Engie (France)

In Europe, the building sector accounts for approximately 35 % of the energy-related emissions. Hybrid systems coordinating heat pumps and gas boilers can avoid greenhouse gas (GHG) emissions from carbonized electricity production by providing demand-side flexibility without any service interruption. This work aimed to develop a control strategy for a fleet of hybrid heat pumps to reduce GHG emissions. The electricity and gas consumption of a fleet of 3000 hybrid heat pumps, heating 100,000 dwellings spread throughout France, was evaluated. A Modelica model of a district archetype was simulated in seven cities representative of the French climatic zones to obtain the national heating demand. The marginal emission factor of the electricity consumption was assessed using a French power system model coupled with marginal emission factors for interconnected power systems, which were assessed through linear regressions. Two types of control strategies (prioritizing the heat pump and fuel switch) are evaluated considering 4 different sizing for the heat pump (120 %, 50 %, 35 %, and 20 %). Between July 2018 and June 2019, a strategy prioritizing the heat pumps would have avoided between 8000 and 26,000 tCO2eq for the power system. A strategy switching between the heat pump and the boiler based on the marginal emission factor of the electricity consumption would have avoided around 38,000 tCO2eq, with a limited influence of the sizing of the heat pump.

Wind energy development is a key component of climate change mitigation. However, birds collide with wind turbines, and this additional mortality may negatively impact populations. Collision risk could be reduced by informed selection of turbine dimensions, but the effects of turbine dimensions are still unknown for many species.As analyses of mortality data have several limitations, we applied a simulation approach based on flight height distributions of six European raptor species. To obtain accurate flight height data, we used high-frequency GPS tracking (GPS tags deployed on 275 individuals). The effects of ground clearance and rotor diameter of wind turbines on collision risk were studied using the Band collision risk model.Five species had a unimodal flight height distribution, with a mode below 25 m above ground level, while Short-toed Eagle showed a more uniform distribution with a weak mode between 120 and 260 m. The proportion of positions within 32–200 m ranged from 11 % in Marsh Harrier to 54 % in Red Kite.With increasing ground clearance (from 20 to 100 m), collision risk decreased in the species with low mode (−56 to −66 %), but increased in Short-toed Eagle (+38 %). With increasing rotor diameter (from 50 to 160 m) at fixed ground clearance, the collision risk per turbine increased in all species (+151 to +558 %), while the collision risk per MW decreased in the species with low mode (−50 % to −57 %).These results underpin that wind turbine dimensions can have substantial effects on the collision risk of raptors. As the effects varied between species, wind energy planning should consider the composition of the local bird community to optimise wind turbine dimensions. For species with a low mode of flight height, the collision risk for a given total power capacity could be reduced by increasing ground clearance, and using fewer turbines with larger diameter.

During in situ biomethanation, microbial communities can convert complex Organic Matter (OM) and H2 into CH4. OM biodegradation was compared between Anaerobic Digestion (AD) and in situ biomethanation, in semi-continuous processes, using two inocula from the digester (D) and the post-digester (PoD) of an AD plant. The impact of H2 on OM degradation was assessed using a fractionation method. Operational parameters included 20 days of hydraulic retention time and 1.5 gVS.L−1.d−1 of organic loading rate. During in situ biomethanation, 485 NmL of H2 were injected for each feeding (3 times a week). Maximum organic COD removal was 0.6 gCOD in AD control and at least 1.6 gCOD for in situ biomethanation. Therefore, COD removal was 2.5 times higher with H2 injections. These results bring out the potential of H2 injections during AD, not only for CO2 consumption but also for better OM degradation.

Assessing a wind farm’s annual energy production (AEP) involves modelling the wind resource and the wind-to-power conversion at the site. The greenfield pre-construction phase generally comprises the installation of wind measurement devices. For repowering projects, the wind data from the pre-construction phase of the existing farm can be used as wind input to assess the energy yield of the repowered wind farm. Indeed, one study demonstrates that when the modelling error correlations are known, the AEP prediction uncertainty of the repowered farm can be reduced by combining the energy production records of the existing farm with the AEP assessment for both farms. Previous studies have successfully identified the correlation structure for certain errors, especially for horizontal flow modelling, but not for vertical flow modelling. However, vertical extrapolation is essential, as the wind measurement heights are generally lower than the hub height on the repowered farm. This paper bridges this research gap and demonstrates that the correlation structure of errors in vertical profile modelling is Gaussian, with parameters dependent on shear values and heights. The distribution is validated against site data from simple to moderately complex sites in France.

By offering promising solutions to two critical issues – the integration of renewable energies into energy systems and the decarbonization of existing hydrogen applications – green hydrogen production through water electrolysis is set to play a crucial role in addressing the major challenges of the energy transition. However, the successful integration of renewable energy sources relies on gaining accurate insights into the impacts that intermittent electrical supply conditions induce on electrolyzers. Despite the rising importance of addressing intermittency issues to accelerate the widespread adoption of renewable energy sources, the state-of-the-art lacks research providing an in-depth understanding of these concerns. This paper endeavors to offer a comprehensive review of existing research, focusing on proton exchange membrane (PEM) and alkaline electrolysis technologies operating under intermittent operation. Despite growing interest over the last ten years, the review underscores the scarcity of industrial-scale databases for quantifying these impacts.

The emergence of increasingly powerful and inexpensive single-board computers has motivated a great deal of work in the field of edge computing. We believe that knowledge graphs will contribute to intelligent edge computing. This requires the ability to efficiently answer queries requiring inferences performed with minimal knowledge accessible on a device at the edge of the network. In this work, we determine the minimum size of the knowledge graph that an edge device needs based on the analysis of its query workload. In the context of a succinct data structures-based store, we also present an incremental update of this knowledge graph when new queries are introduced into the environment. We demonstrate the effectiveness of our solution on real use cases encountered by our industrial partner.

Energy systems are composed of interacting components and subsystems, commonly owned by different actors and stakeholders. For several decades, multi-agent systems (MAS) have been used to effectively model and manage energy systems. However, interactions between heterogeneous constituents of such systems remain one of the key challenges. In this paper, we present an approach that enables such components to interoperate seamlessly. The approach is implemented based on SPARQL-Act, a SPARQL-based agent communication language (ACL), to provide a platform-independent ACL. It also manages agents' knowledge bases and handles messages of different performatives. The use of RDF as the content language of exchanged messages and ontologies contributes to the interoperability of agent interactions. The approach was validated in a use case of building energy management of CityLearn. We modelled and implemented a rule-based MAS for optimising the Heating, Ventilation, and Air-Conditioning (HVAC) system of a building. The results show that our proposed solution allows agents to communicate without ambiguity, to manage their knowledge bases, and to interact effectively to achieve their individual as well as the system's objectives.

We present an approach that autonomously adapts sensor monitoring of an IoT environment. Based on semantic technologies, our solution supports the generation of relevant continuous queries when certain anomalies are identified. The generation consists of a query graph extension which is triggered when some rules are fired. These queries are executed on a graph database system designed for Edge computing. We evaluate the accuracy of the generated queries, the robustness, and the latency of our system in a real use case consisting of a smart building context equipped with multiple sensors.