87results about How to "Solve the selection problem" patented technology

The invention relates to the field of the plan and the design of energy systems, and discloses a multi-energy complementary distributed energy system and an equipment configuration and operation optimization method thereof. According to the multi-energy complementary distributed energy system and the equipment configuration and operation optimization method thereof, energies such as fossil energy, unemployed energy, and renewable energy etc. are converted by the adoption of energy conversion technologies such as the conventional generation technology, the renewable energy technology, the cogeneration technology or the thermal technology etc., the requirements of terminal loads such as electric power, refrigeration, heating, and hot water etc. are satisfied, defects of single energy carrier and single energy supply technology in the prior art are overcome, the coupling utilization mode of a plurality of distributed energy supply technologies is provided, and energy complementation is formed. Further, during system equipment configuration and operation optimization, an established optimization model comprehensively considers equipment performance, policy indicators or economic indicators and combines economic, environmental, and policy boundary conditions, the problem of decoupling of equipment type selection and operation strategy in the prior art is solved, and scientific basis is provided for relevant policy making.

The invention discloses a three-dimensional reconstruction method based on multi-view self-adaptation characteristic registration. The method comprises the following steps that: A, seed points in each two-dimensional picture are selected; B, a region growing method is adopted to perform image segmentation on the two-dimensional picture so as to obtain the boundary point set of each image of each picture; C, through combining the boundary point sets of the segmented images and a polar geometry constraint theory, three-dimensional registration is performed on two images of which the angles are close to each other in a self-adaptation manner; and D, and three-dimensional reconstruction is performed according to the parallax images of boundary points. According to the three-dimensional reconstruction method based on the multi-view self-adaptation characteristic registration of the invention, the region growing method for image segmentation and the polar geometry constraint theory are fully utilized, such that requirements for characteristic point selection and three-dimensional registration can be satisfied. The three-dimensional reconstruction method is applicable to the implementation of three-dimensional reconstruction systems.

The invention discloses a method for identifying partial discharge of a transformer based on a particle swarm optimization kernel nearest neighbor propagation algorithm, which comprises the followingsteps: establishing an oil-paper insulation partial discharge test model of the transformer; Extraction of Moment Features, Typing Features and Texture Feature Parameters from Gray Image; Dimension Reduction of Characteristic Parameters in Principal Component Analysis; The Calculation Formula of Similarity Based on Kernel Function and Shared Nearest Neighbor; Methods and steps of classifier basedon particle swarm optimization kernel nearest neighbor propagation algorithm. The invention improves the shortcoming that the traditional nearest neighbor propagation algorithm is only suitable for dealing with the data clustering problem of a compact hyper-spherical structure, and still has certain validity when the data set is loosely distributed or the structure is complex. The experimental results show that the kernel nearest neighbor propagation algorithm based on particle swarm optimization can significantly improve the recognition rate compared with the traditional nearest neighbor propagation algorithm and improve the recognition rate of part of the model data compared with BP neural network when it is applied to the identification of the four types of partial discharge.

The invention discloses a hook-shaped steel core composite connecting piece prefabricated sandwiched heat insulation wallboard and a manufacturing method thereof .Internal and external acanthuses are formed by pouring concrete, a middle sandwich layer is a heat insulation board, and the three-layer structure is connected into a whole through hook-shaped steel core composite connecting pieces, wherein the hook-shaped steel core composite connecting pieces are hook-shaped reinforcing steel bar core continuous fiber composite connecting pieces and high-strength steel wire core model plastic compression molding steel core composite connecting pieces .The steel core composite connecting pieces of each kind are of a straight insertion type and an oblique insertion type .The internal and external concrete acanthuses are connected together through the straight insertion type hook-shaped steel core composite connecting pieces in a 90-degree orthogonal mode, and hook heads of the connecting pieces can hook an eternal acanthus internal reinforcing steel bar net; the internal and external wall acanthuses are connected together through the oblique insertion type steel core composite connecting pieces in a 45-degree oblique crossing mode .The prefabricated concrete sandwiched heat insulation external wall has the advantages of being high in anti-shear and anti-bending performance, good in heat insulation effect, high in construction efficiency, low in manufacturing cost, safe and reliable to use and the like .

The invention belongs to the field of information, and particularly relates to an influential communicator recognizing and quantifying method in an online social network. The social network is abstracted as a graph G (V, E) on the basis of an SIR family communication model, wherein V represents the set of nodes, E represents the set of edges, and each edge indicates the relationship between the nodes. The communication capacity S of all the nodes in the graph G (V, E) is calculated by using a greedy algorithm based on the percolation theory, and the size of the communication capacity S of allthe nodes is ordered so that the top M vertexes of the highest communication capacity are found. The beneficial effects of the method are that the global network communication influence is measured through the local network structure so that the time complexity of the algorithm can be reduced.

The invention discloses a production device and method for recovering sulfur from sulfur slag. The device comprises a material preparation unit, a conveying unit, a pretreatment unit, a melting unit and a thermal filtration unit. The material preparation unit comprises a crusher, a vibrating screen and a raw material bin. The conveying unit comprises a feeder and a conveyor belt. The pretreatment unit comprises a mixing device and/or a kneading device. The melting unit comprises a sulfur slag stirring and melting tank. The thermal filtration unit comprises a submerged pump and a filtration device. The method includes the steps of (1) material preparation, wherein the sulfur slag is crushed, screened and stored in the raw material bin; (2) pretreatment, wherein the sulfur slag conveyed by the conveying unit is pretreated; (3) melting, wherein the sulfur slag is stirred and molten in the sulfur slag stirring and melting tank; (4) thermal filtration, wherein the molten sulfur slag is filtered, and molten sulfur and filter residues are obtained. By means of the production device and method, direct dipping and flotation sulfur slag structure transformation and device model selection are achieved, and the elemental sulfur content in the obtained molten sulfur is lager than or equal to 99.52%. The production device is easy to operate, high in flexibility and suitable for industrial production.

The invention discloses a method for determining orbit position of a geosynchronous earth orbit satellite, which comprises the steps of determining a to-be-selected arc section and frequency resource demand of the satellite; inputting more than one frequency resource demand, and sequentially forming one or more policies according to importance of the frequency resource demand; determining a special orbit position in the to-be-selected arc section, and further determining frequency resource which can be obtained most probably and is related with the orbit position; selecting a plurality of to-be-selected orbit positions based on an orbit position isolation principle according to the present satellite network information and on-orbit satellite information, wherein influence caused by the special orbit position has to be considered; calculating values of all the to-be-selected orbit positions according to determined policies; carrying out quantitative ordering of all the to-be-selected orbit positions by a grey correlation degree method, and selecting the optimal orbit position according to result of the quantitative ordering. The method for determining orbit position of the geosynchronous earth orbit satellite can effectively determine the optimal orbit position of a satellite system under a multi-target condition, thereby realizing maximum compatibility and accomplishing maximum feasibility of coordination of a to-be-declared satellite network and an existing satellite network.

The embodiment of the invention provides an aerosol secondary inversion method based on PARASOL multi-angle polarization data. The method comprises: constructing an continental aerosol type lookup table and an urban aerosol type lookup table respectively by utilizing a vector 6S radiation transmission model; meanwhile, obtaining different combinations of aerosol component proportions under the background of the continental aerosol and the urban aerosol by utilizing an orthogonal test method; constructing a component lookup table on the basis of the PARASOL multi-angle polarization data, then selecting the preprocessed PARASOL multi-angle polarization data as inversion data, determining an aerosol type closest to each pixel through an aerosol type lookup table, and obtaining a first inversion result; and selecting a corresponding component lookup table according to the aerosol type of each pixel determined in the first inversion to determine the AOD and the component proportion of eachpixel, and obtaining a second inversion result. According to the method, a default aerosol type in a radiation transmission model is replaced by secondary inversion, so that inversion errors caused byaerosol model selection are effectively reduced, and the inversion precision of the aerosol optical thickness is improved.

The invention discloses a modeling and detecting method of a behavior sequence based on HDP-HMM. A logarithmic transition probability of a state trajectory is used as a judgment index, thereby being simpler, two models are established to jointly determine the detection result from the positive and negative angles, and the problem that judgment index parameters are too complex to be selected and calculated in the previous detection method is solved. Compared with the prior art, the modeling and detecting method disclosed by the invention has the advantages of obtaining a better detection resultwhile compensating the defects of the HMM on model definition and parameter estimation, and it can be seen from experiments that the average detection rate of the modeling and detection method in theinvention is 95.3%.

The invention discloses a device and a method for detecting mixed gas decomposition products of perfluoro-isobutyronitrile and carbon dioxide. The device comprises a gas decomposition module and a gasdetection module, wherein the gas decomposition module comprises an alternating current power source, a voltage regulator, a transformer, a resistor, a capacitor, a rectifier diode, a barometer, a closed cavity, a needle plate electrode, a gas valve and a micro-water generating device. The gas detection module mainly includes a gas bottle, a pressure reducing valve, a gas purifier, a GC-MC gas chromatography mass spectrometer, a gas collecting bag, an exhaust gas recovery device and a computer. During the stage of a discharge test, the closed cavity is filled with a small amount of water andoxygen and a proportion of mixed gas of the perfluoro-isobutyronitrile and carbon dioxide, the needle plate electrode of the cavity is subjected to pressure boosting by a high-voltage power source, after the corona discharge occurs, pressure boosting is stopped, and the voltage for 12 hours is maintained; and in the stage of gas detection, the gas collecting bag collects gas to be tested and the gas chromatography mass spectrometer is turned on, the gas in the gas collecting bag is injected into an instrument for detection, and the gas to be tested is subjected to the qualitative and quantitative treatment.

The invention provides a BGA package chip test clamp. The test clamp is a clamp plate formed by a plurality of layers of printed-circuit boards. The top surface and the bottom surface of the clamp plate are respectively provided with a BGA bonding pad consistent with pins of a BGA package chip to be tested; the top surface of the clamp plate is provided with test points; the bonding pads on the top surface and the bottom surface of the clamp plate are in wiring communication through printed-circuit board via holes and the printed-circuit boards; when the BGA chip to be tested is packaged on the top surface of the clamp plate, the bottom surface of the clamp plate is packaged on a main board using the BGA chip to be tested; and the test points and signal lines needing testing in the bonding pads are in wiring communication through the printed-circuit boards. According to the BGA package chip test clamp, the test clamp can realize more accurate measurement result, and the cost is low.

The invention provides a model range value taking method of the excavation numerical value simulation analysis of deep-buried tunnel construction. The model range value taking method comprises the following steps: establishing numerical value calculation models under different model ranges, and carrying out the excavation numerical value simulation of a tunnel; calculating a displacement error, extracting an integral displacement error, adopting a least square method to analyze influence on the integral displacement error by model ranges of different directions, and discussing a function relationship between the integral displacement error and the model ranges; and on the basis of the function relationship between the integral displacement error and the model ranges, limiting a certain integral displacement error allowable value, and taking model range minimum as a target to determine a model range optimal value. The model range value taking method can simultaneously consider the influence of model sizes of different directions so as to determine the optimal model range, can be widely applied to the excavation numerical value simulation of the deep-buried tunnel construction, and establishes and provides guidance for a numerical value model of the deep-buried tunnel.

The invention provides a Switch rectifier multi-objective optimization design method based on an NSGA-II algorithm, and belongs to the technical field of rectifier parameter optimization design. The method comprises the following steps: 1, determining an optimized component and carrying out power and volume modeling; 2, defining a target function; 3, initializing a population; 4, selecting; 5, selecting binary match; 6, crossing; 7, realizing variation; 8, combining the parent population and the offspring population to obtain a first generation population; 9, forming populations by utilizingthe parent population and the offspring population; 10, reaching the maximum evolution algebra gen; and 11, outputting a Pareto optimal solution leading edge and a parameter matrix, according to the selected target performance indexes, obtaining corresponding decision variable values, namely solutions of the optimization problem, and selecting and designing components according to the parameters.

The invention discloses calibration methods for relative positions of a three-dimensional sensor and a mechanical arm. A method comprises the steps of S1, moving the mechanical arm, thereby enabling aflange plate at a tail end of the mechanical arm to face the three-dimensional sensor, recording point cloud information obtained by the three-dimensional sensor at this time, extracting at least onefeature point of the flange plate through utilization of a three-dimensional vision algorithm, and obtaining a three-dimensional coordinate value of the feature point relative to a three-dimensionalsensor coordinate system; S2, for the extracted feature point, obtaining the three-dimensional coordinate value of the feature point relative to a mechanical arm base coordinate system; S3, repeatingthe S1 and S2, extracting a plurality of non-coplanar feature points, obtaining a plurality of groups of three-dimensional coordinate values of each feature point relative to the three-dimensional sensor coordinate system and the mechanical arm base coordinate system, and computing the optimum hand and eye conversion matrix. According to the methods, through adoption of geometrical features of theflange plate at the tail end of an existing mechanical arm, the problem of selecting calibration points is solved, a mark point does not need to be installed in calibration, a calibration process isremarkably simplified, and calibration cost is greatly reduced.

The invention relates to a method for inhibiting an induction voltage and an induction current of a 750kV parallel single-line construction line. Two large-distance parallel and single-line construction 750kV alternating current transmission lines are coupled on an off-the-line line to generate a higher induction voltage and a higher induction current to ensure that the problem of difficult equipment selection caused by exceeding the turning on or off capacity of the existing 750kV earthing switch is solved. The invention provides a transposition and phase sequence arrangement mode to ensure that levels of the induction voltage and the induction current applying the measure are remarkably reduced and are lower than the requirement on technical parameters of the existing equipment, and thus the equipment selection problem is solved.

The invention discloses a medical image labeling method and device, equipment and a storage medium. The method comprises the following steps: selecting a plurality of types of reference images from animage database, carrying out similarity matching on the remaining images in the image database and the selected reference image, classifying the remaining images into the corresponding reference images according to a similarity matching result, constructing to obtain a multi-class reference image set, determining a to-be-labeled image from the reference image set, determining the labeling class of the to-be-labeled image based on the labeling information of the to-be-labeled image, and performing image labeling on the to-be-labeled image according to the labeling class. According to the method, the image content is standardized by adopting the color similarity and the content structure similarity, so that the candidate image selection problem of automatic annotation is solved, the labor time cost of early-stage image cleaning is saved, and the image annotation quality and speed are improved.

The invention provides a packaging structure of N substrate diode half bridges with common anodes in TO-220, which is characterized by comprising an insulating heat conductive aluminium oxide ceramic substrate, wherein the right side of the aluminium oxide ceramic substrate is provided with two mutually independent metal layers; the opposite side of the aluminium oxide ceramic substrate is provided with at least a metal layer; the metal layers at the opposite side of the aluminium oxide ceramic substrate are connected with heating panels; the two metal layers at the right side of the aluminium oxide ceramic substrate are respectively connected with left leads and right leads; the N poles of two diodes are respectively connected on the left leads and the right leads and are two cathodes; and the P poles of the two diodes are respectively connected on the middle leads by aluminium wires and are common anodes. The packaging structure has simple process, low manufacturing difficulty and stable yield and reliability, ensures the diodes to have high effective heat release and is suitable for using the N substrate diodes to assemble the half-bridge devices with common anodes.

The invention relates to a gear box operation state online monitoring method and system, and a storage medium. The method comprises the steps of establishing a frequency characteristic database of allto-be-measured parts of a gear box during normal operation; selecting a to-be-measured point and the to-be-measured part, and obtaining a real-time vibration signal; acquiring a Hilbert spectrogram of the to-be-measured part, and comparing the Hilbert spectrogram with a spectrogram in the frequency characteristic database; and obtaining a fault frequency, and marking the fault frequency on the spectrogram. The method solves the problems that when the parts of the gear box have slight faults, generated impact signals are small in amplitude and wide in energy dispersion, and a traditional signal processing method is difficult to capture the impact signals.

The invention discloses a clean processing equipment system of Pinus taiwanensis sunglow tea, comprising de-enzyming equipment, rolling equipment, baking equipment and shaping equipment. The clean processing equipment system is characterized in that the equipment system sequentially sets green curing equipment, the de-enzyming equipment, cooling equipment, the rolling equipment, primary baking equipment, the shaping equipment and flavor enhancing and sufficient drying equipment according to the clean processing steps of Pinus taiwanensis sunglow tea, i.e. green curing, de-enzyming, cooling, rolling, primary baking, shaping, flavor enhancing and sufficient drying. In aspects of primary and fine clean processing of Pinus taiwanensis sunglow tea, the invention combines a food engineering technology and a green tea processing principle, systematically provides primary and fine tea making processes and choosing and matching and cleaning requirements of machine tools, operates clean control, and realizes clean in the whole course of primary and fine green tea making.

The invention relates to a method for preventing after-combustion of gangue piles, and concretely relates to an after-combustion preventing gangue pile platform with checkerboard-shaped ditches, and agrouting method thereof. The method for preventing after-combustion is characterized in that the easy deforming, fracturing and spontaneous combustion parts of the gangue pile platform and its surrounding gentle slope are trenched in a checkerboard manner after the fire of the spontaneous combustion gangue piles is extinguished, and are grouted. The method has the advantages of easiness in operation, significant effect, long lasting time, solving of the spontaneous combustion prevention problem of the gangue pile platform, improvement of the integral stability of a gangue dump, and further solving of the selection problem of a grouting material and the monitoring of the temperature and the processing of a backfill layer after the grouting.

The invention discloses a cabinet system wind resistance test method comprising the following steps: S1, respectively measuring operation parameters of an air-discharge assembly of the cabinet under a cabinet door open and cabinet door close condition when the cabinet is powered up in operation; S2, comparing the operation parameters of the air-discharge assembly under two work conditions in the S1 with delivery test data of the air-discharge assembly so as to obtain an air quantity Q and a static pressure P of the air-discharge assembly under the cabinet door close condition, thus calculating a resistance coefficient of the cabinet system, wherein K=P/Q2; S3, using the cabinet system resistance coefficient K to draft a Q-P real performance curve in a Q-P delivery performance curve map of the air-discharge assembly, and the static pressure P matched with an intersection point between the Q-P delivery performance curve and the Q-P real performance curve is the cabinet system wind resistance value. The cabinet system wind resistance test method is simple in operation, and can accurately obtain the cabinet system wind resistance value.

The invention relates to the technical field of computer vision, in particular to a convolutional neural network feature decoding system based on FPGA implementation. According to the system, a feature decoding module of the CNN network is also placed in the FPGA for acceleration, the problem that feature extraction and feature decoding rates are not matched in an FPGA acceleration scheme is solved, all functions are realized in the FPGA, namely, the acceleration effect completely depends on the performance of the FPGA, and the maximization of the acceleration effect of the FPGA can be achieved as long as the functions of the two parts are completely designed in a streamlined manner. And meanwhile, a feature extraction result is directly processed in the chip without DDR cache, so that thetime delay of CNN network processing is reduced, and even the CNN network can be accelerated completely without a processor, and a chip-level calculation acceleration effect is achieved.

The invention discloses a screening method of a thermal stability and safety critical unit of a power system, which comprises the following steps: (1) obtaining an active output upper limit, an activeoutput lower limit of a system node of a designated power system and a line impedance value between the system nodes; (2) selecting any number of lines in the system as the system cross-section, andcalculate the sensitivity of all system nodes to these lines; (3) the sensitivity values of each node (n in total) to m lines in the system are treated as one sample point respectively. The mean algorithm clusters the n sample points. (4) the centroid of each cluster is selected as the key unit of this class. On the basis of the existing calculation of the active thermal stability sensitivity, Byusing the idea of clustering, the problem of selecting key units corresponding to key sections in thermal stability problems of power system is solved, which provides a feasible preliminary preparation scheme for the construction of thermal stability security region and thermal stability security control of power system, and provides a powerful reference for subsequent dispatchers to carry out security control.

The invention discloses a video label sorting method based on random walk. The method specifically comprises the following steps: (1) a video label graph: defining the similarity between a video labeli and a label j, calculating the similarity between the labels, constructing the label graph, and finally obtaining a complete graph; (2) vertex density: defining the vertex density according to a random walk algorithm to measure the representativeness of a certain label; (3) vertex density correction: correcting the vertex density through an IDF value, and defining the corrected vertex density;and (4) outputting a sorted video tag sequence by adopting a self-adaptive step length tag sorting algorithm. The method has the beneficial effects that the problem of video label selection in non-personalized recommendation can be solved, so that the video label selection in the non-personalized recommendation is more suitable.