基于泡沫尺寸随机分布的铜粗选药剂量控制

为了稳定铜粗选选矿指标,提高矿产资源的利用水平, 根据铜粗选过程中泡沫尺寸分布随药剂量改变而动态变化的特点, 提出一种基于泡沫尺寸随机分布的铜粗选过程药剂量控制方法.首先, 针对泡沫尺寸分布具有非高斯统计特性, 基于方差和均值的统计参量难以表征该分布形态变化的问题, 提出了B样条估计方法以描述泡沫尺寸的概率密度函数(Probability density function, PDF); 然后, 针对B 样条权值相互关联的特点, 建立多输出最小二乘支持向量机模型(Multi-output least square support vector machine, MLS-SVM)以表征权值和药剂量的动态关系; 最后, 为减少系统的随机性, 采用基于熵的优化算法以确定药剂量, 实现对给定泡沫尺寸分布的跟踪控制.工业数据仿真验证了所提方法的有效性, 能有效稳定铜粗浮选的生产指标.     

Probability density function of bubble size based reagent dosage predictive control for copper roughing flotation

As an effective measurement indicator of bubble stability, bubble size structure is believed to be closely related to flotation performance in copper roughing flotation. Moreover, reagent dosage has a very important influence on bubble size structure. In this paper, a novel reagent dosage predictive control method based on probability density function (PDF) of bubble size is proposed to implement the indices of roughing circuit. Firstly, the froth images captured in the copper roughing are segmented by using a two-pass watershed algorithm. In order to characterize bubble size structure with non-Gaussian feature, an entropy based B-spline estimator is hence investigated to depict the PDF of the bubble size. Since the weights of B-spline are interrelated and related to the reagent dosage, a multi-output least square support vector machine (MLS-SVM) is applied to depict a dynamic relationship between the weights and the reagent dosage. Finally, an entropy based optimization algorithm is proposed to determine reagent dosage in order to implement tracking control for the PDF of the output bubble size. Experimental results can show the effectiveness of the proposed method.     

基于泡沫大小动态分布的浮选生产过程加药量健康状态分析

针对矿物浮选过程泡沫大小分布随着药剂量的改变而动态变化的特点,提出一种基于泡沫大小动态分布特征的具有自学习功能的浮选生产过程加药量健康状态统计模式识别方法.首先,通过泡沫图像分割、气泡尺寸分布核密度估计获得浮选气泡大小的概率密度分布函数,采用无监督的最远邻聚类方法获得典型药剂量添加状态下的气泡尺寸统计分布特征集;然后,采用简单的贝叶斯推理方法获得测试时间段对应的药剂添加健康状态分析识别结果,并根据浮选生产工况状态的动态变化对各典型药剂状态下的气泡大小统计分布特征集进行在线学习修正.实验结果表明,所提出方法能实时获取泡沫尺寸分布的动态变化,实现浮选药剂操作健康状态的自动识别与评价,为进一步实现浮选生产过程的加药量优化控制奠定了基础.     

Statistic PID tracking control for non-Gaussian stochastic systems based on T-S fuzzy model

A new robust proportional-integral-derivative （PID） tracking control framework is considered for stochastic systems with non-Gaussian variable based on B-spline neural network approximation and T-S fuzzy model identification. The tracked object is the statistical information of a given target probability density function （PDF）, rather than a deterministic signal. Following B-spline approximation to the integrated performance function, the concerned problem is transferred into the tracking of given weights. Different from the previous related works, the time delay T-S fuzzy models with the exogenous disturbances are applied to identify the nonlinear weighting dynamics. Meanwhile, the generalized PID controller structure and the improved convex linear matrix inequalities （LMI） algorithms are proposed to fulfil the tracking problem. Furthermore, in order to enhance the robust performance, the peak-to-peak measure index is applied to optimize the tracking performance. Simulations are given to demonstrate the efficiency of the proposed approach.     

A new froth image classification method based on the MRMR-SSGMM hybrid model for recognition of reagent dosage condition in the coal flotation process

Intelligent separation is a core technology in the transformation, upgradation, and high-quality development of coal. Realising the intelligent recognition and accurate classification of coal flotation froth is a key technology of intelligent separation. At present, the coal flotation process relies on artificial recognition of froth features for adjusting the reagent dosage. However, owing to the low accuracy and subjectivity of artificial recognition, some problems arise, such as reagent wastage and unqualified product quality. Thus, this paper proposes a new froth image classification method based on the maximal-relevance-minimal-redundancy (MR MR)-semi-supervised Gaussian mixture model (SSGMM) hybrid model for recognition of reagent dosage condition in the coal flotation process. First, the features of morphology, colour, and texture are extracted, and the optimal froth image features are screened out using the maximal-relevance-minimal-redundancy (MRMR) feature selection algorithm based on class information. Second, the traditional GMM clusterer is improved, called SSGMM, by introducing a small number of marked samples, the traditional GMM’ problems of unclear training goals, invisible clustering results, and artificially judged clustering results are solved. Then a new hybrid classification model is proposed by combining the MRMR with the modified GMM (SSGMM) which can be named as (MRMR - SSGMM). The optimal froth image features are screened by MRMR to provide the SSGMM classifier. In the process of training and learning the feature samples, using the marked feature samples of froth images to guide the unmarked feature samples. The information of marked feature samples of froth images is mapped to the unmarked feature samples, the classification of the froth images were realised. Finally, the accuracy of the SSGMM classifier is used as the evaluation criterion for the screened features by MRMR. By automatically executing the entire learning process to find the best number of froth image features and the optimal image features, so that the classifier achieves the maximum classification accuracy. Experimental results show that the proposed classification method achieves the best results in accuracy and time, compared with other benchmark classification methods. Application results show that the method can provide reliable guidance for the adjustment of the reagent dosage, realize the accurate and timely control of the reagent dosage, reduce the consumption of the reagent and the incidence of production accidents, and stabilize the product quality in the coal flotation production process.

     

PI control based on fuzzy set-point weighting tracking for hydraulic crane boom system

A PI control strategy based on fuzzy set-point weighting following was proposed for the active damp- ing control of a hydraulic crane boom system (HCBS). Two valve-controlled PI controllers, which include a proportional feedforward controller based on fuzzy set-point weighting following and a limited semi-integrator(LSI), are designed re- spectively. LSI is used to limit output signal and to prevent wind up at the low frequency of the spectrum . By using a range camera and an electronic feedback control, the tip damping on the HCBS can be adjusted artificially. A collaborative control simulation technique of HOPSAN and MATLAB/SIMULINK is applied to the controller design. Simulation results show that the proposed PI control system has less overshoot as well as faster response. The tip damping on the HCBS during operation is improved.     

PI control based on fuzzy set-point weighting tracking for hydraulic crane boom system

A PI control strategy based on fuzzy set-point weighting following was proposed for the active damping control of a hydraulic crane boom system （HCBS）. Two valve-controlled PI controllers, which include a proportional feedforward controller based on fuzzy set-point weighting following and a limited semi-integrator（LSI）, are designed respectively. LSI is used to limit output signal and to prevent wind up at the low frequency of the spectrum. By using a range camera and an electronic feedback control, the tip damping on the HCBS can be adjusted artificially. A collaborative control simulation technique of HOPSAN and MATLAB/SIMULINK is applied to the controller design. Simulation results show that the proposed PI control system has less overshoot as well as faster response. The tip damping on the HCBS during operation is improved.     

Multi-objective PID control for non-Gaussian stochastic distribution system based on two-step intelligent models

A new method for controlling the shape of the conditional output probability density function (PDF) for general nonlinear dynamic stochastic systems is proposed based on B-spline neural network (NN) model and T-S fuzzy model. Applying NN approximation to the measured PDFs, we transform the concerned problem into the tracking of given weights. Meanwhile, the complex multi-delay T-S fuzzy model with exogenous disturbances, parametric uncertainties and state constraints is used to represent the nonlinear weigh...     

Self-tuning of 2 DOF control based on evolving fuzzy model

In this paper we present a self-tuning of two degrees-of-freedom control algorithm that is designed for use on a non-linear single-input single-output system. The control algorithm is developed based on the Takagi-Sugeno fuzzy model, and it consists of two loops: a feedforward loop and feedback loop. The feedforward part of the controller should drive the system output to the vicinity of the reference signal. It is developed from the inversion of the T-S fuzzy model. To achieve accurate error-free reference tracking a feedback part of the controller is added. A time-varying error-model predictive controller is used in the feedback loop. The error-model is obtained from the T-S fuzzy model. The T-S fuzzy model of the system, required in the controller, is obtained with evolving fuzzy modelling, which is based on recursive Gustafson-Kessel clustering algorithm and recursive fuzzy least squares. It employs evolving mechanisms for adding, removing, merging and splitting the clusters.The presented control approach was experimentally validated on a non-linear second-order SISO system helio-crane in simulation and real environment. Several criteria functions were defined to evaluate the reference-tracking and disturbance rejection performance of the control algorithm. The presented control approach was compared to another fuzzy control algorithm. The experimental results confirm the applicability of the approach.     

锑浮选过程加药量自适应迭代学习控制

针对现有的加药量控制方法需要浮选过程动态模型或是鲁棒性不足的问题, 提出一种基于自适应动态规划 (ADP) 的浮选过程加药量自适应迭代学习控制方法. 首先, 将药剂量控制问题转化为两级优化问题 (问题 1 和问题 2). 其中, 基于前馈控制原理求解问题 1 得出前馈补偿分量以抑制外界扰动. 然后, 采用基于值迭代的 ADP 算法, 求解问题 2 以得到最优反馈增益, 从而设计一个数据驱动的最优加药量控制策略使最终的生产指标 (精矿品位和尾矿品位) 跟踪给定值, 且药剂量消耗最少. 最后, 通过工业生产数据进行仿真验证, 证明所提方法的收敛性和稳定性.     

Cascaded two-degree-of-freedom control of seeded batch crystallisations based on explicit system inversion

Simple structures and robustness against disturbances are important attributes of chemical productions controllers. The present contribution considers these aspects for seeded batch cooling crystallisations. A new cascaded control scheme is presented. It combines consistent feedforward control with classic feedback control of the main physical process variables, which are the crystalliser temperature, the supersaturation and the crystal mean size. The calculation of the feedforward trajectories uses an explicit inversion of the crystalliser model which is based on the Methods of Moments. A state observer is used to determine online the respective moments of the crystal size distribution. An additional observer is included to estimate unmeasurable heat disturbances and to update the temperature feedforward trajectories. The present contribution summarises the model derivation, system inversion, feedforward and feedback controller design and the design of the observers. Numerical simulations and experimental results from a laboratory plant at BASF Ludwigshafen prove the applicability of the proposed control concept.     

用B样条神经网络设计自适应模糊控制器*

本文提出一种可用于设计自适应模糊控制器的模化Ｂ样条神经网络，并给出了合适的训练算法。由于这种网络在每次训练时仅需对少量权重进行调整，因此构成的模糊控制器学习速率快，可应用于过程控制中。本文最后以电厂中过热汽温的控制为例，说明本文的设计方法是有效的。     

MRMR和SSGMM联合分类模型的煤泥浮选系统药况图像识别

为解决煤泥浮选过程依靠工人肉眼识别泡沫特征来调节药剂用量,造成药剂浪费,产品质量不合格的问题,提出一种MRMR和SSGMM联合分类模型的药况图像识别方法.针对泡沫图像的形态、纹理、颜色特征与泡沫类别具有不同程度的相关性.将精煤灰分作为泡沫的类别信息,利用最大相关最小冗余(MRMR)算法筛选最优特征;针对传统的高斯混合模型(GMM)在聚类时,存在结果需人为判断实现分类的问题,通过引入少量已知加药状况下的泡沫图像特征样本对其改进,构建半监督高斯混合模型(SSGMM)泡沫图像聚类器.将优选的且具有少量先验标签信息的多维泡沫图像特征融合到SSGMM聚类模型中,利用少量的标记样本引导聚类,并将其标签信息映射给聚类结果实现自动分类.实验表明,这种联合分类模型提高了泡沫识别的准确性,为药剂用量的准确控制与精煤产品质量提供了关键技术支持.     

泡沫图像特征驱动的锑粗选加药控制策略

锑粗选工序的加药控制直接影响精选与扫选的性能.通常由人工观察泡沫手动调节药剂.这种方式,存在控制滞后、主观随意性大、易导致浮选性能不稳定甚至恶化的问题.对此,我们提出一种泡沫图像特征驱动的锑粗选加药控制策略.利用概率支持向量回归方法建立基于锑粗选关键泡沫图像特征与加药量的入矿品位估计模型;在此基础上,采用操作模式匹配方法实现加药量的预设定,快速满足入矿品位类型变化后新的控制要求;并采用基于区间II型模糊系统的加药反馈控制器减小泡沫状态与期望的偏差.工业验证结果表明,该方法能有效代替人工加药并改善了锑浮选性能.     

基于LBPV的浮选泡沫图像纹理特征提取*

在浮选生产中,浮选泡沫表面纹理与浮选工况密切相关,直接反映泡沫层的矿化程度(品位高低).为了给浮选操作提供指导,提出了一种基于LBPV( local binary pattern variance)的泡沫图像纹理特征提取方法.该方法通过融合泡沫图像局部空间结构和对比度来提取泡沫图像纹理特征,然后将LBPV纹理特征应用于...     

模糊复合控制方法在焦炉控制系统中的应用研究

针对焦炉温度的大惯性、纯滞后、非线性和时变性等特点。提出了一种新的模糊复合控制方法．它将常规的前馈控制、反馈控制与具有人工智能的模糊控制相结合。吸取了前馈控制改善系统动态响应特性、反馈控制消除稳态误差以及模糊控制能够较好地解决系统难以建立精确数学模型的优点。解决了焦炉温度控制问题．通过模糊复合控制的理论分析和仿真试验。证明了该控制方法的可行性和有效性．     

基于遗传FCM算法和SVM的图像检索

提出基于遗传FCM聚类算法和SVM相关反馈的图像检索方法。首先对图像库提取颜色和纹理特征,采用遗传FCM聚类算法对图像进行聚类,得到每个图像类的聚类中心;最后计算查询示例图像和对应图像类的图像之间的相似度,按照相似度的大小返回检索结果。为了进一步提高检索精度,提出基于SVM的相关反馈算法。实验结果表明,提出的方法具有优良的检索性能。     

Learning feedforward control using a Dilated B-spline network: frequency Domain Analysis and design

This paper presents a frequency-domain analysis and design approach for a learning feedforward controller (LFFC) using a dilated B-spline network. The LFFC acts as an add-on element to the existing feedback controller (FBC). The LFFC signal is updated iteratively based on the FBC signal of the previous iteration as the task repeats. Similar to proportional-integral-derivative controller tuning, there are only two parameters to adjust: The B-spline support width and the learning gain. The effect of dilation in the B-spline network is discussed. Detailed design formulae are given based on a stability analysis. As an illustration, simulation results on the path tracking control of a wheeled mobile robot are presented.     

A nonlinear state‐feedback state‐feedforward tracking control strategy for a boiler‐turbine unit

A boiler‐turbine unit is a primary module for coal‐fired power plants, and an effective automatic control system is needed for the boiler‐turbine unit to track the load changes with the drum water level kept within an acceptable range. The aim of this paper is to develop a nonlinear tracking controller for the Bell‐Åström boiler‐turbine unit. A Takagi‐Sugeno fuzzy control system is introduced for the nonlinear modeling of the Bell‐Åström boiler‐turbine unit. Based on the Takagi‐Sugeno fuzzy models, a nonlinear tracking controller is developed, and the proposed control law is comprised of a state‐feedforward term and a state‐feedback term. The stability of the closed‐loop control system is analyzed on the basis of Lyapunov stability theory via the linear matrix inequality approach and Schur complement. Moreover, model uncertainties are also considered, and it is proved that with the proposed control law the tracking error converges to zero. To assess the performance of the proposed nonlinear state‐feedback state‐feedforward control strategy, a nonlinear model predictive control strategy and a linear strategy are presented as comparisons. The effectiveness and the advantages of the proposed nonlinear state‐feedback state‐feedforward control strategy are demonstrated by simulations.     

Intelligent Setting Control for Clinker Calcination Process

In the process of clinker calcination, the target value of the raw meal decomposition rate (RMDR) is different in the easy and difficult calcination stages because the boundary conditions of raw meal (i.e., raw meal flow, raw meal ingredients, and particle size) change frequently, where RMDR cannot be guaranteed to be within its desirable range. To solve this problem, an intelligent setting control method is proposed in this paper for a clinker calcination process. The proposed approach is realized by on‐line adjustment of the setpoints of control loops in line with the changes of raw meal boundary conditions. This method consists of five modules, namely an RMDR target value setting model, a control loop pre‐setting model, a feedback compensation model based on the fuzzy rules, a feedforward compensation model based on the fuzzy rules, and a soft measurement model for RMDR. Successful application to the clinker calcination process of the Jiuganghongda Cement Plant in China has been made, where the efficiency of the proposed method has been validated by the results of the practical application.