Hi all, I am performing a sensitivity analysis for a topology optimization problem and I wonder if Ansys APDL allows users to calculate the derivatives of the global stiffness matrix with respect to the element density. I read online that I can take derivatives using the *voper command. My question is: can I exploit this operation command for my purpose? And if yes, how can I import the stiffness matrix as a parameter? Thank you for your help. -- Giulia Angelucci Phd student University of Michigan Department of Civil and Environmental Engineering 2350 Hayward, 2148 GG Brown Ann Arbor, Michigan 48109-2125

Hi Giulia, My suggestion (because I have done it like that) is that thru a Shell script run ansys in bach mode to generate a stiffness matrix file, then use something like matlab to read the file and perform the operations you wish to do. Juan Pablo Toledo González General Manager COMPLX "Ingeniería Avanzada para Soluciones Reales" www.complx.com.mx Tel: 229 200 9573 Cel: 229 136 5599 -----Mensaje original----- De: Giulia Angelucci [mailto:gangeluc@umich.edu] Enviado el: miércoles, 24 de mayo de 2017 09:25 a.m. Para: xansys-temp@xansystest.info Asunto: [Xansys] Sensitivity analysis_ansys Hi all, I am performing a sensitivity analysis for a topology optimization problem and I wonder if Ansys APDL allows users to calculate the derivatives of the global stiffness matrix with respect to the element density. I read online that I can take derivatives using the *voper command. My question is: can I exploit this operation command for my purpose? And if yes, how can I import the stiffness matrix as a parameter? Thank you for your help. -- Giulia Angelucci Phd student University of Michigan Department of Civil and Environmental Engineering 2350 Hayward, 2148 GG Brown Ann Arbor, Michigan 48109-2125 _______________________________________________ Xansys-temp mailing list Xansys-temp@xansystest.info http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info If you are receiving too many emails from XANSYS please consider changing account settings to Digest mode which will send a single email per day. Please send administrative requests such as deletion from XANSYS to xansys-mod@tynecomp.co.uk and not to the list --- El software de antivirus Avast ha analizado este correo electrónico en busca de virus. https://www.avast.com/antivirus

Hi, In fact, it's in the FM. Look @ the Optimisation module of Mechanical APDL, I have used it many years ago to optimize a beam weight, it is even one example of the Verification Manual (VM here, not FM...). OK this is not "real" topology optimisation, but I have seen Ansys guyz making disappear quads for real topology optimisation. But it looked like voxels huuh.... Benjamin HAGEGE, Eng. & Ph.D. Arts & Métiers *Own Compagny : NABLA.Expertises *Academic job 1 : Collaborator of french Minister of Research (for Research Tax Credit) *Academic job 2 : Associate prof. @ www.utc.fr / Dpt Mechanical Engineering 2017-05-24 16:25 GMT+02:00 Giulia Angelucci <gangeluc@umich.edu>: > Hi all, > I am performing a sensitivity analysis for a topology optimization problem > and I wonder if Ansys APDL allows users to calculate the derivatives of the > global stiffness matrix with respect to the element density. > I read online that I can take derivatives using the *voper command. My > question is: can I exploit this operation command for my purpose? And if > yes, how can I import the stiffness matrix as a parameter? > > Thank you for your help. > > -- > Giulia Angelucci > Phd student > University of Michigan > Department of Civil and Environmental Engineering > 2350 Hayward, 2148 GG Brown > Ann Arbor, Michigan 48109-2125 > _______________________________________________ > Xansys-temp mailing list > Xansys-temp@xansystest.info > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > If you are receiving too many emails from XANSYS please consider changing > account settings to Digest mode which will send a single email per day. > > Please send administrative requests such as deletion from XANSYS to > xansys-mod@tynecomp.co.uk and not to the list >

Thank you Benjamin for your reply. I read the VM manual, but can't found my topic: I need to write a brand new optimization code because the Ansys TO tool is very restrictive and I need to consider volume as objective function and displacement as constraint. I am writing a Matlab code but want Ansys to provide FEA results (stiffness and displacement), does anyone of you know how can I change the Ansys stiffness equation to incorporate the element density according to the Simp method ([K]=[Ke]*ro^p with ro: element density and p:penalization factor)? Should I try with a Subroutine? Thank you On Fri, May 26, 2017 at 7:35 PM, BenZ Hagege <Benjamin.Hagege@utc.fr> wrote: > Hi, > > In fact, it's in the FM. Look @ the Optimisation module of Mechanical APDL, > I have used it many years ago to optimize a beam weight, it is even one > example of the Verification Manual (VM here, not FM...). OK this is not > "real" topology optimisation, but I have seen Ansys guyz making disappear > quads for real topology optimisation. But it looked like voxels huuh.... > > Benjamin HAGEGE, Eng. & Ph.D. Arts & Métiers > *Own Compagny : NABLA.Expertises > *Academic job 1 : Collaborator of french Minister of Research (for Research > Tax Credit) > *Academic job 2 : Associate prof. @ www.utc.fr / Dpt Mechanical > Engineering > > > > 2017-05-24 16:25 GMT+02:00 Giulia Angelucci <gangeluc@umich.edu>: > > > Hi all, > > I am performing a sensitivity analysis for a topology optimization > problem > > and I wonder if Ansys APDL allows users to calculate the derivatives of > the > > global stiffness matrix with respect to the element density. > > I read online that I can take derivatives using the *voper command. My > > question is: can I exploit this operation command for my purpose? And if > > yes, how can I import the stiffness matrix as a parameter? > > > > Thank you for your help. > > > > -- > > Giulia Angelucci > > Phd student > > University of Michigan > > Department of Civil and Environmental Engineering > > 2350 Hayward, 2148 GG Brown > > Ann Arbor, Michigan 48109-2125 > > _______________________________________________ > > Xansys-temp mailing list > > Xansys-temp@xansystest.info > > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > > If you are receiving too many emails from XANSYS please consider changing > > account settings to Digest mode which will send a single email per day. > > > > Please send administrative requests such as deletion from XANSYS to > > xansys-mod@tynecomp.co.uk and not to the list > > > _______________________________________________ > Xansys-temp mailing list > Xansys-temp@xansystest.info > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > If you are receiving too many emails from XANSYS please consider changing > account settings to Digest mode which will send a single email per day. > > Please send administrative requests such as deletion from XANSYS to > xansys-mod@tynecomp.co.uk and not to the list > -- Giulia Angelucci Phd student University of Michigan Department of Civil and Environmental Engineering 2350 Hayward, 2148 GG Brown Ann Arbor, Michigan 48109-2125

Hi Giulia, "APDL Math" might work for you. *SMAT allows you to access the stiffness matrix without exiting APDL. (Specify EQSLV with WRFULL before SOLVE). *AXPY lets you do [K]=[Ke]*a. Thanks, Sze Kwan (Jason) Cheah Starkey Hearing Technologies -----Original Message----- From: Giulia Angelucci [mailto:gangeluc@umich.edu] Sent: Wednesday, May 31, 2017 8:13 PM To: XANSYS Mailing List Temporary Home <xansys-temp@xansystest.info> Subject: Re: [Xansys] Sensitivity analysis_ansys Thank you Benjamin for your reply. I read the VM manual, but can't found my topic: I need to write a brand new optimization code because the Ansys TO tool is very restrictive and I need to consider volume as objective function and displacement as constraint. I am writing a Matlab code but want Ansys to provide FEA results (stiffness and displacement), does anyone of you know how can I change the Ansys stiffness equation to incorporate the element density according to the Simp method ([K]=[Ke]*ro^p with ro: element density and p:penalization factor)? Should I try with a Subroutine? Thank you On Fri, May 26, 2017 at 7:35 PM, BenZ Hagege <Benjamin.Hagege@utc.fr> wrote: > Hi, > > In fact, it's in the FM. Look @ the Optimisation module of Mechanical > APDL, I have used it many years ago to optimize a beam weight, it is > even one example of the Verification Manual (VM here, not FM...). OK > this is not "real" topology optimisation, but I have seen Ansys guyz > making disappear quads for real topology optimisation. But it looked like voxels huuh.... > > Benjamin HAGEGE, Eng. & Ph.D. Arts & Métiers *Own Compagny : > NABLA.Expertises *Academic job 1 : Collaborator of french Minister of > Research (for Research Tax Credit) *Academic job 2 : Associate prof. @ > www.utc.fr / Dpt Mechanical Engineering > > > > 2017-05-24 16:25 GMT+02:00 Giulia Angelucci <gangeluc@umich.edu>: > > > Hi all, > > I am performing a sensitivity analysis for a topology optimization > problem > > and I wonder if Ansys APDL allows users to calculate the derivatives > > of > the > > global stiffness matrix with respect to the element density. > > I read online that I can take derivatives using the *voper command. > > My question is: can I exploit this operation command for my purpose? > > And if yes, how can I import the stiffness matrix as a parameter? > > > > Thank you for your help. > > > > -- > > Giulia Angelucci > > Phd student > > University of Michigan > > Department of Civil and Environmental Engineering > > 2350 Hayward, 2148 GG Brown > > Ann Arbor, Michigan 48109-2125 > > _______________________________________________ > > Xansys-temp mailing list > > Xansys-temp@xansystest.info > > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > > If you are receiving too many emails from XANSYS please consider > > changing account settings to Digest mode which will send a single email per day. > > > > Please send administrative requests such as deletion from XANSYS to > > xansys-mod@tynecomp.co.uk and not to the list > > > _______________________________________________ > Xansys-temp mailing list > Xansys-temp@xansystest.info > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > If you are receiving too many emails from XANSYS please consider > changing account settings to Digest mode which will send a single email per day. > > Please send administrative requests such as deletion from XANSYS to > xansys-mod@tynecomp.co.uk and not to the list > -- Giulia Angelucci Phd student University of Michigan Department of Civil and Environmental Engineering 2350 Hayward, 2148 GG Brown Ann Arbor, Michigan 48109-2125 _______________________________________________ Xansys-temp mailing list Xansys-temp@xansystest.info http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info If you are receiving too many emails from XANSYS please consider changing account settings to Digest mode which will send a single email per day. Please send administrative requests such as deletion from XANSYS to xansys-mod@tynecomp.co.uk and not to the list NOTICE: This e-mail is intended only for the addressed party and may contain information that is proprietary/confidential or otherwise protected. If you are not the named addressee, you are not authorized to view this message. If received in error, please contact the sender and delete the communication.

I will try the procedure you suggested me and let you know. Thanks for your valuable help! On Thu, Jun 1, 2017 at 2:45 PM, Jason Cheah <Jason_Cheah@starkey.com> wrote: > Hi Giulia, > > "APDL Math" might work for you. > > *SMAT allows you to access the stiffness matrix without exiting APDL. > (Specify EQSLV with WRFULL before SOLVE). *AXPY lets you do [K]=[Ke]*a. > > > Thanks, > Sze Kwan (Jason) Cheah > Starkey Hearing Technologies > > > -----Original Message----- > From: Giulia Angelucci [mailto:gangeluc@umich.edu] > Sent: Wednesday, May 31, 2017 8:13 PM > To: XANSYS Mailing List Temporary Home <xansys-temp@xansystest.info> > Subject: Re: [Xansys] Sensitivity analysis_ansys > > Thank you Benjamin for your reply. I read the VM manual, but can't found my > topic: I need to write a brand new optimization code because the Ansys TO > tool is very restrictive and I need to consider volume as objective > function and displacement as constraint. > I am writing a Matlab code but want Ansys to provide FEA results > (stiffness and displacement), does anyone of you know how can I change the > Ansys stiffness equation to incorporate the element density according to > the Simp method ([K]=[Ke]*ro^p with ro: element density and p:penalization > factor)? > Should I try with a Subroutine? > Thank you > > On Fri, May 26, 2017 at 7:35 PM, BenZ Hagege <Benjamin.Hagege@utc.fr> > wrote: > > > Hi, > > > > In fact, it's in the FM. Look @ the Optimisation module of Mechanical > > APDL, I have used it many years ago to optimize a beam weight, it is > > even one example of the Verification Manual (VM here, not FM...). OK > > this is not "real" topology optimisation, but I have seen Ansys guyz > > making disappear quads for real topology optimisation. But it looked > like voxels huuh.... > > > > Benjamin HAGEGE, Eng. & Ph.D. Arts & Métiers *Own Compagny : > > NABLA.Expertises *Academic job 1 : Collaborator of french Minister of > > Research (for Research Tax Credit) *Academic job 2 : Associate prof. @ > > www.utc.fr / Dpt Mechanical Engineering > > > > > > > > 2017-05-24 16:25 GMT+02:00 Giulia Angelucci <gangeluc@umich.edu>: > > > > > Hi all, > > > I am performing a sensitivity analysis for a topology optimization > > problem > > > and I wonder if Ansys APDL allows users to calculate the derivatives > > > of > > the > > > global stiffness matrix with respect to the element density. > > > I read online that I can take derivatives using the *voper command. > > > My question is: can I exploit this operation command for my purpose? > > > And if yes, how can I import the stiffness matrix as a parameter? > > > > > > Thank you for your help. > > > > > > -- > > > Giulia Angelucci > > > Phd student > > > University of Michigan > > > Department of Civil and Environmental Engineering > > > 2350 Hayward, 2148 GG Brown > > > Ann Arbor, Michigan 48109-2125 > > > _______________________________________________ > > > Xansys-temp mailing list > > > Xansys-temp@xansystest.info > > > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > > > If you are receiving too many emails from XANSYS please consider > > > changing account settings to Digest mode which will send a single > email per day. > > > > > > Please send administrative requests such as deletion from XANSYS to > > > xansys-mod@tynecomp.co.uk and not to the list > > > > > _______________________________________________ > > Xansys-temp mailing list > > Xansys-temp@xansystest.info > > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > > If you are receiving too many emails from XANSYS please consider > > changing account settings to Digest mode which will send a single email > per day. > > > > Please send administrative requests such as deletion from XANSYS to > > xansys-mod@tynecomp.co.uk and not to the list > > > > > > -- > Giulia Angelucci > Phd student > University of Michigan > Department of Civil and Environmental Engineering > 2350 Hayward, 2148 GG Brown > Ann Arbor, Michigan 48109-2125 > _______________________________________________ > Xansys-temp mailing list > Xansys-temp@xansystest.info > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > If you are receiving too many emails from XANSYS please consider changing > account settings to Digest mode which will send a single email per day. > > Please send administrative requests such as deletion from XANSYS to > xansys-mod@tynecomp.co.uk and not to the list > > NOTICE: This e-mail is intended only for the addressed party and may > contain information that is proprietary/confidential or otherwise > protected. If you are not the named addressee, you are not authorized to > view this message. If received in error, please contact the sender and > delete the communication. > _______________________________________________ > Xansys-temp mailing list > Xansys-temp@xansystest.info > http://xansystest.info/mailman/listinfo/xansys-temp_xansystest.info > If you are receiving too many emails from XANSYS please consider changing > account settings to Digest mode which will send a single email per day. > > Please send administrative requests such as deletion from XANSYS to > xansys-mod@tynecomp.co.uk and not to the list > -- Giulia Angelucci Phd student University of Michigan Department of Civil and Environmental Engineering 2350 Hayward, 2148 GG Brown Ann Arbor, Michigan 48109-2125