53 results about "Projection pursuit" patented technology

A method and system of object classification uses measurements for a training set of objects to classify an unknown object as being a member in one of several classes of the training set. The classes are defined by features of the training set, the objects of which have known class memberships. The method comprises performing principal component analysis on the training set measurements to discard features that have negligible information regarding class membership, conducting projection pursuit on the remaining training set measurements to accentuate differences between the classes, estimating a distribution of each accentuated class in the training set, and evaluating measurements of the unknown object to determine the membership of the unknown object in one of the accentuated classes. The system implements the method in a computer program stored in computer memory.

The invention discloses a projection pursuit-based risk evaluation method for meteorological disasters of facility agriculture, belonging to the technical field of the evaluation of the meteorological disasters. The projection pursuit-based risk evaluation method comprises the following steps of: fitting and acquiring historical meteorological data by means of a neural network; counting the annual occurrence frequency of the meteorological disasters in various grade, calculating to acquire disaster comprehensive indexes of evaluation indexes, and establishing disaster comprehensive index grading standards of the evaluation indexes; constructing projection functions of the historical meteorological data, and acquiring the optimal projection vector by an accelerating genetic algorithm; and building a risk evaluation model according to projection values of the meteorological data on the optimal projection vector and the disaster comprehensive index grades of the evaluation indexes. The projection pursuit method is applied to the field of the risk evaluation of the meteorological disasters of the facility agriculture, the projection vectors are optimized by combining with the accelerating genetic algorithm, and the actual meteorological data are taken as input data of the evaluation model, the accuracy of risk evaluation results is high, and new thoughts and methods are provided for the research of the risk evaluation of the meteorological disasters.

The invention discloses an SAR image segmentation system and a segmentation method based on immune clone and projection pursuit. The system comprises an image characteristic-extracting module, an initial label-selecting submodule, a projection direction-selecting submodule and a subspace clustering submodule, wherein the image characteristic-extracting module extracts the gray co-occurrence matrixcharacteristics, the wavelet characteristics, the brushlet characteristics and the contourlet characteristics of an input image; the initial label-selecting submodule clusters the image characteristics to acquire and transmit an initial label to the projection direction-selecting submodule for calculating a linear judgment analysis projection index and acquiring an optimal projection direction; the subspace clustering submodule projects the image characteristics in the optimal projection direction, acquires and clusters an optimal subspace to acquire a clustering label, returns the clusteringlabel to the initial label-selecting submodule for iteration, acquires a final clustering label corresponding to image pixels and acquires a final image segmentation result. The invention has the advantage of high segmentation accuracy and can be applied to civil and industrial fields or as martial reconnaissance means.

The invention discloses a projection pursuit hyperspectral image segmentation method based on transfer learning, belonging to the technical field of image processing. The technical key points are as follows: the multiband characteristic of the hyperspectral image data is utilized to regard each wave band as a gray level image, each image is analyzed and researched; the characteristic that the images of different wave bands are similar but not the same to introduce transfer learning in the projection pursuit clustering algorithm; and a ground object marked graph is used to obtain the label of the source domain image data, and the known label knowledge is utilized to guide the image data without label in the object domain and obtain the optimal projection direction and the optimal subspace, thus increasing the segmenting precision. The method has the advantage that the priori knowledge is utilized to increase the segmenting precision, and can be used for military reconnaissance and in the civil and industrial fields.

The invention discloses a data reconfiguration and decompression method of a power system based on projection pursuit. A signal can be sampled at frequency far below Nyquist frequency on the basis of a compressive sensing theory. The method comprises the following steps of: firstly, constructing an over-complete atom dictionary to ensure that atoms in the atom dictionary are redundant; secondly, searching an optimally-matched atom, updating signal residuals, and re-acquiring sparse vector elements; and finally, acquiring a sparse vector. According to the method, in an iteration process at each time, the atom which is optimally matched with the signal is selected from a sensing matrix to construct sparse approximation, signal residual representation is obtained, and the signal is linearly represented by a certain number of atoms by certain iteration frequency. According to the conventional data compression method, sampling and compression are required to be performed in sequence. According to the data reconfiguration and decompression method, the sampling and the compression are performed concurrently, and an original electric energy quality signal can be well restored by low sampling frequency.

The invention relates to a method for selecting a protective farming mode, which is characterized by comprising the following steps: 1) data collection: soil data, climate data, geographic data and production habit data of a mature area and an area to be implemented in different modes are collected; 2) preparation of mode selection: a decision unit and an index of each protective farming mode are determined according to the data of the mature area; 3) construction of a projection pursuit analysis model; 4) construction of a similarity evaluation model; 5) model selection: the protective farming mode corresponding to the mature area with the highest similarity to the area to be implemented is selected as the protective farming mode of the area to be implemented according to the level of similarity. The invention selects a proper protective farming mode for the area to be implemented by constructing a protective farming mode selection mode, and the whole process is completed by computer analog simulation without field experiment, thereby improving the efficiency of protective farming mode selection, and reducing the cost for popularizing protective farming.

Provided is a cross-district interconnecting large power network homology identification method based on projection pursuit. Rotate speed information of a generator is read from a wide area measurement system to generate a standardized rotate speed sample matrix, a mathematic model of electrical power system unit homology grouping is established according to the rotate speed sample matrix, the mathematic model of the electrical power system unit homology grouping is solved by adopting a genetic algorithm to determine all projection directions and corresponding target evaluation function values, the optimum projection direction, and the projection contribution rate and the accumulated contribution rate of the rotate speed sample matrix are determined according to projection evaluation function values to determine the homology grouping results of an electrical power system. According to the cross-district interconnecting large power network homology identification method based on the projection pursuit, calculation is simple and easy to achieve, due to the fact that the unit homology grouping is carried out based on the wide range actual measured data of the electrical power system, the influences on the electrical power system unit grouping results by element model parameters and system model parameters in the electrical power system are avoided, and on-line identification of the electrical power system homology unit group is achieved.

The invention discloses a projection pursuit dynamic cluster method for a multidimensional index based on particle swarm optimization, which effectively combines the advantages of APH, the projection pursuit, the particle swarm optimization and the like, and provides a new projection pursuit dynamic cluster method for the multidimensional index based on the particle swarm optimization. The new method effectively avoids the problems that the subjectivity of the AHP is too strong and the ordinary projection pursuit neglects index weight. Through the application of the method, cluster evaluations of high-capacity and high-dimensionality multidimensional index data of business can be realized, therefore, the method has high practical value.

A pattern matching system that outputs an likeness values, comprising a designation block, a comparison block, a calculation block and an output block, is provided. The likeness value shows how much a first and second image compares to each other. The designation block designates a first and second image data as data to be compared. The first and second image data comprises 2̂x of area data. The comparison block compares the data level of the area data corresponding to pattern areas at the relatively same location in the first and second images. The calculation block calculates the likeness value. The likeness value varies according to the number of the pattern areas where the absolute value of the difference between the compared data levels of the area data is less than a predetermined standard value.

Provided is a task acceptance simulation method for a labor crowdsourcing platform. The method includes the following steps of clustering member groups according to member information of the crowdsourcing platform and historical participation task information by using projection pursuit evaluation and combination; delimiting simulation areas, determining the number of different member groups, locations and rewards of task points, and randomly generating initial positions; according to the different types of member groups and a cellular automata theory, designing corresponding evolution rules respectively, evolving according to the rules, and simulating a task accepting process; after the simulation is completed, extracting the member information applied of each task point, using an ideal point method for evaluation, and proving suggestions for improvement. The method simulates acceptance conditions of tasks in the crowdsourcing platform from a microscopic point of view, considers a plurality of influencing factors and historical prior information, has high simulation precision and reasonable calculation amount, and can provide reference basis for defining task rewards and task ranges.

The invention provides a data-driven spacecraft state prediction method, first a prediction sample is input, predictive factors are determined, a projection value is calculated, then a projection pursuit auto-regressive model is established, a projection indicator function is built, an acceleration genetic algorithm is used to optimize a projection direction, the projection pursuit regressive model is finally determined, and parameter prediction is performed. The data-driven spacecraft state prediction method provided by the invention can find the structure or characteristic from telemetry data themselves which do not conform to normal distribution or have little prior information, and can solves the problem of a nonlinear structure in linear projection. The data-driven spacecraft state prediction method has a good prediction effect, and has positive significance to spacecraft fault prediction and diagnosis.

The invention discloses a high-spectral image classification method, and relates to the technical field of information processing. An image to be classified is segmented in multiple dimensions, a salient map is extracted from the image after multi-dimension segmentation, features are extracted from the image from which the salient map is extracted, the extracted image features are normalized, and the normalized image features are classified in a nonlinear kernel function method; and the nonlinear kernel function method concretely refers to one selected from a Gaussian process, a support vector machine, kernel principal component analysis, a kernel function Fisher discriminant method and a kernel projection pursuit method. The image data bulk is reduced greatly, and the classification speed is high; and via nonlinear mapping, influence of radiation and geology errors is reduced greatly, the high-spectral image classification precision is higher, and the method of the invention is suitable for application occasions of the high-spectral remote-sensing images.

The invention discloses an evaluation method of a country-city change index. The evaluation method includes the steps of 1, determining study areas, to be specific, according to current situation of the administrative division of each prefecture-level city, laying off municipal districts from cities to represent urban area types; 2, determining a study period, to be specific, the change period from a starting point T1 to an end point T2; 3, determining an evaluation index system of 'country-city' change processes of prefecture-level cities, and collecting index data of the prefecture-level cities of the year T1 and the year T2 and inputting the same in a database; 4, performing projection pursuit classification counting on original matrixes of evaluation indexes of the prefecture-level cities; 5, based on projection values of the 'country-city change processes of the prefecture-level cities, obtaining comprehensive evaluation results of the 'country-city' change processes of the prefecture-level cities. In this way, quantitative evaluation of the country-city change process and feature of the prefecture-level cities in China is developed in the objective and fair manner, evaluation results are high in practical application value and the method has bright application prospect.

The invention relates to a data-driven comprehensive energy efficiency evaluation method of a substation based on data driving, comprising the following steps: dividing the energy efficiency index ofthe substation and collecting various energy consumption data of the substation to be evaluated; establishing the comprehensive energy efficiency evaluation matrix of the substation to be evaluated and obtaining the energy efficiency evaluation index matrix of the substation to be evaluated after the standardization of the same trend; obtaining the evaluation vector of projection value and constructing the distance function and intra-class density function between classes,Constructing maximum information entropy function,based on maximum information entropy and projection pursuit, establishinga substation comprehensive energy efficiency evaluation model, and using the differential evolution algorithm to solve the coupling model, and obtaining the vector of comprehensive energy efficiencyevaluation value of the substation to be evaluated. The invention takes the collected energy consumption data of the substation as a driving force, constructs a coupling model of maximum information entropy and projection pursuit, and adopts a differential evolution algorithm to solve the problem, thereby reducing the uncertainty of the projection vector distribution, and making the obtained comprehensive energy efficiency evaluation result of the substation more objective and reliable.

A method and a system for data imputation and classification are provided. The system includes a database, a historical sample imputation module and a current sample imputation and classification module. In the method, at first, an imputation calculation is performed on each of classified historical sample groups to obtain a basis matrix and a missing value corresponding to each of the classified historical sample groups. Thereafter, a sample classification stage is performed. In the sample classification stage, an IPP (Iterative Projection Pursuit) algorithm and an equation of nonlinear inequality constraints to calculate weighting vectors corresponding to a current sample. Thereafter, plural candidate samples corresponding to different classes are calculated in accordance with the basis matrix and the weighting vectors, and the sample class of the current sample and a prediction value for a missing value of the current sample are determined accordingly.

The invention relates to a water resource development and utilization efficiency evaluation method and application thereof. The water resource development and utilization efficiency evaluation method comprises the steps of: 1, constructing a regional industrial comprehensive level evaluation index system and a water resource development and utilization efficiency evaluation index system; 2, evaluating a regional industrial comprehensive level and water resource development and utilization efficiency respectively by adopting common evaluation methods (AHP, entropy evaluation method, projection pursuit and the like); 3, performing linear weighting on two evaluation results to obtain a final water resource development and utilization efficiency evaluation result. The water resource development and utilization efficiency evaluation method provided by the invention can better reveal the development and utilization degree and level of water resources, further clarifies the potential of water resource development and utilization, and provides a theoretical basis and decision support for the urban water resource optimized configuration and industrial structure adjustment.

The present invention relates to the rail traffic electronic utilization energy consumption analysis field, especially to an electricity utilization energy consumption index assessment method facing rail traffic. The method comprises concretely steps: (1) performing graduation of energy consumption indexes, and generating projection indexes in a grade range according to the random distribution principle; (2) establishing a projection target function (3) through taking the projection indexes as a projection value as a certain direction to solve the maximum value of the projection target function through the genetic algorithm, and solving a projection direction and an optimal projection value; and (4) establishing an energy consumption assessment model according the projection value and the assessment grade scatter diagram. The electricity utilization energy consumption index assessment method facing the rail traffic employs the genetic algorithm and the projection pursuit method to provide an electricity utilization energy consumption index assessment method facing the rail traffic to establish an energy consumption index comprehensive assessment genetic projection pursuit model, employs examples to prove the model feasibility, and performs comprehensive assessment of a transfer station and an operation line, and the assessment result is comprehensive, objective and reasonable so as to provide important basis for assessment of the rail traffic electric energy consumption.

The invention discloses a dimension reduction clustering data analysis method. The method comprises the following steps of: S101, generating sample data, carrying out a dimensionless method for the sample data and constructing projection data needed by a model; S102, constructing a projection index and obtaining an optimal projection direction vector as shown in the specification; and S103, carrying out linear projection on the projection data to obtain a one-dimensional projection characteristic value. Through the abovementioned steps, a projection pursuit technology and a dynamic clustering method can be combined and are applied to a high-dimensional data dimension reduction clustering model, the operation is simple and the model objectivity is also improved; for the possible situation of decision-maker preference, constraint conditions are increased so that the model can comprehensively consider objective weights and the decision-maker preference and the application range of the model is widened.

The invention discloses a risk evaluation method of flood resource utilization of water transfer engineering, and belongs to the technical field of risk evaluation of flood resource utilization of water transfer engineering. The method comprises the following steps: step 1, obtaining basic data of flood resource utilization of the to-be-studied engineering; step 2, constructing a risk evaluation system of flood resource utilization of the water transfer engineering; step 3, determining risk evaluation levels and standards of flood resource utilization of the water transfer engineering; step 4,calculating flood resource utilization risks of water storage engineering, water transfer engineering and water pumping engineering risk indicators; step 5, coupling objective weights calculated andobtained by an entropy weight method and a projection pursuit method and initial weights calculated by an analytic hierarchy process, and using a combined weight method to calculate index weights; step 6, calculating indicator membership degrees through a conditional cloud generator; and step 7, obtaining an overall evaluation level. The combined weight cloud model-based risk evaluation method offlood resource utilization of the water transfer engineering provided by the invention provides a basis for safe flood resource utilization of the water transfer engineering.

The invention discloses a user load adjustment and control value evaluation method participating in orderly power utilization. The method comprises. The method comprises the following steps: S1, building a load adjustment and control value index system through the full consideration of a plurality of impact factors in different modes of rest adjustment, peak avoidance and time staggering; S2, performing dimension reduction processing on the matrix under each potential index based on a projection pursuit dynamic clustering model, thereby realizing quantification of user adjustment potential; and S3, comprehensively evaluating the load adjustment and control value levels of the users participating in orderly power utilization by utilizing the matter-element extension theory. The implementation process is simple, the orderly power utilization value potential of the users can be effectively mined, the users suitable for participating in orderly power utilization enterprises can be screenedby a power grid company, the safety and stability of power grid operation are improved, and the economic value is remarkable.

The invention relates to a method for measuring the humification degree of organic waste and a method for applying matured compost by different quality. A quantitative evaluation method for the humification degree of organic waste compost is provided, so that the best utilization way of the compost is recommended according to the humification degree of the compost, and the recycling efficiency of the organic waste compost is maximized. The method for measuring the humification degree comprises steps as follows: 1, fulvic acid of to-be-utilized compost is extracted and separated; 2, 3D fluorescence data of fulvic acid in step 1 are acquired by use of a fluorescence spectrophotometer; 3, integral ratio indexes of 3D fluorescence areas of visible and ultraviolet regions are calculated through MATLAB programming, and values of components C1%, C2% and C3% of fulvic acid are calculated through parallel factor analysis; 4, the projection value F is calculated by use of a projection pursuit model. The application method comprises steps as follows: the optimum utilization modes of different composts are recommended according to the projection value F.

The invention discloses a post-evaluation method for benefits of dam danger elimination and reinforcement based on an improved cloud model, and belongs to the technical field of reservoir dam danger elimination and reinforcement benefit evaluation. It includes the following steps: Step 1, obtain the relevant index data after the reinforcement of the dams to be studied, and build a post-evaluation system for the benefits of reinforcement of the reservoir dams; Step 2, formulate according to the requirements of different reservoir dams and the development situation of each region For the corresponding expected value, standardize each index value and expected value to improve the universality of the system, and determine the evaluation level and standard of the comprehensive benefit of dam danger reduction and reinforcement; Step 3, calculate the projected value obtained by the projection pursuit method and the entropy weight method The objective weight is coupled with the initial weight calculated by the cloud model, and the index weight is calculated by the combination evaluation method; step 4, the index membership degree is calculated by the conditional cloud generator, and the evaluation level is judged by the confidence criterion evaluation method to obtain the evaluation value of each index and Comprehensive evaluation level. The invention provides a basis for engineering management.

The invention relates to a multilevel index projection pursuit dynamic clustering method and device. The method includes the steps: building a multilevel evaluation index system according to difference of domains and targets; determining the weight of indexes in the multilevel evaluation index system; processing noise in real-time tuple data; building a projection pursuit clustering model based on the processed real-time tuple data and the weight of the indexes; realizing projection pursuit dynamic clustering for the processed real-time tuple data based on a genetic algorithm and the projection pursuit clustering model. The data can be rapidly, efficiently, objectively and dynamically clustered.

The invention relates to the technical field of power grid management, in particular to an electricity quality monitoring and managing method based on genetic projection pursuit evaluation. The electricity quality monitoring and managing method based on genetic projection pursuit evaluation is characterized by including the following steps: (1) a projection index is established; (2) a projection objective function is established; (3) genetic calculation optimization is performed on the projection objective function; (4) a scatter diagram is formed and end point values are determined. Feature vectors corresponding to maximum feature values of matrixes are judged through comparison. The genetic projection pursuit calculation method overcomes affection of artificial subjective factors in a conventional method, objectively reflects internal relations between evaluation indexes and is capable of evaluating voltage of different stages from different levels. The eigenvalue weighting method of the electricity quality monitoring and managing method based on genetic projection pursuit evaluation establishes comparison judgment matrixes through evaluation results obtained by the genetic projection pursuit calculation method and automatically judges relative importance of evaluated subjects. Thus, a determined weight function is more objective.

The invention relates to a data processing method in the technical field of precision processing technologies, and specifically relates to a machine tool thermal error modeling method based on a multi-element projection pursuit clustering algorithm. The method comprises the steps of: (1) carrying out a detection experiment of a machine tool temperature field and thermal deformation; (2) establishing a multi-element projection pursuit clustering model and carrying out parameter calculation; (3) designing and developing a thermal error compensation system; and (4) testing prediction performance of the model. The method provided by the invention has the advantages of high prediction precision, simple calculation and high robustness. The thermal deformation of a main shaft of the machine tool is effectively controlled, and the machining precision of the machine tool is further improved.

The invention belongs to the technical field of radar radiation source signal characteristic evaluation in electronic countermeasure, and discloses a radar radiation source signal intra-pulse characteristic comprehensive evaluation method and system. The method comprises the following steps of firstly, carrying out feature extraction on a received radar radiation source signal, and carrying out feature evaluation index measurement and normalization according to an established evaluation system; carrying out improved interval analytic hierarchy process by combining expert priori knowledge and an actual environment, and establishing a nonlinear equation optimization model by using an improved projection pursuit algorithm; and finally, performing final subjective and objective decision fusionby using a projection spectrum gradient algorithm. The radar radiation source signal intra-pulse characteristic evaluation method and system can reasonably and effectively realize various radar radiation source signal intra-pulse characteristic evaluations based on an actual environment, and scientific and effective evaluation is carried out to help to select characteristics capable of highlighting radar radiation source signals so as to facilitate the subsequent radar radiation source signal sorting and identification.

The present invention provides a projection pursuit and Z-shaped function based assessment method for a marine environmental influence on an offshore military activity. The specific steps are as follows: 1. performing normalization processing on marine environmental factor indexes that influence the security of a military activity; 2. based on a projection pursuit technology, performing dimension reduction on a normalized index sequence; 3. determining a projection vector; 4. establishing a Z-shaped function assessment model of a degree of the marine environment on the security of the military activity and a projection value; 5. establishing a nonlinear optimization model to estimate Z-shaped function parameters; and 6. verifying validity of the new method. According to the method provided by the present invention, a complex nonlinear assessment model is obtained by coupling the projection pursuit technology and the Z-shaped function, the model parameters are all determined by observed data and have relatively strong objectiveness, so that the precision of environmental influence assessment of offshore military activity security is improved. the new model has no requirements on the data volume, and has wide adaptability and practicality.

The invention discloses an optimal scheme selection method based on fuzzy clustering iteration and projection pursuit in the technical field of data clustering mining. The specific steps of the methodare as follows: S1, moderately evaluating each individual in a sample set to find the optimal object; S2: parallel evolution of two spaces; 3, after that main body group space and the belief space are mutually fused, if the evolution index reach an integral multiple of the set value, the receiving operation and the influence operation are carried out; 4, stopping that operation when the operationtermination condition is reached, otherwise, the iteration numb is increased by 1, and the operation is continued; 5, selecting that optimal scheme output according to the short distance between sample parts, The invention can avoid the selection problem of the density window width, intelligently sets all parameters in the operation process, ensures the clustering effect through the projection pursuit clustering and the fuzzy clustering iteration, optimizes the clustering center and the projection direction, improves the optimization speed of the data key vector and the selection time of theoptimal scheme.

The invention discloses an evaluation index system and an evaluation method for comprehensive benefits of water-saving reform of an irrigation district. The method is characterized by comprising the following steps: step a, constructing the evaluation index system by determining five first-grade indexes capable of reflecting the comprehensive benefits of water-saving reform of the irrigation district as a criterion layer and setting twelve second-grade indexes on the basis of the current technical standards and theoretical research results according to an actual project for water-saving reformoptimization of a water transport and distribution system of the irrigation district; step b, determining evaluation standard for the value range of each evolution index according to the evaluation indexes; step c, evaluating the comprehensive benefits of water-saving reform of the irrigation district with a projection pursuit method. The evaluation index system and the evaluation method can evaluate the comprehensive benefits of water-saving reform of the irrigation district comprehensively and wholly, so that the comprehensive benefit level of water-saving reform of certain irrigation district can be reflected objectively and accurately.

The invention discloses a comprehensive performance multi-dimensional evaluation method, which is applied to a rail transit power supply system of a bilateral power supply structure, and comprises thefollowing steps of: establishing multi-dimensional comprehensive evaluation indexes of a system power quality dimension, a system application economy dimension and a system operation stability dimension; adopting different evaluation algorithms for different system evaluation dimensions, specifically, adopting a projection pursuit method based on a genetic algorithm for evaluating the system comprehensive power quality, adopting an approximate ideal solution sorting method based on an application variable coefficient method for evaluating the system application economical efficiency, and adopting an approximate ideal solution sorting method based on a mean square error method for evaluating the system operation stability; and allocating weights to all dimension index evaluation results according to system operation scene requirements to obtain a final comprehensive performance multi-dimensional evaluation result of the rail transit power supply system for the bilateral power supply structure. The comprehensive performance multi-dimensional evaluation method remarkably improves the correctness, rationality, scientificity, effectiveness and comprehensiveness of the evaluation method.