CFD (computational fluid dynamics) modelling of

5 downloads 0 Views 64KB Size Report
The PHOENICS CFD model produced by Concentration, Heat and ... of PHOENICS for modelling waste stabilization pond hydraulics refer to Shilton (2001).
A.N. Shilton* and D.D. Mara* * *Centre for Environmental Technology and Engineering, Massey University, Private Bag 11 222, Palmerston North, New Zealand (E-mail: [email protected]) **School of Civil Engineering, University of Leeds, Leeds LS2 9JT, UK Abstract CFD modelling of the incorporation of two baffles equally spaced along the longitudinal axis of the pond and with a length equal to 70% of the pond breadth, indicated a potential improvement in the removal of E. coli in a 4-day secondary facultative pond at 258C from 5 £ 106 per 100 ml in the effluent from a 1-day anaerobic pond to 4 £ 104 per 100 ml; the reduction in an un-baffled pond was an order of magnitude less effective. The addition of a similarly baffled 4-day primary maturation pond reduced the effluent E. coli count to 340 per 100 ml; the reduction in an un-baffled series was two orders of magnitude less effective. Well designed baffles thus have considerable potential for reducing pond area requirements and hence costs in the hot tropics. These very promising results highlight the need for field studies on baffled pond systems to validate (or allow calibration) of the CFD model used in this study. Keywords Baffles; CFD modelling; design; E. coli; hydraulics; removal; tropical climates; waste stabilization ponds

Introduction

Waste stabilization ponds (WSP) are most efficient in tropical areas where in-pond temperatures are high and decay/reaction rates are fast. The design is typically focused on the reduction of pathogens, as it is not uncommon for the pond effluent to be utilized for crop irrigation. The World Health Organization guidelines for the microbiological quality of treated wastewaters used for this purpose are as follows (WHO, 2005): (a) Restricted irrigation: # 105 E. coli per 100 ml and # 1 human intestinal nematode egg per litre (reduced to # 0.1 egg per litre if children under 15 are exposed). (b) Unrestricted irrigation: # 1000 E. coli per 100 ml and the same egg numbers. The WSP designer’s goal is to optimise pond design by minimising cost and land required while maintaining satisfactory treatment. Using CFD modelling, Shilton and Harrison (2003a) showed that the incorporation of baffles could dramatically improve removal of indicator bacteria in primary facultative ponds (Table 1). Their work considered a single primary facultative pond in a temperate region with a design temperature of 148C, a wastewater flow of 10,000 m3 d21 and a resultant theoretical hydraulic retention time of 31 days. WSP in tropical countries have higher design temperatures, faster decay rates and hence shorter retention times than those investigated by Shilton and Harrison (2003a). Unfortunately, this combination of faster decay rates and shorter retention times make the impact of any hydraulic short-circuiting on treatment efficiency much more significant. This implies that use of effective baffling to mitigate short-circuiting is even more important in tropic regions than it is in the temperate ones previously studied. To date, information on the use of baffling for optimizing the design of tropical WSP has been lacking.

Water Science & Technology Vol 51 No 12 pp 103–106 Q IWA Publishing 2005

CFD (computational fluid dynamics) modelling of baffles for optimizing tropical waste stabilization pond systems

103

Table 1 The effects of baffling on E. coli removal in primary facultative ponds at a design temperature of 148C and for an influent E. coli count of 1 £ 108 per 100 ml (Shilton and Harrison, 2003a) In practice for different ponds the resultant performance values will depend on the influent concentration, retention time, temperature and many other variables Number of baffles

A. N. Shilton and D.D. Mara

None 1 2 4 6 8

E. coli per 100 ml of effluent

6.2 4.1 6.0 3.9 5.7

£ 106 £ 106 £ 103 £ 102 £ 102 10

In this study we have used a CFD model with integrated first-order decay kinetics to evaluate the effect of baffles on E. coli removals in secondary facultative and primary maturation ponds in the hot tropics at a design temperature of 258C.

Methods

The secondary facultative pond was designed for 258C following standard design procedures (Mara, 1997) for a raw wastewater with a BOD and an E. coli count of 300 mg l21 and 5 £ 107 per 100 ml, respectively; these were assumed to have been reduced to 90 mg l21 and 5 £ 106 per 100 ml in a 1-day anaerobic pond. The facultative and maturation pond dimensions were taken as 640 £ 320 £ 1.5 m (the same model dimensions used by Shilton and Harrison, 2003). A wastewater flow of 76,800 m3 d21 was used giving a theoretical hydraulic retention time of 4 days. As was the case in the Shilton and Harrison (2003a) study, this project also used a pond with a length-to-breadth ratio of 2 to 1; a pond depth of 1.5 m; and baffles equally spaced along the longitudinal axis of the pond with a length equal to 70% of the pond width. Both a facultative pond and a maturation pond were modelled with and without the provision of two baffles. Single baffles were not considered as Shilton and Harrison (2003a) had found them to be generally ineffective (Table 1). The provision of more than two baffles in the facultative pond was not considered in order to avoid significant variation in organic loading across this pond. Another option could have been to use an un-baffled facultative pond followed by a maturation pond with four baffles. However, it can be deduced from the work of Shilton and Harrison (2003a) that this would not be nearly as efficient as the provision of two baffles in each pond. The PHOENICS CFD model produced by Concentration, Heat and Momentum Ltd (London) was used in this work. The pond models were solved for fluid flow (via the solution of pressure, momentum in three dimensions and turbulence parameters) and first-order E. coli die-off at 258C. For full details of the use and evaluation of the validity of PHOENICS for modelling waste stabilization pond hydraulics refer to Shilton (2001). Shilton and Harrison (2003b) discuss and validate the technique of integrating decay into a CFD model. Results

104

The CFD modelling of the facultative pond on its own and the facultative and maturation ponds in series, both with and without two 70%-width baffles in each pond and operating at 258C, yielded the following results for the effluent E. coli count per 100 ml for the five pond systems investigated (see Figure 1).

Design A Un-baffled facultative pond: 4 × 105 5 × 106

5 × 106

5 × 106

3.3 × 103

Design E Twin-baffled facultative pond followed by a twinbaffled maturation pond:

5 × 106

3.4 × 102

Design C Un-baffled facultative pond followed by an un-baffled maturation pond:

5 × 106

A. N. Shilton and D.D. Mara

Design B Twin-baffled facultative pond: 4.1 × 10 4

Design D Un-baffled facultative pond followed by a twin-baffled maturation pond:

3.2 × 104

Figure 1 Five pond systems studied

It is possible to calculate the performance of Design A (the un-baffled facultative pond) by simply using the completely mixed flow equation as proposed by Marais (1966). This approach yields a value of 2 £ 105. Considering that the results from different designs vary by orders of magnitude, this result is obviously very similar to that predicted by the CFD simulation. As would be expected, the result from the CFD model, which simulates the actual hydraulic flow pattern in the pond (and its deficiencies), is somewhat higher than that which is predicted by the completely mixed flow equation which assumes perfect mixing. The incorporation of twin 70%-width baffles in the facultative pond achieves an improvement in E. coli reduction of an order of magnitude. The twin-baffled facultative pond produces essentially the same effluent quality as the un-baffled facultative pond and un-baffled maturation pond combined.

Discussion Practical implications of results

The modelling predicted that restricted irrigation requirements (# 105 E. coli per 100 ml) could be met using both Design B (twin-baffled facultative pond system) or Design C (un-baffled facultative pond followed by one un-baffled maturation pond). However, compared with Design C, the use of baffling in Design B reduces the pond area requirement by almost 50%. Where an effluent suitable for unrestricted irrigation is required, the modelling showed that Design E (twin baffled facultative pond followed by a twin-baffled maturation pond) reduces effluent quality to # 1000 E. coli per 100 ml. To achieve this effluent quality without any baffling, the area required would be almost doubled. The use of four baffles in the primary maturation pond was originally considered, but after obtaining the results given above this was clearly not warranted. Installing baffles is not, of course, without cost. However, given the above findings that the use of baffles in tropic climates can dramatically reduce pond area requirements, it can be expected that the cost of baffling would be more than offset by the savings in land and pond construction costs.

105

CFD modelling for pond design assessment

A. N. Shilton and D.D. Mara

CFD models of WSP attempt to represent a complex and dynamic environment. Just as is the case for many other engineering models, CFD pond models are not be intended to produce an exact result, but rather as a tool for exploring various designs alternatives. Like all practical engineering equations and models, the CFD pond model has to make various simplifications and assumptions. With regard to its ability to predict hydraulic behaviour, confidence has been developed by previous studies into the application of CFD modelling to WSP, particularly in terms of its ability to assess large ‘step changes’ in performance that result from different design configurations (Shilton, 2001). A weakness of the model does lie in the simple assumption of uniform first-order decay kinetics throughout the pond volume. However, this assumption is still widely used in pond design equations. Compared with these traditional design equations, the advantage of using the CFD simulation is its ability to assess the likely impact that hydraulic modifications, such as baffles, can have on performance. While only extensive field-testing can conclusively validate these results, it would nevertheless appear that there is certainly large potential for significant cost optimization to be achieved by the intelligent incorporation of baffles in facultative and maturation ponds in tropical countries. Conclusions

This modelling has shown that the provision of two 70%-width baffles in secondary facultative and primary maturation ponds at 258C significantly increases E. coli removals in these ponds. With respect to the particular pond configuration/parameters used in the CFD modelling undertaken for this study, the results showed that the combination of a 1-day anaerobic pond and a 4-day twin-baffled facultative pond could produce an effluent suitable for restricted crop irrigation. Additionally it was shown that an effluent suitable for unrestricted crop irrigation could be achieved by addition of a 4-day twin-baffled primary maturation pond. These findings show that there is significant potential for size reduction and cost optimization to be achieved by the incorporation of properly designed baffles in ponds in tropical climates.

References Mara, D.D. (1997). Design Manual for Waste Stabilization Ponds in India, Lagoon Technology International, Leeds, UK. Marais, G.v.R. (1966). New factors in the design, operation and performance of waste stabilization ponds. Bulletin of the World Health Organization, 34(5), 737 – 763. Shilton, A.N. (2001). Studies into the Hydraulics of Waste Stabilization Ponds PhD thesis, Massey University, Palmerston North, New Zealand. Shilton, A.N. and Harrison, J. (2003a). Guidelines for the Hydraulic Design of Waste Stabilization Ponds, Institute of Technology and Engineering, Massey University, Palmerston North, New Zealand. Shilton, A. and Harrison, J. (2003b). Integration of coliform decay within a CFD model of a waste stabilisation pond. Water Science and Technology, 48(2), 205 –210. WHO (2005). Guideline s for the Safe Use of Wastewater in Agriculture, World Health Organization, Geneva, Switzerland (in press).

106