65 results about "Change prediction" patented technology

The invention relates to the cross technical field of artificial intelligence and environment monitoring, and discloses an algal community structure change prediction algorithm and system based on multi-source data fusion, and the algorithm comprises the steps: obtaining water quality, weather and plankton multi-source time sequence data; performing time alignment and missing value interpolation; eliminating and screening key environment factors through recursive features; performing dynamic weighted fusion on the multi-modal features by using a space-time attention mechanism; inputting a three-layer stacked LSTM network to output future algae dominant species abundance prediction; and model parameters are corrected on line based on measured data. The system comprises a multi-source data acquisition module, a preprocessing module, a key factor extraction module, a space-time attention fusion module, a dynamic prediction module and an adaptive correction module. According to the method, the prediction accuracy and stability are remarkably improved, and algal bloom early warning and ecological regulation are effectively supported.

The invention provides a drought and flood sudden change prediction method and system based on artificial intelligence, and the method comprises the steps: collecting the in-out reservoir runoff observation data of a reservoir in a target region, obtaining a restored natural reservoir runoff series based on a water balance method, calibrating a long-short-term memory model through a minimum batch gradient descent method, and carrying out the prediction of the drought and flood sudden change. Based on the natural reservoir runoff, the actual reservoir runoff and meteorological data, constructing a long-short-term memory model to simulate the influence of water conservancy project regulation and storage on the runoff; based on an earth system mode set and a multivariable deviation correction method, performing spatial downscaling on output data of earth system modes to obtain corrected meteorological variables; driving a multi-member set of an earth system mode, and separating by adopting a detection attribution technology to obtain a contribution rate of man-made climate compulsion to drought and flood sudden turning event evolution; and predicting a daily runoff process under future climate change, and predicting a future drought and flood sudden turning event by adopting deep learning and superposition of regulation and storage influence of a water conservancy project.

The invention relates to a protein conformation change prediction method based on a deep generative model and reinforcement learning. The method comprises the following steps: acquiring a plurality of sample data to form a sample data set; masking the (K + 1)-th frame to the L-th frame of molecular dynamics simulation trajectory data of each piece of sample data in the sample data set to obtain a masked sample data set; inputting the sample data set, the masked sample data set and the first condition guide information into a to-be-trained protein conformation prediction model for training, and performing fine tuning on network parameters of the trained protein conformation prediction model by adopting a reinforcement learning fine tuning algorithm optimized by a protein conformation dynamics strategy, and obtaining a trained protein conformation prediction model. And obtaining L-frame random noise, first condition guidance information and K-frame protein initial conformations, and inputting the L-frame random noise, the first condition guidance information and the K-frame protein initial conformations into the trained protein conformation prediction model for prediction to obtain L-frame protein conformation change trajectories. Therefore, the protein conformation change process can be dynamically analyzed.

The invention provides a nuclear power plant thermal discharge dynamic change prediction analysis method, and relates to the technical field of data processing, and the method comprises the steps: collecting a temperature observation data set, extracting a temperature maximum value sequence and a temperature minimum value sequence in an equal-tide period, calculating a wave amplitude stable factor, extracting a plurality of trend inflection points of a temperature change first-order derivative curve, and calculating the temperature change first-order derivative curve. The method comprises the following steps: determining thermal expansion boundary candidate areas caused by thermal drainage, evaluating a time interval and amplitude difference between a lag response inflection point in each thermal expansion boundary candidate area and a previous dominant temperature rise point to obtain a thermal lag offset value, extracting a boundary sensitive area, performing multipoint communication fitting, determining a maximum space extrapolation boundary influenced by the thermal drainage effect, and performing thermal expansion on the maximum space extrapolation boundary. And carrying out trajectory tracking and change direction regression on the extrapolation boundaries at multiple different time points, and identifying a heat disturbance movement trend to obtain dynamic evolution data. The method can identify the diffusion trend of the warm discharged water and predict the temperature change.

The invention discloses an underground water level change prediction method and system based on machine learning, and relates to the technical field of hydrogeology, and the method comprises the steps: carrying out the standardization processing of multi-source time series data, obtaining a standardized multivariable time series data matrix, and constructing a supervised learning sample set; inputting the underground water level sequence in the supervised learning sample set into a physical information guided variational mode decomposition network; decomposing the underground water level sequence into K intrinsic mode component sequences and residual term sequences through a loss function of physical driving consistency constraint; and for each of the K intrinsic mode component sequences, dynamically assembling a differentiable simulator from the differentiable simplified physical simulator basic library, and carrying out cooperative training by taking approximation to each intrinsic mode component sequence and overall reconstruction of the original water level sequence as targets to obtain K completely trained assembled differentiable simulators. According to the method, a reliable visual prediction result is generated through multi-simulator collaborative deduction and uncertainty quantization.

The invention relates to a capacity change prediction method and device of a lithium iron phosphate battery, equipment, a storage medium and a program product, and relates to the technical field of energy storage battery management. The practicability and flexibility of the scheme can be improved. The method comprises the following steps: acquiring a voltage change curve of a to-be-detected lithium iron phosphate battery in a charging and discharging process, and obtaining a capacity increment curve corresponding to the to-be-detected lithium iron phosphate battery according to the voltage change curve; acquiring a preset number of voltage fragments in the capacity increment curve, determining the maximum voltage drop in the voltage change curve according to the voltage fragments, and constructing a target relation curve between the maximum voltage drop and the capacity change of the lithium iron phosphate battery to be detected; predicting the future change trend of the maximum voltage drop according to the target relation curve by adopting a target numerical method in combination with Gaussian process regression to obtain a target prediction curve of the maximum voltage drop; and obtaining a capacity change prediction curve of the lithium iron phosphate battery to be detected according to the target relation curve and the target prediction curve.

The invention relates to the technical field of remote sensing image processing and mapping knowledge domain, in particular to an earth surface change prediction method for guiding mapping knowledge domain reasoning based on space-time rules and application of the earth surface change prediction method. The change information of the earth surface in time and space is obtained in a mode of mining knowledge graph rules, the characteristics of each earth surface change can be utilized to the greatest extent, the complexity of the earth surface change can be comprehensively considered through sampling entity pair calculation time sequence constraint and space constraint, combined loss optimization embedding vector and reasoning of future earth surface change, and the method has the advantages of being high in practicability and the like. The space-time rule is considered in the reasoning research of the earth surface change knowledge graph, and the own characteristics of the triple in the knowledge graph and the space-time rule constraint are effectively utilized. The objective of the invention is to solve the problem of how to accurately predict future earth surface change by utilizing knowledge graph reasoning.

The present invention implements a tool for determining a situation and predicting changes on the basis of a change principle in binary format by considering time sequence, thereby helping people to accurately determine which situation occurs for themselves and in the vicinity thereof and predict how the situation will change in the future, and provides a service in cooperation with an online server, an artificial intelligence model and a database such that the service is provided from a service unit on the basis of a hexagram change arrangement table in the tool for situation determination and change prediction.

The invention provides a remote sensing change detection method and system based on deep reinforcement learning, and the method comprises the steps: carrying out the feature extraction and difference modeling of an input multi-temporal remote sensing image through a deep learning network, so as to generate a preliminary change prediction map; and introducing a reinforcement learning module, taking the preliminary change prediction map as an initial state, and performing multi-round iterative optimization on the preliminary change prediction map through a decision process comprising a state space, an action space and a reward function so as to generate a final change detection result. And the reward function comprehensively considers an intersection-to-union ratio improvement value, an F1 score improvement value and cross entropy loss, and introduces a misclassification penalty enhancement mechanism to preferentially repair missing report and false alarm areas. According to the method, through iterative optimization of reinforcement learning, errors in a preliminary detection result can be effectively corrected, and the detection precision and robustness are remarkably improved.

The invention provides a karst carbon sink change prediction method and system based on machine learning, and the method comprises the steps: obtaining environment base data and carbon sink actual measurement data of a karst region, and constructing a karst carbon sink dynamic coupling body comprising a real-time association link and other elements; and starting a pre-trained carbon convergence prediction model to perform bidirectional interaction modeling with the coupling body, generating a feature iteration instruction through the carbon convergence prediction model to drive the coupling body to update features, and feeding back a feature change signal to correct model parameters by the coupling body to obtain a multi-round interaction modeling intermediate result. And performing operations such as association link enhancement on the coupling body based on the intermediate result to obtain an evolved coupling body. And performing cross-dimension feature decoding and correlation trend deduction on the evolved coupling body by using the carbon convergence prediction model to generate a karst carbon sink change prediction result. According to the method, the relation between environmental factors and carbon sink is considered, and the prediction accuracy and adaptability are improved through dynamic interaction and optimization.

This invention relates to the field of foundation pit support design technology, and provides a foundation pit support design method and system based on digital twins. The method includes: establishing a digital twin related to the foundation pit and support; constructing a corresponding change curve from the foundation pit excavation data; substituting the curve into the digital twin to predict the mechanical parameters of the foundation pit sidewall at the unexcavated depth, thereby obtaining second soil layer change prediction data; judging the future support requirements changes at various points in the foundation pit based on the second soil layer change prediction data and real-time foundation pit support data, and adaptively reinforcing the foundation pit support using coordinates, thereby reducing foundation pit support costs and improving the efficiency of excavation-as-you-go support.

The application relates to the technical field of building management, in particular to a building abnormal change identification method and device based on an intelligent algorithm, which comprises a building abnormal change identification center, an abnormal change analysis unit, an abnormal change division unit, a cause analysis unit, an abnormal change prediction unit and a visual response unit; the application preliminarily analyzes point cloud data at different time points to determine whether abnormal changes occur in a building, meanwhile, the abnormal building is distinguished in a color marking mode, so that the building which has abnormal changes in a target area can be directly and intuitively understood, the abnormal building is analyzed in an information progressive mode, so that it can be directly and intuitively understood whether the abnormal change of the abnormal building is caused by a sudden event or a slow long-term change accumulation, and the sudden abnormal change and the long-term accumulated abnormal change are analyzed, so that the management efficiency of the sudden abnormal change of the target building is improved, and the structure state in the target building is understood based on the strain change trend, so that advanced early warning management is facilitated.

The invention relates to a method for simulating and predicting the dynamic change of oyster reefs in a marine ranch. The method comprises the following steps: step a, drawing a marine ranch seabed acoustic image; b, identifying spatial distribution of oyster reefs in the marine ranch, and converting the spatial distribution into an element map layer; c, selecting oyster biological growth and development influence factors, and converting the factors into a grid map layer; d, all the grid layers and the element layers are imported into a Maxent model, and model fitting parameters are set and run; e, determining a plurality of influence factors with high contribution degrees according to a model operation result; step f, determining an oyster growth suitable area according to a model operation result; and step g, adjusting influence factors with high contribution degrees, and forming a dynamic change prediction result of the oyster reefs in the marine ranch. According to the method, the difficulty of uncertainty of influence factors of the oyster reef can be overcome, the problem of simulation and prediction of development of the oyster reef in a strong-dynamic and high-complexity marine environment is solved, and technical support is provided for research on marine benthic ecology, oyster reef restoration and regeneration and the like.

This application discloses a land use change prediction method, apparatus, device, and medium based on map representation generation learning, relating to the fields of spatiotemporal big data mining and urban planning technology. The method includes: constructing multi-source map submaps and performing binarization and attention weighting to obtain binarized land use submaps and weighted geographic environment submap features; extracting deep aggregated spatial features from both to generate latent vector representations; flattening the representation into a one-dimensional feature sequence and querying the index of each element in a landscape codebook to form a vectorized index sequence; combining relative distance encoding, using a map sequence generation learning model to predict future vectorized index sequences, and querying the landscape codebook to obtain future vectorized tensors; iteratively decoding the tensor using a map decoder to generate land use change results. This application achieves more accurate land use change prediction by capturing mesoscale spatial local information and reducing unit-level error accumulation.

The application provides a network topology change prediction method and device combined with big data analysis, relates to the technical field of network topology prediction, and comprises the following steps: pre-building a coupled network topology; deploying a plurality of edge fault prediction nodes locally in a plurality of physical devices in a physical topology layer; when a first-hop prediction fault is output, performing topology cascade propagation analysis to locate X associated fault devices; triggering a cooperative prediction mechanism, performing pre-fault prediction update, and outputting X pre-fault prediction faults; performing accompanying entity mapping and outputting a global logical influence domain; locating network topology change distribution; performing service dependence chain stall mapping and outputting service interruption risk. The application solves the technical problem that the network topology change prediction in the prior art is mainly based on static topology structure, the dynamic coupling between the physical topology layer and the logical topology layer is ignored, the propagation and influence of faults between the two layers cannot be accurately evaluated, and the prediction accuracy of network faults is affected.

The present application relates to a change detection method for river and lake management violation based on feature enhancement and refinement, belonging to the technical field of deep learning and remote sensing image change detection. The remote sensing images of two periods are respectively sent into an inflation residual network to extract features, obtaining a double-time feature map; the double-time feature map is sent into the same cross-time feature interaction module in pairs to obtain cross-time features; the obtained cross-time features are sent into a multi-scale feature detail completion module to obtain a preliminary prediction result; then sent into a prediction enhancement module for refinement and enhancement to obtain the final prediction result, and the best model is iteratively trained and saved; the remote sensing images of different periods in the test set are sent into the trained model in pairs to obtain a change prediction map. The present application maximally retains the features of small-size ground objects in the river and lake remote sensing images, and enhances and refines them, ensuring the accuracy of the change detection of river and lake management violations, and being able to solve the problem of missing small-size violations in the river and lake shoreline remote sensing images.

The invention discloses a software defect priority change prediction method and device based on dynamic evolution characteristics, and belongs to the technical field of crossing of software engineering, data mining and artificial intelligence, and the method comprises the following steps: obtaining a defect report and defect historical data of a first defect; if the defect report of the first defect changes at the moment t, based on the defect report and defect historical data, a defect feature vector of the first defect at the moment t is obtained, and the defect feature vector comprises a defect dynamic evolution feature; and predicting a second priority after the first defect is changed by adopting the first defect change prediction model based on the defect feature vector feature of the first defect at the t moment. Wherein the defect dynamic evolution characteristics comprise project evolution characteristics, reporter historical behavior characteristics, comment interaction dynamic characteristics and historical change dynamic characteristics. According to the method, dynamic defect priority change prediction can be accurately realized.

The purpose of the present invention is to implement a tool for situational determination and change prediction on the basis of a binary format change principle that takes time into consideration, to: help people accurately determine situations affecting themselves and their surroundings and predict in advance how the situations will change in the future; and provide a service in cooperation with an online server, artificial intelligence, and a database, i.e., enabling the people to be provided with a service from a service unit on the basis of a situation category arrangement table in the tool for situational determination and change prediction.

This invention discloses a machine learning-based method and system for predicting groundwater level changes, relating to the field of hydrogeology. The method includes: standardizing multi-source time-series data to obtain a standardized multivariate time-series data matrix; constructing a supervised learning sample set; inputting the groundwater level sequence from the supervised learning sample set into a physically-guided variational mode decomposition network; decomposing the groundwater level sequence into K intrinsic mode component sequences and a residual term sequence using a loss function with physical-driven consistency constraints; for each of the K intrinsic mode component sequences, dynamically assembling a differentiable simulator from a library of differentiable simplified physical simulators, and co-training these simulators with the goal of approximating each intrinsic mode component sequence and reconstructing the original water level sequence as a whole, resulting in K fully trained assembled differentiable simulators. This invention generates reliable, visualized prediction results through multi-simulator collaborative extrapolation and uncertainty quantification.

The embodiment of the present application provides a shield tail deformation prediction method, device and equipment and a readable storage medium, wherein the shield tail deformation prediction method comprises the following steps: acquiring a distance change actual value of a detection mark point on an inner wall of a shield tail of a shield tunneling machine and at least one deformation amount of the detection mark point in a deformation state; establishing a change mapping model based on the detection mark point by a radial basis function neural network method according to the distance change actual value and the at least one deformation amount; and calculating a distance change prediction value of the detection mark point corresponding to the at least one deformation amount by the change mapping model. The present application also provides a shield tail deformation prediction device and equipment and a readable storage medium. Through the embodiment, the actual deformation amount can be monitored in real time and accurately, and the occurrence of situations such as misalignment of segment installation caused by collapse or deformation intrusion of the inner wall of the tunnel can be avoided.

The invention relates to the technical field of data analysis, in particular to a microstructure change tracking method based on chip aging analysis, and the method specifically comprises the steps: obtaining chip factory structure data, importing the data into a chip change prediction model, obtaining an initial aggregation stability index, carrying out the evaluation of the aging condition of a chip, and entering a chip maintenance and renovation process if the evaluation result is a to-be-renovated grade; if the evaluation result is a to-be-recycled grade, constructing a physical field coupling migration model, and analyzing and outputting the initial instantaneous migration speed of noble metal atoms on the chip; interference analysis is conducted on adjacent precious metal atoms, correction compensation is conducted on the instantaneous migration speed, the aggregation density condition of the precious metal atoms at all recovery evaluation point positions on the chip is obtained through tracking, and high potential points of the chip are screened out. According to the invention, the problem of low identification efficiency of the high-potential recovery point on the aged chip in the prior art is solved.

To provide a method capable of predicting a change of an object when an action is applied to the object in a relatively short time and with high accuracy.SOLUTION: A prediction method of the present invention is a method of predicting a change of an object when an action is applied to the object, and includes predicting a change of a first object in a region to which the action of the object is applied and a vicinity thereof by a graph neural network, and predicting a change of a second object in at least a part of a region other than the region to which the action of the object is applied and the vicinity thereof by analysis using a finite element method.SELECTED DRAWING: Figure 1

The application is suitable for the field of vehicle lane change prediction technology, and provides a vehicle lane change prediction method, device and electronic equipment, which comprises the following steps: acquiring a motion vector of a target vehicle, the motion vector comprising a tangential velocity and / or a displacement; calculating a vector entropy of the motion vector, the vector entropy being used to describe a stability degree of a motion trend of the target vehicle; acquiring an RSSI of the target vehicle; inputting the motion vector, the vector entropy and the RSSI into a trained decision tree model to obtain a prediction result output by the trained decision tree model, the prediction result being used to indicate whether the target vehicle changes lanes. Through the above method, the accuracy of the prediction result can be improved.

The invention provides a monitoring-free emotion change prediction method and device, and relates to the technical field of psychological model calculation and behavior prediction. The method comprises the following steps: acquiring activity characteristics and executor parameters of the activity, and performing interactive calculation according to the activity characteristics and the executor parameters of the activity to obtain an intermediate variable value; performing correction calculation on the emotion change initial value of the executor according to the intermediate variable value to obtain a first emotion change correction value; performing correction calculation on the first emotion change correction value according to the executor parameters of the activity to obtain a second emotion change correction value; performing correction calculation on the second emotion change correction value according to the activity characteristics and the executor parameters of the activity to obtain a third emotion change correction value; and generating an emotion change prediction result according to the emotion change third correction value. The device executes the method. The method and the device provided by the embodiment of the invention do not need to depend on a sensor and are low in application cost.

The invention discloses a software research and development management method and system based on artificial intelligence, and relates to the technical field of software management, and the method comprises the steps: collecting research and development process data, and carrying out the data preprocessing; according to a preset demand change prediction model and the demand document, predicting the working hours of the related tasks and the influence degree of the task progress, and generating a change influence evaluation report; generating a resource allocation report according to the multi-dimensional data of the human resource data and a preset resource allocation strategy; performing agent quality and defect prediction according to the code data and a preset code prediction model, and generating a code optimization report; predicting a project delay rate according to the task, the task progress data, the multi-dimensional influence factors and a preset prediction delay risk model, and generating a project progress early warning report; and generating a corresponding dimension operation suggestion scheme according to the report, and performing updating processing. By adopting the method and the system, intelligent management of the software research and development process can be realized, so that the overall research and development efficiency and the software product quality are improved.

This invention discloses a remote sensing change detection method based on edge-enhanced cross-attention and multidimensional loss. The method includes: inputting dual-temporal images into a trained Siamese backbone network to extract multi-scale backbone features from the dual-temporal images; inputting the multi-scale backbone features into a trained unsharpened mask cross-attention fusion model to extract spatial and semantic information from the dual-temporal images through cross-attention operations, and extracting edge information from the dual-temporal images through sharpening convolution kernels to obtain multi-level differential features at different scales; inputting the differential features into a trained multi-scale hierarchical connection decoder to obtain preliminary multi-scale prediction results for the changed region; and performing weighted fusion of the preliminary multi-scale prediction results to obtain the change prediction result. This invention can fully utilize the multi-scale information of the Siamese backbone network and improve the ability to extract edge information from dual-temporal images, thereby improving the accuracy of change detection.

The invention discloses a mountain city land utilization change prediction method based on a gray model and deep learning, and the method comprises the following steps: S1, collecting data, and constructing a data set; s2, data preprocessing; s3, constructing a state transition matrix; s4, establishing a gray system model; s5, simulating and predicting the change trend of the land utilization type in combination with a Markov model; s6, training and evaluating the suitability probability of the land utilization type through a deep learning ANN module; s7, simulating the mutual relation among the land use types; s8, constructing a land utilization change prediction model according to the land utilization transfer change and the land utilization change driving factors; s9, performing multi-scene prediction on future land utilization change through the trained model; according to the method, efficient land utilization prediction is realized by integrating multi-source data, introducing a self-adaptive mechanism, combining a decision support tool and providing ecological environment influence assessment.

The invention relates to a KAN-based tropical cyclone strength change mechanism analysis method and device, and belongs to the technical field of TC strength change prediction. The method comprises the steps that a factor recursive pruning algorithm based on a KAN2.0 attribution score is adopted, and a high-impact forecasting factor set for TC intensity change forecasting is screened and obtained; performing factor contribution analysis based on symbolization regression by using KAN2.0 trained by a high-impact forecasting factor set, obtaining a complete linear equation from high-impact forecasting factors to TC intensity change, and calculating the TC intensity change according to the symbols and absolute values of various linear coefficients in the equation; and quantitatively analyzing the contribution direction and contribution degree of each high-influence forecasting factor to the TC intensity change. According to the method, the stability, the accuracy, the generalization ability and the interpretability of TC intensity forecasting can be effectively improved.

To provide a ripeness degree estimation method for estimating a ripeness degree of fruits more appropriately than conventional methods.SOLUTION: A fruit peel, a test specimen, is irradiated with light in visible and near-infrared regions ranging from 400 to 1700 nm. Then, reflection intensity corresponding to the wavelength of the light is measured, and the reflectance is calculated from the reflection intensity. The calculated reflectance value is second-order differentiated to calculate a second-order derivative value. A ripeness degree index of the fruit is then estimated from the calculated secondary differential value on the basis of a ripeness degree prediction model. The ripeness degree prediction model represents a relation between the second derivative value of the reflectance and the ripeness degree index obtained by a sensory test. The ripeness degree prediction model is generated by partial least squares (PLS) regression analysis, with the ripeness degree index defined as a dependent variable and the second derivative of the reflectance defined as an independent variable.SELECTED DRAWING: Figure 5

The invention relates to a data change prediction method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring source data of each source database contained in a measurement database; wherein the source data comprises at least two table files, and each table file comprises at least two fields; constructing a target association graph of the measurement database according to the source data; wherein the target association graph comprises nodes, connecting edges between the nodes and edge weights of the connecting edges; the nodes represent table files or fields, and connection edges between different nodes represent association relationships between different table files and / or fields; performing feature extraction on the target association graph and each piece of source data to obtain multi-dimensional features of the measurement database; and according to the multi-dimensional features and the target association diagram, predicting data change information of the measurement database in a future time period. By adopting the method, the data change information can be efficiently predicted in real time.