68results about How to "Improve division accuracy" patented technology

The invention relates to a harmonic contribution division method and a harmonic contribution division system. The harmonic contribution division method comprises the following steps of acquiring harmonic voltage data of a bus and harmonic current data of a harmonic source to be calculated on a feeder line; calculating background harmonic impedance by using a leading fluctuation quantity method according to the harmonic voltage data and the harmonic current data; and dividing harmonic contributions of the harmonic source to be calculated by using a quantile regression method according to the harmonic voltage data, the harmonic current data and the background harmonic impedance. The background harmonic impedance is estimated by the leading fluctuation quantity method, fluctuation quantity with a leading function is screened out to calculate the background harmonic impedance, influences of background harmonic and measurement noise fluctuation on a background harmonic impedance estimation result are restrained effectively, and the background harmonic impedance is calculated accurately; and the background harmonic current is calculated according to the background harmonic impedance, and quantile regression is performed to obtain the harmonic contributions of the harmonic source. Calculation deviation caused by background harmonic fluctuation can be reduced, division accuracy is improved, and the stability and the data utilization rate are high.

The invention provides a vehicle lower boundary detection method based on multi-sensor fusion. Through measuring information of a millimeter-wave radar and a camera, space alignment points are obtained, an area of interest containing a target vehicle is selected in a camera image according to space alignment point information, by determining k peak values of the area of interest, k-1 threshold values of the area of interest are determined, a minimum threshold value is obtained, and accordingly a vehicle lower boundary shadow area corresponding to a gray level area marked out by the minimum threshold value, finally, a vehicle lower boundary shadow line is obtained in the shadow area, and vehicle lower boundary detection is achieved. By using a certain searching strategy, the target area of interest is determined, the target area of interest can contain the target vehicle, the size of the area can be proper, following computing is convenient, the k-1 threshold values are determined through a particle swarm optimization algorithm, operation steps are simplified, operation speed is improved, and meanwhile area dividing accuracy is improved.

The embodiment of the invention provides a user group partitioning method and device. The method comprises the steps of capturing user identifiers; building a joint behavior relation between the user identifiers; obtaining a user node diagram; in the user node diagram, recognizing one or more core user groups according to the joint behavior relation; in the one or more core user groups, dividing one or more target user groups according to the joint behavior relation. On one hand, manual rule setting is avoided, different user groups have different distribution features, although individual behavior differences are large, the potential relation of the user is more stable, and the user group partitioning accuracy of the user groups is improved in a diagram-based user group partitioning mode; on the other hand, by roughly partitioning the core user groups, the data size is greatly decreased, the partitioning efficiency is improved, and the user group partitioning accuracy is improved.

The invention relates to a wind power plant equivalent modeling method based on the probability coherence of wind power units. Based on the probability of various short circuit faults in the actual operation of an electric system and statistical result of wind resource state in a wind power plant, the dynamic change of the rotating speed of a wind-driven generator in a transient process under the conditions of combinations of various short circuit faults and wind speeds is collected through time-domain simulation, and the wind power units are aggregated and classified by a two-step classification method according to the information. Based on collecting the clustering results of all wind power units, the wind power plant equivalent model is determined according to the probability of appearance of unit group classifying result, and the parameters of the wind-driven generator, current collection system and the like are equalized. The wind power plant equivalent modeling method provided by the invention fully considers the influence of different short circuit faults and unit operation conditions on the classifying result of the coherent unit group, the step of real-time monitoring the change of the coherent unit group in engineering application is effectively avoided, and the engineering application value of the wind power plant equivalent model is improved.

An audio watermarking method of self-synchronization includes converting synchronous information and watermark information being RS coded to be two-way code base band signal then embedding synchronous and watermark information in audio signal frequency spectrum in sequence by means of spread spectrum sequence embedding when watermark is embedded, finalizing frame and data synchronization by picking up synchronous information with manner of moving correlation in communication then carrying out correlation detection on frequency spectrum of received signal for picking up watermark information when watermark is picked up.

The invention discloses a wavelet packet conversion based frequency spectrum imaging method. The method includes steps of performing encrypted wavelet packet conversion on a three-dimensional seismic data volume subjected to offset and dissolving seismic record into frequency-dividing seismic record ranked according to frequency band size from small to large; calculating five three-dimensional attribute data volume including frequency-dividing instantaneous amplitude spectrum, frequency-dividing instantaneous phase spectrum, instantaneous average frequency, instantaneous bandwidth and instantaneous absorption coefficient according to the wavelet packet conversion result; performing attribute slicing under the control of a seismic interpretation layer. According to the invention, influence on seismic wavelet spectrum estimation due to reflection coefficients is eliminated through a complex cepstrum technology, so that the instantaneous absorption coefficient estimated on said basis is more reasonable. A target is made prominent by utilizing a specific effect of color mixing in integrated display of three different frequency attributes and the target distinguishing effect of single frequency can be improved effectively.

The invention provides a sealed integrated smoke collection test device. The sealed integrated smoke collection test device comprises a sampling hole substrate, an outer sleeve, a smoke sampling pipe, a thermoelectric couple and an annular sealing disc, wherein the sampling hole substrate is arranged in a sampling hole of a to-be-tested flue, and the sampling hole substrate is provided with a through hole which is communicated with the to-be-tested flue; the outer diameter of the outer sleeve is smaller than the inner diameter of the through hole of the sampling hole substrate; the smoke sampling pipe is arranged in the outer sleeve; the thermoelectric couple is pre-installed outside the smoke sampling pipe and arranged in the outer sleeve; the annular sealing disc is sleeved outside the outer sleeve, the sealing disc is in sealing connection with the outer sleeve, and the outer diameter of the sealing disc is greater than the inner diameter of the through hole of the sampling hole substrate. By adopting the sealed integrated smoke collection test device, the influence of leaked air from the sampling hole on a test of smoke components and a temperature test can be well solved, the use is convenient, a great amount of in-situ installation and length calibration work can be saved, the test time can be saved, and the test efficiency and the test precision can be improved.

The invention relates to a method for measuring urban space function compactness. The method comprises the steps of S1, obtaining interest point data, night light data and road network data of a to-be-measured city; S2, determining the urban core area of the to-be-detected city according to the distribution density of the acquired interest point data in the to-be-detected city; S3, dividing the urban core area into a plurality of land parcels according to the road network data; S4, determining the type of each land parcel according to the interest point data, and determining the human activityintensity of each land parcel according to the night light data; and S5, determining the space function compactness of the to-be-detected city according to the types of all land parcels in the city core area of the to-be-detected city and the human activity intensity. According to the method, the dimension which cannot be expressed by the urban form compactness can be reflected, the reasonabilityof urban layout is evaluated functionally, and a new view angle is provided for urban compactness measurement.

The invention discloses a direction- sensitive multi-gesture identification system and method based on WiFi equipment, and the system comprises a data collection module, a data preprocessing module, a feature analysis module, an abnormal gesture filtering module and a gesture identification module.The method comprises the following steps: collecting and storing CSI data in a WiFi signal transmitted by a WiFi transmitting end through a WiFi receiving end, and obtaining CSI amplitude information from the collected CSI data; preprocessing the obtained CSI amplitude information to obtain CSI data which is subjected to noise elimination and completely contains gesture information; judging illegal CSI (Channel State Information) data in the preprocessed data of the gesture information by adopting an illegal data judgment algorithm based on an SVM (Support Vector Machine), and filtering abnormal gesture actions to obtain judged legal CSI data; and inputting the judged legal CSI data into a DNN neural network, performing feature analysis on gesture actions in different directions by adopting a method based on a Fresnel zone model, determining different directions of the same gesture action, and then classifying the gesture actions in different directions and outputting a gesture identification result.

The invention provides a hydroelectric generating set multi-dimensional vibration area fine division method based on a decision-making tree model. The method comprises steps of S1, obtaining a workingcondition parameter and a stability parameter of a set state monitoring system, and carrying out data cleaning, S2, determining a unit stability parameter threshold, S3, constructing a judgment matrix and dividing the judgment matrix into a training set and a test set, S4, establishing a decision tree model by using the training set, and S5, completing a precision test by using the test set, andsubstituting the precision test into the decision tree model to predict whether the new working condition belongs to the vibration area. The method is advantaged in that a big data mode is adopted, unit state on-line monitoring system data is fully utilized, the effective data information is mined, the effective data information is applied to unit vibration area full-water-head fine division, a problem of predicting whether uncovered data points are located in the vibration area or not is solved, and the method considers multi-dimensional working conditions, and improves accuracy of the vibration area division model of the unit based on big data mining.

The invention relates to a curved surface template segmentation method, a curved surface template segmentation device, computer equipment and a storage medium. The method comprises the steps of obtaining a to-be-segmented curved surface and a target segmentation unit size; extracting a first direction length and a second direction length of the target segmentation unit in the two curved surface directions from the target segmentation unit size; extracting a first structure line from the to-be-segmented curved surface according to the first direction length, determining a first segmentation point in the first structure line according to the second direction length, and generating a first isoparametric curve according to the first segmentation point; extracting a second structure line from the to-be-segmented curved surface according to the second direction length, determining a second segmentation point in the second structure line according to the first direction length, and generatinga second isoparametric curve according to the second segmentation point; and generating a segmentation line of the to-be-segmented curved surface according to the first isoparametric curve and the second isoparametric curve, obtaining a three-dimensional template segmentation model according to the segmentation line, and outputting the three-dimensional template segmentation model. By adopting the method. The working precision of curved surface template segmentation can be improved.

The invention discloses a crowd behavior recognition method based on sub-group division and dynamic and static feature fusion, and the method comprises the steps: firstly obtaining the spatial and temporal information of a moving target in a video image frame through employing an angular point tracking and background modeling method, and dividing spatially adjacent crowds into a plurality of sub-groups through employing the spatial region information of group distribution; secondly, on the basis of sub-group segmentation, obtaining sub-groups; three momentum characteristics of crowd movement are extracted; fusing the CNN neural network with a static sequence generated through the CNN neural network by utilizing Cholesky transformation; video classification is realized by using a GRU neuralnetwork, feature vectors are converted into crowd behavior tags by using an output function, and training video clips are marked as different description vocabularies by using a manual marking methodaccording to differences of behavior occurrence subjects, behavior occurrence places and behaviors.

The invention discloses a method and device for dividing a video time length, relates to the technical field of videos, and solves errors generated by segmented operation when the video time length is divided. The main technical scheme of the invention comprises: acquiring video time length information and the number of sections to be determined corresponding to video information; determining whether there is a remainder in a result of an pre-calculation between the video time information and the number of sections to be determined; If there is a remainder, configuring the non-remainder portion in the result as the video time length of each section except the last section, and configuring the sum of the non-remainder portion and the remainder as the video time length of the last section. The method and device are mainly used for dividing the time length when the video is segmented.

The present invention discloses a method for dividing edge users and center users in LTE (Long Term Evolution) frequency reuse. The invention mainly aims to solve inaccurate user type division and poor frequency reuse performance in LTE frequency reuse in the prior art. The implementation steps of the method include the following steps that: (1) the frequency bands of LTE sectors are divided; (2)the location areas of the LTE sectors are divided; (3) parameters are configured; (4) UE (user equipment) sends an uplink PRACH preamble; (5) a service sector calculates the uplink propagation time t0(k) of the PRACH preamble of the UE; (6) user types are judged primarily on the basis of the uplink propagation time t0 (k); (7) UE types during service transmission are dynamically updated; and (8)the service sector allocates broadband resources. With the method of the present invention adopted, the accuracy of the division of the edge users and center users in the frequency reuse can be improved, and the performance of the LTE frequency reuse is improved.

A resource multiplexing method and device, a storage medium, a terminal and a base station are provided. The method includes the following steps: determining a first reserved resource reserved for a PDCCH (Physical Downlink Control Channel) in the resources used in the overlapping area of the PDCCH and a PDSCH (Physical Downlink Shared Channel); mapping the first reserved resource from a system grid to a signaling grid to obtain a signaling grid bitmap; and sending the signaling grid bitmap to user equipment, wherein the system grid contains M resource blocks, the signaling grid is constructedon the resources used in the overlapping area of the PDCCH and the PDSCH and contains K resource grids, K and M are positive integers, and K is smaller than M. Through the scheme provided by the invention, the resources which can be multiplexed by the PDSCH in the resources used by the PDCCH can be indicated with less overhead, and the overhead is constant. Moreover, the user equipment can use the resources multiplexed by the PDCCH to the PDSCH more efficiently.

The invention relates to the technical field of audio processing, and aims to provide a song phrase division method and system, electronic equipment and a medium, and the song phrase division method comprises the steps: obtaining audio data corresponding to a song, and processing the audio data to obtain text data corresponding to the audio data; processing the text data to obtain a pinyin array corresponding to the text data; extracting vowels corresponding to each pinyin in the pinyin array, and storing the vowels corresponding to all the pinyin in the pinyin array into a hash table; counting the occurrence frequency of all finals in the hash table to obtain a finals and occurrence frequency comparison table; according to the vowels and the occurrence frequency comparison table, obtaining the vowel with the most occurrence frequency, and judging that the pinyin corresponding to the vowel with the most occurrence frequency is the final of the text data; and according to the position of the final, performing phrase division on the pinyin array to obtain music segments. The efficiency and accuracy of phrase division can be improved.

The invention belongs to the technical field of community division, and particularly relates to a dynamic community division method based on a spectrum method. Aiming at the defects of a structure-based division method and a user-based division method, the invention provides a mode of carrying out community division on users in a social network by adopting a three-way decision division mode aftercarrying out information aggregation by adopting a spectrum method. Such a thought ensures that the structure information and the user attribute information in the social network can be fully utilized, and potential new community groups can be discovered to a certain extent, so that more reasonable communities can be divided, and the community division precision is improved. According to the method, the structure information and the user attribute information in the social network are fully utilized, the users are subjected to attribution division by adopting the three-way decision division strategy, and potential new communities in the social network can be explored, so that the community division accuracy is improved.

The invention discloses an image processing method, device and equipment and a storage medium, and the method comprises the steps: obtaining a lesion lung image, and carrying out the lung region extraction of the lesion lung image based on a preset lung region contour; obtaining a left lung image and a right lung image based on the extracted lung area; and performing region division on the left lung image based on the left lung volume of the left lung image, and performing region division on the right lung image based on the right lung volume of the right lung image. The lung lobe region division precision can be improved, the lung lobe region division time can be shortened, and an auxiliary effect can be provided for quantitative analysis of lung diseases.

The invention relates to a sequence stratigraphic division method, which belongs to the field of seismic exploration seismic data interpretation, and realizes deep fusion of seismic information and logging information, that is, forward modeling is carried out by using a reconstructed wave impedance curve to obtain a seismic synthesis record, and after the seismic synthesis record and well-crossing seismic data are reconstructed, the seismic synthesis record and the well-crossing seismic data are subjected to forward modeling. And a final time-frequency spectrum is obtained through generalized S transformation, sequence stratigraphic division can be performed by using the time-frequency spectrum, the division precision is high, and the time-frequency spectrum can be divided to the fifth level. The sequence stratigraphic division method is high in practicability and can be popularized and applied in the continental lake basin sedimentary environment.