Hydrocarbon Processing - November 2022 - 15

Digital Technologies
several minutes to a few seconds) without loss of significance.
One of the functions assessed during the research was an
economic evaluation of the debutanizer performance. This
evaluation was made from " customer " encrypted production
data and from clear parameters on product prices that can be
set from market prices.4,5
The encrypted inputs are:
* Feed mass flow: F1
* Fuel gas mass flow: F2
* Liquefied petroleum gas (LPG) mass flow: F3
* Stabilized naphtha mass flow: F4
* Hot oil mass flow: F5
* Hot oil inlet temperature: T1
* Hot oil outlet temperature: T2.
The clear market prices inputs are:
* Unstabilized naphtha $/t (ton): K1
* Fuel gas $/t: K2
* LPG $/t: K3
* Stabilized naphtha $/t: K4
* Hot oil properties: K5.
The output O1 (in $) is expressed as (Eq. 1):
O1 = [F2
× K2 + F3
K2 × (T1
× K3 + F4
× K4 - F1
× K1 - F5
× K5
- T2 )] × (Sample Interval / 3,600)
×
(1)
The O1 computation time with the BFV encryption
scheme from the Microsoft SEAL library was 0.033 sec when
encrypting large plain texts (~32 bits long). Note: Two preanalyses
allowed the authors' companies to fix the size of their
plain texts and choose the suited HE algorithm for O1 computation
(i.e., to choose the BFV encryption scheme from
the Microsoft SEAL library). First, O1 was computed over
clear real numbers to get the expected output values and their
ranges. Then, it was computed over clear integers obtained by
rescaling the clear real inputs. Computing O1 with rescaled
integers introduced a loss of precision. The rescaled integers
size was fixed to ensure obtaining a result with two significant
digits. The companies ended up working with clear integers
longer than 32 bits. Then, as O1 computation consists only
in addition and multiplication of integers, the partnership
chose to use an HE scheme supporting modular arithmetic
with large inputs, such as BFV in the Microsoft SEAL library
or BGV in IBM Helib. That is, the companies did not choose
to use libraries working with binary plain texts by encrypting
separately the bits of inputs (such as the TFHE library), as O1
computation in that case would take around 1 min.
However, with the Microsoft SEAL implementation of the
BFV scheme used to reach this performance, it was not possible
to perform comparison on real numbers that also were part of
the evaluation. Therefore, the final demonstrator integrated
two different implementations relying on two different HE
schemes, one with a TFHE scheme and a binary plain text space
and the other with the BFV scheme and covering large numbers
to minimize computation time. The functions evaluated
with TFHE enclosed an encrypted comparison. The functions
computed with BFV required only additions and multiplications.
The outcome of the work performed on the demonstration
test case was that the 20 tested functions were evaluated
every 10 sec without lag with a standard 4-core server.
3
4
2
Takeaways. The research work performed on the applicability
of HE schemes in the energy industry proved successful
thanks to the cooperation between the authors' companies
and their collective knowledge of industry and research. At
the start of the project, the forecast of computation time was
unsatisfactory and seemed to be a showstopper in the current
state of the research. Creative solutions were found to overcome
this initial assessment and ended in a workable demonstrator
hosted in a standard virtual machine.
Online digital services in the energy industry can be made
easier from a contractual standpoint with the use of HE. It secures
actors in their respective confidential production data
and intellectual properties for this kind of project.
HE extends a plant optimization program performed on
selected historical data to a near real-time application. Algorithm
crafted from the historical data in traditional ways can
be converted into an online homomorphically encrypted application
that continues providing operations information
and advice.
LITERATURE CITED
1
Rivest, R. L., L. Adleman and M. L. Dertouzos, " On data banks and privacy
homomorphisms, " Foundations of secure computation, Massachusetts Institute of
Technology, Cambridge, Massachusetts, 1978.
Gentry, C., et al., " Fully homomorphic encryption using ideal lattices, " Stanford
University and IBM Watson, STOC '09: 41st Annual ACM symposium on the
theory of computing, May 2009.
Gentry, C., " A decade (or so) of fully homomorphic encryption, " Eurocrypt 2021,
online: https://eurocrypt.iacr.org/2021/slides/gentry.pdf
Cheon, J. H., A. Kim, M. Kim and Y. Song, " Homomorphic encryption for
arithmetic of approximate numbers, " Cryptology ePrint Archive, 2016, Report
2016/421, online: https://eprint.iacr.org/2016/421
5
Boura, C., N. Gama, M. Georgieva and D. Jetchev, " Chimera: Combining ringLWE-based
fully homomorphic encryption schemes, " Cryptology ePrint Archive,
2018, Report 2018/758, online: https://eprint.iacr.org/2018/758
Following his initial training as a process and chemical
engineer at the Ecole Nationale Supérieure des Industries
Chimiques in Nancy, France, in 2000, FABRICE REY
specialized in process dynamic simulation applied to the
energy industry. He worked for 14 yr at a company dedicated
to process dynamic simulation, where he developed his skills
working as a model developer, a lead engineer, a project
manager and as a technical sales engineer. Rey has more than 20 yr of
experience gathered on five continents, and he joined Technip Energies as
Lead of process dynamic simulation activities to develop the company's
internal capabilities as part of the Expertise and Modeling department.
MATHIEU SANCHEZ has more than 15 yr of experience in
advanced systems design and integration, deploying
added-value solutions for energy industries including FLNG,
refining and offshore platforms. He graduated in 2006 in
process and chemical engineering from the Ecole Nationale
Supérieure en Génie des Technologies Industrielles located
in Pau, France, and started his professional journey at an
automation company specializing in energy industries optimization. In 2012,
Sanchez joined the Advanced Systems Engineering department of Technip
Energies and serves as Lead Engineer of definition, development,
implementation and commissioning of digital infrastructure and solutions
used to enhance asset performance.
AYMEN BOUDGUIGA is a Senior Researcher at CEA, and
works on fully homomorphic encryption applications to
neural networks. Previously, his investigations and work
have included topics related to autonomous vehicles security
or mobile device authentication. Dr. Boudguiga earned
a PhD in 2012 in computer science from University Pierre
and Marie Curie in Paris, France.
Hydrocarbon Processing | NOVEMBER 2022 15
https://eurocrypt.iacr.org/2021/slides/gentry.pdf https://eprint.iacr.org/2016/421 https://eprint.iacr.org/2018/758
Hydrocarbon Processing - November 2022

Table of Contents for the Digital Edition of Hydrocarbon Processing - November 2022
