Short term electricity load forecasting using hybrid prophet-LSTM model optimized by BPNN

Energy Reports, Journal Year: 2022, Volume and Issue: 8, P. 1678 - 1686

Published: Jan. 13, 2022

Electrical load forecasting plays a vital role in the operation and planning of power plants for utility companies policy makers to design stable reliable energy infrastructure. Load is categorized long-term, mid-term short-term. Among them, short term that monitors weekly, daily, hourly even sub-hourly operations gaining lot attention which saves time cost while satisfying consumers' needs without interruption. Different models such as conventional, Artificial Intelligence (AI) hybrid have been developed investigate short-term forecasting. However, these suffers various issues low speed convergence (conventional), high complexity so on. Consequently, this work proposes method using Prophet Long Short Term Memory (LSTM) overcome above limitations an effort predict accurate load. The model utilize linear well non-linear data original but still some residuals are left regarded data. Here, (non-linear data) trained by employing LSTM, finally both forecasted from LSTM Back Propagation Neural Network (BPNN) further enhance prediction accuracy. Elia Grid real quarter hour based electrical 2014 2021 has utilized verify working performance proposed technique computing Mean Absolute Percentage Error (MAPE), Root Square (RMSE), Average (MAE). Results substantiate outperforms standalone Autoregressive Integrated Moving average (ARIMA), on basis reduced errors with least computation time.

Language: Английский

---

An adaptive backpropagation algorithm for long-term electricity load forecasting

Neural Computing and Applications, Journal Year: 2021, Volume and Issue: 34(1), P. 477 - 491

Published: Aug. 11, 2021

Language: Английский

---

District heating load forecasting with a hybrid model based on LightGBM and FB-prophet

Journal of Cleaner Production, Journal Year: 2023, Volume and Issue: 409, P. 137130 - 137130

Published: April 18, 2023

Language: Английский

---

A hybrid long-term industrial electrical load forecasting model using optimized ANFIS with gene expression programming

Energy Reports, Journal Year: 2024, Volume and Issue: 11, P. 5831 - 5844

Published: May 29, 2024

Electric energy demand forecasting is vital in contemporary power systems, especially amidst market deregulation trends and the increasing influence of industrial customers on dynamics. However, existing models encounter challenges such as slow convergence high complexity. Addressing these issues, this study proposes a hybrid model that combines Adaptive Neuro-Fuzzy Inference System (ANFIS) with Gene Expression Programming (GEP) to enhance predictions electrical consumption. Validated using real-time monthly load data from an user Uganda, outperforms individual ANFIS GEP models, demonstrating reduced errors minimal computation time. The application presents promising results, showcasing exceptional predictive capabilities offering potential improvements efficiency precision for consumption evolving

Language: Английский

---

LSTM vs. Prophet: Achieving Superior Accuracy in Dynamic Electricity Demand Forecasting

Energies, Journal Year: 2025, Volume and Issue: 18(2), P. 278 - 278

Published: Jan. 10, 2025

Accurate electricity demand forecasting is critical for improving energy efficiency, maintaining grid stability, reducing operational costs, and promoting sustainability. This study presents a novel hybrid model that integrates Long Short-Term Memory (LSTM) networks Prophet models, leveraging their complementary strengths through dynamic weighted ensemble methodology. The LSTM component captures nonlinear dependencies long-term temporal patterns, while models seasonal trends event-driven fluctuations. was evaluated using comprehensive dataset of hourly consumption from Ontario, Canada, achieving Root Mean Square Error (RMSE) 65.34, Absolute Percentage (MAPE) 7.3%, an R2 0.98. These results demonstrate significant improvements over standalone LSTM, Prophet, other State-of-the-Art methods, highlighting the model’s adaptability superior accuracy. underscores practical implications approach, particularly in management resource optimization, setting new benchmark time series sector.

Language: Английский

---

Novel approaches to optimize the layouts of solar photovoltaic and wind power systems to improve their performance considering limited land availability and site-specific features

Sustainable Energy Technologies and Assessments, Journal Year: 2025, Volume and Issue: 74, P. 104189 - 104189

Published: Jan. 20, 2025

Language: Английский

---

Modeling Energy Demand—A Systematic Literature Review

Energies, Journal Year: 2021, Volume and Issue: 14(23), P. 7859 - 7859

Published: Nov. 23, 2021

In this article, a systematic literature review of 419 articles on energy demand modeling, published between 2015 and 2020, is presented. This provides researchers with an exhaustive overview the examined classification techniques for modeling. Unlike in existing reviews, comprehensive study all following aspects models are analyzed: techniques, prediction accuracy, inputs, carrier, sector, temporal horizon, spatial granularity. Readers benefit from easy access to broad base find decision support when choosing suitable data-model combinations their projects. Results have been compiled figures tables, providing structured summary literature, containing direct references analyzed articles. Drawbacks discussed as well countermeasures. The results show that among articles, machine learning (ML) used most, mainly applied short-term electricity forecasting regional level rely historic load main data source. Engineering-based less dependent cover appliance consumption long horizons. Metaheuristic uncertainty often hybrid models. Statistical frequently modeling serve benchmarks other techniques. Among accuracy measured by mean average percentage error (MAPE) proved be similar levels eases reader into subject matter presenting emphases made current suggesting future research directions, basis quantitative testing hypotheses regarding applicability dominance specific methods sub-categories

Language: Английский

---

Energy Consumption Forecasting for the Digital-Twin Model of the Building

Energies, Journal Year: 2022, Volume and Issue: 15(12), P. 4318 - 4318

Published: June 13, 2022

The aim of the paper is to propose a new approach forecast energy consumption for next day using unique data obtained from digital twin model building. In research, we tested which chosen forecasting methods and set input gave best results. We naive methods, linear regression, LSTM Prophet method. found that information about total real top 10 energy-consuming devices following day. this paper, also presented methodology decision trees conditional attributes understand errors made by model. This was proposed reduce number monitored devices. research described in article carried out context project deals with development

Language: Английский

---

Load forecasting of district heating system based on improved FB-Prophet model

Energy, Journal Year: 2023, Volume and Issue: 278, P. 127637 - 127637

Published: May 3, 2023

Language: Английский

---

Statistical Comparison of Time Series Models for Forecasting Brazilian Monthly Energy Demand Using Economic, Industrial, and Climatic Exogenous Variables

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(13), P. 5846 - 5846

Published: July 4, 2024

Energy demand forecasting is crucial for effective resource management within the energy sector and aligned with objectives of Sustainable Development Goal 7 (SDG7). This study undertakes a comparative analysis different models to predict future trends in Brazil, improve methodologies, achieve sustainable development goals. The evaluation encompasses following models: Seasonal Autoregressive Integrated Moving Average (SARIMA), Exogenous SARIMA (SARIMAX), Facebook Prophet (FB Prophet), Holt–Winters, Trigonometric Seasonality Box–Cox transformation, ARMA errors, Trend, components (TBATS), draws attention their respective strengths limitations. Its findings reveal unique capabilities among models, excelling tracing seasonal patterns, FB demonstrating its potential applicability across various sectors, Holt–Winters adept at managing fluctuations, TBATS offering flexibility albeit requiring significant data inputs. Additionally, investigation explores effect external factors on consumption, by establishing connections through Granger causality test conducting correlation analyses. accuracy these assessed without exogenous variables, categorized as economical, industrial, climatic. Ultimately, this seeks add body knowledge prediction, well allow informed decision-making planning policymaking and, thus, make rapid progress toward SDG7 associated targets. paper concludes that, although achieves best accuracy, most fit model, considering residual autocorrelation, it predicts that Brazil will approximately 70,000 GWh 2033.

Language: Английский

---