With increasing regenerative power generation, adapting the energy supply to the demand becomes increasingly difficult, resulting in network instabilities and utility provider losses due to expensive spot-market (day-ahead) energy.
Vehicle to Grid (V2G) technology allows electric vehicles (EVs) to be connected to the power grid and both send and receive electricity. V2G can help balance the grid by allowing EVs to provide power back to it when demand is high, and to charge their batteries when demand is low.
It also allows energy providers to increase their profits and cut losses by shifting energy loads to times of cheaper market energy. Additionally, V2G can help to increase the utilization and value of EVs by allowing them to be used as a distributed energy storage. This can also help to reduce the need for expensive, centralized energy storage solutions, and promote the use of renewable energy sources.
In this demo, the usage of V2G is implemented as a so-called Feed-in Reward Program (FiRP): the utility provider offers (participating) EV users a reward price for their energy to get them to feed into the grid at specific times. Depending on the prices, the amount of feed-in energy can be controlled by the utility provider. The optimal reward prices are those that maximize the utility provider’s profit.
The Quantum Vehicle-to-Grid (VQG) app maps the Feed-in Reward Program for different scenarios (i.e., energy demands, market prices and user profiles) to an optimization problem that can be solved with classical, quantum and quantum-inspired algorithms.
Why Quantum? Depending on the scenario, this optimization problem is often non-convex making it hard to be solved classically. In these cases, quantum algorithms may provide benefits for finding better solutions. Although current quantum solutions are still not competitive with corresponding classical ones, the full demo showcases the use of several different quantum approaches for solving a relevant optimization problem.
The app benefits from a separation of business and optimization logic by utilizing a custom optimization framework that offers easy integration of various backends and solvers. The full version can access/includes, e.g., Operator-Splitting Quadratic Program (OSQP) solver, Covariance Matrix Adaption Evolution Strategy (CMA-ES) algorithm, Quantum Approximate Optimization Algorithm (QAOA), Quantum Annealing and Quantum-inspired Algorithms (Microsoft QIO) and more.
