Bounded consensus for multiagent systems by event‐triggered data transmission,time delay,and predictor‐based control

This paper investigates the consensus issue of multiagent systems with data transmission time delay. The state measurement of each local agent is directly sent to a private event‐trigger and further authorized to be broadcasted to its neighbors via communication network only when the threshold of the event‐trigger is violated. Since the controller always receives discrete‐time neighbor information with data transmission time delay, a predictor is employed to estimate the continuous‐time neighbor state. Based on the estimated state, a novel consensus protocol is mainly proposed for achieving the bounded consensus of the multiagent systems. By the proposed method, the asynchronous neighbor information is allowed and the margin of data transmission time delay is also given. Furthermore, it has been proved that the unwanted Zeno phenomena can be naturally excluded. Numerical example is provided to demonstrate the effectiveness of the proposed method.     

基于鲁棒自适应Smith预估器的电阻炉温控制

针对炉温设定值变化等原因引起的对象模型参数变化,提出一种鲁棒自适应smith预估器。鲁棒项的引入,提高了smith预估器自适应律对纯迟延参数精度的鲁棒性,在纯迟延时间失配的情况下,自适应算法仍能保证预估误差快速收敛。对预估器和电阻炉的仿真结果表明,在纯迟延时间失配的情况下,鲁棒自适应smith预估器能够保证预估误差收敛于零,进而保证系统的控制性能。     

Robust tuning of PFC (Predictive Functional Control) based on first- and aperiodic second-order plus time delay models

Robust tuning is presented for PFC (Predictive Functional Controller) for stable non-integrative processes modelled as FOPDT and aperiodic SOPDT (First- and Second-Order Plus Dead Time) models, assuming stepwise changes in the reference or disturbance signal. The robustness is improved by using a first-order filter. The filter time constant is designed in the knowledge of the uncertainty of the model parameters (static gain, time constants and time delay). Both simulation results and the control of a laboratory pilot plant are presented.     

基于Fuzzy自整定PID参数的飞机除冰车温度控制

针对新型飞机除冰车加热系统的大迟延特性和模型不确定性,设计了Fuzzy自整定PID参数的Smith预估飞机除冰车温度控制系统.根据实验数据及其工作原理建立了加热系统数学模型,并用Matlab进行仿真,仿真结果表明所设计的控制系统在稳定性、准确性,鲁棒性方面明显优于该除冰车现有的Smith预估控制系统.     

基于量测一致性的联邦滤波故障检测方法研究

针对联邦滤波器的信息融合特点,提出了一种基于测量一致性的传感器故障检验方法.该方法利用融合状态递推器对各个子传感器的量测预报与各传感器量测之间的偏差构造统计量,进行故障检测,可以实现先检测后滤波,在子滤波器启动之前确定对应传感器量测的有效性.仿真结果表明,该方法具有较高的故障检测灵敏度.     

一种用于高速公路的快速有效的车道线识别算法

提出了一种针对高速公路的车道线检测与跟踪方法;在图像预处理中采用基于采样的自适应阈值以满足不同光照条件下的使用要求,采用霍夫变换(HT,Hough Transform)进行车道线初始检测,车道线跟踪利用Kalman预测参数动态建立感兴趣区域(ROI,Region of Interest),用扫描线法搜索车道线边界点,在车道线间断区域利用Kalman预测器定位车道线边界;设计了一个失效判别模块,验证跟踪结果,当跟踪失败时,重新启动初始检测算法进行识别;实验结果表明,对于不同的车道线种类和在大部分车道线被前方车辆遮挡的条件下,该算法均具有较高的实时性和鲁棒性。     

具有离散参数齐次随机场的线性预测(IV)

本文提出和研究了4个具有离散参数齐次随机场的线性预测问题;预测问题12,13,14,15。并 求出了它们的预测值与预测误差。     

Smith Prediction Monitor AGC System Based on Fuzzy Self-tuning PID Control

 In accordance with the feature of pure delay in monitor AGC system for cold rolling mill, a new fuzzy self-tuning PID Smith prediction controller is developed. The position control model is deduced based on a single stand cold rolling mill, and the fuzzy controller for monitor AGC system is designed. The analysis in dynamic performance between traditional PID Smith prediction controller and fuzzy self-tuning PID Smith prediction controller is done by MATLAB toolbox. The simulation results show that fuzzy self-tuning PID Smith controller has stronger robustness, faster response and higher static accuracy than traditional PID Smith controller.     

基于Adaboost算法和BP神经网络的税收预测

针对传统税收预测模型精度较低的问题,提出一种将Adaboost算法和BP神经网络相结合进行税收预测的方法。该方法首先对历年税收数据进行预处理并初始化测试数据分布权值;然后初始化BP神经网络权值和阈值,并将BP神经网络作为弱预测器对税收数据进行反复训练和调整权值;最后使用Adaboost算法将得到的多个BP神经网络弱预测器组成新的强预测器并进行预测。通过对我国1990—2010年税收数据进行仿真实验,结果表明该方法相比传统BP网络预测,平均误差相对值从0.50%减少到0.18%,有效地降低了单个BP陷入局部极小的影响,提高了网络预测精度。