59results about How to "Achieve aggregation" patented technology

The present invention relates to a hyperspectral and laser radar joint classification method of a category-aware fusion network, comprising: obtaining hyperspectral images and laser radar images, slicing the hyperspectral images and laser radar images into spectral patches and laser radar patches, and inputting them into feature extraction branches; the feature extraction branches include: spectral branches, spatial branches and elevation branches; the three feature extraction branches extract features simultaneously; and the features are merged using an uncertain category-aware fusion strategy. The beneficial effects of the present invention are as follows: the present invention utilizes the spectral, spatial and elevation features contained in the hyperspectral and laser radar data to recognize each category, and specifically proposes an uncertain category-aware fusion strategy, that is, according to the recognition capabilities of the spectral, spatial and elevation branches for each category, dynamically assigns different weights to the classification confidence, thereby achieving targeted information aggregation, and thus improving the classification accuracy of the overall model.

The invention belongs to the field of natural language processing technology and false news fact verification, and particularly relates to a novel false news detection method of a knowledge graph assistance and attention mechanism, which comprises the following steps: firstly, extracting data from a target news text, identifying a key entity and linking the key entity to a knowledge graph; obtaining an entity embedding vector based on the knowledge graph through a gating mechanism; then, taking sentences, topics and entities in the target text as input vectors, and extracting high-order knowledge representation of the input vectors by using a Transform encoder; then, obtaining deeper text embedding through a multi-head attention mechanism, including sentence-entity and sentence-theme embedding representation; and finally, comparing all the obtained advanced knowledge representations, and inputting the advanced knowledge representations into the model together to predict the truth score of the news, thereby effectively discriminating potential false news reports.

The invention discloses a method for realizing'active ' / controllable free radical polymerization of hydrophilic and / or hydrophobic monomers under an air condition, relates to the field of'active' / controllable free radical polymerization, and aims to solve the problem of narrow monomer selection range in the prior art. A functional monomer comprising a hydrophobic monomer, a halide, copper bromide, a ligand and a nonionic surfactant-GOx are subjected to monomer polymerization in a solvent through an initiator azodiisobutylimidazoline hydrochloride (VA-044), and a polymer with narrow target molecular weight and molecular weight distribution is formed. According to the method, the glucose oxidase modified by the nonionic surfactant is introduced into the'active ' / controllable free radical polymerization reaction, and an effective way for quickly and efficiently synthesizing a target polymer from a hydrophobic monomer under the mild air condition is provided.

The invention discloses a personalized federal learning method and system based on parameter similarity, relates to artificial intelligence, and provides a scheme for solving the problem of poor personalized effect in the prior art. According to the method, key parameter distribution is determined according to the training condition of each client model in each round of training, and similar clients are determined according to the similarity of the key parameter distribution, so that the similar clients have similar training degrees, and key parameter aggregation is realized. According to the expression similarity of the key parameters of the similar clients, the aggregation proportion of the key parameters is further determined, and accurate personalized aggregation of the key parameters is achieved. And the importance value of each parameter is determined according to the number of iterations of a single round, and the position of the key parameter is further determined, so that the accuracy of determining the position of the key parameter is improved. And on the basis of the similar client set of each client, an optimization target of the key parameter personalized aggregation weight is constructed, so that the personalized aggregation model better fits the local data distribution.

The invention discloses a multi-domain constraint emotion recognition system inspired by an adaptive cognitive map and an emotion recognition method, which are applied to the field of electroencephalogram signal processing. According to a traditional emotion classification method, after electroencephalogram signals are subjected to manual feature extraction, key features are screened and input into traditional classifiers such as a support vector machine (SVM), but professional priori knowledge is needed for manual feature extraction; most of the methods mainly focus on feature selection and time information utilization, so that space information is not fully utilized; according to the method, a space-guided dynamic graph constraint learning module is designed, a topological structure is generated in a self-adaptive mode, the function relation of the topological structure is constrained, and complex spatial domain feature extraction overcoming cognitive priori information is achieved; meanwhile, a time-excited sequence attention module is provided, and global time sequence dependence characteristics are captured by paying attention to more important time in the sequence; furthermore, a new multi-domain interaction attention module is designed, traditional static cross-domain interaction is overcome, and constraint of multi-domain difference information and aggregation of complementary information are achieved. Besides, in order to better utilize information migration between different domains, a constraint is introduced to reduce distribution difference between the source domain and target domain features, and similarity and mobility of cross-domain features are improved.

The invention discloses an enterprise knowledge graph inference completion method and inference completion system, and the method comprises the steps: obtaining a to-be-completed knowledge graph of a target enterprise, inputting the to-be-completed knowledge graph into a trained knowledge graph embedding model, and predicting an incomplete triple to achieve the inference completion of the knowledge graph; wherein the knowledge graph embedding model comprises an entity node coding module, a multilayer graph convolutional neural network module, a Softmax pooling layer and a gating neural network based on a self-attention mechanism. According to the method, the knowledge graph embedding model can provide more accurate semantic information for feature interaction, multi-level feature interaction and aggregation of the triple on sub-tensor kernel granularity and global granularity can be realized, and hidden semantics behind entities and relationships of the knowledge graph can be more efficiently mined. Therefore, the knowledge graph completion accuracy, the semantic modeling capability, the interpretability and the like are obviously improved.

The invention discloses a scene-adaptive unmanned aerial vehicle target detection method and system. The method comprises the following steps: inputting an image into a scene classification network to obtain scene features; inputting the image and the scene features into a multi-scene target detection network to obtain a target identification positioning result; specifically, an image is input to an image feature extraction module to obtain a local feature, the local feature and a preset reference point are input to an offset attention module to obtain an offset attention feature, and the offset attention feature is input to a target query positioning module; inputting the scene features and a preset target query vector into a scene feature guiding module to obtain a scene target query vector, and inputting the scene target query vector into a target query positioning module; the target query positioning module performs target detection based on the offset attention feature and the scene target query vector. The unmanned aerial vehicle target detection tasks of different scenes are unified by adopting one model, a detection algorithm is lightened, computing resources are saved, and unmanned aerial vehicle deployment is facilitated.

The invention discloses a cultivated land non-agrochemical change detection method, system and device and a storage medium, and relates to the technical field of cultivated land non-agrochemical change detection, and the method comprises the steps: collecting two-stage image data of a cultivated land non-agrochemical to-be-detected region; constructing a cultivated land non-agrochemical change detection model; inputting the two-stage image data of the to-be-detected area into the cultivated land non-agrochemical change detection model, and performing multi-scale feature extraction on the input sample image data through a multi-scale feature extraction module; the multi-scale features are calibrated through a difference attention module; capturing context information of different scale features in the multi-scale features through a Tr annform encoder; obtaining a new feature map after low-level spatial information and high-level semantic information of different scale features are fused through a Transfer decoder; and inputting the new feature map into a classification prediction module to obtain a cultivated land non-agrochemical change detection result. According to the method, the accuracy of cultivated land non-agrochemical change detection is improved.

The invention discloses a software-defined science and education resource dynamic collaborative recommendation method and system based on a knowledge graph. The method comprises the following steps: constructing an artificial intelligence knowledge graph; obtaining science and education resource data of a science and education institution, extracting knowledge entities in the science and education resource data, and connecting the knowledge entities to the artificial intelligence knowledge graph to obtain a science and education resource knowledge graph; constructing a user preference network model containing graph updating and time sequence feature fusion, and extracting local resource features / user features to obtain local resource features and user features of each client; constructing a local knowledge graph based on the science and education resource knowledge graph, and performing model training by taking the local resource features of each client as initial features until the model converges; and inputting the users and the resources into the graph neural network for prediction, and recommending the resources with high scores to each user. According to the invention, the data source and the coverage range of the recommendation system are greatly expanded, and multi-source science and education resource recommendation in a real sense is realized.

The invention discloses an intelligent Internet of Things data aggregation method without a trusted third party, and relates to the technical field of Internet of Things, a plurality of resource nodes and a resource node management center are arranged, and the types of Internet of Things data collected by the resource nodes are distinguished through a multi-classification fusion method based on fuzzy measurement; each resource node autonomously generates a public key and a private key and encrypts the Internet of Things data to generate a ciphertext, a plurality of aggregation nodes are arranged to aggregate different types of ciphertexts, an operation center identifies the type of the aggregated ciphertext and selects a port to a corresponding service node, the service node decrypts the aggregated ciphertext to obtain aggregated data, and the aggregated data is transmitted to the aggregation node. According to the method, the distinguishing and aggregation of the multi-type Internet of Things data without a trusted third party are realized, the data transmission efficiency is improved, meanwhile, the security problem caused by external attacks and internal collusion attacks is avoided, and the reliability of Internet of Things data transmission protection is improved.

This invention discloses a vertical cigarette packing device and its working method, which solves the problem of large footprint of existing cigarette packing machines. It has the beneficial effects of reducing the footprint of cigarette packing equipment and expanding the application range of cigarette packing machines. The specific solution is as follows: A vertical cigarette packing device includes a frame, which supports a multi-segment transfer mechanism, a cigarette pack flipping mechanism, and a flat plate opening and closing mechanism. The multi-segment transfer mechanism is located at the top of the frame and is connected to a clamping and suction composite clamp. The cigarette pack flipping mechanism is set on the side of the clamping and suction composite clamp. A cigarette pack clamping mechanism is set on the inner side of the frame of the cigarette pack flipping mechanism. The flat plate opening and closing mechanism is set below the cigarette pack clamping mechanism. The cigarette pack flipping mechanism flips several horizontal cigarette packs into a vertical state. The cigarette pack clamping mechanism clamps the vertical cigarette packs. When the flat plate opening and closing mechanism is closed, it supports the bottom of the vertical cigarette packs.

The invention relates to the technical field of grinding devices, and provides a grinding device for the exterior of a wire rod, which can realize single-time full-ring complete treatment of the wire rod, is relatively stable in treatment quality, has an automatic abrasive cleaning function, is relatively low in manual maintenance frequency and relatively good in practicability, and comprises a transmission mechanism and a main body structure, a main body structure comprises a foundation frame, a processing box is fixedly connected in the foundation frame, a threading opening is formed in the processing box and used for threading of wires, a fixed box is fixedly connected to the top end of the processing box, lifting boxes are slidably connected to the front end and the rear end of the processing box, and a plurality of fixed coils and a plurality of fixed conductors are arranged in the fixed box; a plurality of fixed conductors are arranged in the processing box, a plurality of fixed coils are respectively wound on the plurality of fixed conductors, a plurality of lifting coils and a plurality of lifting conductors are arranged in the two lifting boxes, the plurality of lifting coils are respectively wound on the plurality of lifting conductors, the transmission structure is mounted in the foundation frame, and an inclined arc-shaped supporting plate is fixedly connected in the processing box.

This invention discloses a method and apparatus for intrusion detection based on deep learning-based 3D point cloud. The method acquires a point cloud of the scene to be tested; it then performs target recognition on the point cloud using a personnel target detection network to obtain personnel target detection information. The personnel target detection network is constructed based on the Voxel R-CNN framework, and the feature set of the voxel grid is generated based on the position information of the points in the point cloud contained within the voxel grid and the point cloud features. Based on the detection information and the position information of dangerous areas in the scene to be tested, it determines whether an intrusion event has occurred. This invention compensates for the shortcomings of MeanVFE in feature utilization by reconstructing the voxel grid feature set generation method through the personnel target detection network, calculating the offset, and concatenating it with other features. It also achieves the aggregation of local and global information, utilizing feature information at different levels to improve feature expression quality, thus significantly improving detection accuracy.

The invention belongs to the field of biomedical materials, and provides a preparation method of an ultrasonic response nano composite piezoelectric film, which comprises the following steps: preparation of a KNN nanowire precursor, preparation of a KNN nanowire, preparation of a KNN (at) PDA nanowire, preparation of a PCL / KNN (at) PDA solution and preparation of a PCL / KNN (at) PDA film. According to the preparation method, the piezoelectric film is prepared by using the polycaprolactone and the KNN nanowire which are relatively good in biocompatibility and degradability, so that the risk of a secondary operation is avoided; through the PCL / KNN (at) PDA film, regulation and control of electric energy output by adjusting external ultrasonic pulses are realized; the PDA is modified on the surface of the KNN nanowire, so that the aggregation problem in a PCL matrix is avoided.

The present application relates to the field of data processing, and provides a GIS service aggregation method, device, computer equipment and storage medium. The method includes: receiving a GIS service request, the GIS service request carries a target service identifier; determining the GIS associated service of the target service based on the geographic location data of the target service corresponding to the target service identifier; obtaining the first configuration information of the target service and the second configuration information of the GIS relationship service, and based on the first configuration information and the second configuration information, calling the corresponding service access component to obtain the data information of the target service and the data information of the GIS associated service; performing aggregation processing based on the data information of the target service and the data information of the GIS associated service to obtain the data aggregation information of the target service. The above-mentioned aggregation of GIS services can realize the aggregation between the target service and the GIS associated service without the user manually operating multiple services, thereby saving the user's manual operation and responding to the service items initiated by the user in a timely manner.

This invention discloses a method for constructing a dataset through multimodal data fusion and extraction in the field of solid waste disposal. The method includes: first, converting the original document into a sequence of multimodal objects; second, constructing a document-level context index and entity knowledge base, extracting entity relationships, and constructing a global knowledge base and index; next, using the global knowledge base and index, performing semantic classification and dynamic routing on the multimodal object sequence, performing multimodal divide-and-conquer extraction by assembling contextual information packets, and outputting discrete record fragments; finally, generating feature identifiers based on experimental conditions, reconstructing cross-modal data through a confidence priority mechanism, and generating a standard dataset. This invention effectively solves the attention drift and task confusion problems when large language models process complex long documents, overcomes the challenge of entity referencing resolution in cross-paragraph contexts, and overcomes the difficulties of information incompleteness and alignment in multimodal table and image data, significantly improving the quality of dataset construction.

This disclosure pertains to the field of batteries, specifically a method and system for early warning of thermal runaway in automotive power batteries. It involves collecting data on individual battery cell voltage, total voltage, current, and temperature to construct an original feature sequence. This sequence is then fused using multi-head one-dimensional convolution to integrate local temporal features at different time scales, generating a fused feature sequence. Temporal encoding is performed using a gated recurrent unit with an added anomaly response gate. Anomaly scores are calculated based on the degree to which the input features deviate from the historical mean, and the hidden state is updated accordingly. The output consists of a hidden state sequence and anomaly score sequence. A dual-channel attention module is used to calculate temporal and feature attention weights, which are then coupled through a Hadamard product to generate a coupled attention weight matrix. The hidden states are weighted and aggregated to obtain a context vector, which is decoded by a fully connected network to determine the thermal runaway risk probability. This probability, combined with the rate of change, generates an early warning threshold, enabling real-time and accurate thermal runaway early warning triggering.

The application provides a travel service recommendation method and device, an electronic device and a storage medium. The method comprises the following steps: performing regional clustering based on station position information of a traffic station to obtain a plurality of candidate travel region pairs, and constructing a candidate travel chain; obtaining first historical travel data in a preset historical period, and extracting corresponding second historical travel data for a preset operation time interval; for each candidate travel region pair, determining a maximum number of transfers, and obtaining a target historical passenger flow in a continuous historical time interval; combining a preset screening condition to screen a target travel region pair; extracting third historical travel data, generating a travel inappropriateness evaluation index of each operation time interval based on the first and third historical travel data for each target travel region pair, and recommending a travel service to a travel object according to the travel inappropriateness evaluation index. The application can provide personalized traffic services for travel objects with the same travel demand and provide a good travel experience.

The invention discloses a bath heater device, and belongs to the technical field of bath heaters. The bath heater device comprises a shell and further comprises a light source assembly arranged in the middle of the shell and used for emitting parallel light in the horizontal direction and generating heat radiation; the light reflecting assembly is arranged on the side portion of the shell and used for reflecting light emitted by the light source assembly, and the reflection angle is adjustable; the light emitting assembly is arranged at the bottom of the shell, and the light reflected by the light reflecting assembly is emitted to the outside of the shell through the light emitting assembly; the light-emitting assembly comprises a light-condensing light-emitting cover and a light-diffusing light-emitting cover. According to the bath heater, through the mode that the light source assembly emits light in the horizontal direction and then the light is reflected to the light emitting assembly through the light reflecting assembly to be emitted, the overall thickness of the bath heater can be reduced to a certain degree, and therefore the problem that in the prior art, a bath heater lamp is too thick is solved; the effect of changing the heat radiation area and position can be achieved, and the functionality and applicability of the bath heater are improved.

The invention relates to the technical field of power systems, in particular to a fine adjustment resource power plant dynamic aggregation packaging method based on high-dimensional model expression, and the method comprises the steps: carrying out the aggregation of feasible regions of different power generation factors, and obtaining a feasible region aggregation model of the power generation factors; the distributed load resources are aggregated into virtual units, aggregation and scheduling of power plant dynamic aggregation packaging scheduling objects are dynamically optimized based on power generation characteristics and power generation economy, and dynamic construction and aggregation of the power plant dynamic aggregation packaging scheduling objects are achieved; the mapping relation between the physical parameters and the aggregation model parameters is established through the high-dimensional model expression technology, and real-time online identification of the aggregation unit external characteristic parameters based on the high-dimensional model expression technology is achieved. According to the method, resource aggregation and optimization can be dynamically carried out in real time on the premise of ensuring the calculation efficiency and precision, unified regulation and optimal utilization of flexible resources are ensured, and quick response can be realized under different scheduling requirements.

The application provides a magnetic field control system and method for magnetic micro-cluster multi-branch navigation, which comprises a control module and a magnetic field generation module. The control module obtains a first moving magnetic field gradient, a split magnetic field gradient and a second moving magnetic field gradient through the relationship among the magnetic field gradient, the size and the speed of the magnetic micro-cluster, and obtains the coil current value of the magnetic field generation device through the relationship among the coil current value, the position of the magnetic micro-cluster and the magnetic induction intensity; the magnetic field generation module is configured to generate a uniform rotating magnetic field based on the initial magnetic induction intensity to drive the magnetic micro-particles to gather and form the magnetic micro-cluster, navigate the magnetic micro-cluster to the intersection point of different branch positions based on the first moving magnetic field gradient, split the magnetic micro-cluster into at least two sub-magnetic micro-clusters based on the split magnetic field gradient, and navigate each sub-magnetic micro-cluster to the different branch positions based on the second moving magnetic field gradient, so that the application realizes synchronous targeted drug delivery for multi-branch lesions.

This invention discloses a method for domesticating mandarin fish fry, comprising the following steps: 1) After releasing the fry, feed them naturally for one day, then stop feeding for one day, and then feed them live bait fish at a fixed point on the bottom of the pond for 1-2 days; 2) Feed them live bait fish and dead bait fish in a certain proportion, and feed them together for 2-3 days, feeding them once a day at sunset; during feeding, the proportion of dead fish is 5%-50% and gradually increases; 3) Feed them together with live fish, dead fish and feed for 2-3 days, feeding them once a day, gradually increasing the proportion of dead fish from 50% to 100%, while the amount of feed is 0.5-2.5% of the total weight of the mandarin fish in the pond, and gradually increasing, with the total amount of feed being 2.5-3.5% of the total weight of the mandarin fish in the pond; at the same time, continuously raise the feeding height to the water surface; 4) Stop feeding the bait fish and feed them with pure feed on the water surface, and domestication is completed after 1-2 days. This invention can increase the domestication rate of mandarin fish in a short time and can be widely applied to the artificial breeding of mandarin fish.

The invention provides a new energy power plant centralized monitoring system time period reporting system, and relates to the technical field of new energy power generation, and the system comprises a time period analysis device which carries out the analysis processing of a time period selection instruction received by a monitoring human-computer interface, and forms a target time period parameter set. And the data screening device performs screening processing on the original operation data and the event records corresponding to the target time period to form a target information set, and the screening processing adopts a field white list mode. According to the time period reporting system of the centralized monitoring system of the new energy power plant, original operation data are accurately analyzed and screened through the synergistic effect of a plurality of modules such as the time period analysis device, the data screening device and the state arrangement device, a target information set is formed, and aggregation and arrangement of equipment-level state evolution sequences are achieved. By combining the event records of different sources, aligning the time labels and intensively arranging the cross-station data, the efficiency and accuracy of data processing are improved.

In reactive polycarbonates, there is a problem of poor reactivity in the crosslinking reaction. Furthermore, existing reactive polycarbonates that require minor components such as polymerization initiators tend to be unsuitable for various applications and have poor versatility. The above problems are solved by the polycarbonate resin of the present invention, wherein the polycarbonate resin comprises: structural units (A) derived from monomers represented by the following general formula (1) and / or structural units (B) derived from monomers represented by the following general formula (2); and structural units (C) derived from monomers represented by the following general formula (3), wherein the total proportion of structural units (A) and (B) is 0.5 to 25 mol% relative to the total molar number of structural units (A) to (C) constituting the polycarbonate resin, and the proportion of structural unit (C) is 75 to 99.5 mol%. R in formulas (1) to (3) 11 ~R 38 As shown in this application specification.

This invention provides a data spatial query method, apparatus, electronic device, and computer-readable storage medium. The data spatial query method provided by this invention includes: acquiring target business data; constructing a spatial field based on the latitude and longitude data of the target business data; importing the spatial field into a preset index; acquiring the latitude and longitude data of two diagonal vertices within a visualized map area; determining a data query range based on the latitude and longitude data of the two diagonal vertices; acquiring a geographic grid based on the data query range and the shrinkage ratio of the visualized map; and aggregating and statistically analyzing the target business data within the geographic grid area based on the spatial field to obtain query results. The data spatial query method provided by this invention can aggregate points on a large number of maps and respond promptly to the aggregation.

The application provides a dynamic priority charging and discharging scheduling method and system based on community V2X vehicle network interaction, obtains vehicle battery state, travel regularity, community load and power grid state data through V2X communication, jointly evaluates vehicle dynamic priority and scheduling time window, and forms power cell division based on load demand and vehicle adaptability. Then, the conflict is resolved by combining the deterministic travel plan, an emergency adjustment set is generated, power optimization distribution is completed within the power grid safety boundary, and a scheduling instruction is output. Thus, the aggregation, evaluation, grading and optimized scheduling of community electric vehicle energy storage resources are realized, the electric vehicle cluster is converted into a controllable distributed energy storage system under the premise of guaranteeing the travel demand of the vehicle owner, which not only supports the power grid and stabilizes the fluctuation, but also creates benefits.

The application provides a pyridine amine ligand-scandium complex, a preparation method thereof and application thereof in catalytic preparation of polyisobutylene, and belongs to the technical field of catalysts.The pyridine amine ligand-scandium complex provided by the application has a structure shown in any one of formula I to formula III.The pyridine amine ligand (pyridine imine or pyridine amine) is used to complex with trivalent scandium, the structure of polyisobutylene and the reaction rate can be better controlled by changing the skeleton of the ligand, and the polymerization of isobutylene can be better realized.The results of examples show that, by using the pyridine amine ligand-scandium complex as a catalyst and by using a cocatalyst, the polymerization reaction of isobutylene can be realized at room temperature of 25 DEG C, and low-molecular-weight polyisobutylene is obtained.

The shoe sole supporting piece is used for supporting the shoe sole and comprises a plurality of first supporting parts, a plurality of second supporting parts and a plurality of first supporting arms, and the first supporting parts and the second supporting parts are suitable for being connected with the lower surface and the upper surface of an inner cavity of the shoe sole correspondingly. The first supporting arm is connected between the first supporting part and the second supporting part; on a projection plane perpendicular to the vertical direction, projections of at least three adjacent second supporting parts are evenly distributed around the projection of part of the first supporting parts, the first supporting parts are connected with the adjacent second supporting parts through the first supporting arms, projections of at least three adjacent first supporting parts are evenly distributed around the projection of part of the second supporting parts, and the first supporting parts are connected with the adjacent second supporting parts through the second supporting arms. And the second supporting part is connected with each adjacent first supporting part through the first supporting arm. The shoe comprises the shoe sole provided with the shoe sole supporting piece. By the adoption of the technical scheme, under the action of the shoe sole supporting piece, the shoe can provide stable supporting for the foot and has the cushioning effect, and the shoe sole and the shoe have the good movement effect and comfort.

The invention relates to the field of crushing and grinding equipment, in particular to a micro-powder crusher for diamond processing, which comprises a crushing and grinding machine, the crushing and grinding machine comprises a machine body, protective covers are mounted at both ends of the machine body, uniformly distributed crushing rollers are mounted at the middle lower part in the machine body, and rotating shafts are fixedly connected to the middle parts in the crushing rollers; one end of the forefront rotating shaft is fixedly connected with a servo motor, the periphery of the other end of the forefront rotating shaft is fixedly connected with a driving gear, a plurality of driven gears are arranged on the rear side of the driving gear, the driven gears are connected in a meshed mode, and the driving gear is connected with the rear driven gear in a meshed mode. Diamond powder can be directly guided to the grinding position of the smashing roller through the swing bin, the contact speed of the diamond powder and the smashing roller is increased, and the situation that the needed grinding time is long due to the fact that the diamond powder continuously floats and does not participate in grinding in the grinding process in the smashing and grinding machine is avoided.

The invention relates to the field of steel ladle baking equipment, in particular to an oxygen-enriched combustion type steel ladle baking device which comprises a steel ladle baking heater, the steel ladle baking heater comprises a vertical frame, a transverse frame is arranged on the inner side of the top of the vertical frame, a fixing frame is fixedly connected to one end of the transverse frame, and a plurality of connecting openings are formed in the other end of the transverse frame. The connecting ports are used for being connected with an air source and connected with guide pipes, the guide pipes are installed in the transverse frame, the ends, away from the connecting ports, of the guide pipes are fixedly connected with an air supply pipe and an oxygen supply pipe respectively, the air supply pipe and the oxygen supply pipe both penetrate through the end of the fixing frame, and the bottom of the fixing frame is fixedly connected with an end cover. According to the steel ladle baking device, synchronous baking can be conducted on a large-range area in the steel ladle body, temperature maintaining and averaging work of all the areas in the steel ladle body can be completed easily, and local overheating or supercooling is avoided.