Example: dental hygienist

Optimisation of an Ammonia Synthesis Loop - NTNU

Optimisation of an Ammonia Synthesis Loop Investigation of a Novel Approach for Optimisation of Integrated plants Martin Jonathan Bland Chemical Engineering and Biotechnology Submission date: June 2015. Supervisor: Sigurd Skogestad, IKP. Co-supervisor: Julian Straus, IKP. Norwegian University of Science and Technology Department of Chemical Engineering i Abstract A Hysys model representing the steady state behaviour of a Haber-Bosch Ammonia plant was developed based on a dynamic model and plant data from Norwegian fertiliser company Yara International ASA. The plant model was optimised using matlabs fmincon function with an interior-point met- hod for economical Optimisation with the aim to study the performance of different strategies. The Haber-Bosch Ammonia plant represents an interconnected complex sys- tem with many nested mass and energy streams that make for a challenging Optimisation .

Optimisation of an Ammonia Synthesis Loop. Investigation of a Novel Approach for Optimisation of Integrated plants. Martin Jonathan …

Tags:

  Synthesis, Ammonia, Optimisation, Optimisation of an ammonia synthesis

Information

Domain:

Source:

Link to this page:

Please notify us if you found a problem with this document:

Other abuse

Transcription of Optimisation of an Ammonia Synthesis Loop - NTNU

1 Optimisation of an Ammonia Synthesis Loop Investigation of a Novel Approach for Optimisation of Integrated plants Martin Jonathan Bland Chemical Engineering and Biotechnology Submission date: June 2015. Supervisor: Sigurd Skogestad, IKP. Co-supervisor: Julian Straus, IKP. Norwegian University of Science and Technology Department of Chemical Engineering i Abstract A Hysys model representing the steady state behaviour of a Haber-Bosch Ammonia plant was developed based on a dynamic model and plant data from Norwegian fertiliser company Yara International ASA. The plant model was optimised using matlabs fmincon function with an interior-point met- hod for economical Optimisation with the aim to study the performance of different strategies. The Haber-Bosch Ammonia plant represents an interconnected complex sys- tem with many nested mass and energy streams that make for a challenging Optimisation .

2 Therefore this thesis studies the effect of separating the plant model into units that are optimised separately to reduce the computational effort needed for the Optimisation . These units consists of the Synthesis gas makeup, reaction, separation and refrigeration sections. Two strategies where proposed, one were the plant refrigeration section was separated out from the plant model, which greatly reduced the amount of nested loops in the problem. The 2nd strategy replaced the separation and refrigeration section of the plant with a simplified model in the first opti- misation set. The individual sections of both strategies were independently optimised There were several difficulties with developing a model that allowed for a wide range of operating configurations of the Ammonia plant in Hysys.

3 In addition a few unphysical aspects of the Hysys model were uncovered, such as the compressor lacking an upstream flow effect and poor methods for evaluating pressure drop for varying conditions. These challenges meant that the Optimisation problem was only posed for a limited window of oper- ation for the process. Thus, for application to a real world plant the results should be scrutinised. The Optimisation strategies were successful in improving the plant prof- itability from the nominal operating point. In addition, as expected, the division decreased the computational effort needed for each Optimisation . Further work is required to develop the separation strategy and identify possible limitations. iii Sammendrag En Hysys modell som representer den stasjon re tilstanden for en am- moniakkfabrikk basert p Haber-Bosch prosessen ble utledet basert p an- leggsdata og en dynamisk modell fra den norske gj dselprodusenten Yara International ASA.

4 Hensikten med dette var uf re en konomisk opti- malisering av anleggsdriften ved bruk av matlabs funksjon fmincon med en "interior-point" optimaliseringsmetode for vurdere ulike optimaliser- ingsstrategier. Haber-Bosch-prosessen utgj r en tett koblet og kompleks prosess med flere resirkulasjonsstr mmer og n stede l kker for varmeintegrasjon. Denne sam- menkoblingen gir en utfordrende optimalisering og studien s p ulike op- timaliseringsstrategier for forenkle optimaliseringsproblemet. Konseptet var studere anlegget i ulike deler og optimalisere sett av enhetene hver for seg. Disse enhetene er klargj ring av syntesegassen, reaktor-, separasjon- og kj leseksjonene. Ideen med partisjonene anlegget er redusere bereg- ningsbehovet for finne en optimal l sning for driften av anlegget.

5 To separasjonsstrategier ble foresl tt, n der kj leseksjonen ble separert ut og optimalisert som en egen enhet. Dette reduserer antallet n stede l kker i Hysys-flytskjemaet betraktelig. Den andre strategien erstattet separasjon- og kj leseksjonen med en forenklet modell og optimaliserte syntesegass og reaktor seksjonen med den forenklede modellen, mens separasjon- og kj le- seksjonen ble optimalisert som et sett. Det viste seg v re vanskelig utvikle en stasjon r modell i Hysys som representerte prosessen p en fysikalsk korrekt m te for et bredt spekter av prosessbetingelser. Svakheter knyttet til modelleringen av den stasjon re tilstanden i Hysys ble ogs avdekket, som manglende oppstr ms effekt fra kompressorer og svake metoder for vurdere trykkfallet over prosessenheter, spesielt varmevekslere.

6 Dette resulterte i at prosessen kun ble optimalisert for et begrenset vindu av prosessbetingelser og at resultatene m vurderes n ye for applikasjon p et virkelig anlegg. Optimaliseringsstrategien var vellykket med tanke p ke anleggsinntek- tene for de gitte prosessbetingelsene og som forventet ble beregningsbehovet redusert for hver optimalisering. Mer arbeid er n dvendig for utvikle sep- arasjonsstrategien og for identifisere mulige begrensninger med denne. v Acknowledgements This thesis has been written as the conclusion of the degree in Chem- ical Engineering from the Norwegian University of Science and Technology. The work has been completed at the group for Process Systems Engineering. I would like to thank my main supervisor, Sigurd Skogestad, for the pos- sibility to write this thesis on an interesting and challenging subject at his research group, where the academic level and technical skill of the staff has only grown more impressive with the time spent here.

7 I am forever grateful for the knowledge that has been instilled in me through lectures and working with the professors, PhD-candidates and fellow students on the 2nd floor of Chemistry Block IV. This whole project would not have been possible without the guidance and motivation from PhD-candidate Julian Straus, my co-supervisor. I hope that this work will prove to be of benefit for your doctoral work and I. cannot express my gratitude enough for the effort you have spent with me. These five years of studies have been made into student life with the com- pany and support from my friends and family, I look forward to future adventures! Statement of Compliance I declare that this is an independent work according to the exam regulations of the Norwegian University of Science and Technology (NTNU).

8 Trondheim, 18th June 2015. Martin Jonathan Bland Contents Contents vi 1 Introduction 1. Ammonia .. 1. Modelling and Optimisation .. 2. Scope of work .. 2. Structure of the thesis .. 3. 2 Ammonia Production 5. History of Ammonia production .. 5. Process chemistry .. 6. Plant description .. 8. Syngas makeup .. 9. Reaction section .. 11. Separation section .. 13. Refrigeration section .. 15. 3 Process Modelling 19. Thermodynamic model and kinetics .. 19. Fluid package .. 19. Reaction kinetics .. 20. Unit operations modelling .. 21. Compressors .. 22. Heat exchangers .. 23. Separation tanks .. 24. Valves and relief orifices .. 24. vi Contents vii Reactor .. 25. Air coolers .. 25. Pumps .. 25. Recycle loops .. 26. Alternative modelling approach .. 28. 4 Optimisation 29. Problem definition.

9 29. The optimiser .. 31. Optimisation strategy .. 32. Subset 1 .. 33. Subset 2 .. 34. Optimisation problem .. 34. Setting up Hysys for Optimisation .. 37. 5 Results of Optimisation 41. Subsets 1(a) and 2(a) .. 41. Subsets 1(b) and 2(b) .. 45. Summary of Optimisation results .. 47. 6 Discussion 49. Aspects of the model and optimiser .. 49. Modelling issues .. 49. Alternative modelling approaches .. 50. Alternative Optimisation approach .. 51. Costing .. 51. Optimisation strategy .. 52. Future work .. 53. 7 Conclusion 55. Bibliography 57. A Upstream Syngas Processing 61. Disturbances originating in syngas production .. 62. B Nominal Process Conditions 63. Syngas makeup .. 63. Reaction section .. 64. Separation section .. 65. Refrigeration section .. 66. viii Contents C Compressor Description 69.

10 Syngas compressors .. 69. Refrigeration compressors .. 70. Compressor turbine .. 70. D MATLAB Code 75. Non-linear optimiser fmincon .. 75. Objective function .. 76. Contraints .. 77. HYSYS connectivity .. 77. E HYSYS Settings 81. Chapter 1. Introduction For this master thesis an Ammonia -plant has been investigated with regards to optimal operation. This study also looks into strategies for optimisa- tion of large scale interconnected plants, such as the Haber-Bosch process for Ammonia production. Optimality for a chemical plant in this context is to produce the maximum or desired amount of product for the lowest achievable cost whilst satisfying constraints like product specifications and environmental limits. The plant studied is operated by the Norwegian com- pany Yara, a major global fertiliser manufacturer and the worlds largest producer of Ammonia .


Related search queries