Insulation state prediction of cable based on grey Markov error back stepping model
                                                        ZHANG Ran, XIN Yan li, YU Ze yuan, TANG Wen hu
                   （School of Electric Power Engineering, South China University of Technology, Guangzhou 510640, China）

Abstract: Aiming at the main drawbacks of solely using the grey model or the Markov theory for state prediction with large demand for the amount of data and low prediction accuracies, an accurate approach based on the grey Markov error back stepping model to predict insulation states were proposed, which also employs the fuzzy logic and the evidence reasoning algorithm to derive cable insulation states. A grey model was developed to obtain preliminary state prediction based on online monitoring data, and relative errors were calculated between the preliminary prediction and the online monitoring data. Based on the calculated errors, the proposed grey Markov error back stepping model was employed to calculate relevant errors for future states. Accurate state prediction was obtained using the error back stepping model, and accurate insulation states of cable were evaluated by employing the fuzzy logic and the evidence reasoning algorithm. The validity of the proposed approach was verified using a set of insulation data from a 110 kV XPLEcable. The proposed model were compared with conventional methods and the results indicate that the normal demand for the amount of data and the accuracy of prediction is improved.

Key words: state prediction; insulation state; grey model; markov theory; evidence reasoning; uncertainty