Abstract
In order to improve the prediction accuracy of the relevance vector machine model, an improved method for equipment condition prediction is proposed. First of all, an improved kernel function of variance Gauss kernel (VGKF) is constructed to improve the global performance and generalization ability of the kernel function. Then, by using the method of selecting the number of adjacent points in the chaotic sequence local prediction method, the H-Q criterion was used to optimize the embedding dimension of the training space to avoid the blindness of subjective selection. Through the prediction example of terminal guidance radar equipment test parameters, the effectiveness and superiority of the improved RVM were verified.
Highlights
State prediction technology is the key technology of condition-based maintenance strategy
The research of equipment condition prediction mainly focuses on grey theory, artificial neural network, cloud theory and correlation vector machine[2,3,4,5], but the equipment history test data usually has the characteristics of small sample, nonlinear and dynamic uncertainty, and the traditional theory and method have some limitations
In order to improve the prediction accuracy of equipment status, the RVM model is improved in this paper
Summary
State prediction technology is the key technology of condition-based maintenance strategy. The principle of the regression model is as follows: Let be a training dataset with the sample number of N, xi RD is D dimension input vector, ti R is output scalar. Assuming that they are all independently distributed, the relationship between the two is represented as follows: ti y(xi , ) i (1). In order to facilitate the expression and introduce a super parameter, the likelihood function of the entire training sample data set is represented as:. The training of sample data using RVM is to obtain the posterior distribution of the weight parameter vector w.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
