45results about How to "Reflect characteristics" patented technology

The invention relates to the research field that a demander response participates in large-scale grid-connected wind power consumption, in particular to a large-scale grid-connected wind power consumption based on the demander response. The method comprises the steps of establishing high-energy-consumption enterprises of mining, metal smelting, petroleum processing, heating and the like nearby areas with rich wind power, utilizing adjustable and interruptible high-energy-consumption loads for participating in wind power pitch peaks, locally absorbing wind power resources, and avoiding large-scale and remote wind power transmission. Therefore, power grid losses are reduced, the power utilization cost of high-energy-consumption users is saved, output consumption of a wind power plant is increased, and occurrences of the wind abandoning phenomenon are reduced. The method can objectively and comprehensively reflect the output characteristics of large-scale power wind and the user power utilization load characteristics, the sizes of the adjustable and interruptible high-energy-consumption loads are accurately calculated, then the capacity of the newly-increased wind power consumption scale is determined, and the current large-scale wind power grid-connected consumption problem and the wind abandoning problem are solved.

The invention discloses a popular song emotion integration and classification method, and relates to the field of audio information processing. The method comprises the following steps: firstly, judging a music chorus appearance mode of a piece of music, and confirming different popular music modes; then, segmenting a song into N pieces of segments via a flexible segmenting method, and predictinghappiness and an intense degree of each segment; and implementing emotion integration of the whole piece of music by selecting various classifiers in accordance with the mode of a piece of popular music as well as happiness and intense degree results of the N music segments, so that an emotion label of the whole piece of music is obtained. The method provided by the invention has the advantages that V/A emotional evolution characteristics are extracted via the flexible segmenting technique and the characteristics are separately processed, so that the targeting performance of training on the popular song emotion classifiers of different structures is enhanced; and the method, which conducts song emotion classification via the structures of the popular songs and the emotional evolution characteristics, is better in effect of reflecting emotion cognition processes and features of human beings to music in comparison with methods which simply conduct integration on the basis of statistic characteristics of the whole piece of music.

The invention discloses a wireless frame transmission method and a wireless frame transmission device. The wireless frame transmission method comprises the following steps of: transmitting a null data packet (NDP) to a receiver, wherein the NDP comprises at least two very high throughput (VHT) long training fields (VHT-LTF), and the maximum number of the VHT-LTFs is determined according to the number of the VHT-LTFs which can be supported by at least one receiver. By the transmission method and the transmission device, the maximum number of the VHT-LTFs in the NDP is determined.

The method comprises: a) before infiltrating the tracer agent into the target layer, using a pulsed neutron dual-spectrum saturation logger to collect the well logging base value data; 2) infiltrating the tracer agent containing Gadolinium into stratum, and measuring the data after infiltration; 3) making the tracer to squeeze into the stratum, and measuring the data after squeezing; 4) processing and explaining the data.

The invention relates to an emulation method of high fidelity onboard foresight weather radar low-altitude wind shear signals, which belongs to the technical field of emulation of radar echo signals and comprises the following steps of: (1) establishing a wind shear wind field model by using wind field emulation software; (2) computing reflectivity factors of wind shear wind fields according to density information of an emulation wind field; (3) initializing an emulation scene; (4) determining opposite position information of scattering particulates in the scanning orientation and an airplane; (5) computing the amplitude and the phase position information of echo signals of the scattering particulates of the wind field; (6) superposing echoes of the scattering particulates to form an echo signal in the scanning orientation; and (7) updating a scanning angles of an antenna, and judging scanning range, if the scanning range is exceeded, outputting emulation results, and otherwise, repeating the steps. The invention has the advantages of simple method, good emulation effect, echo emulation of onboard foresight weather radar wind shear targets at high fidelity, and the like.

The invention discloses a powder metallurgy random particle size distribution 3D (three-dimensional) finite element modeling and simulation method which includes the steps: firstly, acquiring the content and the particle size range of all components according to a powder metallurgy formula, calculating the average density of material particles according to the density and the content of different components, and acquiring the sizes and the number of the particles under a particle size belt; secondly, throwing the particles by PFC3D software, discretizing a random model and deriving throwing coordinates and particle sizes; thirdly, leading the coordinates and the particle sizes into ANSYS software and building a random 3D finite element model; fourthly, leading the finite element model into Hypermesh, dividing grids and finally loading a solver for finite element simulation to simulate the random model of a powder metallurgy random particle size distribution process. Main parameters include maximum particle size and minimum particle sizes. By making full use of randomization of discrete elements and strong analysis functions of finite elements, the model is visual, simple and high in practicability and conforms to application and research of present powder metallurgy in terms of sinter molding.

The invention discloses a detecting method for blocking failure of a water drainage pipeline. The detecting method for blocking failure of the water drainage pipeline comprises the following steps that a detecting device for the blocking failure of the water drainage pipeline is installed; a known pipeline failure-free section/a pipeline failure section are selected for detecting, and signals under two working conditions are obtained by a computer from the receiving end; acoustic response signals of the water drainage pipeline under the two working conditions are obtained; SVD is conducted on the acoustic response signals; an energy entropy index and an approximate entropy index are extracted; the extraction results of the two indexes are merged as a initial eigenvector set; a within-class discrete matrix and a between-class discrete matrix are obtained by using a distance separable criterion method for an obtained energy entropy eigenvector set, an approximate entropy eigenvector set and the initial eigenvector set; the weight of the energy entropy index and the approximate entropy index are calculated; all the eigenvector sets are weighted to obtain a new eigenvector set; a random forest classifier is trained with the eigenvector set, and a failure identification model is obtained. According to the detecting method for blocking failure of the water drainage pipeline, accuracy and reliability of the detecting result are improved.

The invention belongs to the technical field of the high-speed railway geological disaster, and specifically relates to a method for determining unknown goaf range and old goaf residual settlement. The method comprises the following steps: firstly collecting goaf related documents such as peripheral coal seam distribution, mining and coverage resource coditions, and then drawing a settlement cloudatlas, and determining a suspicious goaf range; treating the suspicious goaf range by adopting the PSP-InSAR deformation measurement technology; form the suspicious goaf region delineated by the InSAR technology measurement, initially determining a fact that the goaf is existent under the region if the goaf abnormity is detected by using the geophysical prospecting technology measure, and performing the drilling; and performing indoor physical mechanics test for a sample taken out through drilling, and acquiring the mechanics parameter of each stratum covered on the goaf, thereby preparing for the subsequent modelling; performing precise modelling, performing mutual verification on a numerical analysis result and the settlement result of the InSAR, finally obtaining the influence range and the residual settlement of the old goaf. The method disclosed by the invention is scientific in design, accurate in measurement result, friendly in application environment, and wide in market prospect.

The invention discloses a solar collector output probability modeling method based on non-parametric kernel density estimation. The method comprises the following steps of: 1. obtaining hourly meteorological data measured in history and counting according to different years; 2. solving the output value and the bandwidth value of the unit area collector through a non-parametric kernel density estimation model of the output of the solar collector; 3. modifying the nonparametric kernel density estimation model. The method of the present invention can accurately reflect the random characteristicsand variation law of the output of the collector under different time scales, and can be used for planning and reliability research of the photothermal system under the condition of lacking historicaloperation data in the early stage.

The invention provides a linkage early warning management method and system, a storage medium and equipment, and the method comprises the steps: carrying out the grouping of sensors on a target monitoring object according to a preset grouping rule, so as to obtain a plurality of monitoring groups, and obtaining the data of all sensors of all monitoring groups in real time; analyzing the data of all the sensors of the preset monitoring group to obtain fluctuation conditions of the data of all the sensors of the preset monitoring group, association conditions of corresponding influence factors and change rates; according to the fluctuation condition of the data of all the sensors of the preset monitoring group, the association condition of the corresponding influence factors and the change rate, the abnormal condition of the data of all the sensors of the preset monitoring group is judged; and analyzing and judging abnormal conditions of the data of all the sensors of the preset monitoring group to obtain an early warning condition of the preset monitoring group. According to the invention, comprehensive and efficient linkage early warning management can be conveniently carried out on different monitoring items or different monitoring parts on the monitoring object.

The invention relates to a technical method for separating abnormal subdomain fault in chemical production, which is a novel fault separating method and is realized by using statistical analysis theory. The method comprises the following steps: extracting a principal element from data information acquired by a process; classifying data into information related with the principal element and information in weak relation with the principal element according to the correlation of a variable and the principal element; establishing a statistical process control limit for diagnosis of the fault respectively according to the two classes of data; determining a domain of normal data as a normal subdomain and a domain of abnormal data as an abnormal subdomain; and iterating the processes in the abnormal subdomain until the fault is converged to be separated. The method classifies the process data according to the correlation of the process data and a principal element array, can reflect characteristics and presentation of the fault information better, and improve relative accuracy and reliability of fault separation by statistical analysis. The method has the advantages of advanced technique, theoretical principle existence, and strong actual application and operability.

The invention discloses a method for diagnosing multiple faults of hydraulic pumps on the basis of extreme learning machines. The method for diagnosing the multiple faults of the hydraulic pumps is implemented by a feature extraction module, a feature reduction module and a fault diagnosis module. The method includes sequentially carrying out feature extraction, feature reduction and fault diagnosis on vibration signals, outlet flow signals and outlet pressure signals of an optional pump case to ultimately diagnose the multiple faults of the hydraulic pumps. The method for diagnosing the multiple faults of the hydraulic pumps has the advantages of reliable, accurate and simple diagnosis, high robustness and fault tolerance, good generalization capacity and the like.

The invention discloses a self-adaptive multi-channel graph convolutional network method combined with graph contrast learning. Multi-modal big data is often composed of multiple data of different structural forms, and the data often have the characteristics of complementarity, mutual verification and fusion. How to accurately and efficiently extract complementary information among multi-modal data is a main target of multi-modal research. However, most of methods for performing multi-mode fusion only pay attention to complementary information among multiple modes at present, but often neglect specific information under a single mode. In addition, how to extract abundant and distinctive expressions from multiple modes by using a graph convolutional network is rarely researched at present. Therefore, the invention discloses a self-adaptive multi-channel graph convolutional network method combined with graph contrast learning, a framework takes brain network research as a background, time and space information of a brain network can be mined, and unique and shared features of multiple modes can be effectively fused.

One or more embodiments of the invention provide a multi-source heterogeneous disaster data processing and presenting method and device; the method includes determining A data source coding result according to a data source type; according to disaster types in the data, determining a disaster information coding result; extracting target data from the multi-source heterogeneous disaster data; storing the target data, the data source code and the disaster information code in a unified manner; building the relevance among the target data, the data source and the disaster information; according tothe invention, the characteristics and attributes of disaster data of different sources and different disaster types can be reflected more comprehensively; source information and disaster informationof the multi-source heterogeneous disaster data are completely stored as much as possible in a coding mode; unified fusion of the heterogeneous disaster information and the heterogeneous disaster information is achieved, and the multi-source heterogeneous disaster data can be managed and presented more orderly.

The invention belongs to the field of rock slope stability analysis, and particularly relates to a three-dimensional analysis method for the stability of a toppled rock slope, which comprises the following steps: defining a toppled body, performing block division on the toppled body, and establishing a block system which comprises at least one block; establishing a force balance equation and a moment balance equation of the block body; constructing a standard optimization model by taking the force balance equation and the moment balance equation of the block body as constraint conditions, andcalculating the slope stability safety coefficient of the toppling body through the standard optimization model. On the basis of a force and moment balance equation, a boundary meeting a yield condition inequality equation is taken as a constraint condition, and the maximum value of a slope stability safety coefficient is taken as a target function, so that three-dimensional analysis of the stability of the toppled rock slope is converted into optimization solution, errors are reduced, and the mechanical characteristics and the stability characteristics of the three-dimensional analysis of thestability of the toppled rock slope can be better reflected.

The invention relates to a seismic data space-variant generalized wavelet extraction method, and aims to solve the problem that seismic wavelets extracted in the prior art cannot truly reflect energyattenuation, phase distortion and spatial change characteristics of seismic signals, and the method comprises the following steps: (1) estimating initial parameters of space-variant wavelets accordingto a seismic data power spectrum; (2) determining a global generalized wavelet reference frequency; (3) estimating a space-variant fractional order value of the generalized wavelet; (4) optimizing the global generalized wavelet reference frequency; (5) optimizing the space-variant fractional order value of the generalized wavelet; and (6) constructing a space-variant seismic generalized wavelet.The invention has the advantages that the extraction method is stable in operation and high in anti-noise capacity, the extracted seismic generalized wavelet is concise in form, and the extracted wavelet can truly reflect space-variant features such as absorption attenuation of seismic signals in the stratum medium propagation process.

The invention relates to the technical field of artificial intelligence, in particular to a chemical enterprise safety risk early warning method and system based on visual perception. The method comprises the following steps: acquiring a fuzzy smoke image according to ambiguity features, and detecting the smoke image to obtain a key frame image and a danger weight of a smoke color in the image; obtaining a geometric center point of the smoke area, drawing a plurality of straight lines through the geometric center point, obtaining equal division points of each straight line to construct a polygon, and obtaining a volatility index of smoke according to a difference value of interior angles of the polygon; obtaining the diffusivity index of the smoke according to the offset of the coordinates of the geometric center point, and obtaining an energy characteristic difference value according to the energy characteristics of the smoke area; and acquiring the danger degree by a danger evaluation model constructed by the danger weight, the volatility index, the diffusivity index and the energy characteristic difference value. By constructing the polygon of the smoke area, the dynamic change characteristics and edge detail changes of the smoke can be quickly and accurately reflected, the prediction result of the danger degree can be reduced, and dangerous accidents are avoided.