40 results about "Nonparametric regression" patented technology

The invention relates to an intelligent traffic system, in particular to short-time prediction method and system of traffic flow data, which are used for improving the prediction accuracy of the traffic flow data and are suitable for real-time traffic prediction. The short-time prediction method of traffic flow data, which is provided by the invention, improves the accuracy of short-time traffic flow prediction and acquires optimal K and l values and corresponding predicted traffic flow data by further adding a state pattern vector to a traditional K adjacent nonparametric regression prediction method and adopting search methods of variable adjacent numbers K and match numbers l.

The invention discloses a short-term traffic flow weighted combination prediction method, which comprises the following steps of: (1) organizing historical traffic flow data by utilizing a dynamic clustering algorithm; (2) performing short-term traffic flow prediction by using an improved nearest neighbor nonparametric regression method; (3) performing the short-term traffic flow prediction by taking a cluster which is the most similar to a current point in a historical database as a training sample of a fuzzy neural network and using a fuzzy neural network model; and (4) determining the weight of a combined prediction method according to a prediction error of the improved nearest neighbor nonparametric regression method and the fuzzy neural network model in the last time bucket, and outputting a final prediction result in a weighted combination way. A traffic flow in the last time bucket and a traffic flow of related turning at an upstream road junction are taken into account, the training sample of the fuzzy neural network is optimized, and the final prediction result is output in the weighted combination way, so that short-term traffic flow prediction accuracy and real-time performance are improved.

The invention belongs to the field of intelligent traffic, and relates to an urban road vehicle running speed forecasting method based on road network characteristics. The urban road vehicle running speed forecasting method can be applied to forecasting the vehicle running speed on an urban road during a period of time, a model is improved on the basis of a k-nearest neighbor nonparametric regression method, relations between road sections in a road network are considered, and a matched time series state vector is enlarged into a multi-dimensional space-time state matrix. A sample database is built through collected historical data, real-time data are collected for serving as a template to be matched with a sample, and the vehicle speed of a next time series of a target road section in the obtained neighbor sample serves as the forecasted vehicle speed. The gaussian function is used in the model two times for setting weights for the state matrix and forecasting results in an integrated mode so that the forecasting accuracy can be improved. The model provided in the urban road vehicle running speed forecasting method has the advantages that the data of the small road network with the road section to be detected as a center serve as the state matrix; compared with a prior forecasting model which only gives consideration to data of a current forecasted road section, the model provided in the urban road vehicle running speed forecasting method is higher in accuracy of multi-step forecasting; in addition, the method can offset real-time data or can be used for forecasting under the condition that a road section speed detector breaks down.

The invention discloses a road segment traffic state distinguishing method based on multi-source data. The road segment traffic state distinguishing method comprises the following steps: using traffic parameter values acquired by a fixed detector and a floating car as a data source; selecting historical traffic parameter values and historical traffic state grades in multiple time periods to build a sample database, and calculating an average velocity adjustment parameter of the fixed detector in the smooth traffic period of the road segment, and the road segment historical space average velocity of each period; training to obtain a support vector machine model; respectively adopting a direct judgment method, a K-nearest neighbor nonparametric regression method and a data state correlating analysis method to obtain the road segment space average velocity in the current period; adopting the direct judgment method and the support vector machine model to distinguish the road segment traffic state grade in the current period. According to the road segment traffic state discriminating method disclosed by the invention, the fixed detector and the floating car are used as the data source, and sufficient digging and complementary using are performed on the premise of completely considering the characteristics and the applicability of the data, so that the road segment traffic state discrimination precision is further improved.

The invention provides a wind power probability forecasting method based on numerical weather forecasting ensemble forecasting results. Numerical weather forecasting serves as the foundation, basic input data are provided for short-period wind power forecasting through a numerical weather forecasting ensemble forecasting technology, and a short-period forecasting model is established for each ensemble member to obtain a plurality of groups of forecasting results. For the obtained plurality of groups of forecasting results, and different characteristic forecast errors are classified through an ensemble forecasting configuration characteristic classification method and a forecasting power level classification method to obtain future forecast error bands under certain confidence level. According to the wind power probability forecasting method, under the same confidence level, the error band section is narrower, and for power grids containing large-scale wind power integration, under the condition of the same safety margin, the power grid operation cost can be effectively reduced, and the power grid operation economical property can be improved.

The invention discloses a method for selecting the state vector in the nonparametric regression short-time traffic flow prediction, relating to the technical field of short-time traffic flow prediction. At four conditions comprising peak hours, even hours, low hours and all the day, by using the method provided by the invention, the forecast accuracy, the stability, the speed and the transportability are improved, and the operation time is shortened, thus verifying the effectiveness and the necessity of the method provided by the invention.

The invention discloses a medium-and-long term typical daily load curve prediction method based on function type nonparametric regression which comprises the following steps: according to an existing historical daily load curve, based on a functional data analysis theory and a nonparametric kernel density estimation method, establishing a functional nonparametric regression prediction model; and by considering a daily load factor and a minimum load factor of a typical day to be predicted, establishing a quadratic programming model to correct a prediction curve of the functional nonparametric regression prediction model, and finally, obtaining the prediction curve meeting a load characteristic index requirement of the typical day to be predicted. A simulation example based on typical daily load data of a certain provincial power grid in China and PJM (Pennsylvania-New Jersey-Maryland) electric power company in America proves that the method disclosed by the invention is simple and practical, and is accurate in prediction result. The method has a god popularization value and application prospect.

A congestion prediction algorithm based on historical and peripheral intersection data comprises a first step of performing real-time updating and preprocessing on traffic data; a second step of inputting a predicted target intersection and a time interval; a third step of matching the traffic flow data with historical data of the current area to find out a similar time interval and then carryingout prediction. According to the method, the problem of " traffic flow prediction " is solved by utilizing an algorithm idea of non-parameter regression, the traffic flow is predicted according to historical data and peripheral intersection data, and the traffic data flow is effectively predicted. The congestion situation of the target intersection at the demand time is predicted according to therequirement of a user on the traffic jam prediction situation, and the traffic flow can be efficiently and accurately predicted.

There is provided a method including, acquiring brain function data and diffusion tensor data, calculating a connection degree between voxels adjacent to each other based on the diffusion tensor data, constituting a data evaluation value based on the brain function data and the connection degree between the adjacent voxels, subjecting the data evaluation value to nonparametric regression analysis, and forming and displaying images based on results of the analysis.

Some embodiments provide a system that analyzes telemetry data from a computer system. During operation, the system obtains the telemetry data as a set of telemetric signals from the computer system and validates the telemetric signals using a nonlinear, nonparametric regression technique. Next, the system assesses the integrity of a power supply unit (PSU) in the computer system by comparing the telemetric signals to one or more reference telemetric signals associated with the computer system. If the assessed integrity falls below a threshold, the system performs a remedial action for the computer system.

There is provided a method including, acquiring brain function data and diffusion tensor data (S10), calculating a connection degree between voxels adjacent to each other based on the diffusion tensor data (S30), constituting a data evaluation value based on the brain function data and the connection degree between the adjacent voxels (S40), subjecting the data evaluation value to nonparametric regression analysis (S50), and forming and displaying images based on results of the analysis (S60, S70).

Provided are a blood pressure measuring device, a blood pressure measurement method and a blood pressure measurement program that allow calculating blood pressure properly even for ambulatory users. A blood pressure measuring device 1 is provided with: an electrocardiogram acquisition unit 11 that acquires an electrocardiogram of a user; a pulse wave acquisition unit 12 that acquires a pulse wave of the user; a first extraction unit 13 that extracts a heart rate on the basis of the electrocardiogram; a second extraction unit 14 that extracts a pulse wave velocity on the basis of the electrocardiogram and the pulse wave; a third extraction unit 15 that extracts one or a plurality of feature values pertaining to the pulse wave, on the basis of the pulse wave; and a calculation unit 16 that calculates blood pressure of the user from the heart rate, the pulse wave velocity and the one or plurality of feature values extracted for the user, by a learner 17 that has learned, by nonparametric regression analysis, a relationship between blood pressure of each of a plurality of subjects and a heart rate, a pulse wave velocity and one or a plurality of feature values pertaining to a pulse wave extracted for each of the plurality of subjects.

The invention discloses a power load abnormal data recognition and modification method based on nonparametric regression analysis. The method comprises the steps of 1, performing power utilization mode classification on power load data to obtain a common power utilization mode data set and a special power utilization mode data set; 2, extracting a load feature value at each moment from the obtained common power utilization mode data set by adopting a nonparametric regression analysis method; 3, forming an abnormal data field by using the extracted load feature values according to the selected confidence level; 4, performing load abnormal data recognition on load data in the common power utilization mode data set and the special power utilization mode data set by using the abnormal data field formed in step 3; and 5, modifying the recognized load abnormal value by using an improved introduced load level mapping relation and a weighted mean method considering the influence of feature values. The method can recognize and modify power load abnormal data including big industrial power load data, and simultaneously can overcome the defect of the load abnormal data recognition and modification theory on the aspect of power load data processing.

One embodiment of the present invention provides a system that constructs a functional relationship approximation from a set of data points through nonparametric regression. During operation, the system receives a training data set in an n-dimensional space. Next, the system defines a set of regression primitives in the n-dimensional space, wherein a regression primitive in the set passes through N data points in the training data set, wherein N≧n. The system then logically combines the set of regression primitives to produce a convex envelope F, such that for each point p in the n-dimensional space: (1) F(p)=0, if p is on the convex envelope; (2) F(p)<0, if p is inside the convex envelope; and (3) F(p)>0, if p is outside the convex envelope. The system next obtains the functional relationship approximation by computing an argument of the minimum of F in the n-dimensional space. The system subsequently uses the functional relationship approximation to classify data.

The invention discloses a bayonet traffic flow prediction method. The method comprises the following steps: step 1, preprocessing historical bayonet traffic data; step 2, constructing a data structureto store the historical data; 3, acquiring historical data required by the prediction model from the R tree; and step 4, finding K most relevant historical time periods for prediction in combinationwith the relevant checkpoints and the relevant time periods. According to the problem of traffic flow prediction, an algorithm idea of non-parametric regression is utilized, a method of combining an Rtree and a hash function is used for storing and searching data, historical data and surrounding intersection data are combined for traffic flow prediction, and the prediction requirement of trafficdata flow is efficiently met. According to the invention, the congestion condition of the target intersection at the prediction time can be predicted according to the demand of the user for the traffic congestion prediction condition, and the prediction of the traffic flow can be efficiently and accurately realized.

The invention relates to a mobile communication signal interference prevention assessment method and system. According to the mobile communication signal interference prevention assessment method and system of the invention, in a given context, points are reasonably distributed, and then, a professional measuring instrument is adopted to measure actual values; a time domain finite difference algorithm and a non-parametric regression algorithm are adopted to predict signal strength respectively; analysis and correction are performed on a result through optimizing parameters in a prediction algorithm, and an interference-to-signal ratio can be obtained through calculation; the interference-to-signal ratio is compared with an actual interference effect, so that an interference prevention assessment index system can be obtained; full-range prediction and assessment are carried out; and therefore, the interference prevention effect of mobile communication signals can be judged.

The invention discloses a nonparametric regression method, which relates to the field of forecast methods. The method of the invention comprises the following steps: determining a forecast quantity according to a forecast object; acquiring the properties P1-Pn of the forecast quantity from the forecast object, and using the properties P1-Pn as each component of the forecast object state, wherein n is the number of properties; collecting patterns; constructing a pattern library by a KD tree data structure according to the collected patterns; collecting parameters of the state of the forecast object, composing the current state vectors of the forecast object by the parameters, searching for k numbered patterns similar to the current state vectors in the pattern library according to a predetermined criterion, and acquiring the values y1-yn of the quantities to be forecasted corresponding to the k numbered patterns; substituting the acquired values y1-yn of the quantities to be forecasted into a forecast function to obtain the forecast value yforcast; after a time T, collecting the real value yreal of the quantities to be forecasted; calculating the forecast error e according to the forecast value yforcast and the real value yreal; and adjusting the weight in the predetermined criterion and the structure of the pattern library according to the forecast error e. The method improves the calculation speed and the forecast precision of nonparametric regression forecast, and meets the requirement in practical application.

The invention discloses a method for predicting a regional saturation capacity based on a nonparametric model. The method comprises the steps of (1) establishing a nonparametric regression model, introducing a Gaussian kernel weight function, using a local polynomial estimation method to carry out estimation, and determining a mapping relation between the power demand and influence factors, (2) establishing a nonparametric cumulative model, introducing a secondary planning problem, and confirming a cumulative coefficient based on the nonparametric regression model, (3) selecting an impact factor, (4) selecting an order and a bandwidth according to the amount of collected data, and (5) combining data and substituting the data into the nonparametric regression model and the nonparametric cumulative model to predict electricity consumption and saturation power. According to the established nonparametric cumulative model, the precision of regional saturation capacity prediction is greatly improved, the computational complexity is reduced, the regional saturation capacity can be precisely predicted, and the rationality of regional vision power system planning work is improved.

The invention discloses a power distribution network safe distance identifying method and device. The method comprises the steps that a test site is selected, the mutual distances between a pluralityof test points of the test site, the distances between the multiple test points and a high voltage source and the field strength of the test points are obtained; the field strength gradient at the test site is calculated according to the mutual distances between the multiple test points and the field strength of the test points; a non-parametric regression model between the electric field strengthand the safety distance is established according to the field strength gradient and the distances between the multiple test points and the high voltage source; and the field strength of a working point is brought into the non-parametric regression model to obtain the safety distance of the working point. The power distribution network safe distance identifying method solves the problem that the safety distance cannot be accurately identified on the distribution network operation site, and has the advantages of being wide in application range, high in accuracy, convenient to test, safe, high in efficiency and the like.

The invention relates to a function-type non-parameter model and a KNN estimation method adopting the model. Traffic road condition prediction is effective application of car-networking big data, is conductive to urban traffic management and provides effective references for vehicle owners to select driving routes. Non-parametric regression serves as a parameter-free and high-precision algorithm,a prediction effect is more superior to parametric regression, and errors are smaller. In addition, the speed value in a certain period of time is regarded as a continuous function curve and is analyzed from the perspective of function-type data. A K-neighbor estimation method is adopted, and road flow can be predicted in real time only by determining parameters such as optimum window width.

The present invention discloses a traffic flow prediction method based on data-driven k-nearest non-parametric regression. The method is characterized in that, based on the development of a two-step data search algorithm, firstly, in a non-predictive time period, candidate input data is searched and identified from a historical database to be approximated with the current state; optimal decision input data for prediction is identified from the candidate input data at the prediction point; and finally the prediction is generated through the prediction algorithm by using the optimal decision input data. According to the algorithm provided by the present invention, the time for searching historical data can be effectively reduced, the execution time in the system prediction process can be reduced, the prediction efficiency of the prediction system can be improved, and the accuracy for the system prediction can be ensured.

Some embodiments provide a system that analyzes telemetry data from a monitored system. During operation, the system obtains the telemetry data as a set of telemetric signals from the monitored system and groups the telemetry data into one or more clusters of correlated telemetric signals from the telemetric signals. Next, the system increases a bandwidth associated with monitoring the telemetric signals. To increase the bandwidth, the system omits one or more of the correlated telemetric signals from each of the clusters during sampling of the telemetric signals and estimates the omitted correlated telemetric signals by applying a nonlinear, nonparametric regression technique to the sampled telemetric signals.

The invention discloses an electromagnetic wave signal amplitude attenuation and propagation distance relation curve fitting method, which comprises the steps of step 1, signal abnormal value elimination: regarding the distribution of acquired signal data as Gaussian distribution, and eliminating abnormal values distributed outside the twice variance of an average value in the data when the probability density distribution meets N-(mu, sigma2); step 2, filtering by adopting a Kalman filtering algorithm; and step 3, fitting by adopting a semi-parameter regression model. Through the filtering method of the time acquisition sequence, the authenticity of the ultrahigh frequency receiving signal strength is improved, and on the basis, the fitting precision is improved by adopting the nonparametric regression model.

The invention discloses a power load abnormal data recognition and modification method based on nonparametric regression analysis. The method comprises the steps of 1, performing power utilization mode classification on power load data to obtain a common power utilization mode data set and a special power utilization mode data set; 2, extracting a load feature value at each moment from the obtained common power utilization mode data set by adopting a nonparametric regression analysis method; 3, forming an abnormal data field by using the extracted load feature values according to the selected confidence level; 4, performing load abnormal data recognition on load data in the common power utilization mode data set and the special power utilization mode data set by using the abnormal data field formed in step 3; and 5, modifying the recognized load abnormal value by using an improved introduced load level mapping relation and a weighted mean method considering the influence of feature values. The method can recognize and modify power load abnormal data including big industrial power load data, and simultaneously can overcome the defect of the load abnormal data recognition and modification theory on the aspect of power load data processing.

The invention discloses a mix proportion design method of a high-toughness cement-based engineering composite material based on uniform experiments and ACE nonparametric regression. The method comprises the following steps that firstly, the optimal objective of the mix proportion of the high-toughness cement-based engineering composite material is determined; then experimental factors are selected, the levels of each factor are determined, a uniform design table and an application table of the uniform design table are selected to combine the factors, a mix proportion table is obtained, and the experiments are conducted; experimental data is processed and analyzed by using the ACE nonparametric regression method, screening is conducted according to the experimental objective and constraint conditions, the optimal mix proportion of the high-toughness cement-based engineering composite material is obtained, and finally, an experiment is conducted according to the optimal mix proportion so as to verify the reasonability of the experiments. Compared with other experiment design methods, the optimization design method has the advantages that under the condition of the same number of the levels, the method has the fewest experiment times, and the experimental data has more uniformity and representativeness, and the method is simple in operation.

The invention discloses a short-term traffic flow weighted combination prediction method, which comprises the following steps of: (1) organizing historical traffic flow data by utilizing a dynamic clustering algorithm; (2) performing short-term traffic flow prediction by using an improved nearest neighbor nonparametric regression method; (3) performing the short-term traffic flow prediction by taking a cluster which is the most similar to a current point in a historical database as a training sample of a fuzzy neural network and using a fuzzy neural network model; and (4) determining the weight of a combined prediction method according to a prediction error of the improved nearest neighbor nonparametric regression method and the fuzzy neural network model in the last time bucket, and outputting a final prediction result in a weighted combination way. A traffic flow in the last time bucket and a traffic flow of related turning at an upstream road junction are taken into account, the training sample of the fuzzy neural network is optimized, and the final prediction result is output in the weighted combination way, so that short-term traffic flow prediction accuracy and real-time performance are improved.

The invention discloses a traffic flow forecasting method based on data-driven k-neighborhood non-parametric regression, which is developed based on a two-step data search algorithm. First, within a non-prediction period, candidate input data is searched and identified from a historical database to be compared with the current The state is approximated, then the best decision input data for prediction is identified from the candidate input data at the prediction point, and finally the best decision input data is utilized to generate a prediction by a prediction algorithm. This algorithm can effectively reduce the time for searching historical data, thereby reducing the execution time in the system forecasting process, improving the forecasting efficiency of the forecasting system, and at the same time ensuring the accuracy of the system's forecasting.

The invention discloses a medium-and-long term typical daily load curve prediction method based on function type nonparametric regression which comprises the following steps: according to an existing historical daily load curve, based on a functional data analysis theory and a nonparametric kernel density estimation method, establishing a functional nonparametric regression prediction model; and by considering a daily load factor and a minimum load factor of a typical day to be predicted, establishing a quadratic programming model to correct a prediction curve of the functional nonparametric regression prediction model, and finally, obtaining the prediction curve meeting a load characteristic index requirement of the typical day to be predicted. A simulation example based on typical daily load data of a certain provincial power grid in China and PJM (Pennsylvania-New Jersey-Maryland) electric power company in America proves that the method disclosed by the invention is simple and practical, and is accurate in prediction result. The method has a god popularization value and application prospect.

The invention relates to an intelligent traffic system, in particular to short-time prediction method and system of traffic flow data, which are used for improving the prediction accuracy of the traffic flow data and are suitable for real-time traffic prediction. The short-time prediction method of traffic flow data, which is provided by the invention, improves the accuracy of short-time traffic flow prediction and acquires optimal K and l values and corresponding predicted traffic flow data by further adding a state pattern vector to a traditional K adjacent nonparametric regression prediction method and adopting search methods of variable adjacent numbers K and match numbers l.

The invention relates to a mobile communication signal interference prevention assessment method and system. According to the mobile communication signal interference prevention assessment method and system of the invention, in a given context, points are reasonably distributed, and then, a professional measuring instrument is adopted to measure actual values; a time domain finite difference algorithm and a non-parametric regression algorithm are adopted to predict signal strength respectively; analysis and correction are performed on a result through optimizing parameters in a prediction algorithm, and an interference-to-signal ratio can be obtained through calculation; the interference-to-signal ratio is compared with an actual interference effect, so that an interference prevention assessment index system can be obtained; full-range prediction and assessment are carried out; and therefore, the interference prevention effect of mobile communication signals can be judged.