Benchmarking Daily Line Loss Rates of Low Voltage Transformer Regions in Power Grid Based on Robust Neural Network

Line loss is inherent in transmission and distribution stages, which can cause certain impacts on the profits of power-supply corporations. Thus, it is an important indicator and a benchmark value of which is needed to evaluate daily line loss rates in low voltage transformer regions. However, the number of regions is usually very large, and the dataset of line loss rates contains massive outliers. It is critical to develop a regression model with both great robustness and efficiency when trained on big data samples. In this case, a novel method based on robust neural network (RNN) is proposed. It is a multi-path network model with denoising auto-encoder (DAE), which takes the advantages of dropout, L2 regularization and Huber loss function. It can achieve several different outputs, which are utilized to compute benchmark values and reasonable intervals. Based on the comparison results, the proposed RNN possesses both superb robustness and accuracy, which outperforms the testing conventional regression models. According to the benchmark analysis, there are about 13% outliers in the collected dataset and about 45% regions that hold outliers within a month. Hence, the quality of line loss rate data should still be further improved.

[1]  Xianguang Dong,et al.  Analysis of Influencing Factors of Transmission Line Loss Based on GBDT Algorithm , 2019, 2019 International Conference on Communications, Information System and Computer Engineering (CISCE).

[2]  Gautam Bhattacharya,et al.  Outlier detection using neighborhood rank difference , 2015, Pattern Recognit. Lett..

[3]  Subha Chakraborti,et al.  Boxplot-Based Outlier Detection for the Location-Scale Family , 2015, Commun. Stat. Simul. Comput..

[4]  Feng Jiang,et al.  Initialization of K-modes clustering using outlier detection techniques , 2016, Inf. Sci..

[5]  Ioannis Mitliagkas,et al.  Fortified Networks: Improving the Robustness of Deep Networks by Modeling the Manifold of Hidden Representations , 2018, ArXiv.

[6]  Chen Xin,et al.  Low Voltage Distribution Network Line Loss Calculation Based on The Theory of Three-phase Unbalanced Load , 2017 .

[7]  Farokh Marvasti,et al.  A Novel Approach to Quantized Matrix Completion Using Huber Loss Measure , 2018, IEEE Signal Processing Letters.

[8]  Xu Bo,et al.  Research of Typical Line Loss Rate in Transformer District Based on Data-Driven Method , 2019, 2019 IEEE Innovative Smart Grid Technologies - Asia (ISGT Asia).

[9]  Shouxiang Wang,et al.  A Novel Method of Statistical Line Loss Estimation for Distribution Feeders Based on Feeder Cluster and Modified XGBoost , 2017 .

[10]  Heiko Paulheim,et al.  A decomposition of the outlier detection problem into a set of supervised learning problems , 2015, Machine Learning.

[11]  Liu Jun,et al.  Design and application of integrated distribution network line loss analysis system , 2016, 2016 China International Conference on Electricity Distribution (CICED).

[12]  Ashkan Sami,et al.  Entropy-based outlier detection using semi-supervised approach with few positive examples , 2014, Pattern Recognit. Lett..

[13]  Arthur Zimek,et al.  Hierarchical Density Estimates for Data Clustering, Visualization, and Outlier Detection , 2015, ACM Trans. Knowl. Discov. Data.

[14]  D. Wilkes,et al.  A fast MST-inspired kNN-based outlier detection method , 2015, Inf. Syst..

[15]  Maurizio Filippone,et al.  A comparative evaluation of outlier detection algorithms: Experiments and analyses , 2018, Pattern Recognit..

[16]  Arthur Zimek,et al.  On the evaluation of unsupervised outlier detection: measures, datasets, and an empirical study , 2016, Data Mining and Knowledge Discovery.

[17]  M. Sudarma,et al.  Filtering Outlier Data Using Box Whisker Plot Method for Fuzzy Time Series Rainfall Forecasting , 2018, 2018 4th International Conference on Wireless and Telematics (ICWT).

[18]  Peter Filzmoser,et al.  Locally centred Mahalanobis distance: a new distance measure with salient features towards outlier detection. , 2013, Analytica chimica acta.

[19]  Michael Pokojovy,et al.  A Cluster-Based Outlier Detection Scheme for Multivariate Data , 2015 .

[20]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[21]  He Liu,et al.  An improved one-class support vector machine classifier for outlier detection , 2015 .

[22]  Tingquan Deng,et al.  An Adaptive Weighted One-Class SVM for Robust Outlier Detection , 2016 .

[23]  Xueying Zhang,et al.  Robust support vector data description for outlier detection with noise or uncertain data , 2015, Knowl. Based Syst..

[24]  Junjie Li,et al.  Research on Predicting Line Loss Rate in Low Voltage Distribution Network Based on Gradient Boosting Decision Tree , 2019, Energies.

[25]  Abraham Yosipof,et al.  k‐Nearest neighbors optimization‐based outlier removal , 2015, J. Comput. Chem..

[26]  Alexandros Nanopoulos,et al.  Reverse Nearest Neighbors in Unsupervised Distance-Based Outlier Detection , 2015, IEEE Transactions on Knowledge and Data Engineering.

[27]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[28]  S. Peng,et al.  Partial least squares and random sample consensus in outlier detection. , 2012, Analytica chimica acta.

[29]  M. Moghaddam,et al.  Inverse Scattering Using a Joint $L1-L2$ Norm-Based Regularization , 2016, IEEE Transactions on Antennas and Propagation.

[30]  Jiangming Zhang,et al.  A Review of Line Loss Analysis of the Low-Voltage Distribution System , 2019, 2019 IEEE 3rd International Conference on Circuits, Systems and Devices (ICCSD).

[31]  Yu Jianming,et al.  Low-Voltage Distribution Network Theoretical Line Loss Calculation System Based on Dynamic Unbalance in Three Phrases , 2010, 2010 International Conference on Electrical and Control Engineering.

[32]  M. Hubert,et al.  An adjusted boxplot for skewed distributions , 2008, Comput. Stat. Data Anal..

[33]  Sheng-Tsaing Tseng,et al.  Outlier detection in general profiles using penalized regression method , 2014 .

[34]  Li Yang,et al.  Theoretical Line Loss Calculation of Distribution Network Based on the Integrated Electricity and Line Loss Management System , 2018, 2018 China International Conference on Electricity Distribution (CICED).

[35]  Tao Ding,et al.  Ensemble Recurrent Neural Network Based Probabilistic Wind Speed Forecasting Approach , 2018, Energies.

[37]  Qingsheng Zhu,et al.  A novel outlier cluster detection algorithm without top-n parameter , 2017, Knowl. Based Syst..