A Meta-Heuristics Based Input Variable Selection Technique for Hybrid Electrical Energy Demand Prediction Models
|
|
- Pamela Mosley
- 5 years ago
- Views:
Transcription
1 A Mea-Heurisics Based Inpu Variable Selecion Technique for Hybrid Elecrical Energy Demand Predicion Models Badar ul Islam Perumal allagownden Zuhairi Baharudin Masood Rehman Absrac Elecrical energy demand forecasing plays a pivoal role as a decision suppor ool in he modern power indusry. The focus of he paper is o propose a hybrid approach for he selecion of he mos influenial inpu variables for he raining and esing of neural nework based hybrid models. The combined influence of he geneic algorihm and correlaion analysis are used in his echnique. The significance of he seleced inpu variable vecors is sudied o analyze heir effecs on he predicion process. Anoher objecive of he sudy is o develop and compare he predicion models for elecrical energy demand of one day-ahead. These models are developed by inegraing mulilayer percepron neural nework and evoluionary opimizaion echniques. Geneic algorihm and simulaed annealing echniques are used o opimize he conrol parameers of he neural nework. The resuls show ha he neural nework opimized wih geneic algorihm and rained wih an opimally and inelligenly seleced inpu vecor conaining hisorical load and meeorological variables produced he bes predicion accuracy. Keywords - arificial neural nework; mean absolue percenage error; geneic algorihm; simulaed annealing; correlaion analysis I. ITRODUCTIO Elecrical energy demand forecasing is indispensible for he cos effecive, secure and reliable operaion of he power sysems [1, ]. Shor erm load forecas (STLF) of one dayahead faciliaes muliple power sysem operaions, for insance, scheduling of fuel purchase, mainenance of equipmen, adjusmen of ariff and conrac assessmens []. Arificial neural nework (A) is an arificial inelligence based echnique, which provides much beer performance as compared o previously implemened mehod for STLF [3]. This is because of he powerful capabiliy of A o map and memorize he non-linear relaions beween he inpu and he oupu variables during he raining process. The performance of A based models depends upon many facors, such as, neural nework archiecure, ype of raining algorihm, selecion of inpu variables and iniial values of economic variables using hi and rial mehods for heir selecion [8-10]. However some researchers implemened compuaional inelligence in he daa selecion process [11, 1]. In his paper, he imporan issues relaed o he bes inpu variable selecion for A based STLF models are synapic weighs [4]. Among he ohers, he selecion of mos influenial inpu variables and choice of raining echnique have a criical impac on he forecas resuls. Many evoluionary search echniques have been implemened o opimize he raining inpu vecors and conrol parameer of he neural neworks for STLF applicaion. Mos of hese effors emphasize on he correlaion analysis of he daa, however some of he researchers also focused on he combinaion of mahemaical formulaion o resor he derived variables by squaring, averaging, adding or differencing he daa sequences o deermine he suiable inpu variables [5, 6]. Because of heir linear and inconsisen approach, hese mehods are unable o rack he unusual and brisk variaions occurring in he real ime inpu daa, [7]. Many inpu variables have been used for he A based STLF, such as a hisorical load, meeorological variables and addressed. A new echnique is proposed ha inegraes he geneic algorihm (GA) and correlaion coefficien mehods o esablish he supremacy of he cerain inpu variables over he ohers. The opimally seleced inpu variable vecors (IVs) are used o rain he A based hybrid models using GA and simulaed annealing (SA). The analysis of he resuls DOI /IJSSST.a ISS: x online, prin
2 show ha hybrid model based on A and GA rained wih he opimally seleced IV which conains load and merological variables produced beer forecas accuracy as compared o oher IVs. II. DATA AALYSIS AD PRE-PROCESSIG The hisorical daa of he Sae of Vicoria and ew Souh Wales of Ausralia are used in he experimenaion. A comprehensive correlaion analysis is conduced o sudy he maximum relevance of he inpu variables wih load demand [7]. The resuls of he analysis are shown in Table I. All he inpu variables are normalized/scaled beween 0 and 1, before applying hem o he models [13]. TABLE I. CORRELATIO AALYSIS RESULTS Inpu Variables Correlaion Same ime in he previous day load L(h-4) Same day and ime in previous week load L(h-168) Same ime, wo days earlier load L(h-48) 0.80 Same day, wo hours earlier load L(H-) Same ime, previous day emperaure T(h-4) Same ime, previous week emperaure T(h-168) III. RESEARCH METHODOLOGY The framework of he research aciviies are menioned in he following seps. Apply a hybrid scheme based on he GA and correlaion mehod for he selecion of mos appropriae inpu variables for he raining of simulaion models. The hybrid muli-layer percepron (MLP) models are developed and validaed, where he iniial parameers are opimized by GA and SA for one day ahead elecrical energy demand predicion using he seleced inpu variables. Compare he forecass and acual energy demand of he opimized models wih and wihou using opimally seleced IVs. A. Selecion of forecas model inpus The firs sep in he developmen of he GA is o define a chromosome. A fixed lengh chromosome equal o 8 is implemened in he design. The value of each genome in he defined chromosome is he index of inpu variables. These chromosomes correspond o he possible soluions in he selecion process. Once he individual chromosome is defined, an iniial populaion S conaining n number of chromosomes are generaed. The finess funcion is developed ha maximizes he correlaion coefficien and minimizes he mean square error (MSE) beween acual and forecas load. The defined finess funcion is called (mxrmne), maximum correlaion value and minimum error (see equaion. 3). Mean square error (equaion. ) is used as a performance index in his case. The roulee wheel selecion mehod wih a single poin crossover and mulipoin muaion is implemened in his echnique. The muaion operaor is defined as given in equaion 4. This operaor complemens he values in he genome o avoid from he local minima [14]. In he firs sep of his geneic-based algorihm, he chromosome lengh and index of inpu variables are defined. Real coded GA is implemened wih a fixed lengh chromosome i.e. 8 and he value of each genome in a paricular chromosome is he index of he inpu variable. In he second sep, he GA is used for he selecions of he final daa se on he basis of defined finess funcion. The crossover and muaion raes are seleced as 0.8 and 0.1 respecively. The process flow diagram of he proposed mehod is shown in Fig. 1. mu Q, k,qk K Q (1) k () err (A P ) 1 1 max R, min err Q, p x, p.r(x ) i Q xi Q (A P) (3) i ò R finness funcion (4) 1 err where, Q is a subse of inpu variables, R is he correlaion value and p is arge class of he predicion variable. A and P are he acual and prediced load values. Whereas, k is muaion poin and K is he lengh of inpu dimension. A se of inpu variables is iniially generaed conaining hisorical load of differen ime inervals, meeorological variables and dae/ime indicaors. The se is hen splied ino wo subses conaining hisorical load relaed variables and meeorological variables respecievely. The proposed selecion process in implemened on he original se and he subses. Consequenly, hree opimally seleced inpu variable vecors IV1, IV and IV3 are reurned, conaining combinaion of load and meerological variables, load relaed variables and meeorological variables, respecievly. The opimized inpu variable vecor IV1 is composed of; day and ime indicaors; dry bulb emperaure of he same ime in he previous day T(w, d, h), dew poin of he same ime in he previous day D(w,d-1,h), elecrical load of he same ime in he previous day L(w, d-1, h), load of same day and ime in he previous week L(w-1, d, h), load of he previous day minus one hours L (w, d, h-1) and load of he previous day minus wo hours L (w, d, h-). B. Developmen of opimized models As menioned previously, he performance of MLP neural nework models depends on is free parameers, and choice of he raining algorihm. The opimizaion algorihms including GA and SA are developed o une he crucial parameers of he nework via minimizing raining errors and DOI /IJSSST.a ISS: x online, prin
3 validaion errors. In his paper, we choose back-propagaion (BP) as a raining algorihm. In his process an iniial populaion (p) of 30 chromosomes is creaed and he number of ieraions (ier) is se o 100. The finess funcion is defined which is he minimum mean square error (MSE) beween he acual and forecas values of MLP. The MLP is rained and esed for he raining daa and MSE and finess [f(p(ier)] is calculaed for he curren ieraion. The GA operaors crossover and muaion are applied on he seleced pairs in he populaion and a new populaion is generaed and esed on he basis of he finess funcion in he nex ieraion (ier=ier+1). The eliism mehod of selecion is used wih he crossover and muaion rae of 0.8 and 0.1 respecively. error. In his way, he forecas error is periodically reduced afer every ieraion. The iniial values of he criical conrol parameers for GA and SA are summarized in Table II. TABLE II. IITIAL PARAMETERS OF EMPLOYED OPTIMIZATIO ALGORITHMS GA Parameers SA Parameers Populaion size 30 Populaion size 30 Crossover rae 0.8 Minimum Temperaure 0.00 Muaion rae 0.1 Maximum Temperaure umber of ieraions 100 Sep size (ΔT) 0.98 ormal disribuion (σ) 0.5 C. Performance Analysis To analyze he performance and accuracy of hybrid predicion models on he basis of IV1, IV and IV3, roo mean square error (RMSE), mean absolue percenage error (MAPE), mean square error (MSE) and mean absolue error (MAE) are used. These performance evaluaion measures can be compued as follows: 1 RMSE ( A P ) 1 1 A P MAPE 100% A (5) (6) 1 1 ( ) (7) 1 1 (8) 1 MSE A P MAE A P Figure 1. Flow diagram for inpu variable selecion echnique. The opimizaion process of he simulaed annealing algorihm sars wih generaing an iniial soluion (rank marix) R aken as he curren saring soluion. Then a neighbor (rank marix) R* of (rank marix) R is generaed and he difference Δ=F(R * ) F(R) in he objecive funcion values of boh schedules is calculaed. If Δ < 0, he neighbor R* is acceped as he new saring soluion in he nex ieraion since i has a beer funcion value. If he objecive funcion value does no decrease (i.e. Δ 0), he generaed neighbor may also be acceped wih a probabiliy exp( Δ/E), where E is a conrol parameer called where, A and P are he acual and prediced values a ime poin. The MAPE is considered as a benchmark performance index due o is sable performance ha resolves he inconsisency problem in he predicion resuls [10]. IV. RESULTS AD DISCUSSIO The hybrid models are applied o predic he elecrical energy demand. As he predicion ime horizon is one dayahead (wih 30 minue inerval), so he oal numbers of observaions would be fory-eigh. The opimally seleced inpu variable vecors IV1, IV and IV3 are used for he raining and esing of he hybrid models. The predicion resuls of he opimized hybrid models MLP-GA and MLP-SA are compared on he basis DOI /IJSSST.a ISS: x online, prin
4 of MSE, MAE, MAPE and RMSE. These resuls are depiced in Fig., Fig. 3 and Table III. The hybrid model based on GA has shown beer performance as compared o he SA based model for one day-ahead load demand forecass. The bes forecasing performance is observed by using he hisorical load and weaher variables relaed inpu vecor IV1. The MAPE using IV1 for MPL-GA and MLP-SA are observed 1.75% and 1.91%, respecively. TABLE III. Forecas Techniques MLP-GA OE DAY-AHEAD FORECAST RESULTS OF OPTIMIZED MODELS BASED O IPUT VECTORS Opimized inpu variable Performance Index vecors MAE MSE MAPE RMSE Load & Weaher (IV1) Load Relaed (IV) Weaher Relaed (IV3) MLP-SA Load & Weaher (IV1) Load Relaed (IV) Weaher Relaed (IV3) The comparaive analysis of MLP-GA wih ha of he MLP-SA reveals ha, he GA par has improved performance han he SA par. Figure. One day-ahead forecas resuls based on IV1 o IV3 for MLP-GA Figure 3. One day-ahead forecas resuls based on IV1 o IV3 for MLP-SA model. The improved percenage of he MAPE from IV1 o IV3 is observed as; 0.6%, 0.4% and 0.03%, respecively. On he oher hand, IV3 generaed high forecas error because of he weak correlaion of he meeorological inpu variables wih he load demand. The use of load relaed variables (IV) produced reasonable accuracy which is slighly lesser ha he resuls shown by implemening IV1. V. COCLUSIO This paper proposed a hybrid mehod o idenify and selec he bes inpu variables for A based forecas models. In his echnique, he combined effec of minimum MSE and maximum value of correlaion coefficien are used o develop a finess funcion of GA for he selecion of mos influenial IV in STLF. In his way, hree inpu vecors, including, load relaed (IV3), merological relaed (IV) and a combined vecor of hese wo ypes of variables (IV1) are seleced. These opimally seleced inpu vecors are deployed in he raining and esing processes of MLP- GA and MLP-SA models and he resuls are analysed. These resuls show ha he performance of opimized MLP model wih GA using IV1 ouperformed all oher models and inpu vecors used in his research. The obained resuls show ha he research has provided a suiable echnique for he selecion of bes inpu variables for he A based hybrid load demand predicion models. ACKOWLEDGMET The auhors would like o hank and appreciae he suppor of Universiy Technology PETROAS for providing he funding under he gran number (URIF 0153AA-B13) o conduc his research. DOI /IJSSST.a ISS: x online, prin
5 REFERECES [1] K. S. Reddy, M. Kumar, T. K. Mallick, H. Sharon, and S. Lokeswaran, "A review of Inegraion, Conrol, Communicaion and Meering (ICCM) of renewable energy based smar grid," Renewable and Susainable Energy Reviews, vol. 38, pp , 10// 014. [] H. K. Alfares and M. azeeruddin, "Elecric load forecasing: Lieraure survey and classificaion of mehods," Inernaional Journal of Sysems Science, vol. 33, pp. 3-34, 00/01/ [3] L. Suganhi and A. A. Samuel, "Energy models for demand forecasing A review," Renewable and Susainable Energy Reviews, vol. 16, pp , // 01. [4] A. Asar, S. Hassnain, and A. Khaack, "A Muli-agen Approach To Shor Term Load Forecasing Problem," The Inernaional Journal of Inelligen Conrol and Sysems, vol. 10, pp. 5-59, 005. [5] A. P. Alves da Silva, V. H. Ferreira, and R. M. G. Velasquez, "Inpu space o neural nework based load forecasers," Inernaional Journal of Forecasing, vol. 4, pp , 008. [6] T. Mahmoud, D. Habibi, O. Bass, and S. Lachowicz, "Load demand forecasing: Model inpus selecion," in Innovaive Smar Grid Technologies Asia (ISGT), IEEE PES, 011, pp [7] K.-h. Yang, G.-L. Shan, and L.-L. Zhao, "Correlaion coefficien mehod for suppor vecor machine inpu samples," in Inernaional Conference on Machine Learning and Cyberneics,, 006, pp [8] P. J. Sanos, A. G. Marins, and A. J. Pires, "Designing he inpu vecor o A-based models for shor-erm load forecas in elecriciy disribuion sysems," Inernaional Journal of Elecrical Power & Energy Sysems, vol. 9, pp , 007. [9] I. Drezga and S. Rahman, "Inpu variable selecion for A-based shor-erm load forecasing," Power Sysems, IEEE Transacions on, vol. 13, pp , [10] M. Ghayekhloo, M. Menhaj, and M. Ghofrani, "A hybrid shor-erm load forecasing wih a new daa preprocessing framework," Elecric Power Sysems Research, vol. 119, pp , 015. [11] I. Drezga and S. Rahman, "Shor-erm load forecasing wih local A predicors," Power Sysems, IEEE Transacions on, vol. 14, pp , [1] P. R. Khazaee,. Mozayani, and M. J. Molagh, "A geneic-based inpu variable selecion algorihm using muual informaion and wavele nework for ime series predicion," in Sysems, Man and Cyberneics, 008. SMC 008. IEEE Inernaional Conference on, 008, pp [13] L. Xiao, J. Wang, X. Yang, and L. Xiao, "A hybrid model based on daa preprocessing for elecrical power forecasing," Inernaional Journal of Elecrical Power & Energy Sysems, vol. 64, pp , 1// 015. [14] H. Sapahy, "Real-coded GA for parameer opimizaion in shorerm load forecasing," Arificiel eural es Problem Solving Mehods, pp , 003. DOI /IJSSST.a ISS: x online, prin
CHAPTER 10 VALIDATION OF TEST WITH ARTIFICAL NEURAL NETWORK
175 CHAPTER 10 VALIDATION OF TEST WITH ARTIFICAL NEURAL NETWORK 10.1 INTRODUCTION Amongs he research work performed, he bes resuls of experimenal work are validaed wih Arificial Neural Nework. From he
More informationApplication of a Stochastic-Fuzzy Approach to Modeling Optimal Discrete Time Dynamical Systems by Using Large Scale Data Processing
Applicaion of a Sochasic-Fuzzy Approach o Modeling Opimal Discree Time Dynamical Sysems by Using Large Scale Daa Processing AA WALASZE-BABISZEWSA Deparmen of Compuer Engineering Opole Universiy of Technology
More informationApplying Genetic Algorithms for Inventory Lot-Sizing Problem with Supplier Selection under Storage Capacity Constraints
IJCSI Inernaional Journal of Compuer Science Issues, Vol 9, Issue 1, No 1, January 2012 wwwijcsiorg 18 Applying Geneic Algorihms for Invenory Lo-Sizing Problem wih Supplier Selecion under Sorage Capaciy
More informationArticle from. Predictive Analytics and Futurism. July 2016 Issue 13
Aricle from Predicive Analyics and Fuurism July 6 Issue An Inroducion o Incremenal Learning By Qiang Wu and Dave Snell Machine learning provides useful ools for predicive analyics The ypical machine learning
More informationPROC NLP Approach for Optimal Exponential Smoothing Srihari Jaganathan, Cognizant Technology Solutions, Newbury Park, CA.
PROC NLP Approach for Opimal Exponenial Smoohing Srihari Jaganahan, Cognizan Technology Soluions, Newbury Park, CA. ABSTRACT Esimaion of smoohing parameers and iniial values are some of he basic requiremens
More informationParticle Swarm Optimization
Paricle Swarm Opimizaion Speaker: Jeng-Shyang Pan Deparmen of Elecronic Engineering, Kaohsiung Universiy of Applied Science, Taiwan Email: jspan@cc.kuas.edu.w 7/26/2004 ppso 1 Wha is he Paricle Swarm Opimizaion
More informationAir Quality Index Prediction Using Error Back Propagation Algorithm and Improved Particle Swarm Optimization
Air Qualiy Index Predicion Using Error Back Propagaion Algorihm and Improved Paricle Swarm Opimizaion Jia Xu ( ) and Lang Pei College of Compuer Science, Wuhan Qinchuan Universiy, Wuhan, China 461406563@qq.com
More informationLecture 3: Exponential Smoothing
NATCOR: Forecasing & Predicive Analyics Lecure 3: Exponenial Smoohing John Boylan Lancaser Cenre for Forecasing Deparmen of Managemen Science Mehods and Models Forecasing Mehod A (numerical) procedure
More informationNavneet Saini, Mayank Goyal, Vishal Bansal (2013); Term Project AML310; Indian Institute of Technology Delhi
Creep in Viscoelasic Subsances Numerical mehods o calculae he coefficiens of he Prony equaion using creep es daa and Herediary Inegrals Mehod Navnee Saini, Mayank Goyal, Vishal Bansal (23); Term Projec
More informationAn Optimal Dynamic Generation Scheduling for a Wind-Thermal Power System *
Energy and Power Engineering, 2013, 5, 1016-1021 doi:10.4236/epe.2013.54b194 Published Online July 2013 (hp://www.scirp.org/journal/epe) An Opimal Dynamic Generaion Scheduling for a Wind-Thermal Power
More informationA Fusion Model for Day-Ahead Wind Speed Prediction based on the Validity of the Information
A Fusion Model for Day-Ahead Wind Speed Predicion based on he Validiy of he Informaion Jie Wan 1,a, Wenbo Hao 2,b, Guorui Ren 1,c,Leilei Zhao 2,d, Bingliang Xu 2,e, Chengzhi Sun 2,f,Zhigang Zhao 3,g, Chengrui
More informationForecasting Stock Exchange Movements Using Artificial Neural Network Models and Hybrid Models
Forecasing Sock Exchange Movemens Using Arificial Neural Nework Models and Hybrid Models Erkam GÜREEN and Gülgün KAYAKUTLU Isanbul Technical Universiy, Deparmen of Indusrial Engineering, Maçka, 34367 Isanbul,
More informationRecent Developments In Evolutionary Data Assimilation And Model Uncertainty Estimation For Hydrologic Forecasting Hamid Moradkhani
Feb 6-8, 208 Recen Developmens In Evoluionary Daa Assimilaion And Model Uncerainy Esimaion For Hydrologic Forecasing Hamid Moradkhani Cener for Complex Hydrosysems Research Deparmen of Civil, Consrucion
More informationA Framework for Efficient Document Ranking Using Order and Non Order Based Fitness Function
A Framework for Efficien Documen Ranking Using Order and Non Order Based Finess Funcion Hazra Imran, Adii Sharan Absrac One cenral problem of informaion rerieval is o deermine he relevance of documens
More informationRobust estimation based on the first- and third-moment restrictions of the power transformation model
h Inernaional Congress on Modelling and Simulaion, Adelaide, Ausralia, 6 December 3 www.mssanz.org.au/modsim3 Robus esimaion based on he firs- and hird-momen resricions of he power ransformaion Nawaa,
More informationLinear Gaussian State Space Models
Linear Gaussian Sae Space Models Srucural Time Series Models Level and Trend Models Basic Srucural Model (BSM Dynamic Linear Models Sae Space Model Represenaion Level, Trend, and Seasonal Models Time Varying
More informationEnsamble methods: Bagging and Boosting
Lecure 21 Ensamble mehods: Bagging and Boosing Milos Hauskrech milos@cs.pi.edu 5329 Senno Square Ensemble mehods Mixure of expers Muliple base models (classifiers, regressors), each covers a differen par
More informationChapter 15. Time Series: Descriptive Analyses, Models, and Forecasting
Chaper 15 Time Series: Descripive Analyses, Models, and Forecasing Descripive Analysis: Index Numbers Index Number a number ha measures he change in a variable over ime relaive o he value of he variable
More informationAn recursive analytical technique to estimate time dependent physical parameters in the presence of noise processes
WHAT IS A KALMAN FILTER An recursive analyical echnique o esimae ime dependen physical parameers in he presence of noise processes Example of a ime and frequency applicaion: Offse beween wo clocks PREDICTORS,
More informationAir Traffic Forecast Empirical Research Based on the MCMC Method
Compuer and Informaion Science; Vol. 5, No. 5; 0 ISSN 93-8989 E-ISSN 93-8997 Published by Canadian Cener of Science and Educaion Air Traffic Forecas Empirical Research Based on he MCMC Mehod Jian-bo Wang,
More informationEVALUATING FORECASTING MODELS FOR UNEMPLOYMENT RATES BY GENDER IN SELECTED EUROPEAN COUNTRIES
Inerdisciplinary Descripion of Complex Sysems 15(1), 16-35, 217 EVALUATING FORECASTING MODELS FOR UNEMPLOYMENT RATES BY GENDER IN SELECTED EUROPEAN COUNTRIES Ksenija Dumičić*, Berislav Žmuk and Ania Čeh
More informationApplying Genetic Algorithms for Inventory Lot-Sizing Problem with Supplier Selection under Storage Space
Inernaional Journal of Indusrial and Manufacuring Engineering Applying Geneic Algorihms for Invenory Lo-Sizing Problem wih Supplier Selecion under Sorage Space Vichai Rungreunganaun and Chirawa Woarawichai
More informationSingle-Pass-Based Heuristic Algorithms for Group Flexible Flow-shop Scheduling Problems
Single-Pass-Based Heurisic Algorihms for Group Flexible Flow-shop Scheduling Problems PEI-YING HUANG, TZUNG-PEI HONG 2 and CHENG-YAN KAO, 3 Deparmen of Compuer Science and Informaion Engineering Naional
More informationModal identification of structures from roving input data by means of maximum likelihood estimation of the state space model
Modal idenificaion of srucures from roving inpu daa by means of maximum likelihood esimaion of he sae space model J. Cara, J. Juan, E. Alarcón Absrac The usual way o perform a forced vibraion es is o fix
More informationTime series Decomposition method
Time series Decomposiion mehod A ime series is described using a mulifacor model such as = f (rend, cyclical, seasonal, error) = f (T, C, S, e) Long- Iner-mediaed Seasonal Irregular erm erm effec, effec,
More informationEnsamble methods: Boosting
Lecure 21 Ensamble mehods: Boosing Milos Hauskrech milos@cs.pi.edu 5329 Senno Square Schedule Final exam: April 18: 1:00-2:15pm, in-class Term projecs April 23 & April 25: a 1:00-2:30pm in CS seminar room
More informationGeorey E. Hinton. University oftoronto. Technical Report CRG-TR February 22, Abstract
Parameer Esimaion for Linear Dynamical Sysems Zoubin Ghahramani Georey E. Hinon Deparmen of Compuer Science Universiy oftorono 6 King's College Road Torono, Canada M5S A4 Email: zoubin@cs.orono.edu Technical
More informationMODULE - 9 LECTURE NOTES 2 GENETIC ALGORITHMS
1 MODULE - 9 LECTURE NOTES 2 GENETIC ALGORITHMS INTRODUCTION Mos real world opimizaion problems involve complexiies like discree, coninuous or mixed variables, muliple conflicing objecives, non-lineariy,
More informationKeywords Digital Infinite-Impulse Response (IIR) filter, Digital Finite-Impulse Response (FIR) filter, DE, exploratory move
Volume 5, Issue 7, July 2015 ISSN: 2277 128X Inernaional Journal of Advanced Research in Compuer Science and Sofware Engineering Research Paper Available online a: www.ijarcsse.com A Hybrid Differenial
More informationOn Comparison between Evolutionary Programming Network based Learning and Novel Evolution Strategy Algorithm-based Learning
Proceedings of he ICEECE December -4 (003), Dhaka, Bangladesh On Comparison beween Evoluionary Programming Nework based Learning and vel Evoluion Sraegy Algorihm-based Learning M. A. Khayer Azad, Md. Shafiqul
More informationIJSRD - International Journal for Scientific Research & Development Vol. 3, Issue 11, 2016 ISSN (online):
IJSRD - Inernaional Journal for Scienific Research & Developmen Vol. 3, Issue 11, 2016 ISS (online): 2321-0613 Sudy of Differen Techniques for Load Forecasing - A Review Vivek Kumar Verma 1 Yashwan Singh
More informationPhysics 235 Chapter 2. Chapter 2 Newtonian Mechanics Single Particle
Chaper 2 Newonian Mechanics Single Paricle In his Chaper we will review wha Newon s laws of mechanics ell us abou he moion of a single paricle. Newon s laws are only valid in suiable reference frames,
More informationLearning a Class from Examples. Training set X. Class C 1. Class C of a family car. Output: Input representation: x 1 : price, x 2 : engine power
Alpaydin Chaper, Michell Chaper 7 Alpaydin slides are in urquoise. Ehem Alpaydin, copyrigh: The MIT Press, 010. alpaydin@boun.edu.r hp://www.cmpe.boun.edu.r/ ehem/imle All oher slides are based on Michell.
More informationParticle Swarm Optimization Combining Diversification and Intensification for Nonlinear Integer Programming Problems
Paricle Swarm Opimizaion Combining Diversificaion and Inensificaion for Nonlinear Ineger Programming Problems Takeshi Masui, Masaoshi Sakawa, Kosuke Kao and Koichi Masumoo Hiroshima Universiy 1-4-1, Kagamiyama,
More information3.1 More on model selection
3. More on Model selecion 3. Comparing models AIC, BIC, Adjused R squared. 3. Over Fiing problem. 3.3 Sample spliing. 3. More on model selecion crieria Ofen afer model fiing you are lef wih a handful of
More informationSTATE-SPACE MODELLING. A mass balance across the tank gives:
B. Lennox and N.F. Thornhill, 9, Sae Space Modelling, IChemE Process Managemen and Conrol Subjec Group Newsleer STE-SPACE MODELLING Inroducion: Over he pas decade or so here has been an ever increasing
More informationThe Rosenblatt s LMS algorithm for Perceptron (1958) is built around a linear neuron (a neuron with a linear
In The name of God Lecure4: Percepron and AALIE r. Majid MjidGhoshunih Inroducion The Rosenbla s LMS algorihm for Percepron 958 is buil around a linear neuron a neuron ih a linear acivaion funcion. Hoever,
More informationRandom Walk with Anti-Correlated Steps
Random Walk wih Ani-Correlaed Seps John Noga Dirk Wagner 2 Absrac We conjecure he expeced value of random walks wih ani-correlaed seps o be exacly. We suppor his conjecure wih 2 plausibiliy argumens and
More informationDimitri Solomatine. D.P. Solomatine. Data-driven modelling (part 2). 2
Daa-driven modelling. Par. Daa-driven Arificial di Neural modelling. Newors Par Dimiri Solomaine Arificial neural newors D.P. Solomaine. Daa-driven modelling par. 1 Arificial neural newors ANN: main pes
More information20. Applications of the Genetic-Drift Model
0. Applicaions of he Geneic-Drif Model 1) Deermining he probabiliy of forming any paricular combinaion of genoypes in he nex generaion: Example: If he parenal allele frequencies are p 0 = 0.35 and q 0
More informationTime Series Forecasting using CCA and Kohonen Maps - Application to Electricity Consumption
ESANN'2000 proceedings - European Symposium on Arificial Neural Neworks Bruges (Belgium), 26-28 April 2000, D-Faco public., ISBN 2-930307-00-5, pp. 329-334. Time Series Forecasing using CCA and Kohonen
More informationQ.1 Define work and its unit?
CHP # 6 ORK AND ENERGY Q.1 Define work and is uni? A. ORK I can be define as when we applied a force on a body and he body covers a disance in he direcion of force, hen we say ha work is done. I is a scalar
More informationINTRODUCTION TO MACHINE LEARNING 3RD EDITION
ETHEM ALPAYDIN The MIT Press, 2014 Lecure Slides for INTRODUCTION TO MACHINE LEARNING 3RD EDITION alpaydin@boun.edu.r hp://www.cmpe.boun.edu.r/~ehem/i2ml3e CHAPTER 2: SUPERVISED LEARNING Learning a Class
More informationPENALIZED LEAST SQUARES AND PENALIZED LIKELIHOOD
PENALIZED LEAST SQUARES AND PENALIZED LIKELIHOOD HAN XIAO 1. Penalized Leas Squares Lasso solves he following opimizaion problem, ˆβ lasso = arg max β R p+1 1 N y i β 0 N x ij β j β j (1.1) for some 0.
More informationSmoothing. Backward smoother: At any give T, replace the observation yt by a combination of observations at & before T
Smoohing Consan process Separae signal & noise Smooh he daa: Backward smooher: A an give, replace he observaion b a combinaion of observaions a & before Simple smooher : replace he curren observaion wih
More informationMATHEMATICAL DESCRIPTION OF THEORETICAL METHODS OF RESERVE ECONOMY OF CONSIGNMENT STORES
MAHEMAICAL DESCIPION OF HEOEICAL MEHODS OF ESEVE ECONOMY OF CONSIGNMEN SOES Péer elek, József Cselényi, György Demeer Universiy of Miskolc, Deparmen of Maerials Handling and Logisics Absrac: Opimizaion
More informationThe field of mathematics has made tremendous impact on the study of
A Populaion Firing Rae Model of Reverberaory Aciviy in Neuronal Neworks Zofia Koscielniak Carnegie Mellon Universiy Menor: Dr. G. Bard Ermenrou Universiy of Pisburgh Inroducion: The field of mahemaics
More informationLearning a Class from Examples. Training set X. Class C 1. Class C of a family car. Output: Input representation: x 1 : price, x 2 : engine power
Alpaydin Chaper, Michell Chaper 7 Alpaydin slides are in urquoise. Ehem Alpaydin, copyrigh: The MIT Press, 010. alpaydin@boun.edu.r hp://www.cmpe.boun.edu.r/ ehem/imle All oher slides are based on Michell.
More informationThe electromagnetic interference in case of onboard navy ships computers - a new approach
The elecromagneic inerference in case of onboard navy ships compuers - a new approach Prof. dr. ing. Alexandru SOTIR Naval Academy Mircea cel Bărân, Fulgerului Sree, Consanţa, soiralexandru@yahoo.com Absrac.
More informationShiva Akhtarian MSc Student, Department of Computer Engineering and Information Technology, Payame Noor University, Iran
Curren Trends in Technology and Science ISSN : 79-055 8hSASTech 04 Symposium on Advances in Science & Technology-Commission-IV Mashhad, Iran A New for Sofware Reliabiliy Evaluaion Based on NHPP wih Imperfec
More informationFITTING OF A PARTIALLY REPARAMETERIZED GOMPERTZ MODEL TO BROILER DATA
FITTING OF A PARTIALLY REPARAMETERIZED GOMPERTZ MODEL TO BROILER DATA N. Okendro Singh Associae Professor (Ag. Sa.), College of Agriculure, Cenral Agriculural Universiy, Iroisemba 795 004, Imphal, Manipur
More informationNotes on Kalman Filtering
Noes on Kalman Filering Brian Borchers and Rick Aser November 7, Inroducion Daa Assimilaion is he problem of merging model predicions wih acual measuremens of a sysem o produce an opimal esimae of he curren
More informationCellular Automata for Pattern Recognition
Chaper 3 Cellular Auomaa for Paern Recogniion Sarra Wonghanavasu and Jesada Ponkaew Addiional informaion is available a he end of he chaper hp://dx.doi.org/10.5772/52364 1. Inroducion Cellular Auomaa (CA)
More informationDevelopment of a new metrological model for measuring of the water surface evaporation Tovmach L. Tovmach Yr. Abstract Introduction
Developmen of a new merological model for measuring of he waer surface evaporaion Tovmach L. Tovmach Yr. Sae Hydrological Insiue 23 Second Line, 199053 S. Peersburg, Russian Federaion Telephone (812) 323
More informationExcel-Based Solution Method For The Optimal Policy Of The Hadley And Whittin s Exact Model With Arma Demand
Excel-Based Soluion Mehod For The Opimal Policy Of The Hadley And Whiin s Exac Model Wih Arma Demand Kal Nami School of Business and Economics Winson Salem Sae Universiy Winson Salem, NC 27110 Phone: (336)750-2338
More informationModel for forecasting expressway surface temperature in snowy areas
Journal of Naural Disaser Science, Volume 37,Number 2,216,pp49-64 Model for forecasing expressway surface emperaure in snowy areas Masafumi Horii* and Takayuki Hayami** * Deparmen of Civil Engineering,
More informationImproved Approximate Solutions for Nonlinear Evolutions Equations in Mathematical Physics Using the Reduced Differential Transform Method
Journal of Applied Mahemaics & Bioinformaics, vol., no., 01, 1-14 ISSN: 179-660 (prin), 179-699 (online) Scienpress Ld, 01 Improved Approimae Soluions for Nonlinear Evoluions Equaions in Mahemaical Physics
More informationFourier Transformation on Model Fitting for Pakistan Inflation Rate
Fourier Transformaion on Model Fiing for Pakisan Inflaion Rae Anam Iqbal (Corresponding auhor) Dep. of Saisics, Gov. Pos Graduae College (w), Sargodha, Pakisan Tel: 92-321-603-4232 E-mail: anammughal343@gmail.com
More informationLecture 4 Kinetics of a particle Part 3: Impulse and Momentum
MEE Engineering Mechanics II Lecure 4 Lecure 4 Kineics of a paricle Par 3: Impulse and Momenum Linear impulse and momenum Saring from he equaion of moion for a paricle of mass m which is subjeced o an
More informationEnsemble Confidence Estimates Posterior Probability
Ensemble Esimaes Poserior Probabiliy Michael Muhlbaier, Aposolos Topalis, and Robi Polikar Rowan Universiy, Elecrical and Compuer Engineering, Mullica Hill Rd., Glassboro, NJ 88, USA {muhlba6, opali5}@sudens.rowan.edu
More informationForecasting. Summary. Sample StatFolio: tsforecast.sgp. STATGRAPHICS Centurion Rev. 9/16/2013
STATGRAPHICS Cenurion Rev. 9/16/2013 Forecasing Summary... 1 Daa Inpu... 3 Analysis Opions... 5 Forecasing Models... 9 Analysis Summary... 21 Time Sequence Plo... 23 Forecas Table... 24 Forecas Plo...
More informationSTRUCTURAL CHANGE IN TIME SERIES OF THE EXCHANGE RATES BETWEEN YEN-DOLLAR AND YEN-EURO IN
Inernaional Journal of Applied Economerics and Quaniaive Sudies. Vol.1-3(004) STRUCTURAL CHANGE IN TIME SERIES OF THE EXCHANGE RATES BETWEEN YEN-DOLLAR AND YEN-EURO IN 001-004 OBARA, Takashi * Absrac The
More informationAn introduction to the theory of SDDP algorithm
An inroducion o he heory of SDDP algorihm V. Leclère (ENPC) Augus 1, 2014 V. Leclère Inroducion o SDDP Augus 1, 2014 1 / 21 Inroducion Large scale sochasic problem are hard o solve. Two ways of aacking
More informationStochastic Model for Cancer Cell Growth through Single Forward Mutation
Journal of Modern Applied Saisical Mehods Volume 16 Issue 1 Aricle 31 5-1-2017 Sochasic Model for Cancer Cell Growh hrough Single Forward Muaion Jayabharahiraj Jayabalan Pondicherry Universiy, jayabharahi8@gmail.com
More informationLinear Combinations of Volatility Forecasts for the WIG20 and Polish Exchange Rates
Eliza Buszkowska Universiy of Poznań, Poland Linear Combinaions of Volailiy Forecass for he WIG0 and Polish Exchange Raes Absrak. As is known forecas combinaions may be beer forecass hen forecass obained
More informationJournal of Chemical and Pharmaceutical Research, 2014, 6(5): Research Article
Available online www.jocpr.com Journal of Chemical and Pharmaceuical Research, 204, 6(5):70-705 Research Aricle ISSN : 0975-7384 CODEN(USA) : JCPRC5 Cells formaion wih a muli-objecive geneic algorihm Jun
More informationLecture 2 October ε-approximation of 2-player zero-sum games
Opimizaion II Winer 009/10 Lecurer: Khaled Elbassioni Lecure Ocober 19 1 ε-approximaion of -player zero-sum games In his lecure we give a randomized ficiious play algorihm for obaining an approximae soluion
More informationComputation of the Effect of Space Harmonics on Starting Process of Induction Motors Using TSFEM
Journal of elecrical sysems Special Issue N 01 : November 2009 pp: 48-52 Compuaion of he Effec of Space Harmonics on Saring Process of Inducion Moors Using TSFEM Youcef Ouazir USTHB Laboraoire des sysèmes
More informationNature Neuroscience: doi: /nn Supplementary Figure 1. Spike-count autocorrelations in time.
Supplemenary Figure 1 Spike-coun auocorrelaions in ime. Normalized auocorrelaion marices are shown for each area in a daase. The marix shows he mean correlaion of he spike coun in each ime bin wih he spike
More informationA car following model for traffic flow simulation
Inernaional Journal of Applied Mahemaical Sciences ISSN 0973-076 Volume 9, Number (206), pp. -9 Research India Publicaions hp://www.ripublicaion.com A car following model for raffic flow simulaion Doudou
More informationInventory Control of Perishable Items in a Two-Echelon Supply Chain
Journal of Indusrial Engineering, Universiy of ehran, Special Issue,, PP. 69-77 69 Invenory Conrol of Perishable Iems in a wo-echelon Supply Chain Fariborz Jolai *, Elmira Gheisariha and Farnaz Nojavan
More informationNumerical Dispersion
eview of Linear Numerical Sabiliy Numerical Dispersion n he previous lecure, we considered he linear numerical sabiliy of boh advecion and diffusion erms when approimaed wih several spaial and emporal
More informationMulti-scale 2D acoustic full waveform inversion with high frequency impulsive source
Muli-scale D acousic full waveform inversion wih high frequency impulsive source Vladimir N Zubov*, Universiy of Calgary, Calgary AB vzubov@ucalgaryca and Michael P Lamoureux, Universiy of Calgary, Calgary
More informationSub Module 2.6. Measurement of transient temperature
Mechanical Measuremens Prof. S.P.Venkaeshan Sub Module 2.6 Measuremen of ransien emperaure Many processes of engineering relevance involve variaions wih respec o ime. The sysem properies like emperaure,
More informationZápadočeská Univerzita v Plzni, Czech Republic and Groupe ESIEE Paris, France
ADAPTIVE SIGNAL PROCESSING USING MAXIMUM ENTROPY ON THE MEAN METHOD AND MONTE CARLO ANALYSIS Pavla Holejšovsá, Ing. *), Z. Peroua, Ing. **), J.-F. Bercher, Prof. Assis. ***) Západočesá Univerzia v Plzni,
More informationVehicle Arrival Models : Headway
Chaper 12 Vehicle Arrival Models : Headway 12.1 Inroducion Modelling arrival of vehicle a secion of road is an imporan sep in raffic flow modelling. I has imporan applicaion in raffic flow simulaion where
More informationSpring Ammar Abu-Hudrouss Islamic University Gaza
Chaper 7 Reed-Solomon Code Spring 9 Ammar Abu-Hudrouss Islamic Universiy Gaza ١ Inroducion A Reed Solomon code is a special case of a BCH code in which he lengh of he code is one less han he size of he
More informationSimulation-Solving Dynamic Models ABE 5646 Week 2, Spring 2010
Simulaion-Solving Dynamic Models ABE 5646 Week 2, Spring 2010 Week Descripion Reading Maerial 2 Compuer Simulaion of Dynamic Models Finie Difference, coninuous saes, discree ime Simple Mehods Euler Trapezoid
More informationOptimal Design of LQR Weighting Matrices based on Intelligent Optimization Methods
Opimal Design of LQR Weighing Marices based on Inelligen Opimizaion Mehods Inernaional Journal of Inelligen Informaion Processing, Volume, Number, March Opimal Design of LQR Weighing Marices based on Inelligen
More informationData assimilation for local rainfall near Tokyo on 18 July 2013 using EnVAR with observation space localization
Daa assimilaion for local rainfall near Tokyo on 18 July 2013 using EnVAR wih observaion space localizaion *1 Sho Yokoa, 1 Masaru Kunii, 1 Kazumasa Aonashi, 1 Seiji Origuchi, 2,1 Le Duc, 1 Takuya Kawabaa,
More informationScheduling of Crude Oil Movements at Refinery Front-end
Scheduling of Crude Oil Movemens a Refinery Fron-end Ramkumar Karuppiah and Ignacio Grossmann Carnegie Mellon Universiy ExxonMobil Case Sudy: Dr. Kevin Furman Enerprise-wide Opimizaion Projec March 15,
More informationTask-based Configuration Optimization of Modular and Reconfigurable Robots using a Multi-solution Inverse Kinematics Solver
CARV 2007 Task-based Configuraion Opimizaion of Modular and Reconfigurable Robos using a Muli-soluion Inverse Kinemaics Solver S. Tabandeh 1, C. Clark 2, W. Melek 3 Absrac: Modular and Reconfigurable Robos
More informationLAPLACE TRANSFORM AND TRANSFER FUNCTION
CHBE320 LECTURE V LAPLACE TRANSFORM AND TRANSFER FUNCTION Professor Dae Ryook Yang Spring 2018 Dep. of Chemical and Biological Engineering 5-1 Road Map of he Lecure V Laplace Transform and Transfer funcions
More informationSpeaker Adaptation Techniques For Continuous Speech Using Medium and Small Adaptation Data Sets. Constantinos Boulis
Speaker Adapaion Techniques For Coninuous Speech Using Medium and Small Adapaion Daa Ses Consaninos Boulis Ouline of he Presenaion Inroducion o he speaker adapaion problem Maximum Likelihood Sochasic Transformaions
More informationMethodology. -ratios are biased and that the appropriate critical values have to be increased by an amount. that depends on the sample size.
Mehodology. Uni Roo Tess A ime series is inegraed when i has a mean revering propery and a finie variance. I is only emporarily ou of equilibrium and is called saionary in I(0). However a ime series ha
More informationA First Course on Kinetics and Reaction Engineering. Class 19 on Unit 18
A Firs ourse on Kineics and Reacion Engineering lass 19 on Uni 18 Par I - hemical Reacions Par II - hemical Reacion Kineics Where We re Going Par III - hemical Reacion Engineering A. Ideal Reacors B. Perfecly
More informationSection 3.5 Nonhomogeneous Equations; Method of Undetermined Coefficients
Secion 3.5 Nonhomogeneous Equaions; Mehod of Undeermined Coefficiens Key Terms/Ideas: Linear Differenial operaor Nonlinear operaor Second order homogeneous DE Second order nonhomogeneous DE Soluion o homogeneous
More informationEvaluation of Mean Time to System Failure of a Repairable 3-out-of-4 System with Online Preventive Maintenance
American Journal of Applied Mahemaics and Saisics, 0, Vol., No., 9- Available online a hp://pubs.sciepub.com/ajams/// Science and Educaion Publishing DOI:0.69/ajams--- Evaluaion of Mean Time o Sysem Failure
More informationPade and Laguerre Approximations Applied. to the Active Queue Management Model. of Internet Protocol
Applied Mahemaical Sciences, Vol. 7, 013, no. 16, 663-673 HIKARI Ld, www.m-hikari.com hp://dx.doi.org/10.1988/ams.013.39499 Pade and Laguerre Approximaions Applied o he Acive Queue Managemen Model of Inerne
More informationLinear Time-invariant systems, Convolution, and Cross-correlation
Linear Time-invarian sysems, Convoluion, and Cross-correlaion (1) Linear Time-invarian (LTI) sysem A sysem akes in an inpu funcion and reurns an oupu funcion. x() T y() Inpu Sysem Oupu y() = T[x()] An
More informationResource Allocation in Visible Light Communication Networks NOMA vs. OFDMA Transmission Techniques
Resource Allocaion in Visible Ligh Communicaion Neworks NOMA vs. OFDMA Transmission Techniques Eirini Eleni Tsiropoulou, Iakovos Gialagkolidis, Panagiois Vamvakas, and Symeon Papavassiliou Insiue of Communicaions
More informationKINEMATICS IN ONE DIMENSION
KINEMATICS IN ONE DIMENSION PREVIEW Kinemaics is he sudy of how hings move how far (disance and displacemen), how fas (speed and velociy), and how fas ha how fas changes (acceleraion). We say ha an objec
More informationDEVELOPMENT AND ANALYSIS OF GENETIC ALGORITHM FOR TIME SERIES FORECASTING PROBLEM. Leonid Hulianytskyi, Anna Pavlenko
Inernaional Journal "Informaion Models and Analyses" Volume 4 Number 05 3 DEVELOPMENT AND ANALYSIS OF GENETIC ALGORITHM FOR TIME SERIES FORECASTING PROBLEM Leonid Hulianyskyi Anna Pavlenko Absrac: This
More informationSome Basic Information about M-S-D Systems
Some Basic Informaion abou M-S-D Sysems 1 Inroducion We wan o give some summary of he facs concerning unforced (homogeneous) and forced (non-homogeneous) models for linear oscillaors governed by second-order,
More information5.2. The Natural Logarithm. Solution
5.2 The Naural Logarihm The number e is an irraional number, similar in naure o π. Is non-erminaing, non-repeaing value is e 2.718 281 828 59. Like π, e also occurs frequenly in naural phenomena. In fac,
More informationTypes of Exponential Smoothing Methods. Simple Exponential Smoothing. Simple Exponential Smoothing
M Business Forecasing Mehods Exponenial moohing Mehods ecurer : Dr Iris Yeung Room No : P79 Tel No : 788 8 Types of Exponenial moohing Mehods imple Exponenial moohing Double Exponenial moohing Brown s
More informationThe Implementation of Business Decision-Making Tools in Incident Rate Prediction
Session No. 750 The Implemenaion of Business Decision-Making Tools in Inciden Rae Predicion Samuel A. Oyewole, Ph.D. Deparmen of Energy and Mineral Engineering The Pennsylvania Sae Universiy Universiy
More informationSliding Mode Extremum Seeking Control for Linear Quadratic Dynamic Game
Sliding Mode Exremum Seeking Conrol for Linear Quadraic Dynamic Game Yaodong Pan and Ümi Özgüner ITS Research Group, AIST Tsukuba Eas Namiki --, Tsukuba-shi,Ibaraki-ken 5-856, Japan e-mail: pan.yaodong@ais.go.jp
More informationnot to be republished NCERT MATHEMATICAL MODELLING Appendix 2 A.2.1 Introduction A.2.2 Why Mathematical Modelling?
256 MATHEMATICS A.2.1 Inroducion In class XI, we have learn abou mahemaical modelling as an aemp o sudy some par (or form) of some real-life problems in mahemaical erms, i.e., he conversion of a physical
More informationSliding Mode Controller for Unstable Systems
S. SIVARAMAKRISHNAN e al., Sliding Mode Conroller for Unsable Sysems, Chem. Biochem. Eng. Q. 22 (1) 41 47 (28) 41 Sliding Mode Conroller for Unsable Sysems S. Sivaramakrishnan, A. K. Tangirala, and M.
More information