1. Authors

(a)	Trichy M Kethara Pasupathy, Health Science Center-Jacksonville, University of Florida, 655 West 8th Street, Jacksonville Florida, FL 32209, USA,
trichy.pasuathy@gmail.com 
      Phone # (904) 244-5434

(b)   Ravi Teja SeethamRaju, Siemens Medical Solutions MGH-NMR Center, Building 149, Room 2301, 13th Street, Charlestown, MA 02129, USA
ravi.seethamraju@siemens.com
Phone # (617) 726-1627

(c)   Robert G. Wilhelm, Department of Mechanical Engineering, The University of North Carolina at Charlotte, 9201 University City Building, Charlotte, NC 28223, USA
rgwilhel@uncc.edu
Phone # (704) 547-3003


2. Title: 

Evolutionary algorithm for the placement of fluid power valves on a valve stand

Journal Publication: Engineering with Computers. Paper in press but published online
http://www.springerlink.com/app/home/contribution.asp?wasp=b2f436aca8144afaa01dbe8b4fcb7394&referrer=parent&backto=issue,2,11;journal,1,22;linkingpublicationresults,1:102820,1

(A copy of the paper is also attached.)


3. Abstract: 

This paper is about physically realizing a fluidpower circuit. Fluidpower circuits are similar to electronic circuits except that valves, pipes and connectors are used. However, the constraints are very different. The choice of placement of fluid power valves on a valve stand and routing by pipes impacts operational costs. Choosing the right placement that provides optimum routing configuration or determining the optimum routing for a chosen placement, are both computationally hard problems. An evolutionary algorithm (EA), to minimize operational costs while optimizing placement and routing of valves, is developed here. The best practices in the industry are abstracted and implemented in the EA. In this paper, the algorithm and its performance for examples with varying complexities are presented. Our results meet or exceed experienced designers solutions.


4. Statement:

Both D and G are applicable.

(D) The result is publishable in its own right as a new scientific 
result, independent of the fact that the result was mechanically 
created. 

This is a unique problem in Mechanical Engineering that has not been approached formally as an automated task. To date this problem has been generally carried out by human designers using rules of thumb. While they follow guidelines, they necessarily cannot reach optimal solutions with increased complexity. Hence, by automating this process we have established a new scientific result. 


(G) The result solves a problem of indisputable difficulty in its 
field.

This paper formalizes the complexity and establishes several benchmark problems and solutions.  We believe that this concept can be extended to a whole range of plumbing problems prevalent in industries such as petroleum and other chemical industries. Fluidpower systems usually run continuously and energy is lost in every cycle due to resistance to flow. A non-optimal placement will increase these losses. This is recognized as a problem to be dealt to save on energy losses. We have demonstrated that the search space is huge and it is difficult to solve with other analytical methods. Our approach with evolutionary algorithm solves this indisputably difficult problem and expected to be the stepping-stone for future improvements.