Addressing the world's climate emergency is an uphill battle and requires a multifaceted approach including optimal deployment of green-energy alternatives. This often involve time-consuming optimisation of black-box models in a continuous parameter space. Despite recent advances in quantum computing, real-world applications have thus far been mostly confined to problems such as graph partitioning, traffic routing and task scheduling, where parameter space is discrete and graph connectivity is sparse. Here we propose the quantum nonlinear programming (QNLP) framework for casting an NLP problem - in continuous space - as quadratic unconstrained binary optimisation (QUBO), which can be subsequently solved using special-purpose solvers such as quantum annealers (QA) and coherent Ising machines (CIMs). QNLP consists of four steps: quadratic approximation of cost function, discretisation of parameter space, binarisation of discrete space, and solving the resulting QUBO. Linear and nonlinear constraints are incorporated into the resulting QUBO using slack variables and quadratic penalty terms. We apply our QNLP framework to optimisation of the daily feed rate of various biomass types at Nature Energy, the largest biogas producer in Europe. Optimising biomass selection improves the profitability of biomethane production, thus contributing to sustainable carbon-neutral energy production. For solving the QUBO, we use D-Wave's quantum annealers. We observe good performance on the DW-2000Q QPU, and higher sensitivity of performance to number of samples and annealing time for the Advantage QPU. We hope that our proposed QNLP framework provides a meaningful step towards overcoming the computational challenges posed by high-dimensional continuous-optimisation problems, especially those encountered in our battle against man-made climate change.
Addressing the world's climate emergency is an uphill battle and requires a multifaceted approach including optimal deployment of green-energy alternatives. This often involve time-consuming optimisation of black-box models in a continuous parameter space. Despite recent advances in quantum computing, real-world applications have thus far been mostly confined to problems such as graph partitioning, traffic routing and task scheduling, where parameter space is discrete and graph connectivity is sparse. Here we propose the quantum nonlinear programming (QNLP) framework for casting an NLP problem - in continuous space - as quadratic unconstrained binary optimisation (QUBO), which can be subsequently solved using special-purpose solvers such as quantum annealers (QA) and coherent Ising machines (CIMs). QNLP consists of four steps: quadratic approximation of cost function, discretisation of parameter space, binarisation of discrete space, and solving the resulting QUBO. Linear and nonlinear constraints are incorporated into the resulting QUBO using slack variables and quadratic penalty terms. We apply our QNLP framework to optimisation of the daily feed rate of various biomass types at Nature Energy, the largest biogas producer in Europe. Optimising biomass selection improves the profitability of biomethane production, thus contributing to sustainable carbon-neutral energy production. For solving the QUBO, we use D-Wave's quantum annealers. We observe good performance on the DW-2000Q QPU, and higher sensitivity of performance to number of samples and annealing time for the Advantage QPU. We hope that our proposed QNLP framework provides a meaningful step towards overcoming the computational challenges posed by high-dimensional continuous-optimisation problems, especially those encountered in our battle against man-made climate change.
Renewable energy optimisation poses computationally-intensive challenges. Yet, often the continuous nature of the decision space precludes the use of many emerging, non-von-Neumann computing platforms such as quantum annealing, which are limited to discrete problems. We propose Quantum Annealing Continuous Optimisation (QuAnCO), a Trust Region (TR)-based algorithm, where the TR Newton sub-problem is transformed into Quadratic Unconstrained Binary Optimisation (QUBO), thereby allowing the use of Ising solvers such as D-Wave's quantum annealer. This transformation to QUBO is done by 1) using a hyper-rectangular shape for the TR, 2) discrete representation of each continuous dimension using an interval-bounded integer, and 3) binary encoding of the resulting bounded integers. We tackle a real-world challenge of optimising the biomass mix selection for Nature Energy, the largest biogas producer in Europe, thus providing evidence of feasibility and performance advantage in using QuAnCO in green energy production, and beyond.
Quantum annealers can solve QUBO problems efficiently but struggle with continuous optimization tasks like regression due to their discrete nature. We introduce Quadratic Continuous Quantum Optimization (QCQO), an anytime algorithm that approximates solutions to unconstrained quadratic programs via a sequence of QUBO instances. Rather than encoding real variables as binary vectors, QCQO implicitly represents them using continuous QUBO weights and iteratively refines the solution by summing sampled vectors. This allows flexible control over the number of binary variables and adapts well to hardware constraints. We prove convergence properties, introduce a step size adaptation scheme, and validate the method on linear regression. Experiments with simulated and real quantum annealers show that QCQO achieves accurate results with fewer qubits, though convergence slows on noisy hardware. Our approach enables quantum annealing to address a wider class of continuous problems.
As global energy demand continues to rise, efficient energy management is essential to support the growing integration of renewable energy in the energy mix while controlling operational costs. This paper introduces a method that leverages non-exact Quadratic Unconstrained Binary Optimization (QUBO) formulation combined with tailored post-processing to boost the optimization of the energy management problem of 5G base stations. Unlike traditional rule-based approaches, which rely heavily on domain expertise and manual design, we propose a systematic QUBO-based framework suitable for Quantum Annealing (QA). The proposed formulation significantly reduces the number of decision variables (from 1995 in an exact QUBO to approximately 144) making it compatible with current QA hardware while producing high-quality solutions. Furthermore, the post-processing steps further improve solution quality, increasing the global optimal solution achievement rate by 10–15% depending on the solver. Experimental results demonstrate that the proposed method is deployable on current hardware and achieves performance comparable to conventional rule-based systems. Such observations highlight the potential of such designs in solving real-world optimization challenges.
Abstract This chapter explores the multifaceted relationship between quantum computing (QC) and sustainability, with a focus on the Quadratic Unconstrained Binary Optimisation (QUBO) framework. The manuscript delves into the theoretical underpinnings of QUBO and its formulation as a quantum annealing problem, identifying the quantum principles that facilitate the resolution of such optimisation challenges. It offers a critical analysis of the suitability of QUBO for unconstrained problems and its efficacy in consistently locating the global minimum – a pivotal concern in optimisation tasks. Further, this study provides a nuanced discussion on the intersection of QC and sustainability. It delineates the types of optimisation problems within sustainability initiatives that are amenable to formulation as QUBO problems, while also highlighting sustainability challenges that elude the QUBO framework. It argues for the integration of quantum solutions into business operations, highlighting the potential for QC to play a transformative role in achieving sustainability objectives. The critique of the current hype surrounding QC provides a balanced viewpoint, ensuring a grounded approach to the adoption of quantum technologies in tackling pressing global issues.
This study focuses on the integration of quantum computing and artificial intelligence, proposing a multi-task optimisation framework based on the quantum annealing algorithm to address the bottlenecks in computational efficiency and model optimisation in traditional machine learning. The study converts problems such as time series prediction, Support Vector Machine (SVM) classification, and Convolutional Neural Network (CNN) training into Quadratic Unconstrained Binary Optimisation (QUBO) models and achieves parameter optimisation through the simulated annealing algorithm. In time series prediction, an autoregressive model is constructed based on historical data. The problem of minimising prediction error is mapped to a binary optimisation problem through QUBO transformation, and the model parameters are solved using the simulated annealing algorithm to achieve trend prediction of the target variable, validating the effectiveness of quantum optimisation in time series analysis. For classification tasks, the hinge loss and regularisation terms of SVM are converted into a QUBO matrix to optimise the classification hyperplane parameters, thereby enhancing the model's classification performance in the feature space. The results demonstrate that this method exhibits strong geometric interpretability and generalisation capabilities in binary classification problems. In the field of deep learning, L2 regularisation is introduced into the weight optimisation of CNNs, and a QUBO model is constructed to explore the application of quantum optimisation in image feature extraction and network training. Although limited by the scale of variables, this demonstrates the potential value of combining quantum computing with deep learning. The study shows that the quantum optimisation framework can effectively improve solution efficiency in small- to medium-scale problems, providing new technical pathways for time series analysis, classification tasks, and deep learning optimisation.
Optimization is a crucial challenge across various domains, including finance, resource allocation, and mobility. Quantum computing has the potential to redefine the way we handle complex problems by reducing computational complexity and enhancing solution quality. Optimization, particularly of objective functions, stands to benefit significantly from quantum solvers, which leverage principles of quantum mechanics like superposition, entanglement, and tunneling. The Ising and Quadratic Unconstrained Binary Optimization (QUBO) models are the most suitable formulations for these solvers, involving binary variables and constraints treated as penalties within the overall objective function. To harness quantum approaches for optimization, two primary strategies are employed: exploiting quantum annealers—special-purpose optimization devices—and designing algorithms based on quantum circuits. This review provides a comprehensive overview of quantum optimization methods, examining their advantages, challenges, and limitations. It demonstrates their application to real-world scenarios and outlines the steps to convert generic optimization problems into quantum-compliant models. Lastly, it discusses available tools and frameworks that facilitate the exploration of quantum solutions for optimization tasks.