Aug-2018

Control optimisation for petrochemicals (TIA)

Neste Engineering Solutions’ (NES) Dynamic Real Time Optimization (DRTO) technology continues to demonstrate long term cumulated benefit versus traditional real time optimisation (RTO) for plant-wide application subject to the complex process behaviour of an ethylene cracker and the challenging chemical dynamics of a phenol and aromatics (P&A) production facility.

Stefan Bosman
NAPCON

Article Summary

Borealis Polymers ethylene and P&A plants in Porvoo, Finland implemented a plant-wide application of NES’s NAPCON technology, combining DRTO and advanced process control (APC) in a project that resulted in a $12.5 million/y increase in profits for the ethylene plant as well as a 9% increase in the P&A plant’s production rate.

The underlying concept of DRTO, using linearised models for fast execution (ordinarily in the order of 30 seconds up to two minutes) of control optimisation produces a faster response and therefore enables guiding the process towards optimum performance with a high frequency. A traditional closed loop optimiser, such as RTO, on the other hand, was previously tested in the Borealis ethylene plant and, to the detriment of the method’s reliance on steady state process models and its inability to cope with a move to the more dynamic plant operation, was unable to perform with adequate performance within a reasonable response time.

Borealis chose NAPCON as the provider, offering a solution that was calculated to give the best profitability, lowest investment cost, as well as the fastest pay-back time. Following a vendor selection process, the RTO project of the ethylene cracker was initiated as part of the Borealis programme for enhancing plant operations with an emphasis on the economic result. During both the ethylene and P&A plants’ implementation, the importance of staff participation was heavily emphasised.

NAPCON initiated the P&A DRTO project with a study phase, during which the current operating practices were examined, along with the DCS controllers’ performance, a very often neglected means for basic level optimisations. This was followed by the assembly of a team consisting of operations management, process and automation engineers, panel operators, shift foremen, maintenance department representatives and vendor DRTO specialists who examined all relevant process constraints, key performance indicators and manipulated variables. Project execution and commissioning was split into several phases using the ability of DRTO to support a phased implementation and allowing parts of the optimisation application to be commissioned while others were still in the design phase. Final acceptance test runs confirmed the capability of the application to increase the production rates and substantially improve the overall controllability and plant stability.

In the course of the ethylene project’s implementation, two phases accruing maximum benefit as early as possible were incorporated into the schedule while, for each commissioning phase, a commissioning plan was formulated that reached conclusion as the end user took over responsibility following a joint review period. Throughout the final acceptance tests as a result, the offline optimisation benchmark outcomes were verified and found to agree under differing price scenarios, proving the efficacy and scalability of NAPCON DRTO technology.

In both cases, the NAPCON Controller DRTO implementation, as opposed to traditional RTO, was shown in situ to handle the frequent transients of this process while maintaining it in an economical operating region. In the case of the ethylene plant, the DRTO objective function maximises the gross margin by navigating nearly 200 DRTO constraints within their operating window with the help of the underlying APCs while DRTO directly controls 12 variables to their targets for profit optimisation. In the P&A application, from the first process phase, to produce benzene, through cumene production, and finally the more complicated phases (cumene hydroperoxide cleavage to phenol and acetone, excluded from the APC scope), in addition to phenol and acetone distillation (recycle loops, and an optimised heat exchanger network), an optimisation is no trivial task. Therefore, three separate strategies on three different units proved optimal (see Figure 1). The benzene unit was optimised for maximum throughput, exposing several bottlenecks in the extractive column. The cumene unit optimisation aimed to stabilise DCS level control, which proved particularly challenging, while the phenol unit’s optimisation targeted cumene hydroperoxide production and concentration while in parallel focusing on the phenol distillation train and hydrocarbon recovery section.

In summary, the move to an APC and DRTO implementation for plant-wide optimisation is a proven, fast executing, scalable and efficient solution on top of the existing DCS controls for boosting process plant production. NAPCON Optimizer can be applied to a variety of processes.

This short case study originally appeared in PTQ's Technolology in Action feature - Q3 2018.

For more information: venla.kuuluvainen@neste.com