People are increasingly using clean energy sources, contributing to environmental protection according to the general trend of the world. In the form of renewable energy, solar energy has contributed to solving current pressing problems, such as environmental pollution and air pollution, improving people’s quality of life. The design of solar power projects in Vietnam is mainly based on meteorological data sources from Meteonorm and NASA. However, the accuracy assessment of two data sources compared to the actual solar power data in Vietnam is not available, so there is no basis to determine better meteorological data source quality to serve the design of rooftop solar power projects. The content of this paper analyzes the simulation results of a typical rooftop solar power station at the Electric Power University, Hanoi city based on meteorological data sources from Meteonorm and NASA. After that, the simulation results will be compared with the actual operating data of a rooftop solar power station near the Electric Power University and other real PV systems in the world. The study results showed that the amount of electricity production using the Meteonorm meteorological data was closer to the actual data than the NASA data source. Therefore, solar power projects in Vietnam should use Meteonorm data source for the design process to determine the best economic and technical efficiency for investors.

Vietnam’s economy has been growing rapidly in the last 20 years, leading to significant increases in energy consumption as well as in carbon emissions. Most electricity is consumed by loads of industry and construction due to the country’s socio-economic development strategy. An energy saving strategy cannot be achieved if the industry factories lack energy consumption data. The installation of energy monitoring systems can help to improve energy efficiency by supplying daily, monthly, and yearly energy consumption reports. Moreover, major energy-consuming enterprises in Vietnam must implement solutions for energy-efficient use as prescribed in the Law on Energy Efficient Use. Therefore, this study aimed to determine the impact of an energy monitoring system as an improvement solution for energy efficiency in a typical major energy-consuming enterprise in Vietnam. The study’s results, after six months, show that the total saved electricity after installing the power monitoring system was 191,923 kWh. The company saved approximately 19.584 USD and reduced emission to the environment by 139 tons of CO₂. In addition, the return on investment time of power monitoring systems is about 14 months, while the annual energy costs of the factory can be reduced by about 9.62% per year. Therefore, power monitoring systems should be promoted in factories with different scales to control energy wastage in the domestic industry field.

Accurate forecasting of solar power is essential for enhancing the stability and efficiency of power systems with high Photovoltaic (PV) penetration. This paper proposes a novel hybrid model based on a Stacking Ensemble (SE) of XGBoost, LightGBM, and Random Forest (RF), with optimal weights determined using the Limited Memory Broyden–Fletcher Goldfarb Shanno with Box constraints (L-BFGS-B) algorithm. The model is trained and tested on real-world data from a 49.5 MW solar power plant in Vietnam. The experimental results show that the proposed SE-XGB-LGBM-RF-OW model outperforms individual learners and deep learning baselines in both accuracy and training time. It consistently achieves a Normalized Mean Absolute Percentage Error (NMAPE) below 1.2% across all seasons. Compared to LSTM and GRU models, SE reduces Root Mean Square Error (RMSE) by more than 90% and shortens training time by over 20 times. Additionally, it significantly lowers the MAPE and NMAPE values, with improvements exceeding 90% in most seasonal test cases, highlighting the model’s superior accuracy and generalization capability.

Recently, power companies apply optimal algorithms for short-term load forecasting, especially the daily load. However, in Vietnam, the load forecasting of the power system has not focused on this solution. Optimal algorithms and can help experts improve forecasting results including accuracy and the time required for forecasting. To achieve both goals, the combinations of different algorithms are still being studied. This article describes research using a new combination of two optimal algorithms: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). This combination limits the weakness of the convergence speed of GA as well as the weakness of PSO that it easily falls into local optima (thereby reducing accuracy). This new hybrid algorithm was applied to the Southern Power Corporation’s (SPC—a large Power company in Vietnam) daily load forecasting. The results show the algorithm’s potential to provide a solution. The most accurate result was for the forecasting of a normal working day with an average error of 1.15% while the largest error was 3.74% and the smallest was 0.02%. For holidays and weekends, the average error always approximated the allowable limit of 3%. On the other hand, some poor results also provide an opportunity to re-check the real data provided by SPC.

Convolution Neural Networks (CNNs) have become an attractive method for the detection of anomalous behaviors. However, designing an efficient CNN model in terms of classification accuracy remains a challenging problem. Furthermore, the existing datasets for abnormal behavior detection are limited, with each focusing on a certain context. Therefore, a CNN model trained on a certain dataset will be adaptive for a particular context and not suitable for other contexts. This study proposes a CNN framework with an efficient attention mechanism to capture key information from multiple inputs, namely RGB, optical flow, and heatmap. Experiments were carried out on several benchmark datasets and a self-collected dataset, and the evaluation involved both single- and cross-dataset strategies. The results show the superior performance of the proposed frameworks compared to other SOTA methods in detection accuracy.

In Vietnam, the primary technologies employed for banana processing and preservation are drying methods, predominantly used for soft-dried and crispy-dried banana products. However, no prior studies have focused on the drying of whole bananas using advanced heat pump systems. This study aims to determine the optimal drying regime for whole aromatic bananas through experimental investigation utilizing a hybrid drying system that combines heat pump technology with electrical resistance heating. The results reveal that the optimal drying process consists of three distinct stages, each defined by specific temperature and humidity parameters. Among the tested modes, the hybrid heat pump–resistance drying method demonstrated the lowest energy consumption, achieving energy savings of approximately 37% and 24% compared to the standalone resistance and heat pump drying modes, respectively, under identical conditions (i.e., mass, input moisture content, and target product moisture content). The total drying time using the hybrid method was 35 hours, with the drying agent’s relative humidity having been maintained between 20% and 25%, and the energy consumption having been measured at 2.13 kWh per kg of raw material.