Integrid - Bridging the Gap

22 Hybrid Power Electronic Converter for PV and Storage Concept The hybrid power electronic converter for small-scale storage and PV has a power circuit composed of a three-port converter, with interfaces for AC network, batteries and photovoltaic panels. With this configuration it is possible to have the same functionality of a separate PV inverter and battery inverter, with reduced cost, complexity, volume and weight. These benefits come from the use of shared AC-DC converter, controller, auxiliary supply, sensors, DC link, enclosure, etc. Prototype The development of the final prototype was mainly driven by the conclusion of the tests performed on the older version in the DNV GL Flexible Power Grid Laboratory. In these tests, some weaknesses were found, and work was carried out to overcome them. Furthermore, a number of other improvements and upgrades already identified by INESC TEC were also considered. For the battery interface, considerable effort was made to achieve compatibility with commercial battery banks used for residential energy storage. The power level was reduced to 1.5kW to target typical residential market installations. This modification also allows reduced cost, weight, and volume. Innovative features: Single electronic power converter for small-scale storage and PV that reduces investment cost, complexity, volume and weight. Figure 3.7 – Prototype electronic power converter Final TRL: 7 Data-driven Energy Optimization Concept Data-driven predictive control for wastewater variable- speed pumps that minimize power consumption, based on uncertainty forecasts for the wastewater intake rate and information collected by sensors accessible through the SCADA system. A key advantage of data- driven control algorithms is that they avoid the use of simplifications or simulation models for the physical system. This approach is suitable for cases were the physical (or mathematical) cannot be obtained due to lack of information/data or is very expensive to model. Software The following main sub-systems are: a) Emulated environment: relies on data driven methods to estimate several environment relations. b) Wastewater intake rate (WWIR) forecasting: produces a series of probabilistic quantiles to quantify the uncertainty of the wastewater intake rate (WWIR). c) Predictive control algorithm, based on the current system state andWWIR forecasts, is able to define individual set-points for each system pump in order to optimize operation and reduce power consumption. Innovative features: • Applies a model-free and data-driven control approach based on reinforcement learning. • The control philosophy focuses on operating the tank with a variable water level, instead of controlling the frequency increase/decrease rate. • Avoids the use of simplifications or simulation models for the physical system. Final TRL: 8 Business Domain: Energy Services Building Energy Management System Concept EDP Comercial implemented a flexibility aggregation system to maximize thermal inertia from office buildings for Energy Markets. An on-site Building