Non-technical losses: A systematic contemporary article review

Savian, Fernando de Souza; Siluk, Julio Cezar Mairesse; Garlet, Taís Bisognin; Nascimento, Felipe Moraes do; Pinheiro, J.R.; Vale, Zita

doi:10.1016/j.rser.2021.111205

Cited by 60 publications

(24 citation statements)

References 130 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Meteorological factors such as humidity, wind speed, temperature, atmospheric pressure, and other factors are selected. The preprocessing for the aforementioned factors is as follows (Baran et al, 2013;Rudolf et al, 2019;Simões et al, 2020;Savian et al, 2021):…”

Section: Feature Engineeringmentioning

confidence: 99%

Loss Prediction of Ultrahigh Voltage Transmission Lines Based on EEMD–LSTM–SVR Algorithm

et al. 2022

View full text Add to dashboard Cite

Line loss prediction of ultrahigh voltage transmission lines is the key for ensuring the safe, reliable, and economical operation of the power system. However, the strong volatility of line loss brings challenges to the prediction of transmission line loss. For more accurate prediction, this article uses ensemble empirical mode decomposition (EEMD) to decompose the line loss and proposes the EEMD–LSTM–SVR prediction model. First of all, this article performs feature engineering on power flow, electric energy, and meteorological data and extracts the exponentially weighted moving average (EWMA) feature from the line loss. After the integration of the time dimension, this article mines the curve characteristics from the time series and constructs a multidimensional input dataset. Then, through ensemble empirical mode decomposition, the line loss is decomposed into high-frequency, low-frequency, and random IMFs. These IMFs and the standardized multidimensional dataset together constitute the final input dataset. In this article, each IMF fusion dataset is sent to LSTM and support vector regression models for training. In the training process, the incremental cross-validation method is used for model evaluation, and the grid search method is used for hyperparameter optimization. After evaluation, the LSTM algorithm predicts high-frequency IMF1 and 2 and random IMF4 and 5; the SVR algorithm predicts low-frequency IMF6 and 7 and random IMF3. Finally, the output value of each model is superimposed to obtain the final line loss prediction value. Also, the comparative predictions were performed using EEMD–LSTM, EEMD–SVR, LSTM, and SVR. Compared with the independent prediction models EEMD–LSTM and EEMD–SVR, the combined EEMD–LSTM–SVR algorithm has a decrease in mean absolute performance error% by 2.2 and 25.37, respectively, which fully demonstrates that the combined model has better prediction effect than the individual models. Compared with that of SVR, the MAPE% of EEMD–SVR decreases by 11.16. Compared with that of LSTM, the MAPE% of EEMD–LSTM is reduced by 32.72. The results show that EEMD decomposition of line loss series can effectively improve the prediction accuracy and reduce the strong volatility of line loss. Compared with that of the other four algorithms, EEMD–LSTM–SVR has the highest R-square of 0.9878. Therefore, the algorithm proposed in this article has the best effectiveness, accuracy, and robustness.

show abstract

Section: Feature Engineeringmentioning

confidence: 99%

Loss Prediction of Ultrahigh Voltage Transmission Lines Based on EEMD–LSTM–SVR Algorithm

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Many research papers have been published to overcome the challenges outlined above. A gradient boosting approach, for example, is employed to identify power theft in [5,6]. Furthermore, a particular theft window is created during peak hours to detect suspect power consumer activity.…”

Section: Introductionmentioning

confidence: 99%

A novel deep learning technique to detect electricity theft in smart grids using AlexNet

Khan,

Shahid,

Alam

et al. 2024

IET Renewable Power Gen

View full text Add to dashboard Cite

Electricity theft (ET), which endangers public safety, interferes with the regular operation of grid infrastructure, and increases revenue losses, is a significant issue for power companies. To find ET, numerous machine learning, deep learning, and mathematically based algorithms have been published in the literature. However, these models do not yield the greatest results due to issues like the dimensionality curse, class imbalance, inappropriate hyper‐parameter tuning of machine learning, deep learning models etc. A hybrid DL model is presented for effectively detecting electricity thieves in smart grids while considering the abovementioned concerns. Pre‐processing techniques are first employed to clean up the data from the smart meters, and then the feature extraction technique, AlexNet is used to address the curse of dimensionality. An actual dataset of Chinese smart meters is used in simulations to assess the efficacy of the suggested approach. To conduct a comparative analysis, various benchmark models are implemented as well. This proposed model achieves accuracy, precision, recall, and F1‐score, up to 86%, 89%, 86%, and 84%, respectively.

show abstract

“…Non-technical losses (NTLs) correspond to the electricity consumed but not paid by the consumers. This absence of payment by the consumers can be attributed to the inability of the electricity distribution company to collect its debts, the illegal connections to the network, electricity theft and frauds (de Souza Savian et al, 2021, Jamil & Ahmad, 2019. The electricity theft decreases the efficiency of the electricity network due to power outages, damage to transformers and meters.…”

Section: Introductionmentioning

confidence: 99%