Optimization and control of wastewater treatment using a multi-strategy improved red-billed blue magpie optimizer and...

Amid the challenge of balancing energy consumption and effluent quality in wastewater treatment, this study proposes a dynamic optimization control method integrating a Multi-strategy Improved Red-billed Blue Magpie Optimizer (MIRBMO) with a CNN-BiLSTM-Attention neural network to optimize energy consumption (EC) and effluent quality (EQ). The CNN-BiLSTM-Attention neural network models the EQ-EC relationship. MIRBMO, enhanced by chaotic initialization, spiral search, and Cauchy mutation, demonstrates superior performance on multi-objective benchmark functions with 43-98% average IGD reduction and 15-35% average HV increase on the ZDT test suite, and comparative experiments against seven state-of-the-art metaheuristic algorithms (NSGA-II,SBOA, GKSO, BWO, COA, ARO, and AVOA) confirm its overall superiority across all performance metrics. An intelligent decision system selects optimal solutions, tracked by a PI controller. Evaluated on the Benchmark Simulation Model No. 1 (BSM1), the MIRBMO-CNN-BiLSTM-Attention-PI method achieves a 5.56% reduction in energy consumption while ensuring effluent quality compliance, significantly lowering costs and supporting carbon neutrality in wastewater treatment.