A5083366817- Energy Load and Power Forecasting
- Smart Grid Energy Management
- Air Quality Monitoring and Forecasting
- Water Quality Monitoring and Analysis
- Transportation and Mobility Innovations
- Solar Radiation and Photovoltaics
- Building Energy and Comfort Optimization
- Electric Vehicles and Infrastructure
- Energy Efficiency and Management
- Air Quality and Health Impacts
- Image Retrieval and Classification Techniques
- Integrated Energy Systems Optimization
- Hydrological Forecasting Using AI
- Vehicle emissions and performance
- Water Quality Monitoring Technologies
- Advanced Chemical Sensor Technologies
- AI in cancer detection
- Image and Signal Denoising Methods
- Electric Power System Optimization
- Municipal Solid Waste Management
- Optimal Power Flow Distribution
- Forecasting Techniques and Applications
- Power Quality and Harmonics
- Image Processing and 3D Reconstruction
- Statistical and Computational Modeling
University of Eastern Finland
2014-2024
Finland University
2016-2023
Tampere University
2018
International Society for Optics and Photonics
2015
Society for Imaging Science and Technology
2015
Kuopio University Hospital
2005
The effective use of solar photovoltaic (PV) installations implies the integration PV output into overall energy consumption planning, optimization, and control. Moreover, day-ahead trading electricity in Europe makes forecasting utterly important, thus its accuracy becomes particular interest. Data-driven models are typically trained using numerical weather prediction (NWP) data, availability which represents one main obstacles modeling. In this study, we investigate an alternative...
Parametric and nonparametric modeling methods have been widely used for the estimation of forest attributes from airborne laser-scanning data aerial photographs. However, adopted suffered complex remote-sensed structures involving high dimensions, nonlinear relationships, different statistical distributions, outliers. In this context, artificial neural networks (ANNs) are interest as they many clear benefits over conventional could then enhance accuracy current forest-inventory methods. This...
Abstract. In order to have a good estimate of the current forcing by anthropogenic aerosols, knowledge on past aerosol levels is needed. Aerosol optical depth (AOD) measure for loading. However, dedicated measurements AOD are only available from 1990s onward. One option lengthen time series beyond retrieve surface solar radiation (SSR) taken with pyranometers. this work, we evaluated several inversion methods designed task. We compared look-up table method based radiative transfer modelling,...
When identifying and comparing forecasting models, there may be a risk that poorly selected criteria could lead to wrong conclusions. Thus, it is important know how sensitive the results are selection of criteria. This contribution aims study sensitivity identification comparison choice It compares typically applied for tuning performance assessment load methods with estimated costs caused by errors. The focus on short-term loads energy systems. comprise electricity market network costs. We...
Short-term forecasting of electric loads is an essential function required by Smart Grids. Today increasing amount smart metering data available enabling the development enhanced data-driven models for short-term load forecasting. Until now, a plethora have been developed ranging from simple linear regression to more advanced such as (artificial) neural networks (NNs) and support vector machines (SVMs). Despite relatively high accuracy obtained, acceptance purely NN still remained limited...
This paper aims to introduce a predictive weather-based control policy for the microgrid energy management improve resilience of microgrid. relies on application machine learning models prediction load demand and solar production supply interruption in upstream distribution network. The predictions serve as an input multiobjective chance constraint optimization that balances economic objectives based probability interruption. are made with decision-tree-based model can predict upcoming...
Accurate prediction of energy consumption in district heating systems plays an important role supporting effective and clean production distribution dense urban areas. Predictive models are needed for flexible cost-effective operation usage, e.g., using peak shaving or load shifting to compensate heat losses the pipeline. This helps avoid exceedance power plant capacity. The purpose this study is automate process building machine learning (ML) solve a short-term demand problem. dataset...
In this paper, the performance of three Distance Transform on Curved Space-based features derived from digital H&E stained oncopathological images used in breast cancer pattern recognition scheme are compared. The utilized SW-DTOCS, SW-WDTOCS and SW-3-4-DTOCS with different sliding window (SW) sizes. results imply that distance transform yield slightly classification performance. addition, issue computer-aided histopathological diagnosis is discussed.
Electrical energy consumption is undergoing major changes driven by several factors. Trends in electric vehicle (EV) purchases and heating system conversion indicate that electricity demand can be significant between today year 2030. For instance Finland, the target for EVs 250 000 passenger cars At same time, a number of heat pumps (HPs) will installed detached houses replacing old systems such as oil-fired boilers. In this paper, effects HPs on Finnish rural areas are modeled analyzed.
Accurate forecasting of loads is essential for smart grids and energy markets. This paper compares the performance following models in short-term load forecasting: 1) metering data based profile models, 2) a neural network (NN) model, 3) Kalman-filter predictor with input nonlinearities physically main structure. The comparison helps method selection development hybrid control responses. According to results all these three modeling approaches show much better than 4) traditional profiles 5)...
Modelling of controllable loads is a necessary function required by demand side management, and specifically load control smart grids. A large amount metering data other supporting are available, enabling the development new, intelligent data-driven fashions for recognising modelling loads. However, it challenge to extract useful information from this massive, often aggregated in reliable understandable fashion. In paper we present approach, heating small customers. Main computational...
We describe a neural network model of municipal wastewater treatment plant (WWTP) in which on-line total solids (TS) sewer data generated by novel microwave sensor is used as input variable. The predictive performance the compared with and without modelling traditional linear multiple regression (MLR) model. In addition, benefits using networks are discussed. According to our results, based MLP (multilayer perceptron) provides better estimate than corresponding MLR WWTP effluent TS load....
Predicting weather-related outages in electricity networks is an important issue for distribution system operators. In this study, we apply a data-driven approach and train artificial neural to predict faults the network. our experiments, utilize meteorological data fault records collected period of1.1.2011-31.12.2013 central Finland. Assuming that there might be long-term dependencies between weather conditions network, investigate simple recurrent networks, long short-term memory...
The building sector is a major energy consumer and CO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emitter, being responsible for approximately 40% of the total consumption in EU. Active demand side participation electricity customers seen as crucial management reduction sector's emissions. However, today's markets are often lacking strong incentives active participation. Understandable customer specific comparison information...
A hybrid model for short-term forecasting of aggregated thermal loads and their load control responses is studied in this paper using field test data. Inputs include temperature measurement forecast, measured power signals. The comprises 1) partly physically based the controlled non-controlled power, 2) residual Support Vector Machine (SVM). Their summation gives hourly interval forecast. Here means that structure models dynamics houses. response needs as inputs daily heating energy demand...