85results about How to "Reasonable analysis" patented technology

The invention discloses a space-time significance region detection-based human body behavior analysis method. According to the invention, a data set is adopted to rain a Faster R-CNN model. A multi-channel video is input and a single-channel video is segmented into video image frames. The segmented video image frames are subjected to target detection by adopting the Faster R-CNN model. A target detection result is analyzed and a target candidate box is re-calculated. The single-channel video is subjected to box matching for constructing a motion vector field. Based on the motion vector field,the motion vector of the region of interest is calculated. A foreground remarkable motion region is selected based on a probability calculated by utilizing a Gaussian mixture model. According to the target candidate box and the remarkable motion region, a space-time significance region is synthesized. The target space-time significance region is subjected to feature sampling and feature pretreatment. The target space-time significance region of the video is subjected to encoding and pooling. The space-time significance region of the video is subjected to human body behavior analysis and identification. An obtained analysis and identification result is written into a space-time significance region box. According to the invention, the categories of human body behavior activities in a video can be reasonably analyzed.

The invention relates to a quantitative evaluation method for dynamic stability of a creep-type side slope, in particular to a method for determining a dynamic stability coefficient of the creep-type side slope. The method includes the following steps: (1), measuring related parameters of the side slope and determining an initial stability coefficient F0; (2), arranging side slope displacement monitoring points and acquiring displacement monitoring data; (3), determining side slope creep rupture instability time and a stability coefficient Ft in a constant-speed creep stage; (4), determining an initial point tc and a stability coefficient Ftc in a side slope accelerated creep stage; (5), determining the dynamic stability coefficient in an accelerated deformation stage of the creep-type side slope; (6), determining a criterion for early warning of instability of the creep-type side slope. By the method, instability time of the creep-type side slope and landslide safety coefficients of the side slope at any time can be determined clearly, the limitation of incapability of analyzing and evaluating the dynamic stability of the side slop in real time according to conventional balance methods is overcome, and efficiency of dynamic stability evaluation and monitoring and early warning of the complex creep-type side slope can be greatly improved.

The invention provides a data analyzing method and system. The data analyzing method comprises the following steps: receiving and resolving a user processing command, and preprocessing data files of different sources and different forms to obtain a file to be analyzed conforming to a preset format, wherein parameters needing to be analyzed can be selected according to specific demands of a user, and all data are in the same sampling period, so that the times of the data to be analyzed are synchronized; receiving and resolving a user selection command, and selecting one or more test point data groups from the file to be analyzed according to the user selection command; selecting an algorithm corresponding to the user calculation command from a preset algorithm base, and calculating the test point data groups according to the algorithm; and lastly, combining calculation results of test points, and generating an analysis report including a data analysis diagram and/or a data analysis table corresponding to the algorithm. Based on the technical scheme provided by the invention, test data can be processed and analyzed reasonably, scientifically and regularly.

The invention provides a system and method for evaluating database performance load based on machine learning, the method comprises the following steps: Using machine learning algorithm training datato generate performance load learning model, after data processing, these massive data as a training set and use machine learning technology to train, and finally generating performance and load learning model, using this model to predict the new generated feature data, and evaluating the performance and load of the database. On the one hand, the conclusion of the evaluation system is more reasonable than the analysis of the expert model, and it will not miss the important performance load index, and the problem location of database performance and load is more accurate; On the other hand, reducing the knowledge and ability requirements of database operators can greatly save human cost and provide work efficiency.

The invention discloses a comprehensive health status evaluation method for rotation machinery equipment based on influence of use and maintenance. The health status of the equipment is influenced by different factors, related influential factors are analyzed and arranged, and divided into technical status factors and non-technical status factors, and concepts of an inherent health degree and a comprehensive health degree are defined. The inherent health degree of the rotation machinery equipment is evaluated on the basis of WPD (Wavelet Packet Decomposition) and SOM (Self-Organizing Map) by utilizing degeneration state monitoring data of the equipment. A proportion degeneration model is established by considering influence of service age and condition; and on the basis of an evaluation result of the inherent health degree, influence of service age and condition is also taken into evaluation of the comprehensive health degree. A proportion health recovery model is established, and influence of maintenance is further taken into evaluation of the comprehensive health degree. Influence of the service age, condition and maintenance are comprehensively considered on the basis of the proportion degeneration model (PDM) and the proportion health recovery (PHR) model, and evolution rule of the comprehensive health degree is derived.

The invention discloses an analyzing and determining method for a bridge foundation slope stable slope-angle line. On the basis of the inversion of slope rock-soil physics and mechanics parameters, the action width, embedded depth and load effect modes of a foundation are considered, and the specific position of the stable slope-angle line is analyzed, so that arrangement and engineering design demands for the bridge foundation in a steep slope can be met. The method comprises following technical steps: S1, determining a critical slide surface for the slope parameter inversion; S2, working out the comprehensive cohesive force of a slope rock-soil mass through inverse calculation according to a limit equilibrium state; S3, setting the initial position of the stable slope-angle line; S4, recalculating the slope stability under a bridge foundation loading effect according to the obtained cohesive force of the slope rock-soil mass through inversion calculation; S5, determining whether a safety factor is larger than or equal to a designed safety factor; and S6, obtaining the stable slope-angle line through the analysis.

The invention discloses a reactive voltage control simulation system and method based on a power grid regulation and control integrated mode. The reactive voltage control simulation system based on the power grid regulation and control integrated mode comprises a simulation service system and a control system. The simulation service system is used for receiving external control and disturbance, simulating the regulation of power grid equipment, simulating the load flow changes of a power grid, generating real-time simulation power grid data, storing, managing and distributing the real-time simulation power grid data, and supplying the real-time simulation power grid data for the control system. The control system receives the real-time simulation power grid data from the simulation service system and stores the real-time simulation power grid data, formulates a simulation power grid control strategy, generates a control instruction and sends the control instruction to the simulation service system. According to the reactive voltage control simulation system and method, an AVC system and a DTS system can conduct closed-loop control simulation on the power grid according to an actual power grid operation electricity generation load curve in continuous time periods, and the defect that when an existing AVC system conducts verification and analysis on the control strategy, the regulation condition which may occur is not taken into consideration is overcome.

The invention relates to a prediction method of an urban congestion propagation mode based on a cyclic autoregression model. The method comprises the following specific steps: S1 generating a road section connection relation graph N by utilizing a topological connection relation of an urban road network; S2 selecting historical data in a long time period, and constructing a congestion propagationtime sequence knowledge graph G based on the road section connection relation graph N; and S3 training a cyclic autoregression model, learning a congestion propagation mode in the congestion propagation time sequence knowledge graph G, and performing prediction. According to the method, by constructing the corresponding time sequence knowledge graph, space-time prediction can be carried out for different types of congestion propagation modes in a road network.

The invention discloses a non-target biomarker high-throughput screening method based on pollutant metabolism disturbance, and belongs to the field of environmental exposure and health, and the methodcomprises the following steps: (1) extracting to obtain a to-be-detected extracting solution; (2) performing chromatographic analysis to obtain a spectrogram containing chromatographic peaks; (3) identifying and labeling pollutant characteristic peaks, taking chromatographic peaks except the pollutant characteristic peaks as potential metabolite characteristic peaks, and carrying out non-target labeling on the potential metabolite characteristic peaks; (4) establishing a linear regression model by taking the peak area of the potential metabolite characteristic peak as a dependent variable andthe peak area of the pollutant characteristic peak as an independent variable; (5) operating the model, carrying out non-target screening on the biomarkers, and preliminarily obtaining related biomarkers; and (6) identifying the preliminarily obtained primary and secondary spectrograms of the biomarker, and identifying the biomarker related to pollutant exposure. According to the method disclosedby the invention, the accuracy of biomarker screening is remarkably improved, and meanwhile, the flux of biomarker screening is improved.

The invention relates to an oxygenating device for aquaculture and particularly relates to an oxygenator capable of implementing water quality analysis, pre-warning and intelligent control. The oxygenator comprises a motor, a turbine, a sensor module group, a controller module, a high-gloss pre-warning indicating light, a GSM (Global System for Mobile Communications) text message module, a floating body, a power supply module, an insulating shell and a bracket, wherein the sensor module group consists of a dissolved oxygen sensor, a PH value sensor, a temperature sensor, an electric conductivity sensor and a turbidity sensor; a controller is used for acquiring each quality parameter of a water body by the sensor module group, comparing and analyzing the parameters to obtain comprehensive water quality parameters, executing corresponding pre-warning and braking operation according to the comprehensive water quality parameters, sending the water quality parameters and the motor running state to a user by the GSM text message module and also receiving and executing an operation instruction sent in a text message by the user; the motor is used for driving the turbine to rotate to make sprays in order to improve the water quality; the power supply module is used for supplying power to each part of the oxygenator after transforming commercial power. The oxygenator disclosed by the invention is low in cost, simple to operate and wide in applicability.

The invention relates to a method for designing a partially compensatory lens by taking a slope as an optimization objective and belongs to the technical field of aspherical detection. The method comprises the following steps of: setting system parameters, namely setting entrance pupil diameter D and wavelength lambda as required by using ZEMAX optical design simulation software which is widely used in the field of photoelectron as a design platform; calculating the optical parameters of the partially compensatory lens; and arranging a reflected light path, and calculating structural parameters of a partially compensatory detection system to take the slope as the optimization objective. The method is simple and fast, is easy to implement, optimizes light rays in the full aperture range, performs comprehensive and reasonable analysis, realizes the detection and optimization of the maximum slope of residual wave aberration by using the radius of a confusion circle as the optimization objective, simplifies the design difficulty of the partially compensatory lens under the condition of ensuring distinguishable interference fringes, and can expend the detection scope for a compensator so as to reduce detection cost and improve detection speed.

The invention relates to a fire analysis method and system for horizontal multi-layer cable bridges of a nuclear power plant. The method comprises following steps: acquiring initial fire source feature information of fire of the horizontal multi-layer cable bridges of the nuclear power plant; acquiring real-time heat release rate of each layer of cable bridge in the fire process according to the initial fire source feature information, geometrical structure parameters of the horizontal multi-layer cable bridges, the longitudinal spreading speed of a cable flame and feature parameters of cable materials; acquiring fire risk feature parameters in confined space according to the initial fire source feature information and the real-time heat release rate of each layer of cable bridge; comparing the fire risk feature parameters with quantitative indexes and judging the fire risk of the horizontal multi-layer cable bridges of the nuclear power plant. According to the method and the system, the conservative principle of the nuclear power plant is met, the differences between single cable bridges in the horizontal multi-layer cable bridges are fully considered, the analysis accuracy is higher, the fire risk of the horizontal multi-layer cable bridges of the nuclear power plant can be analyzed more reasonably, and great support is provided for fire protection design of the cable bridges.

The invention relates to a method for analyzing song popularity under the influence of a social relationship in a music information network. The method comprises the steps of firstly obtaining an information sub-network and a network mode of the information sub-network from the music information network, and obtaining the social relationship related to the information sub-network; secondly calculating mean edge betweenness of different types of relational edges in the information sub-network; thirdly making a calculation according to the mean edge betweenness to obtain propagation factors of the different types of relational edges; fourthly designing edge weights in the information sub-network according to the social relationship and selecting nodes from nodes of the same type according to the edge weights to perform random walk; and finally performing random walk in the information sub-network in combination with the propagation factors and the edge weights to obtain a node influence ranking in the network so as to obtain the song popularity. According to the method proposed by the invention, the influence of a user group on songs is considered, so that the final popularity analysis is more reasonable.

The invention relates to the technical field of data processing systems or methods specially suitable for administrative, commercial, financial, management, supervision or prediction purposes, and provides a cultivated land ecological sustainability evaluation method and system. The method is used for evaluating county-level cultivated land ecological sustainability in southern red soil hilly regions. The method comprises the following steps: according to a cultivated land ecological current situation of a research region, selecting indexes from four aspects of mine damage, landscape pattern, ecological stability, and ecological space occupation, and constructing a cultivated land ecological sustainability evaluation index system; combining subjective and objective factors, and combining an analytic hierarchy process and an entropy evaluation method based on a Lagrange multiplier method to determine index weights, wherein the cultivated land ecological sustainability evaluation index system comprises a plurality of cultivated land ecological influence factors, and each cultivated land ecological influence factor comprises a plurality of cultivated land ecological sustainability evaluation indexes; according to index values of the cultivated land ecological influence factors and a cultivated land ecological sustainability level, evaluating the cultivated land ecological sustainability.

The invention discloses an appraisal method based on information technology (IT) system application. The method comprises the following steps of: 1) importing data; 2) establishing an appraisal task; 3) starting an automatic scoring system and an expert scoring system according to the established appraisal task; 4) evaluating comprehensively; and 5) publishing a result. The steps in the method are rationally designed, the two scoring systems are independent of each other and act uniformly, each IT index can be rationally analyzed, and a corresponding management method is made and a corresponding technical measure is taken according to the analysis result to realize comparing and evaluation functions. Therefore, the practical degree of an application system is improved.

The invention provides a method for monitoring and analyzing illegal operation of a controller based on air traffic control voice recognition. The method employs the newest artificial intelligence technology to compare the voice communication between the controller and an aircraft and the actual operation behavior of the controller with a standard operation specification, and judges whether the controller has the illegal operation behavior or not. According to the recorded violation operation behaviors of the controller and the analysis result, in combination with a reasonable management system, the violation behaviors of the controller can be more effectively reduced, and unsafe events and accidents of air traffic control are reduced.

The invention discloses an Internet-based mineral product price dynamic monitoring and query service system, including a main workstation and a data acquisition module connected to the main workstation, an intelligent assembly module, an automatic output module of results, an early warning and forecast module, and a quality supervision module. The acquisition module is connected with an OCR system, and the automatic achievement output module is connected with a data query terminal and a printer. The whole process of data traceability is realized by numbering and cleaning the data between each module, and the main workstation is also connected with a network. Database server; the present invention consists of simple price information collection, data clustering, collation, analysis, and report formation, and raises the collected mineral product market price information and other mineral product pricing influencing factors to scientific, reasonable, and rapid analysis. At the same level, it solves the problems of difficulty in collecting mineral product price data, single means, low accuracy, inaccurate statistics, and untimely publication.

The invention relates to the technical field of artificial intelligence and project cost, in particular to a project cost progress supervision system. The system comprises an image acquisition module which obtains an initial image of a construction area and obtains a first construction area of interest; a drainage ditch area division module which detects the edge of the first construction interested area and screens out ground pixel points to obtain a second drainage ditch area; a drainage ditch digging progress detection module which judges whether a material area exists or not by classifying and recognizing personnel distribution information in the initial image, further obtains a personnel distribution area, divides a second drainage ditch area in real time through the personnel distribution area to obtain real-time sub-areas, and obtains the digging progress of a drainage ditch through the depth information of the real-time sub-areas; and a project cost progress management module which predicts the completion period of a project through the real-time information of the drainage ditch digging progress. The drainage ditch digging progress of the real-time sub-areas of the drainage ditch is obtained in real time, and the accuracy of obtaining the construction progress is improved.

The invention relates to a ship simulation test system based on digital simulation; the system is composed of a test module, a performance evaluation module and a function module; the test module is used for carrying out simulation tests on the performance and function of a ship, and the simulation tests comprise a seakeeping test, a maneuverability test, a rapidity test and an intelligent ship test; the performance evaluation module is used for performing hydrodynamic performance evaluation and intelligent ship evaluation, and hydrodynamic performance evaluation is used for evaluating seakeeping performance, rapidity performance and maneuverability performance of a ship; the intelligent ship evaluation is used for evaluating the autonomous berthing and leaving performance, the anti-collision performance, the tracking navigation performance and the autonomous decision-making performance of the ship; the function module is used for optimizing the ship and comprises a ship type and propeller sub-module and an intelligent ship sub-module. And the evaluation result of the performance evaluation module is accurate and reliable, so that subsequent optimization design is facilitated.

The invention provides a data control method, a data control terminal and a data control system. The method comprises the following steps that a monitoring terminal obtains the monitoring data corresponding to a monitored object; the monitoring terminal receives an analysis request sent by the monitored object, selects a corresponding analysis strategy according to the feature information of the monitored object, and carries out the processing of the analysis request. The method achieves the multi-dimensional and differentiated management of online shops, and greatly improves the experience of online shopping.

The invention discloses an image shift compensation calculation method based on an elliptical orbit, which relates to the aerospace field and aims at solving the problem that the existing image shift compensation model can not realize compensation on the image shift of the elliptical orbit. Based on a circular orbit image shift speed vector calculation model and parameters of the elliptical orbit, an image shift speed vector model for the elliptical orbit is built, and the elliptical motion orbit speed of a spacecraft, the perigee orbit speed and the apogee orbit speed are calculated; the included angle between the orbit speed and a centrifugal speed is calculated, the circular speed and the centrifugal speed can be acquired according to the calculated included angle, and the elliptical orbit height is calculated; the image shift speed converted by the circular speed of the spacecraft to an image plane passes through a scenic spot sub-satellite vertical horizontal coordinate system, a planetary coordinate system, a planetary inertial coordinate system, an orbit coordinate system, a camera coordinate system and an image plane coordinate system; and the image shift speed of the scenic spot on the image plane of the camera is acquired, and the elliptical orbit image shift is compensated according to the acquired image plane image shift speed.

The invention discloses an ultrasonic vibration auxiliary milling force modeling method based on finite element simulation, which takes tool parameters, cutting parameters and vibration parameters into account, and comprises the following steps: firstly, dispersing a blade into a series of infinitesimal elements along the axial direction, and calculating instantaneous cutting thicknesses of different cutting positions of the infinitesimal elements; then, using a response surface method for designing a simulation scheme about cutter parameters, cutting parameters and vibration parameters, performing a series of bevel angle cutting simulation through ABAQUS, extracting simulation cutting force, and obtaining a regression equation for calculating infinitesimal cutting force; and finally, calculating the cutting force of the infinitesimal at different cutting positions through a regression equation, and performing coordinate transformation and integral summation to obtain a predicted value of the ultrasonic vibration milling force in the milling process. The method can solve the problems that an existing modeling method is long in experimental period and high in experimental cost, and the applicability and reliability of the milling force model are improved.

The invention discloses an exhibit recommendation method and system for an online exhibition, and the method comprises the steps: building a user portrait through the collection and analysis of the historical behavior data of a user; performing similarity calculation on the word cloud of the user and exhibit information in an exhibit database, and finding out a preset number of exhibit which most conforms to the word cloud of the user for recommendation; performing similarity calculation on the exhibit category preference degree vector of the user and exhibit category preference degree vectors of other users in the user portrait database, finding a preset number of most similar users, and recommending a preset number of exhibits recently browsed by the users; on the basis of the exhibit category browsing trend transfer matrix of the user, generating a sequence only comprising an exhibit category as prediction; for each element in the sequence, finding the hottest exhibit in the current category for recommendation; and mixing the recommendation results to obtain a final result. According to the method, recommendation results are obtained from multiple angles and mixed, so that the quality of targeted service for the user can be improved.