PCRDE

Process Control Differential Evolution Optimization Algorithm

The proposed approach dynamically partitions the evolutionary process into distinct stages based on fitness progression, with each stage employing adaptively selected mutation and crossover strategies. Furthermore, a population stagnation detector is introduced to monitor evolutionary activity; its output is integrated with the current evolutionary stage to guide strategic adjustments. To enhance fine-grained control, smooth quantification factors derived from the exponential distribution function are designed for mutation and crossover operations. Randomized suppression or amplification mechanisms are incorporated to improve the algorithm’s ability to escape local optima.