Peter Skipworth

Rehabilitation strategies for water distribution networks: a literature review with a UK perspective

Abstract Primarily, a rehabilitation strategy should aim to satisfy the regulatory requirements set down in respect of water distribution network operation. However, water companies in the UK have come to recognise that the business needs associated with the improvement of the deteriorating fabric of their distribution networks extend beyond these requirements. Extra economy can be gained by operating the networks efficiently based on a rehabilitation strategy which considers the associated costs over an extended period. Economic, hydraulic, reliability and water quality performance criteria must be optimised as part of an effective strategy. Numerous rehabilitation decision making approaches have been presented. However, many have adopted flawed economic approaches and have been based inadequately on one or two selected performance criteria. Few models have considered the extended planning horizons associated with a whole-life costing approach to this problem. However, the multi-objective optimisation approaches which have been developed recently have the potential to be developed into the required whole-life costing model based on the appropriate economic model and performance criteria.

The first foul flush in combined sewers: an investigation of the causes

A previously developed numerical model, able to simulate the erosion from an organic in-pipe deposit, was used to investigate the physical parameters that control the first foul flush in combined sewers. Systematic adjustment of physical parameters indicated that the accurate characterisation of bed properties, i.e. the surface erosional strength and its variation with depth, was more important than the accurate description of the imposed hydraulic conditions.

Determining maintenance requirements of a water distribution network using whole life costing

A whole life costing (WLC) methodology has been developed for determining long term maintenance expenditure requirements for water distribution networks. The methodology utilises an accounting scheme that ties the costs incurred by the operator and other stakeholders to the attributes or performance that drive the costs. It has specifically been derived with the requirements placed by the regulatory regime on the water companies that operate in England and Wales in mind. Expenditure constraints are implied by the regulator through price caps that companies can charge their customers. Appropriate levels of expenditures included as part of the price cap determinations are required by the regulator to be economically robust and tied to the service received by the customers. Therefore, maintenance decisions must reflect more immediate concerns of meeting performance requirements, but must ensure that such levels are sustainable in the long term. The WLC methodology achieves this through an integrated platform that links costs identified within a structured accounting scheme with their performance based drivers commonly modelled based on historical data. Thus, a robust and fully auditable methodology is provided that can address the requirements of all stakeholders. This methodology is the basis for software (WiLCO) that provides decision support in determining appropriate pipe rehabilitation and operational strategy and thus expenditure levels over extended time horizons.

Measurement of Flow Volume at Small Wastewater Treatment Works using Dosing Syphons

Discharges from small wastewater treatment works (WwTW’s) in England and Wales are governed by consents from the Environment Agency (EA). These do not currently specify the volume of final effluent discharged to receiving waters. However, the EA’s new policy for discharges at small wastewater treatment works states that operators must be able to measure the size of the majority of these discharges by the year 2005 and will be required to report on their daily totalised flow volumes (DTFV’s) to the EA. Each of the ten regional water companies in England and Wales operate several hundred of these remotely located sites. This new regulation has therefore presented considerable challenges to measure these flows reliably and accurately at reasonable cost. Yorkshire Water Services (YWS), one of these water companies, commissioned studies to develop new flow measurement equipment that would achieve the required accuracy and be inexpensive to build, install and maintain. A sample of the small WwTW’s owned by YWS revealed that many of these sites were equipped with a small dosing syphon distributing settled effluent to a granular trickling filter. This paper describes the development of a simple monitoring device to be attached to dosing syphons to measure flowrates and hence DTFV. Laboratory testing demonstrated that the device could estimate flow rate to +/-2.9% over a range of steady flowrates commonly found at small wastewater treatment plants. A series of time varying diurnal flow tests were carried out. It was found that the DTFV could be estimated to an accuracy of +/4.6%. Introduction In 1998, the EA proposed that the amount of treated effluent discharged from small wastewater treatment works to receiving waters should be measured to within an accuracy of +/-8%, where the DTFV is greater than 5m. At plants where the DTFV is greater than 50m/day the outflows must be measured continuously at the same level of accuracy. However, at smaller works (where DTFV<50m/day), these measurements will not be required to be continuous, but monitoring will be required for discrete periods at the request of the EA. An initial scoping study identified that a practical and economical means of satisfying the EA’s requirements was not available. It was therefore decided to carry out a study to either modify existing technologies or develop new devices that could satisfy the EA’s new regulatory requirements. A number of criteria were defined, that any new device would have to meet, following the survey of a number of small WwTW’s. In addition to achieving the measurement performance demanded by the EA, any new measurement device must be: 1 Pennine Water Group, Department of Civil and Structural Engineering, University of Sheffield, Sir Frederick Mappin Building, Mappin Street, Sheffield S1 3JD, UK. Tel: +44(0)114 2225732 Fax: +44(0)114 2225700 e-mail: t.c.howes@sheffield.ac.uk 2 Yorkshire Water Services, Western House, Bradford BD6 2LZ, UK Copyright ASCE 2004 HMEM 2002 2 • Low in capital and installation costs given the large number of sites • Independent of both mains water and electricity supplies as these are unavailable at most of these sites • Capable of recording the value of the DTFV for extended periods without maintenance, in order to minimise operational costs • Capable of being easily attached and calibrated to current installations • Rugged and of low visual impact to reduce the risk of vandalism • Removable and reusable on other sites Dosing syphons distributing settled sewage onto filter beds were observed at many sites. The syphons were commonly located in a circular tank, located at the centre of the filter bed. Distribution arms were attached to the base of the tank, jets located on the distribution arms evenly applied effluent onto the filter beds. These were angled such that the momentum of the sewage released when the syphon operated caused the distribution arms to rotate (Figure 1). Figure 1 Dosing syphon assembly in the field Initially it was postulated that by measuring the capacity of the syphon tank and monitoring the period of the operation cycle of the syphon an estimate of “instantaneous” flowrate and ultimately DTFV, could be made within the accuracy stipulated by the EA. Initial Investigation of the Syphonic Discharge Cycle A small dosing syphon was installed in a tank in the Water Engineering Laboratory at the University of Sheffield. The two distribution arms, through which the syphon would periodically discharge sewage onto a filter bed, were replaced with short outlet pipes with orifice plates at their ends. These were fitted to simulate the range of hydraulic losses induced by the variety of distribution arm lengths, jet configurations and overall condition of the distribution arms that had been observed in the field. Orifice sizes were selected by summating the cross sectional areas of the discharge jets in the field and representing this by Filter Bed Syphon Tank Dosing Syphon Distribution Arm Effluent Inflow Copyright ASCE 2004 HMEM 2002 3 two orifices of equal total cross sectional area, such that similar hydraulic losses were achieved. Five sizes of orifice plate (18mm, 22mm, 25mm, 31mm and 40mm internal diameter) were used. Water was supplied to the syphon tank from the laboratory’s constant head tank. The flow was varied by means of a programmable control valve and measured using an independently monitored pre-calibrated ABB PSM class C 40mm flow meter, located in the inlet pipe to the syphon tank. Before the tests were started, the flow meter was verified as being accurate to within +/-2% of the manufacturer’s calibration by means of a volumetric measurement tank. A simple monitoring device consisting of a vertical array of two float switches was placed in the tank to measure the times at which particular water levels were reached. In the tests described here, the top switch was fixed at a height so that it triggered at a water level just below the maximum water level, to operate the syphon. The second switch was located 100mm below the top switch. The water was discharged from the tank via the dosing syphon’s short outlet pipes to a sump from where it was pumped back to the constant head tank. The switch array, flow meter and control valve were all monitored or controlled by a computer (Figure 2). Results from the flow meter and the switch array could thus be directly compared. Figure 2 Schematic diagram of the laboratory testing apparatus A series of steady flow tests were carried out in order to investigate the characteristics of the syphon cycle, by varying the steady discharge introduced into the syphon tank using the control valve and the distribution arm hydraulic losses by changing the orifice plates. Initially the complete cycle period was investigated. This was defined as the time for the water level in the tank to rise past a fixed point (monitored using the lower float switch), activate the syphon, drop as the syphon discharged, and then rise again to the same water level as the syphon tank refilled. It was found that when the cycle period (TC) was plotted against incoming flowrate, a characteristic “U-shaped” plot was obtained indicating a nonunique solution for flowrate against measured cycle period (Figure 3). The shape of the plot was caused by the syphon filling slowly and emptying quickly at low flowrates, and filling quickly and emptying slowly at high flowrates. It was also noted that the cycle period was dependent on the hydraulic losses within the distribution arms. As these losses increased, the cycle period for a set flowrate also increased. These two factors demonstrate that cycle period Constant Head Tank Control Valve Flow Meter Sump Syphon Tank Computer

Quantification of Mains Failure Behavior in a Whole Life Costing Approach to Distribution System Management

A whole life costing (WLC) approach to distribution network management should aim to achieve the lowest network provision and operating cost when all costs are considered to achieve standards enforced by regulation. Cognisance is to be taken of all relevant costs (direct and indirect, private and societal) in order to balance the needs of the service supplier, the customer, society and the environment in a sustainable manner. If this is achieved a WLC framework should enable least cost decisions to reflect the level of risk that a company is willing to tolerate. This approach necessitates estimation of future network performance including the risk of pipe failure. This paper presents the WLC framework which has been developed and a rationale of the style of analysis required to derive factors to describe the risk of failure based on the configuration, quality and extent of available data. From an analysis of homogeneous pipe groups within a large water company pipe asset database covering tens of thousands of kilometres of pipes relationships of failure rate against pipe material, diameter, density of connection and age are presented. Application of these relationships in the derivation of risk factors for use in the WLC approach is discussed. An outline of how costs associated with pipe failure can be incorporated in the WLC approach is given.