Seite - 6 - in Short-Term Load Forecasting by Artificial Intelligent Technologies

Energies2018,11, 2226 2.2. ChaoticQuantumFruitFlyAlgorithm(CQFOA) FOAisapopulation intelligentevolutionaryalgorithmthatsimulates the foragingbehaviorof fruitflies [26]. Fruitfliesaresuperior tootherspecies insmellandvision. In theprocessof foraging, firstly, fruit flies rely on smell to find the food source. Secondly, they visually locate the specific location of food and the current position of other fruit flies, and then fly to the location of food throughpopulationinteraction.Atpresent,FOAhasbeenappliedtotheforecastingof trafficaccidents, export trade,andotherfields [40]. 2.2.1. FruitFlyOptimizationAlgorithm(FOA) According to the characteristics of fruit flies searching for food, FOA includes the following mainsteps. Step1. Initializerandomlythe fruitflies’ location(X0 andY0)ofpopulation. Step2.Give individual fruitflies therandomdirectionanddistance forsearchingfor foodbysmell, as inEquations (8)and(9) [26]: Xi=X0+RandomValue (8) Yi=Y0+RandomValue. (9) Step 3. Due to the location of food being unknown, firstly, the distance from the origin (Dist) is estimatedas inEquation(10) [25], thenthedeterminationvalueof tasteconcentration(S) is calculated as inEquation(11) [25], i.e., thevalue is the inverseof thedistance. Disti= √ X2i +Y 2 i (10) Si=1/Disti (11) Step4.Thedeterminationvalueoftasteconcentration(S) issubstitutedintothedeterminationfunction of tasteconcentration(orFitness function) todetermine the individualpositionof the fruitfly(Smelli), as showninEquation(12) [26]: Smelli=Function(Si). (12) Step 5. Find the Drosophila species (Best index and Best Smell values) with the highest odor concentrations in thispopulation,as inEquation(13) [26]: max(Smelli)→ (Best_Smelli)and (Best_index). (13) Step6. Theoptimalflavor concentrationvalue (Optimal_Smell) is retainedalongwith the x and y coordinates (withBest_index) as inEquations (14)–(16) [25], thentheDrosophilapopulationusesvision toflyto thisposition. Optimal_Smell=Best_Smelli=current (14) X0=XBest_index (15) Y0=YBest_index (16) Step7.Enter the iterativeoptimization, repeatSteps2 to5andjudgewhether theflavorconcentration isbetter thanthatof theprevious iteration; if so,gobacktoStep6. TheFOAalgorithmishighlyadaptable, so it canefficiently searchwithout calculatingpartial derivativesof the target function. Itovercomes thedisadvantageof trapping into localoptimaeasily. However,asaswarmintelligenceoptimizationalgorithm,FOAstill tends to fall intoa localoptimal solution,dueto thedecliningdiversity in the lateevolutionarypopulation. 6