Abstract:

Based on the adjustment regulation of endocrine hormone, a novel algorithm, called the hormone adjustment based quantum-inspired immune clone algorithm (HAQICA), is proposed to improve the accuracy and stability of quantum-inspired immune clone algorithm (QICA) on global optimization. In HAQICA, the clone size is calculated according to the individual fitness of the current generation and the average fitness of the previous generation, and is adjusted adaptively in terms of the population diversity and the rise law of Hill function that is the basic model of endocrine networks. HAQICA also increases the clone number of better individuals and decreases the clone number of worse individuals. Standard test functions are used to verify the algorithm, and the results of 50 random independent experiments show that the convergence speed of HAQICA is comparative with that of QICA and HAQICA is more efficient in global optimization according to the mean and variance values of optimal solutions.