核电站蒸汽发生器水位模糊预测控制

Fuzzy-model predictive control on water level of U–tube steam generator

在核电站运行过程中,U形管蒸汽发生器的水位作为重要参数需维持在安全的范围内.U形管蒸汽发生器结构复杂、系统逆动态、大范围变工况下具有强非线性,尤其在低负荷下,采用常规控制难以取得良好效果.本文建立了蒸汽发生器水位模糊模型,提出了能够满足系统输入输出约束的基于模糊模型的准–最小–最大预测控制方法.为了减轻在线运算负担,通过线性矩阵不等式离线计算椭圆不变集合及其对应的反馈控制律,然后依据系统的状态,二等分搜索对应的椭圆不变集参数,将在线计算简化为一个简单的优化问题.针对水位设定值跟踪和负荷变化的仿真结果表明了本文所提出控制策略的有效性.

In a nuclear power plant, the water level of the U-tube steam generator （UTSG） is an important process pa- rameter which must be maintained in a safe range. Traditional control methods encounter difficulties due to the complexity, strong nonlinearity and inverse response dynamics, especially at low operating power levels. In this paper, the steam gen- erator water level dynamics is modeled into a set of fuzzy rules to represent its nonlinear dynamics. Quasi-min-max fuzzy model predictive controller （FMPC） with constraints handling is then developed. In order to reduce the online computation- al burden, a bank of ellipsoid invariant sets together with the corresponding feedback control laws are obtained by off-line solving some linear matrix inequalities （LMIs）. Based on the UTSG states, the online part is simplified to a constrained op- timization problem with a bisection search for the corresponding ellipsoid invariant set. Simulations, considering both the water level set-point tracking and the power-level changina, are given to show the effectiveness of the proposed controllers.