Process Fault Diagnosis and Control Using Neural Networks

本项目的主要目标是研究和演示一种基于神经网络的过程故障检测和控制系统设计的新方法。过去故障诊断系统中的方法没有适当地包括人类专家的推理策略和经验，因此不足以有效和正确地进行故障排除。该项目将探索一种使用神经网络的新方法，以解决开发基于知识的过程故障诊断和控制系统的关键问题。我们开发基于知识的系统的方法依赖于神经网络的重要特征，即它们自动分类和学习输入和输出数据之间的关联以及能够处理噪声数据的能力。私家侦探已经在一个涉及CSTR及其支持子系统的典型案例研究中成功地测试了他的方法。他计划通过对化学加工厂的更大原型进行实验，以及诊断动态过程故障来进一步研究这种方法。他还建议分析这种方法对传感器故障、噪声和不确定数据以及症状信息不完整的稳健性。研究了利用神经网络对行为过程单元及其控制进行建模的方法。此外，他还建议研究神经网络与基于知识的技术的整合。

The major goal of this project is to research and demonstrate a new approach based on neural networks towards the design of process fault detection and control systems. Past approaches in fault diagnostic systems did not properly include the human expert's reasoning strategies and experience and hence were not adequate in efficient and correct trouble-shooting. This project, will explore a novel method using neural networks that address the important issues which are central to the development of knowledge-based systems for process fault diagnosis and control. Our method for developing knowledge-based systems rests on the important characteristics of neural networks, namely, their ability to automatically classify and learn associations between input and output data and to be able to handle noisy data. The P.I. has already successfully tested his approach on a prototypical case study involving a CSTR and its support subsystems. He plans to investigate this approach further by experimenting with larger prototypes of chemical process plants as well as for diagnosing dynamic process faults. He also proposes to analyze the robustness of this approach to sensor failures, noisy and uncertain data, and incomplete information on symptoms. It is also proposed to investigate the use of neural networks for modeling the behavior process units and their in control. Furthermore, he proposes to investigate the integration of neural nets with knowledge-based techniques.