Computer Simulations of FIFZN (Fuzzy Inverse-Free Zhang Neurodynamics) with Expected Precision Adaptively Satisfied Handling TVQP (Temporally-Variant Quadratic Programming)

Some models and algorithms have been used to solve temporally-variant (i.e., time-dependent) quadratic programming (TVQP) in the last few years. But most of the models and algorithms contain inverse matrix and temporally-invariant sampling gap, which may lead to too much computation consumption that we do not want. In this paper, a sub-model based on inverse-free (IF) continuous Zhang neurodynamics (a.k.a., zeroing neurodynamics), IFCZN, is developed for solving the TVQP. A fuzzy system sub-model is designed and added to adaptively adjust the sampling gap under the expected precision. Aided with the fuzzy system sub-model, an inverse-free discrete algorithm, i.e., FIFZN (fuzzy inverse-free Zhang neurodynamics) discrete model, a.k.a., FIFZN algorithm, is further presented. In addition, the verification of the effectiveness and superiority of the FDZN (fuzzy discrete Zhang neurodynamics) algorithm via numerical experiments about four specific TVQP examples is conducted.