Recently there has been a significant proliferation in the use of forecasting techniques, mainly due to the increased availability and power of computation systems and, in particular, to the usage of personal computers. This is also true for power network systems, where energy demand forecasting has been an important field in order to allow generation planning and adaptation. Apart from the quantitative progression, there has also been a change in the type of models proposed and used. In the `70s, the usage of non-linear techniques was generally not popular among scientists and engineers. However, in the last two decades they have become very important techniques in solving complex problems which would be very difficult to tackle otherwise. With the recent emergence of smart grids, new environments have appeared capable of integrating demand, generation, and storage. These employ intelligent and adaptive elements that require more advanced techniques for accurate and precise demand and generation forecasting in order to work optimally. This review discusses the most relevant studies on electric demand prediction over the last 40 years, and presents the different models used as well as the future trends. Additionally, it analyzes the latest studies on demand forecasting in the future environments that emerge from the usage of smart grids.

A Survey on Electric Power Demand Forecasting: Future Trends in Smart Grids, Microgrids and Smart Buildings

Energy storage can alleviate the problems that the uncertainty and variability associated with renewable energy sources such as wind and solar create in power systems. Besides applications such as frequency control, temporal arbitrage or the provision of reserve, where the location of storage is not particularly relevant, distributed storage could also be used to alleviate congestion in the transmission network. In such cases, the siting and sizing of this distributed storage is of crucial importance to its cost-effectiveness. This paper describes a three-stage planning procedure to identify the optimal locations and parameters of distributed storage units. In the first stage, the optimal storage locations and parameters are determined for each day of the year individually. In the second stage, a number of storage units is available at the locations that were identified as being optimal in the first stage, and their optimal energy and power ratings are determined. Finally, in the third stage, with both the locations and ratings fixed, the optimal operation of the storage units is simulated to quantify the benefits that they would provide by reducing congestion. The quality of the final solution is assessed by comparing it with the solution obtained at the first stage without constraints on storage sites or size. The approach is numerically tested on the IEEE RTS 96.

Near-Optimal Method for Siting and Sizing of Distributed Storage in a Transmission Network

The recent development of smart meters has allowed the analysis of household electricity consumption in real time. Predicting electricity consumption at such very low scales should help to increase the efficiency of distribution networks and energy pricing. However, this is by no means a trivial task since household-level consumption is much more irregular than at the transmission or distribution levels. In this work, we address the problem of improving consumption forecasting by using the statistical relations between consumption series. This is done both at the household and district scales (hundreds of houses), using various machine learning techniques, such as support vector machine for regression (SVR) and multilayer perceptron (MLP). First, we determine which algorithm is best adapted to each scale, then, we try to find leaders among the time series, to help short-term forecasting. We also improve the forecasting for district consumption by clustering houses according to their consumption profiles.

/pdf/electricity-load-forecasting-for-residential-customers-3g2a1j8ehy.pdf

Electricity load forecasting for residential customers: Exploiting aggregation and correlation between households

This paper introduces the fundamental concept of locational marginal price (LMP) in the electricity markets, and presents some special observations on LMP. Under market environment, LMP based settlement strategy is used to determine the amount of money earned from ISO by the energy sellers and paid to ISO by the energy buyers. Thus, depending on different market designs, four different calculation models and corresponding properties on LMP are discussed in this paper. Some interesting applications of LMP are also presented. In this paper, all studied examples are based on a simple 3-bus network.

Different models and properties on LMP calculations

Risk-averse profit-based optimal operation strategy of a combined wind farm–cascade hydro system in an electricity market

This paper presents a formulation of security-constrained unit commitment (SCUC) problem based on mixed integer programming (MIP) method with considering prohibited operating zone limits of thermal and hydro units. The objective of SCUC problem is to obtain minimum system operating cost while maintaining the system security. Thermal and hydro units in power system studies are assumed to possess smooth and convex input-output (IO) curves. Practically, not all the operating zones of generation units are available for load allocation due to some physical operation limits. This may cause to have a generator with nonsmooth IO curve with equality and inequality constraints which may not be solved easily by conventional mathematical methods. In this paper, non-convex characteristic of generator cost function, representing prohibited operating zones is considered in SCUC. Test results with an eight-bus system show the accuracy of the model and formulations.

Mixed integer programming method to solve security constrained unit commitment with restricted operating zone limits

Long-term load forecasting has a vital role in generation, transmission and distribution network planning. Traditional studies for long-term load forecasting were based on regression method, which could not provide a true representation of power system behavior in a volatile electricity market. The purpose of this paper is to introduce two approaches based regression method and artificial neural network (ANN) for long-term load forecast. We apply fuzzy sets to ANN for modeling long-term uncertainties and compare the enhanced forecasting results with those of traditional methods. The ISO New England market data are used to illustrate the efficiency of each technique.

Long-term load forecasting in electricity market

This paper starts from the fundamental concepts of market clearing price (MCP) and locational marginal price (LMP) in the electricity markets, and presents some interesting observations on MCP and LMP. The point at which demand curve and supply curve intersects establishes the MCP. The non-existence of MCP and existence of multiple MCPs is explored. LMP is the marginal cost of supplying the next MW of load at a specific location. LMPs with the presence of price-responsive demand and multi-period coupling constraints are discussed. For the first time, the cases where LMP is higher than the highest generation bidding price and where LMP is lower than the lowest generation bidding price are discussed, all using examples for a simple 3-bus network.

Some observations on market clearing price and locational marginal price

This paper presents a formulation of fuzzy mixed integer programming (FMIP) solution for solving security-constrained unit commitment (SCUC) problem with emphasis on uncertainties in forecasted parameters in constraints. The proposed approach could be used by vertically integrated utilities as well as the ISOs in restructured power system. In this model, uncertainties in forecasted load demand, ancillary services and wind power are simulated in a fuzzy frame. Problem is finding a solution which satisfies the constraints and the objective with the maximum degree. The objective function will optimize while constraints have varying degree. The proposed model can be solved using a standard mixed integer-programming (MIP) solver. Case studies with the eight-bus system are presented in detail in this paper. We study the impact of uncertainty in forecasted parameters in SCUC solution and compare the results with those of crisp SCUC model.

H. Daneshi

Papers

Mixed integer programming method to solve security constrained unit commitment with restricted operating zone limits

Long-term load forecasting in electricity market

Some observations on market clearing price and locational marginal price

Fuzzy mixed integer programming: Approach to security-constrained unit commitment