Repository logo
 

Blockchain, parameterisation and automated arbitrage applied to the chemical industry


Type

Thesis

Change log

Authors

Sikorski, Janusz Jerzy 

Abstract

This thesis considers three scenarios related to chemical industry where the concepts of eco-industrial parks (EIPs), Industry 4.0, parameterisation, blockchain and arbitrage are brought together to explore the issues of simulation speed and accuracy, machine-to-machine (M2M) communication and automated participation in financial markets. In the first scenario, a biodiesel plant flow sheet model is analysed and parameterised. The relations between 11 inputs typical to a biodiesel plant and its energy requirements are approximated using surrogate models, of which accuracy is assessed. Additionally, the effects of dimensionality, domain size and surrogate type on the accuracy are investigated and global sensitivities of the outputs are computed using High Dimensional Model Representation (HDMR). Most surrogate models achieved at least a reasonable fit regardless of the domain size and number of dimensions. It was observed that in all cases only 4 or fewer inputs have significant influence on any of the outputs and that the interaction terms have only minor effect on any one output. In the second scenario, applications of blockchain technology related to Industry 4.0 are explored and an example where blockchain is employed to facilitate M2M interactions and establish a M2M electricity market in the context of the chemical industry is presented. Successful implementation of two electricity producers and one electricity consumer trading with each other over a blockchain-based network is presented. In the third scenario, an automated arbitrage spotter is developed and applied to two cases: conversion of natural gas to methanol and crude palm oil to biodiesel. The spotter is designed to search for opportunities to make additional profit by analysing the futures market prices for both the reagent and the product. It considers cost of storage and conversion (other feedstock, steam, electricity and other utilities) derived from physical simulations of the chemical process. In a profitable scenario up to 345.17 USD per tonne of biodiesel can be earned by buying contracts for delivery of crude palm oil in September 2018 and selling contracts for delivery of biodiesel in December 2018 in a ratio of 4 to 1.

Description

Date

2019-10-10

Advisors

Kraft, Markus

Keywords

blockchain, parameterisation, arbitrage, chemical industry

Qualification

Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge
Sponsorship
EPSRC (1501675)
Department of Chemical Engineering and Biotechnology in the University of Cambridge and Cambridge CARES Ltd.
Relationships
Is supplemented by: