Quantum Computing Approach for Energy Optimization in a Prosumer Community

Abstract

The efficient management of energy communities relies on the solution of the "prosumer problem", i.e., the problem of scheduling the household loads on the basis of the user needs, the electricity prices, and the availability of local renewable energy, with the aim of reducing costs and energy waste. Quantum computers can offer a significant breakthrough in treating this problem thanks to the intrinsic parallel nature of quantum operations. The most promising approach is to devise variational hybrid algorithms, in which quantum computation is driven by parameters that are optimized classically, in a cycle that aims at finding the best solution with a significant speedup with respect to classical approaches. This paper provides a reformulation of the prosumer problem, allowing to address it with a hybrid quantum algorithm, namely, Quantum Approximate Optimization Algorithm (QAOA), and with a recent variant, the Recursive QAOA. We report on an extensive set of experiments, on simulators and real quantum hardware, for different problem sizes. Results are encouraging in that Recursive QAOA is able, for problems involving up to 10 qubits, to provide optimal and admissible solutions with good probabilities, while the computation time is nearly independent of the system size.

Quantum Computing Management of a Cloud/Edge Architecture

“…We compare the results achieved with QAOA, with a number of repetitions reps between 1 and 5, and with VQE equipped with Ansatz A1. As predicted by the theory, the performance of QAOA improves with more repetitions; however, the improvement is slow, and the number of repetitions must be kept low, since noise and decoherence effects increase with the depth of the circuit, as shown in a previous work, i.e., [36]. These results show that VQE is definitely preferable, due to its ability of reducing the search space, even with the simplest ansatz.…”

“…This procedure has inspired algorithms that can be run on a gate-based quantum computer, such as those provided by major IT companies. As an example, we employed such an approach in a previous work on the optimization of energy exchanges within a prosumer community [36]. The applications of quantum algorithms to edge computing are all very recent.…”