The first part of this paper presents the theoretical model and simulation results for primary chilled water system and secondary system nonlinear characteristics. An integrated control linearization method is developed to use system flow resistance (SFR) for chilled water pump speed control optimization while maintain a stable control performance by applying constant-gain PI controls. The new control algorithm integrates the secondary system flow resistance into water loop differential pressure reset schedules. Both traditional and optimal control algorithm were implemented and compared in two full-scale real building experiments. Field investigation is further conducted to validate the simulation results and performance improvement with the new control algorithm. Compared to the prior art, this innovative optimal algorithm created a system that was more energy efficient, given the desired control performance, and resulted in increased control stability. The linearization algorithm is advantageous in that it results in lower pump head and improves pump efficiency, better coil-valve control stability, and significantly reduces thefrequency of control valve repositioning.