10results about How to "Increased propulsion speed" patented technology

The invention discloses a multi-unmanned aerial vehicle cooperative search method based on dynamic region division and a multi-constraint virtual force field, and aims to solve the problems of low search efficiency, unbalanced load, difficulty in guaranteeing the motion safety of multiple unmanned aerial vehicles and the like in multi-unmanned aerial vehicle coverage search. Firstly, a task area, an unmanned aerial vehicle kinematics model and a sensor sensing model are constructed; then, dynamic region division is carried out based on a weighted Voronoi diagram, the search progress of the unmanned aerial vehicle, the responsibility region area and the obstacle distribution condition are comprehensively considered, and self-adaptive balancing of task loads is achieved; and selecting a target point based on a utility function, and finally designing a multi-constraint virtual force field control method consisting of target attraction, obstacle avoidance force, regional barrier force and exploration force, generating an expected speed meeting speed constraint and obstacle avoidance requirements, and realizing stable and efficient coverage search. The result shows that the multi-machine collaborative search efficiency can be effectively improved, and the method is suitable for disaster search, routing inspection monitoring, autonomous detection in a complex environment and other scenes.

The application provides a kind of deep learning optical imaging system design method based on complex function neural network, comprising the following steps: step one, object distance, system focal length, F number as input, according to the ideal optical system imaging relationship, determine image distance, imaging scale ratio;Step two, according to the imaging scale ratio and the object image of ideal optical system, the image of ideal optical system is calculated;Step three, construct complex function generator network model, step four, construct discriminator network model, step five, after the establishment, training of the above-mentioned complex function neural network deep learning model, step six, use phase.The deep learning optical imaging system design method based on complex function neural network provided by the application solves the correlation between the advantages and disadvantages of the design result and the experience value of the designer, without the designer providing initial structure parameters;Greatly improve the computing ability and optimization speed.

The invention discloses a transformer substation site selection data processing and feedback method, system and device based on unmanned aerial vehicle-AI cooperation and a medium, and belongs to the technical field of intelligent site selection of electric power engineering, and the method comprises the steps: controlling an unmanned aerial vehicle to shoot a transformer substation site selection region according to a preset flight route, and obtaining a site selection region image; performing preprocessing and feature extraction on the site selection area image to obtain key environment element information; inputting the key environment element information into a pre-trained site selection environment evaluation model to obtain an evaluation result, including a terrain suitability grade, infrastructure proximity and a comprehensive site selection feasibility score; according to the evaluation result and a preset engineering constraint rule base, matching an avoidance suggestion; and displaying the avoidance suggestions. According to the method, site selection suggestions can be provided for decision makers in a shorter time, the advancing speed of substation construction projects is increased, the advantages and disadvantages of site selection can be judged more objectively and accurately, and the risk of site selection errors caused by human factors or insufficient data is reduced.

The utility model relates to photovoltaic panel device technical field, and disclose a kind of quick installation type photovoltaic panel device, including mounting seat, four rectangular equidistant distribution slide rail box are fixedly connected in mounting seat outer side wall, the inner wall one end of slide rail box is slidably connected with sliding arm, the one end output of sliding arm extends to the inner wall one end of slide rail box, the top of mounting seat is fixedly connected with two equidistant distribution spring seat, the inner wall bottom of spring seat is fixedly connected with two springs, the top of spring is fixedly connected with pressboard, and the device is realized by the cooperation of above-mentioned structure, the quick, accurate installation of photovoltaic panel is realized, installation efficiency is effectively improved, installation accuracy and stability are guaranteed, the frequency of later maintenance and replacement is reduced, maintenance cost is reduced, with significant economic benefits and application value.

This invention provides a method for accelerating real-time inference of multimodal large models, relating to the field of data processing technology. The method includes: Step 1, generating a minimum envelope ellipse using the peak point of the first visual attention weight, the second visual coordinate point, and the third visual association point as boundary points, and calculating the ratio of the area to the perimeter of the ellipse as a spatial clustering index. This invention effectively improves the efficiency, accuracy, and resource utilization of real-time inference of multimodal large models by selectively enhancing spatial features, evaluating multidimensional constraint computing power and dynamically scheduling resources, lightweight cross-modal fusion, and multi-step inference attention calibration, combined with real-time decision-making and full-process data-driven evaluation parameters and dynamic updates of the model library. This ensures the real-time performance and reliability of inference results, while continuously optimizing system computing power evaluation and model adaptation, making multimodal inference more aligned with the resource constraints of heterogeneous computing and the real-time and accuracy requirements of actual business operations.

The present application belongs to the technical field of railcar management, and specifically relates to a method and system for guiding and assisting before shunting operation of a freight train, which comprises collecting real-time front approach environment data through a multi-modal sensing module deployed at the front end of the shunting operation during the shunting and pushing operation of the freight train; and transmitting the front approach environment data to an edge computing node; the present application realizes accurate matching of visible light images and radar point cloud data through multi-modal fusion collection of a three-lens high-definition camera unit and a millimeter wave radar, combined with millisecond-level space-time synchronization technology, effectively making up for the visual blind area of traditional manual lookout in adverse weather such as rain, snow and fog, and at night and in complex station environment. Among them, the accurate measurement of obstacle distance and relative speed by the millimeter wave radar, combined with image visual feature capture, expands the environmental perception coverage range, reduces the obstacle identification missed detection rate, and completely solves the core pain point that the accuracy of manual judgment is greatly affected by the environment.

The invention discloses a control method and system of a vertical shaft heading machine, and is applied to the field of vertical shaft heading machines, the method comprises the following steps: controlling the vertical shaft heading machine to adjust the working state thereof by using a received control signal, the control signal generation process comprises the following steps: obtaining geographical environment change information, and determining the working state of the vertical shaft heading machine based on the geographical environment change information; determining the stratum sudden change intensity of the working area; determining an energy consumption disturbance index at least based on the stratum sudden change intensity; processing the energy consumption disturbance index to obtain a predicted energy consumption change rate of the working area; determining energy consumption prediction information based on the predicted energy consumption change rate and the corresponding stratum sudden change intensity; acquiring electrical sampling data; verifying initial risk probability data determined based on the electrical sampling data to generate fault risk information; and generating a control signal based on the energy consumption prediction information and the fault risk information. According to the control method and system of the vertical shaft heading machine, the parameter adjustment accuracy is remarkably improved through accurate analysis of data.

This invention discloses an efficiency optimization control method for a high-speed brushless DC motor system based on time-sharing commutation. It establishes a time-sharing commutation mode to determine the variation law of the motor phase current; obtains the turn-on offset angle and turn-off offset angle based on the maximum electromagnetic power using offline numerical solutions; designs the input layer, hidden layer, and output layer structure of an RBF neural network; performs offline pre-training of the RBF neural network using prior information and online training; updates the neural network parameters by acquiring real-time motor system efficiency information; and applies the optimal turn-on offset angle and optimal turn-off offset angle output by the neural network to the time-sharing commutation control, thereby improving the motor's operating efficiency. The advantages of this invention are that it is not limited to the 120° conduction mode of the brushless DC motor windings, but establishes a time-sharing commutation mode; and it utilizes the RBF neural network to quickly optimize the turn-on offset angle and turn-off offset angle, thereby improving the motor system efficiency.

The application provides a two-dimensional separated sparse array, and relates to the technical field of antenna engineering. The two-dimensional separated sparse array is arranged horizontally and vertically, and array arrangement is neat; the sub-arrays in four corners and the sub-arrays in the middle cross region can be integrated according to rows or columns, which is convenient for the design of the rear-end feed network and the modular installation; the two-dimensional separated sparse array is formed by nesting a first basic sparse array and a second basic sparse array; the first basic sparse array and the second basic sparse array can be decomposed into sparse linear arrays for optimization; the number of optimization variables is reduced from 2MxN to 2(M+N) after decomposition, and only the directional diagram of a section needs to be calculated during optimization, which greatly reduces the number of optimization variables and the calculation amount, and greatly improves the optimization speed; the sidelobes of the two basic sparse arrays have a certain cancellation effect, which can further reduce the maximum sidelobe level of the overall sparse array, and has the characteristics of low three-dimensional space sidelobe level.

The invention relates to an irradiation plan optimization method based on novel sparse sampling, belongs to the technical field of radiation irradiation dose optimization, and solves the problems that invalid voxel data volume cannot be effectively simplified, the processing data volume is large and the result is not accurate enough in the prior art. Receiving an initial irradiation plan and generating a full dose deposition matrix corresponding to each radiation field; constructing a gradient image based on the full dose deposition matrix; performing dual-condition voxel screening on the constructed gradient image to obtain a final sampling voxel set, and constructing a sparse dose deposition matrix; and performing irradiation plan optimization and dose normalization on the sparse dose deposition matrix to obtain an optimized irradiation plan.