摘要：AbstractIn this paper, we show that previously studied generalized distributed optimization algorithms based on passivity described by time differential equations can also be realized using a form of time difference equations that satisfy a global stability on optimal solutions by employing the notion of passivity or dissipative property. First, we show that it is not straightforward to transform the continuous-version to the discrete-version without losing the passivity of the original algorithms as input/output systems. Then, by using the Euler approximation of time differentiation and relaxation of the passivity to a dissipative property, we show that an appropriate transformation is possible and derive the generalized discrete time distributed optimization algorithms which guarantees a global stability.
关键词：Keywordsdiscrete time systemdistributed optimization algorithmpassivitydissipative system
