A Similar Day Based Short Term Load Forecasting Method Using Wavelet Transform and <scp>LSTM</scp>

Zhang, Ruixuan; Zhang, Chuyan; Yu, Miao

doi:10.1002/tee.23536

Cited by 12 publications

(14 citation statements)

References 24 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the ConvLSTM-Adaboost model has a high stability. The curves of Zhang et al (2022) and Wang et al (2021) are similar, indicating that their prediction stabilities are also good. However, the wide curve distribution indicates that the prediction accuracy is not as good as the proposed model.…”

Section: Impact Analysis Of Different Electricity Pricesmentioning

confidence: 70%

Forecasting Model of Electricity Sales Market Indicators With Distributed New Energy Access

Yao

Yang

Sun

et al. 2023

International Journal of Information Technologies and Systems Approach

View full text Add to dashboard Cite

It is difficult for the existing electricity sales market to adapt to the vast amount of distributed new energy access. This article proposes an electricity sales market index prediction model for high proportion distributed new energy access under the cloud-side cooperation architecture. First, an index prediction system is designed based on the cloud edge collaboration architecture. The edge computing center processes regional data nearby to improve prediction efficiency. Second, on the edge side, a K-means clustering algorithm is used to classify the data. Third, the power data, distributed power output data, load data, weather data, holiday information, and electricity price data are obtained. Finally, the ConvLSTM-Adaboost prediction model is built in the cloud center. The ConvLSTM is used as the base learner, and the Adaboost-integrated algorithm is used for serial training. At the same time, the prediction results of each base learner are weighted and integrated to obtain the final power and load prediction results of the electricity sales market. Experiments show that the prediction results of MAE, PMSE, and MAPE of the proposed model for daily electricity are 52.539MW, 56.859MW, and 2.063%, respectively. Not only is this superior to other models, but it provides a better analysis of influencing factors.

show abstract

Section: Impact Analysis Of Different Electricity Pricesmentioning

confidence: 70%

Forecasting Model of Electricity Sales Market Indicators With Distributed New Energy Access

Yao

Yang

Sun

et al. 2023

International Journal of Information Technologies and Systems Approach

View full text Add to dashboard Cite

show abstract

“…Other STLF techniques that have been applied to power station load forecasting include fuzzy logic models [26], support vector regression (SVR) models (Hong et al, 2009) [27], and wavelet transform models (Zhang et al, 2022) [28]. These techniques have shown promise in providing accurate load forecasts for short-term objectives, but further research is needed to fully evaluate their effectiveness.…”

Section: Related Workmentioning

confidence: 99%

Short-Term Load Forecasting Using Nonlinear Dynamic Time Series Autoregressive with Feedback for AZADI Control-Station Erbil-Iraq

2023

IJCCCE

View full text Add to dashboard Cite

The load forecasting is a human or computational technique for accurate preanticipation of electrical load to enhance reliable operation and optimal planning control of system plant for electrical energy flowing without facing any economical and technical limitations, therefore appropriate estimation for present and future consumption cost of electrical loads which are necessary to predict the load demand for generating near to accurate power. During advanced technology at the last few decades, artificial neural networks(ANNs) have been extensively employed in electrical system, they are trained using historical data obtained from plant station. This work is intended to be a study of short-term load forecasting (STLF) basis for a power predicted applied to the actual past load data displayed from Azadi station for Feb.2022 were used in training and validation system of neural grid. The result was evaluated by mean square percentage error of (32.7) for the forecasting dynamic time series method to solve the data over hours, days, and weeks in advance, using a kind of non-linear filtering. Short-term load forecasting tried out with main stages; predicted power load data sets, network training, and forecasting. Neural network used has 3-layers: an input, a hidden, and an output layer. The number of hidden layer neurons can be varied for the different network performance. The active power generation faces economical and technical challenges, therefore appropriate evaluation of loads are much needed. Index Terms— ANN, STLF, MW, Neurons.

show abstract

“…Currently, many load forecasting methods based on similar days use the minimum distance between load curves to determine similar days. The most commonly used distance measurement methods are Euclidean distance, Manhattan distance, etc., and then predict based on time series extrapolation or machine learning methods [2][3][4][5]. These methods mainly focus on the similarity between load values, that is, "value similarity", and cannot guarantee the most similar curve shape information.…”

Section: Introductionmentioning

confidence: 99%

Short term load forecasting method based on local shape similarity and SVR

Jiao,

Wang,

et al. 2024

Eighth International Conference on Energy System, Electricity, and Power (ESEP 2023)

View full text Add to dashboard Cite

This article proposes a short-term load forecasting method based on local shape similarity and SVR. This method first calculates the local shape similarity coefficient between the predicted daily load and historical samples, and selects historical samples with smaller shape similarity coefficients as similar days, solving the problem of neglecting the shape information similarity of load series in traditional similar day selection. By calculating the weight of the impact of similar days on the prediction results and the weighted load change rate of historical time points, the intraday prediction of the load series at the predicted points is achieved. In addition, the intraday known point data and the predicted data of the predicted points are concatenated as input to the SVR prediction model, so that the prediction model takes into account the intraday latest load data changes and overcomes the problem of not considering the intraday load changes in the next day's prediction. The experimental results show that the proposed approach can improve the accuracy of shortterm load forecasting and reduce prediction errors, which has certain practical value.

show abstract

A Similar Day Based Short Term Load Forecasting Method Using Wavelet Transform and LSTM

Cited by 12 publications

References 24 publications

Forecasting Model of Electricity Sales Market Indicators With Distributed New Energy Access

Forecasting Model of Electricity Sales Market Indicators With Distributed New Energy Access

Short-Term Load Forecasting Using Nonlinear Dynamic Time Series Autoregressive with Feedback for AZADI Control-Station Erbil-Iraq

Short term load forecasting method based on local shape similarity and SVR

Contact Info

Product

Resources

About