Engineering Commons^™

Deep Neural Networks As Time Series Forecasters Of Energy Demand, Gregory Merkel

Master's Theses (2009 -)

Short-term load forecasting is important for the day-to-day operation of natural gas utilities. Traditionally, short-term load forecasting of natural gas is done using linear regression, autoregressive integrated moving average models, and artificial neural networks. Many purchasing and operating decisions are made using these forecasts, and there can be high cost to both natural gas utilities and their customers if the short-term load forecast is inaccurate. Therefore, the GasDay lab continues to explore new ways to make better forecasts. Recently, deep neural networks (DNNs) have emerged as a powerful tool in machine learning problems. DNNs have been shown to greatly outperform …

Blending As A Multi-Horizon Time Series Forecasting Tool, Tian Gao

Master's Theses (2009 -)

Every day, millions of cubic feet of natural gas is transported through interstate pipelines and consumed by customers all over the United States of America. Gas distributors, responsible for sending natural gas to individual customers, are eager for an estimate of how much natural gas will be used in the near future. GasHour software, a reliable forecasting tool from the Marquette University GasDay lab, has been providing highly accurate hourly forecasts over the past few years. Our goal is to improve current GasHour forecasts, and my thesis presents an approach to achieve that using a blending technique. This thesis includes …

Using Evolutionary Programming To Increase The Accuracy Of An Ensemble Model For Energy Forecasting, James Gramz

Master's Theses (2009 -)

Natural gas companies are always trying to increase the accuracy of their forecasts. We introduce evolutionary programming as an approach to forecast natural gas demand more accurately. The created Evolutionary Programming Engine and Evolutionary Programming Ensemble Model use the current GasDay models, along with weather and historical ow to create an overall forecast for the amount of natural gas a company will need to supply to their customers on a given day. The existing ensemble model uses the GasDay component models and then tunes their individual forecasts and combines them to create an overall forecast. The inputs into the Evolutionary …

Automation Of Energy Demand Forecasting, Sanzad Siddique

Master's Theses (2009 -)

Automation of energy demand forecasting saves time and effort by searching automatically for an appropriate model in a candidate model space without manual intervention. This thesis introduces a search-based approach that improves the performance of the model searching process for econometrics models. Further improvements in the accuracy of the energy demand forecasting are achieved by integrating nonlinear transformations within the models. This thesis introduces machine learning techniques that are capable of modeling such nonlinearity. Algorithms for learning domain knowledge from time series data using the machine learning methods are also presented. The novel search based approach and the machine learning …

Predictive Pattern Discovery In Dynamic Data Systems, Wenjing Zhang

Dissertations (1934 -)

This dissertation presents novel methods for analyzing nonlinear time series in dynamic systems. The purpose of the newly developed methods is to address the event prediction problem through modeling of predictive patterns. Firstly, a novel categorization mechanism is introduced to characterize different underlying states in the system. A new hybrid method was developed utilizing both generative and discriminative models to address the event prediction problem through optimization in multivariate systems.

Secondly, in addition to modeling temporal dynamics, a Bayesian approach is employed to model the first-order Markov behavior in the multivariate data sequences. Experimental evaluations demonstrated superior performance over conventional …

Disaggregating Time Series Data For Energy Consumption By Aggregate And Individual Customer, Steven Vitullo

Dissertations (1934 -)

This dissertation generalizes the problem of disaggregating time series data and describes the disaggregation problem as a mathematical inverse problem that breaks up aggregated (measured) time series data that is accumulated over an interval and estimates its component parts.

We describe five different algorithms for disaggregating time series data: the Naive, Time Series Reconstruction (TSR), Piecewise Linear Optimization (PLO), Time Series Reconstruction with Resampling (RS), and Interpolation (INT). The TSR uses least squares and domain knowledge of underlying correlated variables to generate underlying estimates and handles arbitrarily aggregated time steps and non-uniformly aggregated time steps. The PLO performs an adjustment …