动态多元状态估计算法在火力发电设备智能预警中的应用

Application of dynamic multivariate state estimation technique in intelligent warning of power equipment

火力发电设备的健康状态直接影响机组的安全性、可靠性,因此对设备的状态进行实时监测显得尤为重要。基于设备的历史正常运行工况,提出将动态多元状态估计算法用于实时判断火力发电设备的运行状态,分别对某燃气蒸汽联合循环电厂的高压给水泵、某联合循环电厂的燃气轮机及660 MW超超临界燃煤机组的末级过热器建立动态多元状态估计技术(DMSET)智能预警算法。测试结果表明:当设备正常运行时,估计向量与实时观测向量之间的欧氏距离较短,预测精度较高;当设备异常时,估计向量与实时观测向量之间的欧氏距离显著大于正常状态下的值。研究表明,所建立的DMSET智能预警算法能敏锐捕捉到设备的异常信息。

The health status of power equipment directly affects the safety and reliability of the unit.It is crucial to real-time monitor the equipment status.Based on the normal historical data,the application of dynamic multivariate state estimation technique(DMSET)to real-time operation status judgement of power equipment was proposed.The DMSET-based intelligent warning algorithm of high-pressure feedwater pump in an integrated gas-steam combined cycle power plant,9E gas turbine system,and final superheater in a 660 MW ultra-supercritical coal-fired unit was developed.Result show that when the equipment run normally,the Euclidean distance between estimator and observation vector was very small.However,the Euclidean distance between estimator and observation vector for abnormal equipment was larger than that for normal one.