The electricity load of iron and steel enterprises is greatly affected by the production conditions and physical process parameters. The existing research ignores the coupling effect of material flow and energy flow between processes

resulting in insufficient simulation accuracy. To solve the above problems

this paper presents a short-process power model of steel production driven by mechanism and data. Firstly

based on the coupling characteristics of material flow and energy flow in the transport process

the attenuation law of physical quantities such as mass and temperature value between the successive processes is derived. Secondly

a general expression of the power function is established according to the operation properties of the equipment

and a kernel extreme learning machine is used to fit the physical quantities and characteristic parameters of the power function. Finally

the power of different processes is superimposed in time domain

and the total power curve for the short-process iron and steel enterprise is obtained. The simulation results of a domestic steelmaking enterprise show that the proposed model can reflect power characteristics more accurately.