101 results about "Location model" patented technology

The invention discloses a fault location model generation method, a fault location method and devices, belonging to the technical field of virtualization. The method includes the following steps: determining the time information of alarm information generation in a virtual network; acquiring log data during generation of alarm information in the virtual network according to the time information; acquiring training data according to the log features; and training the training data into a fault location model of the virtual network through a classification algorithm in data mining. According to the method, log data during generation of alarm information is acquired, training data is acquired gradually according to the log data, and a fault location model is obtained through training according to the training data. Alarm information and training data are associated. The reliability of the fault location model generated based on training data is improved.

The invention discloses a cable fault detection and aging analysis method. According to the insulation resistance oscillation theory of the frequency domain, the cable fault detection and aging analysis method is characterized in that a fault location model is established by analyzing the oscillation characteristics of frequency domain impedance on the basis of oscillation wave spectrum analysis of the frequency domain impedance acquired by a non-destructive test system or a destructive test system, and then test oscillation impedance spectrum defect is compared with an internal simulation curve, so that long-distance accurate fault location can be realized, and the fault types can be also effectively distinguished. Compared with the prior art, the method not only is capable of realizing accurate multipoint positioning of faults, but also is capable of carrying out fault type recognition and insulation aging state analysis; in the whole detection and analysis processes, artificial experience is hardly depended; the method is suitable for a cable with the length of 1m-1,000km, and is especially suitable for a cable with the length of more than hundreds of kilometers.

The invention relates to an electric vehicle charging station planning method considering traffic network flow. The method comprises the following steps: referencing an intercepting location model in a traffic field, taking largest captured traffic flow, minimum network loss of a power distribution system and minimum node voltage deviation as an objective, establishing a multi-objective decision model, then determining a reasonable weight coefficient of each objective function after being normalized by using a super efficiency data envelopment analysis evaluation method, and transforming a multi-objective optimization problem to a single objective optimization problem, later putting forward an improved binary particle swarm optimization algorithm to solve a single objective optimization model. The method considers the urban traffic network traffic, electric car driving mileage and location problem of an electric vehicle charging station which affects a distribution network. The example results of the method show that the planned charging station location can provide convenient and fast charging services for more electric cars, and further improve the safety and reliability of the system operation.

The invention discloses a linear integer planning method of power distribution network on-line breakdown fault-tolerant location. The method includes the following steps that: an independent power distribution region and a related equipment set of the independent power distribution region are established; layered decoupling numbering is performed on the independent power distribution region, and a cause and effect equipment set is established; current alarm information is collected, a switching function set is established; an absolute value model of power distribution network breakdown location is established, and an equivalent linear integer planning breakdown location model can be established, and a feeder breakdown section is located through 0-1 linear integer planning; and the isolation of the feeder breakdown section is realized through the location result of the feeder breakdown section. According to the method of the invention, the dependence of a logic relationship modeling-based optimized fault location method on a swarm intelligence algorithm can be gotten rid of; and the conventional 0-1 linear integer planning is adopted to realize the location of the feeder breakdown section. The method is suitable for closed-loop open-loop-operation large-scale power distribution network online breakdown location. The method of the invention has the advantages of convenience, high reliability, high fault tolerance capability, high breakdown location efficiency, high adaptability for multiple breakdowns and the like.

A method of determining and optimizing the location of a new insurance agency is disclosed to increase market penetration of underrepresented markets. The method comprises the use of a scoring algorithm to rank various geographic regions or related zip codes. The scoring algorithm may be implemented by a location modeling system based on variables selected by a user.

The invention discloses an indoor location non line of sight compensation method and an indoor location method. The compensation method comprises the steps of acquiring base station coordinates of each base station arranged indoor and received signal strength indication of a received signal from each base station; obtaining estimated coordinates of a mobile terminal according to the acquired base station coordinates and received signal strength indication; obtaining an estimation value of a non line of sight error for every two base stations by use of the estimated coordinates of the mobile terminal according to the mobile terminal and a time difference of arrival location model having a specific non line of sight error of the two base stations; and replacing an undetermined non line of sight error with the estimated value of the non line of sight error so as to compensate the time difference of arrival location model. Through adoption of the method, the technical problem that the coordinates of the mobile terminal cannot be resolved due to overlarge time difference of arrival (TDOA) caused by the non line of sight error is solved, and accurate coordinates of the mobile terminal are obtained through non line of sight compensation for the TDOA.

A method of determining and optimizing the location of a new insurance agency is disclosed to increase market penetration of underrepresented markets. The method comprises the use of a scoring algorithm to rank various geographic regions or related zip codes. The scoring algorithm may be implemented by a location modeling system based on variables selected by a user. In addition, the various ranked geographic regions or related zip codes may be analyzed for proximity to natural or man made perils.

A system and method are disclosed for detecting and labeling places recognized in a video stream using change-points detection. The system includes a segmentation module and a label learning module. The segmentation module is configured to receive a video stream comprising multiple digital representations of images. The video stream is represented by a measurement stream comprising one or more image histograms of the video stream. The segmentation module segments the measurement stream into multiple segments corresponding to place recognized in the videos stream. The segmentation module detects change-points of the measurement stream and computes probability distributions of the segments over multiple pre-learned place models. The label generation module is configured to generate place labels for the places recognized by the place models.

The present invention discloses a cooperative-type queue driving system based on communication and a team forming and quitting method based on the same. Each vehicle in the queue is provided with onequeue driving system. The system comprises: a turn signal switch, a queue head vehicle application switch, a queue following vehicle application switch, a takeover queue head vehicle switch, a quitting queue head vehicle switch, a quitting queue following vehicle switch, a vehicle speed sensor, a yaw velocity sensor, an automatic driving vehicle speed and steering control module and an ECU calculation module which are connected with a CAN bus, the ECU calculation module obtains switching signals of queue driving, vehicle speeds and the yaw velocity signals from the CAN bus and sends the switching signals of queue driving, the vehicle speeds and the yaw velocity signals to the automatic driving vehicle speed and steering control module; a high-precision location model is employed to send the location information of the vehicles to the ECU calculation module, a wireless communication module is employed to receive and broadcast the queue driving information, and an HMI prompting module isemployed to display the queue state information. The driving safety of team forming and quitting is ensured.

The invention discloses a linear location model based machine body pose monocular-vision measurement method for a tunnel boring machine. The method comprises the steps: acquiring laser beam images bya camera, carrying out linear detection on acquired laser beams by using Hough conversion, so as to separately obtain linear equations of parallel laser beams and horizontal laser beams; modeling a two-point-three-line-based tunnel boring machine pose computing model, obtaining pose information on the camera in a location model coordinate system according to the linear equations of the horizontallaser beams and the linear equations of the parallel laser beams, and obtaining pose information on a machine body of the tunnel boring machine in the location model coordinate system in light of a conversion relationship between a machine body coordinate system of the tunnel boring machine and a coordinate system of the camera; finally, measuring absolute coordinates of laser orientation instruments by adopting a total station to obtain coordinates of the laser orientation instruments in a tunnel coordinate system so as to obtain a conversion relationship between the location model coordinatesystem and the tunnel coordinate system, and finally obtaining a pitching angle, a drift angle, a rolling angle and location information of the machine body of the tunnel boring machine in the tunnelcoordinate system in light of the pose information on the machine body of the tunnel boring machine in the location model coordinate system.

The embodiments of the invention disclose a method and a device for locating a license plate based on a fully convolutional network. The method comprises the following steps: using a license plate location model pre-trained based on a first fully convolutional network structure to determine at least one license plate image preliminary segmentation area contained in a to-be-detected vehicle image; processing the to-be-detected vehicle image containing at least one license plate image preliminary segmentation area to get a candidate license plate area contained in the to-be-detected vehicle image; using a license plate first character recognition model pre-trained based on a second fully convolutional network structure to determine whether the first character in the candidate license plate area is a first character of a license plate; and if the first character in the candidate license plate area is a first character of a license plate, determining that the candidate license plate area is a license plate, and marking the position of the license plate in the to-be-detected vehicle image. According to the embodiments, a license plate area can be located from a to-be-detected vehicle image in a natural scene, and the efficiency and the precision of license plate area locating are improved.

The invention is applicable to the field of communication and provides a mobile terminal location method, a location server and a serving base station. The method comprises the steps of adopting a nonlinear least square integral structure approach location method to obtain initial position coordinates of a mobile terminal according to measured distances between the mobile terminal and at least two location base stations and an actual distance between the at least two location base stations; and enabling the initial position coordinates to serve as initial iteration points, obtaining an optimal solution meeting location model constraint conditions by solving a location model, and enabling the optimal solution to serve as final position coordinates of the mobile terminal to be output, wherein the location base stations are base stations, which are capable of receiving signals, of the mobile terminal, d<ky> represents measured distances between the mobile terminal and the location base stations, x<k> represents position coordinates of a kth location base station, and y represents position coordinates of the mobile terminal. The location precision of the mobile terminal can still be guaranteed when errors of measuring distances between the mobile terminal and the location base stations are large.

The invention relates to a video analysis and positioning information combination-based continuous aircraft tracking method. The method comprises the following steps of: carrying out calculation according to positioning information of a target aircraft so as to obtain an image coordinate of the target aircraft in a video image; carrying out single-target tracking on the target aircraft according to a monitor video; comparing an image coordinate of a target aircraft tracking box for single-target tracking with the image coordinate of the target aircraft, and if a distance therebetween continuously exceeds an effective tracking threshold value, considering that the singe-target tracking loses the target; after confirming that the single-target tracking loses the target, determining a searchbox according to an effectively tracking range according to an image coordinate point, in the video image, of the target aircraft by taking the image coordinate point as a center point, so as to recognize the target aircraft; preferably taking HOG features as feature descriptors for target aircraft detection; and carrying out calculation by adoption of a DSST algorithm so as to obtain a position model and a scale model of a tracking filter. According to the method, loss of target aircrafts can be timely judged, and data processing amount for finding lost targets is small, so that the method can be mainly used for continuous tracking of targets such as aircrafts and similar occasions.

A method constructs a location model for locating and tracking sources of acoustic signals. Acoustic training signals are acquired from an acoustic training source at an unknown location in an environment with an array of microphones placed at unknown positions in the environment. From each acoustic training signal, relative acoustic features are extracted to construct a location model that is trained with the relative acoustic features.

Provided is a single-phase ground fault section positioning method for a 10kV distribution network. The inherent characteristics of the distribution network and the characteristics of fault location conditions and result characteristics in historical fault records are acquired from a power grid electrical information data source; the intrinsic characteristics of power grids are clustered as the input of self-organizing map neural network SOM, and the power grids with similar characteristics are grouped together; the fault location characteristics and the result characteristics of each type ofthe power grids are trained as input of generalized regression neural network GRNN to obtain fault location models of each type of the power grids; when a single-phase ground fault occurs during the operation of the power grids, based on the cluster type to which the power grids in which the fault occurs belongs, the fault location characteristics of each power grid supply section are input into the power grid fault location model corresponding to the type, and the fault section positioning result is obtained. The method has wide data source and high fault tolerance, and can quickly and accurately locate the faulty section.

The invention relates to a behavior identification method based on 3D point cloud and key bone nodes. The method particularly comprises steps that firstly, a 3D point cloud sequence is extracted from a depth map, the 3D point cloud sequence is uniformly divided into N non-overlapped space-time units, a local position model (LPP) of each space-time unit is calculated, and local position model statistics deviation descriptors (SDLPP) are calculated; secondly, a key bone node subset is extracted from the 3D bone nodes by utilizing a node exercise amount algorithm, 3D node position characteristics of the key bone nodes and a local occupation model (LOP) in the corresponding depth map are calculated; lastly, the three heterogeneous characteristics are cascaded, and random determination forest excavation distinguishable characteristics are utilized to realize classification and behavior identification. According to the method, the 3D local geometric characteristics and dynamic time characteristics of human body behaviors are extracted, an identification rate of complex human body behaviors related to man-object interaction is relatively high, and the method is suitable for complex human body behavior identification.

The present invention relates to an outdoor location method based on a LoRa technology, belonging to the field of wireless location. The method comprises the steps of: employing a BP neural network toperform correction of obtained RSSI (Received Signal Strength Indication) measurement values of LoRa signals; employing the corrected RSSI values, and performing dynamic optimization of parameters ofa LoRa path loss model Z to allow the Z to be suitable for different location environments; employing the trained Z to combine with a linear location model G based on the RSSI to establish a LoRa location linear model L3M based on the RSSI and obtain an initial location result set R of a target node; employing an L3M-C location model to perform clustering processing of the R and obtain a locationresult set R1; employing an L3M-MRE location model to perform minimum estimation RSSI error selection for the R and obtain a location result set R2; and employing an L3M-W selection strategy locationmodel to perform minimum mean estimation RSSI error selection for the R1 and the R2 and obtain the optimal estimation position of the target node. In an outdoor environment, the influence of NLOS errors and environment noise on signal propagation is reduced, and the location precision is improved.

The present invention relates to a construction project planning and site selection system and method. The system comprises a basic data preparation module, an intented lot roughing module and a lot estimation, comparison and selection module. Various information such as planning and current situations is compared and analyzed, a construction-available land one-map database and a location model factor database are generated, related limiting factors and limiting conditions are set, satisfactory construction-available lots are rapdily filtered according to planning and construction requirements of a construction project, total points of intented lots are collected and calculated, and a site selection report is thus automatically generated. With adoption of the system and the method, all construction can meet town and country planning and planning layouts, land wastage is reduced, and efficiency of planning, site selection and managing work is promoted.

The embodiment of the invention provides a mobile terminal indoor location method and device. The method comprises the following steps: offline acquiring signal intensity information respectively corresponding to multiple preset sampling points in one spatial position as fingerprint point signal information; according to the fingerprint point signal information, constructing a location fingerprintdatabase; acquiring displacement motion initial state information of the mobile terminal in the currently preset unit time; calculating a reference position of a displacement termination position byutilizing an inertial navigation principle and utilizing the displacement motion initial state information of the mobile terminal; and acquiring signal intensity information, respectively corresponding to preset unit time points on a displacement motion track, of the mobile terminal, constructing a sparse fingerprint location model based on spatial position constraint, correcting the reference position of the displacement termination position, and acquiring a corrected position of the displacement termination position. The embodiment of the invention realizes relatively high location accuracywith relatively low cost.

The invention discloses a bi-static synthetic aperture radar moving target location method. The method comprises the following steps: S1) establishing an airborne BiSAR space geometry structure, and finishing parameter initialization; S2) establishing a mathematical model for moving target echo in a BiSAR imaging area; S3) obtaining a moving target equivalent radial velocity expression; S4) establishing a reception dual-channel echo in track direction; S5) carrying out phase compensation on the dual-channel echo; S6) carrying out interference processing on the dual-channel echo signals; S7) obtaining a moving target location equation through a gradient analytical method; and S8) solving the target location equation through a genetic algorithm to obtain the position of a moving target. By constructing receiver double channels, equivalent radial velocity of the moving target with respect to a receiver is obtained through the echo interference processing; the moving target positioning formula is deduced through an R-D location model; and the moving target positioning formula is calculated based on the genetic algorithm, so that the moving target can be positioned, and positioning deviation is small.

The invention provides a semantic information and edge constraint-based foreground target detection method, and aims at automatically segmenting foregrounds and backgrounds of images through a segmentation energy function model by means of semantic label information so as to realize detection of foreground targets. A flow chart of the method is shown in an abstract drawing. The method comprises the following five steps of: 1, semantic edge constraint-based image layering and segmentation; 2, establishment of a position model; 3, establishment of an appearance model; 4, establishing of a smoothconstraint; and 5, iterative optimization of a segmentation model. Experiments prove that the method has feasibility, correctness and universality, and can be used in various high-level image analysis and image understanding.

The invention relates to a marine diesel engine fault location method based on a union belief rule base and an ant colony algorithm. The method of the invention comprises steps of collecting oil samples from a marine diesel engine oil monitoring system, detecting concentration of elements in the collected oil samples and then obtaining oil characteristics reflecting a wear fault location of the diesel engine; determining a reference values set of input characteristic parameter variables, and establishing a similarity of the input characteristic parameter variables to the reference values; performing fusion reasoning on activated rules in a rule base, determining a reliability value of a fault type, and taking the fault type corresponding to the maximum reliability value as an output of a fault location model; constructing an optimization model and solving the optimization model with an ant colony algorithm to obtain an optimal rule base, taking the optimal rule base as the final marinediesel engine fault location model and using the optimal rule base to obtain a more accurate fault type identification result. In the invention, the union belief rule base has fewer rules, less optimization parameters and short optimization time.

The invention discloses an indoor telephone traffic accurate location method based on machine learning in a cellular network. The method comprises the steps that a cellular network base station system uses idle resources to acquire the signal parameters of user equipment UE step by step; the data are used to design a hybrid location algorithm; information fusion is carried out to acquire preliminary position estimation; residual analysis decision is used for non-line-of-sight identification; a Kalman filter algorithm is used for multi-path suppression; a position estimated by the hybrid location algorithm is corrected to acquire an estimated position; priori information which comprises the habits of a user and the position of a building is used to correct the estimated position, and a reliable sample is extract; a machine learning algorithm is used to train data to acquire the location model of the combination of different parameters; a model trained in the off-line phase is used to estimate the position of the UE; and a particle filter algorithm is used to track the position in real time.

The invention discloses an LSTM based electroencephalogram signal source location method. The method includes the following steps: step (1) generating analog electroencephalogram data; step (2) constructing an LSTM based location model; step (3) training the location model by using the analog electroencephalogram data; step (4) performing pre-processing on real electroencephalogram data; and step(5) performing location on the signal source of the real electroencephalogram data. According to the technical schemes, the position of the signal source in the brain can be conjectured according to collected electroencephalogram signals.

The invention provides a multi-source-data-based location model and application research thereof, which are mainly based on three data sets: a public transport IC passenger card swiping data set, a telecommunication user conversation data set and a point-of-information data set. On the basis of a multi-source data set, the invention provides a location model establishing method based on the multi-source data set. The method is characterized by extracting some characteristic indexes, which may exert an influence on site selection, from the multi-source data set; carrying out learning on the characteristic indexes through the support vector machine technology to obtain a location model; and finally, determining whether candidate locations are appropriate for site selection. A location model establishing system based on a support vector machine is designed and realized; and the method is verified to be feasible through practical cases: the method for establishing the location model based on the multi-source data set is elaborated in details; then, specifically, the location model establishing system based on the support vector machine is designed and developed, and the provided scheme is subjected to test and evaluation through one practical case; and test results show that the location model obtained through the SVM (support vector machine) is higher in accuracy.

The present invention discloses a bistatic synthetic aperture radar location error calculation method. The bistatic synthetic aperture radar configuration mechanism is employed and the gradient analytical method is employed to complete calculation of the bistatic distance and location errors caused by Doppler center frequency shift to derive a bistatic synthetic aperture radar target location error concrete formula. The problem of difficulties in the bistatic synthetic aperture radar location error analysis and the problem that the location errors in a traditional R-D location model cannot be accurately calculated are solved. According to the error calculation formula, the location precision of the bistatic synthetic aperture radar can be calculated so as to realize measurement and assessment of the bistatic synthetic aperture radar location precision.

The invention provides a distribution network fault location method based on an improved quantum cuckoo algorithm. Distribution network fault location is modeled into high-dimensional combination optimization; a distribution network fault location section is expressed by using a quantum double-chain code; with an improved Levy flight algorithm, updating of a value of each dimension of a quantum cuckoo algorithm is optimized; and the quantum double-chain code is processed by variation by combining a differential evolution algorithm, so that an optimal solution to the distribution network fault location combination and an optimal solution is calculated. According to the invention, an optimized objective function of the distribution network fault location model is established; optimization calculation of the objective function is performed based on the improved quantum cuckoo algorithm; a fault section is identified; and verification is performed by using a distribution network example. Moreover, fault point location under the condition of multi-point simultaneous fault occurrence can be processed.

The invention discloses a comprehensive analysis system for electricity payment channels. The system comprises a payment channel information management module, a channel payment condition management module, a user preference analysis module, a channel geographic distribution management module and a channel simulative construction management module. According to the system disclosed by the invention, the development idea of the current big data age is combined; a data mining technology is introduced to analyze mass payment data of electricity customers; an improved KNN algorithm is adopted in the aspect of evaluation index selection, a customer payment behavior analysis model, a channel simulative construction model and a GIS-based visualized location model are innovatively constructed, and a big data-based comprehensive analysis system for payment channels is developed on such basis.