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：
提出一种Bootstrap—自适应混沌克隆网络建立陀螺漂移预测模型.基于统计理论的Bootstrap方法,设计了自适应混沌克隆网络训练策略,以获得统计意义下陀螺漂移时间序列小样本预测模型的最优估计;运用字符串技术生成复合函数抗体,使预测模型具有在复杂函数子空间逼近强非线性函数的能力;构造了一系列自适应混沌克隆网络进化算子,使复合函数抗体在具有混沌特征的进化中自适应性更强;预测建模实验表明：Bootstrap—自适应混沌克隆网络建立的预测模型训练和测试精度稳定,对比几种典型预测算法,其建模方式更灵活,自适应性更强,适于工程应用.
英文摘要：
A Bootstrap—adaptive chaos clone network is put forward for drift forecasting modeling of gyro.Drift times series of gyro has little samples,in order to obtain statistic optimization estimate of the forecasting model, training strategy of adaptive chaos clone network is designed based on Bootstrap method.String techniques are used to produce compound function antibodies,ability of the forecasting model which approach strong nonlinear function in complex function subspace is high.A series evolve operator in adaptive chaos clone network is structured, adaptability of antibodies improve in evolution with chaos character. Forecasting modeling experiments show that forecasting model built by Bootslrap——adaptive chaos clone network is of smooth training and test precision, compared with several typical forecasting algorithm, the modeling approach is flexible, forecasting model has well adaptability and can meet the requirements of engineering application.
同期刊论文项目
同项目期刊论文
Analysis of the smallest detectable leakage flow rate of negative pressure wave based leak detection
Signed directed graph modeling and validation of industrial processes by process knowledge and proce
Cooperative Predictive Maintenance of Repairable Systems with Dependent Failure Modes and Resource C
System Reliability Prediction Model Based on Evidential Reasoning Algorithm with Nonlinear Optimizat
Wang. A new fault detection scheme for networked control systems subject to uncertain time varying d
期刊信息
位置：