29 results about "Particle type" patented technology

The invention discloses a radiation particle identification method based on an improved ResNet-18 network and a CIS transient response technology, and the method comprises the steps: collecting a CIS dark field image in a radiation environment, and generating a multi-modal data set according to neutron, proton and heavy ion radiation experiment sample images; a transient response image data set is obtained through theoretical calculation based on radiation analog simulation software; preprocessing the transient response image data set, and marking particle type, energy and angle information of each sample image as a label of multi-task learning; a pre-trained ResNet-18 model is used, an input layer is modified to adapt to the image size, and a channel attention module is added; and carrying out compression quantification on the model by adopting an optimizer in combination with gradient cutting through a self-adaptive category weight adjustment strategy. By constructing the intelligent feature extraction network, the system realizes automatic identification of transient response geometric features, and breaks through the bottleneck that a traditional method depends on artificially defined features.

The invention provides a cyclotron electrostatic deflection plate beam loss heat source reconstruction method, and relates to the technical field of particle accelerator thermal analysis and multi-physics field coupling calculation, and the method comprises the steps: 1, configuring beam dynamics simulation parameters, multi-particle tracking is carried out under a three-dimensional field graph containing a central area electromagnetic field, a harmonic coil magnetic field and an electrostatic deflection electric field, and a particle loss list is generated; step 2, transferring the particle loss list into a beam source file of a Monte Carlo radiation transport program, configuring Monte Carlo parameters including a deflection plate material, a boundary, a particle type and an energy spectrum, executing particle transport simulation, and obtaining a three-dimensional energy deposition matrix on an electrostatic deflection plate element; according to the method, high-resolution heat source reconstruction is realized, the peak temperature rise calculation error of the deflection plate can be reduced, a temperature field solving result can be fed back to optimize the particle source, and a reliable basis is provided for thermal structure design and cooling optimization of the electrostatic deflection plate.

The invention provides a method for improving accuracy of glass spontaneous explosion risk early warning, and relates to the technical field of glass processing.The particle type, the particle diameter and the depth of particles in the glass thickness direction are recognized through a high-power microscope, in addition, chemical components of the particles can be analyzed through an EDS, then heat treatment experiments are conducted on the particles, and the accuracy of glass spontaneous explosion risk early warning is improved. The relationship between glass self-explosion and particle type, size and depth direction is obtained, and the glass meeting the self-explosion condition can be selected before tempering, so that the situation that the glass is used on the wall after flowing to the next process is avoided, and the situation that the glass is split into sharp fragments during self-explosion after being used on the wall and possibly causes casualties or property losses is avoided; and particularly, the risk is obvious in scenes such as high-rise building curtain walls and bathroom partitions.

The invention provides an adaptive multi-angle scattering ratio five-wavelength aerosol detection method. The adaptive multi-angle scattering ratio five-wavelength aerosol detection method comprises the following steps: S1, acquiring a multi-wavelength scattering light intensity signal; s2, scattering ratio parameter calculation; s3, classifying particle types; s4, angle-wavelength cooperative adjustment is carried out; and S5, fire alarm judgment. According to the method, different particle morphological characteristics are adapted through angle-wavelength dynamic cooperative adjustment, classification robustness is enhanced in combination with multi-algorithm fusion, the accuracy of fire early warning is finally improved, the false alarm rate is reduced, and the method is suitable for aerosol safety monitoring of scenes such as buildings and industrial places.

A system comprises a particle sensor for a vehicle comprising a light source that directs a beam to an interrogation region outside the vehicle; and an optical detector that receives scattered or reflected light from an aerosol particle in the interrogation region. A processor performs a method for particle classification comprising measuring an optical response from a single particle in the interrogation region by identifying an optical response peak for scattered or reflected light from the interrogation region; measuring an amplitude and duration of the optical response peak; correcting the duration of the optical response peak based on a vehicle airspeed; based on the measured amplitude of the optical response peak, generating an expected optical response duration for water droplet using a calibration table; analyzing the measured optical response duration and the expected optical response duration to determine particle classification; and generating classification probabilities for one or more particle types.

The invention relates to the technical field of visual detection of solid waste particles, in particular to an intelligent identification method and system for solid waste particles based on multi-modal fusion, and the method comprises the steps: firstly, carrying out the particle segmentation and particle size statistics of a solid waste infrared image, and calculating a particle adhesion evaluation value; dividing the solid wastes into low, medium and high adhesion grades; then, for different adhesion levels, channel weighted fusion guided by infrared images, or frequency domain fusion guided by RGB images, or instance level alignment fusion based on three-dimensional point cloud space anchor points is adopted to obtain multi-modal fusion features; and finally, the infrared image and the multi-modal fusion feature are mapped to a three-dimensional point cloud model to construct a three-dimensional solid waste particle model, accurate geometric dimensions and material features are obtained at the particle instance level, and the accuracy of solid waste particle type recognition and particle size measurement and the overall recognition efficiency are remarkably improved.

The application relates to a detection method and device for the concentration of particles in a cleaning device, a cleaning device and an apparatus. The method comprises the following steps: obtaining a target voltage value of a detection signal; when the target voltage value of the detection signal is less than a standard voltage value of a standard signal, determining a particle type corresponding to the target voltage value of the detection signal according to a difference between the standard voltage value of the standard signal and the target voltage value of the detection signal; obtaining cumulative values of at least one difference corresponding to the particle type in a preset period; and determining the concentration of the particle type according to the cumulative values, the maximum concentration and / or the minimum concentration of the corresponding type. The different particle concentrations can be accurately and stably obtained by analyzing and processing the data of different particle types in a certain period.

The invention relates to a simulation method for the influence of a radiation environment on the signal transmission characteristics of a microsystem device, and the method comprises the following steps: S1, constructing a three-dimensional model of the microsystem device, and splitting the three-dimensional model into independent grid files according to the types of materials; s2, setting a particle type, energy and an incident region in Geant4, simulating particle bombardment, and outputting single-particle energy deposition data, energy deposition distribution and linear energy transfer (LET) data; s3, establishing a material damage model in MATLAB, and calculating the post-radiation conductivity sigma (t) and relative dielectric constant epsilon r based on particle and substance action mechanism analysis; and S4, updating the material parameters calculated in the step S3 to a device model in the HFSS, simulating a signal transmission characteristic curve in a radiation environment, and analyzing an influence mechanism of radiation on signal transmission. The electrical performance of the micro-system device can be accurately evaluated in different radiation environments, a scientific basis is provided for design and optimization of the micro-system device in a high-radiation environment, and the anti-radiation capability and reliability of the device are effectively improved.

The application provides a simulation prediction method and system for slow positron source intensity in a research reactor, comprising the following steps: S1, reading arrangement information of components in a reactor core of the research reactor; S2, constructing a Monte Carlo model of the reactor core containing a slow positron source device, and obtaining neutron multi-group angular flux and photon multi-group angular flux information near the slow positron source generating device; S3, constructing a slow positron Monte Carlo model; S4, performing energy-angle double sampling to obtain an incident neutron energy probability density distribution function f n,ig,ia , an angle probability density distribution function f n,ia , an incident photon flux energy probability density distribution function f g,ig,ia , an angle probability density distribution function f g,ia , and a neutron-photon weight factor; S5, performing function sampling of particle types, energy and angle; and S6, obtaining slow positron distribution information generated in the slow positron generating device. The application accurately considers the whole life cycle of slow positron generation and transportation through multi-stage coupling, and realizes accurate prediction of slow positron source intensity.

The invention discloses an adjustable-frequency steel rail vibration absorption structure, device and method based on the particle damping technology, and the adjustable-frequency steel rail vibration absorption structure based on the particle damping technology comprises a fastening piece which is fixed to a steel rail through a fixing piece, and the fastening piece is provided with a vibration absorption face making contact with the steel rail; the two foundation mass blocks are symmetrically arranged on the two sides of the fastening piece through cantilever pieces; and the two damping bins are arranged at the ends, away from the fasteners, of the basic mass blocks correspondingly, storage spaces are formed in the damping bins, and the storage spaces are used for containing vibration absorption particles. The inherent frequency of the steel rail vibration absorption device is adjusted by accurately regulating and controlling the mass, rigidity and damping parameters of the vibration absorption structure and adjusting the type and filling rate of particles in the cylinder, effective energy dissipation is carried out on steel rail vibration of a corresponding frequency band generated by interaction between wheel rails, and the effects of restraining steel rail corrugation and reducing wheel rail noise are achieved.

The invention belongs to the technical field of aero-engines, particularly relates to an oil abrasive particle detection device and method for a fuel oil system of a marine gas turbine, and aims to solve the problems of insufficient abrasive particle detection information and easy secondary pollution in the prior art. The device comprises a backwashing system and an abrasive particle separation and detection system, wherein the backwashing system realizes deep cleaning of a filter element through an ultrasonic massage box and a heat exchanger, and prevents secondary pollution of particles at wrinkles by adopting silicone oil paper; the abrasive particle separation system adsorbs ferromagnetic abrasive particles through a high-gradient magnetic field, and the image acquisition system shoots abrasive particle images through a transflective double-color light source and an automatic focusing objective lens. The method comprises the following three steps: firstly, separating abrasive particles through ultrasonic massage vibration and spray backwashing, then depositing the abrasive particles through a magnetic field, and finally, identifying the type of the abrasive particles through image processing and judging the wear degree. The method has the advantages of simplicity and convenience in operation, high detection precision and capability of qualitatively and quantitatively analyzing the abrasive particles, and is suitable for monitoring the wear state of a fuel system.

The invention discloses a particle classification and size measurement method and device based on interference imaging and improved UNet + + +, and belongs to the field of particle measurement and image processing. The method comprises the following steps: acquiring a spherical particle interference fringe pattern and an irregular particle interference speckle pattern, and constructing a data set containing a mask pattern and a category label; dividing into a training set, a verification set and a test set; the training set is used for training a pre-constructed neural network model, the pre-constructed neural network model takes an original UNet + + + network as a basis, a space and channel attention module, a frequency domain dynamic convolution unit and a multi-axis Hadamard convolution attention module are introduced, and the network is trained to output a reconstructed image and a category label at the same time; and inputting a to-be-measured particle image into the trained network, outputting a particle category and a reconstructed image, and measuring the reconstructed image to obtain size information. According to the method, high-precision, real-time and end-to-end prediction of particle types and sizes is realized, and the method is suitable for a mixed scene of spherical droplets and irregular ice crystals in cloud.

A system comprises a particle sensor for a vehicle comprising a light source that directs a beam to an interrogation region outside the vehicle; and an optical detector that receives scattered or reflected light from an aerosol particle in the interrogation region. A processor performs a method for particle classification comprising measuring an optical response from a single particle in the interrogation region by identifying an optical response peak for scattered or reflected light from the interrogation region; measuring an amplitude and duration of the optical response peak; correcting the duration of the optical response peak based on a vehicle airspeed; based on the measured amplitude of the optical response peak, generating an expected optical response duration for water droplet using a calibration table; analyzing the measured optical response duration and the expected optical response duration to determine particle classification; and generating classification probabilities for one or more particle types.

The invention discloses a radiation detection method and device, and belongs to the technical field of nuclear radiation measurement. According to the radiation detection method, pulse signals output by a plurality of photomultipliers can be converted into digital signals; multiple paths of first signals can be processed to extract pulse amplitude information, first integral processing is performed on the pulse amplitude information to determine a first integral value, and second integral processing is performed on the pulse amplitude information to obtain a second integral value; generating a two-dimensional distribution diagram according to the first integral value, the second integral value and the pulse amplitude information; and classifying the data points in the two-dimensional distribution diagram based on the identification curve so as to distinguish at least two particle types. According to the invention, the charge integration method is adopted to discriminate multiple particles, the waveform is subjected to two kinds of integration processing to obtain the first integration value and the second integration value, the peak value of the waveform is extracted, the integration mode is adopted to extract the characteristics of the whole waveform, the waveform acquisition error is reduced, and the particle separation precision is improved.

The invention discloses a transformer oil pollution degree evaluation method based on lensless imaging and machine learning, relates to the technical field of transformer oil detection, and is used for solving the problems that in the existing transformer oil particle pollution detection technology, the three-dimensional morphology of particles cannot be obtained, and a detection result and the pollution degree are lack of quantitative correlation. The method comprises the following steps: S1, obtaining a lens-free diffraction image of particles in a transformer oil sample to be detected; S2, reconstructing the amplitude of the diffraction image; S3, reconstructing the phase of the diffraction image and inverting the thickness; S4, segmenting and extracting a particle region; particle diffraction information is acquired in a lens-free imaging mode, synchronous acquisition of two-dimensional morphology and three-dimensional thickness information of particles is realized by combining amplitude reconstruction and thickness inversion, and automatic identification of particle types is realized by combining structured feature extraction with a machine learning model. And intelligent and quantitative evaluation of the pollution level of the transformer oil is realized.

The invention relates to the technical field of particle physical detection and high-energy density physical diagnosis, in particular to an electron positron ion spectrometer, which comprises a collimation unit, an aiming unit and a shell, and is characterized in that a magnetic field dispersion unit, an electric field separation unit and a detection unit are arranged in the shell; a magnetic field area is formed on one side, opposite to the collimation unit, of the magnetic field dispersion unit so as to perform energy expansion on incident charged particles along a first direction; the electric field separation unit is orthogonally arranged with the first direction and distinguishes particle types along the second direction; the detection unit adopts a curved surface structure conformal with dispersion geometry, so that particles are incident approximately vertically and aberration is suppressed, and a two-dimensional pattern representing energy in an arc length direction and representing particle types in an arc height direction is obtained. Through the synergistic effect of the magnetic field dispersion unit and the electric field separation unit, clear spatial separation of positive electrons, protons and electrons in a wide energy region can be realized; and the detection unit adopts a curved surface structure conformal with dispersion geometry, so that the accuracy of wide-energy-region energy spectrum data can be ensured.

This invention discloses a method, system, and storage medium for identifying the morphological features of lubricating oil abrasive particles, relating to the field of abrasive particle morphology recognition technology. The steps are as follows: acquiring a high-resolution abrasive particle image; performing preliminary feature extraction on the abrasive particle image using a pre-processing multi-channel encoding component to obtain preliminary features; further extracting the preliminary features using deep learning to obtain deep features; labeling the abrasive particle type and marking the saliency of the abrasive particle features; inputting the deep features with saliency markings into a multi-head self-attention mechanism for processing to obtain key features; performing multi-scale feature fusion on the key features using a BiFPN network, and combining it with the abrasive particle type feature map to achieve online detection of multi-scale abrasive particles. This invention can achieve high-precision automatic identification of various abrasive particles with good real-time performance, meeting the needs of intelligent operation and maintenance of mechanical equipment.

The application provides a cyclotron electrostatic deflection plate beam loss heat source reconstruction method, and relates to the technical field of particle accelerator heat analysis and multi-physical field coupling calculation, and the method comprises the following steps: step 1, configuring beam dynamics simulation parameters, performing multi-particle tracking under a three-dimensional field diagram containing a central area electromagnetic field, a harmonic coil magnetic field and an electrostatic deflection electric field, and generating a particle loss list; step 2, converting the particle loss list into a beam source file of a Monte Carlo radiation transport program, configuring Monte Carlo parameters, including deflection plate materials, boundaries, particle types and energy spectra, performing particle transport simulation, and obtaining a three-dimensional energy deposition matrix on the electrostatic deflection plate element. The application realizes high-resolution heat source reconstruction, can reduce the calculation error of the peak temperature rise of the deflection plate, and can feed back the temperature field solving result to optimize the particle source, thereby providing a reliable basis for the thermal structure design and cooling optimization of the electrostatic deflection plate.

The application relates to a transformer oil pollution degree evaluation method based on lens-free imaging and machine learning, relates to the technical field of transformer oil detection, and is used for solving the problems that the existing transformer oil particle pollution detection technology cannot obtain particle three-dimensional morphology and the detection result lacks quantitative correlation with the pollution degree. The application comprises the following steps: S1, acquiring a lens-free diffraction image of particles in a transformer oil sample to be detected; S2, amplitude reconstruction of the diffraction image; S3, phase reconstruction and thickness inversion of the diffraction image; S4, particle region segmentation and extraction; S5, structured feature extraction; S6, particle classification; and S7, pollution degree grade evaluation. The application realizes synchronous acquisition of particle two-dimensional morphology and three-dimensional thickness information by acquiring particle diffraction information in a lens-free imaging mode, combining amplitude reconstruction and thickness inversion, realizes automatic identification of particle types by combining structured feature extraction with a machine learning model, and realizes intelligent and quantitative evaluation of the pollution degree grade of transformer oil.

The invention provides a method and a system for simulating and predicting the intensity of a slow positron source in a research reactor. The method comprises the following steps: S1, reading arrangement information of components in a reactor core of the research reactor; s2, constructing a research reactor core Monte Carlo model containing a slow positron source device, and obtaining neutron multi-group angle flux and photon multi-group angle flux information near a slow positron source generation device; s3, constructing a slow positron source Monte Carlo model; s4, performing energy and angle double sampling to obtain incident neutron energy probability density distribution functions fn, ig and ia, angle probability density distribution functions fn and ia, incident photon flux energy probability density distribution functions fg, ig and ia, angle probability density distribution functions fg and ia and neutron and photon weight factors; s5, function sampling of particle types, energy and angles is carried out; and S6, acquiring slow positron distribution information generated in the slow positron generation device. According to the method, the generation of the slow positron and the full life cycle of transportation are accurately considered through multi-stage coupling, and accurate prediction of the intensity of the slow positron source is realized.

The invention discloses a suspension particle size monitoring and dynamic regulation and control method, which adopts a polarizing microscope to monitor the dissolution state of an auxiliary material in a suspension, combines with a laser particle analyzer to obtain particle size distribution data, and adjusts production process parameters. The method specifically comprises the following steps: step 1, observing the particle size and form of each original sample by a polarizing microscope; step 2, preparing a suspension according to a process formula, sampling in the preparation process, and observing the particle size, shape, distribution condition, agglomeration and caking phenomenon and polarization state of particles in the suspension by using a polarizing microscope; step 3, judging the type of the particles in the sample by comparing the particle size and form of the original sample in the step 1; observing the dissolution state of the auxiliary materials; and 4, testing the particle size and particle size distribution of the sample. The combined method can help to more scientifically adjust process parameters, eliminate the influence of incomplete dissolution of auxiliary materials on the particle size of the finished product, and significantly reduce the particle size of the finished product.

This invention discloses a method and apparatus for determining particle types, a smoke detector, and an electronic device. The method includes: emitting red and blue light towards a target area, and acquiring a first signal value corresponding to the red light and a second signal value corresponding to the blue light; determining the target particle type of the air in the target area using the first and second signal values. This technical solution solves the problem of inaccurately detecting the particle type of airborne particles.

A system and procedure are provided for collecting and analyzing airborne particles at a user's site. The system includes a blower module for mobilizing dust from surfaces, a sampling system with an integrated vacuum pump for collecting particles on a slide for microscopic examination, and a mobile microscope for analyzing the collected particles. The system also includes an artificial intelligence (AI) module for classifying and counting particle types, and an app connectivity module for wireless communication with a mobile device. The procedure involves using the powerful blower to remove dust from surfaces, collecting particles on a slide, and analyzing the collected particles using the mobile microscope and the AI ​​modules.The system enables the analysis of airborne particles and allows users to detect and monitor particle concentrations in their environment. The system is delivered in a shipping / packaging box.

The invention discloses an extremely early fire detection method and device for a cable tunnel and a computer device. The method comprises the following steps: collecting an air sample in a target cable tunnel; the air sample is placed in a preset electric field area, and the movement speed corresponding to particles in the air sample is obtained; on the basis of the corresponding relation between the preset particle type and the speed range, the particle type in the air sample is judged according to the movement speed corresponding to the particles in the air sample; and repeating the above steps, and determining that the target cable tunnel is in the extremely early stage of the fire disaster under the condition that the continuously detected particle types in the air sample comprise that the number of times of cable pyrolysis particles exceeds a preset threshold value. The technical problems that dust particles, water molecules and cable pyrolysis particles are difficult to distinguish in the extremely early fire detection of the cable tunnel at present, and the judgment accuracy is low due to the interference of the environmental factors are solved.

The present disclosure provides a pencil beam dose calculation method, comprising: acquiring human body image data; determining a plurality of particle types for treatment; calculating an initial pencil beam dose distribution corresponding to each particle type; determining a target action layer of each dose distribution in the human body image data based on the particle type, wherein the target action layer positions corresponding to different particle types are different; adjusting each initial dose distribution based on the target action layer corresponding to each particle type to obtain a particle pencil beam dose distribution corresponding to each particle type; superimposing the particle pencil beam dose distributions corresponding to each particle type to obtain a target pencil beam dose distribution; wherein the dose proportion of the particle pencil beam dose distribution corresponding to the same particle type in the target action layer is greater than the dose proportion of the particle pencil beam dose of the particle type in other parts of the human body image data.

The embodiment of the invention provides a parameter determination method, system and equipment of a feeding device and a storage medium, and belongs to the technical field of agricultural machine design. The method comprises the following steps: acquiring input information, wherein the input information comprises a particle type, an optimization target and a constraint condition of a target particle processed by a feeding device; obtaining particle parameters corresponding to the particle type according to the particle type; generating at least one group of initial structure parameters corresponding to the feeding device according to the particle parameters and the constraint conditions; performing a simulation experiment according to the at least one group of initial structure parameters to obtain experiment data; analyzing the experimental data to obtain an initial performance index corresponding to each group of initial structure parameters; iterative optimization is conducted on the at least one set of initial structure parameters and the at least one set of initial performance indexes according to the optimization target and the constraint condition, at least one set of target structure parameters is obtained, and the parameters of the feeding device comprise the at least one set of target structure parameters. According to the embodiment of the invention, the design and optimization of the intelligent feeding device can be realized.

The present application relates to a kind of particle identification and detection method based on deep learning improved algorithm, it includes the following steps: using electric sensitive area method obtains the electric signal of multiple particle;Extract the time domain, frequency domain and time-frequency domain characteristic parameters of electric signal;Characteristic parameters are combined to enhance the difference between characteristics;Screening characteristic parameters to reduce the amount of operation;Neural network and recurrent neural network model are weighted fusion;Using fusion model to test multiple repeat for the particle to be measured, using voting mechanism to determine the final classification result;The concentration, quantity proportion and particle size distribution of each kind of particle are output.The particle identification and detection method provided by the present application can realize the measurement of particle type, concentration and particle size distribution in water, improve the accuracy, accuracy and stability of water quality monitoring, provide more comprehensive information for water plant operation and environmental monitoring, so that the particle identification and detection method can be better applied to practical application.

The application discloses a reactor core meltdown collector design optimization method based on a least square particle method, which is based on a least square moving particle method in a Lagrange method, has higher simulation accuracy and no space consistency limitation, adds a fluid-solid coupling model (PMS model) for calculating the interaction between a coolant and solid melt, and finally calculates the force between solid particles of the melt in the flow field and between the solid particles of the melt and a solid wall through a solid collision model (DEM method). In addition, an optimized particle slip model is used to relieve the particle particle aggregation effect, the calculation density and dynamic viscosity are processed to improve the calculation stability, and the accuracy of particle type determination is improved through a double detection method of algebraic calculation and geometric calculation. The method is a method with the ability of simulating the fluid-solid coupling problem and the solid collision problem in the reactor core meltdown collector and is used for the design optimization of the reactor core meltdown collector.