Chyr Pyng Liou

Pipeline Leak Detection by Impulse Response Extraction

A systems response to an impulse can be used to detect and diagnose abnormalities. The impulse response can be extracted by using cross-correlations between a low amplitude pseudo random binary disturbance input and the systems output. This fact is applied to pipeline hydraulics as a means of real-time non-interruptive integrity monitoring. A method of generating the pseudo random binary disturbance is proposed. The extraction of a pipelines impulse response with the presence of noise is investigated. The features of the response of an intact pipeline and characteristic changes in the impulse response as a result of a leak are established. The feasibility of using impulse response to detect and to locate a leak in real-time is demonstrated.

Recirculation in an Annular-Type Jet Pump

For some annular-type jet pump applications, it is important to avoid formation of a recirculation zone in the mixing region. The goals of this research were to find (i) when recirculation occurs and (ii) the size and location of the resulting recirculation zone. Experiments were performed using air in a straight-walled, annular-type, ducted jet. Area ratio A j /A s varied from 0.39 to 0.89; here, A is flow area, and j and s identify the jet and secondary flows, respectively. Data showed that recirculation correlates with J, where J P j /(P j + P s ), and P is rate of momentum. For the area ratios studied, recirculation begins when J exceeds a value ranging from 0.89 to 0.94. This paper also presents data showing the recirculation zone boundaries and presents a discussion of jet pump design

Mass Imbalance Error of Waterhammer Equations and Leak Detection

The waterhammer equations can be used to compute mass inventory changes for pipeline leak detection. In such applications the pipeline is monitored continuously in time and there is no clear demarcation between steady-state and transients. This paper addresses a mass imbalance error of the waterhammer equations. The authors establish that the net mass influx error for the waterhammer equations at the steady-state with maximum flow-rate bounds the same error during transients. A method to evaluate this bound is presented. Using petroleum products pipelines as an example, they show that the waterhammer equations are not accurate enough for leak detection in some systems.

Maximum Pressure Head Due to Linear Valve Closure

The maximum pressure head resulting fron one-speed closure of wide open valves is investigated. The dimensionless variables formulated in this study make the subtle effect of the initial valve head loss explicit and separate from that of the pipe frictional head loss. The maximum head is related to initial pipe frictional head loss, the initial valve head loss, the inherent flow characteristic of the valve, and the closure period by plots of dimensionless variables.

A model for vertical frazil distribution

A model is presented for the evolution of frazil over depth and with time in a turbulent flow. The net upward migration due to buoyancy of the frazil is opposed by intermittent mixing induced by large energy-containing eddies. A surface renewal model is used to describe the effects of large eddy mixing. Parameters that represent an entire water body are obtained by averaging those of discrete water columns using a probability density function. These parameters include the concentration profile, the surface age, and the surface layer thickness. A dimensionless surface renewal frequency characterizes the frazil distribution at equilibrium. The rate of heat loss from the water surface, the surface renewal frequency, and the critical surface layer thickness determine whether the frazil will evolve toward a well-mixed equilibrium state or a layered state. The model provides a physical basis for understanding the transition between these states, consistent with existing empirical criteria and field data.

Approximation of the Friction Integral in Water Hammer Equations

AbstractA method is proposed to improve the accuracy of the approximation commonly used with the friction term in the water hammer equations. This method partially corrects the truncation error of the friction integral by incorporating the truncation error associated with a linear flow profile into the solution algorithm. The effectiveness and utility of this accuracy enhancement are demonstrated by examples.