Control parameters for searching

A computer implemented method for searching for control parameters (Amn, Bmn, Cmn) that maximize output (Y) of object, the object producing output (Y) responsive to the control parameters (Amn, Bmn, Cmn), said method comprising:

• providing starting values of m said control parameters (Amn, Bmn, Cmn) in accordance with an algorithm that renews generations;
• repeating search cycles for each of said m control parameters (Amn, Bmn, Cmn) and, in each search cycle, adding a periodic function (S1, S2, S3) of a predetermined period and a correction value (V1, V2, V3) obtained in a previous search cycle to the control parameters (Amn, Bmn, Cmn) to provide input parameters (U1, U2, U3) to said search object;
• multiplying, in said each search cycle, an output (Y) obtained from said search object responsive to said input parameters (U1, U2, U3) by said periodic function (S1, S2, S3) and calculating said new correction value (V1, V2, V3) for correcting an integral value of a value (Zi) obtained by the multiplication such that said integral value converges;
• and
• terminating repetition of the search cycle when said control parameters (Amn, Bmn, Cmn) converge or when a predetermined time elapsed, and determining m search values (Amn′, Bmn′, Cmn′) that are values of said m control parameters (Amn, Bmn, Cmn) corrected by said correction values at the termination of repetition of search cycles with respect to each of said m control parameters (Amn, Bmn, Cmn),
• wherein said providing starting values comprises providing starting values for the next generation based on said m search values (Amn′, Bmn′, Cmn′), and when said m search values converge or when a predetermined number of generations has been reached, outputting said m search values as optimum control parameters.

A system, comprising:

• a processor;
• and a memory,
• wherein the processor is configured to
• search for control parameters (Amn, Bmn, Cmn) that maximize output (Y) of object, wherein the object produces output (Y) responsive to the control parameters (Amn, Bmn, Cmn),
• provide starting values of m said control parameters (Amn, Bmn, Cmn) in accordance with an algorithm that renews generations,
• repeat search cycles for each of said m control parameters (Amn, Bmn, Cmn) and, in each search cycle, add a periodic function (S1, S2, S3) of a predetermined period and a correction value (V1, V2, V3) obtained in a previous search cycle to the control parameters (Amn, Bmn, Cmn) to provide input parameters (U1, U2, U3) to said search object,
• multiply, in said each search cycle, an output (Y) obtained from said search object responsive to said input parameters (U1, U2, U3) by said periodic function (S1, S2, S3),
• calculate said new correction value (V1, V2, V3) for correcting an integral value of a value (Zi) obtained by the multiplication such that said integral value converges;
• and
• terminate repetition of the search cycle when said control parameters (Amn, Bmn, Cmn) converge or when a predetermined time elapsed, and determining m search values (Amn′, Bmn′, Cmn′) that are values of said m control parameters (Amn, Bmn, Cmn) corrected by said correction values at the termination of repetition of search cycles with respect to each of said m control parameters (Amn, Bmn, Cmn),
• wherein said providing starting values comprises providing starting values for the next generation based on said m search values (Amn′, Bmn′, Cmn′), and when said m search values converge or when a predetermined number of generations has been reached, outputting said m search values as optimum control parameters.