Measurement uncertainty in water distribution telemetry systems

Hainsworth, GD, 1988. Measurement uncertainty in water distribution telemetry systems. PhD, Nottingham Trent University.

Preview

Text 10290063.pdf - Published version Download (25MB) | Preview

Abstract

Efficient control of a complex water distribution system requires accurate information about its current operating state. At present in the water industry, modern telemetry hardware systems are being installed to meet these needs. Unfortunately, due to financial constraints, it is not practical to measure all variables with such a system. Therefore, the information supplied by the telemetry system must be supplemented by pseudomeasurements, such as predictions of consumption at the nodes of the network, before a full picture of its operating state can be calculated. These pseudomeasurements are only estimates and hence, contain a great deal of uncertainty. The real meters linked to the telemetry system, although not completely accurate, provide more reliable data. Effectively, they strengthen the monitoring system by reducing the reliance on less accurate pseudomeasurements.

Measurement uncertainty clearly has an impact on the accuracy to which state estimates can be calculated. The precise nature and level of this impact is investigated in this thesis. A network model which allows for measurement uncertainty, is presented. From this model, algorithms are derived which quantify the effect of measurement uncertainty on the accuracy of the derived state estimates. Rather than a single, deterministic state estimate, the set of all feasible states, given the level of measurement uncertainty, is calculated. This set is presented in the form of upper and lower bounds for the individual variables, and hence provides limits on the potential error of each variable. A water distribution system simulation program, TCLAS, that calculates state estimates in this way, has been developed. This program is also described in the thesis.

The location of meters about the network strongly influences the accuracy of state estimates. By carefully designing the meter placement in the telemetry system, it is possible to achieve a much higher level of monitoring accuracy. The problem of how best to design this metering - the optimal meter placement problem - is also addressed, and is presented as a mathematical optimisation problem. The mathematical formulation allows a great deal of flexibility in the choice of cost and constraint functions, so that realistic design objectives and telemetry system restrictions can be modelled. Two optimisation algorithms are presented as solutions to this problem. In addition, it is described how TCLAS can be used to experiment with the location of meters and assess the accuracy of these placements.

Item Type:	Thesis
Creators:	Hainsworth, G.D.
Date:	1988
ISBN:	9781369323122
Identifiers:	Number Type PQ10290063 Other
Divisions:	Schools > School of Science and Technology
Record created by:	Linda Sullivan
Date Added:	16 Jun 2021 11:27
Last Modified:	22 Sep 2023 13:29
URI:	https://irep.ntu.ac.uk/id/eprint/43085

Actions (login required)

Edit View

Statistics

Views

Downloads