Abstract: Based on wavelet transformation and chaos theory, this paper presents a condition forecasting method for complex systems.Firstly,using wavelet decomposition theory,the system feature reference data series are decomposed into two parts: low frequency part and high frequency part.Further analysis on the two parts indicates that there exists a chaos feature in the both parts.Then,by using chaos theory,chaotic forecasting models are established to forecast the low frequency and high frequency parts respectively.Finally,forecasting results of the chaotic models are reconstructed based on wavelet theory so as to forecast the system feature reference data series.Case study shows that the proposed method is of high precision,and can be effectively applied to condition forecasting and fault trend forecast analysis of complex systems.