1453results about How to "Wide adaptability" patented technology

The invention provides a robot autonomous learning and adaptive trajectory planning method oriented to man-machine cooperation aiming at a mechanical arm simulating a human teaching sample to obtain a motion skill and generalizing the motion skill to an adaptive obstacle avoidance behavior of a new situation, and specifically, a dynamic motion primitive algorithm is introduced to carry out learning and generalization of a motion trajectory of the mechanical arm. Based on the initial state and the search area of a generalized trajectory improved particle swarm algorithm, rapid search of a local obstacle avoidance trajectory is achieved, and mechanical arm motion trajectory planning with an autonomous obstacle avoidance function is comprehensively achieved. The method has the advantages that the method has high consistency with an original generalization track and wider adaptability, and the path searching efficiency is remarkably improved; a virtual simulation experiment and a physical experiment verify that the method has good stability and practicability.

The invention belongs to the technical field of data processing, and discloses an enterprise multi-type data labeling method and system based on feature engineering, and the method comprises the steps: collecting initial data of an enterprise internal information system, and dividing the initial data to obtain a text data set, a table data set and an image data set; feature vectors of different types of data are fused after features are extracted and screened, and a multi-dimensional feature space is constructed; constructing an annotation model according to the feature information of the feature space, constructing similar sample pairs as positive sample pairs and irrelevant sample pairs as negative sample pairs for unannotated new data, and comparing, learning and optimizing the similarity between the similar sample pairs; and carrying out label inference based on the feature vectors obtained by comparative learning, and generating a labeling result. According to the method, efficient and automatic labeling of various types of data can be achieved, the method is suitable for various heterogeneous data such as texts, tables and images, and high labeling precision and expansibility are achieved.

The invention discloses an integral type automatic directional drilling machine suitable for underground directional construction and a control method of the integral type automatic directional drilling machine. The integral type automatic directional drilling machine comprises a walking mechanism, a drilling main machine, a rod loading and unloading mechanical arm, a rod supplementing device, a suspension arm and a power pump station, in the drill rod loading and unloading construction process, the rod loading and unloading mechanical arm can achieve the rod loading and unloading function at different drilling angles, the rod loading and unloading action within the range of 0-90 degrees can be achieved in the mode that linear motion and rotary motion are combined, accurate positioning at multiple angles is achieved in the mode that mechanical limiting and an angle sensor are combined, and the drill rod loading and unloading efficiency and reliability are improved; during rod supplementing operation, the drilling machine has the rod bin capacity meeting the single-shift footage average value in actual construction, single-shift one-time rod supplementing can be achieved, and auxiliary personnel do not need to wait for rod supplementing basically. The drilling machine is suitable for underground coal mine automatic directional drilling machine construction, can achieve integral layout, is compact in structure, is provided with a power system, and does not need to be matched with a special power pump truck for use.

The invention provides an intelligent refrigeration system efficiency optimization method, system and device based on deep learning, and relates to the technical field of intelligent refrigeration system efficiency optimization. The method comprises the following steps: firstly, collecting historical operation parameters and environmental data of the refrigeration system, preprocessing, and integrating into a sample set according to a timestamp; and based on the hybrid model of the convolutional neural network and the long-short term memory network, constructing a state prediction model for predicting the system operation state. And in combination with the state prediction model, an optimization objective function is defined for data prediction at the next future moment, and an optimal adjustment strategy for system operation is formulated by using a deep Q network model. And finally, monitoring current operation parameters and environment data in real time, calculating a comprehensive safety score, comparing the comprehensive safety score with a safety threshold value, and deciding whether to continue to optimize the efficiency or preferentially carry out troubleshooting. According to the method, through combination of deep learning and physical modeling, efficient and safe refrigeration system operation optimization is realized.

The invention discloses a weather detection method and system based on a weather rainfall radar and a storage medium. The system comprises the weather rainfall radar, a weather data centralized control management center and a radar time sequence chart neural network detection algorithm arranged in the weather data centralized control management center. According to the method, different cloud layer characteristic weathers are collected through a meteorological rain detection radar, and obtained electromagnetic wave scattering parameters of the different cloud layer characteristic weathers are input into a radar time sequence chart neural network detection algorithm for analysis and calculation, so that polarization phase differences corresponding to the different cloud layer characteristic weathers are obtained to detect the height and vertical distribution of cloud layers; and when the height and the vertical distribution of the cloud layer exceed boundary threshold values, sending out risk early warning information of different weather types, and automatically detecting spectral widths of different cloud layer characteristic weather and corresponding polarization phase differences. And the weather detection precision and efficiency are improved.

The invention discloses an automatic long drilling tool loading and unloading system of a directional drilling machine and a control method. The automatic long drilling tool loading and unloading system comprises a drilling main machine, a rod loading and unloading mechanical arm and a rod supplementing device. In the drill rod loading and unloading construction process, the rod loading and unloading mechanical arm can achieve the rod loading and unloading function at different drilling angles, the rod loading and unloading action within the range of 0-90 degrees can be achieved in the mode that linear motion and rotary motion are combined, accurate positioning at multiple angles is achieved in the mode that mechanical limiting and an angle sensor are combined, and the drill rod loading and unloading efficiency and reliability are improved; during rod supplementing operation, the drilling machine has the rod bin capacity meeting the single-shift footage average value in actual construction, single-shift one-time rod supplementing can be achieved, and auxiliary personnel do not need to wait for rod supplementing basically. The drilling machine is suitable for underground coal mine automatic directional drilling machine construction, can achieve integral layout, is compact in structure, is provided with a power system, and does not need to be matched with a special power pump truck for use.

The invention relates to a multifunctional deformation measurement device and a measurement method, and belongs to the technical field of thermal deformation measurement. Manufacturing a multifunctional deformation measuring device; marking the test piece to be tested; calibrating the precision of the measurement system by using a reference standard substance; a temperature sensor is installed on the test platform; vacuumizing treatment is conducted through a vacuum system; temperature control is carried out through a temperature precise regulation and control module; forming a heat radiation boundary; adjusting the visual angle and focal length of the measuring camera; adjusting a monochromatic illumination light source to realize light coverage on the tested piece; spraying speckles on the test piece to be tested; measuring the initial state of the tested piece; temperature control is carried out through a temperature precise regulation and control module; testing and recording the deformation condition of the tested piece at each temperature point in the whole test process; the method can solve the problems that a measurement test piece is only limited to a material-grade sample, the measurement environment simulation application range is small, the measurement error is large and the like in a traditional method, and is used for guaranteeing high-precision measurement of thermal deformation of a large-size composite material structure of a spacecraft.

The invention discloses a heat preservation and thermal insulation project whole industrial chain quality tracing method and system, and the method comprises the steps: obtaining industrial chain data and monitoring data of a heat preservation and thermal insulation project, carrying out the tracing coding based on the industrial chain data and the monitoring data, and constructing a heat preservation and thermal insulation project whole industrial chain quality tracing network, determining industrial chain data corresponding to the to-be-traced monitoring data through the traceability code, and performing correction and anomaly screening on the to-be-traced monitoring data to obtain to-be-traced abnormal monitoring data, and according to the to-be-traced abnormal monitoring data and the thermal insulation project whole industry chain quality traceability network, determining a thermal insulation project whole industry chain quality hidden danger source. The method not only can improve the efficiency and accuracy of quality traceability of the whole industrial chain of the thermal insulation project, but also has good interpretability, and can be directly applied to a quality traceability system of the whole industrial chain of the thermal insulation project.

The invention discloses a multi-element energy storage collaborative distributed energy system operation regulation and control method, and the method comprises the steps: considering an optimal scheduling strategy of a power-energy type electricity storage, heat storage, cold storage and other technologies for cooperatively regulating and controlling a multi-energy load curve, and constructing a generalized technical model of energy equipment and system operation constraints; designing an energy system operation data preprocessing mechanism and a weight determination model under a given historical window; an energy system scheduling sequence of bipartite dynamic regression coordinated electric heating unit operation is provided, and an economic operation planning model is constructed; the model is applied to a historical window, optimal parameters of key regulation and control nodes are searched and determined through a heuristic algorithm so as to guide future time window scheduling and complete scheduling closed loop, and hyper-parameters of the method, such as historical and future window vision fields, time nodes and weights, are determined based on multiple scheduling closed loop behaviors so as to complete method closed loop. According to the method, a commercial solver is not needed, the economic operation rule of the distributed system is directly mined from historical operation, the method is used for determining the on-off and output plan of each device of the system in a future time sequence, and engineering application is facilitated.

The invention discloses a terahertz vortex phase plate based on a dumbbell-shaped metasurface. A plurality of metasurface units arranged on the same plane; the metasurface unit comprises a top basic unit metal structure layer, a middle dielectric layer and a substrate metal reflecting surface layer; the top basic unit metal structure layer is of a dumbbell-shaped structure, and the area is smaller than that of the middle dielectric layer and the base metal reflecting surface layer; the positions of all layers of the multiple metasurface units are correspondingly arranged, the phase distribution of the arranged multiple top layer basic unit metal structure layers on the plane meets the vortex beam phase distribution, and a metasurface array with the target topological charge is formed; the metasurface array is a terahertz wave incident plane. On the basis of improving the working efficiency, the metasurface technology and the geometric phase principle are combined, the whole-column arrangement of the terahertz vortex phase plate and the generation of terahertz vortex light beams with different topological charges are completed by designing a reflection type basic unit structure, and then the phase coverage corresponding to the specified topological charges is realized.

The invention discloses a preparation method of a high-activity hierarchical pore boric acid imprinted hydrogel microsphere adsorbent and application of the adsorbent to selective adsorption of a flavonoid compound naringin NRG. According to the invention, dopamine is used for modifying COFs so as to enhance surface functionalization of COFs; preparing a sulfydryl functionalized covalent organic framework microsphere; synthesizing sulfydryl microspheres by utilizing an emulsion interfacial polymerization method; preparing a COFs boron affinity imprinting microsphere adsorbent based on boron affinity interaction; the adsorbent is applied to selective adsorption of NRG in agricultural wastewater. The prepared adsorbent has a remarkable multilayer adsorption effect, and the adsorption capacity is remarkably improved through the boron affinity effect. The multi-stage pore structure of the COFs microspheres provides abundant sites for adsorption, and the mass transfer rate is remarkably improved. In addition, the prepared adsorbent shows remarkable specificity in the aspects of adsorption and separation of NRG, and a promising adsorption material is provided for industrial separation of NRG.

The invention discloses a method for dynamically correcting off-axis aberration of a liquid mirror telescope, which comprises the following steps of: S1, adjusting a secondary mirror body to perform dynamic full-angle circular motion around the center of a primary mirror by using a secondary mirror rotating system, so that the secondary mirror body acquires an incident beam from any direction; s2, acquiring attitude information of the secondary mirror body, and adjusting the secondary mirror body according to the dynamically changeable off-axis angle based on a preset secondary mirror pose adjustment sub-method to compensate for low-order aberration; s3, the light beams of the distorted wavefront compensated by the secondary mirror are adjusted to be emitted in a determined direction in parallel and enter an adjustable beam expanding and shrinking mirror through two groups of plane reflection systems; s4, the light beam with the light spot size adjusted through the adjustable beam expanding and shrinking mirror is reflected to the magnetic liquid deformable mirror through the reflecting mirror; and S5, the magnetic liquid deformable mirror receives the reflected light beam and performs residual aberration correction. According to the method, the observation view field of the liquid mirror telescope can be effectively improved, the imaging quality is improved, and an aberration correction system is endowed with excellent flexibility and wide adaptability.

The invention relates to a portrait generation and feature similarity recommendation method and system based on a large language model, and the system comprises a data collection module, a feature extraction module, a similarity calculation module and a recommendation module, the data collection module collects multi-modal data from a social media platform, and covers audio, video, text and interactive data information; the feature extraction module performs deep feature analysis and processing based on the collected multi-modal data, and extracts features in multiple dimensions; the similarity calculation module generates a training sample based on the feature matrix generated by the multi-dimensional feature extraction module in combination with a large language model, constructs a similarity matrix of audio, video, text and interactive data features, and calculates similarity scores to determine candidate objects; the recommendation module provides personalized recommendation based on the comprehensive similarity score generated by the similarity determination module. Through combination of a large language model and multi-modal features, the system can accurately analyze behaviors and expressions and generate personalized recommendation.

The invention relates to the field of flight simulation, and discloses a multi-degree-of-freedom motion control method for flight driving rocker and accelerator simulation, which comprises the following steps: receiving attitude data from flight simulation software in real time, constructing a rotation matrix based on the attitude data, and calculating displacement data of platform angular points; the target stroke of the linear electric cylinders is calculated in combination with the displacement data, and the target stroke is optimized, so that movement of the multiple electric cylinders is coordinated, and the mechanical stroke range of the electric cylinders is met; meanwhile, dynamic feedback data including vibration feedback and damping feedback are generated based on the flight environment parameters, the dynamic feedback data are superposed into the control instruction, and an electric cylinder is driven to adjust the attitude of the platform; the platform motion state is detected and updated in real time through a closed-loop control system, and it is ensured that platform motion and a flight simulation software scene are highly synchronized. The system has the advantages of high-precision attitude control, real-time dynamic response and good system stability, can adapt to various flight simulation scenes, and provides real flight simulation experience.

The invention discloses comprehensive rock drift rapid tunneling equipment and a using method thereof.According to the tunneling equipment, a drilling and anchoring integrated multifunctional drilling machine is arranged on one side of a cantilever type tunneling machine body, and a temporary supporting mechanism is arranged above an oscillation cutting part; a deep hole is formed in a to-be-crushed rock by utilizing the multifunctional drilling machine, the deep hole is further expanded and cracked by utilizing the pulsating fracturing system, the strength of a rock mass is weakened, then the rock mass is milled by adopting the oscillating cutting part, and finally, the rock mass is anchored and protected by utilizing the multifunctional drilling machine and the temporary supporting device, so that a guarantee is provided for an operation environment. According to the equipment and the technology, a traditional roadway tunneling procedure is simplified into a continuous procedure, the tunneling efficiency of roadway rock mass is greatly improved, and meanwhile the cost is saved.

The invention relates to synthesis of a heterogeneous Fenton catalyst and application of the heterogeneous Fenton catalyst to degradation of PFAS, and belongs to the technical field of environmental engineering water treatment. The preparation method comprises the following steps: roasting an MOF material in a non-oxidizing atmosphere to obtain a carbonized MOF material, taking the carbonized MOF material as a template, synthesizing FeS2 by taking FeSO4, Na2S2O3 and elemental sulfur as raw materials through a hydrothermal method, and preparing to obtain the heterogeneous Fenton catalyst. Then, a three-electrode system is constructed, FeS2-MOF (at) C is used as a catalyst for efficiently degrading PFAS, and the heterogeneous Fenton catalyst has the advantages of being stable in crystal structure, high in electron transfer efficiency, excellent in catalytic performance and the like and is suitable for being used for removing various organic pollutants difficult to degrade.

The invention relates to the technical field of urban public transportation, and discloses an intelligent passenger flow perception and bus route adaptive adjustment system and device, and the adjustment system is composed of a data collection module, a data processing module, a passenger flow prediction and event detection module, a dynamic scheduling and route adjustment module, and an information release and user interaction module. According to the method, the characteristics of passenger flow density, flow direction, peak period and the like are extracted, the passenger flow change from a few minutes to a few hours in the future is predicted based on real-time data, and the passenger flow trend from a few days to a few weeks in the future is predicted in combination with historical data and special event information such as large-scale activities, traffic congestion and weather change. By automatically identifying the influence of the special event on the passenger flow, the passenger flow peak or the abnormal passenger flow condition can be predicted in advance, the module automatically adjusts the bus route and the departure time according to the passenger flow prediction result, and the beneficial effect of adjusting the bus route in real time according to the passenger flow distribution in the high stage or the special event period is achieved.

The invention discloses a power transmission line insulator defect detection method based on deep learning, and the method comprises the steps: S1, obtaining an aerial image, carrying out the preprocessing of the aerial image, and generating an insulator defect detection data set; s2, based on a YOLOv8n model, introducing a space object attention module (RCS-OSA), and improving the feature extraction capability under a complex background; s3, adding an extrusion excitation network (SENet) in the neck network of the model to enhance the feature expression ability between channels; s4, constructing a small target detection layer, and adding a high-resolution feature map (160 * 160) to capture small defect target information; s5, optimizing a detection head by using multi-scale feature fusion and a dynamic label distribution strategy, and improving the recognition precision of the model on insulator defects under a complex background; and S6, outputting defect positions, types and confidence coefficients based on a detection result, and generating an electric power inspection report. According to the method, the accuracy and robustness of insulator defect detection under the complex background of the power transmission line are effectively improved, and the method has relatively high applicability and real-time performance.

The invention discloses a preparation method of a high-entropy metal oxide for preparing synthesis gas by carbonate pyrolysis carbon dioxide and methane dry reforming. The high-entropy composite metal oxide is obtained by nucleation of a mixed salt solution and an alkali source solution in a nucleation reactor, crystallization, washing, drying, grinding and high-temperature roasting. The high-entropy composite metal oxide prepared by the method is used for ultrahigh-temperature coupling reaction for preparing synthesis gas by taking carbonate pyrolysis industrial tail gas CO2 and CH4 as raw materials, the synthesis gas is efficiently prepared, and meanwhile, full utilization of waste heat in the carbonate pyrolysis process and resource conversion of CO2 are realized; the contribution is made for promoting the development of environment-friendly society and circular economy; the high-entropy composite metal oxide catalyst has excellent activity, selectivity and stability.

The invention discloses an equipment test model code generation method based on a large model, and the method comprises the steps: designing an equipment digital test demand analysis module which is used for constructing an equipment digital test identification framework system through combining demand description to extract a digital test demand generated by an equipment test model modeling code; designing a large model training code collecting and processing module, and generating a high-quality code data set; the design equipment test model code library module is used for carrying out structured storage on the high-quality code data set, constructing a knowledge graph based on structured data and designing a vectorization dynamic query function; and designing a large model pre-training and fine tuning module, and performing secondary pre-training and fine tuning on the base code large model by using the high-quality code set and a vectorization dynamic query function to generate a final equipment test model code. According to the method, the code generation speed and accuracy are remarkably improved, and it is ensured that the generated code has wide adaptability and universality.

The invention belongs to the technical field of new energy materials, and discloses a fused salt auxiliary valence gradient doped modified single crystal positive electrode material and a preparation method and application thereof. The preparation method comprises the following steps: mixing a transition metal source, a lithium source, a manganese source and a molten salt auxiliary agent, sequentially carrying out first sintering and second sintering, and removing the molten salt auxiliary agent to obtain a doped modified single crystal positive electrode material precursor; and tempering the doped modified single-crystal positive electrode material precursor to obtain the fused salt auxiliary valence gradient doped modified single-crystal positive electrode material, the doped modified single crystal positive electrode material prepared by the invention can effectively avoid the capacity loss of the material, significantly reduce lattice mismatch in the lithium intercalation / deintercalation process, and relieve the volumetric strain in the charge-discharge process. Meanwhile, the formed M-O bond can enhance the structural stability of the material and inhibit the formation of microcracks of the material. The preparation method of the positive electrode material is convenient for industrial application, and provides a thought for commercial modification of the positive electrode material.

The invention provides a Trichoderma asperellum TAZ19, a biocontrol inoculant and application of the Trichoderma asperellum TAZ19, the Trichoderma asperellum TAZ19 is preserved in the China Center for Type Culture Collection on October 25, 2024, and the preservation number of the Trichoderma asperellum TAZ19 is CCTCC M 20242341. The active ingredients of the biocontrol inoculant comprise the trichoderma asperellum TAZ19 strain, and the trichoderma asperellum TAZ19 strain is one or more of conidia, hypha, fermentation filtrate or active extracts of the trichoderma asperellum TAZ19 strain of the trichoderma asperellum TAZ19 strain. The trichoderma asperellum TAZ19 has an excellent antagonistic effect on a plurality of plant disease pathogenic bacteria such as dispora pinnatifida, a plurality of fusarium oxysporum, apple black beancurd skin fungi and botryosphaeria dothidea, has broad-spectrum antibacterial activity, is high in growth speed, high in antibacterial activity and high in hyperparasitism ability, has better saline-alkaline tolerance, also has a growth promoting effect on plants, and can be widely applied to the field of plant diseases. The strain can be used as an active component of a biocontrol inoculant to be widely applied to prevention and treatment of various plant diseases such as pine tree branch rot, wheat stem rot and tomato fusarium wilt, and has a wide application prospect.

The invention discloses a flexible clamping jaw capable of automatically adjusting clamping force based on deformation measurement, and belongs to the field of vegetable and fruit picking devices.The flexible clamping jaw comprises a rotating mechanism, three clamping jaw mechanisms are symmetrically installed at the front end of the rotating mechanism through an installation disc, and the rotating mechanism and the clamping jaw mechanisms form a picking mechanism; the rotating mechanism can drive the three clamping jaw mechanisms to rotate through an internal rotating motor, each clamping jaw mechanism comprises a flexible clamping jaw, the tail of each flexible clamping jaw is driven by a connected cam connecting rod mechanism to swing, and the back face of each clamping jaw mechanism is fixedly provided with a monitoring mechanism. The monitoring mechanism monitors the deformation quantity of the flexible clamping jaw through a camera. According to the flexible clamping jaw, the clamping force is accurately controlled through a camera and an algorithm, and fruits are protected against damage; the three-jaw structure is stable, the process is automatic, and the efficiency is improved; ginseng can be flexibly set according to vegetables and fruits, and the adaptability is wide; data after picking is driven and optimized, long-acting operation of equipment is facilitated, and the cost is reduced.

The invention discloses a weather radar echo intensity online calibration method, and relates to the technical field of weather radar detection. The radar collects and processes an environment signal to obtain effective clutter data; determining a relative calibration offset of the weather radar according to the effective clutter data and the relative calibration baseline; and calibrating the weather radar based on the relative calibration offset. According to the method, the weather radar is calibrated online in real time by using the ground clutter creatively, and the dependence on external equipment is reduced. And the precision and timeliness of radar echo intensity measurement are obviously improved. The maintenance cost and the operation difficulty are reduced, and the calibration inconsistency caused by manual operation errors is reduced. The calibration cost of the weather radar is remarkably reduced, continuous high-quality output of data is ensured, the accuracy and reliability of data acquired by the weather radar are remarkably improved, higher requirements of modern meteorological observation and forecasting are met, and powerful support is provided for meteorological observation and early warning and forecasting. And the long-term consistency of the data is ensured. And the method has wide adaptability.

The invention belongs to the technical field of digital measurement, and particularly relates to a high-precision three-dimensional measurement method for aircraft profile local features, which comprises the following steps: firstly, aligning a camera imaging center to an aircraft surface area to be measured, and adjusting the longitudinal position of a scanning system to enable a camera focal plane to be positioned at an effective scanning position; then adjusting the included angle between the measuring light path and the surface to be measured, so that the period number of imaging fringes is within 1; controlling the scanning system to move in a single direction so as to longitudinally scan the surface to be measured; after scanning is completed, a longitudinal light intensity curve is extracted for each pixel point, and an accurate focusing position of the pixel point is obtained by further combining a Fourier transform technology and a Gaussian curve fitting algorithm; and finally performing height mapping by combining the scanning step pitch so as to recover the three-dimensional shape of the to-be-detected area. The technical scheme can improve the measurement precision, supports the three-dimensional measurement of the tiny local features of the aircraft profile, and has the characteristics of high demodulation efficiency and wide adaptability.

The invention belongs to but not limited to the technical field of animal husbandry statistics, and particularly relates to an alpine meadow yak number calculation method and system based on an unmanned aerial vehicle image, and the method comprises the steps: S1, employing an unmanned aerial vehicle to shoot an orthoimage of a sample plot, and carrying out the statistics of the yak number; s2, splicing the shot images by using Pix4D software, screening a proper sample plot, and respectively counting the quantity of wet cow dung and the quantity of dry cow dung in a visual interpretation mode; and S3, carrying out linear fitting on the obtained data by using Origin 2024b, so as to obtain three relation curves. According to the method for estimating the number of the yaks through statistics of the number of the cow dung in the unmanned aerial vehicle image, the workload of data collection is greatly reduced, and the limitation that the number of the yaks cannot be obtained when the yaks are not grazed in the grassland is avoided. According to the invention, the data acquisition frequency of experimenters is effectively reduced, and the research cost is reduced.

The invention relates to the technical field of network security, in particular to realization of self-efficient autonomous defense in an Internet of Things network. Comprising the following steps: constructing a network attack and defense scene, and modeling observation information by using a graph data structure; modeling the action space of the attacker and the defense agent; modeling an attack and defense process in the network attack and defense scene; performing normalization processing on the feature matrix of the nodes by adopting graph normalization; dynamically aggregating the current state information of each device according to the interaction between the network device and the adjacent device by using the self-attention mechanism of the GAT; selectively reserving a node host according to the self-attention score by using a down-graph sampling pooling strategy; and training a defense agent by using a graph attention network enhanced reinforcement learning algorithm. The problems that automatic defense of an existing reinforcement learning algorithm faces performance bottleneck in a complex network scene, convergence to a high return value is difficult, and an effective defense strategy cannot be made are solved.

The utility model discloses a two-stage coal blending system based on coking coal grading and grading crushing, and belongs to the technical field of coal storage and blending. The coal blending system comprises a hard coal silo, a two-stage screening and crushing mechanism, a coal blending bunker and a coal bunker, a discharging port of the hard coal silo is connected with a feeding port of the first-stage screening and crushing mechanism through a belt, and a discharging port of the first-stage screening and crushing mechanism is connected with a feeding port of the coal blending bunker through a belt; a discharging port of the coal blending bin is connected with a feeding port of the second-stage screening and crushing mechanism through a belt, and a discharging port of the second-stage screening and crushing mechanism is connected with a feeding port of the coal bin through a belt. By adopting the coal blending system disclosed by the utility model, the characteristics of various coals are effectively utilized, and various coking coals are matched according to a proportion, screened and then crushed, so that the mixing uniformity of the coals is improved, the phenomenon of coarse coal leakage in the coal blending process is improved, the coke quality is improved, and the coal blending cost is also reduced.

The invention relates to the technical field of catalysts, in particular to a catalyst for fire coal and a preparation process thereof, and the catalyst comprises the following materials in parts by weight: 2-8% of rare earth nitrate, 5-18% of alkaline earth metal salt, 2-5% of alkali metal salt, 0.5-1% of copper-manganese composite oxide, 0.2-1.5% of transition metal salt, 0.1-0.3% of nonionic surfactant, 0.05-0.15% of organosilicon compound and the balance of water. The coal-fired catalyst is specially designed for a power plant, has excellent permeation and adsorption performance, solves the problem that the coal-fired catalyst is difficult to uniformly disperse in blocky coal, promotes full combustion of the coal in the power plant, improves the combustion rate and the burn-off rate, eliminates solid hazardous wastes from the preparation raw materials, ensures safety and environmental protection, and is suitable for industrial production. The power plant can be assisted to save coal resources, emission of harmful pollutants can be reduced from the source, and remarkable economic benefits and environmental benefits are brought to the power plant.

The utility model relates to the technical field of textile equipment, solves the problem that a pan head is troublesome to adjust, and discloses a movable let-off frame for a warp knitting machine, which comprises a movable frame, a yarn winding mechanism, a driving mechanism and a yarn separating mechanism, the yarn winding mechanism comprises a supporting shaft detachably connected to the frame body and a plurality of winding drums arranged on the supporting shaft in a sleeving mode, the inner diameter of each winding drum is larger than the outer diameter of the supporting shaft, and the winding drums are used for winding warp yarn; the driving mechanism comprises a driving roller rotationally connected to the frame body and a driving part for driving the driving roller to rotate, the winding drum slides downwards under the action of self weight until warp yarns abut against the driving roller, and the driving roller can drive the winding drum to rotate when rotating; the yarn separating mechanism is used for carding the warp yarn led out of the winding drum. Warps are wound on the winding drums respectively, the winding drums with the corresponding number are connected to the supporting shaft in a sleeving mode according to production requirements, and compared with a mode that a pan head with a larger size is adjusted, operation is convenient and fast; the whole let-off frame can be moved to the front of different warp knitting machines, and different production requirements are met.