⚡ PPA Class – Renewable Energy Contract Modeling

🎯 Purpose

The PPA class provides a robust framework for modeling Power Purchase Agreements (PPAs) and managing Renewable Energy Source (RES) portfolios. It enables:

Simulation of renewable generation profiles
Quantification of production risks (e.g., p50, p90)
Evaluation of market capture and revenue potential
Valuation of contracts under different pricing structures

🏗️ Class Structure

Each PPA instance includes the following key attributes:

Attribute	Description
`id`	Unique identifier for the PPA
`site_name`	Name of the renewable generation site
`start_date`	Contract start date
`end_date`	Contract end date
`capacity`	Installed capacity in megawatts (MW)
`techno`	Technology type (`solar`, `wind`, etc.)
`pricing_type`	Pricing model (`fixed`, `floating`, etc.)
`country`	Country of operation
`proxy`	Time series representing simulated or historical past generation
`mark`	Time series representing simulated future generation
`p50`	Expected annual generation at a 50% probability level (MWh)

🧮 Method: `buildProxy()`

The buildProxy() method constructs a generation proxy for the asset. It can be built from:

Historical data: Cleaned and time-aligned actual production records
Synthetic data: Modeled generation using weather inputs and power curves

🔍 Why It Matters

The timing of generation is just as important as the volume.

A plant producing during peak price hours generates more revenue than one producing during low-price periods. The proxy enables:

Estimation of expected volumes (e.g., p50)
Backtesting revenue capture using historical price curves
Comparison of pricing scheme impacts (e.g., fixed vs floating)

Figure: Examples of WIND and SOLAR generation proxies

To construct a proxy, a power curve must first be defined for the asset. The full methodology is documented here:

🔗 🌞🌬️ Building Solar & Wind Power Curves

Once the power curve is established, it is applied to the hourly weather history of the site to simulate a historical generation profile — referred to as the proxy.

📈 Method: `buildMark()`

The buildMark() method constructs a generation mark for the asset. It is derived from the proxy time series and corresponds to the average volume per month and per hour. This provides the expected generation profile of the asset.

This time series is a cornerstone for PPA management, as it establishes the P50 — the expected volume — and, when combined with the forward price curve, it enables Mark-to-Market valuation by projecting the expected cash flows of the asset.

🔗 📈⚡ Future Day-Ahead Power Curve Documentation
🔗 📈⚡ Forward Power Curve Hourly Documentation

Figure: MARK WIND