Application of adaptive time delay model in optimal control of a hydropower cascade

Small hydropower plants (SHP) are increasingly constructed in recent years as a substitution of the use of conventional energy materials such as wood and fossil fuel in remote rural regions. Besides new constructed plants, upgrading operational strategies of existed systems for increasing electricity productivity is also significant. Motivated by this issue, a combination of two techniques, simulation and optimization based on models is frequently used to improve the operation regimes of coordinated reservoir cascades. However, the application of a complex hydraulic model consumes a huge computation time. Hence, this paper proposes a replacement for the complex hydraulic model by an adaptive time delay (ATD) model. The cutting- edge point is that the ATD model is able to quickly predict the system dynamics both in simulation and optimization. This ATD model consists of only two parameters: time constant and time delay which are functions of unsteady flow and can be easily derived from complex hydraulic models (HECRAS, MIKE11), or from physical parameters of rivers (flow rate, roughness, bed slope, cross section). The integration of the ATD model into the simulation and optimization techniques will be demonstrated by a case study of a cascade of SHPs. In terms of optimization, a non-linear constrainted optimization algorithm is applied to improve electricity production to meet the scheduled demand.