Boosting Progressive Optimality Algorithm Performance in Optimizing Complex Large-Scale Multi-Reservoir System Operations by Using Discrepant Optimization Windows and Disturbance-Response Strategy

The Progressive Optimality Algorithm (POA) is a powerful technique widely used for optimizing multi-reservoir operations; however, two crucial downsides cumber its application to complex large-scale multi-reservoir systems, which are insufficient search directions and the dimensionality problem—the former limits the POA’s precision, while the latter reduces its efficiency. Although several POA variants have been developed to overcome these downsides, a further balance between the precision and efficiency of the algorithm is required to boost the POA’s capability of optimizing the operation of complex large-scale multi-reservoir systems. In view of this, we made modifications to the original algorithm and developed a new POA variant, referred to as the Direct Search Algorithm Based on Disturbance-Response Strategy (DRDSA). On one hand, we changed the POA’s uniform optimization window for all reservoirs to discrepant optimization windows for varying reservoirs to enrich the search direction set of the algorithm. On the other hand, we introduced a disturbance-response strategy into the solution of sub-problems to handle the POA’s dimensionality problem. Two multi-reservoir operation optimization problems were employed to test the performance of the DRDSA, and seven advanced alternatives including two existing POA variants were used for comparison. The results showed the improved precision and efficiency of the DRDSA. Thus, a new technique is available for optimizing the operation of complex large-scale multi-reservoir systems.