A Forecasting Method of Peak-Cut of Power Demand Using LSTM at A Clinic
- Resource Type
- Conference
- Authors
- Inagata, Tomoya; Mizuno, Yuji; Matsunaga, Keita; Kurokawa, Fujio; Tanaka, Masaharu; Matsui, Nobumasa
- Source
- 2023 12th International Conference on Renewable Energy Research and Applications (ICRERA) Renewable Energy Research and Applications (ICRERA), 2023 12th International Conference on. :1-6 Aug, 2023
- Subject
- Components, Circuits, Devices and Systems
Computing and Processing
Engineering Profession
Power, Energy and Industry Applications
Renewable energy sources
Power demand
Correlation
Costs
Load forecasting
Hospitals
Predictive models
load forecast
LSTM
clinic
- Language
- ISSN
- 2572-6013
Energy prices are rising due to the changing global situation. Rising energy prices have also led to higher electricity prices in Japan. Electricity prices is determined by the contracted power for the past year. Hospitals and clinics would like to reduce electricity costs with decreasing contracted power. They recently have a combination of diesel generators (DGs) and PV for their power systems. DGs and PV can be used for peak-cut. Therefore, DGs and PV has an important role for peak-cut operations. This paper proposes a forecasting method of peak-cut of power demand using LSTM at a clinic. Four cases based on the correlation of the demand data are defined as input data for LSTM in this paper. The results show that Case 3 and Case 4 is a better model on the point of forecasting peak demand. Power demand data for last three months produced better results than using five years of power demand data.