# Exploiting Deep Learning for Wind Power Forecasting Based on Big Data Analytics

^{1}

^{2}

^{3}

^{4}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Related Work

## 3. Contributions

- A novel big data-driven wind power prediction model is proposed that combines the strengths of both the univariate and multivariate wind power forecasting techniques by using decomposed and exogenous inputs for forecasting; consequently, the forecasting accuracy is significantly enhanced.
- The proposed model employs an existing method wavelet packet decomposition and an enhanced method Efficient DCNN (EDCNN) for feature extraction and forecasting, respectively.
- A DSM algorithm is also proposed. The proposed DSM algorithm takes into account the day-ahead demand, day-ahead price and wind power.
- The proposed DSM algorithm reduces the consumption cost and improves the load profile to almost a normal shape.

## 4. Proposed Model

#### 4.1. Data Preprocess

#### 4.2. Feature Engineering

#### 4.3. Efficient DCNN

## 5. System Description

#### 5.1. Case 1

#### 5.2. Case 2

#### 5.3. Case 3

- -
- Load factor maximization
- -
- Consumption cost minimization

## 6. Problem Formulation

Algorithm 1 Algorithm for Demand Side Management (DSM). | ||

Require:Input: [W, L, P] | ||

1: Output: C | ||

2: if W = L then | ▹ Wind power is sufficient to fulfill demand | |

3: ${P}_{new}=0$ | ▹ Wind power is sufficient that has no cost | |

4: ${L}_{new}=L$ | ▹ Load is equal to wind power, so load adjustment is not performed | |

5: $C={P}_{new}\times {L}_{new}$ | ▹ Calculating consumption cost | |

6: else if W > L then | ▹ Wind power is greater than demand | |

7: $W-L\to SG$ | ▹ Excessive wind power is transmitted to the SG | |

8: S = 0.9 | ▹ 10 % reduction in price is subsidiary for next power purchase | |

9: ${P}_{new}=0$ | ▹ Wind power is sufficient that has no cost | |

10: ${L}_{new}=L$ | ▹ Load is lesser than wind power, so load adjustment is not performed | |

11: $C={P}_{new}\times {L}_{new}$ | ▹ Calculating the consumption cost | |

12: else if $W\phantom{\rule{3.33333pt}{0ex}}\ge \phantom{\rule{3.33333pt}{0ex}}0\phantom{\rule{3.33333pt}{0ex}}AND\phantom{\rule{3.33333pt}{0ex}}W\phantom{\rule{3.33333pt}{0ex}}<\phantom{\rule{3.33333pt}{0ex}}L$ then | ▹ Wind power is not sufficient to fulfill the demand | |

13: $DWD\phantom{\rule{3.33333pt}{0ex}}=\phantom{\rule{3.33333pt}{0ex}}L\phantom{\rule{3.33333pt}{0ex}}-\phantom{\rule{3.33333pt}{0ex}}W$ | ▹ Finding demand that have to be fulfilled by the SG | |

14: ${L}_{new}=Manage\_Demand(DWD,\phantom{\rule{3.33333pt}{0ex}}L)$ | ▹ Managing demand to distribute it normally | |

15: if S = 0.9 then | ▹ If there is subsidiary on the price, the price will be adjusted | |

16: ${P}_{new}=P\times S$ | ▹ 10% reduction on price by subsidiary | |

17: $C={P}_{new}\times {L}_{new}$ | ▹ Calculating consumption cost | |

18: else | ||

19: ${P}_{new}=P$ | ▹ If there is no subsidiary on price, price remains same | |

20: $C={P}_{new}\times {L}_{new}$ | ▹ Calculating consumption cost | |

21: end if | ||

22: end if | ||

23: Manage_Demand Function | ||

24: Function ${L}_{new}=Manage\_Demand(DWD,\phantom{\rule{3.33333pt}{0ex}}L)$ | ||

25: $\mu =mean\left(DWD\right)$ | ▹ Average of demand to be fulfilled by the SG | |

26: $\sigma =std\left(DWD\right)$ | ▹ Standard deviation of demand to be fulfilled by the SG | |

27: $SD=sum\left(DWD\right)$ | ▹ Sum of demand to be fulfilled by the SG | |

28: if DWD < $\mu $ then | ▹ Checking each value of demand vector if it is smaller than mean | |

29: ${L}^{\prime}=L+\sigma $ | ▹ When value is smaller, add standard deviation to make it closer to mean | |

30: else if DWD > $\mu $ then | ▹ Checking each value of demand vector if it is greater than mean | |

31: ${L}^{\prime}=L-\sigma $ | ▹ When value is larger, subtract standard deviation to make it closer to mean | |

32: end if | ||

33: $SL=sum\left({L}^{\prime}\right)$ | ▹ Taking sum of all values of new adjusted load vector | |

34: d = SL – SD | ▹ Taking the difference of demanding load and new adjusted load | |

35: ⊳ | ▹ Now the demanded load and new load are adjusted to be equal | |

36: if d > 0 then | ▹ Difference greater than zero means the new adjusted load is more than the demanded load | |

37: $\left[idx\phantom{\rule{3.33333pt}{0ex}}Count\right]=L>\mu $ | ▹ Count is the number of values greater than average and index are their index | |

38: ${L}_{new}=L\left(indx\right)-\frac{d}{count}$ | ▹ Subtracting the difference from all the larger values | |

39: else if d < 0 then | ▹ Difference smaller than zero means the new adjusted load is lesser than the demanded | |

40: load | ||

41: $\left[indx\phantom{\rule{3.33333pt}{0ex}}Count\right]=L<\mu $ | ▹ Count is the number of values that are smaller than average load | |

42: ${L}_{new}=L\left(indx\right)+\frac{d}{count}$ | ▹ Adding the difference in all the smaller values | |

43: end if | ||

44: [index ${L}_{sorted}$] = Sort(${L}_{new}$) | ▹ Sort will sort the ${L}_{new}$ in ascending order and return index of the sorted array | |

45: ${L}_{sorted}$ | ||

46: For i = 1 to 6 | ▹ Shift the peak load to the lowest load | |

47: j = i-1, sf = 5*i a = length(${L}_{new}$) | ▹ Defining shifting factor | |

48: if index(i) > 6 then | ▹ Shift the load to the lowest load that is not late night | |

49: shftFac = $\frac{{L}_{new}\left(index(a-j)\right)}{sf}$ | ||

50: ${L}_{new}$(index(i)) = ${L}_{new}$(index(i)) - shftFac | ▹ Subtracting the shifting factor from the highest load | |

51: ${L}_{new}$(index(a-j)) = ${L}_{new}$(index(a-j)) + shftFac | ▹ Adding the shifting factor to the lowest load | |

52: end if | ||

53: End For | ||

54: End Function |

## 7. Results and Analysis

#### 7.1. Data Description

#### 7.2. Wind Power Analysis

#### 7.3. EDCNN Performance Evaluation

#### 7.4. Statistical Analysis of EDCNN

#### 7.5. Analysis of Proposed DSM Algorithm

## 8. Conclusions and Future Work

## Author Contributions

## Funding

## Conflicts of Interest

## Nomenclature

ABC | Artificial Bee Colony |

ANN | Artificial Neural Networks |

AEMO | Australia Electricity Market Operator |

ARIMA | Autoregressive Integrated Moving Average |

CASIO | California Independent System Operators |

CNN | Convolution Neural Networks |

DWT | Discrete Wavelet Transform |

DM | Diebold–Mariano (statistical test) |

DNN | Deep Neural Networks |

DSM | Demand Side Management |

ELM | Extreme Learning Machine |

EROL | Enhanced Regression Output Layer |

ISO NE | Independent System Operator New England |

LSSVM | Least Square Support Vector Machine |

LSTM | Long Short Term Memory |

MAE | Mean Absolute Error |

MAPE | Mean Absolute Percentage Error |

MISO | Mid-continent Independent System Operator |

NRMSE | Normalized Root Mean Square Error |

ReLU | Rectified Linear Unit |

RNN | Recurrent Neural Network |

SAE | Sparse Auto Encoders |

SCADA | Supervisory Control And Data Acquisition |

STLF | Short-Term Load Forecast |

TV | Time Varying |

WPP | Wind Power Plant |

WPT | Wavelet Packet Transform |

a | Input signal to wavelet transform |

${b}^{m}$ | Bias of mth hidden layer |

C | Power consumption cost |

${d}_{t}$ | Differential loss of forecasting models’ error |

$DWD$ | Demand and wind generation difference |

$\sigma \left(\right)$ | Sigmoid function |

L | Load vector |

$LF$ | Load factor |

P | Price vector |

S | Subsidiary |

${\epsilon}^{FM}$ | Error of forecasting model |

$\varphi \left(\right)$ | Radial base function |

$\psi \left(\right)$ | Wavelet function |

${w}^{m}$ | Weights of mth layer |

${X}_{i}^{m}$ | Feature map of ${X}_{i}$, mth layer |

## References

- Zhao, Y.N.; Ye, L.; Li, Z.; Song, X.R.; Lang, Y.S.; Su, J. A novel bidirectional mechanism based on time series model for wind power forecasting. Appl. Energy
**2016**, 177, 793–803. [Google Scholar] [CrossRef] - U.S. Department of Energy. Staff Report to the Secretary on Electricity Markets and Reliability. 2017. Available online: https://www.energy.gov/downloads/download-staff-report-secretary-electricity-markets-and-reliability (accessed on 8 June 2019).
- Jong, P.; Kiperstok, A.; Sanchez, A.S.; Dargaville, R.; Torres, E.A. Integrating large scale wind power into the electricity grid in the Northeast of Brazil. Energy
**2016**, 100, 401–415. [Google Scholar] [CrossRef] - Global Wind Energy Council. GWEC Global Wind Report 2016. Available online: https://gwec.net/publications/global-wind-report-2/global-wind-report-2016 (accessed on 8 June 2019).
- Shafiee, M.; Brennan, F.; Espinosa, I.A. A parametric whole life cost model for offshore wind farms. Int. J. Life Cycle Assess.
**2016**, 21, 961–975. [Google Scholar] [CrossRef][Green Version] - Tomporowski, A.; Flizikowski, J.; Kasner, R.; Kruszelnicka, W. Environmental control of wind power technology. Rocznik Ochrona Srodowiska
**2017**, 19, 694–714. [Google Scholar] - U.S. Department of Energy, 20% Wind Energy by 2030: Increasing Wind Energy’s Contribution to US Electricity Supply, Energy Efficiency and Renewable Energy (EERE). 2008. Available online: https://www.energy.gov/eere/wind (accessed on 8 June 2019).
- Athari, M.H.; Wang, Z. Impacts of Wind Power Uncertainty on Grid Vulnerability to Cascading Overload Failures. IEEE Trans. Sustain. Energy
**2018**, 9, 128–137. [Google Scholar] [CrossRef] - Wang, Q.; Martinez-Anido, C.B.; Wu, H.; Florita, A.R.; Hodge, B.M. Quantifying the economic and grid reliability impacts of improved wind power forecasting. IEEE Trans. Sustain. Energy
**2016**, 7, 1525–1537. [Google Scholar] [CrossRef] - Swinand, G.P.; Mahoney, A.O. Estimating the impact of wind generation and wind forecast errors on energy prices and costs in Ireland. Renew. Energy
**2015**, 75, 468–473. [Google Scholar] [CrossRef][Green Version] - Chen, Z. Wind power in modern power systems. J. Mod. Power Syst. Clean Energy
**2013**, 1, 2–13. [Google Scholar] [CrossRef][Green Version] - Haque, A.U.; Nehrir, M.H.; Mandal, P. A hybrid intelligent model for deterministic and quantile regression approach for probabilistic wind power forecasting. IEEE Trans. Power Syst.
**2014**, 29, 1663–1672. [Google Scholar] [CrossRef] - Juban, J.; Siebert, N.; Kariniotakis, G.N. Probabilistic short-term wind power forecasting for the optimal management of wind generation. In Proceedings of the 2007 IEEE Lausanne Power Tech, Lausanne, Switzerland, 1–5 July 2007; pp. 683–688. [Google Scholar]
- Zhou, K.; Fu, C.; Yang, S. Big data driven smart energy management: From big data to big insights. Renew. Sustain. Energy Rev.
**2016**, 56, 215–225. [Google Scholar] [CrossRef] - Haupt, S.E.; Kosovic, B. Variable generation power forecasting as a big data problem. IEEE Trans. Sustain. Energy
**2016**, 8, 725–732. [Google Scholar] [CrossRef] - Zhang, J.; Cui, M.; Hodge, B.M.; Florita, A.; Freedman, J. Ramp forecasting performance from improved short-term wind power forecasting over multiple spatial and temporal scales. Energy
**2017**, 122, 528–541. [Google Scholar] [CrossRef][Green Version] - Zhang, Q.; Yang, L.T.; Chen, Z.; Li, P. A survey on deep learning for big data. Inf. Fusion
**2018**, 42, 146–157. [Google Scholar] [CrossRef] - Najafabadi, M.M.; Villanustre, F.; Khoshgoftaar, T.M.; Seliya, N.; Wald, R.; Muharemagic, E. Deep learning applications and challenges in big data analytics. J. Big Data
**2015**, 2, 1–15. [Google Scholar] [CrossRef] - Mujeeb, S.; Javaid, N.; Gul, H.; Daood, N.; Shabbir, S.; Arif, A. Wind Power Forecasting Based on Efficient Deep Convolution Neural Networks. In Proceedings of the 3PGCIC Conference, Antwerp, Belgium, 7–9 November 2019. [Google Scholar]
- Xu, Q.; He, D.; Zhang, N.; Kang, C.; Xia, Q.; Bai, J.; Huang, J. A short-term wind power forecasting approach with adjustment of numerical weather prediction input by data mining. IEEE Trans. Sustain. Energy
**2015**, 6, 1283–1291. [Google Scholar] [CrossRef] - Zhang, G.; Wu, Y.; Wong, K.P.; Xu, Z.; Dong, Z.Y.; Herbert, H.C. An advanced approach for construction of optimal wind power prediction intervals. IEEE Trans. Power Syst.
**2015**, 30, 2706–2715. [Google Scholar] [CrossRef] - Wang, J.; Niu, T.; Lu, H.; Yang, W.; Du, P. A Novel Framework of Reservoir Computing for Deterministic and Probabilistic Wind Power Forecasting. IEEE Trans. Sustain. Energy
**2019**. [Google Scholar] [CrossRef] - Yang, S.; Xu, X.; Xiang, J.; Chen, R.; Jiang, W. The efficient market operation for wind energy trading based on the dynamic improved power forecasting. J. Renew. Sustain. Energy
**2018**, 10, 45908–45921. [Google Scholar] [CrossRef] - Cavalcante, L.; Bessa, R.J.; Reis, M.; Browell, J. LASSO vector autoregression structures for very short-term wind power forecasting. Wind Energy
**2017**, 20, 657–675. [Google Scholar] [CrossRef] - Yan, J.; Li, K.; Bai, E.; Zhao, X.; Xue, Y.; Foley, A.M. Analytical Iterative Multistep Interval Forecasts of Wind Generation Based on TLGP. IEEE Trans. Sustain. Energy
**2018**, 10, 625–636. [Google Scholar] [CrossRef][Green Version] - Lin, Y.; Yang, M.; Wan, C.; Wang, J.; Song, Y. A multi-model combination approach for probabilistic wind power forecasting. IEEE Trans. Sustain. Energy
**2019**, 10, 226–237. [Google Scholar] [CrossRef] - Ellis, N.; Davy, R.; Troccoli, A. Predicting wind power variability events using different statistical methods driven by regional atmospheric model output. Wind Energy
**2015**, 18, 1611–1628. [Google Scholar] [CrossRef] - Yan, J.; Li, K.; Bai, E.W.; Deng, J.; Foley, A.M. Hybrid probabilistic wind power forecasting using temporally local Gaussian process. IEEE Trans. Sustain. Energy
**2016**, 7, 87–95. [Google Scholar] [CrossRef] - Li, S.; Wang, P.; Goel, L. Wind power forecasting using neural network ensembles with feature selection. IEEE Trans. Sustain Energy
**2015**, 6, 1447–1456. [Google Scholar] [CrossRef] - Lee, D.; Baldick, R. Short-term wind power ensemble prediction based on Gaussian processes and neural networks. IEEE Trans. Smart Grid
**2014**, 5, 501–510. [Google Scholar] [CrossRef] - Chitsaz, H.; Amjady, N.; Zareipour, H. Wind power forecast using wavelet neural network trained by improved Clonal selection algorithm. Energy Convers. Manag.
**2015**, 89, 588–598. [Google Scholar] [CrossRef] - Wu, W.; Peng, M. A Data Mining Approach Combining K-Means Clustering With Bagging Neural Network for Short-Term Wind Power Forecasting. IEEE Internet Things J.
**2017**, 4, 979–986. [Google Scholar] [CrossRef] - Qureshi, A.S.; Khan, A.; Zameer, A.; Usman, A. Wind power prediction using deep neural network based meta regression and transfer learning. Appl. Soft Comput.
**2017**, 58, 742–755. [Google Scholar] [CrossRef] - Shao, H.; Deng, X.; Jiang, Y. A novel deep learning approach for short-term wind power forecasting based on infinite feature selection and recurrent neural network. J. Renew. Sustain. Energy
**2018**, 10, 43303–43314. [Google Scholar] [CrossRef] - Torres, J.M.; Aguilar, R.M.; Zuniga-Meneses, K.V. Deep learning to predict the generation of a wind farm. J. Renew. Sustain. Energy
**2018**, 10, 013305. [Google Scholar] [CrossRef] - Wang, H.Z.; Li, G.Q.; Wang, G.B.; Peng, J.C.; Jiang, H.; Liu, Y.T. Deep learning based ensemble approach for probabilistic wind power forecasting. Appl. Energy
**2017**, 188, 56–70. [Google Scholar] [CrossRef] - Torres, J.M.; Aguilar, R.M. Using deep learning to predict complex systems: A case study in wind farm generation. Complexity
**2018**, 2018, 9327536. [Google Scholar] [CrossRef] - Stephens, E.R.; Smith, D.B.; Mahanti, A. Game theoretic model predictive control for distributed energy demand-side management. IEEE Trans. Smart Grid
**2015**, 6, 1394–1402. [Google Scholar] [CrossRef] - Ghasemi, A.; Shayeghi, H.; Moradzadeh, M.; Nooshyar, M. A novel hybrid algorithm for electricity price and load forecasting in smart grids with demand-side management. Appl. Energy
**2016**, 177, 40–59. [Google Scholar] [CrossRef] - Pascual, J.; Barricarte, J.; Sanchis, P.; Marroyo, L. Energy management strategy for a renewable-based residential microgrid with generation and demand forecasting. Appl. Energy
**2015**, 158, 12–25. [Google Scholar] [CrossRef] - Mujeeb, S.; Javaid, N.; Ilahi, M.; Wadud, Z.; Ishmanov, F.; Afzal, M.K. Deep long short-term memory: A new price and load forecasting scheme for big data in smart cities. Sustainability
**2019**, 11, 987. [Google Scholar] [CrossRef] - Mujeeb, S.; Javaid, N. ESAENARX and DE-RELM: Novel schemes for big data predictive analytics of electricity load and price. Sustain. Cities Soc.
**2019**, 51, 101642. [Google Scholar] [CrossRef] - Coifman, R.R.; Wickerhauser, M.V. Entropy-based algorithms for best basis selection. IEEE Trans. Inf. Theory
**1992**, 38, 713–718. [Google Scholar] [CrossRef] - Burrus, C.S.; Gopinath, R.; Guo, H. Introduction to Wavelets and Wavelet Transforms: A Primer; Prentice Hall Press: Upper Saddle River, NJ, USA, 1997. [Google Scholar]
- Zahid, M.; Ahmed, F.; Javaid, N.; Abbasi, R.A.; Kazmi, Z.; Syeda, H.; Javaid, A.; Bilal, M.; Akbar, M.; Ilahi, M. Electricity price and load forecasting using enhanced convolutional neural network and enhanced support vector regression in smart grids. Electronics
**2019**, 8, 122. [Google Scholar] [CrossRef] - ISO NE Market Operations Data. Available online: https://www.iso-ne.com (accessed on 20 January 2019).
- Wang, Y.; Hu, Q.; Srinivasan, D.; Wang, Z. Wind power curve modeling and wind power forecasting with inconsistent data. IEEE Trans. Sustain. Energy
**2018**, 10, 16–25. [Google Scholar] [CrossRef] - Zhao, Y.; Ye, L.; Pinson, P.; Tang, Y.; Lu, P. Correlation-constrained and sparsity-controlled vector autoregressive model for spatio-temporal wind power forecasting. IEEE Trans. Power Syst.
**2018**, 33, 5029–5040. [Google Scholar] [CrossRef] - Martin, P.; Moreno, G.; Rodriguez, F.; Jimenez, J.; Fernandez, I. A Hybrid Approach to Short-Term Load Forecasting Aimed at Bad Data Detection in Secondary Substation Monitoring Equipment. Sensors
**2018**, 18, 3947. [Google Scholar] [CrossRef] - Derrac, J.; Garcia, S.; Molina, D.; Herrera, F. A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms. Swarm Evol. Comput.
**2011**, 1, 3–18. [Google Scholar] [CrossRef] - Diebold, F.X.; Mariano, R.S. Comparing predictive accuracy. J. Bus. Econ. Stat.
**1995**, 13, 253–263. [Google Scholar] - Lago, J.; De Ridder, F.; De Schutter, B. Forecasting spot electricity prices: Deep learning approaches and empirical comparison of traditional algorithms. Appl. Energy
**2018**, 221, 386–405. [Google Scholar] [CrossRef] - Chen, H.; Wan, Q.; Wang, Y. Refined Diebold-Mariano test methods for the evaluation of wind power forecasting models. Energies
**2014**, 7, 4185–4198. [Google Scholar] [CrossRef]

Inputs | Dataset | Algorithms |
---|---|---|

Past wind power | Delaware wind farm data, American National Renewable Energy Laboratory, 2006 | Nelder–Mead simplex optimization algorithm, Bidirectional backward Extreme learning machine [1] |

Wind power, IEEE 118-bus system parameters | Wind Integration National Dataset , National Renewable Energy Laboratory, CASIO, MISO, ISO NE, 2007–2013 | PLEXOS tool, Flexible Energy Scheduling Tool for Integrating Variable generation tool [9] |

Past hourly wind power | 66 wind power plants data, Supervisory Control And Data Acquisition (SCADA) | Vector autoregression model, Least absolute shrinkage and selection operator [24] |

Past wind power | Wind farm, Donegal, North West Ireland, June–July 2004 | Temporally local Gaussian process [25] |

10-min resolution: wind speed, wind power | Global Energy Forecasting Competition (GEFCom) 2014 | Multi-model combination method: Sparse Bayesian learning, Kernel density estimation and Beta distribution fitting method [26] |

5-minute resolution: wind speed, wind power | Wind power data, Australian Energy Market Operator (AEMO), 2005 | Spatial empirical decomposition, Random Forest, Gradient boosting, Support vector machine [27] |

Wind power, wind speed | Wind farm data, Ireland and USA, August 2006, October 2008 | Hybrid deterministic-probabilistic method with Gaussian process [28] |

10-min resolution: wind speed, wind power | National Renewable Energy Laboratory, 2005–2006 | Ensemble method: Wavelet transform, Partial least squares regression, ANN [29] |

Wind speed, wind power | GEFCom 2012 | ANN, Gaussian process [30] |

Past hourly wind power, past weather forecast: wind speed, wind direction, temperature and humidity | Wind power generation, Alberta, Canada | Improved Clonal selection algorithm, Wavelet neural networks, Maximum correntropy criterion [31] |

Wind turbine parameters, wind speed, wind power | 10-min wind farm data, SCADA | K-means clustering, Bagging ANN [32] |

Wind power, weather forecasts | 5 Wind farms data, Europe | Mutual information, Deep auto-encoders, Deep belief network [33] |

Wind speed, wind power | National Renewable Energy Laboratory (NREL), 2004 | Infinite feature selection method, RNN [34] |

Day of the year, hour, wind speed, wind direction, temperature, humidity, pressure, generators out of service | MADE wind farm, ITER, Tenerife Island, Spain, January 2014–April 2016 | Multi-layer perceptron with ReLU, Long short-term memory, Nonlinear autoregressive network with exogenous inputs [35] |

5-min intervals past wind power | SIWF wind farm, China, 2011–2013 | Wavelet transform, Ensemble CNN [36] |

Wind speed, wind direction, temperature, humidity, pressure | MADE wind farm, ITER, Tenerife Island, Spain | Feed Forward ANN, SELU CNN, RNN [37] |

Past consumption, solar radiation | Victorian solar dataset | Game theory model [38] * |

Historic price and load | Hourly load and price data, NYISO, PJM, AEMO, 2010, 2013, 2014 | Flexible wavelet packet transform, Nonlinear least square support vector machine, ARIMA, TV-ABC [39] * |

Historic consumption, wind power, photovoltaic power | Micro grid data, Renewable Energy Laboratory, UPNa, 2014 | Simple moving average, Central moving average [40] * |

Input | Description |
---|---|

Dew point temperature | Past NWP forecast |

Dry bulb temperature | Past NWP forecast |

Wind speed | Past NWP forecast |

Lagged wind power 1 | Wind power (t-24) |

Lagged wind power 2 | Wind power (t-25) |

Decomposed wind power | Wavelet decomposed past wind power |

Hour | Time of the day |

Method | Season | MAPE | NRMSE | MAE |
---|---|---|---|---|

Spring | 8.42 | 2.34 | 3.34 | |

Summer | 8.23 | 2.27 | 3.24 | |

CNN | Autumn | 7.9 | 2.65 | 3.36 |

Winter | 8.1 | 2.71 | 2.89 | |

Spring | 3.47 | 0.12 | 3.1 | |

Summer | 3.62 | 0.13 | 3.3 | |

SELU CNN | Autumn | 3.45 | 0.12 | 3.4 |

Winter | 3.27 | 0.17 | 3.2 | |

Spring | 2.67 | 0.092 | 2.4 | |

Summer | 2.43 | 0.096 | 2.24 | |

EDCNN | Autumn | 2.56 | 0.085 | 2.67 |

Winter | 2.62 | 0.094 | 2.18 |

**Table 4.**Diebold–Mariano test results at a 95% confidence level and 5% significance level of p-value.

DM Score | |||
---|---|---|---|

Season | EDCNN Compared to SELU CNN | SELU CNN Compared to CNN | EDCNN Compared to CNN |

Spring DM-MAE | 1.4252 | 0.0842 | 1.4256 |

Spring p-value | 0.0432 | 0.9242 | 0.1248 |

Summer DM-MAE | 1.3262 | 0.1024 | 1.3692 |

Summer p-valve | 0.0326 | 0.8624 | 0.2142 |

Autumn DM-MAE | 1.2714 | 0.1762 | 1.6728 |

Autumn p-vale | 0.0196 | 0.0242 | 0.9242 |

Winter DM-MAE | 1.4632 | 1.1426 | 1.2464 |

Winter p-value | 0.02762 | 0.9862 | 0.7642 |

Consumption Cost / Day ($) | Reduction / Day | |||
---|---|---|---|---|

Season | Before DSM | After DSM | Amount ($) | Percentage |

Spring | 483,330 | 475,170 | 8153$ | 1.7% |

Summer | 793,930.5 | 784,403 | 7527$ | 1.2% |

Autumn | 417,980.5 | 413,770.5 | 4210$ | 1% |

Winter | 3,347,106 | 3,305,006 | 42,109$ | 1.3% |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Mujeeb, S.; Alghamdi, T.A.; Ullah, S.; Fatima, A.; Javaid, N.; Saba, T.
Exploiting Deep Learning for Wind Power Forecasting Based on Big Data Analytics. *Appl. Sci.* **2019**, *9*, 4417.
https://doi.org/10.3390/app9204417

**AMA Style**

Mujeeb S, Alghamdi TA, Ullah S, Fatima A, Javaid N, Saba T.
Exploiting Deep Learning for Wind Power Forecasting Based on Big Data Analytics. *Applied Sciences*. 2019; 9(20):4417.
https://doi.org/10.3390/app9204417

**Chicago/Turabian Style**

Mujeeb, Sana, Turki Ali Alghamdi, Sameeh Ullah, Aisha Fatima, Nadeem Javaid, and Tanzila Saba.
2019. "Exploiting Deep Learning for Wind Power Forecasting Based on Big Data Analytics" *Applied Sciences* 9, no. 20: 4417.
https://doi.org/10.3390/app9204417