Tigo Energy announced that Ukrainian utility YASNO has begun using the company’s Predict+ forecasting platform. The AI‑driven solution is intended to help the utility manage demand spikes, volatile weather and infrastructure disruptions, which are critical concerns for operators of large, mixed‑generation grids. Ukraine’s power system faces a uniquely demanding environment: frequent temperature swings, rapid changes in wind and solar output, and the risk of physical damage to transmission assets due to ongoing conflict. In this context, a forecasting tool that can ingest real‑time sensor feeds, blend multiple weather models, and adjust instantly to unexpected grid events becomes a strategic asset. Predict+ promises to turn those challenges into manageable variables, giving YASNO the ability to keep lights on for millions while optimizing the balance between renewable and baseload generation.
YASNO Adds Predict+ to Its Grid‑Management Toolkit
YASNO, the leading provider of electricity, gas, and energy‑efficiency services to more than 2.5 million households in Kyiv, Dnipropetrovsk and Donetsk, as well as over 64 000 business customers, is the latest enterprise‑tier customer of Tigo’s Predict+ platform. The rollout follows a successful configuration and testing pilot in the Dnipropetrovsk region—a market singled out in the source material for its “diverse customer base and highly variable weather conditions.” During the pilot, Tigo engineers integrated live data streams from YASNO’s smart‑meter network, historical consumption records, and weather forecasts from several providers. The platform then generated hourly demand scenarios that were continuously refined as new data arrived.
Predict+ currently manages more than 650 GWh of forecasted energy and delivers 97.5 % forecast accuracy for utility customers, according to Tigo. The neural‑network AI model ingests a wide array of inputs: real‑time weather forecasts, current weather observations across regions, historical hourly consumption and generation data, and operational data from previous days. By analyzing these sources simultaneously, the system can produce multiple forecast scenarios that are updated as temperature, wind speed, or solar irradiation change. This dynamic approach is especially valuable in Dnipropetrovsk, where rapid weather shifts and occasional infrastructure damage have historically complicated load forecasting.
“The broader implementation of this platform will continue to improve the accuracy of our hourly electricity demand forecasts and reduce imbalance settlement costs. This means better planning, more efficient resource utilization, and an even more reliable service for our customers,” said Olena Senkina, Head of Electricity Department at YASNO.
How Predict+ Fits Into YASNO’s Operational Model
Predict+ is designed to work with both smart‑metered and non‑smart‑metered loads. When smart meters are present, the platform models each meter individually; without them, it still produces high‑accuracy demand predictions. Its functional domains span market insights, customer insights, profit analysis, regulatory support and real‑time integration of energy spot‑market pricing.
Archie Roboostoff, vice president of software at Tigo, explained that YASNO required the platform to incorporate “disruptions for core grid elements” such as damage to infrastructure. The Predict+ neural network can rapidly ingest these additional variables and adjust forecasts accordingly, making the system “vastly more predictable” for the utility.
The platform’s broader availability includes utilities, energy retailers, traders, independent power producers and large commercial‑industrial customers across the United States and Europe.
Operational Relevance for Enterprise Energy Managers
For enterprise energy leaders, Predict+ offers a concrete method to align renewable and baseload generation while accounting for real‑time grid stressors. By modeling each meter—or aggregating non‑metered loads—the system can support detailed profit and regulatory analyses, which are essential for utilities operating under tight settlement rules. YASNO’s stated goal of reducing imbalance settlement costs directly ties forecast accuracy to financial outcomes.
The platform’s continuous scenario updating, driven by changing temperature, wind speed and solar irradiation, helps operators anticipate short‑term supply‑demand gaps. This capability is particularly relevant for regions like Dnipropetrovsk, where weather variability and infrastructure damage have historically complicated load forecasting.
Key Takeaways
- YASNO has begun deploying Tigo Energy’s Predict+ platform after a pilot in the Dnipropetrovsk region, with broader rollout plans underway.
- Predict+ manages over 650 GWh of forecasted energy and reports 97.5 % forecast accuracy for utility customers.
- The platform supports both smart‑metered and non‑smart‑metered loads and integrates weather, historical consumption and operational data to produce continuously updated hourly demand scenarios.
TechInsyte's Take
The deployment shows how AI‑based forecasting can be integrated into a utility’s existing data ecosystem to address Ukraine’s unique grid challenges. While YASNO highlights expected reductions in imbalance settlement costs, the actual financial impact will depend on the platform’s performance at scale and the stability of data inputs. Enterprise energy leaders should monitor YASNO’s rollout progress and any measurable changes in forecast accuracy or settlement savings before committing to similar solutions.
Source: Businesswire
