Navigation

    CFD中文网

    CFD中文网

    • Login
    • Search
    • 最新

    关于稳态欧拉-拉格朗日计算速度的探讨

    OpenFOAM
    2
    4
    630
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • 液
      液固两相流 last edited by

      目前我在采用稳态的欧拉拉格朗日求解器,求解水轮机的磨损问题,颗粒数为50000,网格数为1100w左右。颗粒文件的设置如下:```
      code_text

      solution
      {
          active          true;
          
          calcFrequency   15;
      
          maxTrackTime    6;
      
          maxCo           0.6;
      
          coupled         true;
      
          cellValueSourceCorrection off; //suggested to be off
      
          interpolationSchemes
          {
          
             Urel      	cellPoint;		//U for water
             rho     		cell;	 		//rho of water
             mu      		cell;			 //mu of water
             DUcDt        cell;	 		//DUcDt of water 
      
          }
      
          integrationSchemes
          {
              U               Euler;
          }
      
          sourceTerms
          {
           resetOnStartup false;
              schemes
              {
                  U semiImplicit 1;
              }
          }
      }
      
      constantProperties
      {
          rho0 2650;	//the density for the particle
      }
      
      subModels
      {
          particleForces
          {
              sphereDrag; //drag forces
      
              gravity;  // gravity force
      
      		pressureGradient //pressure gradient force
      		{
      		  U   Urel;
      		}
      		
      		virtualMass    //virtual mass force
      		{
      		  Cvm   0.5;
      		}
      		
      		MRF
      		{
      		 omega (0 0 -104.72);
      		 centreOfRotation (0 0 0);
      		 Ctemp	MRFC;
      		}
      	
           }
      
          injectionModels
          {
      
              model1
              {
      		 type					patchInjection; 		
               parcelBasisType   		fixed;					
               patchName        		INLET_SV;					
               U0                		(0 0 0);				
      		 nParticle          	420;
      		 duration            	1;					
      		 parcelsPerSecond		50000;
               flowRateProfile 		constant 1;
      		 massFlowRate     		 0;       			
      		 //positionsFile   		"kinematicCloudPositions";		
                  sizeDistribution
                  {
                      type        fixedValue;
                      fixedValueDistribution
                      {
                          value   0.0002;//m
                      }
                  }
              }
          }
      
          dispersionModel stochasticDispersionRAS;
      
          patchInteractionModel UserLocalInteraction;
          //using the tabakoff rebound model 
      
          UserLocalInteractionCoeffs
          {
              patches
              (
                  "(.*)"
                  {
                      type rebound;
      				reboundModel 2;
                  }
      
                  OUTLET_SC
                  {
                      type none;
                  }
                  INLET_SV
                  {
                      type escape;
                  }
                  OUTLET_SV
                  {
                      type none;
                  }
                  INLET_GV
                  {
                      type none;
                  }
                  OUTLET_GV
                  {
                      type none;
                  }
                  INLET_RUN
                  {
                      type none;
                  }
                  OUTLET_RUN
                  {
                      type none;
                  }
                  INLET_DT
                  {
                      type none;
                  }
      
                  INLET_SC
                  {
                      type none;
                     
                  }
                  OUTLET_DT
                  {
                      type escape;
                     
                  }
              );
          }
      
         
          heatTransferModel none;
      
          surfaceFilmModel none;
      
          collisionModel none;
      
          stochasticCollisionModel none;
      
          radiation off;
      }
      
      

      流场迭代15步之后,会进行颗粒的迭代,颗粒计算平均耗时5min。
      但每次计算至210步时,颗粒就算不动了(前面的颗粒计算均是正常的)。
      我如果把颗粒数降至10000,计算至750步,颗粒就算不动了。

      请问一下大家又碰到这类问题吗?有什么比较好的解决办法吗?

      1 Reply Last reply Reply Quote
      • 李东岳
        李东岳 管理员 last edited by

        发代表性网格截面看一下

        CFD课程 改成线上了 http://dyfluid.com/class.html
        CFD高性能服务器 http://dyfluid.com/servers.html

        液 1 Reply Last reply Reply Quote
        • 液
          液固两相流 @李东岳 last edited by

          @东岳
          我的网格有多个交界面(用的是cyclicAMI),交界面对于拉格朗日的计算会有影响吗?

          1 Reply Last reply Reply Quote
          • 李东岳
            李东岳 管理员 last edited by

            很有可能,cyclicAMI有可能把粒子算丢了

            CFD课程 改成线上了 http://dyfluid.com/class.html
            CFD高性能服务器 http://dyfluid.com/servers.html

            1 Reply Last reply Reply Quote
            • First post
              Last post

            CFD中文网 | 东岳流体 | 京ICP备15017992号-2