: This thesis applies neural network feature selection techniques to multivariate time series data to improve prediction of a target time series. Two approaches to feature selection are used. First, a subset enumeration method is used to determine which financial indicators are most useful for aiding in prediction of the S&P 500 futures daily price. The candidate indicators evaluated include RSI, Stochastics and several moving averages. Results indicate that the Stochastics and RSI indicators result in better prediction results than the moving averages. The second approach to feature selection is calculation of individual saliency metrics. A new decision boundary-based individual saliency metric, and a classifier independent saliency metric are developed and tested. Ruck's saliency metric, the decision boundary based saliency metric, and the classifier independent saliency metric are compared for a data set consisting of the RSI and Stochastics indicators as well as delayed closing price values. The decision based metric and the Ruck metric results are similar, but the classifier independent metric agrees with neither of the other metrics. The nine most salient features, determined by the decision boundary based metric, are used to train a neural network and the results are presented and compared to other published results. (AN)

/pdf/nonlinear-time-series-analysis-1j3a35n4y3.pdf

Nonlinear Time Series Analysis.

The progress of technology on energy and IoT fields has led to an increasingly complicated electric environment in low-voltage local microgrid, along with the extensions of electric vehicle, micro-generation, and local storage. It is required to establish a home energy management system (HEMS) to efficiently integrate and manage household energy micro-generation, consumption and storage, in order to realize decentralized local energy systems at the community level. Domestic power demand prediction is of great importance for establishing HEMS on realizing load balancing as well as other smart energy solutions with the support of IoT techniques. Artificial neural networks with various network types (e.g., DNN, LSTM/GRU based RNN) and other configurations are widely utilized on energy predictions. However, the selection of network configuration for each research is generally a case by case study achieved through empirical or enumerative approaches. Moreover, the commonly utilized network initialization methods assign parameter values based on random numbers, which cause diversity on model performance, including learning efficiency, forecast accuracy, etc. In this paper, an evolutionary ensemble neural network pool (EENNP) method is proposed to achieve a population of well-performing networks with proper combinations of configuration and initialization automatically. In the experimental study, power demand predictions of multiple households are explored in three application scenarios: optimizing potential network configuration set, forecasting single household power demand, and refilling missing data. The impacts of evolutionary parameters on model performance are investigated. The experimental results illustrate that the proposed method achieves better solutions on the considered scenarios. The optimized potential network configuration set using EENNP achieves a similar result to manual optimization. The results of household demand prediction and missing data refilling perform better than the naive and simple predictors.

https://www.mdpi.com/1424-8220/19/3/721/pdf

Household Power Demand Prediction Using Evolutionary Ensemble Neural Network Pool with Multiple Network Structures.

The global popularization of electric vehicles (EVs) poses an opportunity for the construction of micro-grid and smart community within energy internet on competent the massive and concentrated energy switching and routing in the local environment. Smart EV charging is a promising solution to manage EV charging load that relies on an accurate prediction of EV charging demands. Evaluation of household EV charging demand is a primary factor for designing smart EV charging solutions on household and neighbourhood level. However, this subject has not been adequately discussed in the research community. In this paper, several widely used machine learning algorithms are adopted to forecast the household day-ahead EV charging occurrence-time and the "no charge" day respectively. The performance of the algorithms are evaluated and compared. A two-layer hybrid stacking ensemble learning method is also proposed, that combines multiple machine learning algorithms. This proposed model is demonstrated to achieve a better classification performance than individual algorithms.

Household EV Charging Demand Prediction Using Machine and Ensemble Learning

The aim of this paper is to present the results of the competition of time series forecasting using soft computing methods that was organized within the IEEE WCCI 2016 congress.

On the results and observations of the time series forecasting competition CIF 2016

With the maturation of technologies such as communication, sensors and networks, very large amounts of data are generated that are available for processing and analysis. With the popularizing process of the Internet of Things (IoT), available data resources will become even more plentiful and diversified in the near future. To provide feasible solutions in data science for the key features of the energy internet, such as energy interconnection and routing, a big data architecture could be utilized in the energy internet infrastructure to provide large-scale analysis of massive various types of data. In this chapter, the utilization of big data in the energy internet infrastructure is explored. A three-layer big data architecture for usage in the energy internet is presented. The characteristics of data utilized in the energy internet and the potential requirements of the energy internet for the big data architecture are studied. Then, analytics methods that could be executed in the energy internet big data infrastructure are introduced. Real-time and offline analyses, as two types of analysis modes for different requirements of application scenarios, are described. Several well-known open-source big data tools are discussed. In addition, the open challenges of utilizing big data in the energy internet are proposed.

Utilization of Big Data in Energy Internet Infrastructure

We propose hybrid ensemble models for time series forecasting. A hybrid ensemble combines the output of several different models by a weighted mean that forms the final forecast. The final hybrid ensemble model consists of several individual models: A nearest neighbor/trajectory ensemble model, a feed-forward neural network ensemble, a trend cycle model, an autoregressive model and an ensemble model based on the returns of the time series. The best performing models with respect to a left-out part of the time series are selected by cross-validation and are combined by taking the inverse SMAPE prediction error as a weight for the combination of the respective single forecasts. We show the application of this approach in the WCCI 2016 CIF challenge.

Joerg D. Wichard

Papers

An adaptive forecasting strategy with hybrid ensemble models