Electric load forecasting for large office building based on radial basis function neural network

Mai, Weijie; Chung, C.Y.; Wu, Ting; Huang, Huazhang

doi:10.1109/pesgm.2014.6939378

Cited by 32 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent research on RBFN modeling applied to time series prediction can be found, for instance, in the work of Chang [44], where RBFN models are used to produce short-term forecasts for wind power generation. Other recent examples include the following: in Sermpinis, Theofilatos, Karathanasopoulos, Georgopoulos, and Dunis [45], RBFN-PSO hybrid models are employed for financial time series prediction; Yin, Zou, and Xu [46] use RBFN models to predict tidal waves on Canada's west coast; Niu and Wang [47] employ gradient-descenttrained RBFNs for financial time series forecasting; Mai, Chung, Wu, and Huang [48] use RBFNs to forecast electric load in office buildings; and Zhu, Cao, and Zhu [49] employ RBFNs to predict traffic flow at some street intersections.…”

Section: How Do We Improve Radial Basis Function Network?mentioning

confidence: 99%

A Literature Review on Some Trends in Artificial Neural Networks for Modeling and Simulation with Time Series

Muñoz-Zavala,

Macías-Díaz,

Alba-Cuéllar

et al. 2024

Algorithms

View full text Add to dashboard Cite

This paper reviews the application of artificial neural network (ANN) models to time series prediction tasks. We begin by briefly introducing some basic concepts and terms related to time series analysis, and by outlining some of the most popular ANN architectures considered in the literature for time series forecasting purposes: feedforward neural networks, radial basis function networks, recurrent neural networks, and self-organizing maps. We analyze the strengths and weaknesses of these architectures in the context of time series modeling. We then summarize some recent time series ANN modeling applications found in the literature, focusing mainly on the previously outlined architectures. In our opinion, these summarized techniques constitute a representative sample of the research and development efforts made in this field. We aim to provide the general reader with a good perspective on how ANNs have been employed for time series modeling and forecasting tasks. Finally, we comment on possible new research directions in this area.

show abstract

Section: How Do We Improve Radial Basis Function Network?mentioning

confidence: 99%

A Literature Review on Some Trends in Artificial Neural Networks for Modeling and Simulation with Time Series

Muñoz-Zavala,

Macías-Díaz,

Alba-Cuéllar

et al. 2024

Algorithms

View full text Add to dashboard Cite

show abstract

“…Macas et al (2016) proposed a technique based on artificial neural networks (ANN) to predict total heating energy consumption, internal air temperature and aggregated thermal discomfort 12 hours ahead, for operational cost reduction of an office building. Mai et al (2014) presented an hourly electric load forecasting model for an office building based on a neural network using outdoor weather data and historical load data as inputs, proposing a simplified parameter tuning procedure. Yang et al (2014) proposed a model to predict energy consumption for a chiller using historic building operation data and weather forecast information.…”

Section: State Of the Artmentioning

confidence: 99%

A Data-Driven Modelling Approach for Large Scale Demand Profiling of Residential Buildings

Tardioli¹,

Kerrigan²,

Oates³

et al. 2017

Building Simulation Conference Proceedings

View full text Add to dashboard Cite

In this paper the traditional use of data-driven models (DDM) as forecasting tools is coupled with parametric simulation to create a building modelling framework for demand profiling of a large number of buildings of the same typology. Most studies to date utilising DDM have been conducted on single buildings, with less evidence of the role that DDM may have as a modelling technique for application at scale. The proposed methodology is based on the use of a simulation-based building energy modelling tool and a parametric simulator to create a large dataset consisting of 4096 different building model scenarios. Three DDM techniques are utilised; Support Vector Machines, Neural Networks and Generalised Linear Models, these are trained and tested using the generated simulation dataset. Results, at an hourly resolution, show that DDM approaches can correctly emulate the outputs of the building simulation software with mean absolute error ranging from 4 to 9 percent for different DDM algorithms.

show abstract

“…This can be particularly useful in economic and financial forecasting. Mai et al (2014) presented an easy-to-implement hourly electrical load forecasting model for large commercial office buildings based on a radial basis function neural network (RBFNN), using outdoor weather data and historical load data as inputs, without tedious trial-and-error parameterisation procedures. Data from a real building under different weather conditions are used to evaluate the performance of the model, and promising results are obtained showing that the proposed method can accurately predict the evolving hourly electricity load of the building.…”

Section: Introductionmentioning

confidence: 99%

Turkey Paper and Paper Products Sector Import Forecasting with Artificial Neural Networks

Eşidir,

Gür

2024

Hitit Sosyal Bilimler Dergisi

View full text Add to dashboard Cite

The paper and paper products sector is a crucial component of the Turkish economy, characterized by significant interactions with various other sectors. Türkiye imports substantial amounts of paper, playing a vital role in the growth and sustainability of this sector. Accurate import forecasting is essential for strategic planning and resource management. This study aims to forecast the imports of the Turkish paper sector for the period from April 2023 to March 2024 using two artificial neural network (ANN) models: Multilayer Perceptron (MLP) and Radial Basis Function (RBF). The dataset, obtained from the Turkish Statistical Institute (TurkStat), covers 219 months of data from 2005 to 2023. The dependent variable is Türkiye’s monthly import value of paper and paper products, while the independent variables include the monthly average US Dollar exchange rate, monthly imports of Türkiye, the Manufacturing Industry Production Index, the Paper Production Index, and the monthly exports of paper and paper products from Türkiye. The MLP model forecasts that the monthly imports of paper and paper products will range between 270 to 300 million USD, while the RBF model predicts values between 268 and 321 million USD. These findings underscore the efficacy of ANNs in providing accurate and reliable forecasts. This study addresses a gap in the literature by applying ANN methods to forecast imports in the paper and paper products sector, presenting a novel approach that can assist companies in making better-informed decisions regarding inventory management, production planning, and marketing strategies. By leveraging the advanced computational power and pattern recognition capabilities of ANNs, the study aims to enhance the strategic planning processes in the paper and paper products industry. The traditional methods often used in trade data analysis and forecasting are limited in capturing the complex and non-linear relationships present in economic data. This study's application of ANNs offers a significant advancement by utilizing models that can better handle such complexities. The accuracy of the MLP and RBF models highlights their potential as valuable tools for economic forecasting, providing insights that are crucial for optimizing supply chain operations and improving market responsiveness. The results indicate that companies can achieve better operational performance and increased customer satisfaction by effectively forecasting future import requirements. The originality of this study lies in its methodological approach, utilizing ANN models to forecast import values in a sector where traditional methods have been predominant. This innovative approach not only contributes to the existing body of knowledge but also offers practical applications for businesses within the sector. The detailed analysis of the data, combined with the robust modeling techniques employed, provides a comprehensive framework for understanding the dynamics of paper imports and making strategic decisions based on accurate predictions. In conclusion, the study demonstrates the significant success of artificial neural networks in predicting import values for the Turkish paper and paper products sector. The findings provide valuable information that can aid companies in strategic planning, enhancing their ability to manage inventory, plan production, and develop effective market strategies. The research contributes to the literature by filling a gap with its innovative approach, offering new perspectives and practical applications for improving decision-making processes in the industry.

show abstract

Electric load forecasting for large office building based on radial basis function neural network

Cited by 32 publications

References 23 publications

A Literature Review on Some Trends in Artificial Neural Networks for Modeling and Simulation with Time Series

A Literature Review on Some Trends in Artificial Neural Networks for Modeling and Simulation with Time Series

A Data-Driven Modelling Approach for Large Scale Demand Profiling of Residential Buildings

Turkey Paper and Paper Products Sector Import Forecasting with Artificial Neural Networks

Contact Info

Product

Resources

About