为了应对超（超）临界锅炉高温受热面管道蒸汽侧氧化腐蚀引起的过热与爆管等安全问题，提出了一种非停机模式下氧化皮在线分离与检测系统。首次应用在电厂中的氧化皮颗粒分离装置实现主蒸汽管道内氧化皮颗粒在线分离，在分离后的氧化皮通过施加强稳磁场，找出磁感应强度变化与管内相应氧化皮堆积状况的耦合关系;提出一种基于混合粒子群灰狼（PSGWO）算法优化支持向量机（SVM）的氧化皮堆积量快速软测量方法；最后采用多指标进行了实验验证。结果表明：所提出的方法在检测精度、鲁棒性及收敛性方面均具有优越性，可准确检测非停机状态下超临界锅炉管内氧化皮堆积量，测量平均相对误差在3%以内。

In order to cope with the safety problems such as overheating and tube explosion caused by the oxidation corrosion of the steam side from the superheated surface in the super (super) critical boiler,A nonstop mode scale separation and detection system is proposed in this paper.For the first time,the scale separation device used in the power plant realizes the online separation of the oxide scale particles in the main steam pipeline.After the separation of the scale,a strong stable magnetic field is applied to find the coupling relationship between the change of the magnetic induction intensity and the corresponding scale accumulation in the tube.A fast softmeasurement method based on hybrid particle swarm optimization (PSGWO) algorithm for support vector machine (SVM) is proposed.Finally,multiple indicators are used to verify the results.The results show that the proposed method is accurate in detection.It has superiority in terms of detection accuracy,robustness,and convergence,and can accurately detect the amount of scale deposit in the supercritical boiler tube under nonstop state,and the measurement error is within 3%.

TK222

华能集团合作项目（TP-18-TYK07）