OPTIMA 2025

IIT Kharagpur, Kharagpur

Case Study: Energy Financial Planning

1.Problem Background
An organisation is planning to establish a data centre that requires a
continuous power supply of 100 MW per hour. In alignment with their
sustainability goals, the organisation aims to operate the data centre
primarily using renewable energy sources. Specifically, they intend to meet
70% of the energy demand through renewable sources, with the remaining
30% supplied by non-renewable sources.
The renewable energy sources under consideration include solar power
and battery storage systems charged by solar energy. For the non-renewable
portion, the organisation plans to use gas energy. Currently, there are no
existing energy sources at the proposed location for the data centre.
Therefore, the organisation needs to install new power plants for solar,
battery storage, and gas to cater to the energy demand.

Objective
The objective is to develop a comprehensive 25-year investment plan that
outlines the creation of a sustainable energy mix for the data centre. This
plan should integrate solar energy, natural gas, and battery storage to
ensure a reliable and cost-effective energy supply. The key goals are to:

 Determine the cost-effective mix of renewable and non-renewable

energy to fulfil power demand for each hour of the day.
 Prioritize the use of renewable energy to meet their sustainability goal.
 Establish the optimal capacity and deployment strategy for battery
storage systems to effectively balance supply and demand.

By achieving these objectives, the organisation aims to ensure the data

centre operates efficiently while adhering to its sustainability commitments
for upcoming 25 years.
 OPTIMA 2025
IIT Kharagpur, Kharagpur

2. Understanding the requirements and data

To ensure the development of a robust and sustainable energy plan for the
data centre, it is essential to break down the key factors and requirements.
This will help create a clear and actionable strategy. The primary goal is to
ensure the data centre operates efficiently and sustainably. The aim is to
meet the energy demand of 100 MW per hour using a mix of renewable and
non-renewable energy sources. The following requirements must be met to
achieve this goal:

2.1. Investment Analysis

 We need to evaluate the initial capital investment, maintenance costs,
and the levelized cost of energy (LCOE). This will help us determine the
most cost-effective investment strategy. The formula to calculate LCOE
is:

Installation∧operations cost for 25 years of all resources+ Land purchase cost

( 1+discount rate )1− year
Total power provided ¿ destination∈25 years
¿
( 1+discount rate )1− year

 Develop a scalable and versatile model capable of conducting scenario

analyses, such as evaluating the impact of adding new data centers at
the site and their future energy requirements.
2.2. Demand Fulfilment
 Energy Demand Fulfilment: It's crucial that the data centre's energy
demand of 100 MW per hour is met at all times. We must ensure there
are no shortages by optimizing the installed capacity.
 Energy Source Combination: The power required can be sourced
from a mix of solar plants, gas plants, and battery systems. We'll need
to consider their availability to ensure a reliable supply.
 OPTIMA 2025
IIT Kharagpur, Kharagpur

 Renewable Energy Proportion: The organization’s goal is to supply

70% of the data centre's energy needs from renewable sources (solar
and battery storage) and the remaining 30% from natural gas.
2.3. Battery constraints
 Battery Charging Time: The minimum charging time for batteries
depends on the technology used. For example, a 0.5C battery
technology will take minimum 2 hours to charge (1/0.5).
 Battery Charging and Discharging Cycles: Batteries will undergo a
full charge and discharge cycle, with a maximum of 2 cycles per day.
The maximum depth of discharge is 95%.
 Round-trip Efficiency: Consider the round-trip efficiency of the
battery system, which is based on the energy required to fully charge
the battery and the energy it provides when discharged.
 Constant Discharge Energy: The battery will discharge at a constant
rate in a cycle
 Battery Self-Discharge: The battery will self-discharge at a rate of
2% per hour after remaining fully charged for 4 hours without
additional charging.
 Battery Replacement: The solution should recommend the optimal
time for battery replacement. According to industry standards, the
battery should be replaced when its capacity drops below 70%
2.4. Solar constraints
 Solar Power Generation Availability: The availability of solar power
per MW installed solar capacity is provided for each hour of a day in a
year in section 3.
3.Data

This section outlines comprehensive details about various energy sources,

including their utilization patterns and degradation over time. Additionally,
the provided data should be used to calculate the power generation and net
present value (NPV) for each energy source.
 OPTIMA 2025
IIT Kharagpur, Kharagpur

Table 1. Details of Solar Energy

Parameter Value Unit Definitions

Plant Life 20 years No. of years the plant

will be operational

Plant CUF 0.3 % Total generation capacity

of an installed plant in a
given day. For e.g.
Suppose we have
installed a 100 MW
system. This means the
average output for any
given hour will be 20
MWh.

Salvage Value 0.01 % Value of the solar asset

at the end of 25 years

Annual Plant 0.01 % Annual degradation in

degradation the output of the plant
due to degradation of
solar panels.

O&M Growth Rate 0.01 % Yearly growth rate of

operations and
maintenance cost- takes
into account inflation

Discount Rate 0.05 % Assumed weighted

average cost of capital
i.e. rate used to calculate
the present value of all
input costs (capex +
opex)

Land requirement 10 Acres/MW

 OPTIMA 2025
IIT Kharagpur, Kharagpur

Capex 800 $/kw

Opex 0.03 % of
Capex

$/MW Annual opex

Land availablity 1000000 Acre Max land available

0.25 percentag Expected percentage

max_solar_out_dev_exp e deviation in input solar
td profile

land_purchase_cost 2000 $/Acre Cost of per acre land

Table 2. Details of Gas Energy

Parameter Valu Unit Definitions

e

Plant 20 years No. of years the plant will be operational

lifetime

Heat Rate 0.00 MMBTU/ Amount of fuel required to produce 1 kwh

55 Kwh of power

Annual 0.01 % Annual decrease in power output due to

Degradation wear and tear of equipment

O&M Growth 0.01 % Yearly growth rate of operations and

rate maintenance cost- takes into account
inflation

Discount 0.05 % Assumed weighted average cost of capital

Rate i.e. rate used to calculate the present value
of all input costs (capex + opex)
 OPTIMA 2025
IIT Kharagpur, Kharagpur

Salvage 0.01 % Annual degradation in the output of the

Value plant due to degradation of solar panels.

Capex 900 $/kw

Opex

FC 1.2 $/
MMBTU

FC2 0.01 $/Kwh

43

Other Opex 0.1 % of Annual Opex

Capex

$/kw

Table 3. Details of Battery

Parameter Value Unit Definitions

Cyclic 0.000 % % Decrease in battery output with every

Degradation 076 cycle

Round Trip 0.88 % Total discharge output of a battery in every

Efficiency cycle

Capex 141

Opex 0.02 % of Annual

Capex

deg_rate_per 0.000 Per hour degradation in battery if not

_hour 5 charging
 OPTIMA 2025
IIT Kharagpur, Kharagpur

batt_charge_ 0.5 Type of battery ex 0.5C

rate

Max_deg_all 0.7 Max allowed degradation after which

owed battery system needs to be replaced

max_cycles_ 5500 Max cycles a battery can sustain before

allowed replacement

battery_cost 0.03 Annual percentage degradation in cost of

_ann_deg battery

O&M Growth 0.01

Rate

Table 4. Demand

Parameter Value Unit Description

1000
Demand Kw Constant Demand at every hour
00

ren_energy_s 0.75 Minimum percentage share of renewable

hare_frac energy in the overall energy demand

Table 5. Solar Availability

Please find attached the excel sheet.

P.S. Please make appropriate assumptions for any missing parameters and
clearly state all your assumptions along with the rationale behind them.
While we are interested in the final solution, our primary focus is on your
approach to solving the problem. All data mentioned above should be
parametrised in the model, not hard-coded. Preferable, develop the model
using open-source solver packages. Your approach will be evaluated based
on adherence to software engineering practices, code organization, and
documentation.