Swarm intelligence for electrical design space modeling and optimization

A computer comprising:

• a processor;
• a memory coupled to the processor;
• processing logic executing on the processor that:
• receives a design space that is an electrical design space of a complex electronic system;
• generates a plurality of particles of a particle swarm, wherein the particle swarm includes an initial position vector;
• assigns a plurality of initial values for each electrical parameter of the design space, wherein the initial values include an initial velocity vector value for each particle;
• calculates a fitness for each particle of the particle swarm;
• performs an optimization of the design space using the plurality of initial values and the calculated fitness for each particle of the particle swarm;
• in response to the completion of the optimization of the design space:
• updates a global best value for the particle swarm based on the calculated fitness;
• updates a personal best value for each particle based on the fitness;
• and
• determines if the optimization has reached a termination criteria, wherein the termination criteria comprises a predetermined amount of times that an optimization can be run for one or more predetermined variables;
• and
• in response to determining that a termination criteria has been reached, creates an optimized design space using an optimized fitness incorporating the global best value for the particle swarm and the personal best value for each particle of the particle swarm, wherein the optimized fitness also includes a global worst value for the particle swarm and a personal worst value for each particle of the particle swarm.

A hardware-based computer-readable storage medium having a plurality of instructions stored therein, wherein the plurality of instructions, when executed by a processing device, allows a machine to:

• receive a design space that is an electrical design space of a complex electronic system;
• generate a plurality of particles of a particle swarm, wherein the particle swarm includes an initial position vector;
• assign a plurality of initial values for each electrical parameter of the design space, wherein the initial values include an initial velocity vector value for each particle;
• calculate a fitness for each particle of the particle swarm;
• perform an optimization of the design space using the plurality of initial values and the calculated fitness for each particle of the particle swarm;
• in response to the completion of the optimization of the design space:
• update a global best value for the particle swarm based on the calculated fitness;
• update a personal best value for each particle based on the fitness;
• and
• determine if the optimization has reached a termination criteria, wherein the termination criteria comprises a predetermined amount of times that an optimization can be run for one or more predetermined variables;
• and
• in response to determining that a termination criteria has been reached, create and output an optimized design space using an optimized fitness incorporating the global best value for the particle swarm and the personal best value for each particle of the particle swarm, wherein the optimized fitness also includes a global worst value for the particle swarm and a personal worst value for each particle of the particle swarm.