欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对一类难以精确建立数学模型的非线性控制系统,提出了协同随机微粒群优化CSPSO的神经网络预测建模方法。CSPSO在协同微粒群算法CPSO执行之后引入随机微粒群优化SPSO的思想,促使CPSO摆脱了伪最小值现象,并且保证其以概率1收敛于全局最优值。通过采集对象输入/输出数据,将CSPSO应用到模型权值的离线训练中,并给出了实现的具体步骤。结果表明在实验的几种算法中,CSPSO训练的神经网络模型精度较高且算法学习的稳定性较佳。
英文摘要：
To study a kind of nonlinear system which is difficult to describe with a precise mathematical model, a neural network prediction modeling method based on cooperative stochastic particle swarm optimization （CSPSO） is proposed. CSPSO carries out the idea of SPSO（Stochastic PSO） after CPSO（Cooperative PSO） is performed, so it promotes CPSO escaping from pseudo-minimums and converging at the global optimum with probability value of 1. The algorithm is employed to the offline training of model weights by sampling the input/ output data of the object, and the realization details are provided also. The results show that the neural network model trained by CSPSO holds a higher precision and the learning stability of the algorithm is also better than the other algorithms experimented with.
同期刊论文项目
同项目期刊论文
Adaptive robust predictive control for hypersonic Vehicles using recurrent functional link artificia
Adaptive recurrent-functional-link-network control for hypersonic vehicles with atmospheric disturba
Robust adaptive backstepping control for a class of uncertain nonlinear systems based on disturbance
位置：