大连理工大学学报
大連理工大學學報
대련리공대학학보
JOURNAL OF DALIAN UNIVERSITY OF TECHNOLOGY
2014年
2期
246-250
,共5页
工业以太网%网络延时预测%延时补偿%小波神经网络
工業以太網%網絡延時預測%延時補償%小波神經網絡
공업이태망%망락연시예측%연시보상%소파신경망락
industrial ethernet%network delay prediction%delay compensation%wavelet neural network
随着网络控制研究的兴起,对工业以太网延时进行补偿成为研究的重点方向。针对网络延时给网络控制系统带来的问题,提出用小波神经网络对工业以太网延时进行预测,根据输入的过去时间延迟序列预测输出下一采样时刻的网络延时值。预测模型的参数通过训练算法实时更新,以保证预测输出的准确性。对实际工业以太网延时数据样本的预测分析表明,该预测模型能够有效预测延时。为进一步说明延时预测效果,将延时预测模型应用于网络控制系统进行延时的预测与补偿,系统仿真结果证明了预测模型预测的准确性及补偿的有效性。
隨著網絡控製研究的興起,對工業以太網延時進行補償成為研究的重點方嚮。針對網絡延時給網絡控製繫統帶來的問題,提齣用小波神經網絡對工業以太網延時進行預測,根據輸入的過去時間延遲序列預測輸齣下一採樣時刻的網絡延時值。預測模型的參數通過訓練算法實時更新,以保證預測輸齣的準確性。對實際工業以太網延時數據樣本的預測分析錶明,該預測模型能夠有效預測延時。為進一步說明延時預測效果,將延時預測模型應用于網絡控製繫統進行延時的預測與補償,繫統倣真結果證明瞭預測模型預測的準確性及補償的有效性。
수착망락공제연구적흥기,대공업이태망연시진행보상성위연구적중점방향。침대망락연시급망락공제계통대래적문제,제출용소파신경망락대공업이태망연시진행예측,근거수입적과거시간연지서렬예측수출하일채양시각적망락연시치。예측모형적삼수통과훈련산법실시경신,이보증예측수출적준학성。대실제공업이태망연시수거양본적예측분석표명,해예측모형능구유효예측연시。위진일보설명연시예측효과,장연시예측모형응용우망락공제계통진행연시적예측여보상,계통방진결과증명료예측모형예측적준학성급보상적유효성。
As the rising of network control study,industrial ethernet delay compensation has become an important research direction.In order to solve problems caused by network delay in network control system,an algorithm is proposed to predict industrial ethernet delay using wavelet neural network.The controller applying this algorithm can output the predicted time delay value of the next sampling by processing inputted past delay sequence,and the prediction model parameters are updated in real time through the training algorithm to ensure the accuracy of the prediction output. Experiments using the real delay data of industrial ethernet are carried out to test the performance of the algorithm.The experimental results indicate that the prediction model can predict the delay effectively.To further verify the effectiveness of delay prediction,simulation experiments have been operated in which the delay prediction model has been applied to the network control system for delay prediction and compensation.The simulation results demonstrate the prediction accuracy of prediction model and the effectiveness of compensation.