Modeling and predicting the electricity production in hydropower using conjunction of wavelet transform, long short-term memory and random forest models

Abstract Electricity is an important pillar for the economic growth and the development of societies. Surveying and predicting the electricity production (EP) is a valuable factor in the hands of electricity industry managers to make strategic decisions, especially if electricity is generated by renewable resources for environmental considerations. However, because the EP series is non-stationary and nonlinear, traditional methods are less robust to predict it. In this study, we offer a hybrid model which combines adaptive wavelet transform (AWT), long short-term memory (LSTM) and random forest (RF) algorithm (AWT-LSTM-RF) to predict the EP in hydroelectric power plant. We also apply the exogenous affecting variables on EP in the structure of hybrid model, which were selected by ant colony optimization (ACO) algorithm. To evaluate the predictive power of the AWT-LSTM-RF model, we compared its predictive results with the benchmark models including RF, ARIMA-GARCH, wavelet transform-feed forward neural network (WT-FFNN), wavelet transform-random forest (WT-RF), wavelet transform-LSTM (WT-LSTM), and WT-FFNN-RF. The empirical results indicate that the hybrid model of AWT-LSTM-RF outperforms the benchmark models. The results also suggest that applying the wavelet transform on input data of the RF algorithm (WT-RF) can improve the predictive power of the RF.

[1]  Abheejeet Mohapatra,et al.  Repeated wavelet transform based ARIMA model for very short-term wind speed forecasting , 2019, Renewable Energy.

[2]  Hui Qin,et al.  Interval prediction method based on Long-Short Term Memory networks for system integrated of hydro, wind and solar power , 2019, Energy Procedia.

[3]  Yinhai Wang,et al.  Stacked Bidirectional and Unidirectional LSTM Recurrent Neural Network for Forecasting Network-wide Traffic State with Missing Values , 2020, Transportation Research Part C: Emerging Technologies.

[4]  Samia Boukir,et al.  Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests , 2011 .

[5]  Y. Dawen.  Prediction of monthly inflow to the Danjiangkou reservoir by distributed hydrological model and hydro-climatic teleconnections , 2013 .

[6]  Yacine Rezgui,et al.  Tree-based ensemble methods for predicting PV power generation and their comparison with support vector regression , 2018, Energy.

[7]  Hannu Olkkonen,et al.  Discrete Wavelet Transforms - Biomedical Applications , 2011 .

[8]  H. L. Resnikoff,et al.  Wavelet analysis: the scalable structure of information , 1998 .

[9]  Javier M. Moguerza,et al.  Electricity Price Forecasting with Dynamic Trees: A Benchmark Against the Random Forest Approach , 2018 .

[10]  Li Zheng,et al.  Research on Small Hydropower Generation Forecasting Method Based on Improved BP Neural Network , 2016 .

[11]  Shahaboddin Shamshirband,et al.  Modeling monthly pan evaporation using wavelet support vector regression and wavelet artificial neural networks in arid and humid climates , 2019, Engineering Applications of Computational Fluid Mechanics.

[12]  Wasim Ahmad,et al.  A new forecasting model with wrapper-based feature selection approach using multi-objective optimization technique for chaotic crude oil time series , 2020 .

[13]  Özgür Kisi,et al.  Predicting river daily flow using wavelet-artificial neural networks based on regression analyses in comparison with artificial neural networks and support vector machine models , 2017, Neural computing & applications (Print).

[14]  Bart J. van der Lugt,et al.  Conditional Forecasting of Water Level Time Series with RNNs , 2019, AALTD@PKDD/ECML.

[15]  Selvam Nallathambi,et al.  Prediction of electricity consumption based on DT and RF: An application on USA country power consumption , 2017, 2017 IEEE International Conference on Electrical, Instrumentation and Communication Engineering (ICEICE).

[16]  Zong-Chang Yang.  Discrete cosine transform-based predictive model extended in the least-squares sense for hourly load forecasting , 2016 .

[17]  Guang Yang,et al.  Multiobjective reservoir operating rules based on cascade reservoir input variable selection method , 2017 .

[18]  Jianzhong Zhou,et al.  Integrated optimization of hydroelectric energy in the upper and middle Yangtze River , 2015 .

[19]  Jianren Fan,et al.  Predictive single-step kinetic model of biomass devolatilization for CFD applications: A comparison study of empirical correlations (EC), artificial neural networks (ANN) and random forest (RF) , 2019, Renewable Energy.

[20]  Hristos Tyralis,et al.  Evaluation of random forests and Prophet for daily streamflow forecasting , 2018, Advances in Geosciences.

[21]  Jeng-Fung Chen,et al.  Forecasting Monthly Electricity Demands by Wavelet Neuro-Fuzzy System Optimized by Heuristic Algorithms , 2018, Inf..

[22]  Seçkin Karasu,et al.  Recognition of COVID-19 disease from X-ray images by hybrid model consisting of 2D curvelet transform, chaotic salp swarm algorithm and deep learning technique , 2020, Chaos, Solitons & Fractals.

[23]  Wei Sun,et al.  A Hybrid Carbon Price Forecasting Model with External and Internal Influencing Factors Considered Comprehensively: A Case Study from China , 2020 .

[24]  Wenbo Chen,et al.  Electricity price prediction based on hybrid model of adam optimized LSTM neural network and wavelet transform , 2019, Energy.

[25]  Ijaz Ahmad,et al.  Rainfall-runoff modeling at Jinsha River basin by integrated neural network with discrete wavelet transform , 2019, Meteorology and Atmospheric Physics.

[26]  G. Notton,et al.  Solar radiation forecasting using artificial neural network and random forest methods: Application to normal beam, horizontal diffuse and global components , 2019, Renewable Energy.

[27]  Aytaç Altan,et al.  Recognition Model for Solar Radiation Time Series based on Random Forest with Feature Selection Approach , 2019, 2019 11th International Conference on Electrical and Electronics Engineering (ELECO).

[28]  J. L. Tena García,et al.  Generating electrical demand time series applying SRA technique to complement NAR and sARIMA models , 2019, Energy Efficiency.

[29]  Lifeng Wu,et al.  Hybrid support vector machines with heuristic algorithms for prediction of daily diffuse solar radiation in air-polluted regions , 2020 .

[30]  Bahram Sahabi,et al.  A hybrid approach to model and forecast the electricity consumption by NeuroWavelet and ARIMAX-GARCH models , 2019 .

[31]  Ali Lahouar,et al.  Hour-ahead wind power forecast based on random forests , 2017 .

[32]  Seema Singh,et al.  Forecasting of Short-Term Electric Load Using Application of Wavelets with Feed-Forward Neural Networks , 2010 .

[33]  José A. S. Sá,et al.  Artificial neural networks approaches for predicting the potential for hydropower generation: a case study for Amazon region , 2019, J. Intell. Fuzzy Syst..

[34]  Quan Cui,et al.  An innovative random forest-based nonlinear ensemble paradigm of improved feature extraction and deep learning for carbon price forecasting. , 2020, The Science of the total environment.

[35]  V. K. Giri,et al.  Solar radiation forecasting using MARS, CART, M5, and random forest model: A case study for India , 2019, Heliyon.

[36]  Mohamed Lazaar,et al.  Integration of Principal Component Analysis and Recurrent Neural Network to Forecast the Stock Price of Casablanca Stock Exchange , 2019, Procedia Computer Science.

[37]  Vishal Kushwaha,et al.  A SARIMA-RVFL hybrid model assisted by wavelet decomposition for very short-term solar PV power generation forecast , 2019, Renewable Energy.

[38]  Yong Xiang,et al.  A double decomposition-based modelling approach to forecast weekly solar radiation , 2020 .

[39]  M. Varol.  CO2 emissions from hydroelectric reservoirs in the Tigris River basin, a semi-arid region of southeastern Turkey , 2019, Journal of Hydrology.

[40]  Ken Nagasaka,et al.  An artificial neural network based micro-hydropower generation scheduling: considering demand profile and environmental impact , 2006 .

[41]  Thomas Stützle,et al.  Ant Colony Optimization , 2009, EMO.

[42]  Jean-François Toubeau,et al.  Recalibration of recurrent neural networks for short-term wind power forecasting , 2021 .

[43]  Araceli Sanchis,et al.  Hydroelectric power plant management relying on neural networks and expert system integration , 2000 .

[44]  Dongxiao Niu,et al.  Short-term photovoltaic power generation forecasting based on random forest feature selection and CEEMD: A case study , 2020, Appl. Soft Comput..

[45]  Ahmed N. Abdalla,et al.  Prediction of small hydropower plant power production in Himreen Lake dam (HLD) using artificial neural network , 2017 .

[46]  Mónica Mite-León,et al.  Statistical Model for the Forecast of Hydropower Production in Ecuador , 2018, International Journal of Renewable Energy Research.

[47]  Robert P. Sheridan,et al.  Random Forest: A Classification and Regression Tool for Compound Classification and QSAR Modeling , 2003, J. Chem. Inf. Comput. Sci..

[48]  M. Golabi,et al.  Biochemical oxygen demand prediction: development of hybrid wavelet-random forest and M5 model tree approach using feature selection algorithms , 2020, Environmental Science and Pollution Research.

[49]  Adriano C. M. Pereira,et al.  Stock market's price movement prediction with LSTM neural networks , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[50]  Olawale Popoola,et al.  Residential lighting load profile modelling: ANFIS approach using weighted and non-weighted data , 2017 .

[51]  Tao Zhang,et al.  A novel multiscale nonlinear ensemble leaning paradigm for carbon price forecasting , 2018 .

[52]  Francisco C. Pereira,et al.  Multi-output bus travel time prediction with convolutional LSTM neural network , 2019, Expert Syst. Appl..

[53]  B. Ghobadian,et al.  Biodiesel production from Norouzak (Salvia lerifolia) seeds as an indigenous source of bio fuel in Iran using ultrasound , 2015 .

[54]  Guoqiang Chen,et al.  Multiple Random Forests Modelling for Urban Water Consumption Forecasting , 2017, Water Resources Management.

[55]  Yanhong Luo,et al.  Wind Power Prediction Based on LSTM Networks and Nonparametric Kernel Density Estimation , 2019, IEEE Access.

[56]  A. Kazemi,et al.  Selection of the best ARMAX model for forecasting energy demand: case study of the residential and commercial sectors in Iran , 2016 .

[57]  Farid Mohammad Maalek Ghaini,et al.  Improving DWT-RNN model via B-spline wavelet multiresolution to forecast a high-frequency time series , 2019, Expert Syst. Appl..

[58]  Mohammad Hossein Niksokhan,et al.  Development of an indirect method for modelling the water footprint of electricity using wavelet transform coupled with the random forest model , 2020 .

[59]  Yong He,et al.  Short-Term Power Generation Energy Forecasting Model for Small Hydropower Stations Using GA-SVM , 2014 .

[60]  Murat Kankal,et al.  Estimates of hydroelectric generation using neural networks with the artificial bee colony algorithm for Turkey , 2014 .

[61]  Taesam Lee,et al.  Deep Learning-Based Maximum Temperature Forecasting Assisted with Meta-Learning for Hyperparameter Optimization , 2020 .

[62]  Jing Shi,et al.  Applying ARMA–GARCH approaches to forecasting short-term electricity prices , 2013 .

[63]  Licheng Liu,et al.  Analyzing big time series data in solar engineering using features and PCA , 2017 .

[64]  Hao Chen,et al.  Refined Diebold-Mariano Test Methods for the Evaluation of Wind Power Forecasting Models , 2014 .

[65]  N. Arshad,et al.  Complementing hydroelectric power with floating solar PV for daytime peak electricity demand , 2020 .

[66]  Yaxing Wei,et al.  Projecting changes in annual hydropower generation using regional runoff data: An assessment of the United States federal hydropower plants , 2015 .

[67]  Sílvio Mariano,et al.  A bat optimized neural network and wavelet transform approach for short-term price forecasting , 2018 .

[68]  F. Diebold,et al.  Comparing Predictive Accuracy , 1994, Business Cycles.

[69]  Patrick J. Van Fleet,et al.  Wavelet Theory: An Elementary Approach with Applications , 2009 .

[70]  Ranjeeta Bisoi,et al.  Forecasting financial time series using a low complexity recurrent neural network and evolutionary learning approach , 2017, J. King Saud Univ. Comput. Inf. Sci..

[71]  Ping-Feng Pai,et al.  A Survey of Machine Learning Models in Renewable Energy Predictions , 2020, Applied Sciences.

[72]  R. P. Hasabe,et al.  A Comparison of Different Mother Wavelet for Fault Detection & Classification of Series Compensated Transmission Line , 2015 .

[73]  Changjun Zhou,et al.  Forecasting stock prices with long-short term memory neural network based on attention mechanism , 2020, PloS one.

[74]  Yugang Niu,et al.  Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM , 2018 .

[75]  Nikolaos Doulamis,et al.  Deep Learning for Computer Vision: A Brief Review , 2018, Comput. Intell. Neurosci..

[76]  Ozgur Kisi,et al.  Applications of hybrid wavelet–Artificial Intelligence models in hydrology: A review , 2014 .

[77]  Truong Q. Nguyen,et al.  Forecasting of Solar Photovoltaic System Power Generation Using Wavelet Decomposition and Bias-Compensated Random Forest , 2017, 2017 Ninth Annual IEEE Green Technologies Conference (GreenTech).

[78]  Sadegh Sadeghi,et al.  Thermodynamic analysis and optimization of a geothermal Kalina cycle system using Artificial Bee Colony algorithm , 2016 .

[79]  Jianzhong Wu,et al.  Short-term wind power forecasting using wavelet-based neural network , 2017 .

[80]  Enrico Zio,et al.  A hybrid hourly natural gas demand forecasting method based on the integration of wavelet transform and enhanced Deep-RNN model , 2019, Energy.

[81]  Farshid Keynia,et al.  A new short-term wind speed forecasting method based on fine-tuned LSTM neural network and optimal input sets , 2020, Energy Conversion and Management.

[82]  V. Rodriguez-Galiano,et al.  Predictive modeling of groundwater nitrate pollution using Random Forest and multisource variables related to intrinsic and specific vulnerability: a case study in an agricultural setting (Southern Spain). , 2014, The Science of the total environment.

[83]  J. Ben Hadj Slama,et al.  Day-ahead load forecast using random forest and expert input selection , 2015 .

[84]  Marcella Busilacchio,et al.  Recursive neural network model for analysis and forecast of PM10 and PM2.5 , 2017 .

[85]  Chuan Li,et al.  Improving forecasting accuracy of daily enterprise electricity consumption using a random forest based on ensemble empirical mode decomposition , 2018, Energy.

[86]  Jian-zhong Zhou,et al.  Future hydropower generation prediction of large-scale reservoirs in the upper Yangtze River basin under climate change , 2020 .

[87]  Arun Agarwal,et al.  Recurrent neural network and a hybrid model for prediction of stock returns , 2015, Expert Syst. Appl..

[88]  Joao P. S. Catalao,et al.  Short-term wind power forecasting in Portugal by neural networks and wavelet transform , 2011 .

[89]  Eenjun Hwang,et al.  Hybrid Short-Term Load Forecasting Scheme Using Random Forest and Multilayer Perceptron , 2018, Energies.

[90]  Bellie Sivakumar,et al.  Forecasting river water temperature time series using a wavelet–neural network hybrid modelling approach , 2019, Journal of Hydrology.

[91]  Hydroelectricity water footprint in Parana Hydrograph Region, Brazil , 2020 .

[92]  F. Hossain,et al.  Maximizing energy production from hydropower dams using short-term weather forecasts , 2020 .

[93]  Irena Koprinska,et al.  Forecasting electricity load with advanced wavelet neural networks , 2016, Neurocomputing.

[94]  Gwo-Ching Liao,et al.  Application of a fuzzy neural network combined with a chaos genetic algorithm and simulated annealing to short-term load forecasting , 2006, IEEE Transactions on Evolutionary Computation.