Pith. sign in

REVIEW 4 major objections 5 minor 16 references

Reliability-Based Fault Analysis and Modeling of Satellite Electrical Power Subsystems Using Fault Tree and Simulation Tools

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read The paper claims that a modeled satellite electrical power subsystem, built from battery, solar-array, and distribution-unit fault events, reaches 0.999 reliability and 0.998 mission reliability before launch.

desk verdict A textbook FTA application to a small satellite EPS, but the headline reliability numbers are a black box because the failure rates, fault tree, and mission time are never specified. read the letter →

arxiv 2507.15708 v1 pith:QLU26RA6 submitted 2025-07-21 eess.SY cs.SY

classification eess.SYcs.SY
keywords satelliteelectricalpowersystemfaulttreeanalysisreliabilityassessmentsimulationMATLABWindchillsoftwareriskmatrixLEOnano-satellite
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to predict, before launch, whether a satellite electrical power subsystem will survive its mission by combining fault simulation with fault tree analysis. It builds MATLAB models of the battery, solar array, and power distribution units in healthy and artificially faulted states, then feeds the likely faults as events into a fault tree and computes system-level probabilities. The central result is that the modeled power system has a reliability of about 0.999 and the overall mission a reliability of about 0.998, which the authors read as evidence that the simulated design is highly reliable. If true, this would show that pre-mission reliability assessment can be built from component fault data and a system diagram rather than from on-orbit failure statistics.

What carries the argument

Fault tree analysis is the load-bearing mechanism: a top-down Boolean model that connects basic component failures through AND and OR gates to the top event of electrical power subsystem failure. The tree is built from 11 events, including battery, solar array, and distribution-unit failures, and it converts per-component failure rates into system reliability, mission reliability, and a risk matrix. The MATLAB healthy and faulty simulations supply the set of fault modes and their current and voltage signatures; the failure-rate table supplies the quantitative inputs to the gates.

What would settle it

Re-run the reported fault tree with the upper bound of every Table 1 failure-rate range, such as battery at $300\times10^{-9}\,\mathrm{h^{-1}}$ and solar arrays at $200\times10^{-9}\,\mathrm{h^{-1}}$, and compare the resulting system reliability with 0.999. If the figure drops, the unpublished point-value choices, rather than the design, are carrying the result.

Watch

Extended reading notes

Core claim

The paper's central claim is that the designed satellite electric power system, with 11 fault-tree events spanning battery, solar array, and electrical distribution faults, has an overall reliability of 0.999 and supports a mission reliability of 0.998. The claim is obtained by treating every fault as a failure of the power subsystem, assuming at least one fault occurs and at least one power source remains, counting about 2048 combined fault scenarios ($N=2^M$ with $M=11$), and computing $P(S)=F/N$ for complete failure. Feeding the component failure rates from Table 1 into the fault tree yields the reported output, with a green-yellow-red risk matrix showing the design in the low-risk region.

Load-bearing premise

The 0.999 result hinges on the unstated point values chosen from the Table 1 failure-rate ranges, and on those automotive-derived rates applying to satellite components in orbit.

Editorial extensions

If this is right

  • A designer who trusts the component failure rates and the fault tree can use the computed 0.998 mission reliability as a pre-mission assurance figure for the modeled LEO nano-satellite.
  • The fault tree's event structure shows which of the three studied subsystems contributes most to the top event, so redesign effort can be placed where it lowers system failure probability.
  • Because the fault tree output includes a risk matrix, the same analysis can flag any future design change that pushes an event from the low-risk to the medium- or high-risk region.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The failure-rate ranges in Table 1 come from a 40-degree-Celsius drive-by-wire automotive study, not from space-qualified parts data, and the paper does not state which point in each range was entered into the reliability software; the 0.999 and 0.998 figures should therefore be read as conditional on that database, not as an orbital prediction.
  • The assumption that all faults lead to power subsystem failure makes the tree conservative in one direction but ignores partial degradation and fault tolerance; adding redundancy through AND gates could raise computed reliability and change the dominant failure paths.
  • A direct extension would run the healthy and faulty MATLAB models for the full two-year mission profile and count how often each fault event would actually trigger, comparing the empirical reliability with the fault-tree value.
  • Telemetry from operating LEO satellites on battery, solar-array, and distribution-unit fault counts could calibrate or replace the automotive-derived failure rates and show how much the reported reliability would move.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The manuscript presents a reliability analysis of a satellite electrical power subsystem (EPS) using MATLAB simulations of healthy and faulty battery, solar array, and power distribution components, followed by fault tree analysis (FTA) performed in Windchill software. The paper reports a power system reliability of 0.999 and a mission reliability of 0.998, with the abstract claiming a total mission assurance of 0.999. The analysis relies on component failure rates sourced from an automotive drive-by-wire reference, a fault tree with 11 events, and a risk matrix. The paper includes equations for battery and solar array sizing, descriptions of fault types, and definitions of fault tree symbols.

Significance. If the reported reliability figures were reproducible, this study would offer a useful demonstration of FTA applied to satellite EPS design, combining MATLAB-based fault simulations with commercial Windchill analysis. The paper has notable strengths: it provides explicit sizing equations (Eqs. 1–6), a clear taxonomy of EPS faults (open circuit, grounding, line-to-line, mismatch), and a structured fault tree formation process. However, the central quantitative contribution—the 0.999/0.998 reliability values—is not verifiable from the manuscript because the fault tree topology, basic-event probabilities, and mission time basis are not reported. The only quantitative input table (Table 1) gives broad failure-rate ranges from an automotive context at 40°C without justification for applicability to space hardware, and no point values are stated as having been entered into Windchill. As a result, the paper's main claim rests on unstated and possibly inappropriate input choices, limiting its contribution to the reliability engineering literature.

major comments (4)
  1. [Fault tree formation and drawing for satellite electrical power system, Table 1] Table 1 lists component failure rates only as ranges (e.g., transistors 1–70×10^-9 h^-1, Li-ion battery 200–300×10^-9 h^-1) and cites reference [20], a fault-tolerant drive-by-wire study conducted at 40°C. The authors never state which point value within each range was selected for the Windchill analysis, nor do they justify transferring automotive failure rates to a space environment characterized by vacuum, thermal cycling, and radiation. This is load-bearing: the computed system reliability is a direct function of these λ values, and different selections within the ranges could plausibly change the result by an order of magnitude or more.
  2. [Computational analysis, Eqs. (8)–(9)] The paper introduces P(S)=F/N with N=2^M and M=11 events (approximately 2048 scenarios), but never connects this combinatorial counting to an exponential reliability model such as R(t)=∏exp(−λ_i t) or to the Windchill output. The fault tree topology, minimal cut sets, and the mission time over which reliability is evaluated are not reported. Without these elements, the stated reliability values of 0.998 and 0.999 cannot be reproduced or independently audited, undermining the central claim.
  3. [Abstract and Conclusion] There is an unresolved inconsistency in the central result: the abstract reports 'a total mission assurance of 0.999,' while the conclusion states 'the overall reliability of the satellite mission (0.998) and the reliability of the satellite power system (0.999).' The manuscript should specify which number corresponds to which quantity and ensure the abstract matches the body. As written, the variability in the headline number reduces confidence in the reported analysis.
  4. [Computational analysis, 'For the drawn fault tree, 11 events...'] The statement that 11 events yield approximately 2048 combined faults is not derived from the presented fault tree diagram. The mapping from 2^11 scenarios to the actual fault tree basic events and logic gates is absent, and the relationship between P(S)=F/N and the reliability block diagram in Windchill is unclear. This disconnect makes it impossible to verify the probabilistic calculation or the final reliability figures.
minor comments (5)
  1. [Figure numbering and cross-references] The figure references are inconsistent: the text states 'Figure 5 shows the fault tree' and 'Figure 7 frequency-power diagram,' but the captions indicate Figure 6 is the fault tree graph and Figure 7 is the frequency diagram; captions for Figures 8–10 also appear misaligned with the in-text mentions. These mismatches should be corrected in a thorough revision.
  2. [Equation (7)] Equation (7) defines λ = (1/t_int)(N_f/N_t) but does not clarify the units or how this empirical relation connects to the standard exponential failure-rate interpretation used in reliability analysis; a brief derivation or citation would improve clarity.
  3. [Table 1, Diodes entry] The entry for diodes, '9-10×6-1', appears to be a typographical error; it likely should be '9-10×10^-6 h^-1' or a similar range. This needs correction.
  4. [Mission duration] The battery design discussion mentions a 2-year mission, but the reliability calculation does not explicitly state the mission time over which the 0.998/0.999 values apply; this time basis should be specified.
  5. [References] Several references are incomplete (e.g., [1], [7], and [20] lack full publication details), and reference [20] is cited for both Table 1 and Figure 10 without indicating which parts of the source correspond to which reuse.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: reliability output is a propagation of externally sourced failure-rate inputs, not a fit or self-referential definition.

full rationale

The derivation chain is: component failure-rate ranges from Ref. [20] (Table 1) are entered as basic-event data into a fault tree built in Windchill; the software then outputs system and mission reliability values. Nothing in the paper defines the output reliability in terms of the reported 0.999/0.998 numbers, nor fits any parameter to those numbers. The reported values are a standard fault-tree propagation of the assumed failure rates, so circularity would require showing that the inputs were chosen or defined from the outputs, which the paper does not do. The only self-citation, Ref. [4], supports a background statement about EPS components and is not load-bearing for the reliability computation. The main weakness is reproducibility (unspecified point values, mission time, and gate structure), which is a completeness/external-validity concern, not a circularity concern.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central reliability figure is entirely determined by assumed component failure rates and a fault tree structure, with no empirical calibration. These assumptions are the entire content of the estimate.

free parameters (2)
  • Component failure rates (lambda) for transistor, thyristor, IC, switch, amplifier, diode, battery, solar array = Not specified; ranges given in Table 1
    The final reliability is a function of these input rates. The paper does not state which point values are used, so the 0.999 result is not reproducible.
  • Number of events in the fault tree = 11 events (stated)
    The fault tree topology and number of events are modeling choices that determine the propagation of failure probabilities.
assumptions (4)
  • domain assumption All faults in the power subsystem lead to failure of the power subsystem.
    Stated in the Computational analysis section; this is a conservative simplification that shapes the fault tree.
  • domain assumption At least one fault occurs and at least one power source is available.
    Stated in the Computational analysis section; this defines the scenario space but is not justified physically.
  • ad hoc to paper The component failure rates from Table 1, sourced from a drive-by-wire reference at 40 degrees C, apply to satellite EPS components in space.
    The paper uses these rates without adjustment for space thermal, vacuum, or radiation conditions; this is a load-bearing but unjustified transfer.
  • standard math Component failures are statistically independent and the fault tree logic correctly represents the system.
    Standard FTA assumption, not stated explicitly but required for the product rule.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reliability-Based Fault Analysis and Modeling of Satellite Electrical Power Subsystems Using Fault Tree and Simulation Tools." pith.science (2026). https://pith.science/paper/QLU26RA6

@misc{pith2026250715708,
  author       = {Pith},
  title        = {Pith review of: Reliability-Based Fault Analysis and Modeling of Satellite Electrical Power Subsystems Using Fault Tree and Simulation Tools},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QLU26RA6}},
  note         = {Machine review of arXiv:2507.15708}
}
read the original abstract

One of the most important satellite subsystems is its electric power subsystem. The occurrence of a fault in the satellite power system causes the failure of all or part of the satellite. Calculating the overall reliability of the power system before the mission is crucial in improving the design of the satellite power system. Each component of the power system may malfunction due to pressure, launch pressure, and operating conditions. Accordingly, in this paper, first, a healthy and faulty system for the components of the electrical power system is simulated with MATLAB. Finally, by drawing a fault tree to analyze the reliability of the power subsystem, overall mission reliability, power system fault rate, and overall fault rate of the mission are calculated by Windchill software. Finally, a total mission assurance of 0.999 was achieved, indicating the high reliability of the simulated system.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

  1. [20]

    Fault-tolerant drive-by-wire systems

    R Isermann, R Schwarz, S Stolzl “Fault-tolerant drive-by-wire systems”- IEEE Control Systems, 2002

  2. [1]

    Zhenhua Wang” Directed factor graph based fault diagnosis model construction for mode switching satellite power system” Xiaolei Zhang

    Yi Shen. Zhenhua Wang” Directed factor graph based fault diagnosis model construction for mode switching satellite power system” Xiaolei Zhang. *School of Astronautics, Harbin Institute of Technology, Harbin Institute of Technology,

  3. [2]

    Health diagnosis and remaining useful life prognostics of lithium-ion batteries using data-driven methods

    Nuhic, Adnan, et al. "Health diagnosis and remaining useful life prognostics of lithium-ion batteries using data-driven methods." Journal of Power Sources 239 (2013): 680-688

  4. [3]

    TPK Nguyen, J Beugin, J Marais “method for evaluating an extended Fault Tree to analyse the dependability of complex systems: Application to a satellite-based railway system “- Reliability Engineering & System Safety, 2015

  5. [4]

    Smart fault detection in satellite electrical power system

    N. Nobahari and A. Rezaee, “Smart fault detection in satellite electrical power system,” arXiv preprint arXiv:2507.14004, 2025. [Online]. Available: https://arxiv.org/abs/2507.14004

  6. [5]

    Prognostic Modelling Options for Remaining Useful Life Estimation by Industry ,

    J. Z. Sikorska, M. Hodkiewicz and L. Ma, " Prognostic Modelling Options for Remaining Useful Life Estimation by Industry ," Mechanical Systems and Signal Processing, vol. 25, no. 5, pp. 1803-1836, 2011

  7. [6]

    Spacecraft power systems

    Patel, Mukund 2004; R. Spacecraft power systems. CRC Press,

  8. [7]

    Power System Reliability Evaluation Using Fault Tree Approach Based on Generalized Fuzzy Number Available online at www.ispacs.com/jfsva Volume 2012

Show all 16 references
  1. [8]

    K-Means+ ID3: A novel method for supervised anomaly detection by cascading K -Means clustering and ID3 decision tree learning methods

    Gaddam, Shekhar R., Vir V. Phoha, and Kiran S. Balagani2007; "K-Means+ ID3: A novel method for supervised anomaly detection by cascading K -Means clustering and ID3 decision tree learning methods." IEEE Transactions on Knowledge and Data Engineering 19.3: 345-354. [9] Fault Tr...

  2. [10]

    Spacecraft power systems

    Patel, Mukund R. Spacecraft power systems. CRC press, 2004

  3. [11]

    Decision tree-based fault detection and classification in solar photovoltaic arrays

    Zhao, Ye, et al. 2012; "Decision tree-based fault detection and classification in solar photovoltaic arrays." Twenty-Seventh Annual IEEE Applied Power Electronics Conference and Exposition (APEC). IEEE, 2012. [12]Survivability Analysis of the Satellite Electrical Power Subsyst...

  4. [13]

    Fault detection and reconstruction for micro -satellite power subsystem based on PCA

    Lee, Bo, and Xinsheng Wang 2010; “Fault detection and reconstruction for micro -satellite power subsystem based on PCA." Systems and Control in Aeronautics and Astronautics (ISSCAA), 2010 3rd International Symposium on. IEEE, 2010

  5. [14]

    Genetic symbiosis algorithm generating test data for constraint automata A Rezaee Applied and Computational Mathematics 6 (1), 126-137 [15]using Bayesian Network for robot behavior with sensor fault Alireza Rezaee, Abolghasem A. Raie, Abolfazlnadi, saeed shiry ghidary [16]Elec...

  6. [17]

    Analysis of Faulted Power Systems

    Paul M. Anderson, “Analysis of Faulted Power Systems”, The Institute of Electrical and Electronics Engineers, Inc., 1995

  7. [18]

    Analysis Of Transmission System Faults the Phase Domain

    Jun Zhu. “Analysis Of Transmission System Faults the Phase Domain”, Texas A&M University. Master Thesis, 2004

  8. [19]

    Power System Analysis,

    C. Gross, “Power System Analysis,” second edition, Wiley, 1986

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.