14results about How to "Prevent falling into" patented technology

The invention discloses an energy management balance scheduling optimization method and system based on deep learning, and the method comprises the steps: collecting multi-source feature data, carrying out the standardization and preprocessing, and obtaining structured multi-dimensional time series data; constructing a node relation graph by using an improved dynamic graph attention model, and generating a global attention matrix; node features are updated according to the global attention matrix, optimized features are formed, and a scheduling strategy is generated; optimizing the scheduling strategy by adopting an improved Harlisia eagle algorithm to obtain an optimized scheduling scheme; a node operation strategy is adjusted according to the optimization scheme, and power and load distribution is completed; and a final operation scheme is generated and dynamically corrected. According to the method, the improved dynamic graph attention model is constructed, and the Harris eagle optimization algorithm combining a boundary escape mechanism and a dynamic energy regulation and control mechanism is combined, so that intelligent modeling of multi-source data of an energy network, multi-target optimization of a scheduling strategy and dynamic balance scheduling of energy operation are realized.

The application relates to the technical field of pond sludge cleaning, in particular to a pond sludge cleaning robot, which comprises a vehicle frame, a pressure-resistant cabin is installed at the end of the vehicle frame, a screw rod sliding table is arranged on the surface of the vehicle frame, a rotating screw rod is arranged on the surface of the screw rod sliding table, a screw rod seat is arranged on the surface of the rotating screw rod, a waterproof motor is installed on the surface of the screw rod seat, a pump body is installed in the vehicle frame, a suction pipe is arranged at the end of the pump body, and the robot is placed on the upper portion of the robot running on the bottom of the pond through the pressure-resistant cabin. The motor in the pressure-resistant cabin drives the power wheel to rotate, the caterpillar belt moves around the power wheel, the load wheel and the tensioning wheel, the vehicle frame is driven to move, the waterproof motor drives the reamer at the bottom to rotate during the movement of the robot on the bottom of the pond, the sludge at the bottom of the pond is broken, then the pump body sucks the sludge out of the pond through the suction pipe, and the waterproof motor is installed on the surface of the screw rod seat, and the screw rod seat can be lifted on the surface of the screw rod sliding table.

This invention provides an oxygen supply device for aquatic ecological restoration, relating to the field of oxygen supply technology for aquatic ecological restoration. The device includes an aerator shell with a chassis structure at its lower end. Four bases are arranged in a rectangular pattern on the upper part of the outer wall of the aerator shell. Multiple cyclone structures are arranged vertically on the upper part of the aerator shell. A protective structure is located at the upper end of each cyclone structure. An auxiliary structure is located at the upper end of each cyclone structure inside the protective structure. The auxiliary structure includes an umbrella-shaped metal seat, and a rubber skirt is fixedly connected to the outer wall of the umbrella-shaped metal seat at the upper end of each cyclone structure. When external water pressure is applied downwards, the umbrella-shaped rubber pad is first pressed firmly against the aerator output end, forming a first flexible seal. Simultaneously, the water pressure is transmitted along the shaft to the spring and the cover plate. The greater the water pressure, the tighter the cover plate is pressed under the cooperation of the spring, preventing backflow.

The utility model discloses an injection molding machine convenient to clean, which belongs to the technical field of injection molding machines and comprises a base, the top of the base is fixedly connected with a mold closing box and an injection molding machine body respectively, and a male mold is arranged on one side of the mold closing box. The cleaning assembly is arranged, the first servo motor drives the rotating shaft to rotate, the rotating shaft drives the scraping strip to rotate through the rotating rod, the scraping strip is tightly attached to the inner wall of the hopper, cleaning dead corners do not exist, waste can be thoroughly removed, automatic cleaning is achieved, manpower and time cost are saved, and meanwhile the cleaning assembly can prevent the hopper from being blocked when the injection molding machine works. By arranging the lifting assembly, the lifting assembly drives the hopper and the injection molding machine body to be separated through the connecting block and then ascends, and the hopper is driven to rotate by a certain angle after ascending by a certain distance, so that waste materials fall aside when the hopper is cleaned, the waste materials are effectively prevented from falling into the injection molding machine, and the product quality problem caused by mixing of the waste materials is avoided.

The utility model discloses a kind of spliced floor mats for sand ground, it is related to floor mat technical field, including floor mat main part, the floor mat main part bottom inner wall is fixedly installed with web plate, the web plate bottom is provided with pad plate, the floor mat main part two adjacent sides are all provided with two installation grooves, the floor mat main part other two sides are all fixedly installed with two mounting blocks, connecting component, the connecting component is installed on pad plate, and for connecting pad plate and web plate. The utility model, pad plate is embedded and installed to the bottom of floor mat main part by connecting component, pad plate is separated web plate and sand ground contact, and is used as supporting floor mat main part, can prevent when floor mat main part is trodden, loose fine sand sinks into web plate and forms local cavity, to improve stability, again by setting second drain hole in the top of pad plate, and cooperate with gap drainage, so that ponding can pass through second drain hole and is discharged, avoid ponding too much lead to unstable state of floating.

This invention discloses a suspended fertilization device that integrates ditching and fertilization, relating to the field of ditching and fertilization technology. It includes a ditching device with a suspension rod connecting plate fixedly connected to its rear side. Suspension rods are fixedly connected to the left and right sides of the rear side of the suspension rod connecting plate, and a fertilizer storage bin is fixedly connected to the top of the ditching device. This invention, by incorporating a ditching device, ditching frame, rotating rod, and ditching mechanism, effectively improves the adaptability and stability of ditching and fertilization operations. During tractor turning, the ditching device of the integrated ditching and fertilization machine remains stable, avoiding impact and torsion caused by excessive turning radius, thereby extending the equipment's service life. Specifically, after ditching is completed, the ditching device needs to be retracted. The motor rotates, the turntable drives the track to rotate the second turntable, which in turn drives the rotating rod and rotating rod connecting pipe to rotate, causing the ditching mechanism to rotate along the arc-shaped slide groove. The slider slides, and the ditching wheel is stored inside the outer casing.

The utility model relates to an anti-lamination structure of a blister box, which belongs to the technical field of blister box packaging and comprises a first blister box component, a second blister box component and a third blister box component, the first blister box component is provided with a plurality of accommodating cavities, and the accommodating cavities include at least a first accommodating cavity, a second accommodating cavity and a third accommodating cavity; the second blister box assembly is arranged on the first blister box assembly, the second blister box assembly is provided with adaptation cavities, and the multiple adaptation cavities are at least a first adaptation cavity, a second adaptation cavity and a third adaptation cavity; the distance between the first containing cavity and the second containing cavity is L1, the distance between the second containing cavity and the third containing cavity is L2, the distance between the first adaptive cavity and the second adaptive cavity is L3, and the distance between the second adaptive cavity and the third adaptive cavity is L4. According to the anti-lamination structure of the blister box, L1 is not equal to L3, and L2 is not equal to L4, so that the problems of adhesion between different layers, difficulty in taking and even product damage are effectively avoided.

This invention provides a microfluidic reagent card with a membrane support structure, relating to the field of medical testing. It includes: a reagent card body, a reagent card upper membrane mounted on the surface of the reagent card body, an injection port on the surface of the reagent card upper membrane, a sample compartment body at one end of the top of the reagent card body, and a support structure body mounted on the bottom of the inner wall of the sample compartment body. In this invention, the support structure body mounted on the inner wall of the sample compartment body at one end of the top of the reagent card body supports the reagent card upper membrane located above the sample compartment body, preventing it from sinking into the sample compartment body. Excess sample enters the waste liquid tank through a passage three between the quantitative tank and the waste liquid tank, and residual plasma enters the residual plasma tank through a passage four between the flow tank and the residue tank, thereby treating excess sample and residual plasma.

The utility model discloses a desert nursery stock transport vehicle based on hydraulic self -adaptation chassis, including self -adaptation chassis, car, guide rail frame, snatch mechanical arm and radar etc. Self -adaptation chassis is equipped with the walking wheel of independent height adjustable and independent steering system, can adapt to the complex desert topography, car is used for loading nursery stock, guide rail frame and snatch mechanical arm realize the automation nursery stock snatch, and radar provides accurate positioning. The utility model solves the problem that the traditional transport vehicle is poor in desert environment, and the positioning is not accurate, and the operation efficiency is low, and has the advantages of strong topographic adaptability, accurate positioning, high degree of automation operation, good nursery stock protection, especially suitable for the nursery stock transport operation in desert greening engineering.

The invention discloses a cue word generation method and system based on large language model semantic reflection and multi-objective evolution. The method comprises the following steps: initializing a cue word population; calculating a performance index of each cue word, and performing semantic segmentation by using a large language model to calculate an abstraction degree index; performing comprehensive evaluation based on the performance and the abstraction degree, and screening elite individuals by adopting a hierarchical greedy strategy based on a Pareto level; performing semantic reflection variation or parent fusion crossover on elite individuals by using a large language model to generate a next generation population; and the above process is iterated until a termination condition is met. According to the method, the abstraction degree index and the semantic reflection mechanism are introduced, so that the problems of easy overfitting and low variation efficiency in the traditional cue word optimization are solved, and the cue word with high performance and strong generalization ability can be automatically generated.

The utility model belongs to the field of drying and storage of battery raw materials, and particularly relates to a lithium iron phosphate battery raw material storage tank which comprises a material storage tank, a heater, a dust falling structure and a stirring structure, raw materials can be attached to and accumulated on the inner wall of the material storage tank in the stirring and drying process, and when the accumulation thickness exceeds a gap between an inner wall scraping thorn and the inner wall of the material storage tank, the dust falling structure can fall off. The tips of the scraping thorns on the inner wall can scrape off raw materials attached to the inner wall of the storage tank, the multiple layers of dust falling nets are arranged in a stacked mode, when rising raw material particles pass through all the layers of net holes, large particles can be intercepted by the lower layer of dust falling net, and small particles continue to rise along with airflow and are intercepted by the upper layer of dust falling net step by step. The risk that raw material particles are discharged into the working environment and sucked by workers is avoided, the sharp ends of the net scraping thorns hit meshes of the dust falling net, so that the dust falling net vibrates, the raw material particles on the dust falling net are shaken off into the material storage tank, and waste of raw materials caused by escaping of the raw material particles is prevented.

This invention provides a method for cogeneration load allocation in cogeneration power plants based on an improved gray wolf algorithm, belonging to the field of cogeneration system optimization and control technology. The method includes: S1: setting parameters; S2: initializing the population based on chaotic mapping; S3: calculating fitness values ​​and selecting superior individuals; S4: stratifying the population based on the fitness calculated in step S3, dividing the population into different levels, and adopting different position update methods according to the characteristics of different levels; S5: iteration and outputting results. This method exhibits strong real-time performance in cogeneration load allocation, possesses a robust multi-objective trade-off mechanism, can balance local and global search capabilities, and effectively solves optimization problems in complex scenarios of cogeneration power plants.