Soft sensor based dynamic flow rate estimation in low speed radial pumps

Automatic controller tuning for soft sensor based flow rate control

“…Both limiting cases are captured by boundary curves, which are used to estimate steady state parameters. The boundary curves can be determined automatically by a series of measurements carried out by the pump itself (Leonow and Mönnigmann, 2013). In addition to measuring and estimating steady state parameters, step responses are used to determine a dynamic model of the motor current sensor, i.e., the relation between the rotational speed n(t) and the measured motor current i meas (t).…”

“…K (n, q) is a nonlinear function that is estimated from steady state data. The constant parameters T L and T D can be identified from step responses (see Leonow and Mönnigmann (2013) for details).…”

“…by Ahonen et al (2012). We reviewed this and other methods in Leonow and Mönnigmann (2013) and proposed the boundary curve method as a new flow rate estimation algorithm with improved steady state and transient performance. We briefly summarize this method, since it provides the motor current sensor model (1) required by the process identification in section 3.…”

“…Automatic control of pumps is unpopular among practitioners, since it requires sensors, which result in unacceptable additional costs. We have recently reviewed existing flow rate estimation methods and presented a dynamic flow rate estimation method that uses soft sensors in Leonow and Mönnigmann (2013). The proposed soft sensors can easily be integrated into the pumping unit and a flow rate sensor is therefore obsolete.…”

“…We present an automatic PID tuning method that is based on the root locus plot and handles deadtime explicitly for strictly delayed processes of arbitrary degree. We apply the tuning approach in combination with the flow rate estimation algorithm presented in Leonow and Mönnigmann (2013) and show that the tuning method ⋆ This research was funded by the Deutsche Bundesstiftung Umwelt DBU.…”