Using Transfer-Entropy TOPSIS and BP Neural Networks Model to early Warning of External Environments of Life Cycle Asset Management

Abstract

The Life Cycle Asset Management (LCAM) was implemented by the development of smart grid promoted Grid Corporation to change the style of asset management. The external environment of LCAM is a development and changing system which had some characteristic with multi-agent, multi-level and multi-dimensional structure. Therefore, it is imperative requirement of implementing LCAM to .correctly understand and grasp the changing trends of external environment. The purpose of this paper is to propose an external environment early-warning model with LCAM to address the problem with how to grasp the change trends of external environment. This paper first calculated the weight of external environments by TOPSIS based transfer-entropy with survey data. And some significant external environment factors would be selected to be as the input vector of BP Neural Networks. Then the BP Neural Networks model was employed to early warning the situation of external environment. The results showed the superiority of the above two approaches in external environment early warning.