초록 ▼

The invention relates to method for self-optimizing, with closed-loop feed-back control but without human interaction, an object relative to a specific criterion or activity during the operation of said object and in response to variations on a prescribed number m of variables, comprising: planning

The invention relates to method for self-optimizing, with closed-loop feed-back control but without human interaction, an object relative to a specific criterion or activity during the operation of said object and in response to variations on a prescribed number m of variables, comprising: planning without human control, guidance, supervision, and intervention a statistical design matrix for testing said m variables in n tests; according to the design matrix performing said n designed tests on said very object in actual operation; determining the instantaneous combination of said m variables which optimizes said specific criterion or activity during said actual operation; setting the conditions of said m variables at the thus-determined optimal combination before the combination significantly changes; and feeding back information on the status of optimization to achieve closed-loop feed-back control. The back-feeding step may be substantially continuous to achieve substantially continuous self-optimizing operation of the object. A machine in the form of programmed microprocessor embodying this close-looped self-optimizing method is also disclosed.

대표청구항 ▼

A machine for real-time self-optimizing, without human interaction, an object relative to a specific criterion or activity in response to variations on a prescribed number m of variabes, comprising: means for automatically planning a statistical design matrix for testing the m variables in n tests;

A machine for real-time self-optimizing, without human interaction, an object relative to a specific criterion or activity in response to variations on a prescribed number m of variabes, comprising: means for automatically planning a statistical design matrix for testing the m variables in n tests; means for performing said n designed tests according to the design matrix on said object; means for determining the instantaneous combination of said m variables which optimizes said specific criterion or activity; means for setting the conditions of said m variables at the thus-determined optimal combination: and means for coupling the planning, performing, determining, and setting means for substantially continuous recycling to achieve real-time self-optimizing control.