Study on the stochastic dynamic programming method based on the expected benefit

According to mathematical statistic´s mathematic expectation theory, the stochastic dynamic programming method based on the expected benefit is established for the reservoir power plant. The model´s objective function, solution ideas and process are introduced in detail. The calculation results of example indicate that the expected benefit of the power plant increases by 16.4% more than the design benefit, 34.5% more than the actual benefit. The effect of this method is very remarkable, and it can provide reference for the built reservoirs.