Short-term electric load forecasting based on singular spectrum analysis and support vector machine optimized by Cuckoo search algorithm

2017 ◽  
Vol 146 ◽  
pp. 270-285 ◽  
Author(s):  
Xiaobo Zhang ◽  
Jianzhou Wang ◽  
Kequan Zhang
Keyword(s):  
Support Vector Machine ◽  
Spectrum Analysis ◽  
Search Algorithm ◽  
Singular Spectrum Analysis ◽  
Load Forecasting ◽  
Cuckoo Search ◽  
Cuckoo Search Algorithm ◽  
Support Vector ◽  
Electric Load ◽  
Short Term

scholarly journals Intelligent Fuzzy Logic - Cuckoo Search Algorithm Method for Short-Term Electric Load Forecasting in 150 kV Sulselrabar System

2017 ◽  
pp. 154 ◽  
Author(s):  
Muhammad Ruswandi Djalal ◽  
Faisal Faisal
Keyword(s):  
Fuzzy Logic ◽  
Search Algorithm ◽  
Load Forecasting ◽  
Cuckoo Search ◽  
Cuckoo Search Algorithm ◽  
Load Flow ◽  
Time Range ◽  
Electric Load ◽  
Short Term ◽  
Electric Load Forecasting

Forecasting the electrical load becomes important, because it can estimate electricity consumption over a certain time range. Accuracy in electric load forecasting can improve safety and reliability in the operation of power systems such as load flow, maintenance of generating units and scheduling of generating units. In this study used case study system Sulselrabar, which is currently growing, but still not much to discuss about the condition of the current system and which will come. Several methods for predicting electrical loads have been widely used, ranging from conventional to smart-based methods. In this research will be proposed method of artificial intelligence for forecasting Short Term load on Sulselrabar system. The method used is based Fuzzy Logic and Cuckoo Search Algorithm. The combination of Fuzzy logic and Cuckoo Search methods is chosen because the combination of both optimizes optimum fuzzy logic membership, so the forecasting results have a very small error. From the results of the research can be concluded that the result of load forecasting using Fuzzy Logic method optimized using Cuckoo Search Algorithm (FL-CSA) is better than Fuzzy Logic that is not optimized. The analysis results using input data 3 months before day H, to predict the load for one week on January 1 to 7 january 2014, and as a comparison used the predicted day H data. From the simulation results, the mean absolute percentage error (MAPE) is smaller using FLCSA, for the smallest MAPE on 1 January 2014 of 0.06785208%. While the highest MAPE on January 4, 2014 amounted to -0.44973%.

scholarly journals Short-Term Load Forecasting Based on Wavelet Transform and Least Squares Support Vector Machine Optimized by Improved Cuckoo Search

Energies ◽  
2016 ◽  
Vol 9 (10) ◽  
pp. 827 ◽  
Author(s):  
Yi Liang ◽  
Dongxiao Niu ◽  
Minquan Ye ◽  
Wei-Chiang Hong
Keyword(s):  
Support Vector Machine ◽  
Wavelet Transform ◽  
Least Squares ◽  
Load Forecasting ◽  
Cuckoo Search ◽  
Support Vector ◽  
Short Term ◽  
Short Term Load Forecasting

scholarly journals Stability Evaluation of Rock Slope in Hydraulic Engineering Based on Improved Support Vector Machine Algorithm

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-13
Author(s):  
Fei Li ◽  
Hongyun Zhang
Keyword(s):  
Support Vector Machine ◽  
Stability Analysis ◽  
Permeability Coefficient ◽  
Engineering Geology ◽  
Search Algorithm ◽  
Cuckoo Search ◽  
Cuckoo Search Algorithm ◽  
Support Vector ◽  
Svm Algorithm ◽  
Improved Support Vector Machine

The safety problem of the slope has always been an important subject in engineering geology, which has a wide range of application background and practical significance in reality. How to correctly evaluate the stability of the slope and obtain the parameters of the slope has always been the focus of research and production personnel at home and abroad. In recent years, various artificial intelligence calculation methods have been applied to the field of rock engineering and engineering geology, providing some new ideas for the solution of slope stability analysis and parameter back analysis. Support vector machine (SVM) algorithm has unique advantages and generalization in dealing with finite samples and highly complex and nonlinear problems. At present, it has become a research hotspot of intelligent methods and has been widely paid attention to in various application fields of slope engineering. In this paper, a cuckoo search algorithm-improved support vector machine (CS-SVM) method is applied to slope stability analysis and parameter inversion. Aiming at the problem of selecting kernel function parameters and penalty number of SVM, a method of using cuckoo search algorithm to improve support vector machine was proposed, and the global optimization ability of cuckoo search algorithm was used to improve the algorithm. Aiming at the slope samples collected, the classification algorithm of support vector machine (SVM) was used to identify the stable state of the test samples, and the improved SVM algorithm was used to analyze the safety factor of the test samples. The results show that the proposed method is reasonable and reliable. Based on the inversion of the permeability coefficient of the test samples by the improved support vector machine, the comparison between the inversion value and the theoretical value shows that it is basically feasible to invert the permeability coefficient of the dam slope by the improved support vector machine.

Sign in / Sign up

Export Citation Format

Share Document