43results about How to "Small fluctuations in performance" patented technology

The utility model provides a two-way chain type cooling bed. The cooling bed comprises a plurality of forward chains and a plurality of reverse chains, the running direction of the forward chains is the same as the advancing direction of the steel, and the running direction of the reverse chains is opposite to the advancing direction of the steel; a forward chain motor, wherein all the forward chains are connected with the forward chain motor; a reverse chain motor, wherein all the reverse chains are connected with the reverse chain motor; the steel splitting disc is positioned at an inlet of the cooling bed; the fans are arranged on the two sides of the cooling bed correspondingly. According to the cooling bed, steel is cooled on the cooling bed in a temperature monitoring control cooling mode, the overall-length temperature difference deviation of the steel is small, overall-length uniform cooling of the steel is achieved, material performance fluctuation is small, the structure is stable, and meanwhile the performance requirements for strength and toughness are met; and the cooling uniformity of the two ends of the steel is adjusted in real time through the temperature, and the structure uniformity and performance stability of the cooled steel are ensured.

The invention provides a large 3D scene texture dynamic loading and unloading method, medium and system, and belongs to the technical field of 3D scene loading.The large 3D scene texture dynamic loading and unloading method, medium and system comprise the steps that three-dimensional scene parameters are collected, texture frequency parameters are decomposed to obtain basic and fluctuation components, a texture loading model is constructed, and a coefficient matrix is generated; packaging the texture file into a fixed-size block, building a double-layer resource optimization model based on a coefficient matrix, setting global and local optimization targets, and calculating a comprehensive contribution coefficient; an initial resource allocation scheme is obtained through an evolutionary algorithm, and a final scheduling scheme is obtained through dynamic programming; according to the texture blocks, loading the texture blocks, determining a recovery strategy in combination with life cycle, grouping association and state transition prediction, and executing resource recovery; the process comprises texture analysis, model construction, resource optimization, scheduling and recovery, and efficient management of three-dimensional scene textures is realized; the resource scheduling decision-making mechanism can be optimized, and the user experience is improved.

The invention relates to the technical field of fuel cells, and particularly discloses a proton exchange fuel cell system performance prediction method and system, and the method comprises the steps: S1, collecting parameters related to the performance of a proton exchange fuel cell system PEMFC, and obtaining real-time operation data; s2, performing preprocessing and feature extraction on the real-time operation data in the step S1 based on an anomaly detection method of deep learning to obtain standardized data; s3, constructing a machine learning model, training the model by using the standardized data, and performing multi-objective optimization by using a particle swarm optimization algorithm to obtain a sequence prediction model; s4, performing data prediction and verification by using the sequence prediction model in the step S3, and predicting to obtain the service life and performance data of the fuel cell; by optimizing the operation parameters, the performance fluctuation of the fuel cell in a high-power output or high-temperature environment can be reduced, and a plurality of targets are incorporated into the optimization process, so that the method is suitable for various application scenes.

The invention discloses a straw powder filled biodegradable composite material and a preparation method thereof, and relates to the technical field of biodegradable composite materials, the preparation method of the biodegradable composite material comprises the following steps: S1, pretreating straw; s2, impurities are removed through winnowing; s3, removing magnetic impurities; s4, cleaning and drying; s5, performing high-pressure steam treatment; s6, drying in the oven and crushing again; s7, adding into a high-speed mixer and mixing; s8, extruding the biodegradable composite material, and cooling and shaping the biodegradable composite material; and S9, cutting and polishing are conducted, and performance testing is conducted. The biodegradable composite material is prepared from agricultural waste, waste accumulation is reduced, sustainable development is promoted, the straw powder filled biodegradable composite material has biodegradability, high strength and good processability, and the straw powder filled biodegradable composite material is suitable for large-scale production. The problems of reducing agricultural wastes, enhancing mechanical properties and environmental protection performance, improving processability and reducing raw material cost can be solved.

The invention discloses a self-crosslinking type multi-component graft copolymerization paper reinforcing agent and a preparation method, and belongs to the technical field of papermaking. Carboxymethyl cellulose is used as a dispersing agent, isopropanol is used as a chain transfer agent, deionized water is used as a solvent, and the self-crosslinking type multi-component graft copolymerization paper reinforcing agent is prepared. The preparation method comprises the following steps: initiating polymerization reaction of methacryloyloxyethyl trimethyl ammonium chloride, acrylamide, acrylic acid and 2-acrylamide-2-methylpropanesulfonic acid at 50-80 DEG C under the action of an initiator; the mass ratio of the carboxymethyl cellulose to the methacryloyloxyethyl trimethyl ammonium chloride to the acrylamide to the acrylic acid to the 2-acrylamide-2-methylpropanesulfonic acid is (1-10): (1-5): (10-30): (1-10): (1-5), and then adjusting the pH value of the obtained reaction solution to 4.5-5.5 to obtain the self-crosslinking type multi-component graft copolymerization paper reinforcing agent. The stability of a cross-linked structure of the reinforcing agent is improved, the binding power is increased, and the strength and the water resistance of paper can be remarkably improved.

The invention belongs to the technical field of optical fibers, and particularly discloses application of microcrystalline glass in an optical fiber ferrule and an optical fiber connector. The glass ceramic comprises the following components in percentage by weight: 50 to 80 percent of SiO2 (silicon dioxide), 5 to 40 percent of Al2O3 (aluminum oxide), 8 to 30 percent of Li2O (lithium oxide) and 0.1 to 3 percent of a strength additive, the strength additive is one or more of a rare earth metal compound, a transition metal compound, organic fibers, borate, phosphate, magnesium oxide or magnesium fluoride. According to the invention, the microcrystalline glass added with the strength additive is used as an insertion core material of the optical fiber connector, so that the optical fiber connector can show higher environmental stability and higher thermal shock resistance in an extremely high temperature environment, and performance fluctuation caused by high temperature thermal expansion and thermal cycle impact is reduced; and the method can be applied to high-temperature fields such as high-power laser and the like and fields with large environment temperature change.

The invention belongs to the technical field of high-voltage electric appliance material preparation, and particularly relates to a zinc oxide resistor disc and a preparation method and application thereof. The preparation method comprises the following steps: (1) pre-sintering zinc oxide powder; (2) mixing the zinc oxide powder obtained in the step (1), an auxiliary agent and a dispersion liquid, and carrying out ball milling to obtain slurry; the particle size D50 of the slurry is 0.8-1.2 [mu] m, and the dispersity is greater than or equal to 95%; (3) performing spray drying and granulation on the slurry to obtain granulation powder; (4) carrying out molding and first sintering on the granulated powder to obtain a resistor disc semi-finished product; the first sintering adopts sectional sintering; and (5) forming an electrode layer on at least one surface of the resistor disc semi-finished product by adopting a vacuum sputtering and / or electrophoretic deposition process, and performing secondary sintering and annealing. The contact resistance of the zinc oxide resistor disc is less than or equal to 5m omega; and after 1000 hours of aging in a 85 DEG C + 85% RH environment, the leakage current increase is less than or equal to 2 [mu] A, and the performance degradation rate is less than or equal to 5%.

The invention provides a bio-based water-based polymer foam inhibitor and a preparation method thereof, and the bio-based water-based polymer foam inhibitor comprises the following raw materials in parts by weight: 5-15 parts of water-based acrylate polymer, 1-2 parts of bio-based hydrophobic particles (polylactic acid or carnauba wax), and the balance of water. 70-90 parts of bio-based quick bubble-breaking effective substances (soybean oil polyol and polycarbonate diol), and 6-10 parts of bio-based emulsifiers (bio-based polyethylene glycol and lecithin). The invention can provide an environment-friendly, efficient and stable foam inhibition solution, is suitable for various industrial fields, and reduces the negative influence on the environment at the same time.

The invention relates to a transparent software updating method and device based on program entity activity modulation, and the method comprises the steps: obtaining state data and service dependence topology of a software component in real time, the state data comprising CPU usage rate, memory usage rate, network throughput and response time; based on the trained prediction model and the current state data, predicting the state data of the software component at the next moment to obtain predicted state data; calculating the activity degree of each software component based on the prediction state data, and determining the state classification of each software component; outputting an update decision through request analysis, window scoring and sequence planning based on the service dependency topology, the prediction state data and the state classification of each software component; the updating decision comprises updating time, an updating sequence and a resource demand; and executing transparent updating based on the updating decision. According to the invention, through prediction of the service dependence topology and the real-time state data, the stability and reliability of transparent updating and the resource utilization rate are improved.

The present invention discloses an intelligent load balancing method based on a zero-sample hybrid expert model, comprising the steps of: creating a task spectrum mapping relationship through zero-sample learning, generating virtual labels, refining the requirements of new tasks into resource requirement feature spectra through multi-dimensional feature cross-projection, and predicting the multi-dimensional resource requirements of new load balancing tasks; establishing a multi-perspective hybrid expert dynamic integration mechanism, configuring multi-perspective hybrid collaborative experts for tasks, applying a combination of several expert models to different demand dimensions, creating virtual image samples of new tasks, simulating resource requirements and their load effects in a virtual environment, and generating an initial resource allocation plan; in the face of dynamic load fluctuations, distributing task cells to nodes with sufficient resources; the method improves the robustness and efficiency of the system, simplifies the implementation process, and has broad application prospects and practical value.

The invention provides an ultrahigh-temperature-resistant carbon fiber composite filter material and a production system thereof, and relates to the field of composite filter material production. An ultrahigh-temperature-resistant carbon fiber composite filter material production system comprises the following steps: S1, using amount optimization through a computer simulation technology so as to predict an optimal material combination ratio, S2, selecting high-quality carbon fibers and resin as a composite resin material according to a simulation result, S3, pretreating the carbon fibers, modifying the resin, and S4, preparing the ultrahigh-temperature-resistant carbon fiber composite filter material. The method comprises the following steps: S1, preparing a polyimide resin matrix, S2, preparing a polyimide resin matrix, S4, designing a composite structure of the filter material, S5, combining carbon fibers with the polyimide resin matrix through dipping treatment to form a composite material with high strength and light weight, S6, placing the composite material in a mold for curing and cooling, and S7, carrying out surface polishing and grinding on the molded composite filter material, and coating a high-temperature-resistant coating. And S8, obtaining the formed ultrahigh-temperature-resistant carbon fiber composite filter material. The material performance is verified by simulating the ratio of the resin to the carbon fibers, and performance fluctuation caused by material proportion difference is reduced.

This invention relates to an intelligent screen printing device, a screen printing system, a control method, and a computer-readable storage medium, belonging to the field of thin-film battery manufacturing technology. The intelligent screen printing device is used for perovskite thin film screen printing and includes: a screen frame for supporting the screen; a tension sensing module disposed on the screen frame for real-time monitoring of the tension distribution in various areas of the screen; a temperature regulation module disposed on the screen frame for heating or cooling the screen; an electrostatic control module disposed above the intelligent screen printing device for generating an ion flow to eliminate static electricity accumulation during the printing process; and a control unit electrically connected to the tension sensing module, the temperature regulation module, and the electrostatic control module, respectively, for receiving monitoring signals and outputting control commands to achieve coordinated closed-loop regulation of tension, temperature, and electrostatics.

This invention discloses a corona-resistant insulating varnish for enameled wire and its preparation method, belonging to the technical field of insulating coatings. The raw materials for preparing the corona-resistant insulating varnish for enameled wire include aromatic tricarboxylic anhydride, aromatic diisocyanate, and hydroxyl-modified composite nanoparticles; the hydroxyl-modified composite nanoparticles comprise silica nanoparticles and layered double hydroxides. This invention significantly increases the hydroxyl modification density by introducing layered double hydroxides and silica composites, enabling the composite particles to be covalently anchored to the prepolymer backbone through esterification, resulting in uniform release after curing. This achieves a balance between storage stability and curing uniformity, making it suitable for enameled wire applications in high-voltage motors.

This invention relates to the field of heat-resistant alloys and high-temperature alloys, and discloses a high-temperature resistant creep-resistant iron-nickel-chromium-based alloy and its preparation process. The alloy, by weight percentage, comprises Ni 25.0%–28.0%, Cr 18.0%–22.0%, C 0.06%–0.08%, V 0.50%–0.80%, N 0.10%–0.15%, Ce 0.015%–0.025%, B 0.005%–0.010%, Mn 0.3%–0.8%, and Si 0.1%–0.4%, with the balance being Fe and unavoidable trace impurities. The preparation process employs gradient-controlled pressure reverse nitrogen purging, delayed addition of a nickel-cerium master alloy and a micro-carbon boron-iron alloy before tapping, combined with a double-step isothermal carbon consumption treatment. This approach improves the effective utilization rate of nitrogen and rare earth elements, and enhances the high-temperature creep life, fracture elongation, and cross-sectional property consistency of large-section forgings.

The present invention relates to the technical field of charging piles, and discloses a cooling device for a charging pile of a new energy vehicle, including a charging pile box and a charging gun connected to the outer wall thereof and detachable for use, and also including a charging module arranged in the charging pile box, the outer walls on both sides of the charging pile box are provided with vents, the inner wall of the charging pile box is fixedly installed with a bottom plate and a top plate, the bottom plate, the top plate and the outer wall of the charging pile box together form a space, and a heat insulation box is fixedly installed on the top of the bottom plate. The present invention switches the heat dissipation mode under different temperature environments. In a hot environment, the charging module can be directly cooled and heat-exchanged to cool the charging module in a high temperature environment, thereby reducing performance fluctuations caused by excessively high temperatures; when working in a low temperature environment, it can prevent the charging speed from being too slow due to low charging efficiency and affecting the user experience, and indirect heat dissipation of the charging module can increase its service life and reduce safety hazards.

The invention discloses a preparation method of a polyvinylidene fluoride based composite material of a magnetic drive pump key component, and relates to the technical field of polymer composites.The preparation method comprises the specific steps of raw material modification treatment, component premixing, melt blending extrusion, mold pressing cooling forming and plasma aftertreatment. The PVDF resin is subjected to graft modification, the carbon fibers are sequentially subjected to oxidation and silane coupling agent dipping treatment, and the filler with the specific function is matched and premixed, so that bonding of adaptive interfaces of a matrix, a reinforced phase and the functional filler is promoted, and it is guaranteed that components are uniformly distributed and consistent with the internal structure; a compact structure is formed through melt blending, gradient pressure mold pressing and segmented cooling forming, and the surface state is optimized in combination with plasma aftertreatment, so that the material adapts to the stress requirement of key components of the magnetic drive pump and resists severe working conditions, performance fluctuation caused by environmental factors is reduced, and long-term stable operation of the components is guaranteed.

The application relates to the technical field of sodium ion battery negative electrode materials, in particular to a closed pore directional regulation and control bamboo-based hard carbon negative electrode material and a preparation method thereof. The method is used for preparing the bamboo-based hard carbon negative electrode material through bamboo material selection and pretreatment, three-stage gradient pre-oxidation, plasma activation, molten salt auxiliary step-by-step carbonization, deep acid pickling purification, and grading ball milling and surface carbonization coating steps, accurately regulates and controls a closed pore structure, reduces defects, improves a carbon yield and batch consistency, is applied to a sodium ion battery, and realizes synchronous improvement of a first circle coulomb efficiency, sodium storage capacity, rate performance and cycle stability.

The application discloses a preparation method and application of a medical sterile gel coupling agent and belongs to the technical field of gel coupling agents for medical examination. First, the in-situ neutralization of acrylic acid is carried out by using the chitosan quaternary ammonium salt in the formula to generate a hydrophilic monomer pre-composite; then the pre-emulsion of the acrylic ester monomer is mixed, and the main network is constructed by initiating copolymerization under an inert atmosphere; subsequently, the tannic acid slurry is introduced in the polymerization later period, and physical crosslinking is realized through the polyphenol structure, so that the medical sterile gel coupling agent is finally prepared. The preparation method and application of the medical sterile gel coupling agent are adopted, through the endogenous neutralization-synchronous complex integrated process, the powerful antibacterial function and the tissue repair function are deeply integrated in the stable gel network, the process makes the key active ingredients produce structure-function synergistic effect, forms an indivisible precision system, and finally the product realizes active antibacterial protection while maintaining excellent imaging performance, and the biocompatibility is excellent.

This invention relates to the field of functional food processing technology, specifically disclosing a functional asparagus powder with high saponin content and its preparation method. The method includes the following steps: asparagus is sliced ​​and microwaved to deactivate enzyme activity; after pulping, a composite regulator aqueous dispersion is added; the pH is adjusted and a cellulose complex enzyme is added for enzymatic hydrolysis; then, it is pre-frozen and vacuum freeze-dried; finally, it is pulverized, sieved, and packaged under a nitrogen atmosphere. The composite regulator can dynamically stabilize and protect saponin molecules through temperature-responsive behavior during processing, significantly reducing thermal degradation. The resulting asparagus powder has a saponin content of 72-85 mg / g and is rich in flavonoids, phenols, and soluble dietary fiber. This product can be used in functional foods that improve sleep. When added at a 4% level to bread, noodles, or meal replacement porridge, animal experiments have verified that it can significantly prolong sleep time and increase the incidence of sleep in mice, exhibiting high nutrient retention and good stability.

The application discloses a sleep apnea severity prediction method based on a single-channel blood oxygen signal and belongs to the technical field of medical artificial intelligence. The prediction method comprises the following steps: acquiring a single-channel blood oxygen saturation signal and demographic characteristics of a target object; performing preprocessing and signal feature extraction operations on the single-channel blood oxygen saturation signal to obtain a signal feature set of the target object; inputting the feature set of the target object into a trained double-branch joint sleep apnea severity prediction model based on multi-scale width learning; outputting a preliminary apnea hypopnea index estimation value through a regression branch; outputting an interval probability distribution through a classification branch; fusing features through a joint regression unit to obtain a final prediction value and a severity classification. The application can realize accurate prediction of sleep apnea severity based on a single-channel blood oxygen signal, has the advantages of low model complexity, strong interpretability and efficient operation on a CPU, and is suitable for clinical portable screening scenes.

The utility model provides an elastic pressing and fixing mechanism for an ion source device. The elastic pressing and fixing mechanism comprises a supporting and assembling component, a matching and sealing component, a bearing and positioning component, a containing and covering component, an elastic stretching component, a switching and extending component and a pressing and positioning component, and the supporting and assembling component, the bearing and positioning component, the containing and covering component and the like of the pressing and fixing mechanism. A precise assembling and positioning environment is jointly provided for the arcing cavity, so that performance fluctuation caused by position deviation or instability is reduced, due to the design of the containing cover protection assembly and the elastic stretching assembly, the elastic stretching assembly is effectively shielded and protected when stretching action is conducted, interference with external equipment is avoided, and the service life of the arcing cavity is prolonged. The press-fit positioning assembly is connected with the arcing cavity through the elastic stretching assembly and the switching extension assembly, the arcing cavity is subjected to press-fit positioning through the elastic force of a stretching spring, the arcing cavity can be properly buffered and adjusted when subjected to pressure, and therefore stress concentration and unstable factors generated by rigid connection are reduced.

The invention provides a cemented filling material with a large mixing amount of fly ash and high shear bearing capacity and a preparation method of the cemented filling material, and relates to the technical field of mine filling. The cemented filling material comprises the following raw materials in percentage by mass: 70-80% of a solid material and 20-30% of water, wherein the total mass of the raw materials is 100%; the solid material comprises the following components in parts by weight: 55-70 parts of fly ash, 2-5 parts of cement, 5-15 parts of slag, 2-5 parts of carbide slag, 21-35 parts of tailings, 0.1-1 part of a fiber material and 1-3 parts of a composite activator. The cementing filling material breaks through three aspects of improving the activity of the fly ash through synergistic excitation modification, optimizing the structure through aggregate particle gradient grading design and reducing performance fluctuation through intelligent process control, so that efficient activation excitation of the volcanic ash activity of the fly ash can be realized, the mixing amount of the fly ash is increased, and the cost of the filling material is reduced; the problem that an existing filling material is low in anti-shearing bearing capacity is solved.

This invention provides a short-process manufacturing technology for high-performance titanium-aluminum alloys, relating to the technical field of titanium-aluminum alloy material preparation. The manufacturing technology uses hydrogenated dehydrogenated powder, aluminum-vanadium alloy, and aluminum-niobium alloy as raw materials. Through hydrogenated dehydrogenation powder preparation, cold isostatic pressing, vacuum sintering, and subsequent processing, a uniformly composed titanium-aluminum alloy is prepared. Utilizing the intrinsic brittleness of titanium-aluminum after its self-propagating reaction, and combining the high purity and easily breakable characteristics of aluminum-vanadium and aluminum-niobium master alloys, the technology achieves thorough mixing and refinement of multiple elements under solid-state conditions. This avoids the microstructure inhomogeneity caused by diffusion limitation in high-vanadium and high-niobium systems, and overcomes the problems of component segregation, coarse microstructure, and difficulty in removing inclusions associated with traditional casting methods for preparing titanium-aluminum alloys. Through the powder metallurgy short-process manufacturing design, this technology offers significant technical advantages in reducing raw material and manufacturing costs, improving compositional and microstructure uniformity, enhancing comprehensive mechanical properties, and achieving large-scale production.

The utility model is applicable to the related technical field of transformers, and provides a copper-clad aluminum high-current output planar transformer, which comprises a magnetic core and a base embedded in the magnetic core, a coil and a PCB (printed circuit board) are further arranged in the magnetic core, and an output piece connected with the PCB is arranged on the base. The total cost of the product is 40% lower than that of a copper pin; the bottom is suspended, and the coil is directly contacted with the air to easily dissipate heat; the secondary output end adopts two rows of copper-clad aluminum pins to realize shunting, so that eddy-current loss can be effectively reduced; used materials are simple, assembly is convenient, and large product performance fluctuation caused by inconsistent materials and manual operation can be reduced to a great extent; the coil and the PCB are alternately combined for four times, so that coupling between the primary and the secondary is enhanced, and the working efficiency of a product is improved; and the output end adopts two rows of copper-clad aluminum pins, the inner row is a straight pin, the outer row is a bent pin, and large-current output is realized by adopting shunt connection.

This invention belongs to the field of intelligent robotics and visual navigation technology, and discloses a spatiotemporal graph-based object navigation method based on target correlation. It constructs a target-related object relationship graph and combines local relationship modeling with a global serialized information propagation mechanism to effectively model object relationships in the environment. Simultaneously, by introducing target correlation to control the information propagation order, node features can be updated progressively according to target correlation, thereby enhancing the modeling ability of long-distance dependencies. Combining temporal feature modeling with action decision-making processes, it achieves efficient robot navigation in complex environments. This invention can utilize the relationship information between objects in target object navigation tasks to improve target orientation capabilities; enhance long-distance dependency modeling capabilities through a serialized information propagation mechanism to improve navigation success rate; and exhibit stronger decision-making stability and environmental adaptability in complex environments and long-path tasks, thus improving overall navigation performance.

This application relates to the field of artificial intelligence technology, and discloses a method and apparatus for collaboratively adjusting the quality of artificial intelligence datasets and model performance. The method includes: one or more iterative operations; the iterative operations include: constructing a preprocessed dataset according to model construction requirements and dataset quality indicator target parameters; performing quality assessment on the preprocessed dataset to obtain dataset quality indicator assessment parameters, and determining a target dataset that passes the quality assessment based on the dataset quality indicator assessment parameters; training the model using the target dataset to obtain a pre-trained model, and performing performance assessment on the pre-trained model to obtain model performance indicator assessment parameters; if the model performance indicator assessment parameters do not meet the preset standard, performing a correlation analysis between dataset quality and model performance based on the model performance indicator assessment parameters, and generating updated dataset quality indicator target parameters based on the analysis results. This achieves collaborative optimization of dataset quality and model performance.

The invention discloses a manufacturing method of a thin-film acoustic metamaterial composite sound insulation board, relates to the technical field of noise control, and aims to solve the problems that an existing thin-film acoustic metamaterial sound insulation board is narrow in sound insulation frequency band, core components are prone to damage, the structure is unstable, the machining precision is low, and the process adaptability is poor. The sound insulation plate comprises a porous sound absorption plate, a micro-perforated plate, a reinforcing grating and a protective layer, the reinforcing grating is of a frame type structure, the porous sound absorption plate is mounted in the reinforcing grating in a filling manner, the porosity of the porous sound absorption plate is linearly decreased from outside to inside, and the micro-perforated plate and the protective layer are sequentially mounted on the outer surface of the reinforcing grating filled with the porous sound absorption plate; the base plate comprises protective layers, clamping plates and a resonance film set, the resonance film set is composed of 2-3 layers of sound insulation films, an elastic damping layer is arranged between every two adjacent sound insulation films, the resonance film set is arranged between the two clamping plates, and the protective layers are arranged on the outer walls of the two clamping plates; the base layer plate is arranged between the two layers of sound insulation plates which are symmetrical in structure, and the reinforcing grating filled with the multi-hole sound absorption plates is connected with the protective layer of the base layer plate. 20-1600Hz broadband efficient sound insulation is realized, the structure is stable and durable, the product consistency is good, the weight is light, and the construction is easy.

The application discloses a finishing method of machine-washable silk fabric, and relates to the technical field of textile fabric finishing control, which is executed by a continuous silk fabric finishing production line comprising a feeding section, a padding section, a pre-drying section, a baking section and a cooling section. The surface texture image of the fabric to be finished in the feeding section is collected, the warp and weft interlacing feature points are extracted and mapped to the padding section roller pressure control grid to generate initial pressure distribution data; the initial pressure distribution data is used to calculate the start-stop state code of the infrared radiation array emission unit in the pre-drying section to control the heat energy output mode; the frequency of the hot air circulating fan in the baking section is adjusted in response to the code, the air pressure feedback value is recorded, the rotating speed parameter of the cooling section cold water roller group is set based on the value, and the whole process is started. The method realizes accurate control and parameter linkage of the finishing section, and ensures the uniform and stable finishing state of the silk fabric.

The invention relates to a steel structure performance prediction method and system based on a multi-modal large model, and belongs to the technical field of steel manufacturing. The method solves the problem that traditional cooling parameter control depends on artificial experience and is low in precision. According to the technical scheme, a historical knowledge base is constructed by collecting steel whole-process technological parameters and mechanical property data, the similarity between to-be-predicted data and historical cases is calculated to screen reference cases, a prompt is constructed and input into a large language model for training prediction, and optimized water tank parameters are output. The system comprises a data collection module, a data processing module, a similarity calculation module, a case screening module, a prompt construction module, a model training prediction module and a deployment maintenance module. According to the method, accurate prediction of the cooling parameters can be realized, and the steel performance consistency and the production efficiency are remarkably improved.