8results about How to "Reliability advantage" patented technology

The invention relates to the technical field of electric power Internet of Things, in particular to a time-space hierarchical data compression method and system for multi-source heterogeneous data on a power distribution network side, and the method comprises the following steps: S1, carrying out the classification and lightweight compression processing of original data, and outputting intermediate data with a priority identifier; s2, based on the intermediate data obtained in the step S1, a gateway layer carries out topological grouping according to the electrical distance, and then carries out time alignment and weighted fusion on the data in the group to generate a fusion data packet; and S3, based on the fusion data packet obtained in the step S2, the cloud platform performs panoramic insight and value evaluation, generates and dynamically issues an optimization strategy instruction, and the strategy instruction is dynamically issued back to the edge layer and the gateway layer through a control channel. According to the method, the real-time advantage of edge calculation and the big data processing capability of a cloud platform can be fully exerted, modules of all levels work cooperatively, a set of complete power distribution network multi-source heterogeneous data processing solution is formed, and the overall efficiency of the system can be effectively improved.

The application provides a two-stage MPC control method and system for power smoothing output, and relates to the field of new energy control. A discrete state space prediction model is constructed; all state variables are obtained in the current sampling period and used as the input of the first-stage MPC control; the mechanical and electrical state prediction sequence and the optimal control input are obtained by combining the constraint condition and the dynamic optimization objective function of the second-stage MPC control feedback in the last sampling period; in the second-stage MPC control, the optimal power trajectory meeting the smoothing requirement is obtained by combining the smoothing optimization objective function, and is mapped to the control constraint interval of the next sampling period and fed back to the first-stage MPC control; the optimal control input is applied to the execution equipment, and after the constraint condition is fed back to the first-stage MPC control, the next sampling period is entered. The application can inhibit the rapid fluctuation of the generated power without energy storage devices, improve the grid connection quality, and enable the renewable energy power generation system to still stably output under the severe changes of the external environment.

The application discloses a multi-complexity behavior recognition system and method based on adaptive feature extraction, and relates to the technical field of artificial intelligence, comprising a human body behavior data acquisition module, a human body behavior data transmission module, a human body behavior data storage module, a human body behavior data preprocessing module, an MFAEF simple human body behavior recognition network module, an LSGRA complex behavior recognition network module and a human body behavior information application module; the MFAEF simple behavior recognition network module comprises a feature pre-extraction unit, a parallel multi-dimensional space-time feature extraction and reuse unit and an adaptive multi-dimensional space-time feature extraction unit, a feature fusion unit and a simple behavior discrimination output unit which are sequentially connected; and the LSGRA complex behavior recognition network module comprises a single-window simple behavior feature acquisition unit, a cyclic multi-window attention unit and a complex behavior discrimination output unit which are sequentially connected. The application adopts the above structure to make up for the deficiencies of high cost, easy interference and poor privacy of the behavior recognition based on vision.

The application discloses a graphite-based three-dimensional heat-conducting structure with continuous Z-direction heat-conducting framework and a preparation method thereof. The structure comprises a plurality of graphite structure units and a high-thermal-conductivity interface layer, and the in-plane high-thermal-conductivity direction of the graphite structure units is perpendicular to the main plane of the structure, and the adjacent units are bonded and integrated through the high-thermal-conductivity interface layer. The plurality of graphite structure units are arranged in a bricklaying staggered manner in the main plane, and the staggered amount is 25% to 75% of the width of the unit. The preparation method comprises the following steps: rotating the graphite material by 90 degrees after processing into a strip to orient the in-plane high-thermal-conductivity direction to the Z direction, coating the interface material, and then stacking in a bricklaying staggered manner, and bonding into a whole through hot pressing or segmented sintering. The application realizes efficient Z-direction heat conduction by directly utilizing the intrinsic in-plane thermal conductivity of the graphite material, and the Z-direction thermal conductivity can reach 500 to 800 W / m·K. The bricklaying staggered structure gives excellent mechanical interlocking effect and thermal reliability, and is suitable for the thermal management scene of high-power-density electronic devices.

The utility model discloses a kind of electrically conductive guide rail devices suitable for different thickness layer board, it is related to electrically conductive relevant field, including electrically conductive guide rail assembly, cabinet body assembly, fixed assembly, electrically conductive guide rail assembly is fixedly connected in cabinet body assembly inside, cabinet body assembly rear end is fixedly connected with fixed assembly, electrically conductive guide rail assembly further includes: guide rail frame, guide rail frame is fixedly connected in cabinet body assembly inside;Movable sliding slot, movable sliding slot is located in guide rail frame guide rail frame inside;By setting movable sliding slot and spring structure, the elastic floating of electrically conductive strip is realized, and the insertion of self-adapting different thickness layer board and keep stable electric connection;Fixed assembly uses electric push rod drive limit frame, realizes the quick locking and release of layer board, improves installation efficiency and operation convenience, overall structure has remarkable advantage in mechanical adaptability, electrical reliability and automation operation aspect, provides efficient solution for the flexible power supply of layer board in cabinet.

The utility model discloses a child seat with an adjustable backrest length. The child seat comprises a seat assembly. The backrest piece can rotate relative to the seat assembly so as to adjust the inclination angle of the backrest; the backrest telescopic part is mounted on the backrest part in a sliding manner and is used for adjusting the effective length of the backrest; one end of the connecting belt is connected with the backrest telescopic piece, and the other end of the connecting belt is connected with the seat assembly; one end of the reset piece is connected with the backrest telescopic piece, and the other end of the reset piece is connected with the backrest piece or the seat assembly; the connecting point of the connecting belt and the seat assembly is different from the rotating point of the backrest piece and the seat assembly; when the backrest piece rotates, the backrest telescopic piece can slide relative to the backrest piece through cooperation of the connecting belt and the reset piece, and the effective length of the backrest is automatically adjusted. Through the synergistic effect of the connecting belt and the reset piece, the backrest telescopic piece can automatically slide along with rotation of the backrest piece, the effective length of the backrest is automatically adjusted, and comfort and convenience are improved.

This invention relates to a multi-state transport device with a synchronous containment structure, comprising a main structure, deformable units respectively disposed on both sides of the main structure, movable support components respectively associated with the deformable units, and an additional synchronous structure for applying synchronous retraction drive to the deformable units on both sides during containment. When transitioning from an expanded state to a containment state, the synchronous structure applies synchronous retraction drive to the deformable units on both sides in the containment direction, forming motion coupling and establishing a motion linkage retraction relationship between the deformable units on both sides, forcing the deformable units on both sides to contain synchronously, and the movable support components on both sides synchronously retract from the ground support position to the containment position. The synchronous structure does not apply synchronous constraints to the deformable units on both sides during deployment; it only functions during containment. This invention achieves smooth and coordinated containment through a two-way mechanism of natural gravity synchronization during deployment and forced synchronization by the synchronous structure during containment, resulting in a simple and reliable structure.

The invention relates to the technical field of spectral analysis and ore component detection, and discloses an LIBS (laser-induced breakdown spectroscopy) spectrum grade prediction method based on multi-head attention. A multi-scale extraction module is constructed based on one-dimensional convolution of different convolution kernel sizes, a feature extraction module is constructed based on rotation position coding and multi-head attention, and a spectrum grade prediction model is constructed based on the multi-scale extraction module, the feature extraction module, residual connection and a multi-layer perception mechanism; the method comprises the following steps: acquiring spectral data of an ore component detection site, preprocessing the spectral data to obtain spectral features, inputting the spectral features into a spectral grade detection model, extracting local multi-scale features by a multi-scale extraction module based on the spectral features, acquiring global features from the local multi-scale features by a feature extraction module, and extracting the global features from the local multi-scale features; obtaining fusion features based on the global features and the local multi-scale features in combination with residual connection, and obtaining an ore grade prediction result based on the fusion features and the multi-layer perceptron; the problem that an existing calculation model is difficult to carry out grade prediction based on spectral data is solved.