Photovoltaics (PV) have attracted considerable attention owing to their longer lifespans and higher generation potentials compared with other renewable energy sources. However, the intermittent nature of PV systems can degrade the power quality, hindering their widespread adoption. To mitigate the power-quality degradation resulting from the proliferation of PV, high forecasting accuracy is essential. However, missing data during the development of forecasting models can degrade performance. Therefore, appropriate imputation procedures are required. Typically, linear imputation is used. However, there is a tendency for the performance of the forecasting model to decline owing to errors between the actual and imputed values. In this study, we addressed missing PV power data using direct deletion, linear imputation, k-nearest neighbors imputation, and Generative Adversarial Imputation Nets. Subsequently, to assess the impact of weather variability on the imputation performance, we employed the “sky status” to categorize the replaced data and analyze whether differences in imputation performance emerged. Finally, we developed a PV forecasting model using the replaced data and evaluated its forecasting performance.

Photovoltaic (PV) power is subject to variability, influenced by factors such as meteorological conditions. This variability introduces uncertainties in forecasting, underscoring the necessity for enhanced forecasting models to support the large-scale integration of PV systems. Moreover, the presence of missing data during the model development process significantly impairs model performance. To address this, it is essential to impute missing data from the collected datasets before advancing with model development. Recent advances in imputation methods, including Multivariate Imputation by Chained Equations (MICEs), K-Nearest Neighbors (KNNs), and Generative Adversarial Imputation Networks (GAINs), have exhibited commendable efficacy. Nonetheless, models derived solely from a single imputation method often exhibit diminished performance under varying weather conditions. Consequently, this study introduces a weighted average ensemble model that combines multiple imputation-based models. This innovative approach adjusts the weights according to “sky status” and evaluates the performance of single-imputation models using criteria such as sky status, root mean square error (RMSE), and mean absolute error (MAE), integrating them into a comprehensive weighted ensemble model. This model demonstrates improved RMSE values, ranging from 74.805 to 74.973, which corresponds to performance enhancements of 3.293–3.799% for KNN and 3.190–4.782% for MICE, thereby affirming its effectiveness in scenarios characterized by missing data.

With advancements in information and communication technology (ICT), the use of big data has become increasingly prevalent, particularly in the power industry. This study proposes a novel data synthesis methodology to enable the secure sharing and utilization of data containing personal information while retaining the utility of the original data. To model the key components of electricity consumption data, a segmented seasonal trend decomposition using loess (STL) synthesis methodology is proposed, focusing on capturing temporal patterns and characteristics. Segmented STL synthesizes electricity consumption data by segmenting into smaller units and using Gaussian process regression (GPR) and generalized extreme value (GEV) models to effectively capture temporal patterns while enhancing privacy protection. A comparative analysis was conducted on row-wise shuffling, segment-unit row-wise shuffling, and the proposed model. The effectiveness of the proposed method was evaluated using various metrics, including population stability index (PSI), principal component analysis (PCA), and t-distributed stochastic neighbor embedding (t-SNE). In addition, a case study was conducted to determine whether the proposed method properly estimates the potential of a demand response service. The experimental results demonstrate that the proposed methodology effectively reflects the time-series characteristics and electricity usage patterns of the actual data with an error margin within 0.1% while providing a high level of privacy protection.

Amid evolving energy policies and increasingly diversified electricity consumption patterns, the operational complexity of distribution systems has intensified, leading to amplified voltage variability and heightened operational uncertainty. Time-varying load characteristics and dynamic network operating conditions contribute to inefficient system operation and pose challenges to voltage stability. As voltage directly influences the dynamic stability of a system, accurate voltage forecasting is essential for improving the reliability of system operation. Traditionally, voltage forecasting of individual buses relies primarily on load forecasting followed by power flow calculations. However, more efficient voltage forecasting methods are needed to account for the dynamic nature of voltage variations. In this study, we propose a voltage-forecasting model that integrates load forecasting based on graph convolutional network-long short-term memory (GCN-LSTM) with power-flow calculations. By considering the spatial characteristics derived from the physical configuration of the distribution system and the temporal characteristics, we predict the voltage magnitude and phase angle of each bus. A case study of a distribution system using real data from South Korea is conducted to evaluate the prediction accuracy and computational efficiency of the proposed method. When evaluated on a dataset characterized by seasonal variability, the proposed model reduced prediction errors by 7.8% to 29.9% compared with the benchmark models. These results demonstrate that, in scenarios where conventional approaches suffer from error propagation—originating from load forecasting and subsequently amplified through the power flow calculation stage—the proposed unified framework plays a significant role in enhancing the reliability of voltage forecasting.