նѧ
JOURNAL OF AEROSPACE POWER
1999ꡡ141ڡVOL.14No.11999



ѹҶƬƵһ·
Ԫнϯ
ժҪӦŴ㷨缼ͨԺѡҶƬٶȷֲĿٶȷֲ֮ļСٶȷֲµѹҶƬߡҶƬBezier߲ʾѡҶƬߵBPɡķһѹƽֱѹĶԪȺҶƬƣʾ÷ЧͿеġ
ʣѹ硡ҶƬ
ţTH452
A NEW INVERSE DESIGN METHOD
OF DIFFUSER BLADES PROFILE
Fan HuiyuanWang ShangjinXi Guang
(Xian Jiaotong University, Xian, 710049)
ABSTRACTA genetic algorithm combined with a neural network technique is proposed to search for turbomachinery diffuser blade profile under the condition of a given velocity distribution on its surface.Such a new inverse method works through minimizing the error between surface velocity distribution of candidate blades and the target velocity distribution.For ease in employing genetic algorithm,the blade profiles searched are parameterized with Bezier spline curves.The flow analysis of the candidate blades is carried out by BP neural networks which have been trained with a set of samples.The proposed approach is demonstrated through the application to a diffuser with two-dimensional height-and thickness-constant blades.The simulations show that the new method is not only feasible,but also reliable and effective.
KeyWordsDiffusersNeural networksBladesDesign
1
͸ƽеѹҶƬһǸҶƬٶȷֲ³SVDSurface Velocity Distribution))ѹֲͨһֱӹҶƬⷽǱǱ任෽ǽڴƽ⡣϶౻õⷽſƺŻ෽12ݡЩΪݶ෽Ч༫ֵͷĿ꺯(е)оһֱѰҸõƷŴ㷨(GA)һģȻƵŻС³ԡԡȫԵص㣬GA㷺Ӧ๤оرǽеоӦҲ35ݡ缼ģԪṹԻƵһּԣĿǰ幤еӦҲߣ67ݡӦGA缼ʵ͸ƽеѹԪҶƬơҶƬĿSVD£һȾɫʾĸ(ѡҶƬ)ɵȺGAӵ³ĿSVDѡҶƬSVD֮ļСþĿSVDҶƬߡGAУѵ򴫲(BP)ɶԺѡҶƬѹ㡣BPѵռΪҶƬԼгӦзáķӦһʵѹƽֱѹĵȺҶƬơʾֻҪθһSVD÷ԳɹҪҶƬߡ
2GAӦ
GAͨɸɵȺû׼н⡣־ѹҶƬ⣬ӦGAⷽ
21ҶƬ߲
GAһʵʵʱ뽫ÿһƱ޳ַ롣ƱĳͷβɵĳΪĳһ壨ѡƣȾɫ塣иΪѡҶƬߣҶƬ߱ΪɺٵıʾĲBezierҶƬߡΪˣƽϵĻѹҶդ(1)ʽǵر任Ϊӳƽ


ͼ1ӳҶƬ߼ʾ
ϵֱҶդ
Z=log(Z䣯R1)log(R2/R1)(1)
ʽУZƽ棬R1R2ѹںͳڰ뾶Zӳƽ档ѡҶƬӳƽȻ(1)ʽ任صƽΪʵʺѡҶƬߡӳƽҶƬ4Bezier߿̻״ӦBezier5ȷͼ1ʾBezier5εĶΪP0(x0y0)P1(x1y1)P4(x4y4)Bezier߿һ4BernsteinʽʾΪQ(t)=Bn,iPiBn,i=cinti(1-t)n-itֵ߲ȡΪ0t1Ϊ˼ȱ֤ߵּ򻯼㣬P0̶ԭ㣬P0(00)P2P4x̶yɱ䣬x2=0.5x4=1P1P3xɱ䣬yҶƬڽǺͳڽԼҶƬڽǺͳڽǷֱΪ1ͦ2ɵy1=(tg1)x1y3=(tg2)x3+tg2-y4Bezier߱ҶƬ߱ζеĸδ֪x1y2x3y4ȷ
ĸҶƬ߲һάX=(x1x3y2y4)һ塣öƱ뷽ÿ룬ɹɸȾɫ塣
22BPģ
BPھõĺƽԶڽ֪ҶƬµΪṹӶʡѵʱ䣬ѡҶƬSVDΪѹֿǡSVDϵĲֿԷֱ5ζʽϡɴˣĽBPֱҶƬѹSVD4-8-8-6ṹʽһ㣬һ㣩4뵥ԪֱѡҶƬ߲x1x3y2y46ԪֱѹSVD5ζʽϵڱ½ѹṹߴ£øıҶƬ߲ķҶƬ߹200ۺͱ߽ԪҶդ϶ЩҶƬзóÿҶƬӦٶȷֲ(ٶȷֲĶʽϵ)ҶƬ߲ӦѹSVDʽϵԸ200ÿиȡ100ΪѵֱѵBP㷨Ȩֵѵһ586΢ϽУѵϵƽ͡0.20ʱѵøԵ100ʣԶѵм飬ƽ;1.0֤ѵЧáѵBPֻҪһҶƬ߲(ֻҪòѵųɵĿռ)ӦѹSVDĶʽϵ
23Ӧֵ
GAУӦֵϵҪԹơGAСĿ
꺯K(X)R2R1(Vt-Va)2dR(2)
ʽУVtĿSVDVaѡҶƬߵʵSVDKļС׵תΪӦFļ⣺
F(X)11+K(X)ݡ(3)
FڸӦֵ㡣ӦֵʱȶȾɫн룬BPӦѡҶƬ߱SVDȻ(2)(3)ʽӦֵ
24GAĻ
GAһģȻŴɵѭ̡һѭ԰ȻŴѧеʹĲѡӽͱ졣༴GAһѡӽͱɡѡǰӦֵѡĹ̡ӦֵԽߵĸ屻ѡϵĿԽӽ̷ֶɣһѡĸȾɫ尴һӽʵԣڶÿһȾɫ尴λ໥Բ֣˲һеһȾɫ(Ӹ)δѡԵĸȾɫͬʱֱӿΪӸ塣Ӹ尴ձʶԻ(ֵֽ)ĸı䡣ȾɫֵֽǷҪı䣬ı01෴
3Ӧʵ
ӦһʵѹѹϡƽøѹĻṹߴڲ£ֱD1=1300 mmֱD2=1760 mm(Ҷ)b1=b2=40 mmǦ1=19㣬Ǧ2=29㣬Gin=10.185 kg/s¶t1=61.7棬ѹp1=141200 PaȡѹҶƬN=20ҶƬtn=8 mm
Ϊ˻ÿҶƬ״߲ȡֵ綨Ϊ0x10.50.5x31.00y21.20.8y42.0ĳȡҪľȡһʵɳΪlĶ룬ñ䱻ȷΪ2l-1䡣lĶһһӦԲx1Ҫ㾫ΪС3λζ00.5ӦٱȷΪ(0.5-0)100=50䡣Ȼx1ҪĶΪ6(Ϊ26=6450)Ϊ򵥼ƣҶƬ߲óΪ8Ķ롣GAĲԼѡȡӰ㷨̬졣еѡ̲ñ׼̶ķӦֵȺиӦֵ͵ıֵڽȨ֡ʽתȷǷѡΪӽ4ӽģʽÿ߲Ļͬʱӽ㡣GAá顱ȷͨ仯۲GAֲ̬ȡֵӡȨ۲Ⱥģn=40ӽpc=0.8ͱpm=0.02ʱõϺGA̬
BPѵųɵĿռУΪɲͬҶƬ(³ԭʼҶƬ)ȡBPѵͬһӦSVDЩSVDΪĿSVDñķЩٶȷֲҶƬߡMatlabһ586˼ʵ֡ڲͬGAɹҵٶȷֲҶƬߡѡȡĸҶƬ(Case1Case2Case3Case4)ԱĸҶƬ״ڱ1



ͼ2GAҶƬԭʼҶƬ߶Ա
1ԭʼҶƬ߲
Casex1x3y2y4
1
2
3
40.20
0.10
0.23
0.080.90
0.80
0.93
0.931.00
0.60
1.15
0.600.90
1.10
1.85
1.85



ͼ3GAҶƬSVDʵSVDԱ
GA150õҶƬԭʼҶƬ߶Աͼ2ʱGAҶƬSVDӦĿSVDʾͼ3ͼ3ΪҶƬٶȣVʵٶֵV0ѹٶֵΪѹؾ࣬LҶƬǰԵʵʽؾ࣬LCǰԵ֮ؾࡣͼеԭʼҶƬGAҶƬԼĿSVDGAҶƬSVDһԣ˵˱ķǿкЧġͼ4ʾ4£GAӦֵȺƽӦֵݱΣпԿŴ㷨


ͼ4GAøӦֵȺƽӦֵݱ
Ⱥ͸
4ᡡ
ѹҶƬ·ҶƬٶȷֲʱӦҶƬ״ķŴ㷨ŵ⣬缼ĲãŴռʱ䣬ֵЧʡԼӸĶƹ㵽ҶƬԪҶƬơеĿ꺯ʽӦ仯˼Ϊӵе⡣
߼飺ԪС35ꡡʿͨѧԴ붯ѧԺ710049
ߵλͨѧ
ο
1ݡVenkataraman P.Inverse Airfoil Design Using Design Optimization. AIAA-96-2503-cp,1996
2ݡLewis J C,Agarwal R K.Airfoil Design Via Control Theory Using Full Potential and Euller Equations.AIAA-96-2483-cp,1996
3ݡObayshi S,Tankanshi S.Genetic Optimization of Target Pressure Distribution for Inverse Design Methods.AIAA Journal,1996,34(5):881-886
4ݡQuagliarella D,Cioppa A D.Genetic Algorithm Applied to the Aerodynamic Design of Transonic Airfoils.Journal of Aircraft,1995,32(4):889-890
5ݡObayashi S,Tsukahara T.Comparison of Optimization Algorithm for Areodynamic Shape Desian.AIAA-96-2394-cp,1996
6ݡPrasnth R K,Whitaker K W. Neuromorphic Approach to Inverse Problems in Aerodynamics.AIAA Journal,1995,33(6):1150-1152
7ݡWhitaker K W.Specifying Exhaust Nozzle Contours with a Neural Network.AIAA Journal,1993,31(2):
ո壺19981
޸壺19984
α༭
