The Backwards Induction Algorithm for Energy Investments

A decision framework for Energy investments

What you'll learn
Contents: We go through the Backwards Induction Framework for Energy Investments
This is a decision framework under uncertainty
You will learn about development of frameworks
Code analytics and uncertainty
Requirements
You must install Matlab. Other than this :There are no prerequisites! We build everything from scratch!
Description
Investments in Energy infrastructure are very expensive. Therefore, many frameworks have been developed for coming up with optimal investment decisions. The problem of selecting optimal investment decisions becomes more challenging when the future is uncertain. A solution to this challenge is the Backwards Induction Algorithm, which is a framework that can yield optimal investment decisions under uncertainty. In these videos, we go through the philosophy of this algorithm and then we build the MATLAB code step by step that simulates this algorithm.
All the calculations are performed manually and also using MATLAB. This allows a full understanding of the algorithm.We are also conducting a sensitivity analysis around key parameters. We apply the problem to an electricity grid, which is about to receive a large number of Electric Vehicles in the future. In other words, the consumers of this grid currently do not have Electric Vehicles but will soon buy. Then these vehicles will have to charge with electricity, thereby increasing the overall electricity demand, and mainly its maximum value (also known as the "peak"). However, the exact number of Electric Vehicles is currently uncertain and can even be zero. In other words, there is uncertainty around the number of Electric Vehicles that will connect to the system in the future. This is a problem that almost all electricity grids face.
If the number of Electric Vehicles that will connect is large, then we will need to invest in the electricity grid by increasing the capacity of the existing lines and/or by installing smart chargers. When electric vehicles connect to the electricity grid, the electricity demand increases. If the Electric Vehicles charge using "smart chargers" then the demand will still increase but this increase will be smaller than without smart chargers. The Backwards Induction algorithm can tell us what investment is optimal to do among all possible candidate investments. For example: is the best investment strategy to invest in smart charging equipment across the electricity grid? Or is the best strategy to invest in upgrading the lines of the electricity grid?Or maybe, is the best strategy to do nothing at all?
The Backwards Induction framework uses a scenario tree approach. We first calculate the total costs (investment costs + operational costs) in the last stage, and then we work our way backwards, using probabilities, until the current time. And then we select the investment strategy that minimizes the total expected costs (i.e. the total probability-weighted average costs). We also find the "Option Value" of Smart Charging.
This is the economic benefit that we get by investing in smart charging. In other words, this means that if we invest in lines is much more expensive than if we invest in both lines and smart charging. In electricity grids, there is always a debate between investing in "smart technologies" (like smart charging) versus investing in traditional / mainstream / conventional technologies (like upgrading the capacity of existing lines). Usually, the optimal thing to do is a mix of the two: to invest in both smart charging and upgrade of lines. And this is actually what the Backwards Induction framework yields in these studies. However, there are cases where it is optimal to invest only in smart charging or only in lines. These concepts are all analysed in this course.
Who this course is for:
Investment Bankers & Portfolio managers
Data Scientists
Quantitative developers
Management



More Info