Photovoltaic System Maximum Power Point Genetic Identification Function Biology Essay

This paper proposes the designation of maximal power point ( MPP ) map for photovoltaic ( PV ) faculty utilizing the familial algorithm ( GA ) . Then tax write-off of the needed map to bring forth the mention values to drive the trailing system in the PV system at MPP is done with the assistance of Artificial Neural Network ( ANN ) . This map deals with the more likely state of affairss for variable values of temperature and irradiance to acquire the corresponding electromotive force and current at maximal power. The mathematical PV faculty patterning depends on Schott ASE-300-DGF PV panel with the assistance of MATLAB environment. The purpose of this paper is to pick extremums of the power curves ( maximal points ) . The simulation consequences at MPP are good depicted in 3-D figures to be used as preparation or acquisition informations for the ANN theoretical account.

Keywords: Maximum power ; PV Module ; Genetic Algorithm ; Neural Network ; MATLAB.

Introduction

Photo-voltaic systems have become progressively popular and are ideally suited for distributed systems. Many authoritiess have provided the much needed inducements to advance the use of renewable energies, promoting a more decentralised attack to power bringing systems. Recent surveies show an exponential addition in the worldwide installed photovoltaic power capacity. There is ongoing research aimed at cut downing the cost and accomplishing higher efficiency. Solar energy is the universe ‘s major renewable energy beginning and is available everyplace in different measures. Photovoltaic panels do non hold any moving parts, operate mutely and bring forth no emanations. Another advantage is that solar engineering is extremely modular and can be easy scaled to supply the needed power for different tonss [ 1 ] , [ 2 ] . A important sum of fuel cell research focuses on cardinal issues of public presentation and cost [ 3-6 ] . Since the power harvested from the photovoltaic faculty is different at assorted runing points it is of import that maximal power is obtained from the photovoltaic faculty [ 7- 9 ] . A PV array is normally oversized to counterbalance for a low power output during winter months. This mismatching between a PV faculty and a burden requires farther over-sizing of the PV array and therefore increases the overall system cost. To extenuate this job, a maximal power point tracker ( MPPT ) can be used to keep the PV faculty ‘s operating point at the MPP. MPPTs can pull out more than 97 % of the PV power when decently optimized [ 7 ] , [ 10 ] . This paper ‘ computations are based on practical PV faculty informations in mention [ 11 ] .

PV Cell Model

The usage of tantamount electric circuits makes it possible to pattern features of a PV cell. The method used here is implemented in MATLAB plans for simulations. The same mold technique is besides applicable for patterning a PV faculty. There are two cardinal parametric quantities often used to qualify a PV cell. Short-changing together the terminuss of the cell, the photon generated current will follow out of the cell as a short-circuit current ( Isc ) . Therefore, Iph = Isc, when there is no connexion to the PV cell ( open-circuit ) , the photon generated current is shunted internally by the intrinsic p-n junction rectifying tube. This gives the unfastened circuit electromotive force ( Voc ) . The PV faculty or cell makers normally provide the values of these parametric quantities in their datasheets [ 11 ] . The ASE-300-DGF/50 is an industrial-grade solar power faculty built to the highest criterions. Highly powerful and dependable, the faculty delivers maximal public presentation in big systems that require higher electromotive forces, including the most ambitious conditions of military, public-service corporation and commercial installings. For superior public presentation, quality and peace of head, the ASE-300-DGF/50 is renowned as the first pick among those who recognize that non all solar faculties are created equal [ 11 ] . The simplest theoretical account of a PV cell tantamount circuit consists of an ideal current beginning in analogue with an ideal rectifying tube. The current beginning represents the current generated by photons ( frequently denoted as Iph or IL ) , and its end product is changeless under changeless temperature and changeless incident radiation of visible radiation. The PV panel is normally represented by the individual exponential theoretical account or the dual exponential theoretical account. The individual exponential theoretical account is shown in fig. 1. The current is expressed in footings of electromotive force, current and temperature as shown in equation 1 [ 12 ] .

Figure 1. Single exponential theoretical account of a PV Cell

( 1 )

Figure 2. Double exponential theoretical account of PV Cell

( 2 )

Where Iph: the exposure generated current ; Io: the dark impregnation current ; Is1: impregnation current due to diffusion ; Is2: is the impregnation current due to recombination in the infinite charge bed ; IRp: current flowing in the shunt opposition ; Rs: cell series opposition ; Rp: the cell ( shunt ) opposition ; A: the rectifying tube quality factor ; Q: the electronic charge, 1.6 i‚? 10 – 19 C ; K: the Boltzmann ‘s changeless, 1.38 i‚? 10 – 23 J/K ; and T: the ambient temperature, in Kelvin.

Eq.1 and Eq.2 are both nonlinear. Furthermore, the parametric quantities ( Iph, Is1, Is2, Rs, Rsh and A ) vary with temperature, irradiance and depend on fabrication tolerance. Numeric methods and swerve adjustment can be used to gauge [ 12 ] , [ 13 ] .

There are three cardinal runing points on the IV curve of a photovoltaic cell. They are the short circuit point, maximal power point and the unfastened circuit point. At the unfastened – circuit point on the IV curve, V = Voc and I = 0. After replacing these values in the individual exponential equation ( 1 ) the equation can be obtained [ 12 ] .

( 3 )

At the short – circuit point on the IV curve, I = Isc and V = 0. Similarly, utilizing equation ( 1 ) , we can obtain.

( 4 )

At the maximal – power point of the IV curve, we have I = Impp and V = Vmpp. We can utilize these values to obtain the followers:

( 5 )

The power transferred to the burden can be expressed as

P = IV ( 6 )

We can gauge the rectifying tube quality factor as:

( 7 )

And

Rp = Rsho ( 8 )

( 9 )

( 10 )

( 11 )

As a really good estimate, the photon generated current, which is equal to Isc, is straight relative to the irradiance, the strength of light, to PV cell [ 14 ] . Therefore, if the value, Isc, is known from the datasheet, under the standard trial status, Go=1000W/m2at the air mass ( AM ) = 1.5, so the photon generated current at any other irradiance, G ( W/m2 ) , is given by:

( 12 )

It should be notified that, in a practical PV cell, there is a series of opposition in a current way through the semiconducting material stuff, the metal grid, contacts, and current collection coach [ 15 ] . These resistive losingss are lumped together as a series obstructionist ( Rs ) . Its consequence becomes really conspicuous in a PV faculty that consists of many series-connected cells, and the value of opposition is multiplied by the figure of cells. Shunt opposition is a loss associated with a little escape of current through a resistive way in analogue with the intrinsic device [ 15 ] . This can be represented by a parallel obstructionist ( Rp ) . Its consequence is much less conspicuous in a PV faculty compared to the series opposition so it may be ignored [ 15 ] , [ 16 ] . The ideality factor denoted as A and takes the value between one and two ( as to make the nominative features ) [ 16 ] .

Photovoltaic Module Modelling

A individual PV cell produces an end product electromotive force less than 1V, therefore a figure of PV cells are connected in series to accomplish a coveted end product electromotive force. When series-connected cells are placed in a frame, it is called as a faculty. When the PV cells are wired together in series, the current end product is the same as the individual cell, but the electromotive force end product is the amount of each cell electromotive force. Besides, multiple faculties can be wired together in series or parallel to present the electromotive force and current degree needed. The group of faculties is called an array. The panel building provides protection for single cells from H2O, dust etc, as the solar cells are placed into an encapsulation of level glass. Our instance here depicts a typical connexion of 216 cells that are connected in series [ 11 ] . The scheme of patterning a PV faculty is no different from patterning a PV cell. It uses the same PV cell theoretical account. The parametric quantities are the all same, but merely a electromotive force parametric quantity ( such as the open-circuit electromotive force ) is different and must be divided by the figure of cells. An electric theoretical account with moderate complexness [ 17 ] is shown in figure 3, and provides reasonably accurate consequences. The theoretical account consists of a current beginning ( Isc ) , a rectifying tube ( D ) , and a series opposition ( Rs ) . The consequence of parallel opposition ( Rp ) is really little in a individual faculty, therefore the theoretical account does non include it. To do a better theoretical account, it besides includes temperature effects on the short-circuit current ( Isc ) and the contrary impregnation current of rectifying tube ( Io ) . It uses a individual rectifying tube with the rectifying tube ideality factor set to accomplish the best I-V curve lucifer.

Figure 3. Equivalent circuit used in the simulations

The equation ( 13 ) describes the current-voltage relationship of the PV cell.

( 13 )

Where: I is the cell current ( the same as the faculty current ) ; V is the cell electromotive force = { faculty electromotive force } ? { No. of cells in series } ; T is the cell temperature in Kelvin ( K ) .

First, cipher the short-circuit current ( Isc ) at a given cell temperature ( T ) :

( 14 )

Where: Isc at Tref is given in the datasheet ( measured under irradiance of 1000W/m2 ) , Tref is the mention temperature of PV cell in Kelvin ( K ) , normally 298K ( 25oC ) , a is the temperature coefficient of Isc in per centum alteration per grade temperature besides given in the datasheet.

The short-circuit current ( Isc ) is relative to the strength of irradiance, therefore Isc at a given irradiance ( G ) is introduced by Eq. 12.

The contrary impregnation current of rectifying tube ( Io ) at the mention temperature ( Tref ) is given by the equation ( 15 ) with the rectifying tube ideality factor added:

( 15 )

The contrary impregnation current ( Io ) is temperature dependent and the Io at a given temperature ( T ) is calculated by the undermentioned equation [ 17 ] .

( 16 )

The rectifying tube ideality factor ( A ) is unknown and must be estimated. It takes a value between one and two ; nevertheless, the more accurate value is estimated by curve suiting [ 17 ] besides, it can be estimated by attempt and mistake until accurate value achieved. Eg is the Band spread energy ( 1.12 V ( Si ) ; 1.42 ( GaAs ) ; 1.5 ( CdTe ) ; 1.75 ( formless Si ) ) .

The series opposition ( Rs ) of the PV faculty has a big impact on the incline of the I-V curve near the open-circuit electromotive force ( Voc ) , therefore the value of Rs is calculated by measuring the incline dI/dV of the I-V curve at the Voc [ 17 ] . The equation for Rs is derived by distinguishing the I-V equation and so rearranging it in footings of Rs as introduced in equation ( 17 ) .

( 17 )

Where: is the incline of the I-V curve at the Voc ( utilizing the I-V curve in the datasheet so split it by the figure of cells in series ) ; Voc is the open-circuit electromotive force of cell ( Dividing Voc in the datasheet by the figure of cells in series ) .

Finally, the equation of I-V features is solved utilizing the Newton ‘s method for rapid convergence of the reply, because the solution of current is recursive by inclusion of a series opposition in the theoretical account [ 17 ] . The Newton ‘s method is described as:

( 18 )

Where: degree Fahrenheit ‘ ( ten ) is the derived function of the map, degree Fahrenheit ( x ) = 0, xn is a present value, and xn+1 is a following value.

( 19 )

By utilizing the above equations the undermentioned end product current ( I ) is computed iteratively.

( 20 )

The undermentioned small figures at assorted faculty temperatures simulated with the MATLAB theoretical account for our PV faculty are shown with the maximal power points identified on them. After that, a more powerful tool is presented ( Genetic Algorithm ) to be used for happening the MPPs over most likely scope.

Figure 4. P-V curves at ( 1KW/m2 ; 0, 25, 50, 75oC )

Figure 5. P-V curves ( 0.75KW/m2 ; 0, 25, 50, 75oC )

Figure 6. P-V curves ( 0.50KW/m2 ; 0, 25, 50, 75oC )

Figure 7. P-V curves ( 0.25KW/m2 ; 0, 25, 50, 75oC )

From the above figures, it is clear that there is a maximal point at each irradiance value with the temperature value. The coming phase presents the familial algorithm map to place this maximal point for more broad scope than before so utilizing the installation of ANN to infer the needed thrust equation.

Familial Algorithm

The familial algorithm is a method for work outing both constrained and unconstrained optimisation jobs that is based on natural choice, the procedure that drives biological development. The familial algorithm repeatedly modifies a population of single solutions. At each measure, the familial algorithm selects persons at random from the current population to be parents and uses them to bring forth the kids for the following coevals. Over consecutive coevalss, the population “ evolves ” toward an optimum solution. We can use the familial algorithm to work out a assortment of optimisation jobs that are non good suited for standard optimisation algorithms, including jobs in which the aim map is discontinuous, non – differentiable, stochastic, or extremely nonlinear [ 18 – 23 ] .

The familial algorithm uses three chief types of regulations at each measure to make the following coevals from the current population:

aˆ? Selection regulations select the persons, called parents, which

contribute to the population at the following coevals.

aˆ? Crossover regulations combine two parents to organize kids for the

following coevals.

aˆ? Mutation regulations apply random alterations to single parents

to organize kids.

Our familial test uses the undermentioned MATLAB prescribed nomenclatures:

Population type: Double Vector with Populations size = 20 ; Creation map, Initial population, Initial Score, and Initial scope: Default ; Fitness grading: Rank ; Selection map: Stochastic unvarying Reproduction ; Elite Count: Default ( 3 ) , Crossover fraction: Default ( 0.8 ) ; Mutation map: Adaptive feasible ( due to its benefits ) ; Crossover map: Scattered Migration ; Direction: Forward, Fraction: Default ( 0.2 ) , Interval: Default ( 20 ) ; Stoping standards ( Defaults ) : Coevalss: 100, Time bound: Inf. , Fitness bound: Inf. , Stall coevalss: 50, Stall clip bound: Inf. , Function Tolerance: 1e-6, nonlinear restraint tolerance: 1e-6.

Besides, this technique is used before by the same writers in the field of green energy in [ 24-26 ] .

Maximal Power GA Function

The purpose of this map is to pick the extremums of PV power curves shown earlier ; as the nonsubjective map and out two variables as statements x ( 1 ) , and x ( 2 ) ( Vmp, and Imp ) .

This efficient map is implemented by maximising the power with the electromotive force and current as optimising variables, and with bounds for them by the values of Voc, and Isc from the PV faculty informations sheet, besides with nonlinear restraints with the assistance of Voc, and Isc obtained from I-V curves for each irradiances, and temperatures values. Both the nonsubjective map and restraint map are implemented utilizing the old mold dealingss in the signifier of MATLAB m-files.

Function MPP = degree Fahrenheit ( ten )

MPP = x ( 1 ) * x ( 2 ) ( 21 )

Function Constraints:

This optimising variable ( x ( 1 ) ) is bounded by [ 0 VocDataSheet ] .

This optimising variable ( x ( 2 ) ) is bounded by [ 0 IscDataSheet ] .

The nonlinear restraint:

Function [ degree Celsius, ceq ] = degree Fahrenheit ( ten )

degree Celsiuss = [ z1-VocModule ( For Each Irradiance & A ; Temperature Values ) ; z2 – IscModule ( For Each Irradiance & A ; Temperature Values ) ] ( 22 )

ceq = [ ]

Familial Algorithm Results

Finally, a set of 3 D figures are proposed to cover the most likely state of affairss at assorted irradiance, assorted temperature with the current, the electromotive force, and the power at the coveted maximal power. These surface faces dealingss will be considered subsequently as the acquisition or preparation informations for the ANN theoretical account. The undermentioned figures cover the most likely scope for both irradiance ( 0.05:1 kW/m2 ) and temperature ( 0:75 i‚°C ) .

Figure 8. Maximal Power relation with Irradiance and Temperature

Figure 9. Voltage at Maximum Power relation with Irradiance and Temperature

Figure 10. Current at Maximum Power relation with Irradiance and Temperature

The nervous web has the ability to cover with all old dealingss as surface or function face, due to this technique ability for insertion between points with each other and besides curves.

ANN PV GA Function with its arrested development map

This theoretical account uses the ANN technique with back-probagation technique which used, described and verified before in the field of renewable energy like in [ 27-33 ] for the same writer. This theoretical account uses the old 3D graphs illustrated before as preparation or acquisition informations for input and desired mark. The inputs in this theoretical account are the Irradiance and Temperature ; the end products are: Module Voltage, and Current at maximal Power. This theoretical account with its hidden and end product beds ‘ suited nerve cells Numberss is depicted in fig. 11. Besides, the preparation province is presented in fig. 12, and 13 severally.

Figure 11. ANN Genetic PV Module Model at Maximum power

Figure 12. Training State

Figure 13. Comparisons samples of existent and ANN-predicted values for Voltage

The arrested development nervous web map is deduced as follow:

The normalized inputs Gn: ( Normalized Irradiance ) ; Tn: ( Normalized Temperature ) are:

( 23 )

( 24 )

The undermentioned equations lead to the required derived outputs equations.

( 25 )

( 26 )

( 27 )

( 28 )

( 29 )

( 30 )

The normalized end products are:

Vn = – 0.6907 F1 + 3.0723 F2 + 2.2439 F3 + 3.2258 F4 – 123.3601 F5 + 0.4641 F6 – 3.9162 ( 31 )

In = 0.1583 F1 + 22.0215 F2 + 36.2987 F3 + 36.3581 F4 + 1.2441 F5 – 3.4176 F6 – 45.3856 ( 32 )

The un- normalized out puts

V = 17.3087 Vn + 37.2067 ( 33 )

I = 1.9128 In + 2.7979 ( 34 )

Decision

Due to the importance of PV systems particularly in green energy field, this paper introduces an efficient designation method for maximal power point ( MPP ) map for photovoltaic ( PV ) faculty utilizing the familial algorithm ( GA ) . The needed map to bring forth the mention values to drive the trailing system in the PV system at MPP is done with the assistance of Artificial Neural Network ( ANN ) . This map uses the most likely state of affairss for variable values of temperature and irradiance to acquire the corresponding electromotive force and current at maximal power. The mathematical PV faculty patterning depends on Schott ASE-300-DGF PV panel with the assistance of MATLAB environment. The purpose of this paper is to pick extremums of the power curves ( maximal points ) to do the Sun tracker works expeditiously. The simulation consequences at MPP are good depicted in 3-D figures to be used as preparation or acquisition informations for the ANN theoretical account. Finally, the ANN arrested development map for this unit is introduced to be used straight without runing the nervous theoretical account each times.

REFRENCES

[ s.n. ] , Trends in photovoltaic applications. study study of selected iea states between 1992 and 2006.

T. Markvart and L. Castaner, Practical Handbook of Photovoltaics, Fundamentals and Applications. Elsevier, 2003.

Ahluwalia, R. , Wang, X. , Lasher, S. , Sinha, J. , Yang, Y. , and Sriramulu, S. “ Performance of automotive fuel cell systems with nanostructured thin movie accelerators. ” Proceedings of the 2007 Fuel Cell Seminar and Exposition, San Antonio, TX, 2007.

Meimei Guo, Baijun Liu, Long Li, Chang Liu, LifengWang, Zhenhua Jiang, “ Preparation of sulfonated poly ( ether ether ketone ) s incorporating amino groups/epoxy rosin composite membranes and their in situ crosslinking for application in fuel cells, ” Journal of Power Sources 195 ( 2010 ) 11-20

Tauqir A. Sherazi, Michael D. Guiver, David Kingston, Shujaat Ahmad, M. Akram Kashmiri, Xinzhong Xue, “ Radiation-grafted membranes based on polythene for direct methyl alcohol fuel cells, ” Journal of Power Sources 195 ( 2010 ) 21-29

Xiu Qing Xing, KahWai Lum, Hee Joo Poh, Yan LingWu, “ Optimization of assembly clamping force per unit area on public presentation of proton-exchange membrane fuel cells, ” Journal of Power Sources 195 ( 2010 ) 62-68

K. Hussein, I. Muta, T. Hoshino, and M. Osakada, “ Maximum photovoltaic power trailing: an algorithm for quickly altering atmospheric conditions, ” Generation, Transmission and Distribution, IEE Proceedings- , vol. 142, pp. 59-64, Jan 1995.

D. Hohm and M. Ropp, “ Comparative survey of maximal power point tracking algorithms utilizing an experimental, programmable, maximal power point tracking trial bed, ” Photovoltaic Specialists Conference, 2000. Conference Record of the Twenty-Eighth IEEE, pp. 1699-1702, 2000.

D. Sera, R. Teodorescu, and P. Rodriguez, “ Pv panel theoretical account based on datasheet values, ” Industrial Electronics, 2007. ISIE 2007. IEEE International Symposium on, pp. 2392-2396, June 2007.

Hohm, D. P. & A ; M. E. Ropp “ Comparative Study of Maximum Power Point Tracking Algorithms ” Advancement in Photovoltaics: Research and Applications November 2002, page 47-62

Schott ASE-300-DGF PV panel informations sheet. Source ( Affordable Solar web site ) ; hypertext transfer protocol: //www.affordable-solar.com/admin/product_doc/Doc_pd-00-009-c_ase_300_20080328114646.pdf

Ali Keyhani, Mohammad N. Marwali, and Min Dai, “ Integration of Green and Renewable Energy in Electric Power Systems, ” Wiley, January 2010

Masters, Gilbert M. Renewable and Efficient Electric Power Systems John Wiley & A ; Sons Ltd, 2004

Messenger, Roger & A ; Jerry Ventre Photovoltaic Systems Engineering 2nd Edition CRC Press, 2003

Castaner, Luis & A ; Santiago Silvestre Modelling Photovoltaic Systems, Using PSpice John Wiley & A ; Sons Ltd, 2002

Green, Martin A. Solar Cells ; Operating Principles, Technology, and System Applications Prentice Hall Inc. , 1982

Walker, Geoff R. “ Evaluating MPPT convertor topologies utilizing a MATLAB PV theoretical account ” Australasian Universities Power Engineering Conference, AUPEC ’00, Brisbane, 2000

A. R. Conn, Gould N. I. M. , and Toint Ph. L. “ A Globally Convergent Augmented Lagrangian Algorithm for Optimization with General Constraints and Simple Bounds, ” SIAM Journal on Numerical Analysis, Volume 28, Number 2, 1991, pages 545-572.

A. R. Conn, Gould N. I. M. , and Toint Ph. L. “ A Globally Convergent Augmented Lagrangian Barrier Algorithm for Optimization with General Inequality Constraints and Simple Bounds. Mathematicss of Computation, ” Volume 66, Number 217, pages 261- 288.

D.E Goldberg, Genetic Algorithms in Search Optim. and Machine Learning. Addison Wesley ; 1989.

G. Rudolph, 1994 “ Convergence Analysis of Canonical Genetic Algorithms, ” IEEE Trans. On N.N. ; Vol. 5 ; No. 1 ; Jan. 1994, PP. 96-101.

H. Holland, “ Adaptation in Natural and Artificial Systems, ” Ann Arbor, Univ. of Michigan Press, 1995.

G. Cvetkovski, Petkovska L. , Cundev M. and Gair S. “ Mathematical theoretical account of a lasting magnet axial field synchronal motor for GA optimization, ” in Proc. ICEM’98, Vol. 2/3, 1998, pp. 1172-1177.

A. El Shahat, A. Keyhani, and H. El Shewy, “ Spacecraft Flywheel High Speed PM Synchronous Motor Design ( Classical & A ; Genetic ) ” , Journal of Theoretical and Applied Information Technology, March 2010, Vol. 13, No. 1, pp. 83-100.

Adel El Shahat, and Hamed El Shewy, “ PM Synchronous Motor Genetic Algorithm Performance Improvement for Renewable Energy Applications ” , MDGEN11, Accepted in the International Conference on Millennium Development Goals ( MDG ) : Function of ICT and other engineerings December 27 – 29, 2009 in Chennai, India.

Adel El Shahat, and Hamed El Shewy, “ PM Synchronous Motor Genetic Algorithm Performance Improvement for Green Energy Applications ” , Paper ID: X792, Accepted in 2nd International Conference on Computer and Electrical Engineering ( ICCEE 2009 ) ; Dubai, UAE, December 28 – 30, 2009.

A. El Shahat, and H. El Shewy, “ PM Synchronous Motor Control Strategies with Their Neural Network Regression Functions ” , Journal of Electrical Systems ( JES ) , Vol. 5, Issue 4, Dec. 2009.

Adel El Shahat and Hamed El Shewy, “ High Fundamental Frequency PM Synchronous Motor Design Neural Regression Function ” , Journal of Electrical Engineering, Vol. 10 / 2010 – Edition 1, Article 10.1.14.

Adel El Shahat, and Hamed El Shewy, “ High Speed Synchronous Motor Basic Sizing Neural Function for Renewable Energy Applications ” , MDGEN05, Accepted in the International Conference on Millennium Development Goals ( MDG ) : Function of ICT and other engineerings December 27 – 29, 2009 in Chennai, India.

Adel El Shahat, and Hamed El Shewy, “ High Speed PM Synchronous Motor Basic Sizing Neural Regression Function for Renewable Energy Applications ” , Paper ID: X304, Accepted in 2nd International Conference on Computer and Electrical Engineering ( ICCEE 2009 ) ; Dubai, UAE, December 28 – 30, 2009.

A. El Shahat, “ Generating Basic Sizing Design Regression Neural Function for HSPMSM in Aircraft ” EP-127, 13th International Conference on Aerospace Science & A ; Aviation Technology, May 26 – 28, 2009, ASAT 2009 – Military Technical College, Cairo, Egypt.

El Shahat, A and El Shewy, H, “ Nervous Unit for PM Synchronous Machine Performance Improvement used for Renewable Energy ” , Ref: 93, The Third Ain Shams University International Conference on Environmental Engineering ( Ascee- 3 ) , April 14-16 2009, Cairo, Egypt.

A. El Shahat, H. El Shewy, “ Nervous Unit for PM Synchronous Machine Performance Improvement used for Renewable Energy ” , Paper Ref. : 910, Global Conference on Renewable and Energy Efficiency for Desert Regions ( GCREEDER2009 ) , Amman, Jordan.