46 results about "Extensive data" patented technology

The invention belongs to the field of artificial intelligence, and particularly relates to a disaster risk early warning algorithm based on multi-modal fusion and a space-time propagation model, which comprises a multi-modal data management module, a cross-modal semantic alignment module, a space-time propagation graph modeling module, a propagation constraint deduction module, a graded early warning generation module and an online closed-loop calibration module. Depth alignment of multi-modal data on a semantic level, dynamic deduction of a disaster propagation process under constraint of a physical mechanism and operable conversion of an early warning result in a business scene are realized, and the whole system takes a unified space-time framework as a base, takes a propagation graph as a link and takes closed-loop calibration as a guarantee, so that the accuracy of the system is improved. The structural defects of extensive data splicing, propagation modeling deficiency and decision chain fracture in a traditional early warning method are fundamentally overcome.

This invention relates to the field of blasting testing technology and discloses a method for quantitatively representing intrusive blasting waves within a double-layer curtain wall space. The method includes establishing an experimental system, creating a basic model of the double-layer curtain wall structure, conducting blasting experiments, verifying the accuracy of the LS-DYNA simulation software's simulation of the experimental content, and parameter analysis. The parameter analysis includes quantitative analysis using blasting wave intensity, intrusion hole height H, and the distance C between the outer and inner curtain walls as variables. Through finite element simulation, combined with extensive data and theoretical analysis, a method for calculating the peak pressure of the intrusive blasting wave inside the structure is derived. This method for quantitatively representing intrusive blasting waves within a double-layer curtain wall space has the advantage of ease of use. It allows for the exploration of the blasting wave pressure inside the structure and the corresponding implementation of structural protection designs, effectively improving the scientific rigor and reference value of internal structural protection designs, and making a significant contribution to the stable management of structural safety.

The invention belongs to the technical field of universal time Transform models, and particularly relates to a universal time Transform model zero sample prediction training method which comprises the following specific steps: S1, data preprocessing and enhancement: firstly collecting and integrating time series data of multiple fields, then performing feature engineering on the time series data, and then enhancing the data; s2, improving a model structure and a training strategy: firstly, adding an adaptive weight adjustment mechanism in an attention module of a universal time Transform model; through the multi-field data fusion and data enhancement technology, types and features of training data are enriched, so that the model can learn wider data modes and laws; therefore, the model can be better generalized when facing zero sample prediction tasks in different fields, the prediction accuracy and stability are improved, and the overfitting problem caused by single data is reduced.

A data processing system and method provides automated extraction, translation, load (ETL) operations. A server coupled with the processor executes the instructions and extracts data from various sources, such as cloud storage, external APIs and direct uploads. The system may include a stream mode processing unit to process large data files in manageable blocks, thereby improving efficiency and reducing memory load. It performs integrity checks to ensure data accuracy and consistency. The system configures and applies predefined and custom transitions, which is facilitated by user-friendly interface and API integration. Custom conversion logic is integrated into the process, allowing adaptive data operations to be performed. The converted data is then verified and formatted to be loaded into different destination systems. The ETL process is efficient, extensible and user-friendly, and thus suitable for wide data processing applications.

This invention belongs to the field of equipment life prediction technology, and discloses a method and related apparatus for predicting the remaining life of a photovoltaic inverter. The method includes: acquiring junction temperature data of the IGBT modules in the photovoltaic inverter under test; determining the number of junction temperature thermal cycles of the IGBT modules in the photovoltaic inverter under test based on the junction temperature data; calculating the cumulative damage of the IGBT modules in the photovoltaic inverter under test based on the number of junction temperature thermal cycles; and calculating the remaining life prediction result of the photovoltaic inverter under test based on the cumulative damage of the IGBT modules. This invention can more accurately predict the remaining life of the IGBT modules, and thus more accurately predict the overall remaining life of the photovoltaic inverter. The prediction process does not require extensive data collection and training of the prediction model, effectively reducing the difficulty of life prediction and improving prediction efficiency.

The present application relates to the field of drone control technology and discloses a method, system and medium for cooperative task allocation among multiple drones. The method comprises: S1, defining the problem of cooperative task allocation among multiple drones; S2, solving the optimal control strategy for the problem of cooperative task allocation among multiple drones; S3, solving the dual optimization problem based on the primal-dual theory; S4, solving the convex dual optimization problem using the Q-function and Schur Complementation theory transforms the convex dual optimization problem into a semidefinite programming problem. S5, based on the properties of matrix congruence, transforms the semidefinite programming problem into a model-free semidefinite programming problem. S6, using a solver, obtains the optimal control strategy. S7, by varying the weight coefficients and repeating S2-S6, obtains the optimal energy loss function bound. This application, based on the Q-learning method, requires only a small amount of data collection, not a precise dynamics model or extensive data, to obtain the optimal strategy for multi-UAV cooperative task allocation.

Systems and methods ingest extensive data regarding parties and contextual data to determine correlation between a variety of data types, parameters related thereto, and member service representative (MSR) contact events and corresponding staffing levels. The appropriate level of staffing, based on real-time and historical context to meet demand for calls, emails, messages, and other contact or servicing can be calculated without overstaffing.

A method for achieving super-resolution in localizing electrical fields measured at the head surface with electroencephalography through a generative model of the cerebral cortex that has a very high resolution of cortical surface dipoles constructed from the known properties of human cerebral cortex and adapted to optimize the Bayesian explanation the individual's cortical surface electrical fields. The iterative optimization of the prior (generative) with the posterior (observed) fields with extensive data from extended recordings provides a probabilistic estimation of the individual's functional brain activity that can be used to train artificial neural network approximations of the individual's mental activity.

This invention discloses a deep learning-based orthodontic image classification model determination system, relating to the field of image data technology. The system includes a construction module for building an orthodontic data annotation library, comprising multiple tooth overlay models and multiple maxillary / mandibular overlay models, and for annotating orthodontic sample images according to the library. A training module, connected to the construction module, determines the feature information of the annotated orthodontic sample images, selects a deep learning model based on the feature information, and trains the selected deep learning model using the annotated orthodontic sample images to obtain a trained deep learning model. This invention enables data integration and annotation of numerous dental problems through the orthodontic data annotation library, achieving extensive data coverage and ensuring the accuracy and consistency of image annotation. It also enables rapid annotation of oral cavity-related images, improving the efficiency of orthodontic analysis.

This invention discloses an image-based method for detecting and reconstructing irregular strip-shaped objects, comprising the following steps: S1: image acquisition; S2: image preprocessing; S3: optimizing the network model; S4: setting network model parameters; S5: inputting the training set into the optimized network model for training, saving the trained weight file, and using the test set to test the model's segmentation effect on the target; S6: reconstructing branch segments. This method can effectively detect and segment branches with similar background colors, overlaps, and occlusions, without requiring extensive data statistics, and achieves high reconstruction accuracy.

This invention relates to a system and method for inverting formation cavity volume during well gas invasion, belonging to the field of petroleum drilling engineering technology. The system includes a wellbore throttling and circulation pressure control section, a throttling response parameter acquisition section, and multiphase flow simulation software. Using this system, three wellbore throttling response parameters—wellhead back pressure, inlet flow rate, and outlet flow rate—can be acquired. Then, based on a multiphase flow model, bottom hole pressure, gas dissolution rate, and gas invasion flow rate are predicted. Finally, the formation cavity volume is inverted using a reservoir model and gas state equations. This method fully utilizes field-obtainable parameters, eliminating the need for extensive data training, and is simple, efficient, and real-time. Furthermore, the method acquires relevant parameters during pressure control measures, which helps to curb gas invasion development. This invention enables the calculation of formation cavity volume after gas invasion occurs in the wellbore, providing theoretical support for decision-making regarding pressure-controlled drilling and well control measures in such formations.

This invention relates to a method for diagnosing keratoconus by analyzing the geometric properties of Purkinje images. The method involves capturing Purkinje images of the cornea using a low-cost and simple device, such as a smartphone attachment, making it more accessible than current diagnostic tools like slit lamps. By comparing these images to reference images from healthy corneas, significant shifts and distortions are identified as indicators of keratoconus. This non-invasive tool provides an accessible and cost-effective means for early detection and ongoing monitoring. The method's reliability is ensured through extensive data collection from both healthy and keratoconic corneas. This approach offers primary assessments, especially in areas lacking access to advanced medical care, potentially reducing the need for more invasive procedures.

This invention relates to a method, apparatus, electronic device, and medium for calculating landslide thrust, belonging to the field of geotechnical engineering technology. The method includes: acquiring soil structure parameters of a slope and constructing a geotechnical model of the slope; calculating the theoretical value of the landslide thrust by the soil structure parameters; simulating the geotechnical model to obtain simulated values, using the ratio of the simulated value to the theoretical value as a correction coefficient; obtaining a linear relationship between the structure parameters and the theoretical, simulated, and correction coefficients based on different soil structure parameter values; determining a three-dimensional relationship between the soil structure parameters and the correction coefficient based on the linear relationship; acquiring the soil structure parameters and theoretical values ​​of the target slope; determining the target correction coefficient for the target slope; and determining the corrected landslide thrust of the target slope. This invention obtains the correction coefficient through extensive data calculation, thereby obtaining the three-dimensional landslide thrust of the target slope when the theoretical value of the target slope is known.

This invention discloses an NLOS identification method based on the GMM-PSO-EM algorithm. First, the time-delay data is preprocessed and denoised, and the distribution quantity is determined by combining AIC and BIC criteria. Then, the GMM parameters are iteratively updated using the PSO-SA-EM algorithm. After establishing the GMM model, the NLOS probability value within the scene is determined by the area of ​​the region determined by the PDF value. This GMM-PSO-EM algorithm exhibits excellent NLOS identification performance, demonstrating strong effectiveness and robustness in handling NLOS problems. It can solve the problems of high complexity in NLOS identification in complex indoor environments, the need for extensive data labeling, and poor performance with small data volumes. Furthermore, it is relatively simple to implement and has a faster computation speed, making it particularly suitable for identifying NLOS signals during UWB system positioning in complex indoor environments.

This disclosure provides a data reading method, device, and storage medium for a database. The method involves receiving query conditions for the Value column of an LSM-Tree database; retrieving first candidate data from the LSM-Tree database whose Values ​​satisfy the query conditions; constructing a right table based on the Keys of the first candidate data; using the full data of the LSM-Tree database as a left table; performing a semi-join between the left and right tables; obtaining second candidate data from the left table that matches each Key in the right table; and retrieving target data from the second candidate data that satisfies the query conditions. This disclosure implements predicate pushdown in a Merge-On-Read scenario for LSM-Tree databases, avoiding extensive data scanning while balancing write and query performance.

The invention relates to the technical field of vocational education, in particular to a vocational education demand prediction system based on LSTM time sequence prediction and a dynamic adjustment method, and the system comprises a multi-source data collection module, a knowledge graph construction module, an intelligent data processing module, an LSTM-reinforcement learning prediction module, a dynamic adjustment module and an execution feedback module. According to the scheme, firstly, three types of data of industrial economy, post requirements and education and learning are integrated, a knowledge graph containing semantic and quantitative information is constructed, abnormal value recognition and missing value filling are optimized, and data suitability is improved; secondly, capturing time sequence dependence through bidirectional LSTM, optimizing model parameters in combination with knowledge attention and reinforcement learning, and achieving demand prediction; and finally, constructing a resource dynamic coefficient matrix, converting constraints into mathematical conditions to solve an optimal adjustment amount, and feeding back an optimization model and a map in real time, thereby effectively solving the problems of extensive data, prediction deviation and experience-dependent adjustment in the prior art, and realizing efficient matching of vocational education resources and demands.

The invention relates to the technical field of credit scoring, in particular to a real-time credit scoring system based on deep learning, which comprises a data collection module, a data preprocessing module, a deep learning model module and a real-time scoring module, and is characterized in that the data preprocessing module is connected with the data collection module, and the deep learning model module is connected with the data preprocessing module; the real-time scoring module is connected with the deep learning model module; an advanced deep learning algorithm is introduced to process real-time data from a plurality of heterogeneous data sources, such as financial transaction records, social media behaviors, online consumption habits and the like of users, so that a comprehensive, dynamic and real-time updated credit score is generated. According to the system, multiple innovations are carried out in the aspects of the flexibility of an algorithm, the universality of data processing and the real-time performance of risk prediction, and a more comprehensive and efficient credit risk assessment tool is provided for financial institutions.

The present invention provides a distribution line wave velocity measurement and calibration method based on a distribution automation master station. Following a remote control switch operation, the master station sends a command to all traveling wave ranging devices on the switch's line to upload their most recent wave recording. Upon receiving the command, the traveling wave ranging devices upload their most recently cached wave recording, i.e., the wave recording data initiated by the switch operation. After receiving the uploaded wave recording, the master station calculates and updates the wave velocity. This invention provides an active and easy-to-use distribution line wave velocity measurement and calibration method. This method measures and calibrates line wave velocity without relying on faults or requiring extensive data statistics. This method addresses the large errors in distribution line fault location caused by inaccurate wave velocity, improves fault location accuracy, and significantly accelerates distribution line fault handling, significantly reducing property losses, personal safety risks, and power supply reliability losses caused by faults.

The invention discloses a passive and green building coupled indoor environment dynamic regulation and control method, and belongs to the field of green buildings and building automation control. According to the method, environment, energy consumption and user behavior data are collected through a distributed sensor, after normalization and other fusion processing, a time sequence algorithm and building energy consumption are used for simulating and predicting a trend, a multi-target decision matrix is constructed, regulation and control are preferentially carried out through passive assemblies such as a sun-shading system and a phase change material, and the energy consumption and user behavior data are obtained. When the comfort level does not reach the comfort level threshold value, active systems such as an air conditioner and photovoltaic-energy storage are linked for cooperative adjustment, and closed-loop optimization is conducted on a regulation and control strategy through algorithms such as reinforcement learning. The problems that in the prior art, a passive system and an active system are separated, data utilization is extensive, and energy distribution is low in efficiency are solved, and cooperative regulation and control of passive priority and active compensation are achieved. According to the method, the building energy consumption is reduced, the comfort reaching rate is increased, the renewable energy utilization rate is increased, and the method is suitable for intelligent regulation and control of the indoor environment of various green buildings.

The system provides real-time safety scores for motor carriers, accurately predicts future safety scores, and allows users to simulate various safety scenarios. Users can input data and perform complex calculations, ensuring accurate and timely updates of safety scores. The system's efficiency lies in accumulating and analyzing extensive data, leading to more precise predictions of future safety scores. The calculated, predicted, or simulated safety scores are presented on diverse devices such as desktop computers, mobile phones, or other compatible devices.

The invention discloses a student physical health data risk assessment method and system. The method comprises the following steps: acquiring physical health data reported by all schools and physical health data generated by spot check and retest; performing data processing and feature extraction on physical health data reported by a school, and establishing a training sample; establishing a physical health data risk assessment dichotomy model based on a Gaussian kernel function, and performing model training by using the obtained training sample to obtain model parameters; and inputting the obtained model parameters into a risk assessment dichotomy model, performing risk prediction on physical health data autonomously reported by a school, and positioning and outputting abnormal data. According to the method, for student groups of different grades and genders, biological differences of different stages of teenager physique development are considered, subdivided data verification is carried out, collaborative anomalies among indexes are considered, and a more accurate and non-extensive data risk inspection mode based on spot check and retest of high-credibility data is provided.

This invention discloses a fault-constrained Dal-UNet stratigraphic surface reconstruction method, applicable to the fields of information processing and oil and gas exploration. Existing deep learning reconstruction algorithms, based on supervised learning, require extensive data and manually labeled training data for pre-training, resulting in high economic and time costs. Data quality directly impacts algorithm performance, and in geological studies with relatively scarce data, supervised deep learning algorithms are prone to overfitting, making them even more difficult to apply directly to stratigraphic surface reconstruction tasks involving complex geological structures with faults. This invention utilizes the Dal-UNet network to capture hidden features from limited and sparse stratigraphic surface data and combines it with a fault-constrained algorithm to reconstruct complex stratigraphic surfaces containing faults, effectively restoring the true topography.

A method for achieving super-resolution in localizing electrical fields measured at the head surface with electroencephalography through a generative model of the cerebral cortex that has a very high resolution of cortical surface dipoles constructed from the known properties of human cerebral cortex and adapted to optimize the Bayesian explanation the individual's cortical surface electrical fields. The iterative optimization of the prior (generative) with the posterior (observed) fields with extensive data from extended recordings provides a probabilistic estimation of the individual's functional brain activity that can be used to train artificial neural network approximations of the individual's mental activity.

The invention discloses a power system data acquisition system which comprises a data acquisition module, a data storage module, a data processing module, a data transmission module, a data analysis module and a data application module. The power system data acquisition system based on the big data has the characteristics of high efficiency, accuracy, real-time performance, stability, safety and the like, can provide powerful support for intelligent upgrading and optimization of a power system, is expansion and upgrading of a traditional data acquisition system, can process wider data sources, and has wide application prospects. And data values are mined by using a big data analysis technology, so that more powerful decision support is provided for safe, economic and efficient operation of a power system.

The invention belongs to the technical field of laser precision machining and intelligent manufacturing, and particularly relates to a laser machining parameter transfer learning method, system and equipment and a medium, and the method comprises the following steps: S1, specifying a machining requirement, and inputting known physical characteristics and an initial laser parameter range of a to-be-machined material; s2, establishing a basic model library containing a multi-physics field coupling model, and constructing a laser parameter-electronic dynamics-processing result mapping database based on historical experimental data and model simulation results; and S3, based on the physical model library and the mapping database, training a machine learning or deep learning model as a prediction model. According to the method, a multi-physics field coupling model and a data driving model can be fused, the interpretability of the model is ensured by utilizing a physical mechanism, and the prediction precision is improved through mass data training; and the three-stage feedback mechanism realizes real-time adaptive adjustment in the machining process, so that the key index deviation of the machining result can be within an effective control range.

The invention provides a feature mixed cross view image matching method based on MambaVision. The images from the unmanned aerial vehicle platform and the satellite platform are matched, and the geographic position information carried by the images of the satellite platform is used as the current geographic position information of the unmanned aerial vehicle, so that the pure vision geographic positioning task of the unmanned aerial vehicle is realized. And inputting the unmanned aerial vehicle platform image and the satellite platform image into the same feature extraction network, and respectively extracting image features: regarding an image matching process as an image retrieval process, and calculating cosine similarity between the current unmanned aerial vehicle image and the satellite image library image through features to obtain a matching result. According to the method, a University-1652 data set is used for training and testing, and the data set is the most widely used data set in the field at present. Compared with other unmanned aerial vehicle cross-view geographic positioning methods based on deep learning, the method has the advantages of high precision, strong generalization and network lightweight.

The application provides a training data determination method, device and equipment for an insurance field large model and a medium, and relates to the technical field of data processing. The method comprises the following steps: obtaining at least one insurance field text data and at least one general text data; the general text data comprises the insurance field text data and text data of other fields except the insurance field text data; performing filtering processing on the insurance field text data and the general text data to obtain first training data; performing clustering processing on the first training data to obtain second training data; determining target training data based on the second training data; and the target training data is used for training the insurance field large model. The application can alleviate the "memory loss" of the large model by obtaining extensive data and performing a series of processing on the data, and improve the generalization ability of the large model.

This invention relates to the field of electrically excited synchronous motor technology, specifically to a method for rapidly estimating the initial position of a main motor based on estimated rotor current characteristic values, comprising: acquiring the three-phase rotor current of the exciter; acquiring the target rotor current of the exciter; acquiring the excitation current of the main motor; and acquiring the stator current. α shaft and β The invention obtains the variance of the excitation current of the main motor corresponding to the stator shaft; obtains the angle compensation value of the main motor rotor; and obtains the estimated initial position value of the main motor rotor. This invention accurately estimates the excitation current of the main motor and quickly estimates the initial position of the main motor rotor even when the initial position of the exciter is unknown, without requiring extensive data processing, thus simplifying the complexity of the initial position identification process for the main motor rotor.

The present invention provides a guided visual diagnostic system and method for use with medical devices. [Solution] When a technician discovers and reports a device malfunction, the diagnostic system automatically generates a diagnostic plan detailing the necessary parts, areas of interest, and actions to be taken. The perceptual module analyzes the technician's actions and recommends the next steps, guiding the technician using augmented reality (AR) indicators. The planning component creates the diagnostic plan using a knowledge graph (KG) and a large-scale language model (LLM), eliminating the need for extensive data labeling or model training. The tracking component enhances 3D detection using perceptual sensors and a 2D nested object detection model. The guiding component simplifies the process for novice technicians by integrating 2D models and AR interactions to ensure efficient and accurate diagnosis.

This invention discloses a wind speed prediction method and system based on Kalman filtering and LSTM, belonging to the field of wind power generation and wind energy prediction technology. The method includes: acquiring data from the wind turbine's SCADA system, including historical wind speed, wind direction, humidity, and temperature; preprocessing the acquired data, including data denoising and normalization, and using mutual information to filter out the required data; dividing the filtered input data into training and testing sets, and inputting them into an LSTM network for training, obtaining the optimal hyperparameters based on extensive data training and a Bayesian optimization algorithm; inputting the LSTM prediction results into the Kalman filter KF update equation, leveraging the optimal state estimation characteristics of the dynamic system using the KF update equation to correct instantaneous meteorological disturbances and inherent model biases in real time, and finally outputting the wind speed prediction result. This invention achieves high-precision wind speed prediction, facilitates real-time wind speed monitoring, and enables optimal control of wind turbine units.