The ability to accurately predict lithium-ion battery life-time already at an early stage of battery usage is critical for ensuring safe operation, accelerating technology development, and enabling battery second-life...
详细信息
The ability to accurately predict lithium-ion battery life-time already at an early stage of battery usage is critical for ensuring safe operation, accelerating technology development, and enabling battery second-life applications. Many models are unable to effectively predict battery life-time at early cycles due to the complex and nonlinear degrading behavior of lithium-ion batteries. In this study, two hybrid data-driven models, incorporating a traditional linear support vector regression (LSVR) and a Gaussian process regression (GPR), were developed to estimate battery life-time at an early stage, before more severe capacity fading, utilizing a data set of 124 battery cells with lifetimes ranging from 150 to 2300 cycles. Two type of hybrid models, here denoted as A and B, were proposed. For each of the models, we achieved 1.1 % (A) and 1.4 % (B) training error, and similarly, 8.3 % (A) and 8.2 % (B) test error. The two key advantages are that the error percentage is kept below 10 % and that very low error values for the training and test sets were observed when utilizing data from only the first 100 *** proposed method thus appears highly promising for predicting battery life during early cycles.
linear support vector regression(SVR) is a popular machine learning algorithm. However, as the amount of data increases, the learning procedure of SVR becomes time consuming. In this paper, we propose a mini-batch qua...
详细信息
linear support vector regression(SVR) is a popular machine learning algorithm. However, as the amount of data increases, the learning procedure of SVR becomes time consuming. In this paper, we propose a mini-batch quasi-Newton optimization algorithm to speed up the training process of linear SVR. The main idea of the proposed optimization method is to use a small set of training data to estimate the first and second order gradient information and incorporate them into the framework of the popular limited memory BFGS quasi-Newton algorithm. Some modifications have been made to the generation of correction pairs of the BFGS algorithm in order to avoid the source of noise. Experimental results show that the proposed method outperforms some state-of-art methods in both training time and generalization ability.
This research paper discusses the growing demand for electricity in the Philippines due to population growth, economic development, and the need for accurate long-term electrical load forecasting to sustain the power ...
详细信息
ISBN:
(纸本)9798350370058;9798350370164
This research paper discusses the growing demand for electricity in the Philippines due to population growth, economic development, and the need for accurate long-term electrical load forecasting to sustain the power system since most load forecasting studies focus on short-term duration. The study aims to optimize, evaluate, and compare accurate load forecasting models for the load demand of Luzon. This study used machine learning techniques for forecasting, specifically linearsupportvector Machine (linear SVM), Multi-layer Perceptron (MLP), and linearregression with XGBoost (LR-XGBoost). The objectives include identifying and constructing relevant features, optimizing hyperparameters, and comparing performances using appropriate evaluation metrics. The study is significant to the power system industry. Electrical load forecasting equipped them to create operational decisions in energy management, generation scheduling, and assessment that maintain supply and demand. The comparison will solely focus on the three models and identify how the models performed using the Luzon hourly load demand from 2013 to 2022. Based on the results, the optimized LR-XGboost is the best-performing Long-Term Load Forecasting Model for the Load Demand of Luzon with a MAPE of 4.61%, outperforming the optimized linear SVR and MLP models. The researchers recommend adding external regressors such as historical weather and economic data, performing Scenario forecasting, and adding holiday indicators to improve the performance of each model.
Three-dimensional fuzzy logic controller (3-D FLC) is a novel FLC developed for spatially distributed parameter systems. In this study, we are concerned with data-based 3-D FLC design. A nearest neighborhood clusterin...
详细信息
ISBN:
(纸本)9781424473175
Three-dimensional fuzzy logic controller (3-D FLC) is a novel FLC developed for spatially distributed parameter systems. In this study, we are concerned with data-based 3-D FLC design. A nearest neighborhood clustering algorithm is employed to extract fuzzy rules from input-output data pairs, and then an optimization algorithm based on geometric similarity measure is used to reduce the obtained rule base. The consequent parameters are estimated using linear support vector regression. Finally, a catalytic packed-bed reactor is taken as an application to demonstrate the effectiveness of the 3-D FLC.
At first, a linear support vector regression feature extraction algorithm was introduced concisely. Then two improvements were presented in order that a simply explicit nonlinear regress function can be gotten easily ...
详细信息
ISBN:
(纸本)9780769535616
At first, a linear support vector regression feature extraction algorithm was introduced concisely. Then two improvements were presented in order that a simply explicit nonlinear regress function can be gotten easily by SVR feature extraction. One improvement was to decrease the dimensions of input space at the expense of regression function accuracy. Another improvement was to map the linear space to polynomial space corresponding to input features. The order of polynomial space depends on practical applications. Experimental result showed the efficiency of the improvements.
At first,a linear support vector regression feature extraction algorithm was introduced *** two improvements were presented in order that a simply explicit nonlinear regress function can be gotten easily by SVR featur...
详细信息
At first,a linear support vector regression feature extraction algorithm was introduced *** two improvements were presented in order that a simply explicit nonlinear regress function can be gotten easily by SVR feature *** improvement was to decrease the dimensions of input space at the expense of regression function *** improvement was to map the linear space to polynomial space corresponding to input *** order of polynomial space depends on practical *** result showed the efficiency of the improvements.
暂无评论