基于多约束条件和广义系统理论的复杂设备故障诊断方法

As the requirement for the security of industrial production becomes higher, the importance of accurate fault diagnosis of the key equipment is also increasing. As a result of complex system structure, poor working environment, and diverse types of faults, the existing methods of fault diagnosis, esp. fault estimation, are not very effective. Finding out the constraints that the state of the system faces and reasonably utilizing the priori information is an effective way to improve the estimation accuracy. Therefore, in this project, the fault estimation problem on complex equipment (such as megawatt wind turbine) is studied with the application of singular system theory. The main researches are: 1) According to the internal mechanism of the system, a mathematical model which meets the constraints condition is to be established.2) Analyze the effect of the constraints on the estimability of the fault, establish the complete criterion of measurement system, and find out the a general method for the unified description of the different types of fault in the framework of the singular system.3) Respectively conduct research on the common system component fault, sensor fault, as well as components and sensors fault to establish a direct diagnosis of fault estimation strategy. The purpose of this project is to dig out the fault information inherent in constraints, to reveal the general rules of fault estimation based on multiple constraints, and provide theoretical basis and technical support for the accurate fault diagnosis of complex equipment.

随着现代社会对工业生产安全性的要求越来越高，对关键设备实现精确的故障诊断变得越来越重要。但复杂的系统结构、未知的外界干扰和多样的故障类型导致现有方法在进行故障诊断特别是故障估计时效果不佳，而找出系统状态满足的约束条件并合理利用其包含的先验信息是提高估计精度的有效途径。为此，本项目拟以复杂设备（如兆瓦级风力发电机组）为研究对象，结合广义系统理论，开展故障估计的基础科学研究。重点研究：1）根据系统的内部机理，建立状态满足多约束条件的数学模型；2）分析约束信息对故障可估性造成的影响，建立测量信息的完备性准则，寻找将不同类型故障统一描述在广义系统框架下的普适方法；3）分别就常见的系统元件故障、传感器故障以及元件与传感器并发故障开展研究，建立直接实现故障估计的诊断策略。本项目旨在挖掘出约束条件蕴含的故障信息，揭示基于多约束条件系统故障估计的一般规律，为实现复杂设备的精确诊断提供理论依据和技术支持。

随着现代社会对工业生产安全性的要求越来越高，对关键设备实现精确的故障诊断变得越来越重要。但复杂的系统结构、未知的外界干扰和多样的故障类型导致现有方法在进行故障诊断特别是故障估计时效果不佳，而找出系统状态满足的约束条件并合理利用其包含的先验信息是提高估计精度的有效途径。为此，本项目以复杂设备（如兆瓦级风力发电机组）为研究对象，结合广义系统理论，开展故障估计的基础科学研究。重点研究：1）根据系统的内部机理，建立状态满足多约束条件的数学模型；2）分别提取等式和不等式约束所包含的信息，实现系统的状态估计，揭示系统内部的运行情况；3）就常见的执行器故障、传感器故障以及并发故障开展研究，结合广义系统理论建立直接实现故障估计的诊断策略。本项目旨在挖掘出约束条件蕴含的故障信息，揭示基于多约束条件系统故障估计的一般规律，为实现复杂设备的精确诊断提供理论依据和技术支持。