Electricity Consumption Forecasting Based on a Bidirectional Long-Short-Term Memory Artificial Neural Network

Abstract

The accurate forecasting of the hourly month-ahead electricity consumption represents a very important aspect for non-household electricity consumers and system operators, and at the same time represents a key factor in what regards energy efficiency and achieving sustainable economic, business, and management operations. In this context, we have devised, developed, and validated within the paper an hourly month ahead electricity consumption forecasting method. This method is based on a bidirectional long-short-term memory (BiLSTM) artificial neural network (ANN) enhanced with a multiple simultaneously decreasing delays approach coupled with function fitting neural networks (FITNETs). The developed method targets the hourly month-ahead total electricity consumption at the level of a commercial center-type consumer and for the hourly month ahead consumption of its refrigerator storage room. The developed approach offers excellent forecasting results, highlighted by the validation stage’s results along with the registered performance metrics, namely 0.0495 for the root mean square error (RMSE) performance metric for the total hourly month-ahead electricity consumption and 0.0284 for the refrigerator storage room. We aimed for and managed to attain an hourly month-ahead consumed electricity prediction without experiencing a significant drop in the forecasting accuracy that usually tends to occur after the first two weeks, therefore achieving a reliable method that satisfies the contractor’s needs, being able to enhance his/her activity from the economic, business, and management perspectives. Even if the devised, developed, and validated forecasting solution for the hourly consumption targets a commercial center-type consumer, based on its accuracy, this solution can also represent a useful tool for other non-household electricity consumers due to its generalization capability.

E-Commerce Sales Revenues Forecasting by Means of Dynamically Designing, Developing and Validating a Directed Acyclic Graph (DAG) Network for Deep Learning

“…During the process of developing our proposed forecasting method, in order to solve the contractor's needs as best as possible, we have adapted and tested several other forecasting methods from the literature that provided very good results in terms of forecasting accuracy, in order to find the one that would provide the best results for our specific case. The hybrid approach developed in [39] uses enhanced BiLSTM ANNs combined with function fitting neural networks (FITNETs), while the hybrid forecasting method developed in [38] is based on LSTM ANNs combined with FITNETs, and the hybrid prediction solution from [37] consists of a mixed non-linear autoregressive with exogenous inputs (NARX) ANNs and FITNETs. Our proposed method has outperformed the other hybrid methods in terms of forecasting accuracy, as depicted by the NRMSE (Table 5).…”

“…In this way, we have obtained a higher degree of generalization of the proposed forecasting approach, making it usable in a large number of practical cases, similar to the one analyzed within our study, even in the ones in which AMVs of input data would be registered. In order to manage the potential AMVs, the proposed method considers a gap-filling, linear interpolation-based approach that we have previously used in other studies and over the course of time has proven its usefulness and efficiency [36][37][38][39]. The preprocessed dataset obtained at the end of this step is denoted as PWDS.…”

Short‐term electricity load forecasting based on improved sparrow search algorithm with optimized BiLSTM

“…The results showed that the root means square error of the forecast result was 0.0495 and was able to provide an accurate prediction result for the forecasting of power consumption. 9 Zhu K et al constructed a BiLSTM-based prediction model for the purpose of solving the daily peak load prediction. 10 For the purpose of improving the precision of STELF, Zhang W et al constructed a prediction model based on regression probability, and the simulation outcome is that the model has better prediction behavior.…”

A hybrid CNN-BiLSTM approach for remaining useful life prediction of EVs lithium-Ion battery

“…BiLSTM is a temporal recurrent neural network. 25 BiLSTM can effectively retain the ability of historical input data and has the ability of memory. Compared with LSTM processing input data in one direction, BiLSTM bi-directionally trains the input data in ascending and descending order, and predicts in combination with context.…”