38results about How to "Solve the problem of inaccurate estimates" patented technology

The present invention provides an advertisement click rate estimation method based on a decision tree and a device. The method comprises a step of obtaining the related characteristic information of a specific historical advertisement in a predetermined historical time period, a step of obtaining the personalized characteristic information of a target user, a step of carrying out decision tree model calculation based on the obtained target user personalized characteristic information and the related characteristic information of the specific historical advertisement so as to determine a cross characteristic vector for estimating the click rate of the specific historical advertisement, wherein each leaf node of the decision tree model represents one cross characteristic, a step of calculating based on the cross characteristic vector and the model training parameter obtained from training in advance so as to estimate the click rate of the specific historical advertisement. The invention also provides an application software recommendation method and a device. According to the method or the device, through the decision tree model, the cross characteristic with a strong classification characteristic is generated, and thus the advertisement click rate is accurately and rapidly estimated.

The invention discloses a multi-starting-point incremental three-dimensional reconstruction method based on large-scale disordered images. The method comprises the steps of performing image matching, constructing a matching relationship chart, selecting a reconstruction starting point, detecting an edge, performing multi-starting-point three-dimensional reconstruction and splicing a submodel. By means of a clustering strategy and an iteration drift strategy, an image pair which is suitable as starting points is adaptively selected from a disordered image set and reconstruction is performed simultaneously; before reconstruction, determining an optimal reconstruction sub-process of each image according to a layered shortest path first algorithm, and determining a reconstruction boundary; and obtaining a whole three-dimensional reconstruction model according to sub-models obtained in different sub-processes and splicing the common parts contained in the sub-models. The three-dimensional reconstruction method according to the invention has advantages of settling a problem of suspension in reconstruction process when a current scene comprises a plurality of non-overlapped parts, obtaining a whole three-dimensional model which is covered by the image set, preventing three-dimensional structure error caused by matching and transmission of error images, realizing parallel processing in reconstruction sub-process, and improving image reconstruction efficiency.

The invention discloses a method and a device for analyzing an internal state of a lithium battery based on an ultrasonic frequency spectrum. The method comprises the steps of enabling an ultrasonic signal with a set frequency to penetrate through the lithium battery, and calculating a charge state of the lithium battery by use of a linear relation between a main lobe power and a charge state of the ultrasonic signal; and judging the health of the lithium battery by use of the ratio of the power of the higher harmonics to the main lobe. Another method comprises the steps of driving an ultrasonic transducer by use of a swept-frequency signal to generate ultrasonic signals of different frequencies to penetrate through the lithium battery to obtain the frequency response function of the lithium battery; and judging the aging degree of the lithium battery by use of the frequency response function. The invention provides the device for analyzing the internal state of the lithium battery based on the ultrasonic frequency spectrum. According to the method and the device for analyzing the internal state of the lithium battery based on the ultrasonic frequency spectrum disclosed by the invention, the frequency spectrum of the ultrasonic signal is deeply analyzed, the problem that the SOC (State Of Charge) and SOH (State of Health) estimation of the lithium battery is not accurate is effectively solved, and the device can be integrated into a battery management system to realize the real-time measurement of the internal state of the battery, and timely make early warning for the possible failure of the battery.

The present invention provides an automobile battery state of charge correction method and device. The method comprises the steps of real-timely calculating the to-be-corrected state of charge values, the temperatures and the open-circuit voltage values of a battery pack during the operation process of an automobile; testing to obtain an open-circuit voltage-state of charge curve of the battery pack under different temperatures, and obtaining the areas satisfying the preset conditions in the open-circuit voltage-state of charge curve as the effective correction voltage areas; selecting the effect correction voltage area under the corresponding temperature according to the temperature of the battery pack during the operation process of the automobile, and obtaining the corresponding state of charge value in the selected effective correction voltage area according to the open-circuit voltage value of the battery pack during the operation process of the automobile. According to the automobile battery state of charge correction method and device provided by the present invention, the problem that the automobile battery state of charge estimation is not accurate enough, is solved.

The invention discloses an FDS-based oilpaper capacitive bushing series polarization model building method. The method comprises the following steps of: 1, enabling each layer of capacitive medium ofan oilpaper capacitive bushing to be equivalent through connecting a capacitor and a resistor in parallel; 2, simulating a relaxation polarization phenomenon of each layer of capacitive medium througha branch circuit connected with the capacitor and the resistor in series, and enabling a relaxation polarization project of each layer of capacitive medium to be equivalent through connecting an RC series branch circuit on the basis of the step 1, so as to obtain a polarization circuit mode of each layer of the oilpaper capacitive bushing; and 3, connecting the n polarization circuit models, obtained in the step 2, of the layers of the oilpaper capacitive bushing to obtain an oilpaper capacitive bushing series polarization model. According to the method, the problem that the model parameter estimation is not correct enough as oilpaper capacitive bushing modeling time domain dielectric spectrum maps are easy to be interfered by environmental noise is solved.

The invention provides a method, device and system for high-speed mobile communication, wherein the method comprises the steps of: normally transmitting uplink data PUSCH/PUCCH/SRS on UE, and recording the current time T0, wherein the PUSCH represents an uplink control channel; the PUCCH represents an uplink shared channel; and the SRS represents a sounding reference signal; according to the uplink data, obtaining TA0 by calculation by a base station, and notifying the UE; obtaining the final TA value TA2 by calculation; and, obtaining the final time synchronization adjustment amount at the T2time according to the TA2, and, on this basis, transmitting the uplink data. According to the method, device and system for high-speed mobile communication provided by the invention, on the basis ofan airborne terminal, calculation is carried out corresponding to the movement distance of the base station; inaccurate TA sent by the base station is corrected; therefore, the problem that TA estimation is inaccurate due to high-speed movement of an LTE system in an ATG network can be solved; and thus, the performance of an LTE ATG system is improved.

The invention discloses an SOC adaptive correcting method for a secondarily utilized battery. The SOC adaptive correcting method for the secondarily utilized battery comprises the following steps: 1)acquiring battery running data in a specific voltage range; 2) analyzing the battery running data, calculating available capacity Ca of a battery under discharging rate current Ia by using an ampere-hour integral method, and calculating equivalent available capacity C'a of the battery under standard current I0 according to a current-capacity curve of the battery; 3) calculating attenuation changerate of the available capacity of the battery, if the attenuation change rate is in a normal range, updating the available capacity C'0=C'a of the battery under the standard current I0, not updating the available capacity C'0=C'a of the battery if the attenuation change rate exceeds the normal range, and reporting abnormity and alarming; and 4) calculating a SOC correcting coefficient k by using the available capacity C'0 of the battery under the standard current I0 and initial available capacity, and updating the SOC correcting coefficient of the battery. The accuracy of SOC estimation of thebattery is improved.

The invention provides a position and speed estimation method and a position and speed estimation device for a PMLSM servo control system, which relate to the technical field of motor control. The position and speed estimation method comprises the steps of: adding a winding resistance factor into a state variable of the PMLSM servo control system; subjecting a state equation of the PMLSM servo control system to disperse linearization so as to obtain a disperse linearization model; and estimating a position and a speed by adopting an extended Kalman filtering method according to the disperse linearization model. According to the position and speed estimation method and the position and speed estimation device, the winding resistance factor is added into the state variable to form the state augmented extended Kalman filtering method (AEKF), the position and speed of a mover of a PMLSM are estimated, and the influence on resistance parameters of temperature changes when a linear motor operates at a low speed can be reduced, thereby improving the estimation precision of the algorithm.

The invention discloses an estimation method for mobile points of a software reliability growth model. The estimation method comprises the first step of drawing a quality control chart according to software system actual failure data, the second step of obtaining initial mobile points according to the quality control chart and a mobile point judging criterion, and the third step of obtaining the accurate positions of the mobile points by using a gradual adjustment method. According to the estimation method for the mobile points of the software reliability growth model, the mobile point judging criterion is provided, the preliminary judgment of the mobile points are more accurate, the number, obtained in an estimated mode, of the mobile points is accurate, and positional deviation is small. The estimation method for the mobile points of the software reliability growth model is a new mobile point estimation method, and the problem that the position estimation of the mobile points is not accurate is solved by using the gradual adjustment method on the basis that the number and the positions of the initial mobile points are obtained by using the quality control chart.

The embodiment of the invention discloses a circulation delay diversity signal channel estimation method, which comprises the following steps: obtaining the delay value of the selected circulation delay diversity CDD; extracting signal channel parameters in the pilot frequency position according to the delay value of the selected CDD, wherein the signal channel parameters are in linear distribution; and obtaining all signal channel parameters according to the extracted signal channel parameters. At the same time, the embodiment of the invention also provides a circulation delay diversity signal channel estimation system and equipment. Through the technical scheme provided by the invention, the signal channel parameters of the CDD signal channels are converted into the linear values, the problem of inaccurate signal channel estimation can be solved, and the frequency selection performance of the signal channel is added.

The invention discloses a method for estimating the sector capacity of a mobile communication system. The method comprises the steps as follows: the average business volume corresponding to each business in each sector in an area is determined; the blocking probability value of each business in the sector is determined; and when at least one business blocking probability value in the blocking probability value of each business in the sector is not smaller than a set specified value, a fixed amount of basic source units are added to the preset average sector capacity, the average sector capacity in which a fixed amount of basic source units are added returns to continuously execute treatment to determine the blocking probability value of each business, the execution is carried out until the blocking probability value of each business in the sector is smaller than the set specified value, and the sector capacity is determined according to the average sector capacity in which a fixed amount of basic source units are added. The invention solves the problem that the sector capacity of the mobile communication system is not estimated accurately in the prior art.

The invention relates to method for estimating the parameters of ZnO-varistor-equivalent circuit model based on FDS, belonging to the technical field of lightning arrester model. The method comprisesthe following steps. S1: The measuring instrument of frequency domain dielectric spectroscopy (FDS) and the ZnO varistors are prepared as a part of the experiment platform. S2: The parameters are properly set on the measuring instrument of frequency domain dielectric spectrum. S3: The FDS of the ZnO varistors are measured multiple times until the final results are obtained. S4: The complex impedance of the equivalent model is calculated using ZnO-varistor-equivalent polarized circuit model. S5: The parameter function of the equivalent model is optimized based on the measured FDS data of ZnO varistors and the results calculated by the model. S6: The iterative calculation on the optimized parameter function is performed using genetic algorithm to finalize the parameter estimation for the equivalent model. The method of the invention satisfactorily describes the dielectric response of ZnO varistors, and remediates the insufficient explanations by R C parallel model and the IEEE model on the microscopic behavior of ZnO varistors in terms of polarization and electric conductance among others.

The invention discloses an image array optical flow estimation method for an artificial compound eye camera. The method comprises the steps of detecting an image comprising a moving target in array images according to structural characteristics of used artificial compound eyes; building an optical flow estimation energy function; and optimizing the energy function by utilizing a variational method, thereby finishing optical flow estimation of artificial compound eye images. Compared with the prior art, the method has the remarkable advantages that an artificial compound eye image sequence canbe effectively subjected to the optical flow estimation, and due to constraint items of spatial correlation among sub-eye images, corresponding regions of adjacent sub-eyes can be constrained to obtain a consistent optical flow estimation result, so that the problem of inaccurate image edge optical flow estimation caused by small view field angle of the sub-eyes in the artificial compound eye camera is solved.

The embodiment of the invention discloses a mobile state estimation method and device of a terminal, electronic equipment and a storage medium, and the method comprises the steps: obtaining the residence time of the terminal in each NTN GEO cell in a non-ground network NTN GEO scene; and estimating the moving state of the terminal according to the relationship between the residence time and a threshold range, wherein the two end points in the threshold range are determined according to the radius ratio of the coverage radius of the NTN GEO cell to the coverage radius of the common land cell. According to the embodiment of the invention, the mobile state of the terminal is estimated by judging the relationship between the residence time of the terminal in each NTN GEO cell and the threshold range and combining the radius ratio of the coverage radius of the NTN GEO cell to the coverage radius of a common land cell; and the problem of inaccurate terminal movement state estimation caused by too large cell coverage radius in the NTN GEO scene is solved.

The invention provides a system load estimation method, device and equipment and a readable storage medium, and the method comprises the steps: obtaining to-be-analyzed sample log data, and extracting a first impact factor of the sample log data; establishing a simulation influence relation criterion according to the first influence factor; obtaining a load threshold value set by the system, performing linear conversion on the simulation influence relation criterion according to the load threshold value, and generating a vector inner product; according to the vector inner product, performing dynamic correction on noise points in the first influence factor to generate a second influence factor; and calling a trained prediction model to predict the second influence factor, and generating a final load. The problem of inaccurate estimation caused by noise interference when the system load is estimated is solved.

The invention relates to a bus arrival time estimation method and device, and belongs to the technical field of intelligent traffic. The method comprises the following steps of segmenting a bus running line, determining the running speed of each section of line according to each section of factor data influencing the bus running speed by using a vehicle running speed model, dividing the running speed of the corresponding section by a segmentation distance to obtain the running time of each section, accumulating to obtain the inter-station running time, and determining the estimated arrival time of each station. Therefore, the arrival time of the bus can be estimated. According to the invention, the condition that the road conditions on each section of the line are different and the types of factors influencing the running speed of the vehicle are changed is considered, the arrival time is estimated by calculating the running time of each section, the prediction process accords with with the actual road conditions, and the obtained estimated time is more accurate.