Deep-Sea Polychaete Optimization Algorithm

To address the common problems of premature convergence, singular individual search behavior, and synchronization of group rhythms in existing swarm intelligence optimization algorithms for high-dimensional continuous optimization problems, this paper proposes a novel bio-inspired optimization algorithm—the Deep-Sea Polychaete Optimization Algorithm (DSPOA). This algorithm uses the peristaltic movement, chemosensory perception, adhesive anchoring, and sparse cooperative behavior of deep-sea polychaetes in extreme marine environments as biological prototypes, and reconstructs the traditional swarm intelligence optimization paradigm from multiple levels, including environment-mediated search, morphologically variable search, path memory mechanisms, and asynchronous evolutionary strategies. Unlike algorithms that rely on direct information interaction between individuals, DSPOA constructs a dynamic chemical field as an environmental information carrier to guide individuals in indirect cooperative search; simultaneously, it introduces a peristaltic morphological switching mechanism based on fitness changes, enabling individuals to automatically adjust their search dynamics at different search stages. Furthermore, the algorithm uses path-level memory rather than single-point memory to characterize the effectiveness of the search trajectory and enhances group diversity and search stability through asynchronous individual update strategies. Theoretical analysis shows that DSPOA has good comprehensive performance in terms of exploration ability, local exploitation ability and resistance to local optima, providing a new modeling idea for bio-inspired optimization algorithms.