# Forecasting Electricity Demand in Thailand with an Artificial Neural Network Approach

## Abstract

**:**

## 1. Introduction

## 2. Historical Data

Year | Population | GDP | SET Index | Export (million baht) | Electricity Consumption (GWh) |
---|---|---|---|---|---|

1986 | 52511000 | 1257177 | 207.2 | 364017.25 | 10162.7 |

1987 | 53427000 | 1376847 | 284.94 | 455991.43 | 11319.4 |

1988 | 54326000 | 1559804 | 386.73 | 462426.83 | 11942.38 |

1989 | 55214000 | 1749952 | 879.19 | 562426.76 | 14328.1 |

1990 | 55839000 | 1945372 | 612.86 | 683946.13 | 16717.23 |

1991 | 56574000 | 2111862 | 711.36 | 725448.79 | 19406.02 |

1992 | 57294000 | 2282572 | 893.42 | 824643.29 | 21641.01 |

1993 | 58010000 | 2470908 | 1682.85 | 940862.59 | 24321.28 |

1994 | 58713000 | 2692973 | 1360.09 | 1137601.65 | 27758.43 |

1995 | 59401000 | 2941736 | 1280.81 | 1153489 | 31870.37 |

1996 | 60003000 | 3115338 | 831.57 | 1153894.61 | 34607.29 |

1997 | 60602000 | 3072615 | 372.69 | 1492331.29 | 36981.24 |

1998 | 61201000 | 2749684 | 355.81 | 1854500.09 | 35154.99 |

1999 | 61806000 | 2871980 | 481.92 | 1871544.78 | 36275.13 |

2000 | 62236000 | 3008401 | 269.19 | 2378191.26 | 39546.26 |

2001 | 62836000 | 3073601 | 303.85 | 2454987.54 | 41658.51 |

2002 | 63419000 | 3237042 | 356.48 | 2506442.96 | 44805.66 |

2003 | 63982000 | 3468166 | 772.15 | 2857191.85 | 48293.79 |

2004 | 64531000 | 3688189 | 668.1 | 3361360.69 | 50810.54 |

2005 | 65099000 | 3858019 | 713.73 | 3897247.1 | 53894.12 |

2006 | 65574000 | 4054504 | 679.84 | 4305406.71 | 56994.75 |

2007 | 66041000 | 4259026 | 858.1 | 4691207.01 | 59436.12 |

2008 | 66482000 | 4364833 | 449.96 | 5149902.76 | 60266.29 |

2009 | 66903000 | 4263139 | 734.54 | 4619810.05 | 59401.92 |

2010 | 67209942.8 | 4595809 | 1032.76 | 5476766.65 | 60315.04 |

## 3. Data Analysis

#### 3.1. ARIMA Model

Model | MAPE |
---|---|

ARIMA (0,2,2) | 2.80981 |

ARIMA (1,2,1) | 3.02891 |

ARIMA (1,1,0) | 3.34578 |

ARIMA (0,2,0) | 3.30197 |

_{1}in the ARIMA (p, 0, q) model: were calculated by a Box Jenkins method and AIC criterion where . Afterwards, let , and , the parameter of ARIMA (0,d,0) model: was estimated by applying the log Gaussian likelihood function as: where R = Covariance matrix of . The ARIMA (0,2,2) model coefficients are given in Table 3.

Parameter | Estimate |
---|---|

MA(1) | 0.434155 |

MA(2) | 0.488944 |

#### 3.2. Artificial Neural Network

_{j}is the output of node j, f (.) is the transfer function, w

_{ij}the connection weight between node j and node i in the lower layer and X

_{ij}is the input signal from the node i in the lower layer to node j.

Model | MAPE |
---|---|

MLP (4,10,1) | 2.770 |

RBF (4,6,1) | 3.033 |

MLP (4,8,1) | 2.598 |

MLP (4,6,1) | 0.996 |

MLP (4,5,1) | 3.2938 |

#### 3.3. Multiple Linear Regression

_{i}is the dependent variable, x

_{.i}is the independent variable, β

_{i}is the regression coefficient of x

_{.i}and ε

_{i}is the random error. In order to construct the regression model, the independent variables (x

_{.i}) were population, SET index, GDP and Export, while the dependent variable (y

_{i}) was GWh. In order to estimate the coefficients of the model, the predicted response was shown in Equation (4):

_{i}and let it equal to zero. This yielded Equation (7):

_{1}, y

_{1}), … , (x

_{n}, y

_{n}). The coefficients b

_{0}, b

_{1}, b

_{2}, …, b

_{k}were obtained by solving Equation (7). As a result, the regression equation was computed as follows:

## 4. Results

Model | MAPE |
---|---|

ARIMA (0,2,2) | 2.80981 |

MLP (4,6,1) | 0.996 |

MLR | 3.2604527 |

Pairs of Methods | p-value |
---|---|

ANN-MLR | 0.819095 |

ANN-ARIMA | 0.784289 |

Pairs of Methods | p-value |
---|---|

ANN-MLR | 0.785697 |

ANN-ARIMA | 0.927594 |

## 5. Discussion

## 6. Conclusions

## References

- Box, G.E.P.; Jenkins, G.M. Time Series Analysis: Forecasting and Control; Holden-Day: San Francisco, CA, USA, 1970; pp. 5–10. [Google Scholar]
- Montgomery, D.C.; Keats, J.B.; Runger, G.C.; Messina, W.S. Integrating statistical process control and engineering process control. J. Qual. Technol.
**1994**, 26, 79–87. [Google Scholar] - Box, G.E.P.; Luceno, A. Statistical Control by Monitoring and Feedback Adjustment; John Wiley & Sons: New York, NY, USA, 1997. [Google Scholar]
- Ediger, V.S.; Akar, S. ARIMA forecasting of primary energy demand by fuel in Turkey. Energy Policy
**2007**, 35, 1701–1708. [Google Scholar] [CrossRef] - Abdel-Aal, R.E.; Al-Garni, A.Z. Forecasting monthly electric energy consumption in eastern saudi arabia using univariate time-series analysis. Energy
**1997**, 22, 1059–1069. [Google Scholar] [CrossRef] - Zhou, P.; Ang, B.W.; Poh, K.L. A trigonometry grey prediction approach to forecasting electricity demand. Energy
**2006**, 31, 2839–2847. [Google Scholar] [CrossRef] - Cho, M.Y.; Hwang, J.C.; Chen, C.S. Customer Short Term Load Forecasting by Using ARIMA Transfer Function Model. In Proceedings of the International Conference on Energy Management and Power Delivery, EMPD’95, Singapore, 20–23 November 1995; pp. 317–322.
- Hsu, C.-C.; Chen, C.-Y. Regional load forecasting in Taiwan-applications of artificial neural networks. Energy Convers. Manag.
**2003**, 44, 1941–1949. [Google Scholar] [CrossRef] - Catalao, J.P.S.; Mariano, S.J.P.S.; Mendes, V.M.F.; Ferreira, L.A.F.M. Short-term electricity prices forecasting in a competitive market: A neural network approach. Electr. Power Syst. Res.
**2007**, 77, 1297–1304. [Google Scholar] [CrossRef] - Bakirtzis, A.G.; Petridis, V.; Klartzis, S.J.; Alexladis, M.C. A neural network short term load forecasting model for the greek power system. IEEE Trans. Power Syst.
**1996**, 11, 858–863. [Google Scholar] [CrossRef] - Mohamed, Z.; Bodger, P. Forecasting electricity consumption in New Zealand using economic and demographic variables. Energy
**2005**, 30, 1833–1843. [Google Scholar] [CrossRef] - Azadeh, A.; Ghaderi, S.F.; Tarverdian, S.; Saberi, M. Intergraion of artificial neural networks and genetic algorithm to predict electrical energy consumption. Appl. Math. Comput.
**2007**, 186, 1731–1741. [Google Scholar] [CrossRef] - Azadeh, A.; Ghaderi, S.F.; Sohrabkhani, S. A simulated-based neural network algorithm for forecasting electrical energy consumption in Iran. Energy Policy
**2008**, 36, 2637–2644. [Google Scholar] [CrossRef] - Hong, W.-C. Electric load forecasting by support vector model. Appl. Math. Model.
**2009**, 33, 2444–2454. [Google Scholar] [CrossRef] - Ekonomou, L. Greek long-term energy consumption prediction using artificial neural networks. Energy
**2010**, 35, 512–517. [Google Scholar] [CrossRef] - Pappas, S.S.; Ekonomou, L.; Karamousantas, D.C.; Chatzarakis, G.E.; Katsikas, S.K.; Liatsis, P. Electricity demand loads modeling using autoregressive moving average (ARIMA) models. Energy
**2008**, 33, 1353–1360. [Google Scholar] [CrossRef]

© 2011 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 license (http://creativecommons.org/licenses/by/3.0/).

## Share and Cite

**MDPI and ACS Style**

Kandananond, K.
Forecasting Electricity Demand in Thailand with an Artificial Neural Network Approach. *Energies* **2011**, *4*, 1246-1257.
https://doi.org/10.3390/en4081246

**AMA Style**

Kandananond K.
Forecasting Electricity Demand in Thailand with an Artificial Neural Network Approach. *Energies*. 2011; 4(8):1246-1257.
https://doi.org/10.3390/en4081246

**Chicago/Turabian Style**

Kandananond, Karin.
2011. "Forecasting Electricity Demand in Thailand with an Artificial Neural Network Approach" *Energies* 4, no. 8: 1246-1257.
https://doi.org/10.3390/en4081246