Compare commits
5 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| fcd720101a | |||
|
|
cbc2467219 | ||
|
|
77f361734b | ||
|
|
ea81aee8ca | ||
| 1820c38cae |
10
.gitignore
vendored
10
.gitignore
vendored
@ -38,3 +38,13 @@ Thumbs.db
|
||||
.idea/
|
||||
*.sty
|
||||
.DS_Store
|
||||
*.aux
|
||||
*.bbl
|
||||
*.blg
|
||||
*.fdb_latexmk
|
||||
*.fls
|
||||
*.log
|
||||
*.pdf
|
||||
*.toc
|
||||
*.out
|
||||
*.synctex.gz
|
||||
|
||||
@ -586,115 +586,6 @@ mode approach is sound for nuclear failures. Shows safety can be proven even
|
||||
when controller deviates from nominal (pp.85-107, UCA 1
|
||||
analysis).}
|
||||
|
||||
\subsection{Preliminary Results}
|
||||
|
||||
In order to demonstrate the feasibility of this research proposal, we applied
|
||||
this approach to HAHACS creation to a simple nuclear reactor model, with limited
|
||||
procedural guidance. We created a simple point kinetics reactor model with
|
||||
a two-node thermal hydraulics model. The reactor assumes plant parameters and
|
||||
material properties consistent with a Uranium-235 pressurized water reactor. For
|
||||
the reactor core, six precursor groups are included, and fuel temperature and
|
||||
moderator temperature feedback are implemented. For the thermal hydraulics,
|
||||
basic convection assumptions are made about the heat generated from the fuel's
|
||||
reactivity, while the coolant is assumed to be single phase water. The
|
||||
differential equations representing each part of the system are described as
|
||||
follows.
|
||||
|
||||
The point kinetics equations with delayed neutron precursors govern the
|
||||
neutron population $n(t)$ and precursor concentrations $C_i(t)$:
|
||||
%
|
||||
\begin{equation}
|
||||
\frac{dn}{dt} = \frac{\rho(t) - \beta}{\Lambda} n(t)
|
||||
+ \sum_{i=1}^{6} \lambda_i C_i(t)
|
||||
\label{eq:pke}
|
||||
\end{equation}
|
||||
%
|
||||
\begin{equation}
|
||||
\frac{dC_i}{dt} = \frac{\beta_i}{\Lambda} n(t) - \lambda_i C_i(t),
|
||||
\quad i = 1, \ldots, 6
|
||||
\label{eq:precursors}
|
||||
\end{equation}
|
||||
%
|
||||
where $\rho$ is the total reactivity, $\beta = \sum \beta_i$ is the total
|
||||
delayed neutron fraction, $\Lambda$ is the prompt neutron generation time,
|
||||
$\lambda_i$ are the precursor decay constants, and $\beta_i$ are the
|
||||
individual group delayed neutron fractions.
|
||||
|
||||
The two-node thermal hydraulics model couples fuel temperature $T_f$ and
|
||||
coolant temperature $T_c$:
|
||||
%
|
||||
\begin{equation}
|
||||
m_f c_{p,f} \frac{dT_f}{dt} = P(t) - h_{gap} A_s (T_f - T_c)
|
||||
\label{eq:fuel_temp}
|
||||
\end{equation}
|
||||
%
|
||||
\begin{equation}
|
||||
m_c c_{p,c} \frac{dT_c}{dt} = h_{gap} A_s (T_f - T_c)
|
||||
- \dot{m} c_{p,c} (T_c - T_{inlet})
|
||||
\label{eq:coolant_temp}
|
||||
\end{equation}
|
||||
%
|
||||
where $P(t)$ is the fission power proportional to $n(t)$, $h_{gap}$ is the
|
||||
gap heat transfer coefficient, $A_s$ is the heat transfer surface area,
|
||||
$\dot{m}$ is the coolant mass flow rate, and $T_{inlet}$ is the coolant
|
||||
inlet temperature.
|
||||
|
||||
Reactivity is a sum of several different sources. First, reactivity is
|
||||
a factor of the fuel itself. Second, temperature feedback effects from both
|
||||
fuel and moderator will change the total reactivity. Finally, control rods
|
||||
directly add or subtract reactivity in the system by their movement:
|
||||
%
|
||||
\begin{equation}
|
||||
\rho(t) = \rho_{ext}(t) + \alpha_f (T_f - T_{f,0})
|
||||
+ \alpha_m (T_c - T_{c,0})
|
||||
\label{eq:reactivity}
|
||||
\end{equation}
|
||||
%
|
||||
where $\rho_{ext}$ is the externally applied reactivity from control rod
|
||||
position, $\alpha_f$ and $\alpha_m$ are the fuel and moderator temperature
|
||||
coefficients of reactivity, and $T_{f,0}$ and $T_{c,0}$ are reference
|
||||
temperatures. These equations were implemented in a MATLAB code to simulate
|
||||
the simple reactor model.
|
||||
|
||||
First, a discrete controller was created by writing FRETish specifications
|
||||
according to the diagram shown in Figure~\ref{fig:hybrid_automaton}. A full
|
||||
list of the specifications created is included in
|
||||
Appendix~\ref{APPENDIX}.\dasnote{need to actually create the appendix}. This
|
||||
was then synthesized into an automaton using reactive synthesis
|
||||
tools.\dasnote{This will depend once reactive synthesis is done---JKind?
|
||||
Kind2? Strix?} Once this automaton was created, continuous controllers were
|
||||
developed in MATLAB for each discrete control mode. The end result was a
|
||||
hybrid system with discrete modes and continuous controllers created based
|
||||
on logical specifications, but were not yet verified.
|
||||
|
||||
Once the continuous controllers were created, reachability analysis was
|
||||
performed on the transitory modes. For this example, there is only one
|
||||
transitory mode: the heatup mode. For this mode, the entry temperature
|
||||
$T_{MIN}$ was included as an entry condition, while other states were given
|
||||
physically reasonable bounds. Reachability analysis confirmed that the
|
||||
continuous controller satisfied the transitory mode requirement from
|
||||
Section~3.2.1:\dasnote{Need to reference the specific reachability
|
||||
condition equation once earlier sections are finalized}
|
||||
%
|
||||
\[
|
||||
\text{Reach}(\mathcal{X}_{entry}, f_{heatup}, [0,T]) \subseteq
|
||||
\mathcal{X}_{safe} \land \text{Reach}(\mathcal{X}_{entry}, f_{heatup},
|
||||
[0,T]) \cap \mathcal{X}_{exit} \neq \emptyset
|
||||
\]
|
||||
|
||||
Once the transitory mode was complete, the stabilizing mode of power operation
|
||||
was analyzed using a barrier certificate search.\dasnote{More needed here:
|
||||
what are the bounds of the invariant set? What tool was used for the SOS
|
||||
search? What degree polynomial for the barrier function?}
|
||||
|
||||
Finally, the expulsory SCRAM mode was considered. For this example, the failure
|
||||
modes were kept relatively simple. We assumed possible failure modes of coolant
|
||||
fouling, where the coolant thermal properties have changed drastically, and a
|
||||
secondary system coolant leak, where power demand increases dramatically. For
|
||||
both of these cases, reachability analysis was performed to evaluate that the
|
||||
dynamics of the system could safely reach a low power, low temperature state for
|
||||
the possible operating range of the reactor.
|
||||
|
||||
\subsection{Industrial Implementation}
|
||||
|
||||
The methodology described above must be validated on realistic systems using
|
||||
@ -721,4 +612,3 @@ system experts at Emerson ensures that implementation details of the Ovation
|
||||
platform are handled correctly. Direct industry collaboration also provides an
|
||||
immediate pathway for technology transfer and alignment with practical
|
||||
deployment requirements.
|
||||
|
||||
|
||||
6
main.tex
6
main.tex
@ -11,7 +11,7 @@
|
||||
% === SPLIT'S EDITING COMMENTS ===
|
||||
% Set to 1 for edit mode (wider margins, visible comments)
|
||||
% Set to 0 for final mode (normal margins, comments hidden)
|
||||
\newcommand{\editmode}{1}
|
||||
\newcommand{\editmode}{0}
|
||||
|
||||
\ifnum\editmode=1
|
||||
% Edit mode: load todonotes, adjust margins
|
||||
@ -59,9 +59,9 @@
|
||||
|
||||
\begin{document}
|
||||
|
||||
\pagenumbering{roman}
|
||||
% \pagenumbering{roman}
|
||||
\maketitle
|
||||
\input{1-goals-and-outcomes/research-statement.tex}
|
||||
% \input{1-goals-and-outcomes/research-statement.tex}
|
||||
\newpage
|
||||
\tableofcontents
|
||||
\newpage
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user