Abstract: Based on the gray GM(1,1) model of trend prediction for the nuclear power plant, the Markov chains prediction method was presented and the grey Markov GM(1,1) model for predicting the trend of the nuclear power plant was built. The model combines together two kinds of inherent features, meaning that the model takes account both the feature of the GM(1,1) model for dealing with the numbers of strong monotonous series and the feature of random wave response in extracting data through state transfer probability matrix, which make it more scientific and rigorous. So it opens up applicable scope of grey GM(1,1) model prediction. The example shows that the precision of grey Markov GM(1,1) model for predicting the trend is better, not only dealing with the data of strong monotonous series, but also the data sequence of random response.