A5038525019- Energy Load and Power Forecasting
- Neural Networks and Applications
- Stock Market Forecasting Methods
- Machine Learning and ELM
- Forecasting Techniques and Applications
- Grey System Theory Applications
- Image and Signal Denoising Methods
- Renewable energy and sustainable power systems
- Electric Power System Optimization
- Face and Expression Recognition
- Time Series Analysis and Forecasting
- Domain Adaptation and Few-Shot Learning
- Smart Grid Energy Management
- Machine Learning and Data Classification
- Hydrological Forecasting Using AI
- Mining and Industrial Processes
- Metaheuristic Optimization Algorithms Research
- Artificial Immune Systems Applications
- Water Systems and Optimization
- Anomaly Detection Techniques and Applications
- Optimal Power Flow Distribution
- Transportation Systems and Safety
- Geoscience and Mining Technology
- Non-Destructive Testing Techniques
- Geotechnical and Mining Engineering
University of Łódź
2024-2025
Częstochowa University of Technology
2016-2025
Kazimierz Wielki University in Bydgoszcz
2020
TOMSAD Tomasz Sadowski (Poland)
2020
Institute of Power Engineering
2014-2018
Institute of Computer Science
2017
This work presents a hybrid and hierarchical deep learning model for midterm load forecasting. The combines exponential smoothing (ETS), advanced long short-term memory (LSTM), ensembling. ETS extracts dynamically the main components of each individual time series enables to learn their representation. Multilayer LSTM is equipped with dilated recurrent skip connections spatial shortcut path from lower layers allow better capture long-term seasonal relationships ensure more efficient...
Random forest (RF) is one of the most popular machine learning (ML) models used for both classification and regression problems. As an ensemble model, it demonstrates high predictive accuracy low variance, while being easy to learn optimize. In this study, we use RF short-term load forecasting (STLF), focusing on data representation training modes. We consider seven methods defining input patterns three modes: local, global extended global. also investigate key hyperparameters about their...
The decomposition of a time series is an essential task that helps to understand its very nature. It facilitates the analysis and forecasting complex expressing various hidden components such as trend, seasonal components, cyclic irregular fluctuations. Therefore, it crucial in many fields for decision-making processes. In recent years, methods have been developed, which extract reveal different properties. Unfortunately, they neglect important property, i.e., variance. To deal with...
Many forecasting models are built on neural networks. The key issues in these models, which strongly translate into the accuracy of forecasts, data representation and decomposition problem. In this work, we consider both problems using short-term electricity load demand as an example. A time series expresses trend multiple seasonal cycles. To deal with multi-seasonality, four methods problem decomposition. Depending degree, is split local subproblems modeled We move from global model,...
Pattern similarity-based frameworks are widely used for classification and regression problems. Repeated, similar-shaped cycles observed in seasonal time series encourage the use of such forecasting. In this paper, we pattern models mid-term load An integral part these is patterns sequences representation. representation ensures input output data unification through trend filtering variance equalization. This simplifies forecasting problem allows us to based on similarity. We consider four...
Short-term load forecasting (STLF) is challenging due to complex time series (TS) which express three seasonal patterns and a nonlinear trend. This article proposes novel hybrid hierarchical deep-learning (DL) model that deals with multiple seasonality produces both point forecasts predictive intervals (PIs). It combines exponential smoothing (ES) recurrent neural network (RNN). ES extracts dynamically the main components of each individual TS enables on-the-fly deseasonalization,...
Forecasting cryptocurrency volatility can help investors make better-informed investment decisions in order to minimize risks and maximize potential profits. Accurate forecasting of price fluctuations is crucial for effective portfolio management contributes the stability financial system by identifying threats developing risk strategies. The objective this paper provide a comprehensive study statistical machine learning methods predicting daily weekly following four cryptocurrencies:...
A new multiclass classifier based on immune system principles is proposed. The unique feature of this the embedded property local selection. This method selection was inspired by binding an antibody to antigen, which occurs between amino acid residues forming epitope and a paratope. Only certain selected (so-called energetic residues) take part in binding. Antibody receptors are formed during clonal process. Antibodies (recognizing) with most antigens (instances) create memory set. set can...
In this paper, a new forecasting model based on artificial immune system (AIS) is proposed. The used for short-term electrical load as an example of time series with multiple seasonal cycles. Artificial learns to recognize antigens (AGs) representing two fragments the series: 1) fragment preceding forecast (input vector) and 2) forecasted (output vector). Antibodies recognition units AGs by selected features input vectors learn output vectors. test procedure, AG only vector recognized some...
This paper introduces the Hierarchical Kolmogorov-Arnold Network (HKAN), a novel network architecture that offers competitive alternative to recently proposed (KAN). Unlike KAN, which relies on backpropagation, HKAN adopts randomized learning approach, where parameters of its basis functions are fixed, and linear aggregations optimized using least-squares regression. utilizes hierarchical multi-stacking framework, with each layer refining predictions from previous one by solving series...
Computational intelligence (CI) and machine learning (ML) have evolved into foundational pillars of modern data-driven research, with growing impacts across domains such as engineering, medicine, finance, environmental science [...]
A modern power system is a complex network of interconnected components, such as generators, transmission lines, and distribution subsystems, that are designed to provide electricity consumers in an efficient reliable manner [...]
In this paper, we propose a new short-term load forecasting (STLF) model based on contextually enhanced hybrid and hierarchical architecture combining exponential smoothing (ES) recurrent neural network (RNN). The is composed of two simultaneously trained tracks: the context track main track. introduces additional information to It extracted from representative series dynamically modulated adjust individual forecasted by RNN consists multiple layers stacked with dilations equipped recently...
Rheumatoid arthritis (RA) is an autoimmune disease characterized by chronic inflammation affecting up to 2.0% of adults around the world. The molecular background RA has not yet been fully elucidated, but classified as a in which genetic one most significant risk factors. One hallmark impaired DNA repair observed patient-derived peripheral blood mononuclear cells (PBMCs). aim this study was correlate phenotype defined efficiency double-strand break (DSB) with genotype limited...