Page 14 - Process Equipment and Plant Design Principles and Practices by Subhabrata Ray Gargi Das

8 Chapter 1 General aspects of process design

• Requirement of auxiliary facilities like utilities and their levels
• Waste disposal/environmental considerations
• Ease of operation, maintenance, erection, commissioning
• Availability of resources for technology, materials, manpower and skill for erection,
commissioning, operation, and maintenance
• Project completion time
• Designs with proven implementation and performance are preferred
• Financing of the project: Capital availabilitydamount and its layout in a time scale

In practice, the first step of the designer is the collection of information and previous documents on
similar design projects. Implementation success, performance and limitations of these are scrutinized.
The design alternatives emerging from this step are often the improved versions of the previous de-
signs and the experience of the designer is an important factor.

Quantitative considerations
Quantitative selection of the best alternative is arrived at through a process of optimization. The
objective function to be optimized can be an economic parameter like payout period, internal rate of
return, the total annualized cost for the plant, etc. Such economic parameters are usually used in case
of large equipment, process systems/plants or projects.
This requires a mathematical model describing the process in terms of the process design variables
namely (1) operating conditions (temperature, pressure, flow, etc.) and (2) equipment specification
parameters (capacity, number of separation stages, etc.). The mathematical model describing the
design will be relationships among the design variables expressed as equations and inequations (“>”or
“<” expressions). Such expressions can be linear as well as nonlinear. The objective function is
expressed in terms of these variables. Set of numerical values of the variables define the process design
output completely.
A multivariable constrained optimization needs to be solved mathematically for optimizing the
objective function without violating the constraints of the model (these expressions are in the form of
“equalities” and “in-equalities”). Design software commonly uses various linear and nonlinear pro-
gramming solvers for optimization. Some of the algorithms used for nonlinear optimization are Nelder
& Mead, MarquardteLevenberg, Sequential Quadratic Programming. These methods may be used
independently or with other convergence techniques like Wegstein method, NewtoneRaphson, etc.,
which are common for flowsheet simulation convergence.
In the case of large plants/complexes, there can be cases of multiple solutions for the optimization
problem, out of which only one can be implemented. The alternative solutions have the same objective
function value, but the combination of the value of the process parameters (variables) differ. Selection
of the final choice requires the introduction of new criteria that is often qualitative like the ease of
maintenance/operation/etc.

Optimum design
Generation of optimum design of specific equipment requires fixing up a suitable technical criterion
like efficiency (of a furnace, etc.) or specific energy consumption rate that may be optimized where
these technical parameters used as objective function are directly linked to the economics. This leads