Search

This study proposes a deep learning-based long short-term memory (LSTM) model to predict the state of charge (SOC) of lithium-ion batteries. The purpose of the research is to accurately model the complex nonlinear behavior that occurs during the charging and discharging processes of batteries to predict the SOC. The LSTM model was trained using battery data collected under various temperature and load conditions. To evaluate the performance of the artificial intelligence model, measurement data from the CS2 lithium-ion battery provided by the University of Maryland College of Engineering was utilized. The LSTM model excels in learning long-term dependencies from sequence data, effectively modeling temporal patterns in battery data. The study trained the LSTM model based on battery data collected from various charge and discharge cycles and evaluated the model’s performance by epoch to determine the optimal configuration. The proposed model demonstrated high SOC estimation accuracy for various charging and discharging profiles. As training progressed, the model’s predictive performance improved, with the predicted SOC moving from 14.8400% at epoch 10 to 12.4968% at epoch 60, approaching the actual SOC value of 13.5441%. Simultaneously, the mean absolute error (MAE) and root mean squared error (RMSE) decreased from 0.9185% and 1.3009% at epoch 10 to 0.2333% and 0.5682% at epoch 60, respectively, indicating continuous improvement in predictive performance. In conclusion, this study demonstrates the effectiveness of the LSTM model for predicting the SOC of lithium-ion batteries and its potential to enhance the performance of battery management systems.

According to the previous research, proper demand forecasting could help construction-related firms in effective planning for future market changes. However, existing market demand forecasting models are somewhat limited, and most of them bear some critical shortcomings. This research aims to develop a forecasting model for the Korean residential construction industry using system dynamics. In developing the market forecasting model, this research uses variables that significantly impact future construction market change. Many of the existing models do not include as many variables as this model, and none of them have considered complex interlocking effects among these variables. This model is also the first model using a system-based approach by looking at the target industry as a ‘one complex system’ rather than focusing on individual variables’ impact on future market changes. By employing system dynamics, it is possible to consider qualitative and quantitative aspects and produce long-term market forecasting results. The developed market forecasting model consists of two main modules, the first being a prediction module for the grassroots construction market and the second for operation and maintenance (O&M) and the demolition market. Sixteen input variables are grouped into four categories: social, economy, regulation, and past market size among over 25 identified variables. The model utilizes a mathematical function system using the designed feedback loops in producing future market forecasts. Based on the validation tests with past market data, it turns out that the model is reliable, with the determination coefficient (R2) being over 0.7 on all tested occasions. According to the model’s forecasting results, the Korean construction market’s size is expected to be 231 billion won in 2015 and 286 billion won in 2030. However, the O&M market’s growth rate is expected to be higher than 180%, which is much bigger than those of the grass-root and demolition markets. Thus, this research model is realistic according to the construction paradigm change. This research is considered one of the pioneering studies in construction market forecasting by employing dynamic inter-relationships among various input variables. Therefore, the market forecasting results can be interpreted as more practical and can provide more insights to the construction industry stakeholders. The model is envisioned to provide the public sector with useful guidelines in preparing future public market supply strategies such as construction budget allocations. It would also be helpful for the private sector to develop more proactive and accurate demand strategies for timely decision-making.

In this paper, we propose an intelligent battery control system that incorporates an active balancing technique and a fuzzy controller to manage and extend the life of an agricultural drone battery efficiently. The control system includes the following key features: A battery pack balancing algorithm based on a bidirectional DC/DC converter; a cell balancing system to prevent overcharge/discharge and ensure equal control of the cell voltages; and a monitoring system and wireless link to track the real-time status of the battery, temperature, and acceleration while the drone is in operation. Each battery pack consists of six lithium polymer batteries, one fuzzy controller per cell, and supports active balancing using the developed balancing technique. The capacity of each of the battery packs in the system is 11,000 mAh, and two can be combined to provide a total capacity of 22,000 mAh.