2715 results about "Covariance matrix" patented technology

The invention relates to the technical field of online calibration, in particular to a multi-parameter fusion intelligent electric energy meter online calibration method and system, and the method comprises the following steps: collecting voltage waveforms, current harmonics and active power data, dividing windows, calculating a covariance matrix, and generating a feature set; a power factor curvature extreme value and a temperature inflection point offset are analyzed to generate an interference identifier, current density distribution and a voltage distortion spectrum are jointly analyzed to extract a harmonic energy ratio to generate a feature vector, and phase compensation is performed on a pulse sequence to generate a calibration instruction set. According to the method, the covariance matrix is constructed by synchronously collecting voltage and current power parameters, the abnormal mark section is generated by combining the temperature change and the covariance difference value, the environment disturbance and the real deviation are effectively distinguished, and the temperature hysteresis effect is identified through the power factor curvature extreme value and the temperature inflection point offset. And combining current density and voltage distortion spectrum analysis to extract a fundamental wave and harmonic wave energy ratio, establishing a composite calibration reference, and dynamically adjusting a pulse duty ratio to realize harmonic wave energy compensation.

The invention discloses a three-dimensional Gaussian sputtering scene reconstruction method based on structure perception refined Gaussian, and aims to solve the problems of Gaussian drift, edge blur, structure artifacts and the like of a reconstruction model due to the fact that sparse point cloud contains outliers, Gaussian morphology and normal are mismatched and a multi-dimensional optimization target is lacked in an existing three-dimensional Gaussian sputtering reconstruction method. A key frame is extracted by collecting target scene video data, sparse three-dimensional point clouds are reconstructed by using an SfM algorithm, a depth map and a normal map are generated through a Lotus model, three-dimensional Gaussian distribution is initialized after the sparse point clouds are filtered, a Gaussian covariance matrix is adjusted by using a normal consistency regular term, and the sparse point clouds are extracted. And after structure attribute analysis is carried out, a comprehensive scoring function is constructed to screen Gaussian points, and finally, a combined training framework including luminosity, normal consistency and structure continuity loss is adopted to optimize and generate a three-dimensional Gaussian scene model. The method is mainly applied to the field of three-dimensional reconstruction and multi-view rendering, and scene reconstruction precision and geometric consistency can be improved.

The invention provides an automobile magnesium alloy pre-twin crystal deformation data analysis system, and relates to the technical field of data processing, and the system comprises a twin crystal correction module which is provided with a first target monitoring point and a second target monitoring point in a pre-twin crystal area, the first target monitoring point is located at a crystal boundary intersection, and the second target monitoring point is located at the center of a twin crystal zone; acquiring orientation difference angle and strain energy density difference data of the two monitoring points in real time, calculating a covariance matrix of change characteristics of the two monitoring points, and correcting a proliferation rate parameter and an orientation rotation parameter in the dynamic evolution matrix according to a correlation coefficient of the covariance matrix to obtain a corrected dynamic evolution matrix; and the optimization module is used for adjusting a hot extrusion process parameter group according to the twin crystal density gradient distribution output by the corrected dynamic evolution matrix, the parameter group comprises a mold temperature, an extrusion speed and a strain path, and the twin crystal density is dynamically controlled in a target interval. According to the invention, the complex stress state of the automobile part under the actual working condition can be accurately simulated.

The invention relates to the technical field of three-dimensional modeling in engineering surveying, in particular to a point cloud building component modeling method and system based on feature extraction. Three-dimensional point cloud data of building components are obtained, a multi-scale local neighborhood calculation covariance matrix is constructed with each point as the center, eigenvalues are decomposed, an edge probability graph is generated, corresponding component categories are semantically output, and corresponding key feature point sets are screened; taking the feature point cloud as a control point, constructing a three-level B-spline surface model, and adjusting the corresponding spatial distribution density; a weighted complete graph is formed through a key feature point set, a non-planar area is identified and processed through a Kurtowski theorem, a QEM algorithm is applied to carry out lightweight processing on a three-dimensional model, screening and optimization are carried out through automatically extracting key feature points of building components, and a concise and accurate lightweight model is constructed based on feature point topological optimization and a graph theory algorithm. The automation degree and efficiency of modeling are improved, and the contradiction between model lightweight and precision is effectively solved.

The invention relates to the technical field of cross-modal data fusion analysis of soil component detection and improvement agent preparation, in particular to a preparation method of a soil improvement agent based on component detection, which comprises the following steps: acquiring soil component data through a near infrared spectrometer, a high performance liquid chromatograph and an inductively coupled plasma mass spectrometer; an incremental principal component analysis algorithm is combined with an oscillation suppression function to update a covariance matrix in real time, a dynamic defect factor priority list is generated, a multi-objective optimization model improved based on NSGA-II is guided to integrate the soil pH value, humidity and organic matter content to construct a dose response curved surface, and a Pareto optimal solution set is solved. A sensor array collects environment feedback data, drives a transfer learning algorithm to calibrate parameters, a dynamic incidence matrix attenuation rate and optimal weight adaptive adjustment, iteratively outputs a modifier synergistic effect solution set through a closed-loop control mechanism, and solves the problems of difficult data fusion, principal component weight offset and component synergistic deviation. And the proportioning accuracy and stability are improved.

The invention provides a breakwater monitoring data preprocessing method and system based on Kalman filtering, and relates to the technical field of breakwater structure safety monitoring. The method comprises the following steps: acquiring original motion data of acceleration, inclination and displacement through a motion attitude sensor to obtain an original data sequence; initializing a state vector and an error covariance matrix; dynamically correcting the state transition matrix and calculating a prediction state vector and a prediction error covariance matrix; a Kalman gain is generated; updating a state vector and an error covariance matrix; and extracting the filtered motion data as a preprocessing result. According to the method, the state transition matrix is dynamically corrected by introducing the wave force feedback, so that the Kalman filtering algorithm can adapt to the wave impact environment, noise interference in monitoring data is effectively inhibited, and the accuracy and reliability of key motion parameter data of the breakwater are remarkably improved.

The invention discloses a vehicle state estimation method based on adaptive strong tracking extended Kalman filtering. The method comprises the following steps: constructing a nonlinear three-degree-of-freedom dynamic model containing a state equation and an observation equation based on an extended Kalman filtering algorithm to describe longitudinal, lateral and yaw states of a vehicle; state variables of the state equation are a side slip angle, a yaw velocity and a longitudinal vehicle speed; constructing a time-varying measurement noise statistical estimator based on a Sage-Husa algorithm to adaptively correct a measurement noise covariance matrix in the extended Kalman filtering algorithm; state prediction is carried out on the vehicle through the state equation, and a state prediction covariance matrix is calculated; and calculating the ratio of the sum of quadratic terms of the information sequence to the trace of the variance matrix of the information sequence based on an extended Kalman filtering algorithm, and judging whether filtering is really divergent or not. According to the method, the state quantity which is difficult to measure in the vehicle driving process is estimated in real time by establishing an adaptive strong tracking extended Kalman filter, and the vehicle state is accurately estimated.

The invention discloses a PPP-RTK time delay processing method and device in conversion from SSR to OSR, and relates to the technical field of global navigation satellite system positioning, and the method comprises the steps: converting orbit, clock error, ionosphere and troposphere delay SSR parameters provided by a server into OSR virtual observation values through geometric distance calculation and projection function mapping; based on a constant velocity model and a static random process, predicting a correction amount of a future moment, and correcting an OSR virtual observation value through state transition matrix compensation time delay; combining server side parameters and user side observation data, recursively generating an error covariance matrix, dynamically adjusting a state prediction weight, and obtaining covariance information through adaptive filtering optimization positioning calculation; and realizing ambiguity rapid fixation and centimeter-level positioning by using the corrected OSR observation value and covariance information. According to the method, real-time centimeter-level positioning in a high-delay scene is realized through a dynamic correction value prediction method and a recursive covariance estimation method.

The invention discloses a sparse dynamic graph convolution-based electroencephalogram signal decoding method and system. The method comprises the following steps of: acquiring a multi-channel electroencephalogram signal and preprocessing the multi-channel electroencephalogram signal; performing multi-band filtering on each channel signal, extracting statistical characteristics on each band signal, calculating a covariance matrix of a task electroencephalogram signal, constructing image electroencephalogram data by taking an electroencephalogram channel as an image node, the multi-band spliced statistical characteristics on the channel as a node feature vector, and the covariance matrix between the channels as an adjacent matrix; finally, a dynamic graph convolutional neural network model is constructed, the model constructs a graph convolutional neural network based on an autoregression moving average filter, graph electroencephalogram data is used as input, the category of electroencephalogram signals is used as output, an adjacency matrix is dynamically generated in combination with bilinear mapping, fuzzy label learning and sparse constraint are added to improve the decoding capacity of the model, and the dynamic graph convolutional neural network model is obtained. And the frequency domain response capability and robustness of the model to the graph structure are enhanced.

A GNSS positioning slowly-varying fault detection method based on residual-SVR regression comprises the steps that an observation information sequence is acquired based on a Kalman filter, and a covariance matrix of the observation information sequence is calculated; accumulating multi-step information through a sliding window, and constructing chi-square statistics; based on the fault-free data, constructing a training set by taking an innovation sequence as input and chi-square statistics as output, and generating an innovation-statistics mapping function; and fitting a normal slope threshold value based on an SVR predicted value, carrying out least square fitting on an observation statistic curve by sliding a window in real time, and judging whether to start a slow change fault alarm or not. According to the method, the residual error sequence is directly used as model input, and the dynamic chi-square statistical magnitude is used for replacing a traditional dichotomy label, so that the detection delay is reduced; an SVR detection model based on grid search and cross validation collaborative optimization is utilized, and an optimal parameter combination of a minimum mean square error (MSE) is screened through logarithm uniform sampling, interval linear sampling and five-fold cross validation, so that the average absolute error of slowly varying fault detection is reduced.

The invention provides a dual-antenna attitude and orientation and robust self-adaption method based on integrated navigation, which comprises the following steps: combining an INS (inertial navigation system) and a dual-antenna GNSS (global navigation satellite system), and enabling the system to output continuous and stable high-precision carrier attitude information when a GNSS signal is interfered by using dual-antenna integrated navigation. According to the method, most of satellite end clock correction, ionosphere and troposphere errors and receiver end clock correction are eliminated by using double-antenna GNSS pseudo-range and carrier phase double difference, robust statistics are calculated through heading prediction residual vectors to obtain robust factors, an observation noise covariance matrix is expanded to reduce the influence of system noise and observation noise, and the system performance is improved. A Mahalanobis distance based on a prediction residual vector is introduced to detect whether a system is abnormal or not, a state prediction covariance matrix is adjusted through a self-adaptive factor, weight reduction of an abnormal INS dynamic model is achieved, and the stability and reliability of system attitude information are improved through a noise covariance self-adaptive control mechanism.

The invention relates to the technical field of automobile part processing, and discloses an automobile part process parameter real-time optimization method and system based on machine learning, and the method comprises the steps: collecting the temperature gradient, pressure distribution, cutting speed and other multi-source process parameter data, extracting a process feature sequence through a self-attention mechanism and a time convolution network, and carrying out the real-time optimization of the process feature sequence; calculating a process fluctuation coefficient; and generating a process correlation weight through covariance matrix characteristic decomposition and Sigmoid function transformation, dynamically adjusting a reference parameter to generate an optimized parameter, and regulating and controlling equipment operation. The system comprises a multi-source data acquisition module, a process feature extraction module, a fluctuation coefficient calculation module and the like. And a multi-dimensional process parameter space is also constructed to monitor an abnormal state, and a joint optimization model is established to realize collaborative optimization of process and equipment control parameters. The real-time performance and accuracy of technological parameter optimization are improved, machining error accumulation is effectively restrained, the method is suitable for intelligent machining of automobile parts, and the machining quality and efficiency are guaranteed.

A system for extracting glacier boundaries using multiparametric analysis, consisting of: a data acquisition module configured to acquire Synthetic Aperture Radar (SAR) data of a glacial region; a data processing unit configured to process the SAR data to generate coherence images and backscatter intensity maps, to extract terrain parameters including slope and curvature information from a digital elevation model (DEM), to apply thresholds to the coherence images, backscatter intensity maps, and terrain parameters, to generate a multi-band stack comprising the thresholded coherence images, the thresholded backscatter intensity maps, the thresholded slope information, and the curvature information, and a principal component analysis module configured to standardize the multi-band stack, calculate a covariance matrix from the standardized data, calculate eigenvalues and eigenvectors of the covariance matrix, and transform the original data into a principal component space with reduced dimensionality; a texture feature extraction module configured to calculate first- and second-order statistics from the output of the principal component analysis; generating a texture feature set comprising at least one of the following: sum average, entropy, difference entropy, sum entropy, variance, difference variance, inverse difference moment, contrast, correlation, information measures of correlation, and maximum correlation coefficient; a connected component segmentation module configured to: convert the set of textural features into a binary format that distinguishes glacier ice from the background; group spatially connected pixels of similar intensity into segments; extract a vector shape file corresponding to the glacier boundary; an output generation module configured to generate a glacier boundary delineation output; and a user interface having a display configured to display the output generated by the output generation module.

The invention discloses a heterogeneous sensor fusion method and system for dynamic covariance optimization, and the method comprises the steps: constructing a multi-source heterogeneous sensor network which comprises an underwater inertial navigation unit, a Doppler velocity sensor, an ultra-short baseline positioning system, a submarine topography matching device and a geomagnetic gradient detector; the method comprises the following steps: designing a distributed integrated navigation system by adopting a hierarchical fusion architecture, establishing a sensor confidence coefficient dynamic evaluation model based on online estimated error covariance matrix eigenvalue analysis, carrying out state prediction on a local filter level by applying Kalman filtering, and then carrying out measurement updating by adopting a novel variational Bayesian algorithm; a sensor autonomous switching and adaptive weighted dual-mode adjustment mechanism is developed on the main filter level, and self-healing improvement of navigation precision in a non-stationary environment is realized, so that the problems of dynamic reliability evaluation and fusion efficiency optimization in underwater multi-sensor multi-source navigation are solved.

The invention discloses a marine environment forecasting method based on combination of machine learning and numerical forecasting, and particularly relates to the field of marine environment forecasting, which comprises the following steps: extracting marginal region features based on real-time observation data, constructing a forecasting field through a double-loss function, extracting extreme event features through space-time decoupling, and constructing a forecasting field through a double-loss function; and outputting a forecasting result of the target sea area. According to the marine environment forecasting method based on the combination of machine learning and numerical forecasting, the problem of initial field optimization failure caused by covariance matrix estimation deviation when marginal sea area data are sparse is relieved by embedding vorticity conservation equation residual calculation in a hidden layer; the dependence on the number of extreme event samples is reduced by realizing the targeted extraction of the extreme ocean event features; through a physical hard constraint layer technology of a residual error correction network, an initial field error of a data sparse region is corrected in combination with sea surface height and a flow velocity field constraint residual error feature map, and the problem of a large prediction error of an edge sea area is relieved.

The invention discloses a CAN bus intrusion detection method based on adaptive unscented Kalman filtering, and the method comprises the steps: obtaining real-time message data, and carrying out the preprocessing of the real-time message data, and obtaining a time sequence feature vector; constructing a nonlinear state space model based on the time sequence feature vector; performing unscented Kalman filtering state prediction based on the nonlinear state space model; inputting the time sequence feature vector as an actual observation value, calculating a Kalman gain to correct an unscented Kalman filtering state prediction result, and outputting a state estimation residual error; dynamically updating a process noise covariance matrix through exponentially weighted moving average based on the state estimation residual, and adjusting a measurement noise covariance matrix according to the measurement innovation sequence; calculating the mahalanobis distance of the state estimation residual error, comparing the mahalanobis distance with a self-adaptive anomaly detection threshold value, and judging whether an intrusion behavior occurs or not; and if the abnormal score exceeds a threshold value, triggering a multi-level alarm mechanism, recording a suspicious message and executing a safety protection operation.

The invention discloses a knowledge cross validation question-answering method and system for reducing hallusion of a large language model, and belongs to the technical field of artificial intelligence, and the method is implemented by the following steps: generating results through multiple times of sampling: when a user puts forward a question, controlling the large model to perform multiple times of sampling, and generating a specified number of results; calculating hidden state related indexes: extracting the hidden state of the last token of the middle layer of the large model corresponding to the result, and calculating covariance matrixes and answer discrete feature values of the hidden states; mLP model prediction: inputting the discrete feature value of the answer and the length of the answer into a multilayer perceptron MLP, and outputting a hallucination-free probability; querying and summarizing a knowledge graph; and calculating a final illusion-free score and outputting a result. According to the method, the answer quality and credibility of a large language model can be remarkably improved, and the method is particularly suitable for application scenes with extremely high requirements on the accuracy of single-mode text generation contents.

The invention provides a polarization / inertial navigation integrated navigation method based on a deep Kalman filter network, and belongs to the field of bionic polarization navigation. Establishing a priori prediction error covariance matrix estimation network and a measurement residual covariance matrix estimation network in polarization / inertial navigation filtering, and training the priori prediction error covariance matrix estimation network and the measurement residual covariance matrix estimation network at the same time by adopting a two-stage training strategy; the state and measurement are predicted, the optimal gain in polarization / inertial navigation filtering is calculated, state prediction and innovation are fused according to the calculated optimal gain, and estimation of the state at the current moment is obtained. According to the invention, high-precision navigation of the polarization / inertial navigation integrated navigation system is realized, and the environment adaptability of the system is improved.

The invention provides a coherent signal arrival direction estimation method and device based on a deep convolutional network, and belongs to the field of array signal processing. The method comprises the following steps: receiving a to-be-detected signal containing a coherent signal by using a uniform linear array antenna to obtain an array receiving data matrix and extract a covariance matrix; forming an input feature vector by right upper triangular elements divided from a diagonal line in the covariance matrix, inputting the input feature vector into a covariance estimation model formed by a deep convolutional network, obtaining an estimation value of the right upper triangular elements under an ideal incoherent condition, and reconstructing the estimation value to obtain a covariance matrix estimation value; and performing characteristic decomposition on the covariance matrix estimation value, and generating a spatial spectrum by using a MUSIC algorithm to obtain an estimation result of the signal arrival direction. According to the method, the noise-containing mixed signal covariance matrix is mapped into the ideal incoherent noise-free signal covariance matrix through a physical constraint supervised learning framework, so that the estimation precision and robustness of the MUSIC algorithm in a coherent scene are improved.

The invention discloses a self-adaptive battery life calibration method for cloud-side cooperation, and belongs to the crossing field of an energy storage system and cloud-side cooperation calculation. According to the invention, a cloud-edge double-layer collaborative framework is provided; an edge end estimates the health state and the residual life of a battery in real time through a recursive least square extended Kalman filtering model; the error observer calculates a prediction error based on a sliding window, and a dynamic threshold triggers an uploading mechanism; the edge end adopts an auto-encoder to compress original time sequence features into abstract vectors, and the abstract vectors and error statistics are uploaded together; the cloud performs incremental learning by using a deep sequential network, and only finely adjusts tail level parameters of which the gradient sensitivity exceeds a threshold value to generate a correction value; and the correction value is compressed and issued to an edge end, local model parameters are updated through weighted fusion, and a covariance matrix is adjusted. The method realizes high-precision life prediction and dynamic calibration, remarkably reduces the communication load, and is suitable for electric vehicles, power grid energy storage and other scenes.

The invention provides a self-adaptive robust vehicle navigation method and device based on LSTM assistance, and the method comprises the steps: calculating a process noise scale factor through process noise covariance estimation, and dynamically adjusting a process noise covariance matrix Q in a Kalman filter; executing a prediction process of Kalman filtering to obtain a one-step prediction state value and a state prediction covariance matrix; calculating a robust factor and a mahalanobis distance adaptive factor based on the prediction residual vector, and performing expansion adjustment on the prior measurement noise covariance matrix R and the predicted state prediction covariance matrix P; executing an updating process of Kalman filtering to obtain a state optimal estimation and a covariance matrix at the current moment; and when the GNSS loses lock, pseudo GNSS extended Kalman filtering is carried out by using a pseudo GNSS measurement value generated by the LSTM neural network and combining a process noise covariance estimation method. The method can better adapt to the dynamic characteristics of the navigation system in different environments, and provides powerful support for the stability and reliability of the combined navigation system.

The invention relates to the technical field of unmanned aerial vehicle positioning, discloses an unmanned aerial vehicle accurate positioning method and device fusing multi-source positioning data, electronic equipment, a computer readable storage medium and a computer program product, and is used for solving the technical problem that accurate and stable positioning of an unmanned aerial vehicle cannot be realized in an indoor weak texture scene in the prior art. The method comprises the following steps: acquiring initial positioning data of an unmanned aerial vehicle in real time, the initial positioning data being at least from two information sources; constructing a positioning fusion algorithm based on an unscented Kalman filtering algorithm, evaluating the credibility of initial positioning data from each information source in real time, and calculating a positioning covariance matrix; and adjusting the fusion weight of each initial positioning data based on the positioning covariance matrix, and calling a positioning fusion algorithm to perform weighted fusion on the initial positioning data of each information source to obtain fusion positioning data of the unmanned aerial vehicle. According to the method, the positioning accuracy and stability of the unmanned aerial vehicle in a scene in which GPS signals are lacked and environment textures are weak can be improved.

The invention discloses an aircraft target tracking method and system based on compensation prediction, which are used for improving the target tracking precision in an image transmission delay scene. The method comprises the following steps: firstly, acquiring an image frame sequence of a target aircraft by using an airborne monocular camera, extracting a target center coordinate through a small target detection algorithm, and constructing a position sequence; the method comprises the following steps: extracting current high-frequency I MU data aiming at the condition that an image frame has transmission delay, inputting the current high-frequency I MU data into an LSTM-DKF model constructed by fusing LSTM and a delay Kalman filter, and predicting and generating a process noise and observation noise covariance matrix; and initializing a delay Kalman filter by using the matrix, and recursively predicting the target position during the delay period. And when the delayed image frame is received, backtracking and updating the state of the filter, recurring to the current moment again, and outputting the compensated target position. And finally, pixel deviation is calculated according to the compensation position, an aircraft tracking control instruction is generated, and high-precision target tracking is realized.

The invention discloses automatic complete equipment for intelligent predictive maintenance of a primary and secondary fusion circuit breaker. The automatic complete equipment comprises a multi-sensor fusion unit, an edge calculation and analysis module; a parameter interaction module; a predictive maintenance decision unit; the primary and secondary converged communication architecture is used for managing control information and state information on the basis of an IEC61850 (International Electrotechnical Commission 61850) standard; wherein a noise covariance matrix and a feature weight coefficient of the adaptive Kalman filtering health assessment algorithm are dynamically adjusted according to a data quality index and prediction error feedback, and input features of the residual life prediction algorithm based on the LSTM comprise a health index, a change rate and component-level health state information from the health assessment algorithm. Accurate evaluation of the health state of the circuit breaker and accurate prediction of the residual life are achieved, the optimal maintenance strategy is generated, the operation reliability of the circuit breaker is improved, and the maintenance cost is reduced.

The invention relates to the technical field of dam safety monitoring, in particular to a method and system for improving dam GNSS deformation monitoring precision through base station and observation station combined network adjustment. The method comprises the following steps: arranging base stations and monitoring points to form a GNSS monitoring network; gNSS original observation data are collected in real time and preprocessed; constructing a joint network adjustment model, and taking base station coordinates as constraints and monitoring point coordinates as to-be-estimated parameters; performing baseline resolving based on the double-difference carrier phase observed quantity to obtain a baseline vector and a covariance matrix thereof; carrying out overall adjustment on the baseline vector by adopting a robust estimation method, and solving an optimal coordinate estimated value and precision information of the monitoring point; carrying out deformation analysis on the basis of the adjusted coordinate time sequence, and extracting tendency, periodicity and abnormal deformation; and outputting a deformation monitoring result, and carrying out visual display and early warning. According to the method, the precision and reliability of dam deformation monitoring are effectively improved through combination of network adjustment and robust estimation.

The invention relates to the technical field of temperature control, in particular to a temperature sensor fusion compensation method and system based on Kalman filtering, and the method comprises the steps: collecting an original measurement signal, and carrying out the preprocessing; constructing a feature vector matrix and a diagonal feature value matrix of each sensor according to a preprocessing result; updating the state vector and the covariance matrix through a Kalman filtering algorithm to generate an optimal state estimation value; according to the residual distribution of each sensor and KL divergence evaluation, adjusting the weight of each sensor; and after the sensor weights of the normal sensor and the abnormal sensor are adjusted, weighted summation is carried out on the weight of each sensor and the optimal state estimation value, and a final temperature measurement value is generated. Through the Kalman filtering algorithm, the judgment of people on the influence degree of high-frequency noise and equipment errors on the sensor is improved; and the weight of each sensor is dynamically adjusted through residual calculation and KL divergence evaluation, so that the robustness and compensation precision of the system are remarkably improved.

The invention discloses a multi-modal heterogeneous data real-time fusion and intelligent decision-making method and system based on cloud computing, and is suitable for industrial internet scenes. According to the method, multi-modal data such as physical sensors, audio and video, physiological signals and the like are collected, unified preprocessing and feature extraction are carried out, and a covariance matrix is constructed to represent a coupling relation between modals; further constructing the multi-period state into a graph structure, and realizing high-robustness state modeling by using a graph neural network and self-supervised learning; and identifying the operation state through a Riemannian geometric classification method, and carrying out risk scoring and grade judgment in combination with an emotional state and an environment index. The system supports AR visual prompt and control linkage, and the man-machine cooperation intelligent decision-making ability in an industrial scene is improved.

The invention provides a slope displacement monitoring method and system based on reinforcement learning and enhanced Kalman filtering, and the method comprises the steps: carrying out the preprocessing of displacement data collected by Beidou, and carrying out the abnormal value elimination, missing value interpolation and time consistency inspection; establishing a Kalman filtering model containing displacement and speed state vectors, and initializing a process noise covariance matrix Q and an observation noise covariance matrix R as initial filtering parameters; q and R matrixes are dynamically optimized through a PPO reinforcement learning algorithm, and parameter self-adaptive adjustment is achieved; carrying out displacement trend analysis on the filtered output data, marking abnormal trend data by adopting a statistical test and trend inflection point recognition algorithm, and feeding back a root-mean-square error of the abnormal trend data to a PPO algorithm to carry out parameter readjustment; data stage changes are analyzed based on a sliding window technology, independent experience playback buffer areas are set for data in different stages in PPO, and associated updating of filtering parameters is achieved. According to the invention, the precision and reliability of slope displacement monitoring are improved.

The invention relates to the technical field of test bed operation monitoring, in particular to a fuel pump test bed operation monitoring method and system. The method comprises the following steps: acquiring monitoring time sequence data of operation of a fuel pump test bed, performing feature extraction, constructing a coupling prediction model of a space-time diagram attention network-unscented Kalman filter, and predicting a process noise covariance matrix and a measurement noise covariance matrix according to the coupling prediction model, and performing state estimation by using an unscented Kalman filter to obtain vector posterior probability distribution, constructing a fault evolution trajectory manifold, calculating a mahalanobis distance between the vector posterior probability distribution and the fault evolution trajectory manifold, and taking the mahalanobis distance as a monitoring index. According to the scheme of the invention, the deep features which can better reflect the inherent nonlinear and complex dynamic characteristics of the system can be extracted from the multi-source data, the estimation accuracy of the filter on the potential health state of the system is improved, and the defects that model parameters are fixed and gradual change faults are difficult to capture in a traditional method are overcome.

A DoA estimation method and device based on steering vector matrix reconstruction, related to the field of array signal processing. The method includes: obtaining an array sampling covariance matrix according to an array received signal; setting a target variable, and limiting a feasible domain of the target variable by using two operators to determine a first constraint condition; characterizing an estimation error based on the target variable and the array sampling covariance matrix, and using the characterized estimation error as a second constraint condition; establishing an initial optimization model according to a preset norm based on partial sum of singular values and constraint conditions of the target variable; determining a multivariable optimization model according to the initial optimization model; and solving the multivariable optimization model to obtain an optimal result; analyzing the optimal result to obtain a DoA of the target incident signal.