Abstract
Many day-to-day operation decisions in a smart city need short term load forecasting (STLF) of its customers. STLF is a challenging task because the forecasting accuracy is affected by external factors whose relationships are usually complex and nonlinear. In this paper, a novel hybrid forecasting algorithm is proposed. The proposed hybrid forecasting method is based on locally weighted support vector regression (LWSVR) and the modified grasshopper optimization algorithm (MGOA). Obtaining the appropriate values of LWSVR parameters is vital to achieving satisfactory forecasting accuracy. Therefore, the MGOA is proposed in this paper to optimally select the LWSVR’s parameters. The proposed MGOA can be derived by presenting two modifications on the conventional GOA in which the chaotic initialization and the sigmoid decreasing criterion are employed to treat the drawbacks of the conventional GOA. Then the hybrid LWSVR-MGOA method is used to solve the STLF problem. The performance of the proposed LWSVR-MGOA method is assessed using six different real-world datasets. The results reveal that the proposed forecasting method gives a much better forecasting performance in comparison with some published forecasting methods in all cases.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig8_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig9_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig10_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig11_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig12_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig13_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig14_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig15_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig16_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig17_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig18_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10489-020-01728-x/MediaObjects/10489_2020_1728_Fig19_HTML.png)
Similar content being viewed by others
Abbreviations
- \(\left \{x_{i}, y_{i}\right \}_{i=1}^{N_{t}}\) :
-
time series dataset
- N t :
-
number of training points
- xi ∈ Rn :
-
input space of the dataset, i = 1,...,Nt
- yi ∈ R :
-
target value
- Q(xi,x) = φ(xi) ⋅ φ(x):
-
Gaussian kernel function
- φ(x):
-
high dimensional feature space
- w :
-
vector contains the weight coefficients
- b :
-
a real bias constant
- \(\zeta _{i}, \zeta _{i}^{*}\) :
-
upper and lower training errors, respectively
- C :
-
regularization constant that determines the trade-off between upper and lower training errors and the flatness of function f
- \(\eta _{i}, \eta _{i}^{*}\) :
-
Lagrange parameters
- x q :
-
the query point
- K(1 < K ≪ Nt):
-
number of nearest neighbors
- \(k_{\max \limits }\) :
-
the maximum number of nearest neighbors
- Dk(xi):
-
distance between each training point x and its nearest neighbors
- \(D_{\max \limits }\) :
-
the maximum distance
- ν :
-
a constant number
- W i :
-
the weighting function
- h :
-
bandwidth parameter which performs an essential role in local modelling
- C i :
-
weighted regularization constant
- X i :
-
position of the ith grasshopper
- S i :
-
the social interaction
- G i :
-
the gravity force
- A i :
-
the wind advection
- n g :
-
number of population
- m :
-
dimension of the population position variables
- U, L :
-
upper and lower bound of Xi, respectively
- d :
-
distance between two individuals
- T :
-
the individual who has the best fitness value
- c1, c2 :
-
the decreasing coefficient
- it :
-
number of current iteration
- \(it_{\max \limits }\) :
-
the maximum number of iterations
- \(c_{\max \limits }, c_{\min \limits }\) :
-
the maximum and minimum value of the parameter c
- u :
-
a constant that fits the sigmoid function sharpness
- n :
-
a constant that adjusts partition of the sigmoid function
- \(A_{i}, F_{i}, \overline {A}\) :
-
actual value, forecasted value and mean of the actual values, respectively
- N :
-
size of testing dataset
References
Mujeeb S, Javaid N, Ilahi M, Wadud Z, Ishmanov F, Afzal M (2019) Deep long short-term memory: a new price and load forecasting scheme for big data in smart cities. Sustainability 11(4):1–29
Ahmad A, Javaid N, Mateen A, Awais M, Khan Z (2019) Short-term load forecasting in smart grids: an intelligent modular approach. Energies 12(164):1–21
Elattar E, Goulermas J, Wu QH (2010) Electric load forecasting based on locally weighted support vector regression. IEEE Transactions on Systems, Man, and Cybernetics-Part C: Applications and Reviews 40(4):438–447
Papalekopulos AD, Hesterberg TC (1990) A regression-based approach to short-term system load forecasting. IEEE Transactions on Power Systems 5(4):1535–1547
Taylor JW, de Menezes LM, McSharry PE (2006) A comparison of univariate methods for forecasting electricity demand up to a day ahead. International Journal of Forecasting 22(1):1–16
Hippert HS, Pedreira CE, Souza RC (2001) Neural networks for shortterm load forecasting: a review and evaluation. IEEE Transactions on Power Systems 16(1):44–55
Khan A, Javaid N, Nazeer O, Zahid M, Akbar M, Khan MH (2019) Hourly electricity load forecasting in smart grid using deep learning techniques. In: 13th innovative mobile and internet services in ubiquitous computing. Springer, Cham, pp 185–196
Coelho V, et al. (2016) A self-adaptive evolutionary fuzzy model for load forecasting problems on smart grid environment. Applied Energy 169:567–584
Smola AJ, Scholkopf B (1998) A tutorial on support vector regression. NeuroCOLT Technical Report NC-TR-98-030, Royal Holloway College, University of London
Vrablecova P, Ezzeddine A, Rozinajova V, Sarik S, Sangaiah A (2018) Smart grid load forecasting using online support vector regression. Computers and Electrical Engineering 65:102–117
Ayub N, Javaid N, Mujeeb S, Zahid M, Khan W, Khattak M (2019) Electricity load forecasting in smart grids using support vector machine. In: International conference on advanced information networking and applications. Springer, Cham, pp 1–13
Twanabasu S, Bremdal B (2013) Load forecasting in a smart grid oriented building. In: 22nd international conference and exhibition on electricity distribution (CIRED 2013). Sweden, Stockholm, pp 1–4
Lau KW, Wu QH (2008) Local prediction of non-linear time series using support vector regression. Pattern Recognition 41:1539–1547
Elattar EE, Goulermas JY, Wu QH (2009) Forecasting electric daily peak load based on local prediction. In: IEEE power and energy soc. General Meeting (PESGM09). Canada, pp 1–6
Elattar EE, Goulermas JY, Wu QH (2010) Integrating KPCA and locally weighted support vector regression for short-term load forecasting. In: The 15th IEEE miditerranean electrotechnical conf. (MELECOn 2010). Valletta, Malta, pp 1528–1533
Banjoko A, Yahya W, Garba M, Abdulazeez K (2019) Weighted support vector machine algorithm for efficient classification and prediction of binary response data. Journal of Physics: Conference Series 1366(1):1–7
Zhang J, Teng Y, Chen W (2019) Support vector regression with modified firefly algorithm for stock price forecasting. Applied Intelligence 49(5):1658–1674
Barkalov K, Polovinkin A, Meyerov I, Sidorov S, Zolotykh NSVM (2019) Regression parameters optimization using parallel global search algorithm. In: International conference on parallel computing technologies. Almaty, Kazakhstan, pp 154–166
Wang Y et al (2019) A method based on improved ant lion optimization and support vector regression for remaining useful life estimation of lithium-ion batteries. Energy Science and Engineering 7:2797–2813
Saremi S, Mirjalili S, Lewis A (2017) Grasshopper optimisation algorithm: theory and application. Advances in Engineering Software 105:30–47
Elsayed A, Mishref M, Farrag S (2019) Optimal allocation and control of fixed and switched capacitor banks on distribution systems using grasshopper optimisation algorithm with power loss sensitivity and rough set theory. IET Generation, Transmission and Distribution 13(17):3863–3878
Liang H, Jia H, Xing Z, Ma J, Peng X (2019) Modified grasshopper algorithm-based multilevel thresholding for color image segmentation. IEEE Access 7:11258–11295
Li G, Wang N, Liu X (2019) Enhanced grasshopper optimization algorithm for modeling multivariable systems with SVR. In: Proceedings of the 38th Chinese control conference. Guangzhou, China, pp 1569–1574
Arora S, Anand P (2019) Chaotic grasshopper optimization algorithm for global optimization. Neural Computing and Applications 31:4385–4405
Saxena A, Shekhawat S, Kumar R (2018) Application and development of enhanced chaotic grasshopper optimization algorithms. Modelling and Simulation in Engineering 2018:1–14
Wolpert D, Macready W (1997) No free lunch theorems for optimization. IEEE Trans on Evolutionary Computaion 1:67–82
Suresh S, Lal S, Reddy C, Kiran M (2017) A novel adaptive cuckoo search algorithm for contrast enhancement of satellite images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 10(8):3665–3676
Mirjalili S, et al. (2017) Salp swarm algorithm: a bio-inspired optimizer for engineering design problems. Advances in Engineering Software 114:193–191
Wang H, Fan C, Tu X (2016) AFSAOCP: a novel artificial fish swarm optimization algorithm aided by ocean current power. Applied Intelligence 45:1–16
New York Independent System Operator (NYISO) (Last access: October 2019). Available from: http://www.nyiso.com
Elattar E, Goulermas J, Wu QH (2012) Generalized Locally Weighted, GMDH for short term load forecasting. IEEE Transactions on Systems, Man, and Cybernetics-Part C: Applications and Reviews 42(3):345–356
Mandal P, Senjyu T, Funabashi T (2006) Neural networks approach to forecast several hour ahead electricity prices and loads in deregulated market. Energy Conversions and Management 47:2128–2142
Ferreira VH, da Silva APA (2007) Toward estimation autonomous neural network-based electric load forecasters. IEEE Transactions on Power Systems 22(4):1554–1562
Australian Energy Market Operator (AEMO) Website (Last access: October 2019). Available from: http://www.aemo.com.au/
PJM electricity market Website (Last access: October 2019). Available from: https://www.pjm.com
Liu N, Tang Q, Zhang J, WF W, Liu J (2014) A hybrid forecasting model with parameter optimization for short-term load forecasting of micro-grids. Applied Energy 129:336–345
GEFCom 2012 Dataset (Last access: January 2020);. Available from: http://blog.drhongtao.com/2016/07/gefcom2012-load-forecasting-data.html
ISO New England Data (Last access: January 2020). Available from: http://www.iso-ne.com/isoexpress/web/reports/pricing/-/tree/zone-info
Amjaday N, Keynia F (2009) Day-ahead price forecasting of electricity markets by mutual information technique and cascaded neuro-evolutionary algorithm. IEEE Transactions on Power Systems 24:306–318
Amjaday N, Keynia F, Zareipour H (2014) Short-term load forecast of microgrids by a new bilevel prediction strategy. IEEE Transactions on Smart Grid 1:286–294
Kong X, Li C, Wang C, Zhang Y, Zhang J (2020) Short-term electrical load forecasting based on error correction using dynamic mode decomposition. Applied Energy 261:1–13
Zhou M, Jin M (2019) Holographic ensemble forecasting method for short-term power load. IEEE Transactions on Smart Grid 10(1):425–434
Chen F, Han C (2007) Time series forecasting based on wavelet KPCA and support vector machine. In: IEEE international conference on automation and logistics. Jinan, China, pp 1487–1491
Vapink VN (1998) Statistical learning theory. Wiley, New York
Acknowledgements
This study was funded by the Deanship of Scientific Research, Taif University, KSA [Research project number 6147-440-1].
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A: Phase space reconstruction based on KPCA
Due to the complication of the historical datasets, the use of time series reconstruction technique has become necessary in STLF. So, the KPCA method is utilized in this paper to reconstruct the non-linear time series used in STLF which leads to overcoming the drawbacks of the conventional reconstruction methods. KPCA is an unsupervised method that executes the principal component analysis in the feature space of a kernel which is non-linearly related to the input space [43]. The merit of KPCA over other non-linear methods comes from making the time series structure more linear by mapping the original inputs into a high dimensional feature space via a kernel map.
In this paper, the generally used Gaussian kernel is utilized and can be defined as follows:
The details of the basic KPCA can be found in [15, 43].
Appendix B: Support vector regression (SVR)
The quadratic programming and kernel functions are the main elements used to implement the SVR algorithm. Where the SVR’s parameters can be calculated by solving the quadratic programming problem with different constraints. Also, the flexibility of kernel functions helps the algorithm to enhance its searching ability [9].
The main idea of SVR is to transfer the input data x into a high dimensional feature space through a nonlinear mapping and execute a linear regression in that feature space [9] as:
To get the weight coefficients and the bias, the following optimization problem is solved [44]:
where xi is transferred to higher dimensional space via φ. \(\zeta _{i}, \zeta _{i}^{*}\) depend on the ε-insensitive tube |y − (wTφ(x) + b)|≤ ε.
Based on the Lagrangian and the Karush-Kuhn-Tucker conditions, the solution of the model can be found in dual representation instead of high dimension as follows [9]:
By solving this dual representation problem, the regression output can be written by [9]:
The most commonly employed kernel function is the Gaussian kernel defined in (A.1).
Rights and permissions
About this article
Cite this article
Elattar, E.E., Sabiha, N.A., Alsharef, M. et al. Short term electric load forecasting using hybrid algorithm for smart cities. Appl Intell 50, 3379–3399 (2020). https://doi.org/10.1007/s10489-020-01728-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-020-01728-x