The total of 200 patterns has been calculated for optimal couple sizing (CAop, CSop) as described above. From this set 180 patterns were used for the training of the network and 20 were used as for testing and validation for the model. These data have been randomly select. The architecture that gave the best results is shown in figure 4, which has two neurons in the input layer and two neurons in the output layer. However, the number of the neurons in the hidden layer must be adjusted during the learning phase, so that the network can be trained efficiently. Developed model can be generating the optimal sizing coefficients from only the geographical coordinate. These coefficients allow calculating the PV-array area (APV) and the useful accumulator capacity (Cu). The diagram block of developed model is provides in Fig.5. Note that the input/output data are the altitude, the longitude, optimal PV capacity (Caop) and optimal storage capacity (CSop) corresponding respectively ui(k), u2(k), yi(k) and y2(k).

Results

Once a satisfactory degree of input-output mapping has been reached, the MLP-IIR network training is frozen and the set of completely is an unknown test data that w. as applied for validation. After simulation of many different structures, we found that the best performance is obtained with a one hidden layer with 8 neurons. Table 2 displays the statistical features (mean, variance and correlation coefficient) between the measured coefficients and those estimated by our model, it is found that there is no significant difference between the estimated and the measured coefficient from the statistical features point of view. The correlation coefficient obtained for the validation data set is 97.9% for CAOP and 98.9% for CSOP. In this respect, the closer to unity these values are the better the prediction accuracy.

Table 3 lists an example of the results obtained after conducting several simulation comparisons in terms of performance among different neural network structures. The performance the model significantly as the number of hidden neurons is increased until 8 neurons. At this point, adding more hidden neurons to the networks results in a slight
improvement in performance. The MLP-IIR model present good results and take less iteration compared between other neural structures. Figure 6, shows clearly there is almost a complete agreement between the measured (numerical model) and estimated coefficients by our model MLP-IIR, also a contribution with the others neural networks.

In this part, we present an example in order to illustrate how one uses this model to determine the PV-array area and useful capacity. Firstly, you were to give in input of the model the geographical coordinate of the site, which you want install the PV system. Then, from the model we obtain the Caop, and Csop, for given consumption, Eq(1) allows to calculate the APV and the Cu, the number of solar modules and batteries are determining according to unit dimension of module and the storage capacity of the battery. Table 4 shows the results obtained for some sites from the north towards the south of Algeria.

Fig.6. Comparison between measured and others estimated models

Conclusion

The objective of this work is to train the MLP-IIR model to learn the estimation and modeling of the optimal sizing coefficients of stand-alone PV system with a minimum of input data. Once trained, the MLP-IIR estimates these coefficients faster. The validation of the model was performed with unknown sizing coefficient, which the network has not seen before. The ability of the network to make acceptable estimations even in an unusual day is an advantage of the present method. It should be stressed that the training of the network required about 1 minute on a Pentium III 800MHz machine. The estimation with correlation coefficient of 98 % was obtained. This accuracy is well within the acceptable level used by design engineers. The traditional methods of sizing PV system (empirical, analytical, numerical and hybrid) allows to estimate the sizing of PV system for one given site, and requires the availability of several parameters such as the daily solar radiation data, altitude, longitude, the load, the characteristics of stand alone PV system, the inclination of the panels and to take very much computing time for estimation of optimal coefficients. On the other hand, the model that we developed allows estimating the PV-array area and the storage capacity from a minimum input data (altitude, longitude) based on the optimal sizing coefficients and does not take much time for simulation. The advantage of this model is to estimate of the PV-array area and the storage capacity in any site in Algeria particularly in isolated sites, where the global solar radiation data is not always available. Also, this presents a good result compared between other neural network architecture. The results have been obtained for Algerian meteorological data, but the methodology can be applied to any geographical area.

References

[1] D. L.Evans. "Simplified method for predicting array output”, Solar Energy, pp. 27, 55, 1980

[2] D. R Clark, S. A. Klein and W. A.Bckman. "A method for estimating the performance PV systems”. Solar Energy, 33, vol. 6, pp. 551-555, 1984

[3] S. A. Klein and W. A.Beckman. "Loss of load probabilities for stand alone PV”. Solar Energy, Eng, 39, pp. 499, 1987.

[4] L. L.Jr. Bucciarelli. "Estimating loss of load probabilities of stand alone photovoltaic solar energy system”. Solar Energy, 32, pp. 205-211,1984

[5] H. L. Macomber, J. B. Ruzek and Staff of Bird Engineering. "Stand-alone photovoltaic system. Preliminary engineering, Design handbook”, NAZA, CR165352, NASA Lewis Research Center 1981.

[6] Q. Zhang and A. Benveniste,” Wavelet Network,” IEEE Trans Neural Networks, vol.3, n 6,pp. 889-898, 1992

[7] Y. pati and P. Krishnaprasad, " Analysis and synthesis of feed-forward neural networks using discrete affined Wave Trans-formation” IEEE Trans Neural Networks, vol. 4, n 1, pp. 73-85, 1993

[8] A. Mellit, M. Benghanem, A. Hadj Arab, and A Guessoum, "Modelling of sizing the photovoltaic system parameters using artificial neural network”. Proc. of IEEE, CCA, Istanbul vol 1, pp. 353-357, 2003

[9] A. Mellit, M. Benghanem, A. Hadj Arab, and A Guessoum, "Novel technique of sizing the stand-alone photovoltaic systems using the radial basis function neural networks: application in isolated sites”. Improving Building System in Hot and Humid Climate, May, 17-19-2004 renaissances Dallas-Richardson Hotel Texas (Accepted paper)

[10] M. Egido and Lorenzo. The sizing of sand-alone PV systems: A Review and a proposed new method. Solar Energy Materials and Solar Cells, vol. 26, pp. 51-69, 1992.

[11]A. Hadj Arab, B. Ait Driss, R. Amimeur and E. Lorenzo "Photovoltaic System Sizing in Algeria”. Solar Energy, 54, pp 99-104, 1995

[12] M. Benghanem,” An optimal sizing method for stand-alone photovoltaic system for Algeria”. World Renewable Energy Congress IV, Italy 2002; on (CD-ROM)

[13]M. Benghanem, and A. Maafi, "Data acquisition system for photovoltaic systems performance monitoring”. IEEE Trans. on Instru. and Meas., vol. 1, pp. 30-33, 1998.

[14]R. J Aguiar, M. Collares-Perreira and S. P. Conde "Simple procedure for generating of daily radiation values using library of Markov Transition Matrices”. Solar Energy, 49, pp. 229-279, 1988

[15]S. A. Imhoff, D. Y. Rocum, and M. R.Rosick.” New classes of frame wavelets for applications,” Wavelet Application H, H, H. Szu, Ed., Proc. SPIE, vol. 2491,pp. 923­34,1995

[16]S. Haykin,. Neural Networks, A comprehensive foundation, Ed. Macmillan, New York, 1994

[17]X. Ye and N. K Loh., "Dynamic system identification using recurrent radial basis function network”, Proc. of American Control Conference, 3, pp. 2912, 1993

[18] H. H. Szu, B. A. Telfer, and S. Kadambe,” Neural Network Adaptive Wavelets for signal presentation and Classification”, Optimal Engineering, vol. 31, n 9, pp. 1907­1916, 1992

RNTs PKh obtained the average radiation power values for two operational substances using PDISPL model, which are presented in tables 1,2 together with the calculated data. The discrepancy between the calculated and the experimental data is due to both the non-considered in the calculation process inhomogeneity of the operational substance in the laser cell and de-adjustment of resonator due to heating of the whole construction resulting in the interrupted generation.

In the course of the experiments implementing pumping by xenon lamps it was stated that big gradients of the temperature field at the area of laser cell become a serious obstacle for stable laser radiation.

To avoid this the alternative pumping method was studied on the basis of mercury arc lamps featuring high efficiency (about 50%) within the pumping band of the active medium.

The cell having 1,4 cm interior diameter and 67 cm length between the inlet and outlet of the operating gas was used. The cell windows are located at Bruster angle in parallel planes. The pumping was done by 8 low pressure mercury arc lamps. The illuminating element is presented in fig. 5.

Resonator is formed by flat mirrors with reflecting ratios 99,8% and 95%.

The total power of pumping lamps at the resonance mercury line wavelength of 254 nm is not less than 160 W. The effective pumping surface (fig. 6) is about ~270 cm2. The calculated photo-dissociation ratio у is within 0.16-0.18sec-1.

The following results were obtained.

The velocity of gas flux along the cell when one cryogenic pump is used, measured at the absence of generation was 1020 cm/sec. For CF3I composition the generation power of 180 mW was obtained. After activation of the second cryogenic pump (gas velocity was not

specified) the generation power increased up to 245 mW. The optimum generation pressure was 25 mmHg.

Overlightning can be seen at the edges of the display caused by the pumping lamps. Registration sensitivity decreases from periphery to the center.

Relative generation power was specified for C3F7I. The mirror with 2% transmittance and ~ 6 m radius was used as the outlet one. Generation power was 50% of the power recorded for CF3I.

Thus, mercury lamps used as the alternative pumping source offer stable generation, parameters of which provide for applicability of the model for follow-up applied research.

As already indicated, the photovoltaic market is absolutely dominated by silicon. Only a tiny fraction (0.4%) corresponds to non-silicon materials, such as CdTe, ClGS, GaAs, etc. (see Fig. 2). In the last few years, poly-silicon has been gaining share at the expense of monocrystalline silicon. Both approaches altogether, taking 84% of the market, can be

called wafer technologies. The 16% remaining is basically shared by amorphous silicon (both for indoor and outdoor applications), ribbon silicon and silicon heterojunctions.

It is quite clear that p-i-n amorphous silicon cells, which once reached nearly 30% of the PV market, have been loosing share not only in favour of wafer-based devices, but also, and very particularly in favour of silicon heterojunctions, who have conquered 6% of the total PV market in very short time. Very remarkable is the fact that this whole share corresponds to a single company: Sanyo (Japan). The group, who named their cells HIT® (Heterojunction with Intrinsic Thin layer) achieved already in 1994 20.0% efficiency in a

1- cm2 cell of this kind, and have reached 17.3% for 100-cm2 cells at industrial level and 15.2% for production modules4. These figures are unambiguously indicative of the enormous potential of SHJ technology.

In contrast to the well-defined and well-funded Japanese PV programme, the situation in Europe is characterised by a strong diversity in national programmes, some of which are individually strong, but with a high degree of fragmentation. On the other hand, a successful European RTD strategy imposes the collaboration of a large number of public and private institutions. In the 6th Framework Programme (FP6), positive actions have been launched to increase collaboration and co-ordination. A number of research groups and companies are working hard on SHJ cells in Europe. Some projects are now running under FP5 based on hybrid technologies and low-temperature formation of the junction as a main issue, such as MOPHET, METEOR and ADVOCATE7. Their individual results are promising and reveal an excellent scientific level. The need to address the fragmentation of European R&D in this field must however be recognised by creating a permanent structure ensuring the harmonisation of the whole R&D on SHJ cells. Here are just some non-exhaustive highlights of the ongoing activity of European groups involved in SHJ-cell research, with indication of some of their main lines of action (given in alphabetical order of organisations):

v CNR-IMIP (I): PECVD deposition of a-Si, pc-Si, SiC alloys; Er-doped silicon thin films; fully pc-Si thin films with very low H-content even at low temperatures; growth of pc-Si on plastic substrates; optimisation of c-Si/pc-Si heterojunction solar cells (p=15%)

ч/ CNR-IMM (I): SHJ cells using single and double microcrystalline / amorphous emitters (Л=14%).

* ECN(NL): Processing of inorganic thin-film cells: development of methods to grow silicon films by plasma sprayed silicon on ceramics. Advanced surface and bulk passivation and silicon nitride anti-reflection. Development of special ceramic substrates for film silicon applications and "adapted” solar cell concepts.

* Fh-ISE (D): c-Si thin-film solar cells (high-temperature approach) on low-cost silicon and ceramics. Methods: Si-APCVD (using SiHCl3 at ~1200°C), zone melting recrystallisation Reference cells: one-side contacting (Si thickness ~45pm) q=19.2%, direct epi (37pm) on p+-Cz: 17.6%,: Activity foreseen for the future: Transfer of results on reference cells to "realistic” substrates, scaling of the substrates / cells to 10x10cm2. Bringing c-Si thin-film to a state where pilot production can begin.

* HMI (D): Deposition of amorphous and thin-film crystalline-silicon solar cells by PECVD and ECR-CVD; defect spectroscopy by optical, electrical and electron-spin resonance methods; seed-layer approach for a poly-Si thin solar cells on glass:metal-induced crystallisation techniques for seed-layer formation, epitaxial growth a T<600°C by ion — assisted deposition techniques, low-temperature emitter technology. Solar cells using heterostructures such as a-Si:H / c-Si. Results: q=17.1% for a-Si:H(n)/c-Si (p) cells using FZ — cSi wafer. Hall-mobility measurements, in-situ lifetime measurement during plasma process.

* LPICM (F): study of plasma processes and growth of thin films through the use of in­house in situ diagnostic techniques such as UV-visible ellipsometry; Kelvin probe and Time Resolved Microwave Conductivity; stable single-junction p-i-n solar cells based on polymorphous silicon with efficiencies close to 10%; simple dry process to passivate silicon wafers with a quality comparable to that achieved by using standard HF cleaning procedures.

* TU-Delft (NL): microcrystalline, protocrystalline Si and a-SiGe:H by PECVD and Expanding Thermal (Plasma ETP-CVD CVD; studies of defect-state distribution in a — Si:H and pc-Si:H, PECVD; single — junction a-Si:H solar cells ninit=10.3% (no back reflector); tandem a-Si:H/a-SiGe:H solar cell r|init=8.7% (no back reflector); Rd(i)=0.1-

0. 2 nm/s; ETP CVD: single junction a-Si:H solar cell ninit=6.7% (no back reflector), Rd(i)=0.85 nm/s.

* Univ. Barcelona (E): HWCVD microcrystalline p-type emitters, HWCVD microcrystalline cells at low temperature (<200°C). Plans for next future: development of HIT structures with HWCVD-deposited emitters.

* Univ. Ljubljana (SL) and Rome (I) "La Sapienza”: modelling and simulation together with characterisation of thin-film semiconductor materials and electronic devices; simulator with electrical model for heterostructures and with optical model for multilayer structures with smooth or rough interfaces. Particularly remarkable are the activities done by Rome University in collaboration with ENEA by using chromium silicide as the activated layer in high-conductivity emitters in a-Si / c-Si HJ. In particular an efficiency of 17% was obtained on a CZ Si based device of 2.25cm 2 total area.

* Univ. Patras — Plasma Technology Laboratory (GR): Process development, Scale-up, plasma modelling, characterization and control for high rate deposition of amorphous and microcrystalline silicon.

The institutions of the authors of the present paper, who hold a mutual collaboration on

several projects, and are deeply involved in SHJ, do in turn develop the following activity:

v CIEMAT DER (E): applications of deposited silicon (amorphous, microcrystalline and hybrid silicon). PECVD of amorphous silicon at high growth rates. Dry etching and passivation of c-Si surfaces. Wide-bandgap, highly conductive emitters. Development of silicon p-i-n and HJ cells and position sensors.

v ENEA (I): CR Portici: amorphous silicon solar cells using a wide range of technological options (Si, SiN, SiC, SiGe, and a-Si multijunctions, Hot-Wire CVD and VHF-PECVD), laser-scribing full technological process for large-area modules, seed layer by standard LPCVD for epitaxial growth of thick polysilicon films, Solid-Phase Crystallisation (SPC), and Laser-Induced Crystallisation (LIC), development of processes suitable for industrial applications (screen printing, TCO), passivation and dry / wet conditioning of silicon surfaces for a-Si / c-Si HJ technology (n=17% 2,25 cm[1] [2] [3] on CZ Si based devices in collaboration with Rome Univ.), and in collaboration with CR ENEA Casaccia PV lab, Laser doping and screen printed contacts in cSi Technology.

CONCLUSIONS

Photovoltaic research is looking for breakthroughs which can make solar cells competitive as soon as possible. Silicon absolutely dominates the market and will continue to be in a preferential position for a long time. The evolution of wafer technology largely depends on our capacity to develop new cheaper and thinner silicon, whereas thin-film silicon research is focused on improving material quality and crystallinity. In this scenario, a novel, hybrid approach, that of silicon-heterojunction cells, has already demonstrated its capacity to introduce an important milestone in the search of new approaches for photovoltaics. This is possible thanks to the combination of the best of two worlds: the good carrier lifetimes of wafer and ribbon silicon, and the versatility, accuracy and good passivation properties of thin films. The best industrial PV modules so far are silicon-heterojunction devices. This technology has already conquered 6% of the global PV market in a very short time. An important effort should me made in Europe to support and co-ordinate the excellent work done by the European groups involved in this research. This is the only way to quickly fill the present gap with Japan.

ACKNOWLEDGEMENTS

The authors would like to thank for the collaboration, the information received or the discussions held with: J. Andreu & J. Bertomeu (Univ. Barcelona, E), G. Bruno &

M. Losurdo (CNR-IMIP, I), W. Fuhs, S. Gall & F. Wunsch (HMI, D), D. Mataras (Uni-Patras, GR), F. Palma & G. De Cesare (Rome University, I), S. Reber (FhISE, D), P. Roca i Cabarrocas (PICM, F), F. Zignani & C. Summonte (CnR-IMM, I), M. Topic (Univ. Ljubljana), M. Zeman (TU Delft, NL), W. J. Soppe (ECN, NL), and last but not least, J. J.Gandia (CIEMAT, E), M. Tucci & E. Bobeico (ENEA, I). The financial contribution to silicon — heterojunction research by the European Commission by way of the Mophet project is also acknowleded.

The data acquisition has begun since October 2003; the observations consent to determine:

—

the efficiency of the PV system

— the efficiency of the

electrical generator set: 1 liter of gasoline produces

1,2 kWh electrical energy on average;

— the dispersed energy for

charge and discharge of the battery pack:

approximately 25 % of the total produced electrical energy.

On the basis of these experimental data and knowing the monthly average solar irradiation and the

monthly average electrical Monthly energy flows experimental data

loads, we can preview with

good confidence the electrical energy that has to be produced by electrical generator set and the fuel consumption month by month during the year.

These distribution diagram allows us to recalculate the optimum size of the photovoltaic field as a function also of the cost of the fuel. Maintaining constant the electrical load but

varying the power of the photovoltaic field, the annual energy the electrical generator have to produce has been calculated and, as a consequence, the annual cost of the fuel has been determined. The optimal size of the photovoltaic field is the one that determine the minimum total annual cost of the system, addition of annual fuel cost and annual PV field cost.

In the two following diagrams the annual total

Monthly energy flows calculated data cost of the system is

reported as a function of the photovoltaic field size

using as a fuel, respectively, gasoline or Liquefied Petroleum Gas (LPG).

The annual PV field cost has been evaluated supposing an effective life of 25 years and a depreciation rate of 3%; moreover in this period three replacements of the battery pack have been considered. Fuel costs have been determined taking into account the cost of a liter of gasoline equal to 1,05€ and the cost of a liter of LPG equal to 0,55€; moreover with the use of the LPG a diminution of the electrical generator efficiency of approximately 15% has been considered. The optimal size of the photovoltaic field outcomes to be 2300 Wp utilising gasoline and 1900 Wp utilising LPG.

1. Conclusions

All costs of three kinds of PV system for domestic user in stand-alone are reported in the following table:

1. classic PV system

2. hybrid PV-Fuel system, gasoline feeded

3. hybrid PV-Fuel system, LPG feeded.

Using an hybrid PV-Fuel system, an investment cost reduction of approximately 28% is possible by supplying the electrical generator with gasoline, and of approximately 35% in the case of LPG supply. The corresponding cost

reductions of electrical kWh are, respectively of approximately 23%, with gasoline supply and of 28% with LPG supply, confirming in such a way the objective of this project.

Moreover using an electrical generator of back up an increase of the reliability and the continuity of electrical energy service has been obtained.

A second experimental activity will follow in which the Sunny Island will work as a normal inverter connected to the grid but able, in case of main voltage interruption, to behave as a back-up source, guaranteeing to the loads need it a high continuity of electrical energy service.

4. References

— Francesco P. Califano, Vittorio Silvestrini, Gianfranco Vitale. The planning of PV Systems, Publisher Liguori, Napoli 1988.

— Francesco Paolo Vivoli. Electric power from the sun, ISES Italy-ENEA, 1998

— G. Cramer, J. Reekers, M. Rother, M. Wollny. The future of village electrification — More than two years of experiences with AC-Coupled Hybrid Systems -, SMA Regelsysteme GmbH — Niestetal

Sioe Yao Kan, Sjoerd van Beers and Han Brezet

Delft University of Technology Faculty of Industrial Design Engineering Section Design for Sustainability program Landbergstraat 15, 2628 CE Delft, the Netherlands 31 (0)15 278 2956,

S. Y.Kan@IO. TUDelft. nl.. S. vanBeers@IO. TUDelft. nl , J. C.Brezet@IO. TUDelft. nl Keywords:

Mature Design, Photovoltaic (PV), Curved PV Surfaces, Coloured PV modules, Sustainable Products, Product Appearance, Synergy

Abstract

The appearance and the actual convenience of photovoltaic (PV) powered products is often not properly considered in sustainable product design. We undertook an endeavour to identify and chart the main factors that need to be improved in order to reduce those neglects and to stimulate a broader use of PV cells in products. Moreover this paper will stir up the discussion on mature design quality of sustainable products. The word ‘mature’ can for instance mean timeless design. Some solutions and synergetic proper novel technologies are proposed for achieving a sound integration of PV cells into products. An introduction of mature design by optimizing ergonomical aspects, appearance and product — user context relevance is given. In particular the emphasis will be on good matching of the overall design with the PV cell characteristics. The uses of curved coloured PV surfaces will be highlighted. Although the appearance of PV powered products obviously contributes towards the diffusion of such products, it still remains a point of discussion what kind of appearance is decisive in this process.

1. Introduction

Today there are quite a lot ‘PV Powered’ products, in combination with some kind of energy storage media [e. g. CET SOLAR Products and Photon Technologies; 2004]. In these PV powered products however, quite often the PV cells are just add-on units to give the product a ‘green image’. Because of this ‘just add-on’ approach, the PV cells remain foreign bodies not well-integrated into the total product design. In other words, a sub-optimal matching between the PV cells characteristics and the product user contexts. To circumvent this sub­optimal matching, in particular in designing PV powered products with aesthetics and ergonomics in mind, one should be aware of the following observations:

• Appearance and user context have become a selling point. The products have to be practical and state of the art, it has to be a product of the future, not just an ad-hoc prototype. As a direct result, people will be inclined to buy a product because it looks appealing and it is actually useful and meaningful and clearly demonstrates its added value. Not only a funny gadget.

• Mobile and wireless products have become almost a necessity in modern daily life (e. g. mobile phones, mp3 players, PDAs, notebooks). As a side effect, working with appealing products has proven to be a productivity improvement factor. More products have become equal in functionality and price but distinctive in design and image.

• Matching between product perception and user contexts becomes an issue.

An argument stated at Sustainable Product Design conferences [e. g. Sustainable Innovation Conference; 2003] is that: ‘the way to achieve sustainable innovation is by designing cool and sexy products’. This statement would overemphasize the need for trendy appearance
of products, especially in view of the actual contradiction between designing trendy, cool and sexy products and designing for sustainability. Most products that are today trendy will be outdated tomorrow ergo not sustainable.

To establish a baseline for sustainable design of PV powered products, in a recent article [Kan et al. 2004] the proper design methods of mobile and wireless products were analyzed by exploiting the possible synergy that could be gained between the links of their energy chains. The combining of all the links in an optimal manner, including the positive side effects outside the energy chain, is called SYN-Energy. In that article these side effects were mentioned to be aesthetics and ergonomics. In this paper these side effects will be analysed further by introducing the concept of ‘Mature Design of Sustainable PV powered Product’.

Mature design of sustainable products will not only distinguish itself by its appearance, convenience of use and its functionality but will also take into account the environment for the long term by being ‘timeless’. To accommodate these mature designs, some novel technologies are presented. In addition this paper will stir up the discussion on the proper appearance of sustainable products and plea for more international standardization and compatibility of interface elements such as batteries, voltages and connectors. The latter in view of reducing cost, waste and enhance interchange ability and convenience of use of electrical products in general and PV powered product in particular.