Terra is a software platform designed to improve the profitability of renewable energy asset owners through its superior grid-wide modeling, forecasting, insights, and bid recommendations. Terra combines the advanced capabilities of artificial intelligence (AI) with traditional physics-based grid modeling and trading optimization into one comprehensive software solution for Gaiascope clients.
Under management
Added to clientsbottom line
Data points forecasted
Terra’s accuracy and performance are tested and proven every hour of every day through Gaiascope’s managed trading fund, which generated 47% gross returns in 2023 as a direct result of Terra’s capabilities.
Terra is currently available for renewable energy asset owners in ERCOT with expansion into MISO and CAISO in 2024.
Overnight and into the morning, automatically ingesting, cleaning and parsing a myriad of raw data files across data types into our databases, amounting to approximately 3 million data points related to weather, grid topology and transmission, load, generation, outages, electricity markets and other signals. This dataset has more than doubled each y ear as we identify new sources to improve our forecasts.
For any datapoints that are predictive, such as temperature forecasts, calculating the implicit uncertainty within said forecast, creating a range of probable scenarios . For example, if the temperature forecast for San Antonio is 72 degrees Fahrenheit at 3 PM CST, Terra might determine the probable range for this hour is actually 69 to 73 degrees Fahrenheit. Or as another example, Terra might determine the probability of an outage at natural gas-fired generator on the Gulf Coast at 3 PM CST is 30%, resulting in 30 of 100 of its scenarios including an outage at this generator, and 70 scenarios assuming the generator is online.
Building up a digital replica of how the electric grid will be configured at each hour, taking into account estimated location, specifications and interconnections of over 455,000 bus, transformers and lines on the grid, including any potential outages forecasted for that hour, such as downed power lines. Terra calculates every shift factor, load and branch outage distribution factor, resulting in a comprehensive understanding of how power will flow for any change on the grid, such as inflections in individual generator output. This allows Terra to forecast changes to electricity prices for essentially infinite potential scenarios on the grid. The key, of course, is determining which scenarios are most probable.
Building up a digital replica of how the electric grid will be configured at each hour, taking into account estimated location, specifications and interconnections of over 455,000 bus, transformers and lines on the grid, including any potential outages forecasted for that hour, such as downed power lines. Terra calculates every shift factor, load and branch outage distribution factor, resulting in a comprehensive understanding of how power will flow for any change on the grid, such as inflections in individual generator output. This allows Terra to forecast changes to electricity prices for essentially infinite potential scenarios on the grid. The key, of course, is determining which scenarios are most probable.
Running each scenario of generation and load forecasts through the digital replica of the grid for each hour to generate 100 scenarios of what grid operations will look like.
Running contingency analysis, choosing a subset of more than 2,700 potential contingencies each hour, in order to determine the probability of congestion at each bus driven by forecasted constraints expected to occur on the grid. This results in an understanding of the potential impact of shadow prices and ultimately what congestion price impact will be at every node, hub and load zone in ERCOT.
Determining the probability and magnitude of price adders to grid-wide pricing.
Determining the expected bid/offer behavior of the entire generation stack in ERCOT, using 100 scenarios for each hour to forecast generation dispatch and grid-wide system lambda pricing. This process is underpinned by a deep understanding of individual generator bid/offer behavior and how it shifts based on different input costs and and net load expectations.
Flowing system lambda, price adder and congestion price forecasts into individual strategies and risk management parameters specified by the client, in order to generate bid/offer recommendations that reflect the client’s revenue goals, risk policy and trading and dispatch constraints. Bid/offer recommendations are integrated into the client’s workflow to ensure timely delivery for evaluation and submission to ERCOT.
Flowing inputs and outputs into visualizations for the client to gain an understanding of key drivers and forecasts generated by the Terra platform.
The Terra platform, including all pipelines and back-end infrastructure used to run its massive processes in parallel, in the cloud and within a short timeframe to ensure timely delivery to clients, is maintained by a full-time, 7-person technical team to monitor all live processes in real-time, 365 days a year, tasked with addressing (within minutes or even seconds) any issues that threaten to result in downtime, and streamlining the existing codebase to ensure the most efficient runtime possible. Greater efficiency allows for more data to be ingested and more complex analysis and forecasting, resulting in higher accuracy of price forecasts.
Take the first step to realizing the returns possible from your assets. Schedule a Backtest to discover just how much revenue Gaiascope and the Terra Platform can unlock for you.