23results about How to "Stable capture" patented technology

The invention provides an anti-interference measurement method and system for grounding grid parameters in a strong interference environment, and belongs to the technical field of grounding device parameter measurement. The method comprises the following steps: constructing a multi-physics field coupling theoretical model, and integrating a high-precision clock synchronization and wireless communication high-fidelity sensor network; phase and amplitude errors in signal transmission are effectively eliminated through an anti-interference and fidelity processing mechanism, and finally the grounding resistance and related electrical characteristics are calculated by solving a parameter equation set derived from a theoretical model. According to the invention, by constructing a multi-physics field coupling theory model and designing a high-fidelity sensing network, and combining a distributed clock synchronization technology and an anti-interference signal processing mechanism, accurate, portable and safe measurement of the grounding resistance of the tower at a strong interference power transmission line site without power failure and disconnection of a down lead is realized. According to the invention, the risk of operating high-voltage equipment and the complexity of burying the auxiliary electrode in the traditional measurement are thoroughly avoided.

This invention discloses a magnetic solid-phase extraction method for detecting trace amounts of glucocorticoids in hydrophilic media, belonging to the field of sample pretreatment and chromatographic analysis technology. The method includes: adjusting the pH of a hydrophilic medium sample containing glucocorticoids and adding an ionic strength modifier to obtain a pretreated sample; mixing a magnetic covalent organic framework-based composite nanomaterial with the pretreated sample and performing oscillatory extraction under a dynamic gradient magnetic field to adsorb the glucocorticoids onto the material surface; separating the material adsorbed with glucocorticoids using an external magnetic field and discarding the supernatant; performing ultrasonic-assisted elution of the material using an elution solvent and collecting the eluent; and detecting the eluent using high-performance liquid chromatography-diode array detector (HPLC-DID) and calculating the glucocorticoid concentration based on a standard curve.

The invention relates to a three-dimensional optical control method based on a metasurface, and the method comprises the steps: scanning a nano-column unit structure to construct a phase library, selecting high-efficiency half-wave plate nano-columns with phase coverage from-pi to pi from the phase library for the construction of the metasurface, cooperatively introducing a geometric phase and transmission phase regulation and control mechanism, and combining spatial multiplexing arrangement to construct a metasurface. Independent regulation and control of different circular polarization light fields under different wavelengths are realized, so that four switchable optical control focuses are generated in a three-dimensional space; by switching the polarization state or the working wavelength of the incident light, stable capture and controllable transfer of the particles are realized in the space. Compared with the prior art, the method has the advantages that three-dimensional optical control over particles is achieved in an on-chip structure, the control freedom degree is high, and the control adjusting mode is simple and efficient.

A large language model quantization method based on channel arrangement and activation adaptive smoothing belongs to the technical field of large language model compression and efficient inference. In order to solve the problem that the existing post-training quantization method is difficult to simultaneously consider the difference of different channel activation outliers, the group mode quantization has large dynamic range in the group, and the low bit weight and high bit activation alignment efficiency is low, which leads to the problem of precision decline and inference efficiency limitation, the scheme calibrates the maximum activation amplitude of the input channel by statistical data, divides the significant and non significant activation outlier channels according to the percentile threshold, and respectively uses the logarithmic domain and power function domain smoothing strategy to generate channel smoothing scaling factor, and fuses it into the model parameter to extract the weight multidimensional feature to construct the channel quantization sensitivity index and execute the input channel rearrangement, and further generate the joint quantization scaling factor to realize the integer domain alignment inference. It is suitable for low bit compression deployment and efficient inference scene of large language model in resource limited environment.

The invention discloses a high-precision optical isolator reverse loss detection device, which belongs to the technical field of optical element manufacturing, and comprises an equipment bracket, one side of the bracket is detachably connected with a reverse optical path connecting pipe, and the other side of the bracket is detachably connected with a forward optical path connecting pipe. Notch parts are arranged on the reverse light path connecting pipe and the forward light path connecting pipe, depolarization beam splitting prisms are detachably connected to the position, close to the upper portion, of the forward light path connecting pipe and the position, close to the lower portion, of the reverse light path connecting pipe, moving assemblies are installed on the notch parts, and clamping assemblies are arranged on the moving assemblies. The optical isolator is clamped in the clamping assembly, through the matching design of the forward light path and the reverse light path, cross detection and mutual verification of two groups of data can be achieved through the forward light path and the reverse light path, and accidental errors of a single light path are effectively eliminated.

The application discloses a blood flow velocity measurement method based on a light flow method area tracking, relates to the blood flow velocity measurement technical field, and comprises the following steps: extracting multiple frames of blood vessel images from a blood vessel video, and determining a ROI area in each frame of the blood vessel images according to a predetermined ROI area division rule; performing movement estimation on adjacent two frames of the blood vessel images by using a DIS light flow method to obtain blood flow data of the adjacent two frames of the blood vessel images; performing area tracking on the blood flow data by using an affine transformation to obtain affine transformation parameters of each frame of the blood vessel images; inputting the affine transformation parameters into a preset movement amount prediction model, predicting movement amounts of ROI areas of multiple groups of adjacent frames of the blood vessel images, and performing velocity estimation according to the multiple groups of the movement amounts to output blood flow velocities; the application performs movement estimation on the adjacent two frames of the blood vessel images by using the DIS light flow method, improves the response capability of a system to dynamic changes of blood flow in a complex blood vessel structure, and thus ensures accurate measurement under different blood vessel conditions.

The application discloses a liquid level height automatic calculation method and system based on spectral characteristics, a medium and equipment, and particularly relates to the technical field of container liquid level automatic calculation and dynamic monitoring. The method improves the algorithm through spectral curve change point detection, accurately captures the spectral feature difference between gas phase and liquid phase through spectral sequence modeling, introduces the running length index and dynamic updating mechanism, filters out the real gas-liquid interface change point through the multiple check mechanism to filter the false change point, infers the type according to the mean difference before and after the change point and the running length change, takes the clear mutation point or the midpoint of the gradual transition zone as the interface, combines the pixel resolution and the bottom starting position, and obtains the actual liquid level height through formula calculation, establishes the conversion relationship between the pixel quantity and the actual liquid level height through the proportional relationship between the image axis pixel number and the absolute height of the container, and finally takes the average value of the calculation results of three wave bands as the absolute height of the liquid level based on the pixel number of the spectral curve feature point.

This invention relates to the field of aerospace Earth observation and remote sensing imaging technology, and discloses a target-driven dual-camera heterogeneous payload dynamic cooperative aiming method and device. The device includes a base, a wide-area search unit, a high-precision staring unit, a field-of-view deflector, and a cooperative controller. The wide-area search unit and the high-precision staring unit are distributed forward and backward along the flight direction. The cooperative controller identifies potential targets and calculates their geographical locations. It calculates the delay time based on the physical distance and real-time velocity. The cooperative controller calculates the track-crossing swing angle based on the target's position and attitude and sends commands. The field-of-view deflector drives the high-precision staring unit to deflect to the track-crossing swing angle within the delay time. The cooperative controller triggers imaging at the end of the delay time. This invention achieves pointing adjustment and exposure synchronization through heterogeneous payload spatiotemporal coordination and a delay time window, realizing wide-area detection and high-resolution fine imaging of potential targets during a single flight.

This invention discloses an end effector for unmanned aerial vehicles (UAVs), comprising: a robotic gripper module, a motion drive module, and a power output module. The motion drive module is connected to the robotic gripper module, and the motion drive module is also connected to the power output module. The robotic gripper module includes a fixed base and multiple robotic grippers. Each robotic gripper includes a rear gripper piece, a front gripper piece, and a connecting chain. The rear gripper piece and the front gripper piece are hinged together by a connecting pin to form a motion joint. A protective cover is provided outside the motion joint. The connecting chain sequentially connects the fixed base, the rear gripper piece, the motion joint, and the front gripper piece. The end effector of this invention is lightweight and compact, allowing it to be mounted on a UAV. The encased shell greatly reduces interference, occupies little space, and is adaptable to various application environments. The robotic gripper of this invention utilizes the elastic potential energy of a spring during retraction, requiring no energy consumption and effectively extending the device's operating time.

This invention belongs to the fields of bioinformatics and computational biology, specifically relating to a protein stability mutation prediction system and method based on structure and sequence synergy. The structure prediction model outputs structural prediction values ​​based on structural features, while the sequence prediction model outputs sequence prediction values ​​based on sequence features. The fusion prediction module employs a mean fusion strategy to perform decision-level fusion of the structure and sequence prediction values, obtaining the final protein stability change prediction result. This invention adopts a framework of parallel and independent modeling of structural and sequence branches, and finally integrates the outputs of the two branches using a prediction-level mean fusion strategy to achieve complementary information fusion. Experimental results show that this method outperforms existing methods under various mutation scenarios, can reliably capture biophysical constraints in different directions during stability changes caused by mutations, and can effectively improve the accuracy and robustness of protein stability prediction.

This invention discloses a water-land image segmentation method based on local standard deviation texture features, comprising: first, preprocessing the input waterfront image to reduce noise and computational load; second, quantizing texture features using local standard deviation to distinguish between textured real land and smooth water surfaces; third, generating candidate water surface masks using an adaptive Otsu threshold, and combining morphological operations to remove noise and fill holes; finally, statistically analyzing the percentage of water surface pixels row by row to locate the land-water boundary and outputting a high-precision water surface segmentation result. This invention uses local standard deviation as a measure of texture complexity to solve the problem of misjudgment of water reflections, and combines the percentage of water surface pixels to locate the land-water boundary. It features unsupervised operation, strong robustness, and accurate boundaries, and is suitable for strong reflections and complex shoreline scenes such as rivers and lakes, providing reliable technical support for water environment monitoring and unmanned vessel navigation.

The invention discloses an animal delivery automatic counting method and system based on image recognition and threshold adjustment, and the method comprises the steps: firstly collecting a delivery segment video in real time, and obtaining a video detection result through employing a target detection model in combination with a bounding box similarity algorithm; performing preliminary segmentation on the detection frame sequence by adopting a preset interval value, and calculating an interval threshold value based on the interval between the candidate segments; then identifying abnormal short interval segments by using the extracted interval threshold, and merging the abnormal short interval segments to adjacent segments to generate an effective childbirth segment set corresponding to one actual childbirth event; calculating a confidence coefficient dynamic threshold value based on the local sliding window average and a preset minimum confidence coefficient, and setting continuous frame requirements according to confidence coefficient level grading; and finally, in combination with a delivery section division result, a confidence coefficient dynamic threshold value and a continuous frame requirement, carrying out validity judgment on each candidate delivery section, and carrying out statistics on effective delivery times so as to calculate the total number of cubs. The practicability and the intelligent level of the automatic counting system for animal delivery are remarkably improved.

The utility model provides a self-adjusting wind turbine generator set pitch bearing crack monitoring device, which comprises an outer bearing, an inner bearing, bearing holes and an external control device, the inner bearing rotates in the outer bearing, the bearing holes are arranged in the inner bearing and the bearing holes in an annular array, and the wind turbine generator set pitch bearing crack monitoring device is used for monitoring the pitch bearing crack of the wind turbine generator set. The position of the detector can be dynamically adjusted according to the diameter and the distance of the bearing hole through sliding fit of the driving plate and the telescopic plate of the adjusting mechanism, when the size of the bearing hole changes, the external control device can drive the driving plate to slide in the first fixing plate, and meanwhile the telescopic plate stretches out and draws back synchronously, so that the detector is accurately aligned with the bearing holes with different hole diameters. Besides, the size of the detection window is matched with the distance between the bearing holes, a second fixing plate can be elastically extruded through a buffer pad in cooperation with an adjusting structure of a threaded rod and a bolt, it is ensured that the upper surfaces and the lower surfaces of the bearings of different sizes are stably clamped, and the problem that a traditional device cannot be matched with the bearing holes of multiple sizes due to a fixing structure is solved.

The application belongs to the technical field of catalyst adsorption energy value prediction, and particularly relates to a surface relaxation catalyst adsorption energy prediction method based on a graph neural network, which comprises the following steps: constructing a data set based on a molecular graph; constructing an adsorption energy prediction model, wherein the adsorption energy prediction model is established based on an isometric graph neural network, and the adsorption energy prediction model comprises a feature distribution layer, an embedding layer, a plurality of connected Transformer layers, a scalar gate jump fusion module and a third normalization layer connected in sequence, and the scalar gate jump fusion module is used for feature fusion by using the features of initial layers, intermediate layers and final layers; constructing a multi-objective loss function in a training process, updating the parameters of the adsorption energy prediction model by using the multi-objective loss function; and inputting a test set into the trained adsorption energy prediction model to output an adsorption energy prediction result. The application saves the time for calculating the energy of a binary alloy base material and the energy of an adsorbate by DFT, and improves the prediction accuracy of H adsorption energy values.

This invention discloses a high-precision optical isolator reverse loss detection device, belonging to the field of optical component manufacturing technology. The device includes an equipment bracket, on one side of which a reverse optical path connecting tube is detachably connected, and on the other side of which a forward optical path connecting tube is detachably connected. Notches are provided on the reverse and forward optical path connecting tubes. Depolarizing beam splitters are detachably connected near the top of the forward optical path connecting tube and near the bottom of the reverse optical path connecting tube. A moving component is installed in the notch, and a clamping component is provided on the moving component. The optical isolator is clamped inside the clamping component. Through the design of the forward and reverse optical paths, cross detection can be achieved by using two sets of optical paths, one forward and one reverse. The two sets of data are mutually verified, effectively eliminating the random errors of a single optical path.

This invention relates to the field of fluorescence lifetime imaging, and discloses a wide-field fluorescence lifetime imaging system and method based on a streak camera. The system includes an excitation source, a wide-field excitation module, a synchronous detection module, and a synchronous matching module. The wide-field excitation module is positioned along the transmission path of a pulsed laser. The synchronous detection module includes a collection and beam splitting component, a reference camera, and a streak camera component. The input end of the collection and beam splitting component receives the fluorescence beam emitted by the sample, and the output end of the component splits the fluorescence beam into a first detection optical path and a second detection optical path. The reference camera is positioned in the first detection optical path, and the streak camera component is positioned in the second detection optical path. The synchronous matching module is electrically connected to the excitation source, the reference camera, and the streak camera component, respectively. This invention avoids the complex optical encoding scheme and resulting fluorescence photon loss problems of traditional DMD systems, and eliminates the need for complex reconstruction algorithms, thus improving imaging efficiency.

The application provides a particle trapping device and method based on a full dielectric super surface, the particle trapping device based on the full dielectric super surface comprises a medium substrate layer and a resonance layer on the surface of the medium substrate layer, the resonance layer comprises a plurality of resonance units arranged in an array, the medium substrate layer is made of quartz material, the resonance units are made of silicon material, the resonance units are standard cylinders, the height of the standard cylinder is 2.5-3.1 times the radius of the bottom surface of the standard cylinder, and the array period of the resonance units is 2.7-3.3 times the radius of the bottom surface of the standard cylinder. The particle trapping device and method based on the full dielectric super surface can realize high Q value and stable particle trapping under the influence of different incident angles and processing errors.

This application discloses a non-contact particle grasping device and method based on focused vortex light. The device includes a laser, an attenuator, a beam expander and collimator, a half-wave plate, an achromatic cemented lens, a superlens, a focusing and collimator, a dichroic mirror, an objective lens, a sample stage, and an imaging mechanism arranged sequentially along the optical path. The laser light generated by the laser passes sequentially through the attenuator, the beam expander and collimator, the half-wave plate, and the achromatic cemented lens to reach the superlens. The superlens converts the laser light into topologically adjustable focused vortex light, which then passes sequentially through the focusing and collimator, the dichroic mirror, and the objective lens to reach the sample stage and grasp the particles on the sample stage. The imaging mechanism records the grasping process of the focused vortex light on the particles. This device can use the optical force of vortex light to grasp and manipulate tiny particles non-contactly, performing high-precision and complex manipulation of individual particles in three directions, avoiding damage to the sample caused by high-power lasers.

This invention discloses an integrated probe for air-induced ultraviolet fluorescence detection. A detachable snap-in filter (1) is installed at the light inlet of the front end of the cylindrical metal shield shell. A photomultiplier tube (2) is installed close to the back of the filter. A photomultiplier tube base (3) is set at the bottom of the photomultiplier tube (2). A photomultiplier tube base fixing plate (4) is set at the bottom of the photomultiplier tube base (3). The photomultiplier tube base fixing plate (4), the high voltage divider plate (7), the signal processing readout and power supply board (8), and the hollow fixing plate (9) are connected by support columns (6) and fixed with support column fixing screws (5) on both sides. This invention solves the technical problem that existing photon counting detection equipment is difficult to balance low cost, low noise, and compact structure in the application scenario of ultraviolet fluorescence remote sensing induced by radioactive nuclides in the air.

The invention discloses a flexible sensor array based on transient ground voltage. The flexible sensor array comprises a shell which is made of a flexible polymer material and is adaptive to a curved surface structure of a cable terminal surface; the plurality of sensor units are arranged on the surface of the shell at intervals and detect TEV signals on the surface of the cable terminal in a capacitive coupling mode; and one end of the signal cable is electrically connected with the sensor unit on the shell, and the other end of the signal cable is connected with the signal processing module and the wireless communication module at the rear end, so that real-time transmission and remote monitoring of partial discharge signals are realized. The flexible polymer material is adopted to prepare the sensor array, so that the sensor can be tightly attached to the semi-conductive curved surface of the cable terminal, efficient, low-cost, stable and real-time online partial discharge monitoring of the cable terminal is realized, and the safety and reliability of a power system are improved.

The application discloses an ion cage connected covalent organic framework compound and a preparation method and application thereof. The preparation method comprises the following steps: mixing an iron source, an oxime base derivative, a boric acid base derivative, a first solvent and a second solvent, and performing a liquid nitrogen freezing-pump air extraction-freezing cycle degassing treatment, then sealing and performing a condensation reaction to prepare a metal chelate connected covalent organic framework compound; and performing acid treatment on the metal chelate connected covalent organic framework compound to prepare the ion cage connected covalent organic framework compound. The preparation method provided by the application has the advantages of easy raw material, mild conditions, simple operation, and the like, and the prepared covalent organic framework compound has a clear cavity structure and a periodic arrangement of coordination sites, exhibits excellent chemical stability and regular channel characteristics, and has a wide application prospect in the fields of ion adsorption separation, heterogeneous catalysis and the like.

This invention discloses a traffic flow prediction method based on graph structure feedback and spatial prior. The method includes: acquiring target area data; embedding features into historical traffic flow time series data to generate an initial spatiotemporal feature matrix; inputting the initial spatiotemporal feature matrix into a temporal self-attention module to calculate the global dependencies between different time steps in the time dimension and outputting time-encoded features; constructing a bidirectional adjacency matrix through graph structure spatial modeling to output preliminary spatial features; performing spatial linear feedback processing on the preliminary spatial features and time-encoded features to obtain spatial linear feedback features; combining the spatial linear feedback features with spatial prior constraints and aggregating local neighborhood information through graph convolution operations with shared parameters to obtain spatial prior constraint features; and outputting traffic flow prediction values ​​for future preset time steps through a linear mapping layer based on the spatial prior constraint features. This invention achieves high-precision traffic flow prediction.

The application relates to the technical field of bale unpacking, and provides a bale unpacking device. The bale unpacking device comprises a rack, a dismounting roller rotatably arranged on the rack, a movable frame slidably arranged on the rack and spaced apart from the dismounting roller, and a cutting mechanism arranged on the movable frame. The dismounting roller is provided with a piercing member extending outward in the radial direction, and the end of the piercing member is provided with a hook. The bale can be lapped on the dismounting roller and the movable frame, the end of the piercing member can be inserted into the bale and penetrate the net belt on the bale, and the cutting mechanism can cut off the net belt. The bale unpacking device disclosed by the application can drive the bale to rotate and push the movable frame to move by means of the rotation of the dismounting roller, so that the bale without support falls below the rack, the piercing member is inserted into the bale and penetrates the outer net belt of the bale, and the cutting mechanism cuts off the net belt. The net belt can be hooked on the dismounting roller, so that the separation of the forage and the net belt is realized, and the efficiency of bale unpacking is improved.