Considering the time variations, models can be classified

• Steady-State Models
• Dynamic Models

To solve a dynamic model, the chemical engineer has to specify the dependence of parameters with time the time interval for computation the time step to be considered. For each step in time, the algorithm applied to steady state solving, will be used to compute the outputs from inputs. At the end of the time interval, the outputs vector will be a function of inputs, and time: Y = F(X, t).

apiMAX™ is used for Steady State modeling. That is, is can be used for the representation of a process during a specific period/instance of time via the use of various equations/algorithms/process knowledge.

Bellow it is described the solving of a steady state model.

Solving a fundamental model

A fundamental model is a model specified as Y = F(X). For this kind of models, to solve them means to compute every variable of the Y vector considering one or more variables from X vector. This is done in one step, and the solving algorithm is complete.

Solving a composed model

To be able to solve a composed model, the chemical engineer must first specify the order in which the composing models must be considered. This is called an iteration sequence. For every sub-model instance in the model, in the order specified by the sequence, its outputs are determined from its inputs (it is solved). The algorithm starts over and over again, with the first sub-model instance in the sequence, until all input/state/output variables in the model reach a stationary state (the algorithm does not check on all variables involved in the models, only those who are of importance to convergence. This is necessary for efficiency reasons).

The stationary state is defined as the state when the variation of the variable is under a specified threshold.