22 results about "Long-term prediction" patented technology

This invention belongs to the fields of new energy technology and artificial intelligence technology, and provides a high-resolution long-term photovoltaic power prediction method and system based on a dual-branch architecture. By constructing a parallel dual-branch architecture, the baseline branch extracts multi-scale features, and the ramp branch identifies power ramping events. The features from both branches are embedded through multiple channels and then fed into a BiGRU-Enhanced Transformer model with shared weights. By strategically deploying BiGRU layers at the input and between the encoder and decoder, local instantaneous dependencies and global long-term trends are captured simultaneously. A multi-task joint optimization loss function is designed for the ramp branch, using weighted cross-entropy and... L The probability, temporal location, and fluctuation amplitude of slope occurrence are jointly optimized using the L1 norm. Finally, nonlinear residual correction is applied to the outputs of the two branches to achieve the final prediction. This invention effectively overcomes the oversmoothing effect of deep learning models, demonstrating extremely high capture accuracy and robustness in long-term prediction of one month's worth of data at 1-minute resolution.

The present application relates to the technical field of marine data processing and spatio-temporal prediction, and specifically discloses a sea surface core variable prediction method and system based on a wavelet-enhanced graph neural network.The method comprises the following steps: obtaining multivariate graph sequence data of a target sea surface; performing multilevel wavelet decomposition and gated fusion on each variable through a variable-level multiscale wavelet gated fusion module, and outputting enhanced multiscale time series features; inputting the enhanced multiscale time series features into a KAN-LSTM encoder module, performing recursive updating through gated fusion of a conventional convolution and a KAN convolution branch, and outputting encoder spatio-temporal features; inputting the encoder spatio-temporal features into a signed adaptive spatial graph convolution module, learning a signed sparse adaptive adjacency matrix and performing multi-order diffusion aggregation, and outputting a prediction result.The present application realizes long-term prediction of a target sea surface with high precision and high stability.

The application relates to a nuclear power plant accident early warning method and system, and the method comprises the following steps: acquiring real-time operation data of a nuclear power plant; predicting operation data in a future first set time according to the real-time operation data to obtain short-term future data; performing accident diagnosis according to the short-term operation data to obtain an accident prediction type; generating operation data in a future second set time according to the accident prediction type and the real-time operation data to obtain long-term future data; and the second set time is greater than the first set time. The application can realize early prediction of an accident risk, guarantee the confidence of long-term prediction results, and enable staff to realize early intervention by using long-term future data.

This invention discloses a method for predicting the cooling load of public buildings based on spatiotemporal attention transfer learning, comprising the following steps: collecting multi-source heterogeneous data of the target building and constructing a multimodal spatiotemporal dataset; then filtering and classifying the multimodal spatiotemporal dataset to obtain core features of different building types; calculating the subjective and objective weights corresponding to each core feature, and combining the subjective weights to calculate the dynamic weight vector of each core feature; then weighting the dynamic weight vectors of each core feature, and outputting the temporal features affecting the cooling load value based on the temporal dependency of the cooling load; finally, optimizing the weights and enhancing the features of the temporal features, and predicting the enhanced temporal features through a fully connected output layer to output the predicted cooling load value. This invention can achieve universal adaptation for predicting the cooling load of buildings of various types and improve the prediction stability of building cooling load in long-term prediction.

This invention discloses a long-term prediction method for deep-sea creep based on in-situ pore pressure observation using Bayesian-LSTM, relating to the fields of sediment dynamics and marine engineering. The method includes: deploying pore pressure sensors to collect pore water pressure data in real time, performing standardization processing, and calculating the pore pressure change rate; constructing a creep rate calculation model based on the effective stress principle and power law; building a time-series feature dataset using sliding window technology, constructing and training a Bayesian-LSTM network model, and introducing a Monte Carlo Dropout layer to assess prediction uncertainty; outputting a high-precision creep rate using a multiple sampling averaging strategy; calculating the cumulative creep distance through numerical integration, and constructing confidence intervals to achieve medium- and long-term trend prediction and risk analysis. This invention integrates physical mechanisms and data-driven approaches, effectively handling data noise and prediction uncertainty in the complex environment of the deep sea, significantly improving the accuracy and reliability of long-term creep prediction, and providing a scientific basis for deep-sea engineering safety assessment.

The present application relates to the field of real-time monitoring and coordinated control of power grid, in particular to a kind of distributed photovoltaic grid-connected regional power grid real-time monitoring and coordinated control system, system includes sensing acquisition, multi-time scale prediction, self-evolution calibration and distributed coordination control module;Sensing acquisition module is collected wide frequency domain electrical quantity and multidimensional meteorological quantity by double-layer sensing network, and the repair data is obtained by abnormal detection;Multi-time scale prediction module integrates ultra-short-term, short-term and medium and long-term prediction, realizes cross-scale cooperation based on joint state space model etc.;Self-evolution calibration module is based on deep reinforcement learning to compensate the residual error of multi-scale prediction online;Distributed coordination control module combines improved alternating direction multiplier method and non-dominated sorting genetic algorithm, obtains optimal power instruction that meets voltage, frequency and harmonic ternary constraint;The present application improves the stability of power grid operation under high penetration rate photovoltaic grid-connected scene.

This application provides a method and system for regional population dynamic monitoring and prediction based on spatiotemporal big data. The method includes: acquiring historical population characteristic data for a preset region, including monthly and annual historical population characteristic data; acquiring a preset prediction model set, filtering and processing it to obtain preset prediction model sets that meet prediction requirements, and marking them as a preset short-term prediction model set and a preset medium-to-long-term prediction model set, respectively; collecting real-time population characteristic data for the preset region; and combining it with the monthly and annual historical population characteristic data, processing it separately using the preset short-term and medium-to-long-term prediction model sets to obtain corresponding monthly and annual population prediction value sets; further processing to obtain the final monthly and annual population prediction values, thereby realizing the technology for regional population dynamic monitoring and prediction based on spatiotemporal big data.

The application discloses a hierarchical operation control method and system for a bio-natural gas production and storage integrated system, and the method comprises the following steps: acquiring system operation states and external environment information; constructing a monthly planning layer, optimizing cross-seasonal gas storage targets based on long-term prediction, and considering slow process dynamics of anaerobic fermentation; constructing an hourly scheduling layer, formulating hydrogen production and power generation plans based on short-term prediction and model prediction control, and introducing a hydrogen synthesis hydrogen injection rate change rate penalty; constructing a minute-level real-time control layer, rolling tracking plans based on ultra-short-term prediction, and designing a rapid response loop to respond to sudden green electricity consumption, peak regulation and frequency regulation demands; and feeding back real-time operation data to the upper layer for adaptive updating of model parameters. The application integrates multi-scale prediction information, coordinates fast and slow processes, realizes efficient consumption of green electricity and flexible support of the power grid, and can be widely applied to the optimal operation control of bio-natural gas production and storage systems.

The application discloses a long time sequence prediction method based on time-varying periodic coding and hierarchical channel fusion. The method comprises the following steps: decomposing input multivariate time series into trend components and seasonal components; performing mapping processing on the trend components to obtain trend output; performing time-varying periodic coding processing and hierarchical channel fusion processing on the seasonal components to obtain seasonal output, wherein the time-varying periodic coding processing is used for adaptively capturing time-varying periodic characteristics in the seasonal components, and the hierarchical channel fusion processing is used for dynamically capturing correlation difference between different channels in the seasonal components; and generating a prediction result of a future time step based on the trend output and the seasonal output. The method can not only adaptively model time-varying periodic characteristics, but also dynamically capture strong and weak correlations between channels, and has excellent long-term prediction performance and generalization.

The application discloses a kind of meteorological space-time prediction methods based on multi-branch frequency perception, belong to meteorological prediction technical field, steps are as follows: obtaining meteorological space-time data, and constructs multi-branch meteorological space-time dataset;Multi-branch meteorological space-time dataset is preprocessed;The branch sample in each branch dataset is input into the space-time prediction model based on multi-branch frequency network, and the prediction output result of branch sample is obtained;Meteorological prediction total loss is constructed;Space-time prediction model is trained and tested, and the trained multi-branch meteorological prediction model is obtained;Meteorological space-time data newly obtained are used to train the multi-branch meteorological prediction model corresponding branch meteorological prediction, and the corresponding meteorological prediction result is obtained.The application solves the problems of error accumulation, insufficient stability and label autocorrelation in long-term meteorological prediction.

This invention relates to the field of meteorological forecasting technology and discloses a method for predicting surface temperature and evaporation. The invention first performs multi-scale mode decomposition on the raw meteorological data, separating high-frequency noise components from mid- and low-frequency effective information components. This eliminates interference from high-frequency random disturbances at the data preprocessing level, effectively solving the inherent non-stationarity and strong volatility problems of meteorological data. Subsequently, a CNN-BiLSTM-Attention hybrid model is constructed to perform parallel predictions on the selected mode components. The core of this invention lies in introducing a barrel theory optimization algorithm to adaptively and jointly optimize the mode decomposition parameters and deep learning hyperparameters, overcoming the limitations of manual parameter tuning. Finally, the prediction results of each component are reconstructed, and long-term predicted values ​​are output. This effectively solves the problems of insufficient accuracy, difficulty in parameter tuning, and weak cross-site generalization ability of traditional methods and existing combined models when dealing with non-stationary, multi-scale meteorological sequences, significantly improving the accuracy, stability, and automation level of prediction.

The application discloses a kind of based on parameterization model's precision retention evaluation method of roller linear guide pair, belong to high-end numerical control machine tool and precision equipment manufacturing technical field.It includes: step 1: establish the precision degradation theory model of roller linear guide pair.Step 2: set the parameter system required for evaluation.Step 3: execute parameterized simulation analysis.Step 4: extract precision degradation law and establish mapping relationship.Step 5: generate evaluation conclusion and optimization strategy.Compared with prior art, the present application has the following significant advantages: efficient prediction, low cost;The established degradation theory model couples elastic deformation, wear and creep, more in line with the actual complex failure physical process of guide pair, improve the accuracy of long-term prediction.The present application can not only be used for performance estimation and scheme comparison in new product design stage, but also can be used for state evaluation, residual life prediction and maintenance planning of in-service machine tool guide, and the application scenario covers the whole life cycle.

The application discloses a social media popularity prediction method, system and device. First, a first-stage data set is established, and then a two-stage retrieval regression prediction network based on retrieval and advanced time modeling is constructed; early time regression pre-training in the first stage of the two-stage retrieval regression prediction network is carried out through the first-stage data set. Then, the regression result is saved back to the data set, so that a second-stage data set is established. Then, graph-guided hybrid retrieval and final time regression in the second stage of the two-stage retrieval regression prediction network are carried out through the second-stage training set. Finally, test set samples are input into the trained two-stage retrieval regression prediction network for verification. The application innovatively proposes the two-stage retrieval regression prediction network, regards cross-sample retrieval features as intermediate representation, explicitly decouples perception modeling and long-term prediction, and effectively introduces cross-sample popularity patterns into the prediction process.

This invention belongs to the field of time series forecasting technology, specifically a method and system for effective wave height prediction based on dynamic constraints and interactive learning. The method includes: acquiring wave time series data, segmenting it according to the parity of the time index to obtain odd and even sequences; determining the boundary frequency based on the spectral distribution, and performing frequency domain decomposition on the two sub-sequences to obtain swell components and wind wave components; performing evolutionary prediction using a linear Koopman predictor and a nonlinear Koopman predictor respectively; and then performing recursive interactive fusion and time series reorganization based on the prediction results to obtain fused features and outputting an effective wave height prediction sequence for a preset future time period. This invention can improve the accuracy, stability, and generalization ability of medium- and long-term predictions.

The present application belongs to the technical field of intelligent transportation, and particularly relates to a traffic state prediction method based on an adaptive dynamic space-time graph convolution network, which comprises the following steps: firstly, adaptively inferring a graph structure from macro and micro perspectives; secondly, extracting time dependence by using a multi-scale gated TCN; thirdly, perceiving a historical change trend hidden in a traffic state data sequence by a time trend perception self-attention mechanism, so as to realize accurate long-term prediction; and finally, generating a dynamic filter at each time step to filter node embedding and generate a dynamic graph, and capturing dynamic spatial dependence by combining an adaptive adjacency matrix. The present application can well mine complex space-time correlation in traffic state data, so as to reveal potential space-time correlation of a dynamic traffic system. Extensive experiments are conducted on two real traffic state data sets, and the experimental results show that the present application has achieved a good prediction level.

The application belongs to the technical field of atmospheric density prediction, and discloses a thermosphere atmospheric density layered progressive full-scale prediction system and method based on cross-source data fusion, which utilizes respective advantages of different gradient data sources, constructs a three-level layered fusion refinement architecture of a benchmark density field, a correction density field and an instantaneous refined density field, and realizes month-year scale long-term prediction, day-week scale medium and short-term prediction and short-term prediction of thermosphere atmospheric density by combining cross-source data calibration and physical constraint modeling. Through the three-level layered refinement architecture, the application fully excavates the long-term coverage advantage of TLE data, the mesoscale variability description advantage of precise orbit data and the high-frequency high-precision advantage of accelerometer data, avoids defects of a single data source, realizes accurate description of full-scale and full-area density, and can improve refinement accuracy and prediction adaptability of thermosphere atmospheric density and reduce data cost.

The application discloses a kind of based on depth condition generation replay's continuous learning soft measurement method, comprising: the historical time series data of industrial process is collected, historical dataset is constructed, to obtain preprocessed dataset;According to StreamKM++ streaming clustering algorithm, the preprocessed dataset is clustered, obtains C cluster;For each cluster, introduce the working condition identification c j Of one-hot coding, build working condition dataset;Using working condition dataset, the prediction model is trained, and the trained prediction model is obtained;Using working condition dataset, the depth condition generation model is trained, and the trained generation model is obtained;Set data buffer, for receiving the preprocessed data sample in online data stream, each data sample includes process variable;Real-time monitoring data buffer, judge whether the data sample stored in data buffer meets condition A or meets condition B;When any condition is met, trigger adaptive mechanism, realize quality variable prediction under continuous learning.The method effectively solves the problem that existing soft measurement model is difficult to overcome concept drift and catastrophic forgetting simultaneously in dynamic industrial environment, improves the adaptive ability and long-term prediction accuracy of model.

This application proposes a three-dimensional sea surface temperature (SST) long-term intelligent forecasting method that considers the influence of multiple factors, belonging to the field of intelligent marine environmental forecasting. Addressing the shortcomings of existing three-dimensional SST forecasting models, such as insufficient consideration of multi-factor coupling effects, high computational complexity, strong dependence on computing resources, and insufficient long-term forecast stability, this application designs a deep learning algorithm by integrating multi-scale convolutional feature extraction, spatiotemporal decoupling Transformer structure, multi-factor interactive attention mechanism, and depth-sensing attention mechanism. Furthermore, a marine physical constraint loss function is introduced during model training. This significantly reduces the computational complexity and parameter scale of the model while improving long-term prediction accuracy, meeting the needs of three-dimensional SST long-term intelligent forecasting under computationally limited conditions. Finally, based on the optimal performance model, a three-dimensional SST long-term forecast product for the next 30 days for the target sea area is generated by inputting multi-source sea surface data from the past 30 days.

The application belongs to the technical field of artificial intelligence and meteorological prediction, and aims at the technical problems that the existing AI meteorological prediction large model lacks ocean memory modeling, ocean-atmosphere coupling feature calculation, memory injection mode is simple, and the ocean memory module and the AI meteorological prediction large model are insufficient in cooperation when dealing with ocean-atmosphere coupling problems, and provides an ocean memory reinforcement method and related device and a meteorological prediction method. The ocean state sequence including the current ocean state and the historical ocean state is acquired, the ocean-atmosphere coupling feature is calculated in combination with the atmospheric state field, the historical ocean state is encoded to obtain the ocean memory vector, and then the current atmospheric state is fused to form the ocean memory reinforcement result and input the AI meteorological prediction large model. The method can enhance the representation ability of the slow change characteristics, memory effect and ocean-atmosphere interaction of the ocean, and improve the medium and long term prediction ability and prediction accuracy of the AI meteorological prediction large model.

The present application belongs to the technical field of watershed sediment control of hydraulic engineering, and particularly relates to a reservoir sediment dynamic balance method and system based on multi-source data fusion and intelligent regulation. In view of the problems of fragmented monitoring, insufficient prediction accuracy and lagging regulation response in the existing reservoir sediment management, the present application constructs an integrated monitoring network of "sky-ground-water" and a multi-source data spatio-temporal fusion mechanism, establishes a machine learning prediction model based on physical-data fusion, generates a Pareto optimal scheduling scheme combined with a multi-objective optimization algorithm, realizes model self-evolution and knowledge base self-update through a closed-loop feedback learning mechanism, and forms a whole-process intelligent system of "monitoring-prediction-regulation-feedback". The present application can realize accurate medium and long term prediction and self-adaptive regulation of reservoir siltation trend, take into account multiple targets such as sediment discharge and power generation, effectively reduce the cost of dredging, and prolong the service life of the reservoir.

The application discloses a wind power optimization prediction model and an application method thereof. The model comprises a multi-source data acquisition module, which is used for synchronously collecting meteorological forecast data, fan equipment data and real-time and historical operation data of a wind power plant; a model construction and optimization module, which is used for constructing a wind power prediction model and adaptively optimizing model parameters through a double-dimension closed-loop calibration mechanism. The mechanism comprises a first dimension of dynamically adjusting meteorological factor weight coefficients based on historical deviation and a second dimension of correcting performance attenuation functions based on equipment fault and performance attenuation data. The two-dimension results are used for acting on model parameters after weighted fusion. A prediction output module is used for outputting wind power prediction data and deviation correction control instructions. The application method comprises a prediction method, a transaction optimization method, an equipment intelligent maintenance method and a closed-loop power control method based on the model. The application solves the problems of weak generalization, uncompensated equipment attenuation and low medium and long-term prediction accuracy of existing models.