PROJECT TITLE :

Hidden Behavior Prediction of Complex Systems Under Testing Influence Based on Semiquantitative Information and Belief Rule Base

ABSTRACT:

Compared with the observable behavior, it's tough to predict the hidden behavior of a complicated system. In the existing strategies for predicting the hidden behavior, a lot of testing data (usually quantitative info) are required to be sampled. But, some advanced engineering systems have the subsequent characteristics: one) The systems cannot be tested periodically, and the observable information is incomplete; two) the change process of hidden behavior may be stricken by the take a look at; and three) solely half of quantitative information and qualitative knowledge (i.e., semiquantitative data) might be obtained. These characteristics all connected to the take a look at are named as testing influence for simplicity. Though a model and a corresponding optimal algorithm for training the model parameters are proposed to predict the hidden behavior on the idea of semiquantitative data and belief rule base (BRB), the testing influence has not been considered. In order to solve the above problems, a new BRB-based mostly model, which will use the semiquantitative information, is proposed below testing influence in this paper. In the newly proposed forecasting model, there are some parameters of that the initial values are sometimes assigned by experts and may not be accurate, which can cause the inaccurate prediction results. As such, an improved optimal algorithm for training the parameters of the forecasting model is any developed on the basis of the expectation-maximization idea and also the covariance matrix adaption evolution strategy (CMA-ES). By using the semiquantitative info, the proposed BRB-based mostly model and also the improved CMA-ES algorithm can operate together in an integrated manner so as to boost the forecasting precision. A case study is examined to demonstrate the power and applicability of the newly proposed BRB-primarily based forecasting model and therefore the improved CMA-ES algorithm.