Statistical Analysis of Neural Networks as Applied to Building Energy Prediction

Dodier, Robert H.; Henze, Gregor P.

doi:10.1115/1.1637640

Cited by 60 publications

(20 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their output is a linear o non-linear function of the inputs and, therefore, they have been widely used for predicting non-linear data (as in STLF [8], [3], [6], [9], [10], [11]). After many tests, we obtained the best results with a NN design including the temperature-related variables, the value of the previous hour (independently of the day type), and the value of the same hour in the previous same-type day.…”

Section: ) Neural Network (Nn)mentioning

confidence: 99%

Short-term load forecasting in non-residential Buildings

2011

View full text Add to dashboard Cite

Abstract-Short-term load forecasting (STLF) has become an essential tool in the electricity sector. It has been object of vast research since energy load is known to be non-linear and, therefore, very difficult to predict with accuracy. We focus here on non-residential building STLF, an special case of STLF where weather shows smaller influence on the load than in normal scenarios and forecast models, contrary to those on the literature, are required to be simple, avoiding dull and complicated trialand-error parametrisation or setting-up processes. Under these premises, we have used a two-step methodology comprising a classification and a adjustment steps. Since the non-linearity of the load is associated to the activity in the building, we have demonstrated that the best way to deal with it is using the work day schedule as day-type classifier. Moreover, we have evaluated a number of statistical methods and Artificial Intelligence methods to adjust the typical hourly consumption curve, concluding that an autoregressive time series suffices to fulfil the requirements, even in a 5 day-ahead horizon.

show abstract

Section: ) Neural Network (Nn)mentioning

confidence: 99%

Short-term load forecasting in non-residential Buildings

2011

View full text Add to dashboard Cite

show abstract

“…A universally-binding standard for modular buildings, that factors geographic location, is clearly needed. Although often pricy, integrative 3D modeling software and project management software, which enable prompt sharing of designs, information, and results, are crucial to the success of prefab; more so are multi-objective algorithms that use mathematical approaches to solve real-time challenges [74], such as artificial neural networks used to predict the energy use of buildings [75]. Numerous projects incorporating prefab (on various levels) have already been completed successfully, and many more are planned.…”

Section: Future Pathwaysmentioning

confidence: 99%

Performance of Modular Prefabricated Architecture: Case Study-Based Review and Future Pathways

2016

View full text Add to dashboard Cite

Even though tightened building energy efficiency standards are implemented periodically in many countries, existing buildings continually consume a momentous quota of the total primary energy. Energy efficiency solutions range from material components to bulk systems. A technique of building construction, referred to as prefabricated architecture (prefab), is increasing in reputation. Prefab encompasses the offsite fabrication of building components to a greater degree of finish as bulk building structures and systems, and their assembly on-site. In this context, prefab improves the speed of construction, quality of architecture, efficiency of materials, and worker safety, while limiting environmental impacts of construction, as compared to conventional site-built construction practices. Quite recently, a 57 story skyscraper was built in 19 days using prefabricated modules. From the building physics point of view, the bulk systems and tighter integration method of prefab minimizes thermal bridges. This study seeks to clearly characterize the levels of prefab and to investigate the performance of modular prefab; considering acoustic constrain, seismic resistance, thermal behavior, energy consumption, and life cycle analysis of existing prefab cases and, thus, provides a dynamic case study-based review. Generally, prefab can be categorized into components, panels (2D), modules (3D), hybrids, and unitized whole buildings. On average, greenhouse gas emissions from conventional construction were higher than for modular construction, not discounting some individual discrepancies. Few studies have focused on monitored data on prefab and occupants' comfort but additional studies are required to understand the public's perception of the technology. The scope of the work examined will be of interest to building engineers, manufacturers, and energy experts, as well as serve as a foundational reference for future study.

show abstract

“…Their output is a linear o non-linear function of the inputs and, therefore, they have been widely used for predicting non-linear data (as in STLF [1], [2], [20], [21], [25], [26]). After many tests, we obtained the best results with a NN design including the temperature-related variables, the value of the previous hour (independently of the day type), and the value of the same hour in the previous same-type day.…”

Section: A Modelsmentioning

confidence: 99%

“…Regarding artificial intelligence methods, [24] addressed a SVM for predicting the load of a building complex, [21] proposed a NN tuned up by Automatic Relevance Determination in order to optimise the selected input. Moreover, [25] put forward an NN in which the input variables where selected by a version of the Wald's test.…”

Section: Introductionmentioning

confidence: 99%

Short-term load forecasting in air-conditioned non-residential Buildings

Penya

Borges

Agote

et al. 2011

2011 IEEE International Symposium on Industrial Electronics

View full text Add to dashboard Cite

Abstract-Short-term load forecasting (STLF) has become an essential tool in the electricity sector. It has been classically object of vast research since energy load prediction is known to be nonlinear. In a previous work, we focused on non-residential building STLF, an special case of STLF where weather has negligible influence on the load. Now we tackle more modern buildings in which the temperature does alter its energy consumption. This is, we address here fully-HVAC (Heating, Ventilating, and Air Conditioning) ones. Still, in this problem domain, the forecasting method selected must be simple, without tedious trial-and-error configuring or parametrising procedures, work with scarce (or any) training data and be able to predict an evolving demand curve. Following our preceding research, we have avoided the inherent non-linearity by using the work day schedule as day-type classifier. We have evaluated the most popular STLF systems in the literature, namely ARIMA (autoregressive integrated moving average) time series and Neural networks (NN), together with an Autoregressive Model (AR) time series and a Bayesian network (BN), concluding that the autoregressive time series outperforms its counterparts and suffices to fulfil the addressed requirements, even in a 6 day-ahead horizon.

show abstract

Statistical Analysis of Neural Networks as Applied to Building Energy Prediction

Cited by 60 publications

References 7 publications

Short-term load forecasting in non-residential Buildings

Short-term load forecasting in non-residential Buildings

Performance of Modular Prefabricated Architecture: Case Study-Based Review and Future Pathways

Short-term load forecasting in air-conditioned non-residential Buildings

Contact Info

Product

Resources

About