数据驱动的复杂磨矿过程运行优化控制方法

针对赤铁矿磨矿过程的磨矿粒度（Grinding particle size,GPS）与控制回路输出之间的动态特性难以用数学模型描述,且磨矿粒度不能在线测量,并受矿石成分与性质频繁波动干扰,难以采用已有运行优化方法的难题,结合磨矿过程的特点,利用数据,采用神经网络,提出由回路预设定值优化、性能指标估计、优化设定值评价以及磨矿粒度软测量组成的数据驱动的磨矿过程运行优化控制方法. 该方法由磨矿粒度软测量估计矿浆粒度,通过回路预设定值优化模块求得使性能指标估计值接近最优值的回路预设定值,经优化设定值评估产生回路设定值,最后通过控制回路跟踪设定值,将矿浆粒度控制在目标值范围内并尽可能的接近目标值. 通过研制的运行优化与控制研究平台,采用实际运行数据进行仿真实验,表明所提方法的有效性.     

基于案例推理增强学习的磨矿过程设定值优化

磨矿粒度和循环负荷是磨矿过程产品质量与生产效率的关键运行指标,相对于底层控制偏差,回路设定值对其影响要严重的多.然而,磨矿过程受矿石成分与性质、设备状态等变化因素影响,运行工况动态时变,难以建立模型,因此难以通过传统的模型方法优化回路设定值.本文将增强学习与案例推理相结合,提出一种数据驱动的磨矿过程设定值优化方法.首先根据当前运行工况,采用基于Prey-Predator优化的案例推理方法,决策出可行的基于Elman神经网络的Q函数网络模型;然后利用实际运行数据,在增强学习的框架下,根据Q函数网络模型优化回路设定值.在基于METSIM的磨矿流程模拟系统上进行实验研究,结果表明所提方法可根据工况变化在线优化回路设定值,实现磨矿运行指标的优化控制.     

典型赤铁矿磨矿过程智能运行反馈控制

冶金磨矿是典型的高能耗、低效率过程,其控制与优化不仅仅是使常规过程控制系统尽可能好地跟踪期望设定值,而且要控制整个过程运行,实现表征磨矿整体运行性能的磨矿粒度与生产效率等运行指标的优化.针对我国广泛使用的赤铁矿两阶段全闭路磨矿,由于其原矿石性质与成份复杂且不稳定、粒级波动大,磨矿运行指标不能在线测量,工况时变,难以建立过程数学模型,提出基于数据与知识的智能运行反馈控制方法,包括基于案例推理的控制回路预设定、磨矿粒度动态神经网络软测量以及多变量模糊动态调节器.为了验证所提方法的有效性,将所题方法应用于中国某大型赤铁矿选厂,取得显著应用成效.     

工业过程运行的解耦内模控制方法

工业过程运行控制的目的是实现反映过程整体运行性能的工艺指标. 将常规解耦内模控制(Internal model control, IMC)进行推广, 提出了优化过程运行的解耦IMC方法. 通过对广义解耦内模控制器的设计获得了具有高维解耦能力、鲁棒稳定性和抗干扰能力强的回路设定模型. 该模型能够响应系统的各种不确定性和干扰, 对回路设定值进行调整, 通过控制回路的输出跟踪调整后的设定值, 从而实现期望的工艺指标. 磨矿回路运行的解耦IMC设计实例及仿真验证了所提方法的有效性.     

实现粒度指标的磨矿过程智能优化控制系统

磨矿过程具有多变量、非线性、强耦合、大时滞、时变等综合复杂特征，其关键工艺指标磨矿粒度难以用常规控制方法进行直接控制。将基础控制、智能控制以及软测量技术相结合，构建了实现磨矿粒度指标的磨矿过程智能优化控制系统。该系统由基础控制系统、过程监控系统以及智能优化设定系统组成，具有设备顺序逻辑控制、回路控制、过程监控以及回路智能优化设定等功能。该系统已成功应用在某大型选矿厂的磨矿过程中，取得了显著效益，具有推广应用的前景。     

磨矿过程磨机负荷的智能监测与控制

磨机过负荷是磨矿过程的常见故障工况, 如果不及时、准确处理, 就会造成磨矿产品质量变坏甚至磨矿生产的停顿. 采用规则推理(RBR)和统计过程控制(SPC)技术, 提出了由SPC机制、过负荷监测模块和监督控制器构成的磨机负荷智能监测与控制方法. 该方法通过对磨机过负荷的智能监测与诊断, 由监督控制器自动修改控制回路的设定值, 通过控制回路的输出跟踪修改后的设定值, 使磨机负荷逐渐远离过负荷状态. 工业应用表明, 该方法能够实现磨矿生产的安全、稳定和连续运行.     

模糊智能控制在磨矿分级给矿系统中的应用

磨矿分级给矿系统是一个具有纯滞后的多变量系统,采用常规控制方法来控制磨矿分级给矿系统中的给矿设定值,很难获得满意的控制效果.针对磨矿分级给矿系统的特性及控制要求,提出了用模糊优化算法得到给矿设定值和用传统PID控制器实现闭环稳定给矿的控制方案,最后给出了控制系统的实现.实践表明,采用模糊控制与传统PID控制相结合的串级给矿控制系统能很好地优化给矿设定值,控制效果良好.     

再磨过程泵池液位区间与给矿压力模糊切换控制

赤铁矿再磨过程泵池液位受到大的随机干扰的影响, 造成泵池液位波动大, 采用已有的再磨过程泵池液位定值闭环控制方法, 必然会造成矿浆泵转速在大范围内频繁变化, 从而使给矿压力频繁波动在工艺规定的范围之外, 降低了旋流器的分级效率. 本文提出了由泵池液位区间控制和给矿压力回路控制组成的模糊切换控制方法, 泵池液位区间控制通过对给矿压力设定值保持器和模糊补偿器的切换, 将给矿压力设定值控制在所允许的波动范围内; 通过给矿压力PI回路控制器跟踪其设定值, 从而将泵池液位控制在目标值范围内, 并将给矿压力的波动控制在允许的范围内. 在国内某大型赤铁矿选矿厂的成功应用, 表明采用该方法有效减少了泵池液位和给矿压力的波动, 使得再磨过程安全运行, 提高了旋流器分级效率.     

磨矿粒度动态过程的一种快速Monte Carlo仿真方法

磨矿是降低矿物粒度的工业过程,产品粒度是磨矿过程的关键质量指标. 由于磨矿粒度难以在线检测且磨矿生产过程具有综合复杂特性,难以采用传统控制方法实现磨矿粒度的控制. 因此,建立磨矿粒度和关键工艺参数的动态模型对于磨矿运行控制和优化具有重要意义. 采用总量平衡原理获得磨矿粒度的微分方程模型多数情况下无法获得解析解. 而基于Monte Carlo （MC）方法的磨矿粒度模型能够精确模拟磨矿粒度分布的动态变化,但是其仿真效率低难以实用. 本文针对这一问题提出一种新的MC仿真方法: 在定总量方法的基础上引入新的颗粒移除机制,在移除过程中动态地分配各个粒级颗粒数目并保持破裂前后各个粒级颗粒所占总颗粒数的百分比不变,避免颗粒移除过程中由于粒级差异导致的抽样误差,且避免MC仿真速度随着仿真推进下降的问题. 仿真实验验证表明,本方法能够在保证一定精度前提下显著提高磨矿粒度MC仿真的计算速度. 最后,通过一个实例介绍了本文仿真模型在磨矿优化控制中的应用.     

磨矿过程基础回路优化控制方法

以磨矿过程基础回路重要工艺参数——旋流器给矿浓度控制为研究对象,针对无模型自适应控制(MFAC)的参数自适应性差的问题,引入模糊控制,提出了模糊MFAC方法,给出了该方法的理论推导步骤,设计了模糊MFAC控制器。将模糊MFAC方法、基本MFAC方法和PID方法进行对比仿真实验,结果表明,模糊MFAC方法能够快速跟踪期望值,具有更小的超调量和跟踪误差,且抗干扰能力强。     

Intelligent optimal control system for ball mill grinding process

Operation aim of ball mill grinding process is to control grinding particle size and circulation load to ball mill into their objective limits respectively, while guaranteeing producing safely and stably. The grinding process is essentially a multi-input multi-output system (MIMO) with large inertia, strong coupling and uncertainty characteristics. Furthermore, being unable to monitor the particle size online in most of concentrator plants, it is difficult to realize the optimal control by adopting traditional control methods based on mathematical models. In this paper, an intelligent optimal control method with two-layer hierarchical construction is presented. Based on fuzzy and rule-based reasoning (RBR) algorithms, the intelligent optimal setting layer generates the loops setpoints of the basic control layer, and the latter can track their setpoints with decentralized PID algorithms. With the distributed control system (DCS) platform, the proposed control method has been built and implemented in a concentration plant in Gansu province, China. The industrial application indicates the validation and effectiveness of the proposed method.     

Intelligent Optimal-Setting Control for Grinding Circuits of Mineral Processing Process

During the operation of a grinding circuit (GC) in mineral processing plant the main purpose of control and optimal operation is to control the product quality index, namely the product particle size, into its technically desired ranges. Moreover, the grinding production rate needs to be maximized. However, due to the complex dynamic characteristics between the above two indices and the control loops, such control objectives are difficult to achieve using existing control methods. The complexity is reflected by the existence of process heavy nonlinearities, strong coupling and large time variations. As a result, the lower level loop control with human supervision is still widely used in practice. However, since the setpoints to the involved control loops cannot be accurately adjusted under the variations of the boundary conditions, the manual setpoints control cannot ensure that the actual production indices meet with technical requirements all the time. In this paper, an intelligent optimal-setting control (IOSC) approach is developed for a typical two-stage GC so as to optimize the production indices by auto-adjusting on line the setpoints of the control loops in response to the changes in boundary conditions. This IOSC approach integrates case-based reasoning (CBR) pre-setting controlling, neural network (NN)-based soft-sensor and fuzzy adjusting into one efficient control model. Although each control element is well known, their innovative combination can generate better and more reliable performance. Both industrial experiments and applications show the validity and effectiveness of the proposed IOSC approach and its bright application foreground in industrial processes with similar features.     

Expert system based adaptive dynamic matrix control for ball mill grinding circuit

Ball mill grinding circuit is a multiple-input multiple-output (MIMO) system characterized with couplings and nonlinearities. Stable control of grinding circuit is usually interrupted by great disturbances, such as ore hardness and feed particle size, etc. Conventional model predictive control usually cannot capture the nonlinearities caused by the disturbances in real practice. Multiple models based adaptive dynamic matrix control (ADMC) is proposed for the control of ball mill grinding circuit. The novelty of the strategy lies in that intelligent expert system is developed to identify the current ore hardness and then select a proper model for ADMC. Compared with the various nonlinear DMC strategies, the approach can synthesize and analyze as many variables and status as possible to adequately and reliably identify the process conditions, and it does not introduce additional computational complexity, which makes it readily available to the industrial practitioner. Simulation results and industrial applications demonstrate the effectiveness and practicality of this control strategy.     

新颖的模糊认知图学习方法解决工业控制问题

介绍的是基于量子粒子群算法模糊认知图的学习方法。其主要的思路是更新模糊认知图中能够使之趋向所要求的稳定状态的非零权值。将所研究的方法运用到工业控制问题,具有很大的现实意义。实验的结果表明,该方法是有效的,并优于传统的粒子群算法。     

Disturbance observer based multi-variable control of ball mill grinding circuits

Ball mill grinding circuits are essentially multi-variable systems characterized with couplings, time-varying parameters and time delays. The control schemes in previous literatures, including detuned multi-loop PID control, model predictive control (MPC), robust control, adaptive control, and so on, demonstrate limited abilities in control ball mill grinding process in the presence of strong disturbances. The reason is that they do not handle the disturbances directly by controller design. To this end, a disturbance observer based multi-variable control (DOMC) scheme is developed to control a two-input-two-output ball mill grinding circuit. The systems considered here are with lumped disturbances which include external disturbances, such as the variations of ore hardness and feed particle size, and internal disturbances, such as model mismatches and coupling effects. The proposed control scheme consists of two compound controllers, one for the loop of product particle size and the other for the loop of circulating load. Each controller includes a PI feedback part and a feed-forward compensation part for the disturbances by using a disturbance observer (DOB). A rigorous analysis is also given to show the reason why the DOB can effectively suppress the disturbances. Performance of the proposed scheme is compared with those of the MPC and multi-loop PI schemes in the cases of model mismatches and strong external disturbances, respectively. The simulation results demonstrate that the proposed method has a better disturbance rejection property than those of the MPC and PI methods in controlling ball mill grinding circuits.     

基于RBF神经网络的磨矿粒度预测模型

磨矿分级作业是选矿生产过程中至关重要的环节,磨矿粒度的好坏直接影响到浮选的精矿品位和回收率;通过分析实际磨矿过程的生产状况和基本的生产数据,磨矿粒度存在在线检测成本高、滞后时间长、实现困难等问题;在分析RBF神经网络结构特点的基础上,提出用RBF网络建立磨矿粒度预测模型,网络中心的选取采用可以在线学习的最近邻聚类算法;仿真结果表明,该网络非线性处理能力和逼近能力强,学习时间短,网络运算速度快,模型精度满足工艺要求。     

改进的RBF神经网络在磨矿指标预测中的应用

针对选矿生产过程中磨矿生产率和磨矿浓度、介质填充率和料球比3个指标之间的关系,分析了磨矿生产率对磨矿过程产品质量和工作效率的影响,提出了磨矿过程预测模型,采用了模糊聚类与改进的遗传算法相结合训练RBF神经网络的新方法。仿真结果表明,该模型可以快速准确地预报生产状况,对于企业进行实时生产控制具有较高的实用价值。