In order to solve the non-linear and high-dimensional optimization problems more effectively, an improved self-adaptive membrane computing(ISMC) optimization algorithm was proposed. The proposed ISMC algorithm applied...In order to solve the non-linear and high-dimensional optimization problems more effectively, an improved self-adaptive membrane computing(ISMC) optimization algorithm was proposed. The proposed ISMC algorithm applied improved self-adaptive crossover and mutation formulae that can provide appropriate crossover operator and mutation operator based on different functions of the objects and the number of iterations. The performance of ISMC was tested by the benchmark functions. The simulation results for residue hydrogenating kinetics model parameter estimation show that the proposed method is superior to the traditional intelligent algorithms in terms of convergence accuracy and stability in solving the complex parameter optimization problems.展开更多
文摘提出了一种未知信源数的高分辨 DOA 估计算法。该算法在未知信源数的情况下,利用线性预测(LP)法或 Pisarenko 法与 ASPECT 技术相结合来实现高分辨谱估计。该算法在消除 LP 法或 Pisarenko 法 DOA 估计中存在的伪峰的同时可判断出入射信源数,明显减小 DOA 估计算法的运算量,并可提高谱分辨力。计算机仿真结果证明了新算法理论的正确性和有效性。
基金Projects(61203020,61403190)supported by the National Natural Science Foundation of ChinaProject(BK20141461)supported by the Jiangsu Province Natural Science Foundation,China
文摘In order to solve the non-linear and high-dimensional optimization problems more effectively, an improved self-adaptive membrane computing(ISMC) optimization algorithm was proposed. The proposed ISMC algorithm applied improved self-adaptive crossover and mutation formulae that can provide appropriate crossover operator and mutation operator based on different functions of the objects and the number of iterations. The performance of ISMC was tested by the benchmark functions. The simulation results for residue hydrogenating kinetics model parameter estimation show that the proposed method is superior to the traditional intelligent algorithms in terms of convergence accuracy and stability in solving the complex parameter optimization problems.
