Method for optimizing the design of micro-fluidic devices

A system for designing micro-fluidic devices, comprising:

• one or more processors including memory;
• a cost function calculator for determining a target cost function value based on selected input information relating to device design parameters including one or more of task, particles, system and design layout and for determining cost function values for design candidates;
• a design candidate selector that, based upon the defined system parameters, selects one or more micro-fluidic device designs as design candidates based on a comparison between the design parameters of the one or more micro-fluidic device designs and the present device design parameters, or which accepts a user-input design candidate;
• a simulation model operator that runs simulation models on one or more design candidates and identifies a best initial design candidate based on computed cost function;
• an optimization routine operator that modifies the best initial design candidate and optimizes same by performing an optimization routine including:
• modifying design parameters of the best initial design candidate to provide a modified design candidate having parameters that differ from the parameters of the best initial design candidate, running a simulation model on the modified design candidate, checking the cost function of the modified design candidate against the target cost function, forwarding optimized design parameters of an optimized device design as system output when the computed cost function of the modified design candidate meets the target cost function value;
• and performing further optimization routines when the cost function of the modified design candidate does not meet the target cost function value.

A computer program product for designing micro-fluidic devices, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions being executable by a computer to cause the computer to perform a method comprising:

• determining a target cost function based on one or more of task, particles, system features, and design type as device design parameters;
• simulating the performance of one or more chosen design candidates with a selected simulation model and identifying a design candidate with a cost function closest to the target cost function as a best initial design candidate;
• on the best initial design candidate, running an optimization routine to modify design parameters of the best initial design candidate to provide a modified design candidate having design parameters differing from the design parameters of the best initial design candidate, and computing the cost function for the modified initial design candidate;
• returning optimized device design parameters of an optimized device design derived from a modified design candidate having a computed cost function that meets the target cost function;
• repeating, if necessary, the optimization routine on the modified design candidate one or more times until the computed cost function for the modified design candidate meets the determined target cost function.