According to the system engineering theory

simultaneous optimization of technological processes and heat exchanger networks can bring great economic benefits

and it indicates the direction of development in this area. In this paper

a new method called "Superstructure Regression Model" is put forward

it can be used to obtain the optimum flowsheet structure and optimum operation parameters simultaneously. The Superstructure Regression Model is a Mixed Integer Nonlinear Programming (MINLP). In order to overcome the difficulties brought by the discrete variables in the process of optimization

the author proposes a "Integer Variables Continuation Technique" to solve the MINLP problems. At the same time a simplified regression model for ammonia synthesis reactor is brought out in this paper. This method has been used successfully in a fertilizer plants revamping and expansion project.