The paper concerns the problem of optimal control of a natural gas transmission system consisting of a compressor station and adjacent pipeline sections. Natural gas is supplied with two types of compressors, namely gas turbine driven centrifugal compressors and motor-compressors. For a given simulation scenario, the suction pressure, suction temperature, discharge pressure, and total compressor station mass flow are predicted from the non-isothermal transient gas flow model. Next the nonlinear programming problem with continuous and, in case of motor-compressors, discrete variables is solved to evaluate the type and the number of simultaneously operating compressors, while determining such a distribution of the capacity that the total unit fuel consumption in each time interval is minimized subject to the constraints imposed. The paper presents an algorithm of automatic search for the optimal values of the operating parameters of the compressor station. The method presented has been verified experimentally on the telemetry data.
Natural gas is usually transported by pipeline networks which serve as the most cost effective transportation means. Transmission systems usually have a linear topology corresponding to a linear arrangement of compressor stations. The fuel consumption of compressors is responsible for a large fraction of the costs of gas network operation. Luongo et al. (1989) reported that AGA estimates the operating cost of running the compressor stations to vary between 25% and 50% of the total company's operating budget, therefore minimizing fuel usage is a major objective in the control of gas transmission costs.
This work is concerned with the optimization of a single compressor station operated under transient conditions. More specifically, we consider variable boundary conditions, i.e. unsteady inlet and outlet pressures together with a variable flowrate through the compressor, and search for the optimal values of the operating parameters that minimize the running costs of the compressor station.
