Management of algae bloom based on cbr-oss model

The outbreak process of algal bloom is a complex ecological problem of system engineering involving various factors such as water parameters, surrounding environment and human activity. For this ecological problem, the strict restriction and requirement limit the development of management about algae bloom. To select the most suitable strategy from various algae control methods, we propose case-based reasoning-optimal strategy selection (CBR-OSS) model. It builds case library and complex network by extracting the factors of algae management. This model regards the complex network as a directive network to reflect dynamic characteristic and weights of key factors. To improve decision efficiency, it defines the restriction slots and condition slots in directive network. As the inference engine, these slots exclude the unsuitable cases and avoid the redundancy computation so that the model can calculate the similarity between the target water body and screen cases in the process of decision case matcher. This process finds the best matching case and recommended measures by intuitionistic fuzzy rough sets. To verify the model, Kunming Lake and other 20 lakes are simulated with the proposed method. The results accord with expert advice and the model outperforms in accuracy, operation time, expert participation and flexibility.