Dynamic constrained optimization of chemical manufacturing

Abstract

System and method for chemical manufacture utilizing a dynamic optimizer for a chemical process including upstream and downstream processes. The dynamic optimizer includes a maximum feed calculator, operable to receive one or more local constraints on the downstream processes and one or more model offsets, and execute steady state models for the downstream processes in accordance with the local constraints and the offsets to determine maximum feed capacities of the downstream processes; and a feed coordinator, operable to receive the maximum feed capacities, and execute steady state models for the upstream processes in accordance with the maximum feed capacities and a specified objective function, subject to global constraints, to determine upstream production parameters for the upstream processes, which are usable to control the upstream processes to provide feeds to the downstream processes in accordance with the determined maximum feeds and the objective function subject to the global constraints.

Description

FIELD OF THE INVENTION [0001] The present invention generally relates to the field of chemical production. More particularly, the present invention relates to systems and methods for optimizing chemical production in a manufacturing process with downstream and / or upstream constraints using predictive control methodologies. DESCRIPTION OF THE RELATED ART [0002] Like any other commercial enterprise, those in the business of producing chemical products desire to maximize efficiencies and profitability, while meeting various constraints, such as, for example, raw material and energy costs, plant equipment limitations, product prices, and so forth. The ability to produce chemicals in such a manner may be further complicated for chemical plants producing more than one grade or type of chemical product. [0003] As shown in prior art FIG. 1, a chemical plant 104 may produce chemicals, including, for example, olefin, gasoline, and fuel oil, among others, of varying grades, from feedstock, e.g...

AI Technical Summary

Benefits of technology

[0049] Thus, the maximum feed calculation preferably uses steady-state models from the cold side section ICOs. For example, a downstream optimizer may execute the steady state models under a variety of conditions or states to determine a solution (the maximum feeds), as is well known in the art of constrained optimization. Note that in various embodiments, respective ICOs preferably control respective cold side units (e.g., distillation columns), although in other embodiments, one or more ICOs may cover multiple units, or, an ICO may cover the entire cold section. Each ICO application is preferably configurable to improve operation of individual units subject to local process constraints (e.g., maximum feed rates, etc.).

[0069] Then, the plurality of upstream processes in the chemical plant may be controlled in accordance with the determined upstream production parameters, thereby facilitating production of the optimal product mix by the chemical plant in accordance with the specified objective. In other words, the upstream process, e.g., furnaces, may be operated in a manner that makes maximum use of the downstream processes in attempting to meet the specified objective subject to various constraints (e.g., local and / or global). Said another way, the determining maximum feed capacities, the determining upstream production parameters, and the controlling the plurality of upstream processes may implement overall chemical plant steady state optimization via one or more multivariable predictive dynamic controllers adjusting operational targets in the reactors, including target feed rates, to achieve optimum feed rates and product mix in the separation and purification processes. The hot-side ICOs may then operate in accordance with the provided targets, moving hot-side MVs as needed to meet the targets. In this manner, the furnaces may be operated in such as way as to maximize feeds to the cold side sections in accordance with the maximum feed calculations of the RAE.

Problems solved by technology

The ability to produce chemicals in such a manner may be further complicated for chemical plants producing more than one grade or type of chemical product.

However, the complexity of most real world systems generally precludes the possibility of arriving at such solutions analytically, i.e., in closed form.

Conventional computer fundamental models have significant limitations, such as:

(1) They may be difficult to create since the process may be described at the level of scientific understanding, which is usually very detailed;

(3) Some product properties may not be adequately described by the results of the computer fundamental models; and

(4) The number of skilled computer model builders is limited, and the cost associated with building such models is thus quite high.

These problems result in computer fundamental models being practical only in some cases where measurement is difficult or impossible to achieve.

Such models typically use known information about process to determine desired information that may not be easily or effectively measured.

This is very difficult to measure directly, and takes considerable time to perform.

However, there may be significant problems associated with computer statistical models, which include the following:

(1) Computer statistical models require a good design of the model relationships (i.e., the equations) or the predictions may be poor;

(2) Statistical methods used to adjust the constants typically may be difficult to use;

(3) Good adjustment of the constants may not always be achieved in such statistical models; and

(4) As is the case with fundamental models, the number of skilled statistical model builders is limited, and thus the cost of creating and maintaining such statistical models is high.

The resulting error is often used to adjust weights or coefficients in the model until the model generates the correct output (within some error margin) for each set of training data.

Setting such constraints may realistically restrict the allowable values for the manipulated variables.

In addition to the cost involved in developing the models, this approach suffers from the following deficiencies:

a) The optimizer models rarely produce the same results as the APC models, causing conflict between optimization goals and APC goals.

This means the models can only be executed when the unit is at steady state, which is a rare condition for an olefins unit.

c) Most traditional optimization models are very susceptible to instrumentation error.

d) Optimum conditions are not fully implemented due to the conflicts between the optimizer models and the APC models.

This is a more complex optimization problem requiring knowledge of column constraints and interactions with other process equipment.

Prior art approaches to process control and optimization have not adequately addressed these issues.

Typically, even stream compositions that are not fixed by product specifications have some practical limit.

The actual difference may deviate from zero as individual furnaces become limited but there may be some penalty for allowing the difference to become very large.

Images