Data-driven optimization of industrial processes in time-varying environments

Industry is responsible for a large part of the global energy consumption and emissions. To help meet energy demands and reduce climate change, Extreme Efficiency aims to make industrial processes more energy efficient. By measuring the effect of small changes in operating conditions, known as perturbations, it is possible to discover if similar larger changes will contribute to a more efficient process. If done systematically, perturbations enable us to find the best operating settings. This form of optimization has been tested on a wide variety of industrial equipment such as ovens, grinding mills, wind turbines, solar arrays, generators, energy converters, and drilling rigs. Up until now, perturbations are applied continuously to be able to track any changes in optimal settings due to environmental changes. Examples of environmental changes are changes in weather conditions and variations in the composition of raw materials. Continuous perturbations have several disadvantages. Even small changes in operating settings may correspond to actions such as heating or cooling of large areas, or movement of heavy machine parts. These actions may bring about a large direct cost or a significant contribution to wear of machinery. Moreover, swings in sensor readings that result from perturbations make it difficult to detect safety-critical faults. In this project, we address these disadvantages by developing algorithms that only turn on perturbations when needed. These algorithms weigh the cost of perturbing against the predicted gain in efficiency. In addition, contemporary algorithms do not handle known process changes well. Large efficiency losses can arise due to slow reactions to changes like scheduled machine adjustments and switches in setpoint values. We aim to give advance warnings to the optimization algorithms such that they can better prepare for known process changes and reduce performance losses.

With the effects of climate change and an energy crisis in Europe intensifying, the incentive to maximize energy efficiency and minimize emissions is strong. To help meet energy demands and mitigate climate change, the project EXTREME EFFICIENCY will make energy-intensive industrial processes of the future more efficient with a smaller environmental footprint by operating existing machinery and equipment more efficiently using data-driven performance optimization. Process uncertainty and unknown, changing environmental conditions (e.g., weather influences and compositions of raw materials) limit the available information for the purpose of optimization. Perturbation-based optimization methods are among the few methods that can be used to finetune performance. They excite the process to obtain information-rich measurement data. Therefore, more knowledge can be exploited to optimize the process. However, current perturbation-based optimization methods apply an unnecessary amount of perturbation, leading to high direct costs, excessive wear of machinery, and increased safety risks. Moreover, they have no mechanism to efficiently handle known process changes (e.g., machine adjustments), resulting in avoidable performance losses. In the EXTREME EFFICIENCY project, we aim to address the following challenges: • How to monitor the need for process excitation and redesign perturbations such that the adverse consequences of perturbation-based optimization are alleviated, while a similar level of performance improvement as for current perturbation-based optimization methods is achieved? • How to exploit existing process models and data-driven models to compute an efficient response to known process changes, leading to minimal performance losses? The project will develop novel algorithms that actively control the quality of the measured data by systematic excitation of the process and combine measurement data and process models in an intuitive manner for easy deployment.