Reasoning under uncertainty: A data and model-based methodology for process monitoring

现在，在很长一段时间内每秒采样数千个传感器的存档历史是很常见的。然而，我们经常有工艺工程师抱怨：“......我们淹没在数据中，却渴望信息... ". 如何使用这些丰富的数据集？本研究建议关注的是从数据中提取信息和知识的问题，重点是安全过程监控。 大多数主要的工厂、流程、设备和工具中断都是可以避免的，但可预防的故障检测和诊断策略在大多数行业中并不常见。常见的情况是，一些简单且可预防的故障会破坏整个集成制造设施的运行。例如，传感器或执行器故障、报警系统故障、控制器调整或配置不良等故障可能会使最复杂的控制系统变得无用。这种计划外的中断每天可能造成超过100万美元的损失，平均而言，它们夺走了工厂年产能的7%。 新奇：虽然统计推断一直是搜索工具的主要引擎，但这些相同的方法在流程工业中并没有得到有效的使用。 该提案的主要目标是开发一个理论框架，用于使用统计推断方案从存档的过程数据中获取模型，并随后使用这些模型在不确定条件下做出决策。具体目标是开发如何处理特定故障的方法，并根据过程连接性信息制定危害可操作性指南，这些信息可以从过程数据中获取，并通过管道和仪表图中的信息进行验证。 重要性：该提案的社会经济效益将是一种安全、高效和节能的过程监测方法，用于难以建模和操作的过程。

It is now common to have archival history of thousands of sensors sampled every second over long time periods. Yet we frequently have process engineers complain: "....We are drowning in data but starving for information...". How can these rich data sets be put to use? This research proposal is concerned with the issue of information and knowledge extraction from data with emphasis on safe process monitoring. Most of the major plant, factory, process, equipment and tool disruptions are avoidable, and yet preventable fault detection and diagnosis strategies are not the norm in most industries. It is not uncommon to see simple and preventable faults disrupt the operation of an entire integrated manufacturing facility. For example, faults such as malfunctioning sensors or actuators, inoperative alarm systems, poor controller tuning or configuration can render the most sophisticated control systems useless. Such unplanned disruptions can cost in the excess of $1 million per day and on the average they rob the plant of 7% of its annual capacity. Novelty : Whereas statistical inferencing has been the main engine in search tools, these same methodologies are not being used as effectively in the process industry. The main goal of this proposal is to develop a theoretical framework for obtaining models from archived process data using statistical inference schemes and subsequently use such models to make decisions under uncertain conditions. The specific objective is to develop methodologies on how to deal with a specific fault and develop hazard operability guidelines from the process connectivity information that can be captured from process data and validated with information from the piping and instrument diagrams. Significance: The socio-economic benefits of this proposal would be a safe, highly productive and energy efficient process monitoring methodology for processes that are otherwise difficult to model and operate.