Modelling Heat Transfer in the Cooling Pass of a Refuse-Fired Fluidised Bed Combustor

Detta är en Uppsats för yrkesexamina på avancerad nivå från Chalmers tekniska högskola/Institutionen för energi och miljö

Författare: Louise Axelsson; [2012]

Nyckelord: Model; Heat Transfer; Radiation; CFD; Power Boiler; EfW; FBC;

Sammanfattning: In EfW (Energy from Waste) boilers, flue gases can create corrosion of the superheaters and economisers, caused by deposits of the melted ashes. Due to this, the flue gases are cooled with water wall panels and water tube walls in a cooling pass before reaching the heat transfer surfaces, thus limiting the power output. One way to increase the efficiency of the boiler is to introduce superheater screens in the cooling zone instead of using water wall panels to cool the flue gases. This new superheater would be covered in a ceramic material to protect its tubes from the above-cited fouling. The purpose of the present work was to create a 1D model that describes the heat transfer from the flue gases to the coolant (steam and water) in the flue gas cooling pass of a power boiler. The 1D model, which is validated against CFD simulations using the Discrete Ordinates (DO) radiation model, will be used in the retrofit of EfW boilers. Furthermore, CFD simulations are used to study the pressure loss and mass flow distribution in the cooling pass under different situations and to provide design correlations. Also, the influence of different parameters on the computational cost of the CFD simulations is examined. The 1D model agreed fairly well with the CFD simulation, especially in the lower part of the cooling pass. The flue gas outlet temperature was 4.5% lower in the 1D model than in the CFD simulation. In the 1D model, the geometry could be approached as rectangular canals, neglecting the inclination of the ceiling, and modelling the inlet and the outlet of the cooling pass as water tube walls. Regarding the flow field, the mass flow distribution in the cooling pass was found to be independent of factors such as the mass flow rate and the number of screens. It is sufficient to calculate the relative depths of the canals to get the mass flow distribution. The pressure loss can be found as a function of the average flue gas velocity in the cooling pass. The CFD mesh required approximately half a million cells, and the computational cost was decreased by using a symmetry plane in the middle of the cooling pass, costing just over an hour.

  KLICKA HÄR FÖR ATT SE UPPSATSEN I FULLTEXT. (PDF-format)