, and fault detection is realized by comparing statistics and the threshold.

The invention provides a recovery method for autonomous diagnosis of a marine communication link fault, which belongs to the technical field of marine communication, and comprises the following steps: establishing a link stability prediction mechanism based on a nonlinear dynamics theory through four steps of multi-dimensional parameter acquisition, nonlinear dynamic characteristic analysis, stability prediction and early warning and preventive recovery control; the method realizes advanced prediction, early warning and preventive recovery of a maritime communication link fault, and specifically comprises the following steps: acquiring multi-dimensional parameter data of a physical layer, a link layer and a transmission layer; constructing a nonlinear dynamic characteristic model representing a link stability state; calculating a stability index and a fluctuation index based on the hierarchical prediction framework, and generating fault early warning information; according to the method, a maritime communication link is regarded as a nonlinear dynamic system, and measures can be taken before a fault actually occurs.

The invention discloses a traction system fault detection method based on a neighborhood limitation generalized auto-encoder, and belongs to the technical field of fault diagnosis. Aiming at the defects of large information loss, insufficient potential variable interpretation, poor adaptability to a nonlinear dynamic system and the like in the existing high-speed train traction system fault detection method, the method provided by the invention realizes high-precision fault detection through two stages of offline learning and online detection. An off-line learning stage: determining a sample neighborhood set and a weight by using a mahalanobis distance, fusing a local linear reconstruction error and a mutual information regular term to construct a neighborhood limitation generalized auto-encoder loss function, and training to obtain an optimal encoder and an optimal decoder; and finally, calculating a normal state residual error, and determining a fault detection threshold value based on the statistical magnitude. And an online detection stage: collecting data in real time and constructing stacked data, calculating a real-time residual error by using the trained neighborhood limitation generalized auto-encoder, and realizing fault detection by comparing a statistical magnitude with a threshold value.

This invention relates to the fields of computer graphics and image processing technology, specifically to a 3D reconstruction system and method for a 3D unfolded image. The system includes a topological feature extraction module, a folding axis positioning module, a coordinate mapping module, a mesh generation module, an error warning module, and a 3D rendering module. First, it acquires images of the folded paper image and the target unfolded image, extracts topologically invariant feature points, and determines the position parameters of the folding axis in 3D space based on the feature point matching results. It then establishes a mapping relationship from 2D coordinates to 3D coordinates, constructs a triangular mesh, and detects fold lines. The mesh is classified, and the relationship between the normal vector and the center point is analyzed to identify erroneous folding regions. Finally, it generates a 3D model visualization result. Topologically invariant feature extraction and manifold mapping techniques are used to improve the accuracy of feature matching. Lie group transformation theory is introduced to construct an accurate coordinate mapping relationship, and an error warning mechanism based on nonlinear dynamic system theory is developed, effectively improving reconstruction accuracy and efficiency.