Sponsor: RTE
Context
The role of the power transmission grid is to transport electrical energy from production areas (thermal, hydroelectric, wind, solar) to consumption areas (cities, industrial parks). It can be simplified as a set of substations interconnected by power lines. As the transmission network operator, RTE (Réseau de Transport d'Électricité) is responsible for ensuring the safety of the electrical system and maintaining a constant balance between electricity production and consumption. RTE is also responsible for developing the network, in particular by adding new lines to facilitate the integration of renewable energies, as well as maintaining its infrastructure (lines, transformers, pylons, etc.) to ensure that it continues to function properly. However, some maintenance operations can result in the temporary unavailability of a structure, which can weaken the network. This can sometimes lead to complex and costly situations, forcing managers to postpone certain interventions until a later date. RTE's infrastructure is ageing, which requires an increasing volume of maintenance operations. These interventions must be carefully planned throughout the year. In addition, the increased integration of renewable energies into the network introduces further uncertainties into its operation. In this emerging context, planning maintenance operations is becoming a major technical challenge.
Description and objectives
The objective of this project would therefore be to evaluate the ability of quantum algorithms to solve a planning problem under uncertainty. Planning problems are common in many fields, such as logistics, industrial production and event planning. This involves efficiently organising a series of activities over time, taking into account constraints such as available resources, dependencies between tasks and objectives to be achieved. Its resolution is the subject of extensive literature, particularly with the use of quantum algorithms.
In the energy sector, maintaining the electricity transmission network raises complex planning issues that are specific and highly combinatorial. In addition to the usual resource and time constraints, some are specific to electricity networks. For example, it is often impossible to shut down two critical or nearby power lines at the same time, as this could pose a significant risk to the safety of the electricity system. Furthermore, the occurrence of various events can lead to changes in the schedule. The unexpected loss of a power line could prevent neighbouring lines from being shut down due to the fragility of the network. The need to maintain energy production evacuation lines following the unexpected unavailability of other production sources. Such contingencies must be taken into account while limiting their impact on the overall planning.
The objective of this work will therefore be to model and construct a schedule for the lockout of electrical installations, taking into account several constraints: the availability of human resources, the workload, the electrical risk associated with the fragility of the network, and the uncertainties associated with the future state of the system. This uncertainty is mainly caused by weather factors that affect electricity production and consumption, but also by changes in the network structure (availability of facilities, connection topology, etc.).