16results about How to "Improve capture accuracy" patented technology

This invention relates to the field of power distribution system resilience enhancement technology, specifically a method for enhancing the resilience of a mobile hydrogen energy storage power distribution system. This method includes data acquisition and parameter initialization; constructing physical constraints for the power-transportation coupled system based on a multi-microgrid reconfiguration strategy; expanding the original typhoon scenario using an enhancement strategy to generate an expanded sample set including wind and solar power output prediction errors; constructing a fuzzy uncertainty set covering multi-source uncertainties; establishing a sub-Bruker optimal scheduling model with the objective of minimizing the overall system operating cost; and performing dual transformation and linearization on the sub-Bruker optimal scheduling model to obtain the trajectory of the mobile hydrogen energy storage vehicle, charging and discharging commands, and the action sequences of distribution network segment switches and tie switches. This invention effectively solves the problems of high difficulty in multi-grid resource coordination and difficulty in accurately obtaining data distribution characteristics under extreme disasters, significantly improving the economic efficiency of post-disaster system recovery and achieving an effective balance between robustness and economy.

This application provides a software defect prediction method based on a dynamic path adaptive graph convolutional network. The method includes: constructing graph structure data of software modules; inputting the graph structure data into a graph convolutional network to learn the dependencies between modules; adaptively assigning weights to multi-hop paths through a dynamic path weighted convolutional layer and suppressing long path noise interference based on a path length decay coefficient; optimizing the adjacency matrix in real time based on a multi-head attention mechanism, enhancing key dependency edges and pruning redundant connections; embedding dynamic path scoring and an adaptive graph update algorithm into a classification objective function, updating training weights with a balancing strategy, and outputting defect prediction results through the optimized model. This application can effectively solve the performance degradation problems of traditional graph convolutional networks in scenarios with insufficient dependency modeling, noise accumulation, and class imbalance under the static graph assumption.

The invention discloses an event-driven adaptive sampling and hierarchical transmission method based on a data change rate. The event-driven adaptive sampling and hierarchical transmission method specifically comprises the steps that S1, an equipment end collects industrial equipment operation data at a fixed frequency and generates a data frame; s2, the server side calculates the data change rate of adjacent data frames from the same equipment, and generates an abnormal score based on the change rate; s3, the server side divides the data frame into event levels at least including a heartbeat level, a conventional level, an abnormal level and an accident snapshot level according to the abnormal score Score; s4, the server performs adaptive adjustment on an issuing strategy to the client according to the event level; s5, the server triggers an accident snapshot level event, and reliably transmits and issues the accident snapshot level event to the client through a TCP; and S6, the server side enables the data of different event levels to enter different priority queues and dispatches and issues the data according to the priorities. According to the invention, through event grading, sampling self-adaption and grading transmission, priority arrival of key abnormal data and complete reservation of accident data are realized.

This application provides a multimodal zoned induction and precise trapping method based on airflow field reconstruction, relating to the field of pest control technology. The method includes constructing a peripheral light-odor composite attraction zone and pre-optimizing the odor concentration using AI; forming a circumferential swirling locking zone using an annular guide hood and monitoring the residence time in real time; increasing the fan speed to form an axial laminar flow trapping zone after target detection and using a trajectory prediction algorithm; periodically and intelligently executing reverse micro-blowing to break up the stacked structure of the sticky traps; dynamically adjusting the light-attracting and airflow modes according to environmental parameters and handling external interference. This application achieves high trapping accuracy and low energy consumption.

This invention discloses a method, apparatus, storage medium, and terminal for capturing facial expressions. The method includes: acquiring a face image to be recognized and extracting a target implicit code from the face image; inputting the face image to be recognized and the target implicit code into a pre-trained facial expression capture model, and outputting facial expression information corresponding to the face image to be recognized; wherein the facial expression information is generated based on facial expression features, and the facial expression features are generated by fusing multiple extracted features. Because this application extracts multiple features and fuses them into new features for model training, it can effectively improve the model's accuracy in capturing facial expression information.

This invention discloses a dynamic start-stop and pressure stabilization control method for water pumps based on operating condition identification, relating to the fields of mechatronics and automation control technology. The method includes the following steps: S1, outputting the predicted arrival time node of the reflected wave; S2, outputting operating condition attribute labels; S3, parsing the operating condition attribute labels; S4, inputting the dynamic fluctuation characteristics and graph topology space into an improved DCRNN model, introducing a bandgap filter based on acoustic impedance constraints and a resonant coupling diffusion mechanism to perform frequency domain filtering suppression and impedance matching coupling; S5, outputting a dynamic start-stop delay reference; S6, outputting the primary water hammer wave; S7, outputting the error compensation coefficient; S8, performing adaptive linkage correction on the dynamic start-stop delay reference and wave amplitude evolution reference. This invention overcomes the limitations of traditional methods, such as static threshold fixation, semantic simplicity, and neglect of spatiotemporal coupling topological constraints, providing an efficient solution for proactive defense against transient flow in fluid pipeline networks.

The invention relates to the field of small target detection, in particular to an improved unmanned aerial vehicle small target detection method based on YOLO26s, and the method comprises the steps: constructing a YOLO26s-SOD model for the small target detection of an unmanned aerial vehicle, replacing a first C3k2 module and a second C3k2 module of an original backbone network with a context anchor point residual attention feature fusion module, replacing a third C3k2 module with an A2C2f module, and carrying out the detection of the small target of the unmanned aerial vehicle. Deleting modules behind the windmill-shaped wavelet transform convolution module and connecting the modules with a first convolution module; a first Upsample module in an original neck network is replaced by a first connection module, a context anchor point residual attention feature fusion module and a DySample module which are connected in sequence, a second Upsample module is replaced by the DySample module, and a C3k2 module in the neck network is replaced by the context anchor point residual attention feature fusion module. A DySample module, a second connection module and a context anchor point residual attention feature fusion module which are connected in sequence are added behind the second C3k2 module; and the space focusing capability on a tiny target is obviously improved.

This invention discloses a method for constructing a vegetation drought index that integrates soil information and multi-source remote sensing data. The method includes: constructing an index dataset of multi-source remote sensing data for a given historical year, including a precipitation condition index dataset, a temperature condition index dataset, a vegetation condition index dataset, a water body index dataset, a temperature-vegetation drought index dataset, and a soil moisture drought index dataset; performing Z-score centering on the index dataset, and classifying pixels with Z-scores greater than a threshold as outliers, thereby removing outliers to obtain a preprocessed dataset; performing principal component analysis on the preprocessed dataset, extracting the first principal component as a comprehensive drought index, and filling the result with outliers from the original image to obtain a regional vegetation comprehensive drought index; and assessing the drought risk of a target area based on the regional vegetation comprehensive drought index. This invention improves the predictive ability and adaptability for drought occurrence.

The application discloses a high-fidelity single-sample DOM particle-fluorescent component synchronous characterization system and method, and belongs to the field of water environment monitoring.The system comprises a gradient ultrafiltration enrichment unit, an intelligent FlFFF separation unit, a time sequence-particle size linkage collection unit, a multi-parameter collaborative correction EEM detection unit and a multi-dimensional data fusion analysis module, and realizes synchronous characterization through the steps of gradient ultrafiltration enrichment, intelligent separation, differential collection, collaborative correction detection and multi-dimensional analysis.The application has the advantages of high fidelity and high precision through the technical improvements of double-stage ultrafiltration pressure control and concentration self-adaptive flow field regulation, is suitable for various water environment samples, can output a DOM biological activity score, and provides support for ecological function research.

The present application relates to the technical field of weather forecast, especially a strong convective rainfall short-term forecast method and system based on multi-band radar observation, the method comprising the following steps: obtaining multi-band radar observation data, and performing feature extraction on the multi-band radar observation data to construct a multi-band radar differentiated feature set; performing convective system development stage identification based on the multi-band radar differentiated feature set to obtain a convective evolution stage identification result; dynamically and differentially fusing the multi-band radar differentiated feature set according to the convective evolution stage identification result to generate a multi-band radar fusion feature set; constructing an adaptive short-term forecast model, and inputting the multi-band radar fusion feature set into the adaptive short-term forecast model to output a strong convective rainfall short-term forecast result. The present application supports regional emergency scheduling, and matches the fine and time-sensitive business needs of the short-term warning of strong convective rainfall in coastal urban clusters.

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.

The application discloses a kind of based on big data analysis's radiological medical equipment workload prediction method, it is related to medical equipment internet of things technical field, including the following steps: S1, constructs multidimensional feature portrait tensor;S2, by improving TabTransformer model, introduce the SEIR state evolution mechanism based on space epidemiology modeling, generate time-consuming interference coefficient mapping relationship;S3, output individual time-consuming interference coefficient;S4, output each patient's microcosmic real occupancy time;S5, output single-node net occupancy time after timing compensation;S6, generate the real timing occupancy sequence of radiological medical equipment;S7, output workload prediction distribution atlas.The application overcomes the limitation that traditional method isocoupled simple, ignores queuing ripple effect and static mapping deviation, provides efficient solution for radiological medical equipment workload prediction.

The invention relates to the technical field of machine tool spindles, in particular to a multifunctional combined machining spindle of an electric drive axle shell. A hydraulic steering machine rotates to drive a second lower rotating support to rotate around the axis of a motor, and on the premise that a second upper rotating support is fixed, a whole planet wheel assembly is subjected to deformation transmission; the lower rotating bracket II drives the piece Latin mechanism and the knife handle II to rotate by a specified angle respectively; after the upper portion of a hydraulic steering machine is stressed to rotate, the middle portion of the hydraulic steering machine drives a side shaft driving mechanism to conduct bending machining, the lower portion of the hydraulic steering machine drives a driving gear to rotate, and the driving gear sequentially drives a driven gear and a lower rotating support to rotate. The lower rotating support drives the hydraulic cylinder, the supporting plate, the hydraulic rod, the second lower gear, the knife handle support, the Latin barrel, the Latin clamping jaw and the first knife handle to rotate by a specified angle. The multi-angle machining requirement for the new energy reducer shell oil duct is met through rotation of the two split shaft mechanisms of the lower shaft driving mechanism and the side shaft driving mechanism.

The invention relates to the technical field of power distribution area inverter cooperative control, and discloses a distributed cooperative control method, system and device for a power distribution area and a medium, and the method comprises the steps: calculating the real-time risk values of a plurality of risk indexes corresponding to the multi-dimensional operation data of the power distribution area; a risk adjustment mechanism based on operation scene features is introduced for correction, and a comprehensive risk index and a dominant risk type of the power distribution area are obtained; when the comprehensive risk index exceeds a preset high-risk threshold value, controlling each inverter node in the power distribution area to exchange state information so as to determine a master node and a plurality of slave nodes through a distributed election algorithm; then, the master node is controlled to calculate a total adjustment demand according to the dominant risk type, and an optimal resource allocation scheme is solved and issued to the corresponding slave nodes in combination with the maximum safety adjustment amount and the expected contribution degree received from the slave nodes so as to control execution of the slave nodes; quick response to real-time risks is realized, and the distributed cooperative control efficiency is improved.

This invention provides a method for evaluating the uniformity of the flow field in a wind tunnel test section, belonging to the field of wind tunnel construction technology. This invention acquires three-dimensional full-field instantaneous velocity data and wall shear stress distribution, and uses the Stoer-Wagner minimum cut algorithm based on the edge weight function of velocity gradient and turbulence intensity to divide the flow field into a high-uniformity core region and a low-uniformity boundary region. A transition prediction model based on the Monte Carlo method is established to quantify the random influence of wall roughness. The uniformity of the flow field is determined based on velocity uniformity indices, turbulence indices, and flow deflection angle indices, and the contraction section profile parameters and rectifying grid configuration parameters are iteratively optimized. This solves the technical problem of inaccurate flow field boundary identification caused by the lack of a three-dimensional full-field real-time measurement and numerical simulation collaborative verification mechanism in the evaluation of the flow field uniformity in wind tunnel test sections.

The invention discloses a Qinghai-Tibet Plateau global climate mode rainfall downscaling method based on double attention, and relates to the technical field of rainfall downscaling, and the method comprises the following steps: obtaining coarse resolution global climate mode CMIP6 rainfall data, high resolution reference rainfall data and multi-source auxiliary data, and carrying out the standardization preprocessing of all data; a generative adversarial network model is constructed, and a generator of the generative adversarial network model adopts a dual attention mechanism embedded with a channel attention module and a space attention module and is connected in series with a dense block to form a feature extraction main body; training the generative adversarial network model by adopting a two-stage sequential downscaling strategy; and inputting to-be-downscaled CMIP6 rainfall data into the final model, and outputting to obtain a high-resolution downscaled rainfall product. According to the method, the generative adversarial network fusing dense connection and a double attention mechanism is combined with the multi-scale discriminator, so that high-precision downscaling of the CMIP6 rainfall data is realized.