Application of a fuzzy neural network combined with a chaos genetic algorithm and simulated annealing to short-term load forecasting

Liao, Gwo-Ching; Tsao, Ta‐Peng

doi:10.1109/tevc.2005.857075

Cited by 164 publications

(17 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several machine learning or computational intelligence techniques have been applied in the field of Short Term Load Forecasting. For example, decision trees [10], Fuzzy Logic systems [11,12], and Neural Networks [13][14][15][16][17][18][19][20]. In this paper, we propose the using of a particular set of supervised machine learning techniques (called ensemble methods based on decision trees) to predict the hourly electricity consumption of university buildings.…”

Section: Introductionmentioning

confidence: 99%

Load Forecasting for a Campus University Using Ensemble Methods Based on Regression Trees

2018

View full text Add to dashboard Cite

Load forecasting models are of great importance in Electricity Markets and a wide range of techniques have been developed according to the objective being pursued. The increase of smart meters in different sectors (residential, commercial, universities, etc.) allows accessing the electricity consumption nearly in real time and provides those customers with large datasets that contain valuable information. In this context, supervised machine learning methods play an essential role. The purpose of the present study is to evaluate the effectiveness of using ensemble methods based on regression trees in short-term load forecasting. To illustrate this task, four methods (bagging, random forest, conditional forest, and boosting) are applied to historical load data of a campus university in Cartagena (Spain). In addition to temperature, calendar variables as well as different types of special days are considered as predictors to improve the predictions. Finally, a real application to the Spanish Electricity Market is developed: 48-h-ahead predictions are used to evaluate the economical savings that the consumer (the campus university) can obtain through the participation as a direct market consumer instead of purchasing the electricity from a retailer.

show abstract

Section: Introductionmentioning

confidence: 99%

Load Forecasting for a Campus University Using Ensemble Methods Based on Regression Trees

2018

View full text Add to dashboard Cite

show abstract

“…Because STLF is a nonlinear problem, many previous works have utilized ANN for STLF, e.g., Reference [31] applied a feed-forward backpropagation ANN for STLF. Hybrids of ANN and other techniques are also common for STLF, e.g., regression tree model [32], time series analysis [33], genetic algorithm [34], chaos genetic algorithm and simulated annealing [35].…”

Section: Related Work On Short-term Load Forecastingmentioning

confidence: 99%

“…STLF, e.g., regression tree model [32], time series analysis [33], genetic algorithm [34], chaos genetic algorithm and simulated annealing [35].…”

Section: Related Work On Short-term Load Forecastingmentioning

confidence: 99%

“…The LSTM contains layers of memory cells, where the interaction between the LSTM layers is shown in Figure 1, and the architecture of a LSTM cell is shown in Figure 2. STLF, e.g., regression tree model [32], time series analysis [33], genetic algorithm [34], chaos genetic algorithm and simulated annealing [35]. With the advance of computing power, deep neural network (DNN) has gained much popularity and been applied to STLF in recent years [36].…”

Section: Long Short-term Memory For Short-term Load Forecastingmentioning

confidence: 99%

See 1 more Smart Citation

Integrating Long Short-Term Memory and Genetic Algorithm for Short-Term Load Forecasting

Santra

Lin

2019

Energies

View full text Add to dashboard Cite

Electricity load forecasting is an important task for enhancing energy efficiency and operation reliability of the power system. Forecasting the hourly electricity load of the next day assists in optimizing the resources and minimizing the energy wastage. The main motivation of this study was to improve the robustness of short-term load forecasting (STLF) by utilizing long shortterm memory (LSTM) and genetic algorithm (GA). The proposed method is novel: LSTM networks are designed to avoid the problem of long-term dependencies, and GA is used to obtain the optimal LSTM's parameters, which are then applied to predict the hourly electricity load for the next day. The proposed method was trained using actual load and weather data, and the performance results showed that it yielded small mean absolute percentage error on the test data.With the advance of computing power, deep neural network (DNN) has gained much popularity and been applied to STLF in recent years [36]. LSTM is a special type of DNN that is suitable for time series prediction due to its capability of remembering both the short-term and the long-term behavior in time series data. In Reference [37], two types of LSTM were compared against other deep learning techniques for predicting electricity load of every hour or every minute. Their results showed that LSTM outperformed the other techniques. However, all of the neural network approaches above require setting up the initial weightings of the links in the network, but poorly chosen weightings could lead the searching process trapped in the local optimum. A hybrid of LSTM and GA is proposed in Section 4 to resolve this problem.Prior to presenting our approach in Section 4, this section gives a detailed description of LSTM. LSTM is an augmented recurrent neural network model. It learns sequential information with long term dependencies, and preserves information for a long period of time. Traditional recurrent neural network suffers from the vanishing gradient problem. That is, as the number of layers using the same activation function increases, the gradients of the loss function approaches zero, making it difficult to train the network through backpropagation of errors. To prevent the vanishing gradient problem, LSTM utilizes memory cells, where each cell maintains a cell state and a hidden cell state, and uses three gates (namely, input gate, output gate, and forget gate) to control the flow of information into or out of the cell. A formal explanation of the LSTM model is given below.LSTM is for time series modeling, which maps an input sequence x = {x 1 , x 2 , . . . , x n } to an output sequence y = y 1 , y 2 , . . . , y n . For the STLF problem under study, each x i represents the hourly electricity load and the weather data of day i, and each y i represents the hourly electricity load of day i + 1, indicating a look-ahead parameter of value one. The LSTM contains layers of memory cells, where the interaction between the LSTM layers is shown in Figure 1, and the architecture of a LSTM cell is shown...

show abstract

“…Moreover, the integration of ANN and other methods has become a research hotspot. ANN was integrated with fuzzy logic [38], genetic algorithm [39], wavelet analysis [40,41], chaos theory [42], grey system [43], etc. However, ANN has its limitations: it is hard to avoid learning deficiency or over-fit phenomena, and the convergence speed is slow and easy to fall into local minima.…”

Section: Literature Reviewmentioning

confidence: 99%

The Recurrence Interval Difference of Power Load in Heavy/Light Industries of China

Zhang

2018

Energies

View full text Add to dashboard Cite

Abstract:The significant fluctuation of industrial electricity consumption has a high impact on power load, which makes the research on recurrence intervals between extreme events of theoretical and practical significance. The study uses a high-frequency data of heavy and light industries and employs recurrence interval analysis in different thresholds. We find that the reoccurrence interval of volatility can fit with the stretched exponential function and the probability density functions of recurrence intervals in various thresholds shows a scaling behavior. Then, the conditional probability density function and the multifractal detrended fluctuation analysis demonstrate the existence of short-range correlation, long-range correlation, and multifractal properties, respectively. We further construct a hazard function, introduce recurrence intervals into VaR calculation and establish a functional relationship between average recurrence interval and threshold. Following this result, we also shed light on policy discussion for multi-industrial electricity supply management.

show abstract

Application of a fuzzy neural network combined with a chaos genetic algorithm and simulated annealing to short-term load forecasting

Cited by 164 publications

References 43 publications

Load Forecasting for a Campus University Using Ensemble Methods Based on Regression Trees

Load Forecasting for a Campus University Using Ensemble Methods Based on Regression Trees

Integrating Long Short-Term Memory and Genetic Algorithm for Short-Term Load Forecasting

The Recurrence Interval Difference of Power Load in Heavy/Light Industries of China

Contact Info

Product

Resources

About