Page - VII - in Adaptive and Intelligent Temperature Control of Microwave Heating Systems with Multiple Sources

[B(k)] Temperaturechangingmatrix includingallBn,m(k)with 1≤m≤M and1≤n≤N K Φn(k) Temperature increasingmatrixat then-th location K Ψ(k) Temperature increasingvectorof MIMOform K Incontrollerdesign: J(k) Cost/objective functiontobeoptimizedat timek [Γ] Positivedefinite tuningmatrixusedin thecost function Vci The i-th initializedcontrol sequence in theGA algorithm Yci PredictedtemperaturesusingV c i w Weightsvector/matrixusedin theNN S(k) Thestateof theMDPat timek R(k) Therewardreceivedfromtheplantat timek pi(s,u) ControlpolicyusedinRL Vpi(s) Thestatevalue functionunder thestatesusing thepolicypi Qpi(s,u) Thestate-actionvalue functionof thestate-action pair (s,u)usingthepolicypi ∆td Thetemporarydifferenceerror e(s,u) Theeligibility tracedefinedinRL vii ListofSymbols