Application Of Kalman Filter Based On Colored Noise In Bus Load Forecasting

From the structure and the characteristics of the bus load, it was found that the bus load changed very exquisitely and stochastically. Practically such randomness has obvious characteristics of colored noise, instead of white noise. Most previous methods for the bus load forecast considered the noise as white ones, as a result, the sensitivity and accuracy of the load forecast models inevitably decreased. This paper brings forward a method applying Kalman filter to bus load forecast based on colored noise. It first introduces how to transform Markov colored noises sequence to white ones, and then deduces extended Kalman filter equations that are suitable for the situation where the system noise and the measured noise are both colored noise. In the algorithm the state expansion and the measurement expansion are used. The calculation and emulation prove its validity and feasibility.