Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty
Of late, the exponential rise in the global population is driving higher energy demand. However, the rapid depletion of conventional fossil fuels and growing environmental concerns have prompted the evolution of alternative energy sources. To this end, Microgrid (MG) with Renewable Energy Sources (R...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-06-01
|
| Series: | Results in Control and Optimization |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666720724000377 |
| _version_ | 1797261787766194176 |
|---|---|
| author | Sukriti Patty Tanmoy Malakar |
| author_facet | Sukriti Patty Tanmoy Malakar |
| author_sort | Sukriti Patty |
| collection | DOAJ |
| description | Of late, the exponential rise in the global population is driving higher energy demand. However, the rapid depletion of conventional fossil fuels and growing environmental concerns have prompted the evolution of alternative energy sources. To this end, Microgrid (MG) with Renewable Energy Sources (RES) has emerged as popular means of small-scale localized power grid. However, planning of MG operation poses challenges due to the inherent variability and stochasticity in RES power output and energy demand. On account of this, the present study introduces a Stochastic Energy Management Strategy (SEMS) for a grid-connected MG incorporating Micro-Turbine, Fuel-Cell, RES, Battery Energy Storage, and electrical and heat energy demand. The stochasticity of RES is forecasted through a hybrid prediction model (sARIMA-GRU) and the uncertain demand is estimated via 'Monte Carlo Simulation.' The proposed problem is formulated as a dynamic non-linear stochastic optimization problem. It seeks to minimize the expected value of MG operational cost satisfying the practical constraints. Addressing this, a newly developed ‘Artificial Electric Field Algorithm (AEFA)' is utilized. Several case studies are performed to assess MG operation under varied operating conditions. Moreover, the present study analyses the impact of uncertainty on energy contribution from DER, grid dependency, and MG operation cost. Comparative analysis reveals that sARIMA-GRU outperforms other contemporary prediction models. It is noteworthy that the superior prediction accuracy of sARIMA-GRU leads to lower MG operation costs. Moreover, statistical analysis and convergence confirm the proficiency of applied AEFA over state-of-the-art Grey Wolf Optimization and Firefly Algorithm in solving the proposed problem. |
| first_indexed | 2024-04-24T23:46:46Z |
| format | Article |
| id | doaj.art-9a007c34eafd433f9c1f877dd2c08019 |
| institution | Directory Open Access Journal |
| issn | 2666-7207 |
| language | English |
| last_indexed | 2024-04-24T23:46:46Z |
| publishDate | 2024-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Results in Control and Optimization |
| spelling | doaj.art-9a007c34eafd433f9c1f877dd2c080192024-03-15T04:44:42ZengElsevierResults in Control and Optimization2666-72072024-06-0115100407Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertaintySukriti Patty0Tanmoy Malakar1Corresponding author.; Department of Electrical Engineering,National Institute of Technology, Silchar, Assam 788010, IndiaDepartment of Electrical Engineering,National Institute of Technology, Silchar, Assam 788010, IndiaOf late, the exponential rise in the global population is driving higher energy demand. However, the rapid depletion of conventional fossil fuels and growing environmental concerns have prompted the evolution of alternative energy sources. To this end, Microgrid (MG) with Renewable Energy Sources (RES) has emerged as popular means of small-scale localized power grid. However, planning of MG operation poses challenges due to the inherent variability and stochasticity in RES power output and energy demand. On account of this, the present study introduces a Stochastic Energy Management Strategy (SEMS) for a grid-connected MG incorporating Micro-Turbine, Fuel-Cell, RES, Battery Energy Storage, and electrical and heat energy demand. The stochasticity of RES is forecasted through a hybrid prediction model (sARIMA-GRU) and the uncertain demand is estimated via 'Monte Carlo Simulation.' The proposed problem is formulated as a dynamic non-linear stochastic optimization problem. It seeks to minimize the expected value of MG operational cost satisfying the practical constraints. Addressing this, a newly developed ‘Artificial Electric Field Algorithm (AEFA)' is utilized. Several case studies are performed to assess MG operation under varied operating conditions. Moreover, the present study analyses the impact of uncertainty on energy contribution from DER, grid dependency, and MG operation cost. Comparative analysis reveals that sARIMA-GRU outperforms other contemporary prediction models. It is noteworthy that the superior prediction accuracy of sARIMA-GRU leads to lower MG operation costs. Moreover, statistical analysis and convergence confirm the proficiency of applied AEFA over state-of-the-art Grey Wolf Optimization and Firefly Algorithm in solving the proposed problem.http://www.sciencedirect.com/science/article/pii/S2666720724000377Machine learningUncertaintyMicrogridOptimizationRenewable energy |
| spellingShingle | Sukriti Patty Tanmoy Malakar Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty Results in Control and Optimization Machine learning Uncertainty Microgrid Optimization Renewable energy |
| title | Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty |
| title_full | Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty |
| title_fullStr | Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty |
| title_full_unstemmed | Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty |
| title_short | Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty |
| title_sort | performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty |
| topic | Machine learning Uncertainty Microgrid Optimization Renewable energy |
| url | http://www.sciencedirect.com/science/article/pii/S2666720724000377 |
| work_keys_str_mv | AT sukritipatty performanceanalysisofmachinelearningbasedpredictionmodelsinassessingoptimaloperationofmicrogridunderuncertainty AT tanmoymalakar performanceanalysisofmachinelearningbasedpredictionmodelsinassessingoptimaloperationofmicrogridunderuncertainty |