Inter national J our nal of Electrical and Computer Engineering (IJECE) V ol. 8, No. 4, August 2018, pp. 2029 2037 ISSN: 2088-8708 2029       I ns t it u t e  o f  A d v a nce d  Eng ine e r i ng  a nd  S cie nce   w     w     w       i                       l       c       m     Study and Dimensioning of the T anks Dedicated to a Compr essed Air Ener gy Storage system (CAES) I. Rais 1 and H. Mahmoudi 2 1 Mohamadia Engineering School, Mohammed V uni v ersity , Morocco 2 Po wer Electronic and Control T eam (EPCT), Department of Electrical Engineering, Morocco Article Inf o Article history: Recei v ed: October 2, 2017 Re vised: April 14, 2018 Accepted: May 3, 2018 K eyw ord: CAES CAM Compressor Reserv oir ener gy density char ging time dischar ging time ABSTRA CT The fundamental idea of storage is to transfer a v ailable ener gy During periods of lo w demand , using only a fraction of the fuel that w ould be consumed by the standard production machine (g as turbine, thermal engine, etc.). The main role of ener gy storage is therefore to introduce an ener gy de gree of freedom to decouple Consumers and the producer by supplying or Deli v ering the dif ference between these tw o po wers. In this paper is this paper presents a brief study and dimensioning of compressed air storage tanks to a h ybrid system wind-PV . adopts the CAES system as a storage agent. starting with the technical criteria on which the choice of reserv oirs is based and the mechanical constraints that must be tak en into consideration for dimensioning of the reserv oirs Copyright c 2018 Institute of Advanced Engineering and Science . All rights r eserved. Corresponding A uthor: Ilham Rais Po wer Electronic and Control T eam (EPCT) Rabat, Morocco Email: ilhamrais@research.emi.ac.ma 1. INTR ODUCTION In most isolated areas, the diesel generator is the m ain source of electrical po wer .that poses immense technical challenges And financial. This generation of electricity is relati v ely inef ficient, v ery costly and respon- sible for the emission of lar ge quantities of greenhouse g ases (GHGs). The use of wind-solar (JES) twinning in these autonomous netw orks could Thus reducing operating deficits [5][7]. Ho we v er , the profitability of the JES is achie v ed on the condition of obt aining a high penetration rate of wind and/or solar ener gy , which is possible only by using storage systems [8] the solution proposed which meets all the technical and financial requirements while ensuring a reliability of electricity supply to the isolated sites. It is the h ybrid wind-photo v oltaic system with storage of compressed air [1]. The use of compressed air as an ener gy storage agent i s as well suited to wind and solar production as it is to di esels. The principal idea for the storage of electrical ener gy By wind turbine plant and Photo v oltaic array as a source of ener gy coupled with tw o Pneumatic machines: The first is a compressor dri v en by an electric motor and the second i s an compressed air motor which dri v es in turn an alternator [1]. During windy period the turbine directly supplies the isolated site on electricity and Surplus ener gy will be used by the electri c motor to dri v e the compressor to rechar ge the compressed air in the tri vial tanks. In the absence of wind turbine the photo v oltaics ener gy will follo w the same principle. In the absence of the tw o ener gy sources compressed, the air will be relax ed in the compressed air motor which will dri v e in turn the generator to supply electricity at the isolated site . This h ybrid system w ould act in real time in order to maintain the balance between the po wer generated and consumed by achie ving a remarkable reduction in fuel consumption whate v er the le v el of the po wer demanded. the follo wing paper will present a brief study of the selection criteria of the reserv oirs and the dimensioning of this latters as the most interesting elements in the compression chain [1][12]. J ournal Homepage: http://iaescor e .com/journals/inde x.php/IJECE       I ns t it u t e  o f  A d v a nce d  Eng ine e r i ng  a nd  S cie nce   w     w     w       i                       l       c       m     DOI:  10.11591/ijece.v8i4.pp2029-2037 Evaluation Warning : The document was created with Spire.PDF for Python.
2030 ISSN: 2088-8708 2. TECHNICAL CHARA CTERISTIC FOR THE CHOICE OF RESER V OIRS The choice of tanks suitable for the proposed compressed air storage system does not come at random b ut there are se v eral technical criteria which characterize them we cite among them [4][2]: ener gy density of storage The fle xibility of location and specific site requirements. Storage capacity . The self-dischar ge. The yield of the storage system 2.1. Ener gy density of storage An open g as c ycle system allo ws complete polytropic e xpansion of air compressed from the maximum pressure to atmospheric pressure. this allo ws full e xploitation of the ener gy stored as compressed air . F or a 1 m3 unit of v olume, the storage ener gy density can be e xpressed by the follo wing equation [3]: W = K nN P r n 1 (1 ( P a P r ) n 1 nN ) (1) with K = 2 : 777810 6 is the ener gy con v ersion cons tant in kWh, N is the number of e xpansion stages of the CAM, P a is the atmospheric pressure and P r is the storage pressure. The figure abo v e gi v es an idea of the amount of 0 50 100 150 200 250 0 0.5 1 1.5 2 2.5 3 3.5 x 10 −3 maximum storage pressure (bar) Energy density (wh / m    3)     N=1 N=2 N=3 N=4 N=5 Figure 1. v ariation of the ener gy density as a function of the number of stage and the storage pressure ener gy stored in a gi v en v olume of air . It can be noticed that by increasing the maximum allo w able pressure in the tank this quantity can be increased. Thus, the higher the numbe r of st ages of the air e xpansion in the CAM increases, the more ener gy density per m 3 increases and consequently the mechanical w ork (electrical idem) de v elops closer to its maximum v alue. In practice, the compressed air e xpansion process must be stopped once the pressure in the tank drops belo w the minimum pressure, P r m , required for the operation of the system. Indeed, belo w this pressure limit, the po wer deli v ered becomes so lo w that the operation of the system becomes inef fecti v e. The v alue of P r m depends on the nature of the application. Therefore, in the case where P r m is greater than atmospheric pressure P a , une xploited ener gy , W unex , will remain in the air reserv oir . this ener gy can be e xpressed as follo ws: W unex = K nN P r m n 1 (1 ( P a P r m ) n 1 nN ) (2) Consequently , the ef fecti v e ener gy density W ef will be reduced and equal to the dif ference between the total a v ailable ener gy density W and the unused ener gy density W unex , hence: W ef = W W unex (3) IJECE V ol. 8, No. 4, August 2018: 2029 2037 Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE ISSN: 2088-8708 2031 The Figure 2 sho ws the ef fecti v e ener gy density as a function of the pressure minimum of operation, P r m , for dif ferent v alues of the maximum pressure P r . It is that the lo wer the minimum pressure, the higher the ef fecti v e ener gy density . 0 50 100 1 5 0 2 0 0 2 5 0 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 Minimum  oper a t ing   pr e ssu r e   (bar ) E!ective energy density(KWh/ m 3 )     P=   5 0   bar P= 1 0 0  bar P= 1 5 0  bar P= 2 0 0 bar P= 2 5 0  bar P= 3 0 0 bar Figure 2. influence of minimum operating pressure on the density of ener gy F or a good e v aluation of the unused ener gy , the pressure utilization f actor (PUF) for an open g as c ycle can be defined as: P U F = 1 W unex W (4) It is easy to deduce from this equation that PUF = 1 if P r m = P a and PUF = 0 if P r m = P . The Figure 3 sho ws the v ariations of PUF as a function of the minimum storage pressure for a relaxation which is carried out in 5 step . 0 50 100 150 200 250 300 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Minimum operating pressure (bar) Pressure utilization factor     P=50bar P=100bar P=150bar P=200bar P=250bar P=300bar Figure 3. V ariation of PUF as a function of the minimum operating pressure 3. MECHANICAL CONSTRAINTS Currently a v ailable materials can store a high density up to pressures of the order of 500 bar; for the proposed study we wil l considered a spherical tanks, relati v ely slim whose pressure inside has a small thickness is supposed constant, because the storage pressure P r depends on the maximum breaking stress, , and of t he dimensions of the reserv oir , the e xpression that connects these parameters is the follo wing [9][6]: e r D r = P r 4 (5) W ith: e r and D r are, respecti v ely , the thickness and the diameter of the storage tank. kno wing that R int = D r 2 is the Study and Dimensioning of the T anks Dedicated to a Compr essed Air ... (I. Rais) Evaluation Warning : The document was created with Spire.PDF for Python.
2032 ISSN: 2088-8708 internal radius of the reserv oir , R ext is the outer radius of the reserv oir of storage and e r = R ext R int , the v olumes of stored air and the reserv oir are respecti v ely : V = 4 3 R 3 int (6) V r = 4 3 R 3 ext (7) The v olume of the material used in the manuf acture of the tank is then: V matr = V V r = 4 3 ( R ext R int )( R 2 ext + R ext R int + R 2 int ) (8) The h ypothesis that the reserv oir has a small t hickness mak es it possible to simplify the second term of the pre vious equation in 3 R 2 int and the e xpression of V matr becomes as follo ws: V matr = 4 3 e r (3 R 2 int ) (9) From the abo v e equations, a relation between the v olume of air stored, V , as wel l as the v olume of materials used in the manuf acture of the tank, V matr , can be deduced, hence: V matr V = 4 3 e r (3 R 2 int ) 4 3 ( R 3 int ) = 3 e r R int = 6 e r D r = 3 P r 2 (10) The relati onship between the mass of the ma terials of a spherical reserv oir , M matr , and the ener gy stored, E, can then be e xpressed by the follo wing relation M matr E st = V matr n ( n 1) N (1 ( P P a ) 1 n nN ) = 3( n 1) 2 N nN n (1 ( P P a ) 1 n nN ) (11) is the o v erall ef ficienc y of the con v ersion chain between the PV -wind h ybrid system and the storage tank and E st is the stored ener gy which can be defined by: E st = V W = E (12) The follo wing figure sho ws ho w the stored ener gy can v ary as a function of the compression ratio (equal to the storage pres sure) and the properties of the dif ferent materials. The ratio (K= ) is the ra tio of the tensile st rength to the density of the material. Figure 4. ratio (reserv oir mass / stored ener gy) as a function of the compression ratio for dif ferent properties of the materials It is easy to notice that the stored ener gy is higher the higher the compression ratio. The best performance (maximum stored ener gy and minimum reserv oir mass) is obtained with the highest ratio (K= ), for high tensile strength b ut relati v ely lo w density . IJECE V ol. 8, No. 4, August 2018: 2029 2037 Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE ISSN: 2088-8708 2033 4. DIMENSIONING OF THE T ANKS The dimensioning of reserv oirs is a v ery important point in this study . The y should not be too small or too lar ge to ensure that e xcess ener gy can be stored and to limit congestion and engender a disproportionate in v estment. The sizing of the reserv oirs is conditioned by the double study of the flo w (v ariable) of the compressor (and therefore of the v olume that can be stored in a gi v en time) and the compressed air flo w required by the CAM to dri v e the associated alternator and supply the isolated site with electricity [2][10]. Then to calculate the v olume of the tanks tw o methods will be presented 4.1. First method The first, a con v entional method, is a function of: the maximum pressures, P max , and minimal, P min , allo w able by the CAM, the desired autonomy a, maximum air flo w required to po wer the compressed air motor , . The v olume will be calculated using the follo wing formula: V = P a a P max P min (13) 4.2. second method The second method for dimensioning the storage v olume is to calculate the loading and dischar ge time of the tanks, gi v en that the amount of air injecte d into the tank is v ariable and is a function of the po wer of the h ybrid system absorbed by the compressor and of the air consumption of the CAM [11][15]. 4.2.1. Char ging time T o f acilitate calculation, it is considered that the po wer absorbed to compress the air is constant. The instantaneous air flo w rate can then be e xpressed as follo ws: " = p c E c (14) W ith: P c is the po wer consumed by the compressor . E c the ener gy per unit mass necessary to compress the air at a gi v en pressure. On the other hand, by ne glect ing the losses by leakage of compressed air , the equation of conserv ation of the mass and the la w of perfect g ases mak e it possible to e xpress the flo w of air entering the v olume V r of the storage tank as follo ws: " = dm dt = V r T r dp dt (15) By inte grating the equation obtained from the preceding equations after ha ving replaced each term ( P c and E c ) by its v alue, the char ge time of a compressed air reserv oir can be calculated from the follo wing equation t ch = ( ( ch + 1) 1 + ch ) N ch (16) W ith : ch = n 1 N n , = P P a and is the time constant during the char ging phase, defined by: = P a V P c C p r (17) V is the v olume of compressed air produced, C p is the mass heat of the compressed air . From the equation (17), The equation of the time constant Becomes: = t ch N ( ( ch +1) 1+ ch ) (18) By replacing the v alue of in the equation (16), the e xpression of the product v olume of compressed air becomes : V = P c r t ch P a C p N ( ( ch +1) 1+ ch ) (19) Study and Dimensioning of the T anks Dedicated to a Compr essed Air ... (I. Rais) Evaluation Warning : The document was created with Spire.PDF for Python.
2034 ISSN: 2088-8708 4.2.2. Dischar ging time The dischar ge time of a compressed air reserv oir may be calculated from the same the char ging time, replacing " , P and E respecti v ely with the parameters of the compressed air motor [13]. " M , the mass flo w of compressed air consumed by the MA C. P M , the po wer pro vided by the MA C. E M , the ener gy resulting from the e xpansion of the compressed air in the MA C. The e xpression obtained from the dischar ge time is then written as follo ws [16][14]. t dch = ( 1+ dch M 1 + dch + dch 1 + dch M ) N dch (20) W ith: dch = n 1 nN , M = P M in P M ou ;N is the number of e xpansion stages in the CAM and dch is the time constant during the dischar ge phase, defined by the follo wing e xpression dch = P a V r P M C p r (21) V r is the v olume of compressed air tank, P M is the po wer produced by the compressed air motor . The dischar ge time between 2 pressure le v els can be calculated, as follo ws : t dch P 1 P 2 = t dch P 1 t dch P 2 = [ ( 1+ dch P 1 1+ dch P 2 ) 1 + dch + ( P 2 P 1 )] N dch (22) 5. RESUL T AND DISCUSSION 5.1. The first method Figure 8 gi v es an idea of the dimensioning of the tank calculated from the equation 13 It sho ws that o v er the desired range is greater the greater the v olume of air stored in the reserv oir must be lar ge. Thus, for an autonomy of 2 days, the v olume necessary to store compressed air at 300 bar , will be of the order of 34 m3. On the other hand, this v olume is enormous and o v ersized and it will be dif ficult to transport, to install a tank ha ving this v olume in an isolated sit e. In addition, the MA C will rarely operate at maximum flo w when tank pressure is too lo w . Thi s results from the f act that the tank will be rechar ged, once the pressure drops, using the e xcess ener gy that is a v ailable in this site in a f airly re gular manner . Therefore, this method of sizing the reserv oir will not be adopted for the rest of the calculation. 50 100 150 200 250 300 0 50 100 150 200 250 300 350 400 450 maximum pression of storage (bar) reservoir vomume (m 3 )     a=1jours a= 2jours a=3jours a=4jours Figure 5. ratio (v ariation of the tank v olume as a function of the maximum storage pressure and the autonomy IJECE V ol. 8, No. 4, August 2018: 2029 2037 Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE ISSN: 2088-8708 2035 5.2. The second method 5.2.1. Char ging time Figure 6 and Figure 7 sho w the char ging time of a tank of 300 L v olume, respecti v ely , as a function of the number of stages of a compressor with 5 kW of po wer and the po wer absorbed by one compressor ha ving 5 compression stages. These figures sho w that the more the number of compressor stages increases or the more the electric po wer consumed to compress the air increases, the f aster the char ging time decreases. A single-stage compressor can fill the 300 L v olume in 3.5 hours, while approximately 2 hours will be suf ficient to fill the same v olume if the compression of air is done bay 5 stages. A time sa ving of about 43pc. Thus, with a po wer of 20 kW , it will tak e half an hour to fill a tank of 300 L whereas approximately 2 hours are required to fill the same v olume when the e xcess po wer is 5 kW . Precious time w as sa v ed, about 75pc. This results from the f act that the flo w of compressed air injected into the reserv oir increases proportionally with the increase in po wer . Ho we v er , these results still justify the choice of 5 stages of compression which has the adv antages of increasing the reserv e of compressed air and also prolonging the autonomy of the o v erall system. 0 50 100 150 200 250 300 0 0.5 1 1.5 2 2.5 3 3.5 4 compression ratio The filling time of a 300L (h)     N=1 N=2 N=3 N=4 N=5 Figure 6. v ariation of char ging time as a function of compress ion ratio and number of stages for a compressor of 5KW of po wer 0 50 100 150 200 250 300 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 compression ratio The filling time of a 300L (h)     P=5KW P=10KW P=15KW P=20KW Figure 7. v ari ation of the char ging time as function of compression ratio and po wer consumed by the compressor 5 floor It is easy to see from this figure that the amount of compressed air produced increases with the increase in the number of compression stages.A remarkable v olume g ain can be achie v ed if the compressor has 5 floor of compression is used instead of a single stage compressor . Study and Dimensioning of the T anks Dedicated to a Compr essed Air ... (I. Rais) Evaluation Warning : The document was created with Spire.PDF for Python.
2036 ISSN: 2088-8708 0 5 10 15 20 25 30 0 100 200 300 400 500 600 700 800 900 POWER ABSORBED BY THE COMPRESSOR (KW) The volume of compressed air produced (L)     N=5 N=4 N=3 N=2 N=1 Figure 8. v ariation of v olume of air produced as a function of the number of floors and of the po wer consumed by the compressor 5.2.2. Dischar ging time The figures 9 and 10 sho w the v ariations of the di schar ge time, respecti v ely , as a function of the max- imum storage pressure and the number of e xpansion stages in the compressed air motor . The analysis of these figures sho ws that increasing the number of stages of a CAM serv es to reduce the dischar ge time of the reserv oir and consequently to accelerate the restoration of t h e stored ener gy in the form of compressed air , Thus, the higher the storage pressure, the longer the dischar ge time. 0 0.5 1 1.5 2 2.5 3 50 100 150 200 250 300 Discharge time (h) Pressure in a tank of 300bar     P=250bar P=200bar P=150 bar P=100bar P=50 bar Figure 9. v ariation of pressure of the tank as a function of maximum storage pressure and dischar ge time 0 0.5 1 1.5 2 2.5 3 50 100 150 200 250 300 Discharge time (h) Pressure in a reservoir of 300bar     N=1 N=2 N=3 N=4 N=5 Figure 10. v ariation of pressure of the tank as a function of the dischar ge time and the number of stages 6. CONCLUSION The fluctuating and intermittent nature of rene w able ener gies requires the strengthening of the control of ener gy flo ws between supply and demand for electricity . ener gy storage then constitute a rele v ant response to IJECE V ol. 8, No. 4, August 2018: 2029 2037 Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE ISSN: 2088-8708 2037 this problematic,currently v arious solutions for storing green electricity e xist (batteries, storage by compressed air or STEP - Stations of T ransfer of Ener gy by Pumping). Ho we v er , the y do not permit massi v e storage of the intermittent ener gy produced o v er a long period of time. this study presented one of the main e xisting means of storing, the CAES (Compressed Air Ener gy Storage) or ener gy storage by compressing air which consists of storing ener gy in the form of compressed air , in an under ground ca vity (for a po wer of more than 100 MW), or in tri vial tanks for small-scale storage, this is the case presented in this paper and then restore via a turbine producing electricity ag ain. REFERENCES [1] Ilham rais and Hassan Mahmoudi, The control strate gy for a h ybrid windphoto v oltaicsystem with com- pressed air storage element, 2nd International Conference on Electrical and Information T echnologies ICEIT2016,(ICEIT) , T angiers, 2016, pp. 89-92. [2] Hussein ibrahim , etude et conception dun generateur h ybride delectricite de type olien-diesel a v ec lment de stockage dair comprim , uni v ersit du quebec chicoutimi, 2010. [3] Minh Huynh Quang , Optimisation de la production de llectricit renouv elable pour site isol,Uni v ersity of Reims Champagne-Ardenne, 2013 [4] S. Rotthuser , V erf ahren zur Berechnung und Untersuchung h ydropneumatischer Speicher . PhD thesis, Rheinisch-W estflischen T echnischen Hochshule, Aachen, 1993. [5] K. W . Li, Applied Thermodynamics: A v ailability Method and Ener gy Con v ersion. T aylor and Francis, 1996. [6] J. Lefvre, Air Comprim; T ome 1: Production. P aris - France: Enc yclopdie industrielle, 1978. [7] J. F aisandier and Coll., Mcanismes Hydrauliques et Pneumatiques. P aris France: T echnique et Ingnierie, 8 ed., 1999. [8] Sylv ain LEMOFOUET - GA TSI,in v estig ation and optimisation of h ybrid electricity storage systems based on compressed air and supercapacitors, COLE POL YTECHNIQ UE FDRALE DE LA USANNE,2006 [9] Ben Slama Sami;An Intelligent Po wer Management In v estig ation for StandAlone Hybrid System Using Short-T ime Ener gy Storage;International Journal of Po wer Electronics and Dri v e System (IJPEDS);V ol. 8, No. 1, March 2017, pp. 367 375. [10] K.L. Sireesha, G. K esa v a Rao ;Droop Characteristics of Doubly Fed Induction Generator Ener gy Storage Systems within Micro Grids,International Journal of Po wer Electronics and Dri v e System (IJPEDS),V ol. 6, No. 3, September 2015, pp. 429 432 [11] D. Ganesh*, S. Moorthi**, H. Sudheer* ,D. Ganesh*, S. Moorthi**, H. Sudheer* ;A V oltage Controller in Photo-V oltaic System with Batte ry Storage for Stand-Alone Applications ;International Journal of Po wer Electronics and Dri v e System (IJPEDS),V ol.2, No.1, March 2012, pp. 9 18 [12] U. MON ASH, The pioneers: An anthology: V ictor tatin (1843 - 1913), http://www .ctie.monash.edu.au/har gra v e/tatin.html, v ol. Access: january 2006. [13] C. Sutton, Ucla study suggests air h ybrid car could impro v e fuel ef ficienc y , UCLA Engineer , v ol. 10, pp. 4 5, 2003. [14] Hussein Ibrahim, Mariya Dimitro v a, Adrian Ilinca, Jean Perron, Systme h ybride oliendiesel a v ec stockage d’air comprim pour l’lectrification d’une station de tlcommunications isole. European Journal of Electrical Engineering, V olume 12/5-6 - 2009 -pp.701-731 [15] S. Lemofouet, In v estig ation and optimisation of h ybrid electricit y storage systems based on compressed air and supercapacitors, Thse de doctorat, cole Polytechnique Fdrale de Lausanne, Suisse, 2006. [16] Adas Copco, LBZ Air Motors Manual, T ools Nr . 9833 8998 03, 2002, www .adascopcoairmotors. com Study and Dimensioning of the T anks Dedicated to a Compr essed Air ... (I. Rais) Evaluation Warning : The document was created with Spire.PDF for Python.