You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A key element of plasma control systems (PCS) in tokamak reactors is the prediction and avoidance of disruptions, sudden losses of the thermal and magnetic energy stored within the plasma that can occur when tokamaks operate near regions of plasma instability or because of system malfunctions.
9
-
The energy released during disruptions can cause severe damage to plasma-facing components, limiting experimental operation or even the device lifetime.
10
-
This poses a serious challenge to next-step fusion experiments such as SPARC, which will have to operate near some of the limits of plasma stability to achieve its intended performance and will do so at for long and frequent intervals.
11
-
Previous work has shown the promise of machine-learning (ML) algorithms for disruption prediction in both DIII-D and EAST -- the Experimental Advanced Superconducting Tokamak in China -- PCS.
12
-
This is also due to the fact that fusion science currently lacks first-principle, theoretical solutions to fully predict and avoid disruptions.
DisruptionPy is an open-source Scientific Python package for fast retrieval of experimental Fusion data from [MDSplus](https://www.mdsplus.org/) servers.
29
+
The library allows an efficient database preparation for downstream analysis and/or ML model development for disruption studies.
30
+
At present, the main supported machines are [Alcator C-Mod](https://en.wikipedia.org/wiki/Alcator_C-Mod) and [DIII-D](https://en.wikipedia.org/wiki/DIII-D_(tokamak)).
13
31
14
-
DisruptionPy is an open-source python package for training, updating, and evaluating algorithms for disruption prediction and avoidance that can be applied to Alcator C-Mod and DIII-D data, and can deploy models in DIII-D and EAST (TBD) PCSs.
15
32
16
33
## Overview
17
34
18
-
DisruptionPy makes it easy to retrieve tabular data from MDSplus databases efficiently.
19
-
Users can create their own methods and/or use built-in methods that retrieve and derive a variety of important parameters from experimental data for disruption analysis.
20
-
These methods are run across all provided sets of discharges (or shot ids), outputting tabular data in customizable formats.
35
+
### Background
36
+
37
+
A key element to ensure steady state operations in magnetically confined tokamak devices is the prediction and avoidance of disruptions.
38
+
These are sudden losses of the thermal and magnetic energy stored within the plasma, which can occur when tokamaks operate near stability boundaries or because of hardware anomalies.
39
+
The energy stored in the plasma and released during disruptions over milliseconds can cause severe damage to plasma-facing components, limiting experimental operations and the device's lifespan [[1](https://doi.org/10.1080/15361055.2023.2229675)].
40
+
Disruptions still pose a serious challenge to next-generation fusion devices such as ITER or SPARC, which will have to operate near some of the limits of plasma stability to achieve intended performance and will do so at for long and frequent intervals.
41
+
Fusion science currently lacks first-principle, theoretical solutions to fully predict and avoid disruptions.
42
+
However, previous work [[2](https://doi.org/10.1088/1741-4326/ab28bf), [3](https://doi.org/10.1088/1741-4326/abf74d)] has shown the usefulness of machine-learning (ML) algorithms for disruption prevention for both DIII-D and EAST -- the Experimental Advanced Superconducting Tokamak in China -- operations.
43
+
DisruptionPy provides a standardized analysis pipeline across different fusion devices to build ML-ready datasets.
44
+
45
+
### Workflow
46
+
47
+
DisruptionPy makes it easy to retrieve experimental data from [MDSplus](https://www.mdsplus.org/) fusion repositories efficiently.
48
+
Users can create their own routines and/or use built-in ones that retrieve and derive a variety of important signals from experimental data for disruption analysis.
49
+
These routines are then interpolated on a requested timebase across the specified set of plasma discharges (or shots) to assemble a dataset and save it under a variety of available formats.
50
+
51
+
<imgsrc="docs/workflow.png"alt="Schematic flowchart of a typical DisruptionPy workflow. By Y Wei (2024)"width="400"onerror="this.onerror=null;this.src='workflow.png';" />
52
+
53
+
_Figure: Schematic flowchart of a typical DisruptionPy workflow. By Y Wei (2024) [6]._
54
+
55
+
### Acknowledgments
56
+
57
+
The most recent revamp of DisruptionPy [4, 5, 6] was partially supported by DOE FES under Award DE-SC0024368, "Open and FAIR Fusion for Machine Learning Applications" [7].
58
+
59
+
### References
60
+
61
+
1. AD Maris, A Wang, C Rea, RS Granetz, E Marmar (2023), _"The Impact of Disruptions on the Economics of a Tokamak Power Plant"_, **Fusion Science and Technology** 80(5) 636-652, [DOI:10.1080/15361055.2023.2229675](https://doi.org/10.1080/15361055.2023.2229675).
62
+
63
+
2. C Rea, KJ Montes, KG Erickson, RS Granetz & RA Tinguely (2019), _"A real-time machine learning-based disruption predictor in DIII-D"_, **Nuclear Fusion** 59 096016, [DOI:10.1088/1741-4326/ab28bf](https://doi.org/10.1088/1741-4326/ab28bf).
64
+
65
+
3. WH Hu, C Rea, et al. (2021), _"Real-time prediction of high-density EAST disruptions using random forest"_, **Nuclear Fusion** 61 066034, [DOI:10.1088/1741-4326/abf74d](https://doi.org/10.1088/1741-4326/abf74d).
66
+
67
+
4. C Rea, et al. (2024), _"Open and FAIR Fusion for Machine Learning Applications"_, 66th APS Division of Plasma Physics Meeting, [PP12.27](https://meetings.aps.org/Meeting/DPP24/Session/PP12.27).
68
+
69
+
5. GL Trevisan, et al. (2024), _"Functional Improvements and Technical Developments of a Community-driven and Physics-informed Numerical Library for Disruption Studies"_, 66th APS Division of Plasma Physics Meeting, [PP12.9](https://meetings.aps.org/Meeting/DPP24/Session/PP12.9).
70
+
71
+
6. Y Wei, et al. (2024), _"Physics validation of parameter methods in DisruptionPy"_, 66th APS Division of Plasma Physics Meeting, [PP12.10](https://meetings.aps.org/Meeting/DPP24/Session/PP12.10).
72
+
73
+
7. C Rea, et al. (2023), _"Open and FAIR Fusion for Machine Learning Applications"_, [Project website](https://crea-psfc.github.io/open-fair-fusion/).
74
+
75
+
76
+
## Repository layout
77
+
78
+
Notable branches:
79
+
80
+
-`main`, the [stable branch](https://github.com/MIT-PSFC/disruption-py/tree/main),
81
+
-`dev`, the [development branch](https://github.com/MIT-PSFC/disruption-py/tree/dev),
82
+
-`matlab`, the [historical branch](https://github.com/MIT-PSFC/disruption-py/tree/matlab).
83
+
21
84
22
85
## Project layout
23
86
24
-
```python
25
-
disruption_py/# source code
26
-
docs/# documentation
27
-
examples/# example workflows
28
-
scripts/# miscellaneous scripts
29
-
tests/# automated testing
30
-
```
87
+
Brief description of the folders in our project:
88
+
89
+
-`disruption_py/`, package source code,
90
+
-`docs/`, documentation sources,
91
+
-`drafts/`, experimental scripts,
92
+
-`examples/`, example workflows,
93
+
-`scripts/`, miscellaneous scripts,
94
+
-`tests/`, testing workflows.
31
95
32
-
The original Matlab scripts are now stored in the `matlab`[protected branch](https://github.com/MIT-PSFC/disruption-py/tree/matlab).
33
96
34
97
## Installation
35
98
36
99
DisruptionPy is now open-source and [available at PyPI](https://pypi.org/project/disruption-py/)!
100
+
37
101
For standard installations, please follow the usual way:
38
102
39
103
```bash
@@ -46,14 +110,17 @@ pip install disruption-py
46
110
47
111
For custom installations, please refer to our [Installation guide](docs/INSTALL.md).
48
112
113
+
49
114
## Getting Started
50
115
51
-
Please see the project [quickstart](https://mit-psfc.github.io/disruption-py/quickstart/usage_quickstart/).
116
+
Please see the [project quickstart](https://mit-psfc.github.io/disruption-py/quickstart/usage_quickstart/).
52
117
53
-
## Issues
54
118
55
-
If you have an issue please crate an issue on the GitHub repository
119
+
## Contributing
56
120
57
-
## Development
121
+
> [!IMPORTANT]
122
+
> Make sure you refer to the latest version of our [development branch](https://github.com/MIT-PSFC/disruption-py/tree/dev)!
58
123
59
-
Please create a pull request if you have something to contribute!
124
+
- If you encounter any problems, please [create a new issue](https://github.com/MIT-PSFC/disruption-py/issues/new).
125
+
- If you would like to contribute, please [submit a pull request](https://github.com/MIT-PSFC/disruption-py/compare/dev...).
126
+
- If you have general questions, please [start a new discussion](https://github.com/MIT-PSFC/disruption-py/discussions/new?category=q-a).
0 commit comments