Alicia Troncoso Lora

Electricity Market Price Forecasting Based on Weighted Nearest Neighbors Techniques

This paper presents a simple technique to forecast next-day electricity market prices based on the weighted nearest neighbors methodology. First, it is explained how the relevant parameters defining the adopted model are obtained. Such parameters have to do with the window length of the time series and with the number of neighbors chosen for the prediction. Then, results corresponding to the Spanish electricity market during 2002 are presented and discussed. Finally, the performance of the proposed method is compared with that of recently published techniques.

Time-Series Prediction: Application to the Short-Term Electric Energy Demand

This paper describes a time-series prediction method based on the kNN technique. The proposed methodology is applied to the 24-hour load forecasting problem. Also, based on recorded data, an alternative model is developed by means of a conventional dynamic regression technique, where the parameters are estimated by solving a least squares problem. Finally, results obtained from the application of both techniques to the Spanish transmission system are compared in terms of maximum, average and minimum forecasting errors.

Electricity Market Price Forecasting: Neural Networks versus Weighted-Distance k Nearest Neighbours

In todays deregulated markets, forecasting energy prices is becoming more and more important. In the short term, expected price profiles help market participants to determine their bidding strategies. Consequently, accuracy in forecasting hourly prices is crucial for generation companies (GENCOs) to reduce the risk of over/underestimating the revenue obtained by selling energy. This paper presents and compares two techniques to deal with energy price forecasting time series: an Artificial Neural Network (ANN) and a combined k Nearest Neighbours (kNN) and Genetic algorithm (GA). First, a customized recurrent Multi-layer Perceptron is developed and applied to the 24-hour energy price forecasting problem, and the expected errors are quantified. Second, a k nearest neighbours algorithm is proposed using a Weighted-Euclidean distance. The weights are estimated by using a genetic algorithm. The performance of both methods on electricity market energy price forecasting is compared.

A Comparison of Two Techniques for Next-Day Electricity Price Forecasting

In the framework of competitive markets, the markets participants need energy price forecasts in order to determine their optimal bidding strategies and maximize their benefits. Therefore, if generation companies have a good accuracy in forecasting hourly prices they can reduce the risk of over/underestimating the income obtained by selling energy. This paper presents and compares two energy price forecasting tools for day-ahead electricity market: a k Weighted Nearest Neighbours (kWNN) the weights being estimated by a genetic algorithm and a Dynamic Regression (DR). Results from realistic cases based on Spanish electricity market energy price forecasting are reported.

Finding improved local minima of power system optimization problems by interior-point methods

This paper presents a simple heuristic technique to deal with multiple local minima in nonconvex, nonlinear power system optimization problems by solving a sequence of interior-point subproblems. Both the real-valued and the mixed-integer cases are separately discussed. The method is then applied to the unit commitment problem and its performance on realistic cases is compared with that of a genetic algorithm (GA).

Influence of kNN-Based Load Forecasting Errors on Optimal Energy Production

This paper presents a study of the influence of the accuracy of hourly load forecasting on the energy planning and operation of electric generation utilities. First, a k Nearest Neighbours (kNN) classification technique is proposed for hourly load forecasting. Then, obtained prediction errors are compared with those obtained results by using a M5’. Second, the obtained kNN-based load forecast is used to compute the optimal on/off status and generation scheduling of the units. Finally, the influence of forecasting errors on both the status and generation level of the units over the scheduling period is studied.

Application of Evolutionary Computation Techniques to the Optimal Short-Term Scheduling of the Electrical Energy Production

In this paper, an evolutionary technique applied to the optimal short-term scheduling (24 hours) of the electric energy production is presented. The equations that define the problem lead to a nonlinear mixed-integer programming problem with a high number of real and integer variables. Consequently, the resolution of the problem based on combinatorial methods is rather complex. The required heuristics, introduced to assure the feasibility of the constraints, are analyzed, along with a brief description of the proposed genetic algorithm. Finally, results from realistic cases based on the Spanish power system are reported, revealing the good performance of the proposed algorithm, taking into account the complexity and dimension of the problem.