Page 213 - 2024-Vol20-Issue2
P. 213
209 | Gochhait, Sharma & Bachute
short-term load forecasting, setting the stage for the subse- A. About the OPTCL and NRLDC Load
quent analysis. In Section III, the data is thoroughly examined
and the problem formulation is presented to establish a solid The research paper involves an in-depth analysis of electricity
foundation for the study. The proposed BI-LSTM model for load data for two regions:Odisha and Delhi. For Odisha, the
1D CNN is then detailed in Section IV, presenting a novel study examines the monthly electricity load data from January
approach to tackle the forecasting challenge. Lastly, Section 2019 to December 2022, with measurements taken at 1-hour
V presents a thorough analysis of the obtained results and intervals. The objective is to calculate the electricity load
initiates a meaningful discussion surrounding them. demand in megawatts (MW) and investigate the seasonal vari-
ations in consumption. The findings demonstrate distinct load
II. FORECASTING METHOD PROPOSAL demand patterns across four seasons: Spring (March to May),
Summer (June to August), Fall (September to November), and
Winter (December to February).
Efficient power system management relies on a comprehen- B. Load Analysis for Odisha
sive understanding of various factors, including weather condi- The research paper involves an in-depth analysis of the monthly
tions, economic conditions, and appliance usage, to accurately electricity load data for Odisha from January 2019 to De-
estimate the power demand in a district. Fluctuating loads cember 2022, with measurements taken at 1- hourintervals.
resulting from unstable patterns can pose challenges for distri- The objective is to calculate the electricity load demand in
bution companies. However, previous studies have shown that megawatts (MW) and investigate the seasonal variations in
forecasting aggregated power load is relatively easier and can consumption. The findings demonstrate distinct load demand
help in smoothing load shapes with moderate to low RMSE patterns across the four seasons: Spring (March to May),
errors. Summer (June to August), Fall (September to November),
In this research paper, we present the findings of a compara- and Winter (December to February).The insights gained from
tive long-term electricity forecasting analysis conducted in the this study hold immense value for Odisha Power Transmis-
states of Orissa and Delhi. Using data from the OPTCL and sion. Corporation Limited (OPTCL) in effectively managing
NRLDC, we aim to forecast longterm load levels for individ- the electricity load during different seasons. The study re-
ual distribution companies (discoms) in these states. Consid- sults can assist OPTCL in accurately predicting the electricity
ering the extensive number of consumers in the OPTCL and demand in Odisha during each season and implementing nec-
NRLDC databases, it is not feasible to con- sider all of them. essary measures to ensure an uninterrupted power supply to
Therefore, we select a subset of the OPTCL and NRLDC customers. By leveraging this information, OPTCL can en-
datasets to develop our load forecasting approach and derive hance the overall reliability and efficiency of the power grid
meaningful insights. By analyzing the selected dataset, we in Odisha.
provide a comparative analysis of long-term electricity fore-
casting in Orissa and Delhi, highlighting the differences and C. Load Analysis for Delhi
similarities in load patterns, seasonal variations, and forecast- Similarly, the research paper also investigates the electricity
ing accuracy. The proposed load forecasting method offers load data for Delhi ona monthly basis, from January 2019 to
valuable insights for efficient power system management and December 2022, with measurements taken at 24-hour inter-
can aid in enhancing the reliability and efficiency of the power vals. The objective is to calculate the electricity load demand
grids in both states. in MWand analyze the seasonal variations in consumption.
The primary objective of this research is to identify and assess The results reveal distinctive load demand patterns across the
the electricity consumption requirements within a particular four seasons: Spring (March to May), Summer (June to Au-
sector of the utilities industry in order to estimate the over- gust), Fall (September to November), and Winter (December
all power demand for that sector. A significant aspect of to February). The insights gained from this study have sig-
this study revolves around utilizing the data collected during nificant implications for effectively managing the electricity
the project for shortterm forecasting of the electric load lev- load in Delhi throughout the year, considering the seasonal
els of individual distribution companies (discoms). In this variations. The study findings can be utilized to accurately
section, we present pertinent background information on the predict the electricity demand during each season and imple-
electricity load patterns observed in Odisha and Delhi states. ment appropriate measures to ensure a consistent and uninter-
Additionally, we introduce our proposed approach for load rupted power supply to customers. By doing so, the overall
forecasting, which aims to address the forecasting challenges reliability and efficiency of the power grid in Delhi can be
in this context. improved. The insights gained from this study hold immense
value for Odisha Power Transmission Corporation Limited
