pith. sign in

arxiv: 2606.31210 · v1 · pith:DYUGFPBUnew · submitted 2026-06-30 · 📡 eess.SY · cs.SY

Due-to-Heatwaves Faults in Urban Distribution System: An Identification Approach

Andrea Mazza , Haoke Wu This is my paper

Pith reviewed 2026-07-01 04:44 UTC · model grok-4.3

classification 📡 eess.SY cs.SY
keywords heatwave faultsdistribution system reliabilityfault attributioncovariance analysisurban power networksexcess heat factortime-between-failuresnon-exponential failure times
0
0 comments X

The pith

A covariance method distinguishes heatwave-driven faults from other summer faults in urban distribution networks.

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

The paper develops a framework to separate faults whose timing aligns with heatwave thermal stress from those driven by unrelated causes such as ageing or mechanical issues. Heatwaves are first identified with the Excess Heat Factor index. A covariance criterion then attributes faults only when their occurrence pattern matches the expected thermal mechanism, and a time-delay model estimates the lag between heatwave onset and fault appearance. When applied to six years of real medium-voltage network data, the method shows that heatwave-attributable faults form a notable but fluctuating share of total summer faults. This distinction matters because treating every heatwave-period fault as heat-related inflates estimates of system vulnerability and distorts asset-management and reliability planning.

Core claim

The authors claim that not all faults during heatwaves result from thermal stress; a covariance-based attribution criterion, paired with Excess Heat Factor characterization of heatwaves and a normalized-covariance time-delay estimator, isolates the subset directly consistent with heatwave mechanisms. Application to operational data reveals that these Due-to-HW faults represent a significant yet variable fraction of summer faults, while also showing seasonal deterioration in fault rate and mean time between failures plus clear departure of inter-failure times from exponential distributions.

What carries the argument

The covariance-based attribution criterion that flags a fault as Due-to-HW only when its occurrence statistics align with HW-driven thermal mechanisms rather than independent factors.

If this is right

  • Due-to-HW faults form a significant but variable share of total summer faults.
  • Fault rates and mean time between failures worsen across all seasons under recurrent heatwave exposure.
  • Inter-failure time distributions deviate from exponential, limiting the accuracy of Poisson-based reliability models.
  • Asset-management decisions that lump all heatwave-period faults together will overestimate thermal vulnerability.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Utilities could use the same covariance filter on future seasons to update real-time vulnerability maps without assuming every hot day produces thermal faults.
  • If the time-delay model reveals consistent lags, grid operators might schedule inspections or load reductions at those specific offsets before faults occur.
  • The departure from exponential failure times suggests that reliability planning should incorporate clustered or non-memoryless models when heatwaves recur.
  • Extending the framework to other weather stressors such as wind or ice would require only redefining the reference index while keeping the covariance test intact.

Load-bearing premise

Statistical alignment between fault timing and heatwave temperature series via covariance is sufficient to establish that the heatwave itself caused the fault.

What would settle it

Applying the attribution criterion to a set of faults during heatwaves that are independently documented as non-thermal (for example, animal contact or excavation damage) and finding that a large share are still classified as Due-to-HW would falsify the method.

read the original abstract

Distribution system faults occurring during heatwaves (HWs) are not all caused by the HW itself: concurrent factors such as asset ageing, mechanical defects, soil contamination, and operational constraints contribute independently. Hence, indiscriminately attributing all HW-period faults to thermal stress overestimates system vulnerability and misleads asset-management decisions. This paper proposes a systematic framework to identify and quantify the subset of summer faults directly attributable to HW occurrences (denoted Due-to-HW faults), by distinguishing them from Due-to-Others faults. HW events are first characterised through the Excess Heat Factor index. A covariance-based attribution criterion is then developed to distinguish faults whose occurrence is statistically consistent with HW-driven thermal mechanisms from those attributable to independent causes. Complementing the attribution framework, a time-delay model is introduced to estimate the lag between the beginning of a HW and fault occurrence by maximising the normalised covariance between hourly temperature series and shifted fault-duration series. Applied to six years of operational data from a real MV distribution network, the simulation results show that Due-to-HW faults constitute a significant yet variable proportion of total summer faults, underscoring the non-negligible impact of HW occurrences on summer fault statistics. Beyond documenting the deterioration of fault rate and Mean Time Between Failures across all seasons, the analysis confirms that Time-Between-Failures distributions depart significantly from the exponential assumption, with direct implications for the applicability of Poisson-based reliability models to distribution systems subject to recurrent HW stress.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper proposes a framework to identify the subset of distribution system faults during heatwaves that are directly attributable to thermal stress (Due-to-HW faults) rather than independent factors. Heatwaves are characterized via the Excess Heat Factor; a covariance-based attribution criterion distinguishes faults statistically aligned with HW-driven mechanisms from others; a time-delay model estimates the lag between HW onset and fault occurrence by maximizing normalized covariance between hourly temperature and shifted fault-duration series. Applied to six years of real MV network operational data, the results indicate that Due-to-HW faults form a significant yet variable share of summer faults, with accompanying deterioration in fault rates and MTBF across seasons and significant departures from exponential time-between-failures distributions (implying limited applicability of Poisson reliability models).

Significance. If the attribution method proves robust to confounders, the work would offer a data-driven way to quantify HW-specific impacts on urban distribution reliability and would strengthen the case against routine use of memoryless Poisson models under recurrent thermal stress. The use of actual multi-year operational records from a real network is a positive feature.

major comments (3)
  1. [methods (covariance attribution)] The covariance-based attribution criterion (described in the methods) treats statistical alignment between temperature and fault timing as evidence of direct thermal causation, yet the manuscript provides no explicit partialling-out of confounders such as coincident load or demand spikes that also covary with temperature. This assumption is load-bearing for the central claim that the attributed proportion reflects HW thermal impact rather than seasonal correlation.
  2. [time-delay model] The time-delay model is defined by maximizing the normalized covariance between the temperature series and the shifted fault-duration series; because the lag itself is chosen by the same fitting step used for attribution, it is unclear whether the resulting lag or the Due-to-HW fraction is independent of the fitting process (see also the circularity concern in the time-delay construction).
  3. [results and validation] No validation against known thermal failure modes (e.g., cable joint overheating records or asset temperature measurements) or error quantification (confidence intervals on the attributed proportions) is reported, leaving the reliability of the distinction between Due-to-HW and Due-to-Others untested.
minor comments (2)
  1. [methods] Notation for the normalized covariance and the Excess Heat Factor threshold should be defined explicitly with equation numbers rather than inline text.
  2. [figures] Figure captions should state the exact number of HW events and total summer faults analyzed so that the variability claim can be assessed at a glance.

Simulated Author's Rebuttal

3 responses · 1 unresolved

We thank the referee for the thorough and constructive review of our manuscript. We provide point-by-point responses to each major comment below.

read point-by-point responses

  1. Referee: [methods (covariance attribution)] The covariance-based attribution criterion (described in the methods) treats statistical alignment between temperature and fault timing as evidence of direct thermal causation, yet the manuscript provides no explicit partialling-out of confounders such as coincident load or demand spikes that also covary with temperature. This assumption is load-bearing for the central claim that the attributed proportion reflects HW thermal impact rather than seasonal correlation.

    Authors: The covariance-based approach identifies faults whose timing aligns statistically with temperature variations during heatwaves, which we interpret as indicative of thermal stress mechanisms. While we recognize that this does not explicitly partial out all possible confounders such as load, many such factors are themselves temperature-dependent during heatwaves. In the revised manuscript, we will expand the discussion section to acknowledge this limitation and note that future work could incorporate additional covariates if data permits. revision: partial

  2. Referee: [time-delay model] The time-delay model is defined by maximizing the normalized covariance between the temperature series and the shifted fault-duration series; because the lag itself is chosen by the same fitting step used for attribution, it is unclear whether the resulting lag or the Due-to-HW fraction is independent of the fitting process (see also the circularity concern in the time-delay construction).

    Authors: The procedure involves first determining the optimal lag by maximizing the normalized covariance across possible shifts, and subsequently applying the attribution criterion with this fixed lag to compute the Due-to-HW proportion. This sequential approach ensures the lag selection is independent of the final attribution calculation. We will revise the methods description to explicitly outline this two-step process and include a sensitivity analysis to demonstrate robustness. revision: yes

  3. Referee: [results and validation] No validation against known thermal failure modes (e.g., cable joint overheating records or asset temperature measurements) or error quantification (confidence intervals on the attributed proportions) is reported, leaving the reliability of the distinction between Due-to-HW and Due-to-Others untested.

    Authors: We agree that direct validation with asset-level data would be valuable; however, the available operational dataset does not include asset temperature measurements or specific failure mode records. We will incorporate bootstrap-derived confidence intervals for the attributed proportions in the results. Additionally, we will add a limitations subsection discussing the reliance on statistical attribution without direct physical validation. revision: partial

standing simulated objections not resolved
  • Direct validation of the attribution against physical asset measurements or known thermal failure records, as these are not present in the six-year operational fault dataset.

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper proposes a data-driven identification method: HWs are characterized via Excess Heat Factor, faults are attributed via a covariance-based criterion that labels them Due-to-HW when occurrence is statistically consistent with thermal mechanisms (using maximized normalized covariance to set the time lag), and the resulting proportion is reported as the direct empirical output of applying that criterion to six years of network data. No mathematical derivation chain exists in which an independent prediction or first-principles result is claimed and then shown to equal its inputs by construction; the attribution fractions are explicitly the output of the defined fitting criterion rather than a separate prediction. No self-citation load-bearing steps, uniqueness theorems, or ansatz smuggling appear in the provided text. The analysis is therefore self-contained as an application of the proposed statistical procedure.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Abstract-only review limits visibility; the covariance criterion and time-delay maximization introduce at least one fitted parameter (the lag) and a domain assumption that statistical covariance alignment equals thermal causation.

free parameters (1)
  • time lag in delay model
    Estimated by maximizing normalized covariance between temperature and shifted fault series, therefore fitted to the observed data.
axioms (1)
  • domain assumption Statistical consistency via covariance with HW thermal mechanisms implies direct attribution to heatwave stress rather than coincidence or confounding factors.
    This premise underpins the entire attribution criterion described in the abstract.

pith-pipeline@v0.9.1-grok · 5791 in / 1326 out tokens · 47422 ms · 2026-07-01T04:44:20.611982+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

43 extracted references · 18 canonical work pages

  1. [1]

    Hence, indiscriminately attributing all HW-period faults to thermal stress overestimates system vulnerability and misleads asset-management decisions

    Due-to-Heatwaves Faults in Urban Distribution System: An Identification Approach Andrea Mazza, Haoke Wu Abstract Distribution system faults occurring during heatwaves (HWs) are not all caused by the HW itself: concurrent factors such as asset ageing, mechanical defects, soil contamination, and operational constraints contribute independently. Hence, indis...

    2017

  2. [2]

    and multi-parameter sensing installations on underground lines (Peretto et al. 2018). • Fault prediction: these studies aim to forecast fault occurrence to support DSO asset management and proactive operations. For example, (Atrigna et al

    2018

  3. [3]

    compare machine-learning models for predicting MV faults under HW conditions, ranging from data-centric analyses to operationally oriented frameworks targeting reduced response time during fault events. • System-level analyses: These approaches aim to study resilience and reliability by statistically linking the fault occurrence to meteorological conditio...

    2018

  4. [4]

    The faults occurring during a HW, however, are not necessarily all related to the HW occurence

    and refined in (Tolika 2019), has been used, even because it have demonstrated its effectiveness with respect to other metrics, as for example shown in in (Mazza and Wu 2025). The faults occurring during a HW, however, are not necessarily all related to the HW occurence. In fact, distribution system reliability is influenced by multiple multiple factors, ...

    2019

  5. [5]

    Earlier studies (Amicarelli et al

    Fault-level attribution framework. Earlier studies (Amicarelli et al. 2018)–(Mazza et al

    2018

  6. [6]

    Multi-year system-level evidence. The framework is applied to six years (2019–2024) of operational data from an urban Italian MV network, documenting a deteriorating trend in fault rate and MTBF, and demonstrating that the KS-test rejection of exponential TBF distributions has direct implications for reliability modelling practice. The reminder of the pap...

    2019

  7. [7]

    2025), and aligns with the operational definition of HW-attributable fault adopted by Italian regulatory practice (ARERA 2017)

    and (Ciapessoni et al. 2025), and aligns with the operational definition of HW-attributable fault adopted by Italian regulatory practice (ARERA 2017). 2.2 Identification model of Due-to-HW faults HWs are commonly described as prolonged periods of abnormally high temperatures relative to local climatological conditions. For distribution system resilience a...

    2025

  8. [8]

    and Italy (Andrea Mazza et al. 2021). Alternative indicators, such as the Index of HWs (IHW) used by some distribution system operators (Pompili et al. 2021), have also been reported. A comparative study in (Mazza and Wu

    2021

  9. [9]

    #$,!-⋅max)1,𝐸𝐻𝐼&''(,!- where 𝐸𝐻𝐼

    shows that the EHF outperforms alternative indices in both HW-period identification and statistical correlation with fault occurrences. Accordingly, the EHF is adopted in this work as the primary metric for quantifying HW conditions. The EHF is defined on a daily basis and integrates both long-term climatological characteristics and short-term temperature...

    2009

  10. [10]

    To enable consistent comparison across different shifts, the normalised covariance, equivalent to the Pearson correlation coefficient (Nahler 2009), is employed

    The covariance between each shifted dataset and the temperature vector 𝒆 is then evaluated. To enable consistent comparison across different shifts, the normalised covariance, equivalent to the Pearson correlation coefficient (Nahler 2009), is employed. The normalized covariance 𝐶AB(CADE) ranges from −1 to 1 (Ledoit and Wolf 2022). The optimal time delay ...

    2009

  11. [11]

    (8)–(9) to label it as Due-to-HW or Due-to-Others

    Block A performs the Due-to-HW fault identification loop: after computing the daily EHF series and the HW period, the algorithm iterates over every fault recorded within the HW period, recomputing Cov(𝒉,𝒇):!(#)) after removing each fault in turn and applying the decision rule of Eqs. (8)–(9) to label it as Due-to-HW or Due-to-Others. Block B then takes th...

    2021

  12. [12]

    The HW day is defined as the time when the EHF exceeds

    This historical data enables the identification of HWs during the period from 2019 to 2024 using the EHF. The HW day is defined as the time when the EHF exceeds

    2019

  13. [13]

    4, the year 2019 experienced the highest HW extreme, while years 2022 and 2024 had the longest HWs

    As shown in Fig. 4, the year 2019 experienced the highest HW extreme, while years 2022 and 2024 had the longest HWs. In contrast, the year 2021 recorded the fewest HWs, with only two identified. As for the total number of HW days, 2022 recorded extreme HW events (42 days). In contrast, the year 2020 stands out for having very few HW days. Since the effect...

    2019

  14. [14]

    This definition is physically motivated: underground MV cables and their joints respond to ambient and soil temperature through a thermal time constant governed by the surrounding soil’s volumetric heat capacity and thermal conductivity. During sustained hot-dry conditions, soil moisture depletion is a cumulative process that can persist for several weeks...

    2016

  15. [15]

    Mazza et al

    and with the prior work of the authors (A. Mazza et al. 2021), facilitating regulatory comparability. HWs by EHF during 2019-2024. Number of HW period and Non-HW period Days (2019–2024) Year HW Period (days) Non-HW Period (days) 2019 100 265 2020 66 300 2021 76 289 2022 112 253 2023 102 263 2024 84 282 4.3 Due-to-HW faults identification The filtering of ...

    2021

  16. [16]

    filtering are visualised to demonstrate the effect of the proposed methodology. In the unfiltered case the temperature and fault-duration series show weak and temporally incoherent alignment; after filtering, a better-defined lag structure emerges, with the shifted Due-to-HW fault-duration series aligning markedly better with the hourly temperature signal...

    2021

  17. [17]

    exceed those observed in winter, indicating that under non-extreme thermal conditions, summer operation did not inherently exhibit lower reliability. 4.5.2 Analysis of the Empirical Cumulative Distribution Function (ECDF) Using the calculated MTBF, ECDFs are generated for faults during the non-HW period, Due-to-HW faults during HW period and Due-to-Others...

    2019

  18. [18]

    Assessment of the Resilience of the Electrical Distribution Grid: e-distribuzione Approach

    “Assessment of the Resilience of the Electrical Distribution Grid: e-distribuzione Approach.” 2018 AEIT International Annual Conference (Bari, Italy), 1–6. https://doi.org/10.23919/AEIT.2018.8577322. Anders, G. J

    work page doi:10.23919/aeit.2018.8577322 2018

  19. [19]

    Rating of Electric Power Cables in Unfavorable Thermal Environment

    “Rating of Electric Power Cables in Unfavorable Thermal Environment.” IEEE Electrical Insulation Magazine 21 (6): 44–44. https://doi.org/10.1109/MEI.2005.1541500. Anderson, K. et al

    work page doi:10.1109/mei.2005.1541500 2005

  20. [20]

    https://www.arera.it/atti-e-provvedimenti/dettaglio/17/645-17

    645/2017/R/EEL: Increasing the Resilience of Electricity Transmission and Distribution Networks. https://www.arera.it/atti-e-provvedimenti/dettaglio/17/645-17. ARPA Piemonte

    2017

  21. [21]

    Effects of Heatwaves on the Failure of Power Distribution Grids: a Fault Prediction System Based on Machine Learning

    “Effects of Heatwaves on the Failure of Power Distribution Grids: a Fault Prediction System Based on Machine Learning.” 2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC / i&CPS Europe) (Bari, Italy), 1–5. https://doi.org/10.1109/EEEIC/ICPSEurope51590.2021.95847...

    work page doi:10.1109/eeeic/icpseurope51590.2021.9584751 2021

  22. [22]

    Effects of Heatwaves on the Failure of Power Distribution Grids: A Fault Prediction System Based on Machine Learning

    “Effects of Heatwaves on the Failure of Power Distribution Grids: A Fault Prediction System Based on Machine Learning.” 2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC/i&CPS Europe), 1–5. Bragatto, T. et al

    2021

  23. [23]

    Assessment and Possible Solution to Increase Resilience: Heat Waves in Terni Distribution Grid

    “Assessment and Possible Solution to Increase Resilience: Heat Waves in Terni Distribution Grid.” 2019 AEIT International Annual Conference (AEIT) (Florence, Italy), 1–6. https://doi.org/10.23919/AEIT.2019.8893361. Calcara, L. et al

    work page doi:10.23919/aeit.2019.8893361 2019

  24. [24]

    Heatwaves and Duration of Distribution System Faults: A Comparison of Different Indexes

    “Dielectric Measurement Protocols on Medium Voltage Cable Joints at Variable Temperatures.” 2025 AEIT Conference (Amantea (CS), Italy), 1–4. https://doi.org/10.23919/AEIT67669.2025.11218134. Carpaneto, Enrico, and Gianfranco Chicco

    work page doi:10.23919/aeit67669.2025.11218134 2025

  25. [25]

    IEEE-Wiley

    Power Distribution System Reliability: Practical Methods and Applications. IEEE-Wiley. https://doi.org/10.1002/9780470459355. Ciapessoni, E. et al

    work page doi:10.1002/9780470459355

  26. [26]

    A Vulnerability-Based Approach to Quantify Short-Term Failure Probabilities of MV Cables and Joints in Presence of Heat Waves

    “A Vulnerability-Based Approach to Quantify Short-Term Failure Probabilities of MV Cables and Joints in Presence of Heat Waves.” CIRED 2024 (Chicago, USA), 483–87. https://doi.org/10.1049/icp.2024.2666. European Environment Agency (EEA)

    work page doi:10.1049/icp.2024.2666 2024

  27. [27]

    https://www.eea.europa.eu/en/analysis/publications/climate-change-impacts-and-vulnerability-2016

    European Environment Agency. https://www.eea.europa.eu/en/analysis/publications/climate-change-impacts-and-vulnerability-2016. European Parliament

    2016

  28. [28]

    https://www.europarl.europa.eu/topics/en/article/20180905STO11945/infographic-how-climate-change-is-affecting-europe

    Infographic: how climate change is affecting Europe. https://www.europarl.europa.eu/topics/en/article/20180905STO11945/infographic-how-climate-change-is-affecting-europe. Falabretti, D., L. L. Schiavo, S. Liotta, et al

    work page arXiv

  29. [29]

    The Power of (Non-) Linear Shrinking: A Review and Guide to Covariance Matrix Estimation

    “The Power of (Non-) Linear Shrinking: A Review and Guide to Covariance Matrix Estimation.” Journal of Financial Econometrics 20 (1): 187–218. https://doi.org/10.1093/jjfinec/nbab021. Liu, K., R. Zagorščak, R. J. Sandford, O. N. Cwikowski, A. Yanushkevich, and H. R. Thomas

    work page doi:10.1093/jjfinec/nbab021

  30. [30]

    Malmedal, K., C

    https://doi.org/10.3390/en15238897. Malmedal, K., C. Bates, and D. Cain

    work page doi:10.3390/en15238897

  31. [31]

    The Effect of Underground Cable Diameter on Soil Drying, Soil Thermal Resistivity and Thermal Stability

    “The Effect of Underground Cable Diameter on Soil Drying, Soil Thermal Resistivity and Thermal Stability.” 2016 IEEE Green Technologies Conference (GreenTech), 35–39. https://doi.org/10.1109/GreenTech.2016.14. Mazza, A. et al

    work page doi:10.1109/greentech.2016.14 2016

  32. [32]

    Evaluation of the Impact of Heat-Wave on Distribution System Resilience

    “Evaluation of the Impact of Heat-Wave on Distribution System Resilience.” 2021 International Conference on Smart Energy Systems and Technologies (SEST), 1–6. Mazza, A. et al

    2021

  33. [33]

    Investigation on the Impact of Heat Waves on Distribution System Failures

    “Investigation on the Impact of Heat Waves on Distribution System Failures.” 2024 IEEE 22nd Mediterranean Electrotechnical Conference (MELECON) (Porto, Portugal), 1310–14. Mazza, Andrea, Yang Zhang, Chiara Carrozzo, et al

    2024

  34. [34]

    Evaluation of the Impact of Heat-Wave on Distribution System Resilience

    “Evaluation of the Impact of Heat-Wave on Distribution System Resilience.” 2021 International Conference on Smart Energy Systems and Technologies (SEST), 1–6. Mazza, A., and H. Wu

    2021

  35. [35]

    Heatwaves and Duration of Distribution System Faults: A Comparison of Different Indexes

    “Heatwaves and Duration of Distribution System Faults: A Comparison of Different Indexes.” 2025 AEIT International Annual Conference (AEIT) (Amantea (CS), Italy), 1–6. https://doi.org/10.23919/AEIT67669.2025.11218113. Montalà-Palau, M. et al

    work page doi:10.23919/aeit67669.2025.11218113 2025

  36. [36]

    GIS-Based Approach to Improve the Resilience of the Distribution Network

    “GIS-Based Approach to Improve the Resilience of the Distribution Network.” CIRED Chicago Workshop 2024: Resilience of Electric Distribution Systems (Chicago, USA), 138–42. https://doi.org/10.1049/icp.2024.2582. Muñoz, A. M. et al

    work page doi:10.1049/icp.2024.2582 2024

  37. [37]

    Wireless Self-Powered Monitoring System for Underground Cable Joints: a Real Use-Case

    “Wireless Self-Powered Monitoring System for Underground Cable Joints: a Real Use-Case.” CIRED 2023 (Rome, Italy), 2124–28. https://doi.org/10.1049/icp.2023.1189. Nahler, Gerhard

    work page doi:10.1049/icp.2023.1189 2023

  38. [38]

    Monitoring Cable Current and Laying Environment Parameters for Assessing the Aging Rate of MV Cable Joint Insulation

    “Monitoring Cable Current and Laying Environment Parameters for Assessing the Aging Rate of MV Cable Joint Insulation.” Conference on Electrical Insulation and Dielectric Phenomena (CEIDP) (Cancun, Mexico), 390–93. https://doi.org/10.1109/CEIDP.2018.8544904. Pompili, M. et al

    work page doi:10.1109/ceidp.2018.8544904 2018

  39. [39]

    MV Underground Power Cable Joints Premature Failures

    “MV Underground Power Cable Joints Premature Failures.” 2020 AEIT International Annual Conference (AEIT) (Catania, Italy), 1–4. https://doi.org/10.23919/AEIT50178.2020.9241185. Pompili, Marco, Lorenzo Calcara, and Stefano Sangiovanni

    work page doi:10.23919/aeit50178.2020.9241185 2020

  40. [40]

    Heatwaves and Underground MV Cable Joints Failures

    “Heatwaves and Underground MV Cable Joints Failures.” 2021 AEIT International Annual Conference (AEIT), 1–5. Pompili, M., and L. Calcara

    2021

  41. [41]

    Qualification of MV Cable Joints: Partial Discharges Innovative Tests

    “Qualification of MV Cable Joints: Partial Discharges Innovative Tests.” 2024 IEEE EEEIC / i&CPS Europe (Rome, Italy), 1–4. https://doi.org/10.1109/EEEIC/ICPSEurope61470.2024.10751431. Saxena, S., S. Agrawal, and D. Basu

    work page doi:10.1109/eeeic/icpseurope61470.2024.10751431 2024

  42. [42]

    Thermal Behavior of Distribution MV Underground Cables

    “Thermal Behavior of Distribution MV Underground Cables.” 2015 AEIT International Annual Conference (AEIT) (Naples, Italy), 1–5. https://doi.org/10.1109/AEIT.2015.7415247. Tolika, K

    work page doi:10.1109/aeit.2015.7415247 2015

  43. [43]

    Data-Driven Feature Description of Heat Wave Effect on Distribution System

    “Data-Driven Feature Description of Heat Wave Effect on Distribution System.” 2019 IEEE Milan PowerTech (Milan, Italy), 1–6

    2019