Coal mine gas detection method
The invention discloses a gas concentration detection method. A KPCA algorithm is used for identifying 'large numbers'. The method comprises the steps that firstly, two hybrid kernel functions are structured, a kernel matrix is structured with a vector method, and feature vectors of the ke...
Saved in:
Main Author | |
---|---|
Format | Patent |
Language | Chinese English |
Published |
21.12.2016
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a gas concentration detection method. A KPCA algorithm is used for identifying 'large numbers'. The method comprises the steps that firstly, two hybrid kernel functions are structured, a kernel matrix is structured with a vector method, and feature vectors of the kernel matrix are calculated through main kernel ingredient analysis. The algorithm has a high recognition rate and a high operation speed; according to the algorithm, through a set of orthonormal bases of sub-spaces spanned by the training samples in a feature space, the KPCA process on a training set is converted into the PCA process that coordinates of all kernel training samples under the bases are a data set, the features of training samples are extracted at the same time, nonlinear features of training data can be effectively captured, and the algorithm is widely valued and applied in mode recognition and regression analysis. In the KPCA solving process, feature values are needed for decomposing the M*M kernel matrix (M |
---|---|
Bibliography: | Application Number: CN20161390517 |