Optimization of Graded Arrays of Resonators for Energy Harvesting in Sensors as a Markov Decision Process Solved via Reinforcement Learning
The design optimization of the grading of a resonator array for energy harvesting in sensors is described. Attention is paid to set the resonator heights, possibly removing resonators whenever convenient. Instead of employing time-consuming heuristic approaches that require verifying the physical un...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Engineering Proceedings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2673-4591/27/1/18 |
| _version_ | 1797611971171844096 |
|---|---|
| author | Luca Rosafalco Jacopo Maria De Ponti Luca Iorio Raffaele Ardito Alberto Corigliano |
| author_facet | Luca Rosafalco Jacopo Maria De Ponti Luca Iorio Raffaele Ardito Alberto Corigliano |
| author_sort | Luca Rosafalco |
| collection | DOAJ |
| description | The design optimization of the grading of a resonator array for energy harvesting in sensors is described. Attention is paid to set the resonator heights, possibly removing resonators whenever convenient. Instead of employing time-consuming heuristic approaches that require verifying the physical understanding of the problem and tuning the design ruling parameters, the optimization task is treated as a Markov decision process, in which states describe specific system configurations, and actions represent the modifications to the current design. The physics-based understanding of the problem is exploited to constrain the set of possible modifications to the mechanical system. Finite elements simulations are exploited to evaluate the action effects and to inform the reinforcement learning agent. The proximal policy optimization algorithm is employed to solve the Markov decision problem. The procedure is demonstrated to be able to automatically produce configurations, enhancing the mechanical system performance. The proposed framework is generalizable to a large class of problems involving the design optimization of sensors. |
| first_indexed | 2024-03-11T06:35:48Z |
| format | Article |
| id | doaj.art-5aba2755e58c4cbc8496ca34706ae927 |
| institution | Directory Open Access Journal |
| issn | 2673-4591 |
| language | English |
| last_indexed | 2024-03-11T06:35:48Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Engineering Proceedings |
| spelling | doaj.art-5aba2755e58c4cbc8496ca34706ae9272023-11-17T10:54:34ZengMDPI AGEngineering Proceedings2673-45912022-11-012711810.3390/ecsa-9-13216Optimization of Graded Arrays of Resonators for Energy Harvesting in Sensors as a Markov Decision Process Solved via Reinforcement LearningLuca Rosafalco0Jacopo Maria De Ponti1Luca Iorio2Raffaele Ardito3Alberto Corigliano4Dipartimento di Ingegneria Civile ed Ambientale, Politecnico di Milano, Piazza L. Da Vinci 32, 20133 Milano, ItalyDipartimento di Ingegneria Civile ed Ambientale, Politecnico di Milano, Piazza L. Da Vinci 32, 20133 Milano, ItalyDipartimento di Ingegneria Civile ed Ambientale, Politecnico di Milano, Piazza L. Da Vinci 32, 20133 Milano, ItalyDipartimento di Ingegneria Civile ed Ambientale, Politecnico di Milano, Piazza L. Da Vinci 32, 20133 Milano, ItalyDipartimento di Ingegneria Civile ed Ambientale, Politecnico di Milano, Piazza L. Da Vinci 32, 20133 Milano, ItalyThe design optimization of the grading of a resonator array for energy harvesting in sensors is described. Attention is paid to set the resonator heights, possibly removing resonators whenever convenient. Instead of employing time-consuming heuristic approaches that require verifying the physical understanding of the problem and tuning the design ruling parameters, the optimization task is treated as a Markov decision process, in which states describe specific system configurations, and actions represent the modifications to the current design. The physics-based understanding of the problem is exploited to constrain the set of possible modifications to the mechanical system. Finite elements simulations are exploited to evaluate the action effects and to inform the reinforcement learning agent. The proximal policy optimization algorithm is employed to solve the Markov decision problem. The procedure is demonstrated to be able to automatically produce configurations, enhancing the mechanical system performance. The proposed framework is generalizable to a large class of problems involving the design optimization of sensors.https://www.mdpi.com/2673-4591/27/1/18energy harvesting for sensorsmetamaterialsreinforcement learningMarkov decision process |
| spellingShingle | Luca Rosafalco Jacopo Maria De Ponti Luca Iorio Raffaele Ardito Alberto Corigliano Optimization of Graded Arrays of Resonators for Energy Harvesting in Sensors as a Markov Decision Process Solved via Reinforcement Learning Engineering Proceedings energy harvesting for sensors metamaterials reinforcement learning Markov decision process |
| title | Optimization of Graded Arrays of Resonators for Energy Harvesting in Sensors as a Markov Decision Process Solved via Reinforcement Learning |
| title_full | Optimization of Graded Arrays of Resonators for Energy Harvesting in Sensors as a Markov Decision Process Solved via Reinforcement Learning |
| title_fullStr | Optimization of Graded Arrays of Resonators for Energy Harvesting in Sensors as a Markov Decision Process Solved via Reinforcement Learning |
| title_full_unstemmed | Optimization of Graded Arrays of Resonators for Energy Harvesting in Sensors as a Markov Decision Process Solved via Reinforcement Learning |
| title_short | Optimization of Graded Arrays of Resonators for Energy Harvesting in Sensors as a Markov Decision Process Solved via Reinforcement Learning |
| title_sort | optimization of graded arrays of resonators for energy harvesting in sensors as a markov decision process solved via reinforcement learning |
| topic | energy harvesting for sensors metamaterials reinforcement learning Markov decision process |
| url | https://www.mdpi.com/2673-4591/27/1/18 |
| work_keys_str_mv | AT lucarosafalco optimizationofgradedarraysofresonatorsforenergyharvestinginsensorsasamarkovdecisionprocesssolvedviareinforcementlearning AT jacopomariadeponti optimizationofgradedarraysofresonatorsforenergyharvestinginsensorsasamarkovdecisionprocesssolvedviareinforcementlearning AT lucaiorio optimizationofgradedarraysofresonatorsforenergyharvestinginsensorsasamarkovdecisionprocesssolvedviareinforcementlearning AT raffaeleardito optimizationofgradedarraysofresonatorsforenergyharvestinginsensorsasamarkovdecisionprocesssolvedviareinforcementlearning AT albertocorigliano optimizationofgradedarraysofresonatorsforenergyharvestinginsensorsasamarkovdecisionprocesssolvedviareinforcementlearning |