Improved K-means algorithm and priori knowledge based intelligent recognition of flotation working c

来源 :第26届中国过程控制会议 | 被引量 : 0次 | 上传用户:sflljlove
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
  On the basis of correlation between four typical bubble status and different ore grades,a new recognition method based on the improved k-means algorithm and integrating priori knowledge is proposed for working condition recognition of antimony flotation.In offline classification stage,firstly extract the bubble feature of images under different bubble status to obtain the datasheet.Then combine with the prior knowledge(number of bubble status),cluster the datasheet of bubble feature with the improved k-means algorithm.At last,classify the working condition and their causes under different bubble status and different ore grade..For the online recognition,firstly determine current bubble status with the clustering centers which obtained in improved k-means algorithm and the k-nearest neighbor algorithm.Then recognize current working condition by the working condition classification and the current ore grade.At last,verify the validity of the recognition in flotation process control by analyzing the concentrate grade under different bubble status.
其他文献
The inverse relationship between train mechanical energy consumption and trip time is proved by infinitesimal increment analysis of train speed profile.Firstly,a general optimal model for train drivin
Moving window local outlier probability(MWLoOP)is an outlier detecting method which was proposed for monitoring time-varying industrial processes; however,for the practical industrial processes,beside
Operating at different operation steps,process variable correlations of batch processes present the typical multiphase and transition characteristics.The existing phase partition algorithms have sever
Computation time is the main reason that limits the application of the model predictive control(MPC).In this paper,we propose a strategy combining the offline method and the online optimization to sol
As a classic multivariate statistical method,principal component analysis(PCA)has been widely used in monitoring industrial processes,but it is still necessary to make improvements in having a timely
Complex industrial processes of operational control can be seen as two layers,namely the loop control layer in the bottom and the operational layer on the top.The former contains of loop control for e
An algorithm of data selection kernel least squares regression(DS-KLSR)for fault isolation are proposed in this paper.KLSR can increase the distance be tween two different kinds of faults.To deal with
This paper focuses onfinite-time stability problemfor a class of bilinear system with markov jump parameters.By saturation-dependent Lyapunov function,a new criterion for testing the finite-time stabi
For a wide class of time-delay and parameter uncertainty systems,it is quite often to see that prediction models are not effective and uncertain information cannot be accurately described.In this pape
This paper is concerned with identification and nonlinear predictive control method forMIMO Hammerstein systems with constraints.Firstly,an identification method based on steady-state responses and Su