Optimal Design Of Uncertain Steam Power System Based On Stochastic Programming

Steam power systems contribute importantly to the process industry by consuming large amounts of fuel for steam and power generation. There are uncertain factors in the utility system design and operation, such as steam power demand variation due to process changes and market demand fluctuation, and electricity price changing with seasons, etc. Uncertainties cause the system operation deviating from the optimal design state. Conventional design strategies normally are conservative designs or process without uncertainty included. The optimization study under uncertainty has important theoretical and practical significance in steam power system design. This paper proposes stochastic programming strategy in the optimization design of the uncertain steam power system.The characteristics of uncertainty contained in steam power system design are analyzed. Uncertain factors are classified into two categories based on their effect on the optimization objective and constraints:time-based uncertain factors and probability-based uncertain factors. In the optimization model, the predictable time-based factors are formulated as multi-period model. Probability-based uncertainties are expressed by certain probabilities, including fluctuating process steam and power demands due to the fluctuation of operating temperature, pressure, and other operating parameters. Stochastic programming with recourse is utilized to formulate the model accounting for these uncertainties. A methodology of multi-period stochastic programming with recourse (MP-SPR) has been proposed to determine system configuration and equipment operation conditions. A mixed-integer linear programming model is formulated with recourse item in the objective to determine both system configuration and operation scheduling specified to account for uncertain steam power demands and electricity price fluctuation in the design. Moreover, the waste emissions are taken into account in the optimization to achieve the trade-off between the system economic cost and environmental impact.Case studies illustrate that:the proposed MP-SPR method is an effective tool to determine the system configuration, individual equipment size, operating load and the system operating condition. The design contains operating scheduling including both compensation adjustments and penalties for the shortfall of steam power generation under uncertainty, to achieve the system configuration and operation optimization simultaneously. The case study can achieve the reduction of the annual operating costs by 9.51% and the investment costs by 4.57%. Compared with the conventional deterministic method, the proposed design enables the total cost reducing of 8.72%.The system performance is affected by the superheated steam in terms of steam distribution and power generation as well. The equipment sizes and steam flow rate for process heating would be affected in the system design. When changing the high pressure steam superheat, the capital cost the total cost is reduced and after increased with the superheat increasing, in an appropriate superheat range.This paper also analyzes the effect of uncertain factors and environmental factors on the system optimization.