热工系统实用控制性能评价研究

Practical Control Performance Assessment in Thermal Control System

面对热工系统控制回路多,控制要求高的现状,对其进行有效的性能评价是必要的。基于此,一种基于方差上下界的随机性能指标被提出,通过该指标可以判断控制系统的状态,并能给出其性能优化提升建议。然后针对指标实际应用中面临稳态数据选取、采样速率与控制速率不一致的两个问题,分别给出了基于最小熵的稳态数据选取办法和分段线性插值的数据重采样算法来解决。最后,通过仿真验证了性能指标和算法的有效性,并在某1 000MW火力发电机组的主汽压控制系统中进行了应用。

In the face of the thermal system control loop increaseing and the control for higher requirements, the effective perfor- mance evaluation is necessary. A novel stochastical performance benchmark is defined based on the upper and lower variance bound. Accoriding to the benchmark, the state of control system and the optimal control advice are avaiable. In practical, there are two problems： steady-state data selection and sampling rates faster than that used for data collection. To solving the two problems, steady-state data selection algorithm based on minimum entropy and data resampling algorithm using piecewise linear interpolation are proposed, respectively. Then, the effective of the benchmark and the algorithm is tested by simiulation. At last, the performance assessment method is used in the main steam pressure control system in a 1 0O00MM power plant.