Optimization of neural network architecture using genetic algorithm for load forecasting

Islam, Badar ul; Baharudin, Z.; Raza, Muhammad Qamar; Nallagownden, Perumal

doi:10.1109/icias.2014.6869528

Cited by 29 publications

(17 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Islam et al implemented a genetic algorithm for optimizing the number of layers and neurons per layer for an ANN trained with a backpropagation algorithm. Mean absolute percentage error (MAPE) and model computational time was significantly reduced after optimizing ANN topology [21]. Defilippo et al used GAs to select the appropriate architecture and training parameters using evolutionary simulations of a neural network model for the load forecast for a city in Brazil.…”

Section: Related Workmentioning

confidence: 99%

Multi-Sequence LSTM-RNN Deep Learning and Metaheuristics for Electric Load Forecasting

et al. 2020

View full text Add to dashboard Cite

Short term electric load forecasting plays a crucial role for utility companies, as it allows for the efficient operation and management of power grid networks, optimal balancing between production and demand, as well as reduced production costs. As the volume and variety of energy data provided by building automation systems, smart meters, and other sources are continuously increasing, long short-term memory (LSTM) deep learning models have become an attractive approach for energy load forecasting. These models are characterized by their capabilities of learning long-term dependencies in collected electric data, which lead to accurate prediction results that outperform several alternative statistical and machine learning approaches. Unfortunately, applying LSTM models may not produce acceptable forecasting results, not only because of the noisy electric data but also due to the naive selection of its hyperparameter values. Therefore, an optimal configuration of an LSTM model is necessary to describe the electric consumption patterns and discover the time-series dynamics in the energy domain. Finding such an optimal configuration is, on the one hand, a combinatorial problem where selection is done from a very large space of choices; on the other hand, it is a learning problem where the hyperparameters should reflect the energy consumption domain knowledge, such as the influential time lags, seasonality, periodicity, and other temporal attributes. To handle this problem, we use in this paper metaheuristic-search-based algorithms, known by their ability to alleviate search complexity as well as their capacity to learn from the domain where they are applied, to find optimal or near-optimal values for the set of tunable LSTM hyperparameters in the electrical energy consumption domain. We tailor both a genetic algorithm (GA) and particle swarm optimization (PSO) to learn hyperparameters for load forecasting in the context of energy consumption of big data. The statistical analysis of the obtained result shows that the multi-sequence deep learning model tuned by the metaheuristic search algorithms provides more accurate results than the benchmark machine learning models and the LSTM model whose inputs and hyperparameters were established through limited experience and a discounted number of experimentations.

show abstract

Section: Related Workmentioning

confidence: 99%

Multi-Sequence LSTM-RNN Deep Learning and Metaheuristics for Electric Load Forecasting

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Especially neural networks have become extremely popular in many forecasting applications, such as load forecasting [5,19,23,38,54], as they perform well on non-linear and high-frequency time series. Many applications combine these evolutionary algorithms with traditional forecasting techniques in order to capture both linear and non-linear time series information [2, 6,13,61], or use the same algorithms to optimize the input parameters of existing forecasting methods.…”

Section: Introductionmentioning

confidence: 99%

“…The forecast accuracy of these models has been improved by optimizing their input with evolutionary search heuristics, such as Particle Swarm Optimization [4, 32,60,62,63], Genetic Algorithms [22,30,39,42,43,49,54], Simulated Annealing [23,40], Artificial Bee Colony Algorithm [5, 24,47], Differential Evolution [25,57] and Fruit Fly Optimization [38,41]. These hybrid methodologies have been applied to many different fields in forecasting, including tourism flow forecasting [14], electricity demand forecasting [63], rainfall prediction [60], price forecasting [47] and many others.…”

Section: Introductionmentioning

confidence: 99%

“…Especially the AIC value is often used when manually selecting the parameters of an AR(I)MA model [19,65]. However, the most popular criterion for optimization in forecasting is accuracy, which can take many forms, such as the Mean Squared Error (MSE) [1,4,5,30,34,54,62], the Mean Absolute Percentage Error (MAPE) [14,23,38,39,42,47,57,63] or the Root Mean Squared Error (RMSE) [22,25,27,38]. In this paper, however, we turn to a profit measure for sales forecasting to optimize the order identification of Seasonal ARIMA models.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

ProfARIMA: A profit-driven order identification algorithm for ARIMA models in sales forecasting

Calster

Baesens

Lemahieu

2017

Applied Soft Computing

View full text Add to dashboard Cite

In forecasting, evolutionary algorithms are often linked to existing forecasting methods to optimize their input parameters. Traditionally, the fitness function of these search heuristics is based on an accuracy measure. In this paper, however, we combine forecasting accuracy with business expertise by defining a flexible and easily interpretable profit function for sales forecasting, which is based on the profit margin of a given product, the volume of its sales and the accuracy of the forecast. ProfARIMA is a new procedure that selects the lags of a Seasonal ARIMA model according to the profit of a model's forecasts by taking advantage of search heuristics. This procedure is tested on both publicly available datasets and a real-life application with datasets of The Coca-Cola Company in order to assess its performance, both in profit and accuracy. Three different evolutionary algorithms were implemented during this testing process, i.e. Genetic Algorithms, Particle Swarm Optimization and Simulated Annealing. The results indicate that ProfARIMA always performs at least equally to the Box-Jenkins methodology and often outperforms this traditional procedure. For The Coca-Cola Company, our new algorithm in combination with Genetic Algorithms even leads to a significantly larger profit for out-of-sample forecasts.

show abstract

“…RELATED WORK There were quite many attempts on architecture optimization via evolutionary process (e.g. [6], [7]) in previous decades.…”

Section: Introductionmentioning

confidence: 99%

Evolving KERAS Architectures for Sensor Data Analysis

Vidnerová

Neruda

2017

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-Deep neural networks enjoy high interest and have become the state-of-art methods in many fields of machine learning recently. Still, there is no easy way for a choice of network architecture. However, the choice of architecture can significantly influence the network performance.This work is the first step towards an automatic architecture design. We propose a genetic algorithm for an optimization of a network architecture. The algorithm is inspired by and designed directly for the Keras library [1] that is one of the most common implementations of deep neural networks.The target application is the prediction of air pollution based on sensor measurements. The proposed algorithm is evaluated on experiments on sensor data and compared to several fixed architectures and support vector regression.

show abstract

Optimization of neural network architecture using genetic algorithm for load forecasting

Cited by 29 publications

References 20 publications

Multi-Sequence LSTM-RNN Deep Learning and Metaheuristics for Electric Load Forecasting

Multi-Sequence LSTM-RNN Deep Learning and Metaheuristics for Electric Load Forecasting

ProfARIMA: A profit-driven order identification algorithm for ARIMA models in sales forecasting

Evolving KERAS Architectures for Sensor Data Analysis

Contact Info

Product

Resources

About