Calculating mean temperature in 3 dimensional room whit floor heating system, by using of neural network

In this Study, first, the model of System which was a 3 dimensional room with different boundaryConditions Stimulatedby Gambit& Fluent Softwares. Considering, the inner air of theroom incompressibility,boussinesq model, reformed model of k ??, radioactive model DO and simple algorithm, the stimulation of fluid flow and temperature distribution performed. After change of effective various parameters on the floor heating system in the model (room) such as pipe depth, water temperature, internal emissivity, air absorption Coefficient, window sizes, object size(object was in room) , distance between object and floor are used as educational data for neural network. The neural network produced by two layers: hidden layer, with 20 neural neurons and arbitrary function, outer layer, with 1 neural neuron and liner Function, in the Matlab software with 6 inputs and 1 output. Comparing the achieved results with the Corresponding results by numerical methods which done by the new data showed that The difference between the results was very insignificant So the neural network Canestimate mean temperature for floor heating system of room carefully.