Search results for: individual values

Authors: Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Abstract:

Time series data consist of successive data points collected over a period of time. Accurate prediction of future values is essential for informed decision-making in several real-world applications, including electricity load demand forecasting, lifetime estimation of industrial machinery, traffic planning, weather prediction, and the stock market. Due to their critical relevance and wide application, there has been considerable interest in time series forecasting in recent years. However, the proliferation of sensors and IoT devices, real-time monitoring systems, and high-frequency trading data introduce significant intricate temporal variations, rapid changes, noise, and non-linearities, making time series forecasting more challenging. Classical methods such as Autoregressive integrated moving average (ARIMA) and Exponential Smoothing aim to extract pre-defined temporal variations, such as trends and seasonality. While these methods are effective for capturing well-defined seasonal patterns and trends, they often struggle with more complex, non-linear patterns present in real-world time series data. In recent years, deep learning has made significant contributions to time series forecasting. Recurrent Neural Networks (RNNs) and their variants, such as Long short-term memory (LSTMs) and Gated Recurrent Units (GRUs), have been widely adopted for modeling sequential data. However, they often suffer from the locality, making it difficult to capture local trends and rapid fluctuations. Convolutional Neural Networks (CNNs), particularly Temporal Convolutional Networks (TCNs), leverage convolutional layers to capture temporal dependencies by applying convolutional filters along the temporal dimension. Despite their advantages, TCNs struggle with capturing relationships between distant time points due to the locality of one-dimensional convolution kernels. Transformers have revolutionized time series forecasting with their powerful attention mechanisms, effectively capturing long-term dependencies and relationships between distant time points. However, the attention mechanism may struggle to discern dependencies directly from scattered time points due to intricate temporal patterns. Lastly, Multi-Layer Perceptrons (MLPs) have also been employed, with models like N-BEATS and LightTS demonstrating success. Despite this, MLPs often face high volatility and computational complexity challenges in long-horizon forecasting. To address intricate temporal variations in time series data, this study introduces Times2D, a novel framework that parallelly integrates 2D spectrogram and derivative heatmap techniques. The spectrogram focuses on the frequency domain, capturing periodicity, while the derivative patterns emphasize the time domain, highlighting sharp fluctuations and turning points. This 2D transformation enables the utilization of powerful computer vision techniques to capture various intricate temporal variations. To evaluate the performance of Times2D, extensive experiments were conducted on standard time series datasets and compared with various state-of-the-art algorithms, including DLinear (2023), TimesNet (2023), Non-stationary Transformer (2022), PatchTST (2023), N-HiTS (2023), Crossformer (2023), MICN (2023), LightTS (2022), FEDformer (2022), FiLM (2022), SCINet (2022a), Autoformer (2021), and Informer (2021) under the same modeling conditions. The initial results demonstrated that Times2D achieves consistent state-of-the-art performance in both short-term and long-term forecasting tasks. Furthermore, the generality of the Times2D framework allows it to be applied to various tasks such as time series imputation, clustering, classification, and anomaly detection, offering potential benefits in any domain that involves sequential data analysis.

Keywords: derivative patterns, spectrogram, time series forecasting, times2D, 2D representation

Procedia PDF Downloads 45

Authors: Mina Naeini, Thomas A. Adams II

Abstract:

Solid Oxide Fuel Cells (SOFCs) feature high electrical efficiency and generate substantial amounts of waste heat that make them suitable for integrated community energy systems (ICEs). By harvesting and distributing the waste heat through hot water pipelines, SOFCs can meet thermal demand of the communities. Therefore, they can replace traditional gas boilers and reduce greenhouse gas (GHG) emissions. Despite these advantages of SOFCs over competing power generation units, this technology has not been successfully commercialized in large-scale to replace traditional generators in ICEs. One reason is that SOFC performance deteriorates over long-term operation, which makes it difficult to find the proper sizing of the cells for a particular ICE system. In order to find the optimal sizing and operating conditions of SOFCs in a community, a proper knowledge of degradation mechanisms and effects of operating conditions on SOFCs long-time performance is required. The simplified SOFC models that exist in the current literature usually do not provide realistic results since they usually underestimate rate of performance drop by making too many assumptions or generalizations. In addition, some of these models have been obtained from experimental data by curve-fitting methods. Although these models are valid for the range of operating conditions in which experiments were conducted, they cannot be generalized to other conditions and so have limited use for most ICEs. In the present study, a general, detailed degradation-based model is proposed that predicts the performance of conventional SOFCs over a long period of time at different operating conditions. Conventional SOFCs are composed of Yttria Stabilized Zirconia (YSZ) as electrolyte, Ni-cermet anodes, and LaSr₁₋ₓMnₓO₃ (LSM) cathodes. The following degradation processes are considered in this model: oxidation and coarsening of nickel particles in the Ni-cermet anodes, changes in the pore radius in anode, electrolyte, and anode electrical conductivity degradation, and sulfur poisoning of the anode compartment. This model helps decision makers discover the optimal sizing and operation of the cells for a stable, efficient performance with the fewest assumptions. It is suitable for a wide variety of applications. Sulfur contamination of the anode compartment is an important cause of performance drop in cells supplied with hydrocarbon-based fuel sources. H₂S, which is often added to hydrocarbon fuels as an odorant, can diminish catalytic behavior of Ni-based anodes by lowering their electrochemical activity and hydrocarbon conversion properties. Therefore, the existing models in the literature for H₂-supplied SOFCs cannot be applied to hydrocarbon-fueled SOFCs as they only account for the electrochemical activity reduction. A regression model is developed in the current work for sulfur contamination of the SOFCs fed with hydrocarbon fuel sources. The model is developed as a function of current density and H₂S concentration in the fuel. To the best of authors' knowledge, it is the first model that accounts for impact of current density on sulfur poisoning of cells supplied with hydrocarbon-based fuels. Proposed model has wide validity over a range of parameters and is consistent across multiple studies by different independent groups. Simulations using the degradation-based model illustrated that SOFCs voltage drops significantly in the first 1500 hours of operation. After that, cells exhibit a slower degradation rate. The present analysis allowed us to discover the reason for various degradation rate values reported in literature for conventional SOFCs. In fact, the reason why literature reports very different degradation rates, is that literature is inconsistent in definition of how degradation rate is calculated. In the literature, the degradation rate has been calculated as the slope of voltage versus time plot with the unit of voltage drop percentage per 1000 hours operation. Due to the nonlinear profile of voltage over time, degradation rate magnitude depends on the magnitude of time steps selected to calculate the curve's slope. To avoid this issue, instantaneous rate of performance drop is used in the present work. According to a sensitivity analysis, the current density has the highest impact on degradation rate compared to other operating factors, while temperature and hydrogen partial pressure affect SOFCs performance less. The findings demonstrated that a cell running at lower current density performs better in long-term in terms of total average energy delivered per year, even though initially it generates less power than if it had a higher current density. This is because of the dominant and devastating impact of large current densities on the long-term performance of SOFCs, as explained by the model.

Keywords: degradation rate, long-term performance, optimal operation, solid oxide fuel cells, SOFCs

Procedia PDF Downloads 135