Abstract:

To improve the static and transient stability of disturbances with all scales in power system, this paper provides a control strategy of an online self-turning stabilizer based on the parameters adjusted by a fuzzy neural network. Compared with the conventional fuzzy stabilizers, the strategy adds a fuzzy neural network parameter adjuster which stores the fuzzy rules in the neural network combing a qualitative description and a quantitative numerical calculation. The adjuster makes full use of the neural-network’s associative memory and parallel processing ability, adjusting the conventional fuzzy stabilizers’ parameters of quantitative factors and scale factors on line quickly and dynamically. According to the experiment, compared with IEEE PSS2B and conventional fuzzy stabilizers, the proposed control strategy can increase the damping of the power system effectively, strengthen the ability to withstand different scales of disturbances, with strong adaptability and robustness, and improve the static and transient stability of the power system greatly. The proportional control strategy has a broad prospect in industrial application.

摘要：

为进一步提高电力系统在各种不同大小扰动下的静态稳定性和暂态稳定性，提出了一种模糊神经网络参数在线自校正稳定器的控制策略。该方案是在原有常规模糊稳定器的基础上，增加了一个模糊神经网络参数在线调整器。该调整器将定性的知识表达和定量的数值运算相结合，把模糊推理规则存储于神经元中，充分利用了神经网络联想记忆和大规模并行处理的功能，在线、快速、动态地调整常规模糊稳定器的量化因子和比例因子等参数。通过实验证明，与IEEE PSS2B和常规模糊稳定器相比，提出的控制策略能有效地增加电力系统的阻尼，加强系统承受各类不同大小扰动的能力，具有较强适应性和鲁棒性，较大地提高了电力系统的静态稳定性和暂态稳定性。该控制方法具有较为广阔的工业应用前景。