60results about How to "Improve extraction quality" patented technology

The invention relates to the technical field of intelligent scene recognition, and discloses an intelligent scene recognition method and equipment for an intelligent interaction control unit, and the method comprises the steps: carrying out the collection and preprocessing of an operation sequence, video content and audio data of the intelligent interaction control unit in a digital set-top box, and obtaining a standard multi-mode data set; inputting the standard multi-modal data set into a multi-branch deep neural network model, and extracting time sequence, visual and audio features through parallel processing to obtain a fusion scene feature vector; carrying out dimensionality reduction and semantic mapping on the fused scene feature vector to obtain a compressed semantic feature vector; performing similarity calculation on the compressed semantic feature vector and a predefined scene query vector to obtain a scene matching score; and based on the scene matching score, integrating the output of the plurality of sub-classifiers to obtain a target scene classification result, and the method can be efficiently operated on the digital set top box with limited resources to realize real-time scene recognition.

The invention provides an extraction device containing a penthorum chinense pursh extract and a beverage, and relates to the field of food processing. According to the extraction device containing the penthorum chinense pursh extract, through the design of a closed-loop transmission system, continuous feeding, extraction and discharging of penthorum chinense pursh materials are achieved, a driver provides a stable power source through a bevel gear transmission mechanism, it is ensured that the materials continuously move and are stirred in an extraction pipe, and the extraction efficiency is improved. Penthorum chinense pursh materials can be continuously fed and extracted without shutdown treatment. Compared with the prior art, continuous extraction is achieved, the production efficiency is improved, the operation process is simplified, human intervention is reduced, the ultrasonic extraction effect is enhanced, the extraction efficiency is improved, real-time monitoring and feedback are achieved, and it is ensured that the extraction process is stable.

The invention relates to the technical field of industrial sewage treatment, in particular to an industrial sewage treatment method and system based on big data, and the method comprises the following steps: collecting sewage samples discharged from multiple industrial sources, obtaining chemical component data, carrying out the initialization arrangement of the types and contents of recoverable resources in the samples, and forming a resource feature list. According to the method, through detailed data analysis and implementation of an optimization process, the extraction efficiency and quality of recoverable resources in industrial sewage are remarkably improved, a sewage sample is collected from an industrial source and subjected to chemical component analysis, data support is provided for a subsequent resource recovery strategy, resource classification and recovery are more accurate, and the resource recovery efficiency is improved. The recovery priority is determined by constructing an economic evaluation model, the environmental benefits and the economic benefits are effectively integrated, the intelligent degree of industrial sewage treatment is enhanced, and the resource recovery efficiency and quality are remarkably improved through a real-time adjustment mechanism.

The utility model relates to the technical field of rambutan polysaccharide extraction, and particularly discloses a device for extracting polysaccharide in rambutan pulp. The device comprises a base plate and a box body arranged on the upper side of the base plate, a feeding port is formed in the upper side of the box body, an arc-shaped partition plate is fixedly connected to the inner side wall of the box body, a material guiding port is formed in the lower side of the arc-shaped partition plate, and a smashing mechanism is installed in the area, located between the feeding port and the arc-shaped partition plate, in the box body; a pair of baffles is arranged between the material guiding opening and the centrifugal barrel, adjusting mechanisms for adjusting the baffles to be opened or closed are installed on the two sides of the box body, the upper sides and the lower sides of the baffles abut against the arc-shaped partition plate and the centrifugal barrel respectively when the baffles are closed, a plurality of through holes are formed in the side wall of the centrifugal barrel, and an extrusion mechanism is installed below the centrifugal barrel. According to the rambutan pulp polysaccharide extraction device, the crushing mechanism, the adjusting mechanism, the centrifugal barrel and the extruding mechanism are arranged, so that multiple functions of crushing, centrifuging, extruding and the like are integrated, and efficient and continuous extraction of rambutan pulp polysaccharide is realized.

The invention relates to the field of functional food and pharmaceutical preparations, and discloses an active multienzyme preparation with spleen and stomach strengthening effects. The solid beverage is prepared from the following components in parts by weight: 10 to 20 parts of radix astragali seu hedysari extract, 5 to 10 parts of rhizoma atractylodis macrocephalae extract, 10 to 20 parts of poria cocos extract, 5 to 10 parts of pericarpium citri reticulatae extract, 5 to 10 parts of fructus crataegi extract, 5 to 10 parts of malt extract, 5 to 10 parts of alpha-amylase, 5 to 10 parts of protease, 2 to 5 parts of cellulase, 2 to 5 parts of pectinase, 2 to 5 parts of lipase, 10 to 15 parts of prebiotics, 2 to 5 parts of xylitol and 3 to 8 parts of microcapsule wrapping materials. The compound preparation with high stability and multiple functions is prepared by scientifically combining traditional Chinese medicine extracts and multiple active enzymes, combining with reasonable addition of prebiotics and auxiliary materials, adopting a micro-capsule wrapping technology to protect enzyme activity and optimizing traditional Chinese medicine extraction, component mixing and forming processes.

The invention relates to the field of agricultural pest control, and discloses a composite pest control agent based on a plant extract as a core. Comprising 15%-25% of honeysuckle extract, 10%-20% of podophyllum hexandrum extract, 15%-25% of camellia peel extract, 10%-20% of wikstroemia indica extract, 10%-20% of senecio scandens extract, 10%-20% of lycoris radiate extract, 1%-3% of a synergist, 5%-10% of an emulsifier and a solvent. According to the invention, a plurality of plant extracts are fused, and effective prevention and control of a plurality of fungi, bacterial diseases and pests are realized through accurate optimization of the ratio. Compared with a current technology which only adopts a single plant extract, the method has the advantages that the comprehensive prevention and treatment efficiency is remarkably improved, the problems of limited prevention and treatment range and poor multi-pest control effect are successfully solved, and the blank in the research and application field of the compound plant extract prevention and treatment agent is filled.

The invention discloses a supply chain management optimization method and system based on a large model. The method comprises the steps of data acquisition and processing, large model demand prediction, inventory management optimization, supply chain operation optimization and supply chain management optimization. The invention relates to the technical field of supply chain management, in particular to a supply chain management optimization method and system based on a large model, according to the scheme, process design of demand prediction, inventory management optimization and operation optimization is sequentially carried out, and complex integrated supply chain data can be better processed; a long-short-term memory neural network combined with a transformer model is adopted to carry out large model demand prediction, and the processing performance of time sequence trend data is optimized; inventory management is carried out by adopting a deep enhanced hybrid linear programming method combined with graph convolution enhancement, and the extraction quality of transport dynamic edge features is improved by adding an aggregation embedding mode of node features and edge features; and a generation bidirectional encoder combined with a transformer model is adopted, so that the intelligence of the method is improved.

The utility model relates to an ultrasonic-assisted extraction device which comprises a tank body, a shell is movably mounted at the top of the tank body, an ultrasonic generator is movably mounted in the shell, an ultrasonic vibration rod extending into the tank body is fixedly mounted at the bottom of the ultrasonic generator, and an adjusting mechanism is arranged in the shell. Auxiliary mechanisms are arranged on the two sides of the tank body; the adjusting mechanism comprises a motor, the motor is fixedly installed on the front inner wall of the shell, a circular rotating disc is fixedly installed at the position of an output shaft of the motor, and a fixing rod is fixedly installed between the left inner wall and the right inner wall of the shell. According to the ultrasonic-assisted extraction device, by arranging the adjusting mechanism, the ultrasonic generator can move left and right, and then the ultrasonic vibration rod is driven to move left and right in the tank body, through the design, ginseng and sugar in the tank body can be uniformly extracted, the situation that a super blank area appears is effectively avoided, and therefore the extraction efficiency and quality are improved.

The invention belongs to the technical field of food processing, and particularly relates to an extraction process of turnip seed oil and an effective component composition of the turnip seed oil. The invention relates to an extraction process of turnip seed oil, which comprises the following steps: performing staged fermentation pretreatment on turnip seed broken cell walls by using a plurality of microbial strains, and then extracting grease by combining with a nano SiO2 (at) SDS composite active agent and ionic liquid synergistic ultrasonic treatment. Through staged fermentation pretreatment and the synergistic effect of the nano material and the ionic liquid, the extraction efficiency and quality of the turnip seed oil are greatly improved; firstly, cell walls and internal components are gradually degraded by utilizing aspergillus niger, aspergillus oryzae, bacillus licheniformis, lactic acid bacteria and saccharomycetes, grease is released, and the stability of the grease is improved; then, the extraction process is further optimized by a composite active agent formed by nano silicon dioxide and lauryl sodium sulfate and 1-butyl-3-methylimidazolium acetate, and the efficiency is enhanced through ultrasonic treatment;

The invention discloses a biased binocular visual odometer method and device based on strawberry cuttlefish bionics. The biased binocular visual odometer method comprises the following steps: synchronously acquiring a bright field image and a dark field image collected by a biased binocular camera and inertial data of an inertial measurement unit; constructing a perceptual attention weight grid according to the bright field image and the dark field image, and performing brightness normalization processing on the bright field image and the dark field image; extracting lateral visual features based on the normalized image and the perceptual attention weight grid; and fusing the biased visual features with the inertial data, and solving through nonlinear optimization to obtain a carrier pose estimation result. According to the invention, the dynamic range of the binocular vision system is expanded, and the imaging and sensing capabilities of the system in an HDR scene are improved. And meanwhile, the positioning precision and robustness of the binocular visual odometer in an HDR scene are improved.

The present application belongs to the technical field of industrial internet security, and relates to an industrial internet vulnerability library establishment method, which solves the problems of low construction quality, weak traceability, insufficient continuous evolution ability and difficulty in supporting deep security application of the existing vulnerability library. The present application obtains industrial internet multi-source vulnerability data and external feedback data, obtains extraction results and initial confidence based on multi-source vulnerability data by adopting rule extraction and remote supervision deep learning joint extraction, divides the certainty knowledge base and the to-be-reviewed queue according to the double threshold after the confidence is optimized by the graph convolution network, analyzes the external feedback data of the samples in the to-be-reviewed queue as the instant reward signal, optimizes the sampling strategy and updates the model through deep reinforcement learning, extracts triples from the certainty knowledge base to build a knowledge graph and generate a version hash chain regularly, and obtains a structured vulnerability knowledge base containing a hash chain and confidence evaluation. The present application realizes high-precision construction and dynamic optimization of the vulnerability library.

The application discloses a handwriting standard monitoring method and system based on image enhancement, relates to the technical field of image processing and intelligent evaluation, and comprises the following steps: obtaining a handwriting character image through adaptive light compensation of multiple light sources; after image preprocessing such as geometric correction and adaptive binarization, a convolutional neural network is used to extract character skeleton topological structure features for structural standard analysis; meanwhile, a pen tip motion trajectory is extracted based on an optical flow method and time sequence standard analysis is performed; after the structural and time sequence bimodal features are fused, a scoring model constructed based on a style transfer network is input to calculate a standard score; and a hierarchical error correction report is automatically generated according to the score grade and deviation type. The technical problems of incomplete stroke extraction, insufficient structural analysis precision, lack of dynamic writing feature modeling and lack of hierarchical error correction feedback under the complex illumination and shielding conditions in the prior art are solved.

The invention relates to the technical field of soil sampling, and discloses a mining area soil rapid sampler based on coal mining safety monitoring, which comprises a soil sampling tube, a rotating tube arranged on the top of the soil sampling tube, handles arranged on the two sides of the rotating tube, and a soil sampling unit arranged in the soil sampling tube, the soil sampling unit comprises a communicating component arranged at the top of the soil sampling pipe and a soil sampling component arranged at the bottom of the communicating component; the soil sampling part is used for pushing the soil sampling pipe to rotate to sample soil in the downward pressing process of the rotating pipe, and the communicating part is used for communicating the soil sampling pipe with external air in the soil sampling process of the soil sampling part, so that soil can enter the soil sampling pipe conveniently. According to the soil sampling device, the soil sampling pipe can be continuously pushed to rotate in one direction through the arranged soil sampling component, rotary soil sampling of the soil sampling pipe can be achieved without repeatedly rotating the rotating pipe, the labor intensity of soil sampling personnel is reduced, and the soil sampling efficiency is improved.

The present invention discloses a keyword / sequence extraction method through feature set dimensionality reduction, including processing a text set to obtain a feature set; obtaining a vector representation of the feature set; reducing the dimension of the feature set to obtain a feature set after dimensionality reduction and a mapping relationship between the feature set and the feature set after dimensionality reduction; keyword / sequence extraction. The present invention can not only improve the quality of keyword / sequence extraction, and reduce the computing power and memory overhead required for vector calculation in the keyword / sequence extraction model training process of massive data, but the dimensionality reduction algorithm module therein can also be well applied to the dimensionality reduction of high-dimensional feature sets (similarity cannot be calculated at one time).

The invention relates to the technical field of text classification and grading evaluation, in particular to a specific text classification and grading evaluation system and method based on an information element extraction technology. The method comprises the steps that an extraction unit performs key element field extraction on text data based on a large model fine tuning technology; the standardization unit carries out standardization processing on each extracted element field; and the evaluation unit constructs a text classification and grading evaluation system through a decision tree based on the normalized fields. According to the technology, the accuracy and the coverage rate of key element field extraction can be improved through the large model fine tuning technology and the model, the model can be helped to better understand the context in a text, so that the situations of false report and missing report are reduced, the extraction quality is improved, the fine tuning technology can enable the model to better adapt to a new data set, and the extraction efficiency is improved. Even if the data set is slightly different from the pre-training data set, good performance can be kept, and the robustness of the model facing texts of different styles or formats can be enhanced.

The application discloses a kind of low-illumination condition fast moving target multi-view cooperative tracking method and device, the method includes low-illumination condition fast moving target multi-view tracking algorithm and unmanned aerial vehicle cluster cooperative tracking algorithm;Multi-source environment perception is carried out using millimeter wave radar sensor, RGB vision sensor and infrared vision sensor carried by unmanned aerial vehicle;Multi-source perception data is normalized in time and space dimensions by three space-time heterogeneous alignment algorithm;Multi-source fusion perception data after space-time normalization is input into multi-source fusion perception feature calculation network to calculate the world coordinates of the tracked target under the current frame;The calculation result is compared with the calculation result of last frame, to judge the change of the position and moving direction of the tracked target in world coordinate system;The world coordinate information of the tracked target is shared through cluster communication, and the trajectory of unmanned aerial vehicle cluster is formulated and updated using multi-agent reinforcement learning, to realize unmanned aerial vehicle cluster cooperative tracking.

The utility model provides a biopharmacy liquid medicine extraction device, including: box and extraction cylinder, the bottom of said box is movably connected with the rotating column through the sealing bearing, the lower surface of box is fixedly connected with the main motor, the main motor output shaft is fixedly connected with the rotating column through the connecting piece, and the upper surface of rotating column is fixedly connected with the extraction cylinder, the outer side surface of the upper end of the extraction cylinder is fixedly connected with a positioning ring, the positioning ring is slidably connected into a positioning groove formed in a limiting ring, the outer side surface of the limiting ring is symmetrically connected with fixing plates, and the ends, away from the limiting ring, of the fixing plates are fixedly connected with the inner side surface of the box body. Through cooperation of the extraction barrel, the rotating column, the air cylinder, the pressing plate, the stirring plate and the main motor, the extraction device can conduct corresponding medicine smashing treatment on medicinal materials in the extraction device, liquid medicine left in the medicinal materials can be correspondingly extruded out through extrusion, and therefore the extraction effect of the extraction device is improved.

The utility model discloses a low temperature boiling pot for honeysuckle chlorogenic acid extraction, including the body of pot, the pot cover, heating device, stirring device, filter device and control device. The bottom of body of pot is equipped with the heating cavity of adaptive cone and the heating tube of helical distribution in the heating cavity, is equipped with the openable flap, the observation window and the exhaust hole on the pot cover, and the stirring device adopts the scraper type stirring structure, and the stirring structure is composed of the stirring main shaft, the supporting arm and the elastic scraper, is provided with multiple groups of radial extension supporting arms on the stirring main shaft, and the end of each supporting arm is equipped with the elastic scraper of the adhesion body of pot inner wall surface, the bottom of body of pot is equipped with the discharge gate and double -deck filter device, and control device can accurate control heating temperature, boiling time, stirring speed parameter etc. This boiling pot can even heating, promote the even heating of medicinal liquid, filter impurity and accurate temperature control, effectively avoid honeysuckle pasting and chlorogenic acid high temperature degradation, improve chlorogenic acid extraction efficiency and quality.

The invention discloses a pressure adjusting method for tea component separation, and relates to the technical field of tea processing. Comprising the steps of obtaining a first extraction adjustment coefficient according to a relation between extraction pressure and extraction target quantity; measuring the concentration of the extract through a sensor to determine a second extraction adjustment coefficient; comparing the two extracted adjustment coefficients with a preset target value, and judging whether pressure adjustment is needed or not; if the extraction adjustment coefficient is not within the preset range, automatically adjusting the extraction pressure through a pressure adjustment algorithm in combination with a multi-index comprehensive scoring model; the device can automatically adjust the extraction pressure to meet the requirements of different types of tea leaves and extraction targets, so that the extraction efficiency and the product quality of target components are improved, the limitation of traditional fixed pressure setting is solved, the extraction efficiency and the product quality are improved through real-time monitoring and automatic adjustment of the extraction pressure, and the product quality is improved. And the process parameters can be automatically adjusted according to the characteristics and extraction targets of different tea leaves in the extraction process.

The utility model discloses a traditional Chinese medicine purification separator which comprises a supporting leg I, a shell and a supporting leg II, the number of the first supporting legs is four, the top ends of the four first supporting legs are fixedly connected with the lower surface of a heating barrel, an electric heating wire is arranged in an interlayer of the inner wall of the heating barrel, and an inclined table is arranged on the bottom wall of the heating barrel; the shell is fixedly connected to the top end of the heating barrel, a stirring mechanism is arranged in the shell, and the bottom end of the stirring mechanism extends into the heating barrel; the top ends of the four second supporting legs are fixedly connected with the lower surface of a cooling barrel, a discharging hopper is arranged at the bottom end of the cooling barrel, a condensation pipe is arranged in an interlayer of the inner wall of the cooling barrel, and a connecting pipe is installed between a connecting opening in the middle of the upper surface of the cooling barrel and a connecting opening in the right end of the upper surface of the heating barrel through a flange. According to the traditional Chinese medicine purification separator, traditional Chinese medicine materials and solvents can be fully mixed, substance transfer between the traditional Chinese medicine materials and the solvents is accelerated, and the extraction quality and purity are improved.

The utility model discloses a distillation and extraction device for preparing wormwood essential oil, relates to the technical field of distillation and extraction, and aims to solve the problems that in the prior art, raw materials are suspended through a material containing barrel, so that the raw materials are in full contact with steam, but the raw materials are still accumulated in the material containing barrel, so that the middle parts of the accumulated raw materials are in limited contact with the steam; and the distillation effect is influenced. A separation plate is fixedly connected to the lower end in the distillation device, a material containing barrel is arranged at the position, at the upper end of the separation plate, in the distillation device, a stirring shaft is rotationally connected to the middle in the material containing barrel, a plurality of stirring rods are fixedly connected to the two sides of the stirring shaft, and a steam generator is connected to one side of the distillation device; the other side of the distillation device is provided with a condensing device, and the lower end of the condensing device is connected with an oil-water separator, so that automatic stirring of a wormwood raw material in the material containing barrel is realized, accumulation of the raw material is effectively prevented, and full contact between steam and the wormwood raw material is ensured.

The invention relates to the technical field of spectral analysis, in particular to a pulsed electric field pretreatment method for extracting water-soluble dietary fibers from peanut shells, which comprises the following steps: acquiring spectral data of peanut shell extracts pretreated under different pulsed electric field intensities, and dividing absorption peaks in the spectral data into characteristic peaks and non-characteristic peaks; calculating a characteristic difference parameter between the same characteristic peaks in any two spectral data and a non-characteristic difference parameter between any same non-characteristic peaks in any two spectral data, and further obtaining a difference index between any two spectral data; performing hierarchical clustering by combining difference indexes among the spectral data, and calculating classification indexes of a hierarchical clustering tree; and acquiring the pulse electric field intensity corresponding to the maximum classification index as the optimal pulse electric field intensity. According to the method, the optimal electric field intensity can be accurately found, and the extraction quality of the water-soluble dietary fibers of the peanut shells can be further improved.

The invention provides moutan bark extraction equipment for preparing Jindan attached extending particles, which comprises an extraction assembly for spirally stirring a mixed solution, a slag removal assembly arranged at the bottom of the extraction assembly and communicated with the extraction assembly, an electromagnetic valve at least partially embedded in one side of the bottom of the extraction assembly, and a slag collection assembly arranged in the extraction assembly and close to the slag removal assembly, through spiral stirring of the extraction assembly, a mixed solution of a cortex moutan raw material and an ethanol solution is driven to generate eddy current, so that the cortex moutan raw material and the ethanol solution are fully stirred and contacted, and the extraction quality is improved; cortex moutan dregs in the mixed solution can be guided to gather towards the dreg collecting assembly at the center through the eddy current characteristic, and after the dregs are uniformly collected through the dreg collecting assembly, the dreg removing assembly conducts internal circulation filtration on the solution in the extracting assembly, and the effect of efficiently collecting the dregs is achieved.

The invention relates to a road alignment safety evaluation method based on three-dimensional point cloud spatial modeling, and aims to overcome the defects of poor adaptability of a road surface point cloud segmentation algorithm and strong dependency of vertical line fitting on a road marking in a complex scene by utilizing characteristics of obtained high-density point cloud data. The road key information collaborative extraction method comprises the following steps: firstly, reducing the influence of environmental noise on road surface features through a normalized gradient filtering algorithm and a statistical outlier filtering algorithm; secondly, constructing a joint clustering model and an interactive region growing algorithm considering elevation, normal vector and pavement material reflection heterogeneity, and realizing precise segmentation of pavement point cloud in a complex scene; and then extracting a road surface edge line, obtaining road center line data, and generating a safety evaluation result list based on road center line parameters of a road plane, a longitudinal section and a cross section line shape. The research results provide technical support for application scenes such as road maintenance detection and automatic driving high-precision map construction.

The invention discloses a preparation method, system and application of cordyceps sobolifera active ingredient powder for glomerular injury repair. The preparation method comprises the following steps: S1, performing low-temperature freeze drying treatment on cordyceps sobolifera; s2, the crushed cordyceps sobolifera is sieved by a 120-mesh sieve, and cordyceps sobolifera powder is obtained; s3, mixing the cordyceps sobolifera powder with a buffer solution containing enzyme according to a solid-liquid ratio of 1g: 10ml, carrying out enzymatic hydrolysis for 2-4 hours under the hydrolysis conditions that the pH value is 4.5-5.5 and the temperature is 37-50 DEG C, and continuously stirring at the speed of 100-300rpm during the enzymatic hydrolysis; s4, after hydrolysis is completed, heating to 80-90 DEG C for 5 minutes; s5, filtering the extracting solution; s6, concentrating the extracting solution by using vacuum evaporation; and S7, packaging the extracting solution in a sealed container by using a sterile technology, and storing under a refrigeration condition of 4-8 DEG C. The cordyceps sobolifera active ingredient powder disclosed by the invention has the capability of repairing damaged glomerulus, and is suitable for preparing medicines for various types of chronic kidney diseases, especially those related to glomerular injury.

The present invention discloses an image classification method based on Resnet50 combined with an attention mechanism and a feature pyramid in the field of image processing technology, comprising the following steps: S1, collecting input images and performing data preprocessing on the images; S2, importing the preprocessed images into a classification model, and outputting feature maps of three different convolutional layers; S3, performing classification prediction on each feature map separately, taking a classifier with the maximum credibility, and outputting a final predicted classification result. The present invention has the advantages of high image classification accuracy, high extraction quality, and the like.

This invention discloses an aspect-level sentiment analysis method based on a dual-branch graph structure, belonging to the field of natural language processing technology. Addressing the shortcomings of existing technologies such as semantic noise diffusion, sensitivity to syntactic parsing errors, and difficulty in capturing long-distance sentiment cues, this invention employs a parallel modeling architecture with semantic and syntactic dual branches. The semantic branch constructs a dynamic semantic graph through aspect-aware multi-head attention, filters noise using Top-K sparsity and nearest-neighbor constraints, and extracts semantic features via GATv2. The syntactic branch fuses the initial dependency syntactic graph and the learnable latent syntactic graph, adaptively fuses them using a gating mechanism, and then extracts syntactic features using GCN. The features from both branches are then fused and the sentiment polarity is classified and output. This invention effectively filters semantic noise, corrects syntactic errors, accurately captures sentiment cues, and significantly improves the model's classification accuracy and robustness.

The invention relates to the technical field of coffee production, in particular to a spiral powder distributor and a powder uniform distribution method thereof. The powder distribution device comprises a top suspension support, a transverse linear driving mechanism is installed on the lower side of the top suspension support, a vertical longitudinal support is arranged below a driving sliding block of the transverse linear driving mechanism, a longitudinal linear driving mechanism is arranged on the longitudinal support, and a sliding block of the longitudinal linear driving mechanism is connected with a powder distribution support. A powder distribution head for uniformly distributing coffee powder and a powder distribution motor for driving the powder distribution head to rotate are mounted on the powder distribution bracket; the coffee powder distributing device is simple in structure, the powder distributing head is driven to move into the lower brewing device through double-shaft action, coffee powder can be evenly distributed through rotation of the powder distributing head, the coffee powder is evenly distributed, the surface of the coffee powder is flat, and therefore the extraction quality is improved; and the device can be additionally arranged on the existing equipment without the equipartition function, so that the production effect is improved.

The present invention discloses a method and device for identifying transformer faults by fine-tuning fitting based on text mining, and belongs to the technical field of transformer fault identification. The method for identifying transformer faults by fine-tuning fitting based on text mining of the present invention, during data preprocessing, segments the transformer fault text data, and performs data enhancement processing at the same time, which can effectively avoid semantic information loss on the one hand, and provide as many features as possible to the main model in a limited sample volume on the other hand, so that the fault description text with a small number of samples in the transformer fault type accounts for a larger proportion when the model calculates the fitting loss value, avoiding overfitting and enhancing the model recognition effect; linear dimensionality reduction processing is performed through the fully connected layer to obtain a global semantic vector; the Canopy+K-Means model is used to cluster all calculation results, identify the specific fault location and fault type of the transformer fault text data, and improve the overall semantic understanding ability, and the scheme is scientific, reasonable and feasible.

The utility model relates to the technical field of traditional Chinese medicine extraction and filtration, and discloses a traditional Chinese medicine decoction piece multi-layer extraction and filtration device which comprises a processing table, an extraction barrel is fixedly connected to the left side of the top of the processing table, a rotating motor is fixedly installed at the bottom of the extraction barrel, and the output end of the rotating motor is fixedly connected with a T-shaped installation frame. A fine filter cylinder is arranged at the top of the T-shaped mounting frame, a coarse filter cylinder is arranged at the bottom of the inner side of the fine filter cylinder, a fixing bolt penetrates through the bottom of the inner side of the coarse filter cylinder, and the tail end of the fixing bolt sequentially penetrates through the bottom of the inner side of the coarse filter cylinder and the bottom of the inner side of the fine filter cylinder and is in threaded connection to the center of the top of the T-shaped mounting frame. According to the utility model, the coarse filter cylinder and the fine filter cylinder are driven by the rotating motor to rotate, so that the extracting solution is mixed with the traditional Chinese medicine, and the effects of increasing the contact area of the traditional Chinese medicine and the extracting solution, filtering the extracting solution in multiple layers, improving the extracting efficiency and quality of the traditional Chinese medicine and obtaining purer extracting solution are realized.