Data driven hybrid model identification for control and optimisation of petrochemical and refining plants

用于石化和炼油厂控制和优化的数据驱动混合模型识别

• 批准号: 523634-2018
• 负责人: Mahalec, Vladimir
• 金额: $ 2.91万
• 依托单位: McMaster University
• 依托单位国家: 加拿大
• 项目类别: Collaborative Research and Development Grants
• 财政年份: 2019
• 资助国家: 加拿大
• 起止时间: 2019-01-01 至 2020-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=687616
• 关键词: Data; driven; hybrid; model; identification

Excellence in operation of process plants is attained by determining the best operating conditions (optimisationof operation) and controlling their real-time operation to maintain these best conditions. Such control andoptimisation require accurate process models.Since 1970s there have been two separate streams of model development: (i) rigorous models derived fromthe first principles, and (ii) empirical models identified from plant data via methods developed in theautomatic control community. The former usually require a large effort in deriving the first principlesequations and construction of efficient algorithms for solving them. The latter have proceeded along the pathof identifying empirical models from the plant operating data. Even though significant advances have beenmade, and there are many instances of very successful use of both types of models for optimisation and control,respectively, there are still many opportunities for improvement. For instance, identification of models forprocesses with large time delaysDuring the last decade there have been significant advances in artificial intelligence methods for speechrecognition, image recognition and classification, handwriting recognition etc. Foundation for these advancesare deep neural networks comprised on many layers. It has been found that specific neural network structuresare best at creating models for specific types of applications. This research proposes to identify very accuratemodels from operating data by developing specific model structures for specific types of processes. It willcombine some first principles equations (e.g. mass and energy balances) with deep neural networks or withmodels developed via identification methods from the automatic control field. Models predicting bothsteady-state and dynamic behaviour will be developed, with particular attention devoted to models with largetime delays (e.g. ethane/ethylene splitter). Having developed a "standard form" of the model for specificequipment will enable such models to be readily adjusted to represent specific equipment and be re-used.

通过确定最佳操作条件（操作优化）并控制其实时操作以维持这些最佳条件，可以实现过程工厂的卓越操作。这样的控制和优化需要精确的过程模型，自20世纪70年代以来，有两个独立的模型开发流：（i）严格的模型来自第一原则，和（ii）经验模型确定从工厂数据通过方法开发在theautomatic控制社区。前者通常需要大量的工作来推导第一性原理方程和构造求解它们的有效算法。后者沿着从电厂运行数据中识别经验模型的道路前进。尽管已经取得了重大进展，并且有许多非常成功地使用这两种类型的模型分别进行优化和控制的实例，但仍然有许多改进的机会。例如，识别具有大时间延迟的过程的模型在过去的十年中，语音识别，图像识别和分类，手写识别等人工智能方法取得了重大进展。这些进步的基础是由多层组成的深度神经网络。已经发现，特定的神经网络结构最适合为特定类型的应用程序创建模型。本研究提出了通过为特定类型的过程开发特定的模型结构来从操作数据中识别非常准确的模型。它将结合一些第一原理方程（例如质量和能量平衡）与深度神经网络或通过自动控制领域的识别方法开发的模型。将开发预测稳态和动态行为的模型，特别注意具有大时间延迟的模型（例如乙烷/乙烯分离器）。为具体设备制定一个“标准形式”的模型，将使这些模型能够随时调整，以代表具体设备，并重新使用。