1. Introduction
The increasing adoption of microgrids with renewable energy systems, driven by environmental and socioeconomic factors, faces challenges such as renewable energy variability and dynamic load fluctuations, leading to increased grid consumption. This study addresses these challenges by proposing an advanced Energy Management System (EMS) integrated with a Deep Learning model for load forecasting. The objective is to enhance the efficiency and cost- effectiveness of microgrids by dynamically adjusting to forecasted load demands. The EMS utilizes Long Short-Term Memory (LSTM) networks to predict the load demand of a commercial building, allowing for optimized battery scheduling and reduced reliance on the utility grid. The study conducted a month-long simulation using real historical load and solar power data, comparing the proposed EMS with a standard EMS. Key findings indicate that the proposed EMS significantly reduces grid consumption, resulting in a 9.3% reduction in monthly electricity bills. Integrating deep learning in EMS demonstrates substantial improvements in handling dynamic conditions and optimizing energy usage. These findings imply that deep learning-based EMS can lead to significant cost savings and more efficient microgrid energy management, promoting the broader adoption of renewable energy solutions.
2. Objectives
- To develop load forecasting of commercial building using deep learning model.
- To implement an energy management system microgrid integrated with deep learning load forecasting.
- To evaluate the performance of proposed energy management system by comparing the electricity bill with standard energy management system.
3. Methodology
3.1 Proposed Microgrid Design
The microgrid's proposed block diagram is shown below. The PV panel generates power from sunlight, stored in the Battery Energy Storage System (BESS). The BESS acts as the main supplier, providing power to the load. The utility grid supplies the remaining power if the power stored in the BESS does not fulfill the load demand. A Deep Learning Energy Management System (DL EMS) is integrated to analyze current power conditions and forecast load demand, deciding optimal strategies for charging and discharging the battery in the microgrid, thereby reducing power consumption from the utility grid.
3.2 Proposed EMS Algorithms
Two algorithms proposed are EMS strategies and EMS with Load Forecasting algorithms. Figure below is EMS strategies to ensure that load demand is fulfilled which also abides SOC battery constraints. It determines whether to charge or discharge the battery or to supply power from the grid to the load. to avoid overcharge and over-discharge.
3.3 Proposed EMS Algorithm with LSTM Load forecasting
Fig. below illustrates the proposed EMS integrating load forecasting. The system collects daily load forecasts and identifies the time of the highest predicted load. During peak periods, it prioritizes charging the battery before the next expected peak period. This approach ensures sufficient battery power to meet the load demand during peak times. The EMS also considers battery SOC constraints, charging the battery if the SOC is below SOCmax or discharging if it exceeds SOCmin , and stops charging when SOC reaches SOCmax.
4. Result
Fig. below illustrates the comparison between the total costs of the proposed EMS and the standard EMS. The red circles highlight the days when the total cost of the proposed EMS exceeds that of the standard EMS. There is a total of 9 days where the proposed EMS incurs higher costs than the standard EMS. However, this result is acceptable because there are more days where the proposed EMS's costs are significantly lower, leading to overall savings.
Figure below shows the daily costs for the proposed EMS and the standard EMS during peak and non-peak hours over a month. During non-peak hours, the costs are identical for both EMS because both systems prioritize battery charging during these times due to the lower electricity prices. However, during peak hours, the proposed EMS incurs significantly lower costs than the standard EMS, with a difference of RM12,679.88. Overall, the total monthly cost is calculated for both systems, demonstrating that the proposed EMS achieves a 9.23% reduction in total costs compared to the standard EMS.
