用模糊神经网络模型评估电能质量

Evaluation of Power Quality by Fuzzy Artificial Neural Network

为了综合量化评估各种电能质量,实现按质论价、优质优价的购电原则,建立公平的电力市场,提出了结合人工神经网络和模糊识别理论,构造出模糊神经识别网络;根据电能质量评估的实质,构建了电能质量综合评估的模糊神经网络模型。研究表明,模型提高了识别系统的柔性处理能力,具有很强的自组织、自学习和自适应能力,可以使输出结果唯一、客观;模型不仅可确定电能质量的等级,还能评估某一等级的电能质量,从而识别同一等级电能质量的差异,分类能力和排序能力较强;模型引入隶属度概念,使评估结果更加具体。实例计算结果表明,模糊神经网络模型用于电能质量综合评估,其评估结果更为客观、合理。

To evaluate the power quality properly to deal with the problem of power quality reasonably and economically, the fuzzy neural network recognition which combining NN and FR was put forward for power quality evaluation, then the Fuzzy Neural Network Recognition model （FNNR）was established. This model not only calculates the power quality grade of the observation station, but also compares the differences between observation stations in the same grade. Fuzzy Neural Network Recognition simulates the thinking of the brain, and has very strong self-organizing, selfilearning, self-adapting ability. The model proposed in this paper has essential differences with the actual power quality integrate-evaluating methods; the output result of the model has the characteristic of sole result and objectivity, etc. Case results show that the proposed model is objective and reasonable with a special advantage used in power quality evaluation.