This AI service assigns a quantitative efficiency score to each municipal building by comparing its actual energy performance to similar buildings in the city. The score reflects how efficiently a building uses energy relative to its size, use, and category.

The PV Self-Consumption Optimization service uses AI to maximize the use of locally generated solar energy within municipal buildings. By analyzing real-time consumption, historical PV production and energy consumption data, it identifies optimal load-shifting strategies to increase solar self-consumption and reduce reliance on grid electricity. The service recommends adjustments in building operations and controllable loads to align energy usage with PV generation peaks. This helps municipalities lower energy costs, improve sustainability performance, and enhance building energy autonomy.

Building Energy Consumption Forecasting service uses historical energy bills, building characteristics, real-time data from smart meters and seasonal patterns to forecast electricity consumption for municipal buildings. Leveraging machine learning models, it provides accurate short-term and mid-term forecasts. This service helps municipalities anticipate energy needs, optimize procurement, and identify potential inefficiencies before they escalate, supporting smarter and more sustainable energy management across the building portfolio.

Currently the DSOs have multiple requests for 5-20 MW battery energy storage system (BESS) on the 20kV feeder in the primary transformer station. There exists a potential of activating multiple disturbing elements at once, which in turn can affect the TSO network frequency and voltage. Flexibility management between TSO and DSO networks can be achieved by modelling the DSO network and simulate the effects of distributed energy resources (DER) and BESS on the DSO network and in turn on the TSO network. This data can be used in a system that notifies TSO operators about the potentially disturbing events or even the system itself actively controls the deployed BESSs.

SCADA systems produce massive volumes of event logs during both planned and unplanned outages, making it difficult to isolate meaningful signals. This service uses machine learning and GenAI to analyze logs in real time, filtering out noise and highlighting the most informative events. It helps identify likely root causes and generates concise summaries to assist operators during incidents. By accelerating root cause analysis and reducing cognitive load, the tool enables faster response and recovery, improves operational efficiency, and enhances overall system reliability.

This service will leverage AI technologies for evaluation of the grid stability, based on the grid state graphs and evaluation of likelihood of the short circuit events in the neighbourhood of specific node.

This service will leverage AI technologies for rapid detection, localisation, and classification of faults in transmission grids. The analysis will be based on recordings from protection relays, which are essential for accurate event analysis and verification of relay operation. By applying AI to these disturbance records, the service will enable faster understanding of fault causes and assess the correctness of protection system responses. This contributes to improved reliability, faster decision-making, and greater resilience of the transmission network in the face of unexpected disturbances.

This AI-powered service evaluates the performance of energy forecasts by comparing predicted and actual energy outputs of offshore assets. It supports energy analysts and system operators in identifying forecasting errors and improving future models. The service provides critical KPIs like MAE, RMSE, and deviation trends over time.

This AI-based service enables detection and localization/identification of faults in offshore renewable energy systems. It processes real-time sensor data from offshore assets (e.g., subsea cables, power electronics, buoys) to identify faults in voltage, current, power and operational behaviour. The system supports prioritization of critical events, enabling timely dispatch of maintenance teams and helping minimize operational downtime and service disruption.

This service classifies and manages the operational states of offshore renewable energy assets. It distinguishes between normal generation, derated (reduced output), and safeguard (rest or idle) states using real-time ecosystem data. The AI model can trigger control actions to place the system into safe modes or detect idle conditions. This improves operational efficiency, asset lifetime, and ensures safety under abnormal or maintenance conditions.

This AI-driven service continuously analyses the frequency and patterns in incoming observations from offshore renewable energy systems. It predicts the likelihood of component failure by correlating real-time and historical data, enabling predictive maintenance strategies. This reduces unexpected downtimes and extends the operational life of offshore assets. The system can also prioritize assets based on risk levels and maintenance urgency.

This service leverages advanced AI techniques to forecast energy production from wave power plants. By integrating real-time sensor data, historical generation records, and weather forecasts, the AI models produce short-term (0-48 hours ahead) predictions. The forecasts support operational planning, power trading and grid balancing. Ultimately, this service aims to enhance plant efficiency and optimize energy trading strategies.

This service provides accurate forecasts of wind power generation for a wind farm. It utilizes historical SCADA data, including wind speed, wind direction, turbine operating parameters, and potentially weather forecasts, to predict future power output. The forecasts support optimized energy trading, grid integration, and operational planning.

This service analyzes historical inverter and string data (current, voltage, temperature, power output) to predict potential inverter failures and recommend optimal spare parts procurement strategies. By identifying failure patterns and estimating time-to-failure, the AI models help minimize inventory costs, reduce downtime, and ensure timely availability of critical components.

This service quantifies energy losses in PV systems due to soiling (accumulation of dust, dirt, etc.) on the panels. By analyzing historical data, including PV generation and soiling sensor measurements, the AI models estimate the percentage of power reduction caused by soiling. The service provides insights into the optimal timing of cleaning interventions, maximizing energy yield and minimizing operational costs.

This service optimizes pumpback operations between a Hydro Dam and an adjacent Hydro with storage. By analyzing historical and real-time water level data from the HDAM, along with relevant weather information, the AI system generates optimal pump dispatch signals. This enables efficient water transfer to maximize energy storage and generation at the storage hydro plant while preventing overflow or shortages within the system.

This service employs advanced AI models to predict day ahead energy generation at hydropower plants with daily granularity. Utilizing a historical dataset spanning 2013-2023, encompassing generation and water level data from HDAM, the system forecasts future power output.

This AI service aims at optimizing the flow of resources within a supply chain network*. The sustainability is added on top of more standard attributes, such as time-related metrics, with the help of KPIs such as energy efficiency / consumption and/or CO2 emissions. The input is related to the network structure and the alternatives of the input variables. The output is a good (near optimum) solution for the flow of the resources

With the help of this AI services bundle (X = a specific manufacturing process*), the behaviour of a manufacturing process can be modelled. As such, both the operation and the design of a manufacturing line can be facilitated, as optimization scenarios, as well what-if scenarios, can be considered. The input and the output comprises the profiles of the engaged variables in time. Sustainability can thus be optimized either by Energy efficiency directly, or through considering defects reduction workflows.

This is an AI service that can be used optimize the Process Plan* of a manufacturing plant. This implies the selection of the processes and their sequence, as well as the process parameters’ values. The input comprises of the part-related information, the machines availability and their attributes. The output is the process plan for a given plant and a given part. Energy efficiency will be one of the criteria used in decision making, along-side other manufacturing attributes, such as time, cost and quality.

This service utilizes AI to optimize the scheduling* of a production in manufacturing. The input comprises the orders, the machines available and their data. The output envelopes the schedule plan of the manufacturing itself. Energy efficiency will be one of the criteria used in decision making, along-side time-related criteria. Real-time data on the orders could potentially be used to extend this service towards being an adaptive scheduler.

This AI-powered energy management service optimizes power flow in Fuel Cell Hybrid Electric Vehicles (FCHEVs) by intelligently balancing energy use between the fuel cell and battery. Using real-time and historical driving data, it enhances vehicle efficiency, extends component lifespan, reduces hydrogen consumption, and ensures reliable performance across diverse driving conditions.

This AI-powered service enables intelligent control of hydrogen technologies within microgrids by coordinating real-time operation of electrolyzers, fuel cells, and other energy storage systems. It ensures optimal energy flow between hydrogen and electrical systems, enhancing efficiency, reducing renewable intermittency, and supporting grid stability.

This service provides predictive analytics for fuel cell systems using data-driven models. By analyzing historical and real-time operational data, it forecasts performance degradation and identifies potential failures in advance. The goal is to enable proactive maintenance, reduce downtime, and extend the lifespan of fuel cell systems through intelligent, real-time monitoring and decision support.

This AI-based service orchestrates electric vehicles (EVs) and other flexible assets to deliver grid-supportive ancillary services and local demand optimization, targeting CO₂ emissions reduction or economic savings. It dynamically schedules EV charging and discharging based on real-time market signals, carbon intensity forecasts, and user constraints. The service enables communities and fleet operators to activate EV flexibility for frequency response, peak shaving, and self-consumption enhancement while maintaining mobility needs. Designed for integration with TSOs/DSOs and community EMS platforms, it supports automated participation in flexibility markets and low-carbon grid services.

This service simulates and tests AI-enhanced multi-agent systems for Vehicle-to-Grid (V2G) applications. It enables experimentation with decentralized coordination strategies, agent decision-making, and market participation for fleets of EVs offering grid services.

This AI-driven Energy Management System (EMS) optimizes the charging and discharging of stationary batteries and electric vehicles to maximize on-site PV self-consumption in residential and community energy systems. By forecasting solar generation, household demand, and EV usage patterns, the service intelligently schedules energy flows to reduce grid dependency, avoid peak tariffs, and support local flexibility services. It considers dynamic constraints such as vehicle availability, user preferences, and real-time market signals, enabling the seamless integration of EVs as mobile storage units within the broader EMS strategy.

This AI service provides high-resolution short- and medium-term forecasts of wind power generation within EV-integrated energy communities. By combining localized wind turbine data, meteorological inputs, and grid interaction models, it enables the coordinated use of wind energy for EV charging, battery storage, and household consumption. Forecasts are used to align flexible EV charging with expected wind generation, minimizing grid imports and enhancing system-level efficiency. The service supports community-level energy balancing, demand-shaping, and optimal use of distributed renewable resources under variable wind conditions.

This service delivers high-resolution forecasts of solar PV generation tailored to the specific configurations of energy communities, including individual rooftops, shared PV systems, and EV-coupled infrastructure. Using machine learning models trained on historical generation data, weather forecasts, and system metadata (e.g., tilt, azimuth, shading), it provides short- and medium-term predictions. The forecasts enable intelligent scheduling of battery charging, EV loads, and peer-to-peer energy trading. Designed for real-time operations and long-term planning, it supports enhanced self-consumption, grid-aware dispatch, and community-level flexibility management.

This AI-powered service forecasts electricity demand in energy communities with a strong focus on EV charging behavior. By integrating smart meter data, EV usage patterns, household profiles, and external variables like weather and tariffs, the service predicts aggregate and disaggregated demand across time horizons. The forecasts support smart scheduling of EV charging, grid impact mitigation, and alignment with local renewable generation. Designed for both operational and planning use, this tool enables energy communities to maximize self-consumption, reduce peak loads, and coordinate flexibility offerings from EV fleets.

This service forecasts day-ahead charging behavior and energy demand patterns of private and public EV users based on historical usage data, contextual information (e.g., location, time-of-day), and exogenous variables such as weather. The model enables proactive load balancing, infrastructure planning, and tailored incentives by DSOs or aggregators.

Digital Twin for DHCN Optimization is an AI service that combines artificial intelligence with Digital Twin technology to optimize district heating/cooling networks. It creates virtual simulations of the network to analyze and improve production scenarios, helping managers make better operational decisions.

Energy demand forecasting is a critical AI service that predicts future energy consumption patterns in DHCN networks. It uses historical data, weather forecasts, and building usage patterns to optimize energy distribution and reduce operational costs while maintaining user comfort levels.

The energy consumption optimization service applies AI algorithms to optimize energy consumption at both district level (boilers room) and building level (substation). This service uses advanced technology to manage and improve the energy efficiency of facilities.

The Building Energy Consumption Anomaly Detection service identifies unusual or unexpected patterns in municipal building energy usage by analyzing historical consumption data. Using machine learning models, it detects sudden spikes, drops, or gradual deviations from expected behavior. This service enables municipalities to quickly uncover equipment malfunctions, energy leaks, operational inefficiencies, or billing errors, ensuring faster corrective actions. It supports preventive maintenance, reduces unnecessary energy costs, and improves overall building performance monitoring.

The Building Demand Response Optimization service uses AI to identify opportunities for municipal buildings to reduce or shift electricity consumption during peak demand periods. By analyzing historical energy usage patterns, occupancy schedules, and building flexibility, it recommends optimal strategies to temporarily lower loads without disrupting essential services. This enables municipalities to participate in demand response programs, earn financial incentives, and support grid stability. The service prioritizes minimal occupant impact while maximizing energy savings and operational resilience.

The Battery Storage Optimization & Simulation service uses AI to model how installing battery systems can enhance energy performance in municipal buildings with PV installations. It simulates charge/discharge cycles based on historical consumption, solar generation, and tariff structures to identify optimal battery sizing and usage strategies. The service estimates self-consumption gains, peak load reduction, and financial payback, helping municipalities make data-driven investment decisions. It supports scenario analysis with or without dynamic pricing, enabling tailored recommendations per building.

Estimates utilization, power flow, current flow and the voltage magnitude of all grid elements based on measurements, billing information, grid model and weather data.

Estimates utilization, power flow, current flow and the voltage magnitude of all grid elements based on measurements, billing information, grid model and weather data.

This service allows to improve the accuracy and prediction capability of the dynamic models used in model-based control systems used by DSOs