10results about How to "Logical" patented technology

The application provides a robot, a human-computer interaction method based on a robot emotion level and a memory.The human-computer interaction method based on the robot emotion level takes the robot as a core, first summarizes emotion states that the robot should have, and then describes an emotion state conversion mechanism, so that the robot can better process input speech of a user in an actual interaction process, and the emotion interaction ability of the robot is stronger.

The present application relates to the field of computer vision, and specifically discloses a DHA algal oil soft capsule filling amount insufficient detection method based on a deep learning neural network, comprising: accurately separating each capsule instance from a production line transmission image through a lightweight instance segmentation model; synchronously analyzing the global contour and local density features of a single capsule region using a multi-scale feature extraction network, which innovatively introduces an auxiliary density estimation module to enhance the perception of the content filling state; and fusing the above features to complete the accurate classification of the filling state. The present application effectively solves the practical problems of feature confusion, insensitivity to minor changes, poor dense detection effect, and the difficulty in balancing model efficiency and accuracy, and realizes high-precision and high-efficiency automatic detection of DHA algal oil soft capsule filling amount insufficient.

This invention provides a method and system for table data augmentation based on a conditional inference diffusion model, belonging to the field of table data processing technology. The method first acquires the original table data and preset constraints, and constructs a conditional inference diffusion model composed of a chain inference network and a conditional diffusion network. The chain inference network extracts implicit relationships between data based on the original table data, preset constraints, and a large language model, outputting intermediate table data containing these relationships. Subsequently, the conditional diffusion network uses this intermediate table data as constraints to perform data augmentation operations on the original table data, ultimately generating the target table data. This method achieves data augmentation through the conditional inference diffusion model composed of a chain inference network and a conditional diffusion network, effectively solving the problems of poor generation quality and weak data relationship preservation ability of traditional generation models in complex table scenarios, thereby achieving high-quality table data augmentation.

This invention discloses a wet classification and water resource recycling method and system for water-atomized metal powder, belonging to the field of metal powder preparation technology. The core of the method lies in using wet hydraulic classification to treat the water-atomized slurry, which is carried out entirely in a liquid medium, distinguishing it from dry classification. Specifically, it includes: wet sedimentation and concentration of the slurry, wet hydraulic classification, wet precision filtration, and purified water reuse. This invention, through a wet classification technology route, is naturally adapted to the slurry morphology produced by water atomization. While efficiently separating powders of different particle sizes, it directly addresses subsequent water resource purification and recycling, achieving a unified approach of full powder recovery and closed-loop water resource recycling. This solves industry problems such as high energy consumption during dry-wet conversion, difficulty in collecting fine powders, and water waste.

The invention provides a machining method and device for a cross-flow fan with multiple sections of impellers connected in series. The machining method comprises the steps that at least four end plate blanks guaranteeing the parallelism of the upper face and the lower face are stacked and fixed to a tool plate; synchronously machining a rotating shaft center hole, a positioning pin hole and a directional reference surface corresponding to the axis of the impeller on the fixed end plate blank; the end plates are detached; a cross beam corresponding to each impeller unit is machined, and positioning pin holes in the two sides of the cross beam are machined on the basis that the parallelism of the butt joint plane is guaranteed; two end plates corresponding to a single impeller unit are positioned and placed on the datum planes of the two end plates, a coaxial tool penetrates through rotating shaft center holes of the two end plates, a cross beam is arranged between the two end plates and pre-fastened through screws, and positioning pin holes in the end plates and the cross beam are corrected and finely machined through a reamer; at least two impeller units are connected in series along the axis, and positioning and fixing are conducted at the butt joint positions of the adjacent impeller units. The cross-flow fan manufactured based on the method is small in vibration, low in noise and stable in operation, the product performance is substantially improved, and the service life is substantially prolonged.

The invention provides a household broadband installation and maintenance service work order quality inspection method based on an intelligent agent. In the method, firstly, a quality inspection request is obtained, a quality inspection agent is matched from an agent pool according to the quality inspection request, the quality inspection agent is obtained based on a quality inspection rule and an agent meta-model, and the quality inspection agent comprises quality inspection capability; secondly, the sound recording file is translated into text data, and the sound recording file is used for recording a service call corresponding to the home broadband installation and maintenance service work order; and then, matching a cue word template according to the quality inspection capability of the quality inspection agent, and generating a cue word based on the cue word template and a quality inspection rule corresponding to the quality inspection agent. And finally, reasoning the text data and the cue words through the large model to obtain a quality inspection result. Through the method, the change of the quality inspection service can be quickly responded, and the quality inspection accuracy and the service fitting degree are improved.

The application discloses a kind of tube plate laser cutting equipment and its using method, it is related to laser cutting machine technical field, in tube plate laser cutting equipment, pipe material warehouse, machine tool and plate material warehouse are sequentially arranged along X axis direction, pipe material warehouse is equipped with pipe material feeding mechanism, plate material warehouse is equipped with plate material feeding mechanism, machine tool moves along X axis direction by X axis mechanism;The lower portion of portal frame is also equipped with pipe material lifting mechanism, and the pipe material feeding mechanism is movably connected with the portal frame along the Y axis direction;Pipe material feeding mechanism, plate material feeding mechanism, X axis mechanism, pipe material lifting mechanism, pipe material feeding mechanism, Y axis mechanism, Z axis mechanism and laser cutting mechanism are electrically connected with host computer;The method is: after host computer receives the feeding command of pipe material or plate material, host computer drives machine tool to move to plate material or pipe material feeding station, and the pipe material or plate material feeding mechanism is fed to machine tool.The application can improve the degree of automation of pipe material and plate material feeding, and is beneficial to improve feeding efficiency.

The invention relates to a geotechnical characteristic information parameterization method for intelligent prediction of a slope instability mode. The geotechnical characteristic information parameterization method comprises the following steps: identifying and classifying to-be-processed slope original characteristic information; encoding the text type data and the numeric type data; defining a coding result as an input parameter of the slope instability mode prediction model; obtaining classification labels used for representing different instability modes, and defining the classification labels as output parameters; outputting the data set; the data set is used for constructing and training a slope instability mode prediction model. The method has the advantages that the generated standardized data set is clear in physical significance, rigorous in logic and uniform in format and can be directly called by a machine learning algorithm, a high-quality training basis is provided for an intelligent analysis model, and the training efficiency and prediction accuracy of a slope instability mode intelligent judgment model are remarkably improved.

The present application relates to the field of machine vision and food intelligent detection technology, and discloses a machine vision method and system for French fries appearance detection. The method comprises: applying active thermal excitation to French fries and collecting thermal image sequences; obtaining short-wave infrared hyperspectral data to extract molecular bond vibration spectral features; constructing a temperature decay field model based on thermal images to extract thermodynamic features; associating spectral and thermodynamic features through a multi-modal fusion neural network to generate an endoplasmic fusion feature vector; and inputting the feature vector into a discrimination model to output quality evaluation results. The system comprises a thermal excitation module, a thermal imaging acquisition module, a hyperspectral acquisition module, a central processing unit and an execution mechanism. The present application couples molecular spectra and thermodynamic features, realizes non-destructive perspective detection of the internal structure and composition of French fries, and improves detection accuracy, robustness and automation level.

Embodiments of the present application provide a design method and device for a hydraulic oil tank of a civil passenger aircraft, computer equipment and a medium, wherein the method comprises: determining basic parameters including system inherent volume, oil physical properties and extreme working conditions, calculating system total demand volume and initial oiling volume according to the basic parameters; substituting the initial oiling volume into complete flight profiles under cold and hot temperature working conditions respectively, dynamically analyzing oil tank oil volume change curves caused by temperature changes and actuator actions in each flight stage, and extracting a global minimum oil volume; if the global minimum oil volume does not meet the oil absorption safety margin requirement, iteratively adjusting the initial oiling volume and dynamically simulating and analyzing the oil tank oil volume change curve in combination with the complete flight profile until a new global minimum oil volume meets the oil absorption safety margin requirement, determining the volume of the hydraulic oil tank according to the new global maximum oil volume, and adjusting the structural design of the hydraulic oil tank. The scheme improves the efficiency and accuracy of the hydraulic oil tank design.