Design and modeling of an experimental tilapia and African catfish recirculating aquaculture system

Pretorius, Ruan (2020-03)

Thesis (MEng --Stellenbosch University, 2020.

Thesis

ENGLISH ABSTRACT: This study aimed to design a new warm-water recirculating aquaculture system (RAS) that will have the flexibility to rear both African catfish (Clarias gariepinus) and Mozambique tilapia (Oreochromis mossambicus) using an engineering modeling approach. The facility was to be built within existing buildings and should maintain stable conditions when rearing either African catfish or Mozambique tilapia preferably at their commercial stocking densities of 200 kg/m3 and 100 kg/m3 respectively. Empirical data and verified models were obtained from literature to accurately describe the RAS. Unit operations were designed to achieve biofiltration by nitrification, solids removal, aeration, CO2 degassing, UV sterilization and temperature control. The unit operations were sized for peak loading conditions which occur at the end of a trial when the feeding rate was highest. The stocking density was adjusted to 50 kg/m3 due to limited footprint available for the wastewater treatment equipment required. At a peak biomass of 324 kg, the feeding rate was 10.3 kg/day and resulted in the estimated waste production rates of 505 g TAN/day, 2 333 g TSS/day, 1 256 g DOM/day, 3 490 g CO2/day and oxygen consumptions of 2 538 g/day, 1 930 g/day and 2 028 g/day by fish, heterotrophic bacteria and autotrophic bacteria respectively. The biofilters designed for the RAS were six moving bed bioreactors (MBBRs) of 1.0 m3 each with 65% filling of 300 m2/m3 biomedia to maintain TAN concentrations in the fish tanks below 3.0 mg/L and were sized by considering both ammonium and oxygen reaction rate limiting substrates. A lamella settler with 5.5 m2 settling area was designed to remove solids and was sized using experimentally measured settling rates of African catfish solids, which estimated a total suspended solids removal efficiency of 80%. A submerged in-tank aeration system for the fish tanks (110 std m3/h) and MBBRs (25 std m3/h) was designed using empirical data of various submerged diffuser depths and is estimated to maintain oxygen above the required minimum of 5 mg/L and limit CO2 concentrations below 15 mg/L. The required recirculation flow rate to prevent accumulation of waste products in the fish tanks was estimated to be 12 m3/h. Heat transfer models were used to estimate the heat loss from the process water and building air. The estimated peak heat loss of the water at 27oC was estimated to be 8.9 kW and is to be supplied by a heat pump. The largest heat loss contributor was evaporation at the fish tank surfaces. The capital cost of the proposed RAS was estimated to be R 145 000. The annual operating costs of electricity (55.7% of total), make-up water (44% of total) and sodium bicarbonate for alkalinity control (0.2% of total) resulted in a total annual operating cost of R 79 000. There was concluded that this study presents a thorough comparison of RAS technologies based on their advantages and disadvantages. The study also presents verified modeling methodology for conceptual RAS design and demonstrates the use by proposing a RAS design that is ready for implementation.

AFRIKAANSE OPSOMMING: Hierdie studie het beoog om ’n nuwe warmwater hersirkuleringsakwakultuurstelsel (RAS) te ontwerp wat die buigbaarheid het om beide Afrika-baber (Clarias gariepinus) en Mosambiekse kurper (Oreochromis mossambicus) te kweek deur ’n ingenieursmodelleringsbenadering te gebruik. Die fasiliteit moes gebou word binne bestaande geboue en moet stabiele kondisies behou wanneer óf Afrika-baber óf Mosambiekse kurper verkieslik by hul kommersiële voorraaddigthede van 200 kg/m3 en 100 kg/m3 onderskeidelik gekweek word. Empiriese data en bevestigde modelle is verkry uit literatuur om die RAS akkuraat te beskryf. Bedryfseenhede is ontwerp om biofiltrasie deur nitrifikasie, verwydering van vaste stof, belugting, CO2-ontgassing, UV-sterilisasie en temperatuur kontrole te bereik. Die bedryfseenhede is geskaal vir kondisies van spitslading wat plaasvind teen die einde van ’n toets as die voertempo op sy hoogste was. Die voorraaddigtheid is aangepas na 50 kg/m3 as gevolg van beperkte staanoppervlakte beskikbaar vir die afvalwaterbehandeling se toerusting wat benodig word. By spitsbiomassa van 324 kg, was die voertempo 10.3 kg/dag en het die beraamde afvalproduksietempo’s van 505 g TAN/dag, 2 333g TSS/dag, 1 256 g DOM/dag, 3 490 g CO2/dag en suurstofgebruik 2 538 g/dag, 1 930g/dag en 2 028 g/dag deur visse, heterotrofiese bakterieë en outotrofiese bakterieë onderskeidelik tot gevolg gehad. Die biofilters ontwerp vir die RAS was ses bewegende bed-bioreaktors (MBBR) van 1.0 m3 elk met 65% vulling van 300 m2/m3 biomedia om TAN-konsentrasie in die vistenks onder 3.0 mg/L te behou en is geskaal deur beperkende-substrate van beide ammonium- en suurstof-reaksietempo’s in ag te neem. ’n Lamella besinker met 5.5 m2 besinkingsarea is ontwerp om vaste stowwe te verwyder en is geskaal deur ekperimenteel-gemete besinkingstempo’s van Afrika-baber se vaste stowwe te gebruik, wat ’n totaal verwyderingsdoeltreffendheid van hangende vaste stof van 80% beraam het. ’n Onderdompelde binne-tenk belugtingsisteem vir die vistenks (110 std m3/h) en MBBRe (25 std m3/h) is ontwerp deur empiriese data van verskeie onderdompelde verspreier dieptes te gebruik en is beraam om suurstof bo die vereiste minimum van 5 mg/L te handhaaf en CO2-konsentrasies onder 15 mg/L te beperk. Die vereiste sirkulasie vloeitempo om akkumulasie van afvalprodukte in die vistenks te verhoed, is beraam om 12 m3/h te wees. Hitte-oordragmodelle is gebruik om die hitteverlies van die proseswater en geboulug te beraam. Die beraamde spitshittteverlies van die water by 27°C is beraam om 8.9 kW te wees en is voorsien deur ’n hittepomp. Die grootste bydraer van hitteverlies was verdamping by die vistenkoppervlak. Die kapitaalkoste van die voorgestelde RAS is beraam om R 145 000 te wees. Die jaarlikse koste van elektrisiteit (55.7% van totaal), aanvullingswater (44% van totaal) en koeksoda vir alkaliniteitbeheer (0.2% van totaal) het ’n totale jaarlikse bedryfskoste van R 79 000 tot gevolg gehad. Daar is tot die gevolg gekom dat hierdie studie ’n deeglike vergelyking van RAS-tegnologieë lewer gebaseer op hul voordele en nadele. Die studie het modellering metodologie vir konsepsuele RAS-ontwerp bewys en stel die gebruik daarvan voor deur ’n RAS-ontwerp wat reg is vir implementasie.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/108203
This item appears in the following collections: