智能复杂体系研究

 介绍了体系研究的历程与现状，提出当下该领域的研究重点是复杂度及智能性更高的智能复杂体系。阐述了智能复杂体系的概念、特征、建模、涌现和应用。提出今后的研究方向主要集中在智能复杂体系的建模、智能涌现与牵引控制等。     

Implementation of Direct Torque Control Scheme for Induction Machines with Variable Structure Controllers

A torque control scheme for high-performance induction machine drives was developed to overcome some disadvantages of direct torque control （DTC）. In the improved DTC method, the stator flux and the torque controllers use variable-structure control theory which does not require information about the rotor speed. Space vector modulation is applied to the voltage source inverter to reduce the torque, stator flux, and current ripples. The digital signal processor-based implementation is described in detail. The experimental results show that the system has good torque and stator flux response with small ripples.     

Soft Sensing of Overflow Particle Size Distributions in Hydrocyclones Using a Combined Method

Precise, real-time measurements of overflow particle size distributions in hydrocyclones are necessary for accurate control of the comminution circuits. Soft sensing measurements provide real-time, flexible, and low-cost measurements appropriate for the overflow particle size distributions in hydrocyclones. Three soft sensing methods were investigated for measuring the overflow particle size distributions in hydrocyclones. Simulations show that these methods have various advantages and disadvantages. Optimal Bayesian estimation fusion was then used to combine three methods with the fusion parameters determined according to the performance of each method with validation samples. The combined method compensates for the disadvantages of each method for more precise measurements. Simulations using real operating data show that the absolute root mean square measurement error of the combined method was always about 2% and the method provides the necessary accuracy for beneficiation plants.     

面向大规模数据集的单类支持向量机

提出了一种面向大规模数据集的单类支持向量机(OCSVM)方法.该方法基于k近邻思想得到表征数据集合分布特征的集合内点,并依此生成集合边缘点,而后由二者重新组成数据集合,用于OCSVM训练.该新建数据集不仅极大压缩了原有大规模数据集的样本数量,还可以保留原有大规模数据集的分布特征,从而有效解决了OCSVM在处理大规模数据集时所存在的训练时间长、模型复杂以及预测效率低等问题.最后,通过在典型数据集合上进行的对比实验,表明了所提方法的有效性.     

基于大数据的复杂生产线系统健康度实时评估方法

复杂生产线系统由于设备规模大、影响因素多,导致运行过程存在着较强的不确定性。系统部署的传感器网络产生的实时大数据(流数据)可以作为系统状态监测的来源,然而传统数据处理方法并不能很好地对系统的健康度作出实时评价。该文以复杂生产线系统运行中的实时大数据(流数据)为基础,基于信息熵原理通过大数据分析方法量化分析系统内部属性间的行为模式和相关性关系,提出了一种基于大数据的复杂生产线系统健康度实时评估方法。磨矿生产线案例表明,该方法可以对复杂生产线的系统健康度作出实时评估。     

智能复杂体系研究

 介绍了体系研究的历程与现状，提出当下该领域的研究重点是复杂度及智能性更高的智能复杂体系。阐述了智能复杂体系的概念、特征、建模、涌现和应用。提出今后的研究方向主要集中在智能复杂体系的建模、智能涌现与牵引控制等。     

激光制导炸弹的神经网络鲁棒控制

针对激光制导炸弹滚转控制通道的时变特性,基于神经网络具有定性和定量多模态控制能力能够实现多个常规控制器的功能,融合炸弹投放过程中的多种工作状态参数信息,设计了非线性神经网络控制器,给出神经网络控制器与常规控制器的功能等价性分析。该控制器具有鲁棒性,能适应时变系统参数大范围的变化,而且方法简单,实现容易。利用该方法对某型激光制导航空炸弹进行仿真,并与炸弹的变结构控制器相比,从根本上解决了变结构控制器的抖振问题,结果表明,该神经网络控制器具有良好的控制性能。