Multilayer Process Goose Queue (PGQ) formation adjustment approaches based on model-free adaptive control strategies

The Process Goose Queue (PGQ) approach has been proposed to solve decomposition-coordination optimization problems of large-scale process production systems. At present, the existing multilayer PGQ formation adjustment is generally performed with model-based optimization approaches that strongly rely on rigorous models of processes and suffer difficulties in dealing with online model identifications. In this paper, we introduce a novel data-driven control approach for multilayer PGQ formation adjustments using model-free adaptive control (MFAC) strategies. The individual PGQ is formulated as a multi-objective control problem before a model-free controller is designed for each layer of the PGQs by taking advantage of the feed-forward control idea and the inter-level coordination matrix. This approach enjoys effectively restraining the vibration propagation among PGQs as well as realizing rapid and timely adjustments of the PGQ formation. Case simulations show the effectiveness of the proposed method.