Evolving nearest neighbor time series forecasters

Flores, Juan J.; Gonzalez, Jose R. Cedeno; Farías, Rodrigo López; Calderón, Félix

doi:10.1007/s00500-017-2822-1

Cited by 10 publications

(13 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several methods are developed for multistep forecasting problems, where three main strategies are, 1) Direct Multi-step Forecast Strategy, 2) Recursive Multistep Forecast Strategy, 3) Multiple Output Forecast Strategy [14]. Different ML models have been used to implement the direct strategy for multi-step forecasting tasks, for instance, the nearest neighbors [15], DT and RF [16]. This direct method involves developing a separate model for each forecast time step.…”

Section: B Multi-step Forecasting Problemmentioning

confidence: 99%

LSTM-based Multi-Step SOC Forecasting of Battery Energy Storage in Grid Ancillary Services

Ardiansyah¹,

Kim²,

Choi³

2021

Preprint

View full text Add to dashboard Cite

Battery energy storage (BES) participation in the grid ancillary services markets is increasing rapidly in recent years. To facilitate optimal participation, the need for accurate BES state-of-charge (SOC) forecasting is indispensable. In grid ancillary services, the development of SOC forecasting models should deal with uncertainties and corresponding stochastic processes that determine the BES SOC periodically. Several traditional and state-of-the-art machine learning (ML) techniques, ranging from decision-tree to deep learning methods, were used to solve this problem. However, developing a multi-step SOC forecasting model remains a challenge in this subject that is essential for optimal BES economic dispatch and unit commitment. Taking advantage of the Long short-term memory (LSTM) deep learning and its variants techniques which are proven to be a robust method for predicting sequentially dependent data in the time-series domain, this paper proposes LSTM-based multi-step SOC forecasting for BES operating in frequency regulation. Various developed models, i.e., Vanilla-LSTM, Vanilla-Gated Recurrent Units (GRU), Bidirectional-LSTM (Bi-LSTM), and Bi-GRU, are evaluated using real-world datasets. The evaluation results show that the developed models outperform the existing methods in terms of root mean square error (RMSE) and mean absolute error (MAE) evaluation metrics.

show abstract

Section: B Multi-step Forecasting Problemmentioning

confidence: 99%

LSTM-based Multi-Step SOC Forecasting of Battery Energy Storage in Grid Ancillary Services

Ardiansyah¹,

Kim²,

Choi³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…This strategy is considered deterministic because it uses a fixed neighborhood radius of = 1 × 10 −3 and updating = 1.2 when no nearest neighbors are found. Researchers have used Differential Evolution (DE) to optimize parameters m, τ , and simultaneously; this strategy allows us to build accurate models for time series forecasting [10,11]. Even though DE is relatively simple, multiple parameters need to be defined, i.e., boundaries for m, τ and , population size, scale factor, recombination probability, the maximum number of itera-tions, and the number of executions.…”

Section: Knn Approachesmentioning

confidence: 99%

Forecasting without context problem

Ortiz-Bejar

Zamora-Méndez

et al. 2020

2020 IEEE International Autumn Meeting on Power, Electronics and Computing (ROPEC)

View full text Add to dashboard Cite

This work presents an analysis of four regression systems. Two of them are statistical: the widely used Auto-regressive Integrated Moving Average (ARIMA) and the state-of-the-art Facebook Prophet. From the deep learning school, a Long Short-Term Memory (LSTM) Recurrent Neural Network (RNN) is also evaluated. We finish our quartet with a fine-tuned Nearest Neighbor model. The study is carried out over seventeen benchmarks; fifteen coming from M4-Competition and two more power systems time series, i.e., electricity demand and hydropower generation. For all the models, the regression systems are fitted and optimized to minimize user intervention. The results show that deep learning models obtained the best performance; nonetheless, the performance difference is not statistically significant with the rest of the systems tested.

show abstract

“…Previous experimental results with synthetic chaotic time series [49], indicate that using basic principles to reconstruct the non-linear time series in phase space, perform far better predictions than the basic statistic principles. They show that the prediction quality can be improved even more by using Differential Evolution.…”

Section: Related Workmentioning

confidence: 99%

“…Differential Evolution contributes to obtain the best parameters given an specific fitness function. This hybrid method that combines Nearest Neighbors with Differential Evolution is called NNDE [49].…”

Section: Nearest Neighbors With Differential Evolution Parameter Optimentioning

confidence: 99%

Soft Computing Methods with Phase Space Reconstruction for Wind Speed Forecasting—A Performance Comparison

et al. 2019

Self Cite

View full text Add to dashboard Cite

This article presents a comparison of wind speed forecasting techniques, starting with the Auto-regressive Integrated Moving Average, followed by Artificial Intelligence-based techniques. The objective of this article is to compare these methods and provide readers with an idea of what method(s) to apply to solve their forecasting needs. The Artificial Intelligence-based techniques included in the comparison are Nearest Neighbors (the original method, and a version tuned by Differential Evolution), Fuzzy Forecasting, Artificial Neural Networks (designed and tuned by Genetic Algorithms), and Genetic Programming. These techniques were tested against twenty wind speed time series, obtained from Russian and Mexican weather stations, predicting the wind speed for 10 days, one day at a time. The results show that Nearest Neighbors using Differential Evolution outperforms the other methods. An idea this article delivers to the reader is: what part of the history of the time series to use as input to a forecaster? This question is answered by the reconstruction of phase space. Reconstruction methods approximate the phase space from the available data, yielding m (the system’s dimension) and τ (the sub-sampling constant), which can be used to determine the input for the different forecasting methods.

show abstract

Evolving nearest neighbor time series forecasters

Cited by 10 publications

References 12 publications

LSTM-based Multi-Step SOC Forecasting of Battery Energy Storage in Grid Ancillary Services

LSTM-based Multi-Step SOC Forecasting of Battery Energy Storage in Grid Ancillary Services

Forecasting without context problem

Soft Computing Methods with Phase Space Reconstruction for Wind Speed Forecasting—A Performance Comparison

Contact Info

Product

Resources

About