针对配电网故障预测精度低导致的配网运行不稳定问题，提出一种基于 STL-LSTM 的配电网故障预测模型。 首先对配电网故障数据集进行数据分析，使用 STL（seasonal and trend decomposition using loess，STL）进行时间序列分 解，然后引入长短时记忆网络（long-short memory network，LSTM），对数据进行特征提取，对处理后的数据进行预测。 选择平均绝对百分比误差（MAPE）与决定系数（R2）作为性能评价指标 。实验结果表明，该模型的 MAPE 和 R2 分别达 到 0.06 和 0.78，相较于其他模型具有更优秀的性能，为配电网故障数据预测提供了一种精度可靠的的预测方法。
[1] 郝丽丽，王辉，王国栋，等 . 含分布式电源配电网运行风险 的影响因素溯源[J]. 电力自动化设备，2021，41（01）∶27-37.
[2] 高锋阳，李昭君，袁成，等 . 含特殊负荷的配电网故障定位 与识别[J]. 电力自动化设备，2020，40（08）∶96-108.
[3] 张健磊，高湛军，陈明，等 . 考虑复故障的有源配电网故障 定位方法[J]. 电工技术学报，2021，36（11）∶2265-2276.
[4] Ognjanovic I，Gasevic D，Dawson S.Using institutional data to predict student course selections in higher education[J]. The Internet and Higher Education，2016，29∶49-62.
[5] Khameneh P A，Lavasani S M M，Nodehi R N，et al. Water distribution network failure analysis under uncertainty[J]. In- ternational Journal of Environmental Science and Technology， 2020，17（04）∶421-432.
[6] Fan C，Xiao F，Zhao Y. A short-term building cooling load prediction method using deep learning algorithms[J].Applied energy，2017，195∶222-233.
[7] Chen C H.An arrival time prediction method for bus system
[J].IEEE Internet of Things Journal，2018，5（05）∶4231-4232.
[8] Ren L，Meng Z，Wang X，et al. A Wide-Deep-Sequence Model- Based Quality Prediction Method in Industrial Process Analy- sis[J]. IEEE Transactions on Neural Networks and Learning Systems，2020，31（09）∶3721-3731.
[9] Feng C，Cui M，Hodge B M，et al.A data-driven multi-model methodology with deep feature selection for short-term wind forecasting[J].Applied Energy，2017，190 ∶1245-1257.
[10] Wang H Q，Cai YN，Fu G Y，et al.Data-driven fault prediction and anomaly measurement for complex systems using sup- port vector probability density estimation[J]. Engineering Applications of Artificial Intelligence，2018，67 ∶1-13.
[11] Abiodun O I，Jantan A，Omolara A E，et al. State-of-the-art in artificial neural network applications∶A survey[J]. Heli- yon，2018，4（11）∶e00938.
[12] Goldberg Y. Neural network methods for natural language processing[J].Synthesis Lectures on Human Language Tech- nologies，2017，10（01）∶1-309.
[13] Du J，Liu Y，Yu Y，et al. A prediction of precipitation data based on support vector machine and particleswarm optimiza- tion（PSO-SVM）algorithms[J].Algorithms，2017，10（02），57.
[14] Zhang P Y，Shu S，Zhou M C. An online fault detection model and strategies based on SVM-grid in clouds[J].IEEE/ CAA Journal of Automatica Sinica，2018，5（02）∶445-456.
[15] Shahid F，Zameer A，Muneeb M.Predictions for COVID-19 with deep learning models of LSTM，GRU and Bi-LSTM [J].Chaos，Solitons & Fractals，2020，140 ∶110212.
[16] Zhang S，Wang Y，Liu M，et al. Data-based line trip fault prediction in power systems using LSTM networks and SVM [J].IEEE Access，2017，6∶7675-7686.
[17] Qin X，Wang P，Liu Y，et al. Research on distribution network fault recognition method based on time-frequency character- istics of fault waveforms[J].IEEE Access，2017，6∶7291-7300.
[18] Mahdavinejad M S，Rezvan M，Barekatain M，et al. Machine learning for Internet of Things data analysis∶A survey[J]. Digital Communications and Networks，2018，4（03）∶161-175.
[19] Chen D，Zhang J，Jiang S.Forecasting the Short-Term Metro Ridership With Seasonal and Trend Decomposition Using Loess and LSTM Neural Networks[J]. IEEE Access，2020，8∶ 91181-91187.
[20] Wen Q，Gao J，Song X，et al.RobustSTL ∶A robust seasonal- trend decomposition algorithm for long time series[J]. Pro- ceedings of the AAAI Conference on Artificial Intelligence. 2019，33 ∶5409-5416.
