liquid waste or sewage discharged into a river or the sea". Effluent in the artificial sense is in general considered to be water pollution, such as the outflow from a sewage treatment facility or the wastewater discharge from industrial facilities.
The permeate quality from both membrane plants was both reliable and in line with previous values found from pilot plant work. Feed quality to the RO plant can show large variation with rainfall; for instance, a conductivity of between 800 and 1200 μS cm−1. Permeate quality from the RO plant is 38.5 μS cm−1 representing a rejection of 96.7% on average. Overall rejection of TDS is 93% with specific rejections of sodium and calcium of 95 and 99% respectively (Table 5.20).
The main issue of concern was membrane fouling reducing output from the MF plant and increasing pressure on the feed stream of the RO plant. Initial operation of the plant showed excellent water quality but high RO fouling rates. The RO feed pressure increased at a rate of 1 bar per hour at worst due to rapid build up of . Changes to operation has effectively controlled the problem and membrane cleaning is now required every 6 months on average.
The Flag Fen plant is currently operating well and exceeding the target water qualities set down in the original negotiations. The quality of the delivered water has enabled ultrapure water production at the power station to increase by 20%. Coupled to this is a reduction of over 90% in the costs of ion exchange regeneration due to an increase in the operating cycle of the twin bed demineralisation plant from 8 to 60 hours. Overall, 1250 m3 of tap water per day have been saved which has reduced the station's total water use by 11%. Client receptivity to the scheme is very high as supply is guaranteed and operating costs decreased. The high-purity water plant cost around £1 000 000 ($1 524 000) and each membrane stage has an operating cost of 7.5p m−3 ($0.11 m−3). The success of the scheme has attracted media attention and the project has won a number of prestigious awards including the 2000 Water UK/Environment Agency water, the 2001 IChemE innovation efficiency award and the 2001 Green apple award.In order to demonstrate the objectives of this study two production fields and two nearby TSE plants are considered where the total water demands for field 1 (f1) and field 2 (f2) are 4.25 × 106 m3 and 5.46 × 106 m3 respectively. The model developed as part of this research is able to capture the variation of the crop water demand during the hot and cold seasons, where the output of the optimized feedforward illustrates agreement with the observed data as illustrated in figure 3(a) below. The accuracy of the output is validated using a correlation coefficient (0.983) and error measurement indices, root mean squared percentage error (11.93). After validation, a scenario is assumed for the one-year forecasting. The temperature observations are multiplied by random numbers generated between 1 and 1.1, by taking the assumption that the temperature will vary up to 10% above the previous observations. Comparison of forecasted year with previous years demonstrates that the crop water demand will increase with the maximum periods from April to August