diff --git a/joss.05099/10.21105.joss.05099.crossref.xml b/joss.05099/10.21105.joss.05099.crossref.xml
new file mode 100644
index 0000000000..3664e1e762
--- /dev/null
+++ b/joss.05099/10.21105.joss.05099.crossref.xml
@@ -0,0 +1,260 @@
+
+
+
+ 20230706T155724-945d6fac2343c42fc28b6f52ae11cba266ddac85
+ 20230706155724
+
+ JOSS Admin
+ admin@theoj.org
+
+ The Open Journal
+
+
+
+
+ Journal of Open Source Software
+ JOSS
+ 2475-9066
+
+ 10.21105/joss
+ https://joss.theoj.org/
+
+
+
+
+ 07
+ 2023
+
+
+ 8
+
+ 87
+
+
+
+ ProgPy: Python Packages for Prognostics and Health
+Management of Engineering Systems
+
+
+
+ Christopher
+ Teubert
+ https://orcid.org/0000-0001-6788-4507
+
+
+ Katelyn
+ Jarvis
+
+
+ Matteo
+ Corbetta
+ https://orcid.org/0000-0002-7169-1051
+
+
+ Chetan
+ Kulkarni
+
+
+ Matthew
+ Daigle
+
+
+
+ 07
+ 06
+ 2023
+
+
+ 5099
+
+
+ 10.21105/joss.05099
+
+
+ http://creativecommons.org/licenses/by/4.0/
+ http://creativecommons.org/licenses/by/4.0/
+ http://creativecommons.org/licenses/by/4.0/
+
+
+
+ Software archive
+ 10.5281/zenodo.8097013
+
+
+ GitHub review issue
+ https://github.com/openjournals/joss-reviews/issues/5099
+
+
+
+ 10.21105/joss.05099
+ https://joss.theoj.org/papers/10.21105/joss.05099
+
+
+ https://joss.theoj.org/papers/10.21105/joss.05099.pdf
+
+
+
+
+
+ ProgPy packages
+ Teubert
+ 2022
+ Teubert, C., Jarvis, K., Corbetta,
+M., Kulkarni, C., & Daigle, M. (2022). ProgPy packages (Version
+1.4). https://nasa.github.io/progpy
+
+
+ Prognostics: The science of making
+predictions
+ Goebel
+ 2017
+ Goebel, K., Daigle, M. J., Saxena,
+A., Roychoudhury, I., Sankararaman, S., & Celaya, J. (2017).
+Prognostics: The science of making predictions. Createspace Independent
+Pub.
+
+
+ A generic software architecture for
+prognostics (GSAP)
+ Teubert
+ International Journal of Prognostics and
+Health Management
+ 2
+ 8
+ 10.36001/ijphm.2017.v8i2.2618
+ 2017
+ Teubert, C., Daigle, M. J.,
+Sankararaman, S., Goebel, K., & Watkins, J. (2017). A generic
+software architecture for prognostics (GSAP). International Journal of
+Prognostics and Health Management, 8(2).
+https://doi.org/10.36001/ijphm.2017.v8i2.2618
+
+
+ Prognostics models matlab
+library
+ Daigle
+ 2016
+ Daigle, M. (2016). Prognostics models
+matlab library.
+https://github.com/nasa/PrognosticsModelLibrary
+
+
+ Prognostics algorithm matlab
+library
+ Daigle
+ 2016
+ Daigle, M. (2016). Prognostics
+algorithm matlab library.
+https://github.com/nasa/PrognosticsAlgorithmLibrary
+
+
+ Prognostics metrics matlab
+library
+ Daigle
+ 2016
+ Daigle, M. (2016). Prognostics
+metrics matlab library.
+https://github.com/nasa/PrognosticsMetricsLibrary
+
+
+ Prognostics as-a-service: A scalable cloud
+architecture for prognostics: A scalable cloud architecture for
+prognostics
+ Watkins
+ Annual conference of the PHM
+society
+ 11
+ 2019
+ Watkins, J., Teubert, C., &
+Ossenfort, J. (2019). Prognostics as-a-service: A scalable cloud
+architecture for prognostics: A scalable cloud architecture for
+prognostics. Annual Conference of the PHM Society,
+11.
+
+
+ Enabling in-time prognostics with surrogate
+modeling through physics-enhanced dynamic mode decomposition
+method
+ Jarvis
+ Annual conference of the PHM
+society
+ 14
+ 10.36001/phmconf.2022.v14i1.3238
+ 2022
+ Jarvis, K., Corbetta, M., Teubert,
+C., & Schuet, S. (2022). Enabling in-time prognostics with surrogate
+modeling through physics-enhanced dynamic mode decomposition method.
+Annual Conference of the PHM Society, 14.
+https://doi.org/10.36001/phmconf.2022.v14i1.3238
+
+
+ Randomized battery usage data
+set
+ Bole
+ NASA Ames Research Center, Moffett Field,
+CA
+ 2014
+ Bole, B., Kulkarni, C., & Daigle,
+M. (2014). Randomized battery usage data set. In NASA Ames Research
+Center, Moffett Field, CA. NASA Prognostics Data Repository.
+https://www.nasa.gov/content/prognostics-center-of-excellence-data-set-repository
+
+
+ Turbofan engine degradation
+simulation
+ Saxena
+ NASA Ames Research Center, Moffett Field,
+CA
+ 2008
+ Saxena, A., & Goebel, K. (2008).
+Turbofan engine degradation simulation. In NASA Ames Research Center,
+Moffett Field, CA. NASA Prognostics Data Repository.
+https://www.nasa.gov/content/prognostics-center-of-excellence-data-set-repository
+
+
+ Simantha
+ Dadfarnia
+ NIST
+ 2013
+ Dadfarnia, M., & Drozdov, S.
+(2013). Simantha. In NIST. National Institute of Standards; Technology
+[Software].
+https://github.com/usnistgov/simantha
+
+
+ Lifelines, survival analysis in
+python
+ Davidson-Pilon
+ 10.5281/zenodo.7329096
+ 2022
+ Davidson-Pilon, C. (2022). Lifelines,
+survival analysis in python (Version v0.27.4). Zenodo.
+https://doi.org/10.5281/zenodo.7329096
+
+
+ Pomegranate: Fast and flexible probabilistic
+modeling in python
+ Schreiber
+ The Journal of Machine Learning
+Research
+ 1
+ 18
+ 2017
+ Schreiber, J. (2017). Pomegranate:
+Fast and flexible probabilistic modeling in python. The Journal of
+Machine Learning Research, 18(1), 5992–5997.
+
+
+
+
+
+
diff --git a/joss.05099/10.21105.joss.05099.jats b/joss.05099/10.21105.joss.05099.jats
new file mode 100644
index 0000000000..653542ccd6
--- /dev/null
+++ b/joss.05099/10.21105.joss.05099.jats
@@ -0,0 +1,522 @@
+
+
+
+
+
+
+
+Journal of Open Source Software
+JOSS
+
+2475-9066
+
+Open Journals
+
+
+
+5099
+10.21105/joss.05099
+
+ProgPy: Python Packages for Prognostics and Health
+Management of Engineering Systems
+
+
+
+https://orcid.org/0000-0001-6788-4507
+
+Teubert
+Christopher
+
+
+
+
+
+Jarvis
+Katelyn
+
+
+
+
+https://orcid.org/0000-0002-7169-1051
+
+Corbetta
+Matteo
+
+
+
+
+
+Kulkarni
+Chetan
+
+
+
+
+
+Daigle
+Matthew
+
+
+
+
+
+NASA Ames Research Center, United States
+
+
+
+
+KBR, Inc.
+
+
+
+
+12
+12
+2022
+
+8
+87
+5099
+
+Authors of papers retain copyright and release the
+work under a Creative Commons Attribution 4.0 International License (CC
+BY 4.0)
+2022
+The article authors
+
+Authors of papers retain copyright and release the work under
+a Creative Commons Attribution 4.0 International License (CC BY
+4.0)
+
+
+
+Python
+Prognostics
+Health Management
+Degradation Simulation
+Diagnostics
+State Estimation
+Prediction
+CBM+
+IVHM
+PHM
+
+
+
+
+
+ Summary
+
Prognostics of engineering systems or systems of systems is the
+ prediction of future performance and/or the time at which one or more
+ events of interest occur. Prognostics can be applied in a variety of
+ applications, from spacecraft and aircraft to wind turbines, oil and
+ gas infrastructure, and assembly lines. Prognostic results are used to
+ inform action to extend life or prevent failures through changes in
+ use or predictive maintenance.
+
The
+ NASA
+ Prognostics Python Packages (ProgPy)
+ (Teubert
+ et al., 2022) are a set of open-sourced Python packages
+ supporting research and development of prognostics and health
+ management for engineering systems, as described in
+ (Goebel
+ et al., 2017). ProgPy builds upon the architecture of the
+ Matlab Prognostics Libraries
+ (Daigle,
+ 2016c,
+ 2016a,
+ 2016b),
+ Generic Software Architecture for Prognostics
+ (Teubert
+ et al., 2017), and Prognostics As-A-Service
+ (Watkins
+ et al., 2019). ProgPy implements architectures and common
+ functionalities of prognostics, supporting both researchers and
+ practitioners.
+
+
+ Statement of need
+
Prognostics and Health Management (PHM) is a fast-growing field.
+ Successful PHM application can reduce operational costs and prevent
+ failure, making systems safer. There has been limited application of
+ prognostics in practice. This is partially because prognostics is a
+ multi-faceted and complex problem, including data availability, sensor
+ design and placement, and, of interest to us, software.
+
Often, software is written for an ad-hoc single prognostic
+ application and cannot be transferred to others, or is limited in
+ scope. A few related packages are described here. Simantha is a
+ discrete manufacturing system simulation package that simulates
+ degradation, but it is limited to a Discrete-Time Markov Chain and
+ doesn’t include prognostic capabilities
+ (Dadfarnia
+ & Drozdov, 2013). Lifelines is a survival analysis tool
+ that can be used for reliability analysis to establish fixed-interval
+ maintenance schedules, a different problem than that solved by ProgPy
+ (Davidson-Pilon,
+ 2022). Pomegranate is a Python probabilistic modeling package
+ effective for data science applications
+ (Schreiber,
+ 2017). However, Pomegranate does not include explicit state
+ estimation capabilities, prognostics tools, or physics-based
+ degradation modeling features. Finally, there are a number of general
+ machine-learning packages such as TensorFlow, scikit learn, and
+ PyTorch. These are general tools that can be used for diagnostics and
+ prognostics, but are not designed specifically for that
+ application.
+
There is a need for a foundational set of efficient tools to enable
+ new PHM technologies.
+
ProgPy provides a set of support packages for individuals
+ researching and developing prognostic technologies. ProgPy consists of
+ three packages: prog_models,
+ prog_algs, and
+ prog_server. prog_models
+ provides tools aiding the development, evaluation, simulation, and
+ tuning of prognostic models, whether physics-based or data-driven.
+ prog_models also supports downloading select
+ relevant datasets
+ (Bole
+ et al., 2014;
+ Saxena
+ & Goebel, 2008). prog_algs supports
+ uncertainty representation, state estimation, prognostics, the
+ evaluation and visualization of prognostic results, and the creation
+ of new prognostic algorithms. prog_server is a
+ Service-Oriented Architecture for prognostics and state estimation.
+ prog_server is also distributed with a Python client,
+ prog_client.
+
The following sections describe some ways ProgPy could be used.
+ There are many features and use cases beyond those illustrated here.
+ See the
+ ProgPy
+ documentation for more information.
+
+
+ Selected use case: building and simulating models
+
One of the primary use-cases of ProgPy is building new models.
+ Prognostic models are created by subclassing the
+ PrognosticsModel class. Users can copy the
+ model template as a starting point, replacing the representative
+ member functions with model logic.
+
Prognostic models have inputs (load/control applied to a system),
+ internal states, outputs (measurable quantities), and events of
+ interest (what we’re predicting). Logic of a prognostic model is
+ defined using the state transition (dx or
+ next_state), output,
+ event_state, and
+ threshold_met functions.
+
In the below example, a user creates a physics-based model of a
+ Lithium-ion battery. In this model (see
+ here),
+ state transition equations (i.e., internal states) relate the voltage
+ discharge from the battery (i.e., the output) given an applied current
+ (i.e., the input).
+ class Battery(PrognosticsModel):
+ inputs = [
+ ‘i’ # current applied to battery
+ ]
+ states = [
+ # internal battery model states, e.g., temperature, surface potentials
+ # nasa.github.io/progpy/api_ref/prog_models/IncludedModels.html
+ ‘x_1’, # State 1
+ ‘x_2’, # State 2
+ …
+ ]
+ outputs = [
+ ‘t’, # Battery temperature
+ ‘v’ # Voltage supplied by battery
+ ]
+ events = [
+ 'EOD' # battery end-of-discharge
+ ]
+
+ # Default parameters. Overwritten by passing parameters into constructor
+ default_parameters = {
+ 'x0':{ # Initial State
+ },
+ 'param1':p_1,
+ ….
+ # Include parameters to define battery model
+ # nasa.github.io/progpy/api_ref/prog_models/IncludedModels.html
+ }
+
+ def dx(self, x, u):
+ # calculate derivative of the battery state
+ return self.StateContainer({}) # Return state container with derivative
+
+ def output(self, x):
+ # From the state, calculate temperature and voltage
+ return self.OutputContainer({'t': x['t'], 'v': x['v']})
+
+ def event_state(self, x):
+ # From current state, calculate progress towards EOD
+ return {
+ 'EOD': v_now – v_threshold
+ # EOD occurs when voltage drops below threshold
+ }
+
The resulting model can then be used in simulation:
+ m = Battery()
+def future_load(t, x=None): # system loading
+ return m.InputContainer({‘i’:1}) # Constant 1 amp applied
+
+simulated_results = m.simulate_to_threshold(future_load, dt=0.005)
+
+print(f'EOD was reached in {round(simulated_results.times[-1],2)}seconds')
+
ProgPy also includes data-driven models such as the LSTM State
+ Transition and Dynamic Mode Decomposition models. These are trained
+ using data and then used for simulation or prognostics, as above.
+
+
+ Selected use case: prognostics of battery discharge
+ cycle
+
Models can be used for prognostics with
+ prog_algs. Prognostics is often split into two
+ steps: state estimation and prediction. In state estimation, the
+ system state is estimated, with uncertainty, using the prior state
+ estimate and sensor data. In prediction, the state estimate is
+ predicted forward.
+
This example illustrates predicting the battery discharge. Here
+ data is retrieved from some unspecified source
+ (data_source). This can be a data stream,
+ playback file, or any other source. This is similar to the
+ sim_battery_eol example (see
+ here).
+ batt = Battery()
+x0 = batt.initialize()
+# Create Particle Filter State Estimator
+state_estimator = state_estimators.ParticleFilter(batt, x0)
+# Create Monte Carlo Predictor
+predictor = predictors.MonteCarlo(batt)
+
+# Future loading as function of time (t) and state (x)
+# In this case- constant load
+def future_loading(t, x=None):
+ return batt.InputContainer({'i':2.35})
+
+while RUNNING:
+ u, z = data_source.get_data()
+ # Estimate state using loading (u) and output measurements (z)
+ state_estimator.estimate(t, u, z)
+ eod = batt.event_state(filt.x.mean)['EOD']
+ print(f" - State of charge (mean): {eod}")
+ # Only predict every PREDICTION_UPDATE_FREQ steps
+ if (step%PREDICTION_UPDATE_FREQ==0):
+ mc_results = mc.predict(filt.x, future_loading, t0 = t, dt=TIME_STEP)
+ metrics = mc_results.time_of_event.metrics()
+ eod_mean = metrics['EOD']['mean']
+ eod_std = metrics['EOD']['std']))
+ print(f' - Predicted end of discharge: {eod_mean} (sigma: {eod_std})')
+
+
+ NASA use cases
+
ProgPy has been used in various NASA projects. Two are described
+ below.
+
+ Data and Reasoning Fabric
+
ProgPy functionality predicting battery degradation was
+ implemented to assess the Li-ion batteries state of charge during
+ unmanned aerial vehicle (UAV) flight. Based on planned trajectories,
+ ProgPy provided UAV operators with statistics on expected battery
+ health during flight and helped to ensure safety in the national
+ airspace
+ (Jarvis
+ et al., 2022).
+
+
+ Autonomous Spacecraft Operations
+
ProgPy was used to create models predicting the ISS life support
+ system degradation informing maintenance. Researchers evaluated the
+ performance of multiple potential models with data from the system
+ and ProgPy metrics and visualization. Researchers updated models
+ based on performance results. The selected model will be integrated
+ with ProgPy state estimation and prediction into a prognostic
+ application for crew or ground support.
+
+
+
+ Acknowledgements
+
ProgPy is supported by NASA’s Autonomous Spacecraft Operations,
+ Data and Reasoning Fabric, System-Wide Safety, and Transformative
+ Tools and Technologies projects. Additionally, development is
+ supported by Northrop Grumman Corporation, Vanderbilt University, the
+ German Aerospace Center (DLR), Research Institutes of Sweden and
+ others.
+
+
+
+
+
+
+
+ TeubertChristopher
+ JarvisKatelyn
+ CorbettaMatteo
+ KulkarniChetan
+ DaigleMatthew
+
+ ProgPy packages
+ 2022
+ https://nasa.github.io/progpy
+
+
+
+
+
+ GoebelKai
+ DaigleMatthew John
+ SaxenaAbhinav
+ RoychoudhuryIndranil
+ SankararamanShankar
+ CelayaJose
+
+ Prognostics: The science of making predictions
+ Createspace Independent Pub
+ 2017
+
+
+
+
+
+ TeubertChristopher
+ DaigleMatthew J
+ SankararamanShankar
+ GoebelKai
+ WatkinsJason
+
+ A generic software architecture for prognostics (GSAP)
+ International Journal of Prognostics and Health Management
+ 2017
+ 8
+ 2
+ 10.36001/ijphm.2017.v8i2.2618
+
+
+
+
+
+ DaigleMatthew
+
+ Prognostics models matlab library
+ 2016
+ https://github.com/nasa/PrognosticsModelLibrary
+
+
+
+
+
+ DaigleMatthew
+
+ Prognostics algorithm matlab library
+ 2016
+ https://github.com/nasa/PrognosticsAlgorithmLibrary
+
+
+
+
+
+ DaigleMatthew
+
+ Prognostics metrics matlab library
+ 2016
+ https://github.com/nasa/PrognosticsMetricsLibrary
+
+
+
+
+
+ WatkinsJason
+ TeubertChristopher
+ OssenfortJohn
+
+ Prognostics as-a-service: A scalable cloud architecture for prognostics: A scalable cloud architecture for prognostics
+ Annual conference of the PHM society
+ 2019
+ 11
+
+
+
+
+
+ JarvisKatelyn
+ CorbettaMatteo
+ TeubertChristopher
+ SchuetStefan
+
+ Enabling in-time prognostics with surrogate modeling through physics-enhanced dynamic mode decomposition method
+ Annual conference of the PHM society
+ 2022
+ 14
+ 10.36001/phmconf.2022.v14i1.3238
+
+
+
+
+
+ BoleB.
+ KulkarniC.
+ DaigleM.
+
+ Randomized battery usage data set
+ NASA Ames Research Center, Moffett Field, CA
+ NASA Prognostics Data Repository
+ 2014
+ https://www.nasa.gov/content/prognostics-center-of-excellence-data-set-repository
+
+
+
+
+
+ SaxenaA.
+ GoebelK.
+
+ Turbofan engine degradation simulation
+ NASA Ames Research Center, Moffett Field, CA
+ NASA Prognostics Data Repository
+ 2008
+ https://www.nasa.gov/content/prognostics-center-of-excellence-data-set-repository
+
+
+
+
+
+ DadfarniaMehdi
+ DrozdovSerghei
+
+ Simantha
+ NIST
+ National Institute of Standards; Technology [Software]
+ 2013
+ https://github.com/usnistgov/simantha
+
+
+
+
+
+ Davidson-PilonCameron
+
+ Lifelines, survival analysis in python
+ Zenodo
+ 202211
+ https://doi.org/10.5281/zenodo.7329096
+ 10.5281/zenodo.7329096
+
+
+
+
+
+ SchreiberJacob
+
+ Pomegranate: Fast and flexible probabilistic modeling in python
+ The Journal of Machine Learning Research
+ JMLR. org
+ 2017
+ 18
+ 1
+ 5992
+ 5997
+
+
+
+
+
diff --git a/joss.05099/10.21105.joss.05099.pdf b/joss.05099/10.21105.joss.05099.pdf
new file mode 100644
index 0000000000..3871cf5082
Binary files /dev/null and b/joss.05099/10.21105.joss.05099.pdf differ