Optimal Self-Tuning PID Controller Based on Low Power Consumption for a Server Fan Cooling System

Abstract

Recently, saving the cooling power in servers by controlling the fan speed has attracted considerable attention because of the increasing demand for high-density servers. This paper presents an optimal self-tuning proportional-integral-derivative (PID) controller, combining a PID neural network (PIDNN) with fan-power-based optimization in the transient-state temperature response in the time domain, for a server fan cooling system. Because the thermal model of the cooling system is nonlinear and complex, a server mockup system simulating a 1U rack server was constructed and a fan power model was created using a third-order nonlinear curve fit to determine the cooling power consumption by the fan speed control. PIDNN with a time domain criterion is used to tune all online and optimized PID gains. The proposed controller was validated through experiments of step response when the server operated from the low to high power state. The results show that up to 14% of a server’s fan cooling power can be saved if the fan control permits a slight temperature response overshoot in the electronic components, which may provide a time-saving strategy for tuning the PID controller to control the server fan speed during low fan power consumption.

Self-Tuning Fully-Connected PID Neural Network System for Distributed Temperature Sensing and Control of Instrument with Multi-Modules

“…The structure of the PIDNN controller is a three-layer network whose hidden layer neurons’ activation functions work as a PID controller [15,16,17]. The improved FCPIDNN controller proposed in this paper takes a series of PIDNN controllers as basic controllers and adds a full connection layer between the basic controllers and the cooling fans.…”

“…T i is the actual local temperature. Y i is the output of the PIDNN controller [15]. There are three layers in the PIDNN, including an input layer, a hidden layer, and an output layer.…”

“…The proportional–integral–derivative neural network (PIDNN) controller is independent of system modeling, and it is suitable for fan speed control [15,16,17]. Lee et al [15] presented a PIDNN controller for a server fan cooling system.…”

“…Lee et al [15] presented a PIDNN controller for a server fan cooling system. Rossomando et al [16] used an adaptive neural PID controller for mobile robots’ trajectory tracking control.…”

A simple but energy-efficient HVAC control synthesis for data centers

“…The corresponding numbers of publications and patents are increasing rapidly. 2) From an applied control engineering perspective, on/off and PID controllers seem to be widely used (see, e.g., [9], [14], [15], [32], [37], [40], and the references therein). To a large extent this situation is explained by their conceptual simplicity.…”