Solutions
   Blending optimization
   Additives input optimization
   Ex-line blending optimization
   Crude distillation optimization
   Crude oil management & crude blending
   Steam cracker optimization
   Process units optimization
   Bio fuel
   Upgrading your NIR System

Steam cracker optimization

Anticipate and react on time

Quality variations and switches of the steam cracker liquid feeds are common, frequent and mainly depend on spot imports.  
Conventional analyzers cannot detect such variations on time, leading to important production losses.
The fast reactivity of NIR combined with the high accuracy and robustness of Topnir model insures dramatic improvement of the process efficiency.

Technology summary

The PIONA character (Parafin, Iso Parafin, Naphthenic, Olefin and Aromatic) is very well correlated to some specific wavelengths in the NIR spectrum range.

Topnir predicts the steam cracker liquid feeds key properties in real time.  
For Naphtha feed, the model is able to output the Carbon by Carbon PIONA (Detailed Hydricarbon Analysis, DHA) as well as the density, the molecular weight, the distillation Curve D86, the coking Index , the potential yields of Ethylene and propylene…

In addition to monitoring the feed properties, the system can be configured to also predict the Pyrolisys Gasoline.

 

Increase benefits

Maximize Advanced Process Control efficiency

  • Increase throughput

  • Improve olefins yields

  • Push the process toward measured quality constraints

  • Maintain the process on constraint

  • Minimize Process fluctuation

Maximize Feed switch efficiency

  • Minimize transition time

  • Minimize transition associated risks

 

Increase furnaces run length

  • Use Topnir coke knowledge into the optimizer

Save energy

Improve safety

  Savings: 2- 5 M$/Year

Return on investment capital less than 5 months  

KEY ADVANTAGES

 

Accurate and robust model

The predictions accuracy ensures a tight control on key properties. In addition, the Topnir robustness allows accounting for large feed quality range of variation.

Easy model maintenance

Topnir research and development has accounted for the weakness of traditional Partial Least Square models maintenance, that rapidly becomes a bottleneck when the application involves a lot of properties, each requiring a specific calibration treatment.
Topnir is based on robust and non regressional approach does not require specific model update. Thanks to the self-learning methodology, Topnir model is easily maintained up-to-date only with regular additions of new samples into the database.

Coke index calculation

By measuring the feed coke index, Topnir goes beyond a simple analyzer and brings more added value in the optimization chain.

 

 

Easy integration with existing APC and optimization

Topnir outputs the quality information to the DCS to feed the existing advanced process control and optimization systems.
Moreover, the decoking scheduling of radiation tubes and transfer line exchanger can be done through the coke index delivered by Topnir.
These indices, specific to given feedstocks, are used to predict the coke rate and skin temperature.

User friendly and powerful graphical interface

Topnir provides the operator with a efficient graphical representation to monitor, simulate and anticipate process variation.
Specific spectral aggregates have been defined to handle the spectral discrimination and spectral pattern recognition.

   

 

 

 

REAL CASE STUDY

Ethylene production: 1 MT/year ; Naphtha feed from spot market.

Incremental Benefits on Steam Cracker (real case study)

ITEM

YEARLY SAVINGS ($)
Increase Ethylene production

3 M
By increasing throughput  
By Improving olefin yields  
By optimizing feed and campaign switch efficiency  
By increasing furnace run length  
Save energy

Not estimated

YEARLY GLOBAL SAVING

 3 M $

Back to top Click here to know more