Benchmarking of Load Forecasting Methods Using Residential Smart Meter Data

Sousa, João C.; Bernardo, Hermano

doi:10.3390/app12199844

Cited by 5 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, the training and optimization of complex ANN models demand significant computational resources and time [64]. Sensitivity to hyperparameters necessitates careful tuning, and issues with generalization may arise if the training dataset lacks representativeness [65]. While ANNs can capture correlations, their limited ability to infer causality requires additional analysis.…”

Section: Multi-objective Optimization Of the Outputs Of The Annmentioning

confidence: 99%

Modeling and Optimization of Herb-Fortified Fresh Kombucha Cheese: An Artificial Neural Network Approach for Enhancing Quality Characteristics

Lončar,

Pezo,

Iličić

et al. 2024

Foods

View full text Add to dashboard Cite

In this study, an Artificial Neural Network (ANN) model is used to solve the complex task of producing fresh cheese with the desired quality parameters. The study focuses on kombucha fresh cheese samples fortified with ground wild thyme, supercritical fluid extract of wild thyme, ground sage and supercritical fluid extract of sage and optimizes the parameters of chemical composition, antioxidant potential and microbiological profile. The ANN models demonstrate robust generalization capabilities and accurately predict the observed results based on the input parameters. The optimal neural network model (MLP 6-10-16) with 10 neurons provides high r2 values (0.993 for training, 0.992 for testing, and 0.992 for validation cycles). The ANN model identified the optimal sample, a supercritical fluid extract of sage, on the 20th day of storage, showcasing specific favorable process parameters. These parameters encompass dry matter, fat, ash, proteins, water activity, pH, antioxidant potential (TP, DPPH, ABTS, FRAP), and microbiological profile. These findings offer valuable insights into producing fresh cheese efficiently with the desired quality attributes. Moreover, they highlight the effectiveness of the ANN model in optimizing diverse parameters for enhanced product development in the dairy industry.

show abstract

Section: Multi-objective Optimization Of the Outputs Of The Annmentioning

confidence: 99%

Modeling and Optimization of Herb-Fortified Fresh Kombucha Cheese: An Artificial Neural Network Approach for Enhancing Quality Characteristics

Lončar,

Pezo,

Iličić

et al. 2024

Foods

View full text Add to dashboard Cite

show abstract

“…The benchmark methods serve as crucial reference points in fuel consumption forecasting, providing intuitive approaches for comparison with more sophisticated models [27,28]. Benchmark approaches constitute foundational methodologies in time series forecasting, characterized by their simplicity and practicality.…”

Section: Benchmark Modelsmentioning

confidence: 99%

Fueling the Future: A Comprehensive Analysis and Forecast of Fuel Consumption Trends in U.S. Electricity Generation

Bhuiyan,

Sakib,

Alawee

et al. 2024

Sustainability

View full text Add to dashboard Cite

The U.S. Energy Information Administration (EIA) provides crucial data on monthly and annual fuel consumption for electricity generation. These data cover significant fuels, such as coal, petroleum liquids, petroleum coke, and natural gas. Fuel consumption patterns are highly dynamic and influenced by diverse factors. Understanding these fluctuations is essential for effective energy planning and decision making. This study outlines a comprehensive analysis of fuel consumption trends in electricity generation. Utilizing advanced statistical methods, including time series analysis and autocorrelation, our objective is to uncover intricate patterns and dependencies within the data. This paper aims to forecast fuel consumption trends for electricity generation using data from 2015 to 2022. Several time series forecasting models, including all four benchmark methods (Mean, Naïve, Drift, and seasonal Naïve), Seasonal and Trend Decomposition using Loess (STL), exponential smoothing (ETS), and the Autoregressive Integrated Moving Average (ARIMA) method, have been applied. The best-performing models are determined based on Root Mean Squared Error (RMSE) values. For natural gas (NG) consumption, the ETS model achieves the lowest RMSE of 20,687.46. STL demonstrates the best performance for coal consumption with an RMSE of 5936.203. The seasonal Naïve (SNaïve) model outperforms the others for petroleum coke forecasting, yielding an RMSE of 99.49. Surprisingly, the Mean method has the lowest RMSE of 287.34 for petroleum liquids, but the ARIMA model is reliable for its ability to capture complex patterns. Residual plots are analyzed to assess the models’ performance against statistical parameters. Accurate fuel consumption forecasting is very important for effective energy planning and policymaking. The findings from this study will help policymakers strategically allocate resources, plan infrastructure development, and support economic growth.

show abstract

“…Accurate electricity load forecasting provides scientific theoretical support for the smart grid, like demand response, energy management, and infrastructure planning and investment. Sousa and Bernardo (2022) compared the accuracy of multivariate adaptive regression splines, random forests, and artificial neural networks to predict the load of the next day with 5,567 households' half-hourly readings. Shaukat et al (2021) carried out short-term load forecasting by different models, such as artificial neural networks.…”

Section: Electricity Consumption Patterns and Forecastmentioning

confidence: 99%

Review of application of high frequency smart meter data in energy economics and policy research

Zhang

Qiu

2023

Front. Sustain. Energy Policy

View full text Add to dashboard Cite

The rapid popularization of advanced metering infrastructure (AMI) smart meters produces customer high-frequency energy consumption data. These data provide diverse options for energy economics and policy research. In this review, we examine studies applying high frequency smart meter data to explore the overall impact of household new technology adoption and COVID-19 on energy consumption patterns. We find that high frequency smart meter data boosts the accuracy of forecasting models with various data-driven algorithms. In addition, there is a lack of precise assessment and inclusive understanding of energy poverty in advanced economics. Smart meter data help expand and deepen the energy poverty research. Research on how vulnerable groups exhibit energy poverty can improve society's understanding of energy poverty and help implement related policy assistance programs.

show abstract

Benchmarking of Load Forecasting Methods Using Residential Smart Meter Data

Cited by 5 publications

References 24 publications

Modeling and Optimization of Herb-Fortified Fresh Kombucha Cheese: An Artificial Neural Network Approach for Enhancing Quality Characteristics

Modeling and Optimization of Herb-Fortified Fresh Kombucha Cheese: An Artificial Neural Network Approach for Enhancing Quality Characteristics

Fueling the Future: A Comprehensive Analysis and Forecast of Fuel Consumption Trends in U.S. Electricity Generation

Review of application of high frequency smart meter data in energy economics and policy research

Contact Info

Product

Resources

About