Energy Price Forecasting Research Articles

Day‐ahead price forecasting based on hybrid prediction model

Short‐Term Price Forecast is a key issue for operation of both regulated power systems and electricity markets. Energy price forecast is the key information for generating companies to prepare their bids in the electricity markets. However, this forecasting problem is complex due to nonlinear, nonstationary, and time variant behavior of electricity price time series. So, in this article, the forecast model includes wavelet transform, autoregressive integrated moving average, and radial basis function neural networks (RBFN) is presented. Also, an intelligent algorithm is applied to optimize the RBFN structure, which adapts it to the specified training set, reduce computational complexity and avoids over fitting. Effectiveness of the proposed method is applied for price forecasting of electricity market of mainland Spain and its results are compared with the results of several other price forecast methods. These comparisons confirm the validity of the developed approach. © 2016 Wiley Periodicals, Inc. Complexity 21: 156–164, 2016

Optimal Operation of Combined Heat and Power System Based on Forecasted Energy Prices in Real-Time Markets

This paper develops a discrete operation optimization model for combined heat and powers (CHPs) in deregulated energy markets to maximize owners’ profits, where energy price forecasting is included. First, a single input and multi-output (SIMO) model for typical CHPs is established, considering the varying ratio between heat and electricity outputs at different loading levels. Then, the energy prices are forecasted with a gray forecasting model and revised in real-time based on the actual prices by using the least squares method. At last, a discrete optimization model and corresponding dynamic programming algorithm are developed to design the optimal operation strategies for CHPs in real-time. Based on the forecasted prices, the potential operating strategy which may produce the maximum profits is pre-developed. Dynamic modification is then conducted to adjust the pre-developed operating strategy after the actual prices are known. The proposed method is implemented on a 1 MW CHP on a typical day. Results show the optimized profits comply well with those derived from real-time prices after considering dynamic modification process.

Management of hybrid energy supply systems in buildings using mixed-integer model predictive control

In this paper a mixed-integer model predictive controller for hybrid energy supply systems in buildings is presented. This approach is based on a hierarchical building control concept where the energy supply level is coupled to the energy consumption level only by the heat load. The supply level is characterized by non-linear dynamics due to a stratified water storage tank and a switched heat pump with minimum on/off times. The mixed-integer model predictive controller optimizes the unit commitment problem at minimum costs while satisfying the consumption level’s predicted heat load. The hybrid system is formulated as a piecewise affine model comprising continuous and discrete system inputs. Moreover, the proposed controller is able to manage the stratified storage tank including switching sequences of the heat pump with respect to energy price forecasts. The effectiveness of this approach is shown by a comparison to a model predictive controller with an a priori fixed operation mode profile, where the heat pump is only operating at night, and discussing the effect of the variation of the stratified storage tank size. The proposed concept is able to flexibly manage all sizes of stratified storage tanks with better performance than the reference control strategy, which is only effective for larger tanks. Additionally, a robustness analyses demonstrates that the mixed-integer model predictive controller can handle errors in the heat load prediction from the consumption level. Both analyses show promising results for the practical use of the proposed controller within the hierarchical control concept or as a control module in a similar but more general application.

Testing the rationality of DOE's energy price forecasts under asymmetric loss preferences

This paper examines the rationality of the price forecasts for energy commodities of the United States Department of Energy's (DOE), departing from the common assumption in the literature that DOE's forecasts are based on a symmetric underlying loss function with respect to positive vs. negative forecast errors. Instead, we opt for the methodology of Elliott et al. (2005) that allows testing the joint hypothesis of an asymmetric loss function and rationality and reveals the underlying preferences of the forecaster. Results indicate the existence of asymmetries in the shape of the loss function for most energy categories with preferences leaning towards optimism. Moreover, we also examine whether there is a structural break in those preferences over the examined period, 1997–2012.

Use of Storage Water in a Hydroelectric System

Failure to adequately define or understand the use of storage water in hydroelectric system has created a fragmented economic environment and schizophrenia in the industry that has the potential to limit its development as a global industry. This paper suggests that, by redefining the use of storage of water in hydroelectric system as a means of using a decision support system to obtain the efficient use of storage water during low flow. The purpose of hydroelectric system optimization is to maximize the value of water resources by providing rapid and informative recommendations based on current data and dynamic forecasts of hydrology, energy prices and loads at individual hydro plants and within the overall system.

A new feature selection algorithm and composite neural network for electricity price forecasting

In a competitive electricity market, the forecasting of energy prices is an important activity for all the market participants either for developing bidding strategies or for making investment decisions. In this paper, a new forecast strategy is proposed for day ahead prediction of electricity price, which is a complex signal with nonlinear, volatile and time dependent behavior. Our forecast strategy includes a new two stage feature selection algorithm, a composite neural network (CNN) and a few auxiliary predictors. The feature selection algorithm has two filtering stages to remove irrelevant and redundant candidate inputs, respectively. This algorithm is based on mutual information (MI) criterion and selects the input variables of the CNN among a large set of candidate inputs. The CNN is composed of a few neural networks (NN) with a new data flow among its building blocks. The CNN is the forecast engine of the proposed strategy. A kind of cross-validation technique is also presented to fine-tune the adjustable parameters of the feature selection algorithm and CNN. Moreover, the proposed price forecast strategy is equipped with a few auxiliary predictors to enrich the candidate set of inputs of the forecast engine. The whole proposed strategy is examined on the PJM, Spanish and Californian electricity markets and compared with some of the most recent price forecast methods.

An empirical model of daily highs and lows of West Texas Intermediate crude oil prices

There is a large collection of literature on energy price forecasting, but most studies typically use monthly average or close-to-close daily price data. In practice, the daily price range constructed from the daily high and low also contains useful information on price volatility and is used frequently in technical analysis. The interaction between the daily high and low and the associated daily range has been examined in several recent studies on stock price and exchange rate forecasts. The present paper adopts a similar approach to analyze the behaviour of the West Texas Intermediate (WTI) crude oil price over a ten-year period. We find that daily highs and lows of the WTI oil price are cointegrated, with the error correction term being closely approximated by the daily price range. Two forecasting models, one based on a vector error correction mechanism and the other based on a transfer function framework with the range taken as a driver variable, are presented for forecasting the daily highs and lows. The results show that both of these models offer significant advantages over the naïve random walk and univariate ARIMA models in terms of out-of-sample forecast accuracy. A trading strategy that makes use of the daily high and low forecasts is further developed. It is found that this strategy generally yields very reasonable trading returns over an evaluation period of about two years.

Decision Support System for Optimizing Reservoir Operations Using Ensemble Streamflow Predictions

This paper investigates the value of ensemble streamflow predictions and energy price forecasts as aid to decision makers in scheduling the quantity and timing of reservoir releases for daily, weekly, and seasonal operations while meeting regulatory constraints. A decision support system DSS is described as it incorporates two integrated models of system operation: a simulation model that replicates general operating rules for the hydropower system and an optimization model that refines operations based upon forecasts of state variables. The DSS provides a series of recommendations for the quantity and timing of reservoir releases to optimize the economic value of the electrical energy produced, while balancing requirements and concerns related to flood control, environmental flows, and water supply. The DSS generates a range of optimal reservoir releases using an ensemble streamflow forecast and identifies robust operational solutions. The results indicate the value of the forecasts in improving system operation.

Trends in world energy prices

The correct identification of the time series path of non-renewable energy resources has far reaching consequences for economists and policymakers alike. This study builds on the existing literature by employing a data series that includes a sample period of institutional change and recently developed unit root testing procedures. Besides crude oil, natural gas and coal prices are also examined, aiming to further the knowledge of non-renewable energy resource time paths in order to inform future research and update the conclusions of past studies. The unit root tests allow for structural breaks and are based on the procedures developed by Zivot and Andrews (1992), Lumsdaine and Papell (1997) and Lee and Strazicich (2003). Finally, we investigate whether the trend changes signs in the regimes which are bounded by the structural breaks and quantify the prevalence of the trends over the sample period considered. The results show that the trend is not well represented by a single positive or negative trend. The variability of the trend suggests that forecasting energy prices should not typically occur about a single trend.

Electricity market price spike analysis by a hybrid data model and feature selection technique

In a competitive electricity market, energy price forecasting is an important activity for both suppliers and consumers. For this reason, many techniques have been proposed to predict electricity market prices in the recent years. However, electricity price is a complex volatile signal owning many spikes. Most of electricity price forecast techniques focus on the normal price prediction, while price spike forecast is a different and more complex prediction process. Price spike forecasting has two main aspects: prediction of price spike occurrence and value. In this paper, a novel technique for price spike occurrence prediction is presented composed of a new hybrid data model, a novel feature selection technique and an efficient forecast engine. The hybrid data model includes both wavelet and time domain variables as well as calendar indicators, comprising a large candidate input set. The set is refined by the proposed feature selection technique evaluating both relevancy and redundancy of the candidate inputs. The forecast engine is a probabilistic neural network, which are fed by the selected candidate inputs of the feature selection technique and predict price spike occurrence. The efficiency of the whole proposed method for price spike occurrence forecasting is evaluated by means of real data from the Queensland and PJM electricity markets.

Evaluating information in multiple horizon forecasts: The DOE's energy price forecasts

The United States Department of Energy's (DOE) quarterly price forecasts for energy commodities are examined to determine the incremental information provided at the one-through four-quarter forecast horizons. A direct test for determining information content at alternative forecast horizons, developed by Vuchelen and Gutierrez [Vuchelen, J. and Gutierrez, M.-I. “A Direct Test of the Information Content of the OECD Growth Forecasts.” International Journal of Forecasting. 21(2005):103–117.], is used. The results suggest that the DOE's price forecasts for crude oil, gasoline, and diesel fuel do indeed provide incremental information out to three-quarters ahead, while natural gas and electricity forecasts are informative out to the four-quarter horizon. In contrast, the DOE's coal price forecasts at two-, three-, and four-quarters ahead provide no incremental information beyond that provided for the one-quarter horizon. Recommendations of how to use these results for making forecast adjustments is also provided.

An Adaptive Wavelet Neural Network-Based Energy Price Forecasting in Electricity Markets

In a competitive electricity market, an accurate forecasting of energy prices is an important activity for all the market participants either for developing bidding strategies or for making investment decisions. An adaptive wavelet neural network (AWNN) is proposed in this paper for short-term price forecasting (STPF) in the electricity markets. A commonly used Mexican hat wavelet has been chosen as the activation function for hidden-layer neurons of feed-forward neural network (FFNN). To demonstrate the effectiveness of the proposed approach, day-ahead prediction of market clearing price (MCP) of Spain market, which is a duopoly market with a dominant player, and locational marginal price (LMP) forecasting in PJM electricity market, are considered. The forecasted results clearly show that AWNN has good prediction properties compared to other forecasting techniques, such as wavelet-ARIMA, multi-layer perceptron (MLP) and radial basis function (RBF) neural networks as well as recently proposed fuzzy neural network (FNN).

Energy price forecasting - problems and proposals for such predictions

This paper discusses the value of price forecasting in the electricity market during bidding or hedging against volatility. When bidding in a pool system, the market participants are requested to express their bids in terms of prices and quantities. Since the bids are accepted in order of increasing price until the total demand is met, a company that is able to forecast the pool price can adjust its own price/production schedule depending on hourly pool prices and its own production costs. This paper also discusses the challenges of price forecasting and describes some of the proposed methods for meeting these challenges.

Forecasting of electricity prices with neural networks

Abstract During recent years, the electricity energy market deregulation has led to a new free competition situation in Europe and other countries worldwide. Generators, distributors and qualified clients have some uncertainties about the future evolution of electricity markets. In consequence, feasibility studies of new generation plants, design of new systems and energy management optimization are frequently postponed. The ability of forecasting energy prices, for instance the electricity prices, would be highly appreciated in order to improve the profitability of utility investments. The development of new simulation techniques, such as Artificial Intelligence (AI), has provided a good tool to forecast time series. In this paper, it is demonstrated that the Neural Network (NN) approach can be used to forecast short term hourly electricity pool prices (for the next day and two or three days after). The NN architecture and design for prices forecasting are described in this paper. The results are tested with extensive data sets, and good agreement is found between actual data and NN results. This methodology could help to improve power plant generation capacity management and, certainly, more profitable operation in daily energy pools.

Energy Price Forecasting in the Ontario Competitive Power System Market

This paper introduces a method for forecasting energy prices using artificial intelligence methods, such as neural networks and fuzzy logic, and a combination of the two. The new approach is compared with some of the exiting methods. Various factors affecting the market clearing price are investigated. Results for the Ontario electricity market are presented.

Energy Policies and Their Consequences After 25 Years

Hans Landsberg and Sam Schurr each led research teams that produced two important energy futures policy studies that were published in 1979. The conclusions, policy recommendations, and energy demand, supply, and price forecasts contained in these studies are reviewed. Developments in U.S. energy policy over the last 25 years are discussed and compared with the recommendations contained in the two studies. The projections of energy demand, supply, and prices for 2000 contained in the studies are presented and compared to actual realizations. The nature, magnitudes, and reasons for the differences between the studies’ forecasts and what actually emerged 25 years later are discussed. All things considered, the Landsberg and Schurr studies have stood the test of time very well.

Cost estimation and energy price forecasts for economic evaluation of retrofit projects

A cost estimate can have a major impact both on project profitability and influences the technical solution. This leads to a commitment of substantial amounts of money and manpower over an extended time. Various methods often provide different results. These differences can be rather significant towards the real cost of a project and selection of the right arrangements. It is therefore important to use a proper estimate that generates enough confidence to choose the right alternative. Another source of uncertainty during the economic analysis of design and retrofit projects is the future energy price value as part of operating costs. This value can significantly affect the project viability and profitability. This paper gives a summary of the most common methods used for cost estimation of heat exchange equipment in the process industry and the sources of energy price projections. It shows the relevance of the choice of the right method and the most reliable source of energy price forecast used when choosing between alternative retrofit projects or when trying to determine the viability of a retrofit project. Ten methods for heat exchanger costing procedure are considered. Oil, natural gas and electricity price projection figures published in reports of nine established analytical centres are discussed. Web sites of institutions and companies are also included.

Life‐Cycle Energy, Costs, and Strategies for Improving a Single‐Family House

SummaryThe life‐cycle energy, greenhouse gas emissions, and costs of a contemporary 2,450 sq ft (228 m3) U.S. residential home (the standard home, or SH) were evaluated to study opportunities for conserving energy throughout pre‐use (materials production and construction), use (including maintenance and improvement), and demolition phases. Home construction and maintenance materials and appliances were inventoried totaling 306 metric tons. The use phase accounted for 91% of the total life‐cycle energy consumption over a 50‐year home life. A functionally equivalent energy‐efficient house (EEH) was modeled that incorporated 11 energy efficiency strategies. These strategies led to a dramatic reduction in the EEH total life‐cycle energy; 6,400 GJ for the EEH compared to 16,000 GJ for the SH. For energy‐efficient homes, embodied energy of materials is important; pre‐use energy accounted for 26% of life‐cycle energy. The discounted (4%) life‐cycle cost, consisting of mortgage, energy, maintenance, and improvement payments varied between 426,700 and 454,300 for a SH using four energy price forecast scenarios. In the case of the EEH, energy cost savings were offset by higher mortgage costs, resulting in total life‐cycle cost between 434,100 and 443,200. Life‐cycle greenhouse gas emissions were 1,010 metric tons CO2 equivalent for an SH and 370 metric tons for an EEH.

Forecasting energy prices in a competitive market

Deregulation is intended to encourage competition among utilities and power marketers to reduce energy prices. The pricing system is very critical in restructuring the electrical industry and depends upon market rules and the function of the independent system operator (ISO). The locational marginal pricing (LMP) system is in use in the Pennsylvania, New Jersey, Maryland (PJM) Interconnection ISO and is proposed for the New York ISO, which is expected to be in operation in 1999. This article provides fundamental concepts of the LMP and outlines the basic requirements for a computer simulation tool to accurately forecast prices in an LMP-based market. Examples are used to demonstrate the concepts, and some preliminary results for the Eastern Interconnection are presented.

Hydrogen derived from ammonia: Small-scale costs

A systems study was made to assess the economic prospects for using purchased industrial ammonia as a hydrogen distribution and storage medium for users requiring 0.93–9.34 million std m 3 y −1 of hydrogen (33–330 million std ft 3 y −1). Projected costs to the end user were determined for: the product of dissociated ammonia (N 2+H 2)/NH 3, and the 99.999% pure H 2 obtained by separation of the nitrogen (H 2/NH 3); hydrogen produced by the steam reforming of natural gas (H 2/NG); electrolytic hydrogen (EH 2); purchased (merchant) liquid hydrogen (LH 2); OTEC (ocean thermal energy conversion) LH 2; as well as OTEC NH 3 and the H 2 products derived from it. Future costs are projected as $ GJ −1 and as $ MBTU −1 (1980$ in 1990) using two sets of forecast energy prices. The results show that merchant LH 2 was substantially higher in cost than the other options, and that until EH 2 is available at the projected costs, H 2/NH 3 would be the preferred option for the smallest plant sizes where it is projected to be competitive with H 2/NG of comparable purity. Thus, via state-of-the-art technology, industrial NH 3 can now serve as a viable H 2 carrier for some small-scale users. The availability of OTEC NH 3 at the projected costs would substantially increase the competitive position of H 2/NH 3. Consideration of safety and environmental factors are among the important items listed in recommendations for future work.