Bayesian Optimized Echo State Network Applied to Short-Term Load Forecasting
Load forecasting impacts directly financial returns and information in electrical systems planning. A promising approach to load forecasting is the Echo State Network (ESN), a recurrent neural network for the processing of temporal dependencies. The low computational cost and powerful performance of...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-05-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/13/9/2390 |
| _version_ | 1797568302411677696 |
|---|---|
| author | Gabriel Trierweiler Ribeiro João Guilherme Sauer Naylene Fraccanabbia Viviana Cocco Mariani Leandro dos Santos Coelho |
| author_facet | Gabriel Trierweiler Ribeiro João Guilherme Sauer Naylene Fraccanabbia Viviana Cocco Mariani Leandro dos Santos Coelho |
| author_sort | Gabriel Trierweiler Ribeiro |
| collection | DOAJ |
| description | Load forecasting impacts directly financial returns and information in electrical systems planning. A promising approach to load forecasting is the Echo State Network (ESN), a recurrent neural network for the processing of temporal dependencies. The low computational cost and powerful performance of ESN make it widely used in a range of applications including forecasting tasks and nonlinear modeling. This paper presents a Bayesian optimization algorithm (BOA) of ESN hyperparameters in load forecasting with its main contributions including helping the selection of optimization algorithms for tuning ESN to solve real-world forecasting problems, as well as the evaluation of the performance of Bayesian optimization with different acquisition function settings. For this purpose, the ESN hyperparameters were set as variables to be optimized. Then, the adopted BOA employs a probabilist model using Gaussian process to find the best set of ESN hyperparameters using three different options of acquisition function and a surrogate utility function. Finally, the optimized hyperparameters are used by the ESN for predictions. Two datasets have been used to test the effectiveness of the proposed forecasting ESN model using BOA approaches, one from Poland and another from Brazil. The results of optimization statistics, convergence curves, execution time profile, and the hyperparameters’ best solution frequencies indicate that each problem requires a different setting for the BOA. Simulation results are promising in terms of short-term load forecasting quality and low error predictions may be achieved, given the correct options settings are used. Furthermore, since there is not an optimal global optimization solution known for real-world problems, correlations among certain values of hyperparameters are useful to guide the selection of such a solution. |
| first_indexed | 2024-03-10T19:55:32Z |
| format | Article |
| id | doaj.art-6f9f1702cf5e42b998b438dd271a0803 |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-03-10T19:55:32Z |
| publishDate | 2020-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-6f9f1702cf5e42b998b438dd271a08032023-11-20T00:01:55ZengMDPI AGEnergies1996-10732020-05-01139239010.3390/en13092390Bayesian Optimized Echo State Network Applied to Short-Term Load ForecastingGabriel Trierweiler Ribeiro0João Guilherme Sauer1Naylene Fraccanabbia2Viviana Cocco Mariani3Leandro dos Santos Coelho4Department of Electrical Engineering, Federal University of Parana (UFPR), Av. Coronal Francisco Heráclito dos Santos, 100, Curitiba (PR) 80060-000, BrazilDepartment of Electrical Engineering, Federal University of Parana (UFPR), Av. Coronal Francisco Heráclito dos Santos, 100, Curitiba (PR) 80060-000, BrazilDepartment of Mechanical Engineering, Pontifical Catholic University of Parana (PUCPR), Rua Imaculada Conceição, 1155, Curitiba (PR) 80215-901, BrazilDepartment of Electrical Engineering, Federal University of Parana (UFPR), Av. Coronal Francisco Heráclito dos Santos, 100, Curitiba (PR) 80060-000, BrazilDepartment of Electrical Engineering, Federal University of Parana (UFPR), Av. Coronal Francisco Heráclito dos Santos, 100, Curitiba (PR) 80060-000, BrazilLoad forecasting impacts directly financial returns and information in electrical systems planning. A promising approach to load forecasting is the Echo State Network (ESN), a recurrent neural network for the processing of temporal dependencies. The low computational cost and powerful performance of ESN make it widely used in a range of applications including forecasting tasks and nonlinear modeling. This paper presents a Bayesian optimization algorithm (BOA) of ESN hyperparameters in load forecasting with its main contributions including helping the selection of optimization algorithms for tuning ESN to solve real-world forecasting problems, as well as the evaluation of the performance of Bayesian optimization with different acquisition function settings. For this purpose, the ESN hyperparameters were set as variables to be optimized. Then, the adopted BOA employs a probabilist model using Gaussian process to find the best set of ESN hyperparameters using three different options of acquisition function and a surrogate utility function. Finally, the optimized hyperparameters are used by the ESN for predictions. Two datasets have been used to test the effectiveness of the proposed forecasting ESN model using BOA approaches, one from Poland and another from Brazil. The results of optimization statistics, convergence curves, execution time profile, and the hyperparameters’ best solution frequencies indicate that each problem requires a different setting for the BOA. Simulation results are promising in terms of short-term load forecasting quality and low error predictions may be achieved, given the correct options settings are used. Furthermore, since there is not an optimal global optimization solution known for real-world problems, correlations among certain values of hyperparameters are useful to guide the selection of such a solution.https://www.mdpi.com/1996-1073/13/9/2390Bayesian optimizationecho state networksshort-term load forecasting |
| spellingShingle | Gabriel Trierweiler Ribeiro João Guilherme Sauer Naylene Fraccanabbia Viviana Cocco Mariani Leandro dos Santos Coelho Bayesian Optimized Echo State Network Applied to Short-Term Load Forecasting Energies Bayesian optimization echo state networks short-term load forecasting |
| title | Bayesian Optimized Echo State Network Applied to Short-Term Load Forecasting |
| title_full | Bayesian Optimized Echo State Network Applied to Short-Term Load Forecasting |
| title_fullStr | Bayesian Optimized Echo State Network Applied to Short-Term Load Forecasting |
| title_full_unstemmed | Bayesian Optimized Echo State Network Applied to Short-Term Load Forecasting |
| title_short | Bayesian Optimized Echo State Network Applied to Short-Term Load Forecasting |
| title_sort | bayesian optimized echo state network applied to short term load forecasting |
| topic | Bayesian optimization echo state networks short-term load forecasting |
| url | https://www.mdpi.com/1996-1073/13/9/2390 |
| work_keys_str_mv | AT gabrieltrierweilerribeiro bayesianoptimizedechostatenetworkappliedtoshorttermloadforecasting AT joaoguilhermesauer bayesianoptimizedechostatenetworkappliedtoshorttermloadforecasting AT naylenefraccanabbia bayesianoptimizedechostatenetworkappliedtoshorttermloadforecasting AT vivianacoccomariani bayesianoptimizedechostatenetworkappliedtoshorttermloadforecasting AT leandrodossantoscoelho bayesianoptimizedechostatenetworkappliedtoshorttermloadforecasting |