期刊文献+
共找到3篇文章
< 1 >
每页显示 20 50 100
空间点目标识别的模糊神经网络应用研究 被引量:1
1
作者 郑俊生 张继红 《计算机工程与应用》 CSCD 北大核心 2009年第32期244-248,共5页
首先对空间目标辐射特性进行了研究,指出了用空间目标在3个不同波段的辐射通量作为特征向量进行目标识别。然后研究了进化模糊神经网络(EFuNN)和动态进化神经模糊推理系统(DENFIS),最后用EFuNN和DENFIS进行了仿真实验,并且与BP神经网络... 首先对空间目标辐射特性进行了研究,指出了用空间目标在3个不同波段的辐射通量作为特征向量进行目标识别。然后研究了进化模糊神经网络(EFuNN)和动态进化神经模糊推理系统(DENFIS),最后用EFuNN和DENFIS进行了仿真实验,并且与BP神经网络、遗传算法以及遗传-神经算法进行了比较。仿真结果表明EFuNN尤其是DENFIS具有较好的学习能力和泛化能力,较大地提高了目标识别率,能够较好地进行空间点目标的识别。 展开更多
关键词 模糊神经网络 进化模糊神经网络 动态进化神经模糊推理系统
下载PDF
基于动态模糊推理的舒适温度在线预测 被引量:1
2
作者 白燕 冯壮壮 张玮 《计算机测量与控制》 2020年第7期74-80,共7页
基于热感觉预测的室内热环境自动控制方法为解决基于传统温度设定值控制不满足用户舒适度问题提供新的途径;但建模过程中用户热感觉信息难以获取,因此开发了便捷的移动端智能交互系统以实时采集现场数据,建立用户学习样本;并针对热舒适... 基于热感觉预测的室内热环境自动控制方法为解决基于传统温度设定值控制不满足用户舒适度问题提供新的途径;但建模过程中用户热感觉信息难以获取,因此开发了便捷的移动端智能交互系统以实时采集现场数据,建立用户学习样本;并针对热舒适的动态变化性特征,设计动态进化神经模糊推理系统(DENFIS)以建立用户热舒适在线预测模型;通过实时学习样本数据驱动,系统的模糊规则与模型输出函数系数可动态自校正,推理预测出用户偏好温度;实验结果表明所设计的DENFIS算法预测用户的舒适温度范围准确率高达90.5%,误差极小;证明了该算法所建立的在线预测模型用于智能空调温度控制,可解决现有的温度设定方式带来的温度设定值不合理的问题,在实际应用中具有可行性。 展开更多
关键词 热感觉 智能交互系统 动态进化神经模糊推理系统 在线预测模型 偏好温度
下载PDF
Dynamic stability enhancement of interconnected multi-source power systems using hierarchical ANFIS controller-TCSC based on multi-objective PSO 被引量:1
3
作者 Ali Darvish FALEHI Ali MOSALLANEJAD 《Frontiers of Information Technology & Electronic Engineering》 SCIE EI CSCD 2017年第3期394-409,共16页
Suppression of the dynamic oscillations of tie-line power exchanges and frequency in the affected interconnected power systems due to loading-condition changes has been assigned as a prominent duty of automatic genera... Suppression of the dynamic oscillations of tie-line power exchanges and frequency in the affected interconnected power systems due to loading-condition changes has been assigned as a prominent duty of automatic generation control(AGC). To alleviate the system oscillation resulting from such load changes, implementation of flexible AC transmission systems(FACTSs) can be considered as one of the practical and effective solutions. In this paper, a thyristor-controlled series compensator(TCSC), which is one series type of the FACTS family, is used to augment the overall dynamic performance of a multi-area multi-source interconnected power system. To this end, we have used a hierarchical adaptive neuro-fuzzy inference system controller-TCSC(HANFISC-TCSC) to abate the two important issues in multi-area interconnected power systems, i.e., low-frequency oscillations and tie-line power exchange deviations. For this purpose, a multi-objective optimization technique is inevitable. Multi-objective particle swarm optimization(MOPSO) has been chosen for this optimization problem, owing to its high performance in untangling non-linear objectives. The efficiency of the suggested HANFISC-TCSC has been precisely evaluated and compared with that of the conventional MOPSO-TCSC in two different multi-area interconnected power systems, i.e., two-area hydro-thermal-diesel and three-area hydro-thermal power systems. The simulation results obtained from both power systems have transparently certified the high performance of HANFISC-TCSC compared to the conventional MOPSO-TCSC. 展开更多
关键词 Hierarchical adaptive neuro-fuzzy inference system controller(HANFISC) Thyristor-controlled series compensator(TCSC) Automatic generation control(AGC) Multi-objective particle swarm optimization(MOPSO) Power system dynamic stability Interconnected multi-source power systems
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部