本文成功地将模糊控制理论应用于电力系统稳定器设计，设计了一种基于规则和表格的智能型电力系统稳定器。数字仿真实验和动模实验表明了其有效性。由于采用了基于规则的人工智能技术，避免了因建立数学模型以及控制中测量信号不真实带来的麻烦。以8086为主机的s1'D工业控制机和一台Pc机为基础，研制出一台用微机实现的模糊式电力系统稳定器的样机。

提出了一种利用多目标进化算法设计可控串联补偿（TCSC）非线性稳定控制器的方案，设计目标是选取合适的控制器参数，达到提高电力系统暂态稳定性的效果。将非线性控制器反馈增益的选择归结为参数优化问题，并且在优化过程中同时考虑电力系统响应的多项性能指标。采用该方法优化的非线性控制器参数可以较有效地阻尼电力系统振荡。

针对电力系统经济调度问题，定性研究了粗粒度模型并行遗传算法中迁移策略参数对算法性能的影响。这些参数包括：子种群数目、迁移率、迁移规模、迁移选择策略和通信方式等。所获得的结论具有普遍性，对设计和开发适用于大系统优化的快速鲁棒进化算法具有一定的参考价值。

This paper presents a method to design a new nonlinear power system stabiliser (NPSS), using the nonlinear feedback transformation technique and ITAE (integration of the production of time and absolute error) rule in the optimal control theory of linear systems, to deal with the control problem of multi-mode oscillations in large-scale power systems. The design of NPSS is carried out based on a multi-machine power system model. A decoupled nonlinear control strategy is obtained for individual machines and implemented using purely feedback of local variables. Performance of the proposed nonlinear stabiliser in the multimachine power system with multi-mode oscillations is digitally simulated. The system responses controlled by the NPSS are compared with that obtained using the conventional PSS (CPSS), under different circumstances. The simulation results show that the NPSS can provide effective damping of the power system multi-mode oscillations and satisfactory control performance with good robustness.

A novel approach based on the Lyapunov decomposition-aggregation method is developed to analyze shaft stability in power systems. The power system including the turbine-generator shaft, flux decay and voltage regulator is decomposed to two isolated subsystems: mechanical subsystem and electrical subsystem. The mechanical subsystem is converted to a standard Lure control system that is analyzed by an improved Lure Lyapunov function. The stability of the power system at the equilibrium points is discussed in detail. The stable operation region of the system is obtained. Case study demonstrates the effectiveness of the proposed method. This study has shown the potential benefits and validity of Lyapunov direct method in the shaft torsional oscillation analysis.