Forecasting Energy Consumption Time Series Using Recurrent Neural Network in Tensorflow

Abstract

The environmental issues we are currently facing require long-term prospective efforts for sustainable growth. Renewable energy sources seem to be one of the most practical and efficient alternatives in this regard. Understanding a nation's pattern of energy use and renewable energy production is crucial for developing strategic plans. No previous study has been performed to explore the dynamics of power consumption with the change in renewable energy production on a country-wide scale. In contrast, a number of deep learning algorithms demonstrated acceptable performance while handling sequential data in the era of data-driven predictions. In this study, we developed a scheme to investigate and predict total power consumption and renewable energy production time series for eleven years of data using a Recurrent Neural Network (RNN). The dynamics of the interaction between the total annual power consumption and renewable energy production are investigated through extensive Exploratory Data Analysis (EDA) and a feature engineering framework. The performance of the model is found satisfactory through the comparison of the predicted data with the observed data, visualization of the distribution of the errors and Root Mean Squared Error (RMSE) value of 0.084. Higher performance is achieved through the increase in the number of epochs and hyperparameter tuning. The proposed framework can be used and transferred to investigate the trend of renewable energy production and power consumption and predict the future scenarios for different communities. Incorporation of the cloud-based platform into the proposed pipeline may lead to real-time forecasting.

Demystifying the Relationship Between River Discharge and Suspended Sediment Using Exploratory Analysis and Deep Neural Network Algorithms

“…In terms of the selected performance indices, DNN models outperformed the SRC method in simulating the daily SSC. These findings corroborate researchers' claims, such as Cigizoglu and Kisi, 2005, about using the feed-forward back propagation (FFBP) algorithm, sediment concentration, and streamflow data as inputs for daily or monthly SSC estimation and forecasting [40][41][42]. However, it has been reported that not enough studies have been conducted to evaluate the best performance of training algorithms within DNN modeling techniques [15].…”

“…Because of the decaying error backflow, recurrent backpropagation requires a significant amount of computational time and effort to learn to store long-term information. Hence, the concept of the vanishing gradient problem in recognizing long-term dependency of Recurrent Neural Network (RNN) was introduced [42,46]. The main element of processing and retaining long-term information is LSTM feedback connections, and this characteristic distinguishes it from the conventional feedforward neural network.…”

Using Machine Learning to Improve Performance of a Low-Cost Real-Time Stormwater Control Measure

“…The initial model that was trained on historical data was a multiple linear regression and the second was a long short-term memory network. Both multiple linear regression and long short-term memory have been used by previous researchers as data driven models to predict streamflow, water table depth, and urban flooding; however, there is a need for further investigation on how these techniques can be used to optimize RTC systems for an individual SCM using rainfall data, which is commonly available [36][37][38][39][40][41][42][43]. An exploratory 3 of 18 data analysis approach was used to analyze the four years of historical data at the research site to develop a program to optimize basin performance.…”

Demystifying the Relationship Between River Discharge and Suspended Sediment Using Exploratory Analysis and Deep Neural Network Algorithms

“…Because of the decaying error backflow, recurrent backpropagation requires a significant amount of computational time and effort to learn to store long-term information. Hence, the concept of the vanishing gradient problem in recognizing long-term dependency of Recurrent Neural Network (RNN) was introduced [49,53]. The main element of processing and retaining long-term information is LSTM feedback connections, and this characteristic distinguishes it from the conventional feedforward neural network.…”