56results about How to "Ensure reasonableness" patented technology

This invention relates to the field of artificial intelligence visual inspection technology, and discloses an artificial intelligence visual inspection method, system, and storage medium. The method includes: establishing a spatiotemporal field composed of sequential image frames; constructing dynamic behavioral trajectories of pixels within the spatiotemporal field, analyzing their fluctuation and propagation patterns to form evolutionary features; calculating a correlation matrix based on the evolutionary features, identifying pixel clusters, and performing multi-level feature decoupling to separate stable and transient feature layers; performing topological weaving on the stable feature layers to form a feature topology network and extracting structural invariants; using topological rules defined by the structural invariants to constrain the generative network and construct an enhanced feature expression layer; and fusing and projecting the enhanced features with the original image after removing the transient layer to generate a visual inspection image for surface state discrimination. This invention can separate the essential attributes of objects from environmental interference, and improves the accuracy and robustness of detection through feature enhancement guided by topological rules.

This application provides a map generation method and apparatus based on high-resolution remote sensing imagery. Its purpose is to achieve accurate matching and fusion of small patches and semantically compatible large patches through comprehensive similarity calculation of spectral, texture, and shape three-dimensional features, ensuring the accuracy of the generated map of the target area. Specifically, the map generation method based on high-resolution remote sensing imagery includes: acquiring a plot raster map based on remote sensing imagery of the target area; converting the plot raster map into an initial vector polygon map to obtain vector patches; calculating the comprehensive similarity between small patches and any large patch based on spectral, texture, and shape features in the feature data of the vector patches, wherein the large patch is a patch whose minimum distance to the small patch is within a preset distance threshold; fusing the large patch corresponding to the maximum comprehensive similarity with the small patch to obtain a plot vector file of the target area, and using the plot vector file to generate a map of the target area.

The application provides a steel bridge part manufacturing method, device, equipment and medium, and relates to the technical field of bridge manufacturing. The method comprises the following steps: acquiring basic information corresponding to each plate part in a steel bridge; generating a contour map of the plate part based on a three-dimensional model of the plate part, determining a target allowance corresponding to the plate part by using process attribute information and a preset allowance addition rule library, and modifying the contour map to obtain a target contour map according to the target allowance; generating a layout scheme based on the target contour maps corresponding to all plate parts; generating a processing instruction according to the layout scheme, and sending the processing instruction to a corresponding preset processing equipment, so that the preset processing equipment processes a mother plate according to the processing instruction. The application can realize integrated operation of each manufacturing link of the steel bridge part, and effectively improve the efficiency and quality of the steel bridge part manufacturing.

This invention provides a sample entry / exit method and system. Upon receiving sample entry / exit instructions, it automatically determines the task sequence and generates a task scheduling strategy based on the current number of idle robots. This achieves a fully automated process from task reception to execution, significantly reducing manual intervention and improving work efficiency. It also ensures the rationality and timeliness of task allocation, demonstrating a high level of intelligence. By comprehensively considering the task sequence and the number of idle robots, task allocation can be dynamically adjusted to ensure optimal resource allocation, thereby improving overall entry / exit efficiency. Furthermore, based on warehouse layout, shelf location, sample type and specifications, and task allocation results, precise movement paths can be planned for each target robot. This not only shortens robot travel time and reduces energy consumption but also avoids collisions and path conflicts between robots, ensuring the safety and smoothness of entry / exit operations.

The application discloses a multi-agent material calculation consistency calibration method, belonging to the cross field of material science, artificial intelligence and computational physics. The method adopts a master-slave structure multi-agent system, containing one master agent and at least three slave agents. The master agent is responsible for overall coordination, embedding physical mechanism constraints, consistency calibration and output results. The method steps are: constructing a calculation system, introducing physical mechanism constraints and embedding slave agent calculation models, collecting calculation data, performing bias analysis, adjusting parameters or correcting results through iterative calibration, and finally verifying and outputting the final results. The application introduces physical mechanism constraints, combines bias analysis and iterative calibration, and solves the problems of poor consistency of existing multi-agent material calculation results and easy violation of physical laws. Experiments show that the method reduces the deviation of the calculation results and the experimental measured values to below 3%, improves the calibration efficiency by more than 40%, and improves the calculation accuracy by more than 35% compared with the prior art, thereby improving the accuracy and engineering practicability of high-throughput calculation of material performance.

The present application relates to a kind of flexible needle puncture path planning methods of fusing improved A algorithm and artificial potential field method, comprising: obtaining medical image, extracting the target point of medical image and the obstacle needing to be avoided, generates planning coordinate plane;In the planning coordinate plane, the node is expanded using puncture needle simplified model, based on multilevel heuristic function, and the guide mechanism of introducing artificial potential field is generated evaluation function:Through the evaluation function, the node after expansion is evaluated, according to the evaluation result, the planning path of flexible needle puncture is generated.The present application can obtain high-precision, high-security autonomous path planning solution, so as to better cope with the challenge of non-uniform obstacle distribution and dynamic anatomical structure, optimize the clinical executability of puncture trajectory, improve the calculation efficiency, and improve the success rate of operation and patient safety performance.

This invention discloses an abnormal driving recognition method based on pedal data and facial expression multimodal data. The process is as follows: Step 1: Collect vehicle driving data and facial video data, and label real facial abnormal expressions to obtain original multimodal data composed of multiple original multimodal samples; Step 2: Train an original abnormal driving model based on the original multimodal dataset; Step 3: Perform noise diffusion on each original multimodal sample to obtain diffused multimodal samples, and select effective multimodal samples from the diffused multimodal samples to construct a multimodal dataset; Step 4: Process the multimodal dataset using the original abnormal driving model, and select prediction-consistent multimodal samples to construct a prediction-consistent multimodal dataset; Step 5: Train the original abnormal driving model again using the prediction-consistent multimodal data to obtain an abnormal driving recognition model; Step 6: Use the abnormal driving recognition model to predict the abnormal driving prediction result.

A power distribution network resilience improvement method and system considering electric vehicle V2G willingness, comprising: constructing a linear decay model of weather influence coefficient, a wind-induced road physical damage model and a water depth penalty model, calculating the shortest travel time of mobile energy storage vehicle MESS from an access node to another node based on all models; based on the comprehensive utility, the response probability of electric vehicle EV participating in grid interaction V2G is calculated, and the EV cluster response coefficient of each node is calculated through the response probability and the maximum discharge power; combined with the EV cluster response coefficient, a pre-disaster pre-layout two-stage robust optimization model is constructed to solve the pre-disaster mobile energy storage vehicle MESS pre-configuration position and pre-disaster scheduling variable, and combined with the shortest travel time, a post-disaster multi-source collaborative recovery model is constructed to solve the post-disaster mobile energy storage vehicle MESS position and post-disaster scheduling variable. The present application ensures the timeliness and rationality of post-disaster MESS scheduling, and improves the post-disaster recovery efficiency of the power distribution network.

The invention discloses a customer group data processing method and device, electronic equipment and a storage medium, and relates to the technical field of big data or other related fields, and the method comprises the steps: obtaining transaction data of N financial customers from a financial database, and constructing a financial customer group network according to the transaction data, N being a positive integer; based on a preset threshold value formula and a spectral clustering algorithm, the financial customer group network is classified, the financial customer group network is disassembled according to a classification result, M sub-graphs are obtained, and M is a positive integer; solving a maximum independent subset for each sub-graph by adopting a quantum computing optimization strategy, and determining customer associated information corresponding to the maximum independent subset; and updating the current customer group information in the financial database by using the customer association information corresponding to all the sub-graphs. Through the method and the device, the technical problem that customer group information cannot be updated in time due to low processing efficiency of financial customer relationship data in related technologies is solved.

The invention discloses a cooperative state estimation method and system based on Kalman filtering and deep learning, and relates to the technical field of cluster control, and the method comprises the steps: obtaining a state value of an intelligent agent at a (k-1) th sampling moment and a control value of the intelligent agent from the (k-1) th sampling moment to a kth sampling moment; according to the obtained state value and control value, a predictive state value and predictive uncertainty of the intelligent agent at the kth sampling moment are obtained by using a deterministic kinematics model and a process noise neural network; obtaining observation data of the intelligent agent on a neighbor intelligent agent, calculating a structured residual error based on the prediction state value, the observation data and the prediction state value of the neighbor intelligent agent, and inputting the prediction state value, the prediction uncertainty, the observation data and the structured residual error of the intelligent agent at the kth sampling moment into a posterior neural network to obtain a predicted state value of the intelligent agent at the kth sampling moment; and obtaining a posterior state value and posterior uncertainty of the intelligent agent at the kth sampling moment. According to the method, the state estimation precision and the robustness to environmental interference can be remarkably improved.

The present invention discloses a method and system for simulating and analyzing the impact on the marine resource environment. The method of the present invention includes constructing a two-dimensional plane numerical model for simulating the marine hydrodynamical environment; processing the grid of the two-dimensional plane numerical model to obtain a grid processing result; constructing a large sea area model based on the grid processing result; determining the water boundary input, time step, bed roughness rate and horizontal eddy viscosity coefficient of the large sea area model to obtain a simulation and analysis result. Through fine grid processing, the details of the marine hydrodynamical environment can be better captured, thereby improving the prediction accuracy of the model.

This invention discloses a method for predicting the maneuvering motions of hydrogen fuel cell-powered ships based on multi-source data fusion, relating to the field of intelligent ship navigation. The method collects geodetic coordinate system position and heading data from the BeiDou Navigation Satellite System and ship hull coordinate system acceleration data from the Inertial Measurement Unit (IMU). After differential and integral preprocessing, a neural network model is built. A fusion-type loss function is constructed based on the transformation relationship between the ship hull coordinate system and the geodetic coordinate system. The model is trained using a backpropagation algorithm and can also be extended to fuse data from multiple sensors such as GPS. This invention fully leverages the complementary advantages of multiple sensor sources, effectively compensating for the inherent deficiencies of single-sensor data, significantly improving the accuracy and robustness of ship maneuvering motion prediction, and providing reliable technical support for intelligent navigation decision-making and safety assurance of hydrogen fuel cell-powered ships.

The invention discloses a physical information enhanced cluster cooperative intelligent positioning method and system, and relates to the technical field of cluster control, and the method comprises the steps: obtaining the posterior state estimation of an agent and a corresponding covariance matrix through an extended Kalman filtering algorithm based on the inertial measurement data of the agent; constructing a dynamic space-time diagram based on effective relative position observation information among the intelligent agents; according to the posterior state estimation of each agent and the corresponding covariance matrix, obtaining an input feature vector corresponding to each agent, and inputting the input feature vector and the dynamic space-time diagram into a decoupling double-head diagram neural network model, obtaining a global state correction amount for the whole agent cluster and a local relative state correction amount corresponding to each agent, wherein the global state correction amount and the local relative state correction amount are output by the model; and calculating the final state estimation of each agent according to the posterior state estimation and the local relative state correction of each agent and the global state correction. According to the method, the cooperative positioning precision and robustness can be remarkably improved.

The application provides a data enhancement processing and model training method and device, electronic equipment and a storage medium, comprising: determining at least one enhancement processing category of original data according to the data type of the original data; and performing data enhancement processing on the original data according to at least one enhancement processing parameter of the at least one enhancement processing category and a randomly generated noise parameter to obtain enhanced data of the original data. The application can ensure the rationality and universality of the data enhancement processing result by introducing a random noise parameter, and help improve the model training effect.

The invention discloses a mixed loading measurement equipment data acquisition stability method, system, equipment and medium, and relates to the technical field of industrial internet-of-things edge computing, comprising the steps of performing protocol conversion and preprocessing on original data of mixed loading equipment, performing data standardization, analyzing data fluctuation characteristics to evaluate disturbance, and determining the stability of mixed loading equipment data acquisition. Introducing a disturbance shielding mechanism to accumulate and filter historical disturbance, obtaining a shielding quantification result, evaluating a rhythm state by combining the shielding quantification result and a disturbance intensity index, deducing a sampling rhythm adjustment amount of a next period, superposing the adjustment amount with a basic sampling period, and dynamically controlling a data acquisition frequency. According to the method, a complete closed loop from data access to intelligent decision making is formed, and the data acquisition continuity, the system adaptability and the operation reliability of the mixed loading measurement equipment in a complex industrial environment are remarkably improved on the premise that the hardware cost is not greatly increased.

The application provides a structure perception-based penetrating detection radar target detection method and system, which comprises: an interface position estimation stage, wherein according to three-dimensional detection data acquired by a through-wall life detection radar, the spatial position information of at least one physical structure interface in a detection space is automatically estimated and determined; a layered target detection stage, wherein according to the spatial position information of the physical structure interface, the three-dimensional detection data is adaptively divided into multiple independent sub-data blocks corresponding to different longitudinal physical spaces along the distance dimension; and target detection processing is respectively performed in each independent sub-data block to obtain multi-target detection results distinguishing different depth positions. The application effectively solves the technical problem of confusion and ambiguity of different depth targets in a longitudinal detection scene of a through-wall life detection radar, and significantly improves the multi-target layered recognition capability.

The application relates to an application method, equipment and medium of power distribution network harmonic tracing in responsibility division, wherein the method comprises the following steps: S1, collecting harmonic data at a public coupling point, system side equivalent impedance and user side average fundamental active power, and performing pretreatment; S2, performing harmonic responsibility qualitative judgment through waveform feature analysis based on the harmonic data and the system side equivalent impedance, and generating a responsibility preliminary judgment conclusion; S3, performing responsibility quantitative calculation by using a comprehensive evaluation method based on the harmonic data, the system side equivalent impedance and the user side average fundamental active power, and obtaining responsibility quantitative parameters; S4, generating a responsibility division result based on the responsibility preliminary judgment conclusion and the responsibility quantitative parameters; the equipment and the medium are used for realizing the method. Compared with the prior art, the application significantly improves the accuracy and environmental adaptability of harmonic tracing and responsibility division.

The invention relates to a system and a method for calculating annual peak and deep adjustment capability of a power system in a power spot market, which realize unified evaluation of adjustment capability of a power generation side and a power utilization side through a supply and demand integrated architecture of data input, quotation modeling, market clearing, capacity authentication and result output, and realize accurate evaluation of the adjustment capability of the power generation side and the power utilization side based on refined analysis of 8760 hours in the whole year. And the actual operation characteristics of the power system are accurately reflected. And in combination with fuel, wear, environmental cost and other models, a scientific basis is provided for optimal configuration of electric power resources. And source-load double-side coordinated regulation evaluation is realized, the power utilization side peak load shifting capability is taken into consideration of system flexibility, and the new energy consumption capability and the power grid safety are improved.

This application discloses a method, apparatus, storage medium, and electronic device for positioning testing of freight trains. The method includes: acquiring satellite positioning information, driver-input departure information, identification and message information of the first group of physical transponders, and track installation position information of the first group of physical transponders; verifying the validity of each piece of information to obtain the validity status corresponding to each piece of information; combining the validity statuses of each piece of information to form a validity status combination result; generating a test scenario set based on the validity status combination result, scenario dimension layering rules, operating condition probability weighting strategy, and anomaly coupling rules; adjusting the triggering timing and triggering intensity of the target test operating condition based on the test scenario set and the real-time operating status parameters of the freight train; collecting the operating response data of the onboard system and the interaction status data between the onboard system and the radio block center during the simulation of the target test operating condition; verifying the operating response data and the interaction status data; and generating test results.

The invention discloses a static sounding data acquisition quality management and control method and system fused with abnormal early warning, and relates to the field of data analysis and processing, and the method comprises the steps: an edge end collects original static sounding data through associated static sounding equipment, and carries out edge abnormal early warning through an embedded local geologic control model; based on the edge anomaly early warning result, performing transmission pre-processing on the original static sounding data in combination with a preset packaging strategy, obtaining transmission state static sounding data, and transmitting the transmission state static sounding data to the cloud; the cloud carries out depth anomaly identification on the transmission state static sounding data, and the depth anomaly identification comprises knowledge verification and mode verification; and performing structured storage on the transmission state static sounding data in combination with the depth anomaly recognition result and a preset continuity rule set. The problems that in the prior art, reliability of data transmission and storage is low, the data anomaly detection effect lags behind, and safety verification of data storage is lacked are solved.

The application discloses a kind of intelligent evaluation methods of strain tower jumper laying length, belong to electric digital data processing technical field, this method includes: the parametric digital twin model including rigid body dynamics is established, jumper is dispersed into multiple rigid body-spring-damping system;Input dynamic wind load time series curve drives transient dynamics simulation;At each simulation time step, the directional bounding box of jumper microsection and tower body component is extracted, and the penetration depth is calculated using the separation axis theorem;When the penetration depth is less than the electrical safety clearance threshold, a reverse kinematics solver is triggered, with the jumper hanging points on both sides as fixed constraints and the spatial rotation angle of the microsection connection point as optimization variables, to solve the shortest interference-free spatial path that meets the clearance requirement;The path is integrated by three-dimensional arc length and corrected by adding elastic elongation, and the jumper laying length is output.The application overcomes the interference hidden trouble caused by static calculation without considering dynamic wind bias strategy, and avoids the excessive sag caused by increasing fixed allowance.

This invention relates to the field of charging scheduling technology, specifically to a dynamic scheduling method, device, and medium for embodied charging robots, addressing the problems of existing methods failing to achieve strong consistency in state management, resource conflict coordination, and anomaly closed-loop handling under multi-robot collaboration. The method includes: an edge scheduling server constructing a dual-state mapping relationship and acquiring spatiotemporal characteristic data of each embodied charging robot's arrival at the task point; determining shared resource occupancy conflicts based on the spatiotemporal characteristic data and the dual-state mapping relationship, and filtering a set of available embodied charging robots based on the real-time operating status of each robot; generating a scheduling strategy using the conflict determination results and the set of available robots as constraints, performing atomic change operations to lock shared resources and issuing execution instructions; acquiring stage data reported by the robots, triggering state rollback when an anomaly is determined based on the dual-state mapping relationship, releasing the locked shared resources, and performing scheduling updates corresponding to the anomaly event.

The application relates to a page generation method and device, equipment and a medium, and the method comprises the following steps: in response to a page generation event, determining table structure metadata of a target data table corresponding to the event, wherein the table structure metadata comprises field types, index information and field annotation information; determining target query fields according to the index information, determining front-end component types corresponding to each target query field based on the field types of the target query fields and preset component mapping rules; generating query configuration and display configuration according to the front-end component types and the field annotation information, wherein the query configuration is used for defining a query component containing the target query fields, and the display configuration is used for defining a data table component with the field annotation information as a table header; and rendering and generating a target page according to the query configuration and the display configuration. The application automatically reads database table structure metadata and generates a target page in a configuration-driven manner, thereby realizing automatic construction from a data table to a functional page.

The invention discloses a regional development quality comprehensive evaluation method and device based on weight fusion and a medium, and relates to the field of regional development quality assessment, and the method comprises the steps: obtaining multi-dimensional evaluation index data of a to-be-evaluated region, carrying out the data statistics and standardization processing, and obtaining the processed multi-dimensional evaluation index data, an entropy weight method, a CRITIC method and an MAR-IDW method are respectively utilized to determine an entropy weight, a normalization weight and a linear normalization weight of each index; according to the entropy weight, the normalization weight and the linear normalization weight of each index, using an adaptive weight fusion strategy to determine a fusion weight of each index; and according to the fusion weight of each index, determining a comprehensive evaluation index of the to-be-evaluated region by using a TOPSIS method of improving positive and negative ideal solutions based on quantiles so as to determine the development quality of the to-be-evaluated region. According to the invention, more accurate comprehensive evaluation of regional development quality can be realized.

This invention relates to a bidirectional laser scanning measurement system and method for beach topography in a flume experiment. By using a platform equipped with a bidirectional laser ranging system that moves above the flume, a set of two-dimensional coordinate points determined by horizontal and vertical lasers is accurately acquired. An abnormal terrain response and remeasurement mechanism based on real-time slope judgment can automatically identify and correct data anomalies and outliers caused by transient interference. An innovative underwater laser autonomous refraction correction method under non-drainage conditions is proposed, enabling the direct acquisition of high-precision underwater topographic data without drainage or additional water level sensors. Finally, a bidirectional round-trip scanning data fusion mechanism based on interpolation alignment and averaging is introduced to reconstruct and track the evolution of the coastal profile over time. This invention features high precision, high versatility, automated data processing capabilities, and adaptability to amphibious scenarios, reducing the workload of experimental personnel while improving overall experimental efficiency.

This invention relates to the field of medical informatics technology, specifically to a knowledge graph-based auxiliary system for hematological transplantation diagnosis and treatment. The system includes a knowledge graph structure optimization module, a reasoning path constraint module, a candidate solution generation module, a strategy consistency verification module, and a graph repair guidance module. In this invention, by optimizing the structure of the hematological disease knowledge graph, identifying and correcting the feature density and sparse connection regions between nodes, the adaptability and information density of the knowledge graph are enhanced, improving the accuracy and comprehensiveness of the reasoning path, enhancing the medical rationality of decision-making and the accuracy of reasoning, and ensuring the practicality and accuracy of recommended solutions. Combined with candidate strategy generation and consistency verification, the logical closure between strategies is strengthened, ensuring the feasibility and rationality of transplantation diagnosis and treatment strategies. Through graph repair guidance, the structure and integrity of the knowledge graph are optimized, solving the problems of incomplete information and inaccurate reasoning in traditional systems.

The invention discloses a physical model-driven fragrant sugar material atomization video segmentation method and system, and relates to the cross technical field of computer vision and digital image processing. The method comprises the following steps: acquiring a video sequence of a fragrant sugar material atomization process, and performing frame-by-frame preliminary foreground segmentation on the video sequence by adopting a Gaussian mixture model to obtain an initial foreground mask corresponding to each frame; and for the initial foreground mask of each frame, extracting at least one foreground region, and calculating feature information of each foreground region. And on the basis of the atomization fluid mechanics model, physical constraint conditions for verifying the foreground regions are constructed by using the matching relationship of the foreground regions among continuous frames. And verifying the successfully matched foreground region by using a physical constraint condition, screening out regions which do not conform to a physical rule based on a verification result, and reserving the regions which conform to the physical rule as real foreground regions. And generating a final foreground segmentation result of each frame based on all the real foreground regions.

This application discloses a method and equipment for recommending fracturing construction parameters for horizontal wells in transitional marine-continental shale gas reservoirs. The method includes: establishing a comprehensive geoscientific model of the transitional marine-continental shale gas reservoir based on geological parameters; determining a target fracture network model based on microseismic parameters and current fracturing parameters, and determining a target production simulation model based on current production parameters; determining single-stage mechanism models for each type of shale reservoir based on the target fracture network model, the target production simulation model, and the comprehensive geoscientific model; determining multiple sets of simulated fracturing construction parameters corresponding to the current fracturing construction parameters for each type of shale reservoir; determining multiple sets of simulated production parameters through the single-stage mechanism models for each type of shale reservoir; determining multiple net present values ​​(NPVs) based on the multiple sets of simulated production parameters, and determining the simulated fracturing construction parameter with the largest NPV as the target fracturing construction parameter. The technical solution provided in this application can improve the rationality of fracturing construction parameters for horizontal wells in transitional marine-continental shale gas reservoirs.

The application provides a kind of goods grabbing and placing strategy generation method and system, method includes the sampling processing to original video data, obtains continuous video frame sequence;Video frame sequence is encoded, and the discrete visual vocabulary sequence is generated;The time sequence dependence modeling of visual vocabulary sequence is carried out to learn the continuous change rule of object action in the video content represented by visual vocabulary sequence;Based on the visual vocabulary corresponding to the current state of the goods to be operated and the visual vocabulary corresponding to the expected target state, and using the continuous change rule of object action, the latent action vector representing the action execution sequence required for the transition from the current state to the target state is generated;The real-time physical state information of the robot obtained is mapped with the latent action vector, to generate the control instruction for driving the robotic arm to perform goods grabbing or placing operation, effectively reduces the dependence on a large amount of manually labeled data, and improves the self-adaptation and generalization ability of the model to unseen scenarios.

The present application relates to a kind of SAR satellite vacuum thermal test method based on phased array feed system, first, passive reflector antenna is removed and is verified separately, realizes system decoupling.Phased array feed thermal test system and satellite body thermal test system are constructed, and two are physically isolated in same vacuum environment simulator by cold plate, to realize the parallel test of load and platform.Based on the size of phased array feed, the size of wave-absorbing box containing silicon carbide sharp cone wave-absorbing material is determined;Based on the size of satellite body, the size of its infrared cage is determined;Further, the size of cold plate is determined and appropriate vacuum tank is selected;Finally, the synchronous test of two systems is realized through cable connection.The present application significantly shortens the test cycle, reduces cost and risk through parallel test architecture, and ensures the authenticity and effectiveness of the test through accurate size chain design and independent thermal environment control, providing an efficient and reliable solution for vacuum thermal test of similar complex satellites.