On time-variable BP neural network with application in body-bio-model tracking

A novel structure with time-memory function and time-shift terms of dynamic BP neural network (NDBP) using process-tracking training-algorithm based on the system-identification and statistic theory is developed in the CAD-grid 2# (China-Austria-Data-Grid-cooperation-project) and CEFSP 2008 5# (Chinese Education Foundation Science Project). This NDBP is used specially for real-time modeling for tracking the strong-time-variable dynamic process. Owing to its memory- and forgetting- ability, the newer output errors strongly effect the weight-correction. Both advantages, robust process-function approach and high speed process-characterisytic tracking, are possessed by the NDBP facing the urgent requirement of the research fields such as bio-chemical-process analysis, on-line diagnosis medical measurement, industrial adaptive process control, nature synusiologic development monitoring, and so on. The basic principle, algorithm composition and description, execution effect and typical application in medical measurement are introduced. The entire algorithem is supported by the newly developped Cloud-computing system.