双网环境下工业生产过程运行优化控制方法研究

To improve product quality, enhance production efficiency and economy benefits, and reduce costs, consumptions and environment pollutions, the operational optimal control of synthetical production indices of industrial production processes has attracted great attention from both academic and industrial communities over the years. On the other hand, due to the requirement of large-scale data transmission and remote control, communication networks have been introduced into the modern industrial production process control systems. However, there are also some limitations induced by the very nature of communication networks such as transmission delays, packet dropouts, quantization errors, data rate constraints. This project will investigate the operational optimal control problem of industrial production processes under a novel double layer network-based environment. Firstly, the network-induced limitations will be characterized and data prediction and compensation issues will be also studied. Secondly, different kinds of production index evaluation models will be built to characterize the quality, yield, costs and consumptions. Then a new framework for double layer network-based operational optimal control of industrial production processes, as well as the corresponding optimal tracking control methodologies, will be proposed. Furthermore, we will select a number of benchmark industrial production processes such as metallurgical and petrochemical processes to evaluate the effectiveness and performance of the proposed operational optimal control design approaches. Strengths and weaknesses of the proposed approaches will be critically assessed. The outcome of this project is expected to enrich the theoretical foundation on double layer network-based operational optimal control of complex industrial production processes, and, in turn, the proposed approaches will provide control engineers in practice with more solid analysis and synthesis tools.

为改进产品质量、降低生产成本、提高生产效率和经济效益、减少资源消耗和环境污染、增强企业竞争力，工业生产过程综合生产指标的运行优化控制技术受到了工业界与学术界的普遍重视与关注。本项目以实现我国工业生产过程综合指标优化为目标，结合网络化工业生产的特点与要求，通过研究设备与监控双网环境下系统数据信息的传输机理以及数据的预测与补偿方案，建立运行控制指标评价模型，提出双网环境下工业生产过程运行优化控制系统的结构框架，结合双网环境下系统综合性能要求以及运行指标评价模型，设计指标跟踪优化控制算法，从而实现双网环境下工业生产过程的一体化优化运行控制，并结合大型冶金、石化工业生产过程开展应用研究。本项目致力于解决双网环境下工业生产过程运行优化控制方面的关键科学问题，为我国工业生产过程运行控制提供系统设计方法、高性能的控制与优化核心算法，使我国工业生产过程运行控制研究处于国际先进水平。

本项目严格按照重点项目申请书中年度研究计划针对双网环境下工业生产过程的运行优化控制问题进行了深入研究，提出了双网环境下一体化闭环控制系统的混合智能运行优化控制方法，完成了本项目的所有研究内容，建立了双网环境下工业生产过程运行控制系统半实物仿真实验平台，为复杂生产制造过程的全流程一体化控制奠定了基础。完成了项目在论文、专著、专利、人才培养、获奖等方面的预期目标。在项目执行期间，总共发表SCI收录文章104篇，其中16篇发表在自动控制领域顶级期刊IEEE Trans. Automatic Control和Automatica上，73篇发表在其它IEEE Trans.系列汇刊上；发表国际会议论文22篇，出版英文专著3 部；申请国家发明专利9项，其中授权7项；主办国际学术会议3次。项目执行期间，项目负责人高会军教授当选IEEE Fellow、工业电子领域旗舰期刊IEEE Trans. Industrial Electronics共同主编、IFAC国际自动控制联合会理事会成员(Council Member)和中组部“万人计划”科技创新领军人才，入选全球高被引科学家和汤森路透发布的2014年度世界最具影响力科学家榜单；项目负责人高会军受邀在IECON, IEEE SMC等著名国际会议做大会报告7次，团队成员尹珅、邱剑彬、李鸿一入选全球高被引科学家；团队成员邱剑彬、李鸿一、赵旭东获国家自然科学基金优秀青年基金；获国家自然科学二等奖（排名第一），黑龙江省科学技术奖一等奖3项。培养硕士15 人，博士16 人，其中一人获全国博士后创新人才支持计划，三人分别获中国自动化学会和中国指挥与控制学会优秀博士学位论文。