15results about How to "Improve predictive reliability" patented technology

The invention discloses a drilling pump pressure prediction method based on an artificial neural network, and the method comprises the following steps: S1, data collection; s2, data preprocessing; s3, constructing a neural network model; s4, model training and optimization; and S5, model verification and deployment. According to the method, multi-channel data fusion and time synchronization optimization are innovatively adopted, so that the data quality and consistency are improved; in combination with a deep neural network and a self-adaptive optimization strategy, the precision and generalization ability of pump pressure prediction of the drilling pump are improved; and an online updating mechanism is introduced, so that the model can be dynamically optimized according to real-time data, the defects of low prediction precision, poor adaptability and difficulty in real-time updating of a traditional method are overcome, and an efficient and reliable prediction means is provided for intelligent drilling control.

The invention provides a method for establishing a nonlinear constitutive model of low-temperature marine steel, and relates to the technical field of ships and materials. A heterogeneous time series data set is provided for model construction through multi-type sample multi-axial load tests and full-field strain acquisition; nonlinear mapping of deformation activation energy about temperature and microscopic state variables is established by taking a crystal plasticity theory as a framework, and the defect that a low-temperature dislocation evolution rule cannot be reflected due to the fact that a traditional model solidifies physical parameters into constants is overcome; a damage evolution operator jointly induced by stress triaxial degree and a principal stress threshold is embedded, Bayesian inference is adopted to carry out parameter joint calibration, model dynamic calibration and parameter continuous optimization are realized through a simulation and experiment closed-loop iteration mechanism, and finally a digital design evaluation model capable of being under untested extreme working conditions is produced. And an engineering tool taking physical interpretability and probabilistic safety margin into consideration is provided for the ice-resistant design of the polar region ship.

The invention relates to a fire tracing and diffusion prediction method and system based on multi-source data fusion and a storage medium, and relates to the technical field of disaster monitoring and early warning. The fire tracing and diffusion prediction method comprises the following steps: collecting and cleaning multi-dimensional environment information of a disaster scene, and obtaining multi-modal environment data; constructing a disaster situation multi-dimensional map according to the multi-modal environment data, and dividing a field dangerous area in combination with a preset disaster situation threshold value; analyzing the disaster situation multi-dimensional map according to the real-time environment data collected by the intelligent carrier, determining the disaster situation category, and positioning the disaster situation generation source; monitoring environment change data of a disaster occurrence source, and constructing a disaster time sequence deduction model in combination with disaster categories; according to the disaster situation time sequence deduction model in combination with the disaster situation multi-dimensional map, pre-judging and determining the regional disaster situation level of the field dangerous region, and drawing a disaster situation diffusion rendering graph; and planning a rescue path of a disaster scene according to the position of the intelligent carrier in combination with the regional disaster level, and providing disaster early warning.

The invention provides a citrus pulp feed bionic digestible energy value prediction method and system based on federal learning, and relates to the technical field of machine learning, and the method comprises the steps: obtaining a local data descriptor of each node, constructing digestion dynamic characteristics through chemical composition and three-dimensional representation, and obtaining a digestion dynamic value of each node; and initializing the physical information neural network by solving the partial differential equation of the electrical response constraint. And performing federal aggregation on the distributed parameters by adopting a geometric manifold technology to obtain a global consensus model parameter base. And generating a virtual sample through physical rule constraint, and calculating a physical consistency residual error in combination with local real data to perform model calibration. And deconstructing the predicted physical determinacy and data probabilistic part of each node to obtain final energy value prediction and uncertainty measurement. According to the method, on the premise of protecting data privacy, a physical mechanism and data driving are effectively fused, and the accuracy and reliability of prediction of the digestible energy value of the citrus pulp feed are remarkably improved.

This paper presents a method and system for correcting and predicting supersonic internal flow fields by integrating topology consistency evaluation and physical constraint latent space mapping, belonging to the interdisciplinary fields of fluid aerodynamics design and artificial intelligence. The method first employs a topology consistency evaluation based on a self-organizing mapping grid to realize the nonlinear coupling evolution of parameter sets and spatial errors under supersonic conditions in a hexagonal topological space. Gaussian smoothing and dot product operations are used to quantify the topology consistency between parameters and spatial errors, automatically selecting core parameters. Next, a physical constraint latent space mapping architecture is constructed, introducing a composite physical loss function to reduce the dimensionality of high-dimensional flow field features and establish a mapping model from core parameters to the latent space. Decoder weights are frozen to ensure the continuity of physical laws and derivatives. Finally, an error-driven correction mechanism is used to statistically analyze residuals and generate an error feedback matrix to complete the implicit core parameters, achieving closed-loop reconstruction and corrective prediction of the surrogate model. This method can significantly improve the physical fidelity of supersonic internal flow field reconstruction and effectively solve the problem of large prediction deviations in the flow field behind the gate.

This invention discloses a method and system for predicting power supply risks based on multi-source data fusion, belonging to the field of intelligent operation and maintenance technology for power systems. The method includes: acquiring multi-source raw data from prediction units and constructing a standard fusion feature vector; combining the standard fusion feature vector with a pre-trained spatiotemporal risk-constrained prediction model to output a probabilistic prediction result containing multi-step future photovoltaic output and bus load prediction values; analyzing the probabilistic prediction result to construct a comprehensive risk index, and outputting a kilometer-level gridded risk heat map as the risk prediction result based on the comprehensive risk index. This invention achieves the fusion and probabilistic prediction of multi-source heterogeneous data, improving the ability to identify power supply risks and the level of precision in early warning under extreme weather scenarios.

The invention discloses a staged combustion chamber design method and device, a storage medium and electronic equipment. The method comprises the steps that a one-dimensional model of an axial staged combustion chamber is built, disturbance parameters are set according to the one-dimensional model, the disturbance parameters comprise pressure waves and entropy waves, and system attribute parameters are preset; respectively constructing a pressure pulsation model, a density pulsation model and a speed pulsation model according to the disturbance parameters and the system attribute parameters; constructing a disturbance matrix and a disturbance column vector according to all the pressure pulsation models, the density pulsation models and the speed pulsation models; system simulation parameters are calculated according to the disturbance matrix and the disturbance column vector, the system simulation parameters comprise the oscillation angular frequency, the oscillation growth rate and the vibration mode space distribution, the one-dimensional model and the system attribute parameters are adjusted according to the system simulation parameters, and the corrected system attribute parameters serve as combustion chamber design parameters.

The invention discloses a front fracturing well closing flowback numerical simulation method considering reservoir damage, and belongs to the technical field of petroleum development reservoir transformation. The method comprises the steps that initial parameters of a target oil reservoir are obtained, fracturing fracture propagation simulation is conducted, and an embedded discrete fracture grid is constructed based on the fracture form obtained through simulation; determining a function relationship between the reservoir damage coefficient and the water saturation according to an experimental result; establishing a reservoir parameter dynamic calculation model; constructing a mass conservation equation and carrying out differential discretization to obtain a numerical discretization solving model; and performing iterative solution on the model, dynamically updating reservoir physical property parameters at each time step, and finally obtaining a pressure field, a saturation field and crude oil yield prediction in the whole fracturing process. The problem that reservoir damage caused by fracturing fluid invasion cannot be dynamically reflected through traditional numerical simulation is solved, accurate simulation of the full-period process of preposed fracturing, well closing and flowback is achieved, and a reliable tool is provided for fracturing optimization and productivity prediction.

The invention relates to the technical field of river flow prediction, in particular to a river section flow prediction method and system based on deep learning. The method comprises the following steps: acquiring upstream rainfall data, downstream water level data and weather forecast data of a target drainage basin; the upstream rainfall data, the downstream water level data and the weather forecast data are aligned according to a unified time reference, and a time sequence alignment data set is constructed; constructing a traffic prediction model based on the time sequence alignment data set; wherein the flow prediction model extracts a time sequence correlation feature vector between an upstream rainfall and a downstream water level, identifies a spatial dependency relationship vector between a multi-branch convergence point and a target section, performs feature fusion on the time sequence correlation feature vector and the spatial dependency relationship vector, and outputs a feature fusion vector. According to the method, the defect that a pure data driven model prediction result violates a basic physical rule is fundamentally overcome, and the prediction reliability in an extreme rainfall event and a data missing scene is remarkably improved.

The invention relates to a bridge technical condition prediction method based on fusion of a graph neural network and time sequence modeling, belongs to the technical field of traffic infrastructure maintenance and artificial intelligence application, and solves the problems of insufficient spatial-temporal feature utilization, lack of engineering logic constraints and rough prediction granularity of an existing method. According to the method, firstly, historical technical condition data are collected and preprocessed, and structured space-time sequence data are obtained; the double-branch feature extraction network extracts a spatial correlation feature vector and a time sequence evolution feature vector in parallel; fusing the degradation trend prior feature vector through a gating mechanism to generate a comprehensive feature vector; outputting an initial continuous prediction score of the target bridge in the prediction year by using the double-branch continuous ordinal number prediction head; and finally, performing logic constraint correction by using a time sequence consistency post-processing algorithm to generate a bridge technical condition grade prediction result. According to the method, the spatial-temporal characteristics are effectively utilized, the engineering logic interpretability is high, and continuous and accurate bridge technical condition prediction can be realized.

This application provides a dynamic modeling and simulation method and system for the energy efficiency ratio (EER) of a solar-powered seawater desalination system, belonging to the technical field of seawater desalination and system modeling. First, this application acquires data on membrane surface resistance, selective permeability, and DC bus voltage ripple during the photovoltaic electrodialysis process. Second, the ripple data is Fourier transformed and concatenated with membrane parameters to generate an input matrix. This matrix is ​​then imported into a deep belief network, and a restricted Boltzmann machine is used to extract the unsteady-state ion impedance vector reflecting the influence of voltage fluctuations. Subsequently, a nonlinear regression model of this vector and unit water production energy consumption is established using a least-squares support vector machine. Finally, the water production rate and EER are calculated based on the predicted energy consumption and photovoltaic power, generating a dynamic simulation curve. This application can quantify the nonlinear influence of photovoltaic voltage ripple on membrane impedance through deep learning, significantly improving the accuracy of EER prediction for seawater desalination systems under fluctuating power supply conditions.

The invention relates to a coordination control method and device suitable for an optical storage hybrid system, equipment and a medium. The method is applied to an optical storage hybrid system, and comprises the following steps: collecting multi-source data, and generating a real-time data vector; based on the vector, generating a photovoltaic output prediction sequence, a load demand prediction sequence and a battery life attenuation rate prediction sequence, and constructing a time alignment prediction data set; according to an energy storage system charge state value and a power grid real-time electricity price in the vector, combining a battery life attenuation rate prediction sequence in the time alignment prediction data set, and constructing a life-economy collaborative optimization objective function; and performing model prediction control optimization processing through the function and the time alignment prediction data set, constructing a strategy library, starting a corresponding operation mode, and outputting an equipment control instruction. According to the method, through multi-dimensional data acquisition and dynamic optimization control, the prediction precision and the full life cycle benefit of the optical storage hybrid system are improved, and the operation stability and reliability of the system are enhanced.

The application discloses a hypersonic vehicle water-entry three-phase flow-structure strong coupling numerical prediction method. The method is based on the physical and causal relationship of 'pressure wave propagation-super-cavitation initial generation-pressure reloading-structure displacement feedback', takes the pressure prediction result as the driving source of the super-cavitation generation and extinction criterion and the fluid-solid interface load updating, realizes the synchronous prediction of pressure evolution, super-cavitation behavior and structure response in the same calculation framework, thereby avoiding the problems of pressure peak distortion, super-cavitation starting time deviation and structure load prediction lag under the conditions of weak coupling and incompressible assumption, and improving the prediction reliability of the pressure peak and its time sequence evolution under high-pressure impact working conditions.

The invention relates to the technical field of environment monitoring, in particular to a land resource dynamic monitoring system which comprises a resource sensing module, a feature extraction module, a land coverage classification module, a dynamic simulation module, a heat island effect analysis module, a decision optimization module, an information interpretation module and an early warning response module. According to the invention, the remote sensing technology and the ground monitoring network are adopted, real-time collection and synchronization of environmental perception data are realized, the principal component analysis and random forest method are applied to feature extraction, the efficiency and accuracy of feature screening are improved, the accuracy of land cover classification is improved, and the deep learning classification technology introduced by the convolutional neural network is improved. According to the method, a multi-agent simulation and system dynamics model is adopted, the practicality of land utilization change dynamic simulation is enhanced, a scientific basis is provided for a heat island effect identification and mitigation strategy by an urban climate model and a geographic space analysis method, and land planning and management decisions are optimized through integration of a multi-objective optimization algorithm and a decision support system.

This invention relates to a method for predicting crop irrigation demand based on an intelligent integrated prefabricated pumping station, belonging to the field of artificial intelligence technology. It includes the following steps: acquiring crop images in an irrigation scenario using a prefabricated pumping station, obtaining auxiliary data, and calculating daily cumulative reference evapotranspiration values ​​to obtain a time-synchronized multimodal dataset; labeling and partitioning the multimodal dataset; constructing an irrigation demand prediction model, including a spectral reflectance enhancement module, a two-stream separation convolution module, a cross-gated fusion module, a region-sensitive pyramid module, and a coupled prediction head module; inputting the partitioned training set data into the model, training the model using a weighted total loss to obtain a trained model; inputting the image to be detected into the trained model to obtain soil moisture content and evapotranspiration prediction results; and making irrigation decisions based on the soil moisture content and evapotranspiration prediction results. This invention can improve the accuracy of irrigation prediction.