Data-Driven Power Quality Evaluation in a Hospital Electrical System
| Facility | University of Lampung Hospital, Indonesia — large university teaching hospital |
| Measurement point | Main Distribution Panel (MDP) — four datasets, high-sampling-rate electrical measurements |
| Standards applied | IEEE 1159 (definitions) · IEC 61000-4-30 (measurement methods) · IEEE 519 (compliance limits) |
| Voltage & frequency | COMPLIANT — stable, within nominal limits as expected from utility supply |
| Current TDD | NON-COMPLIANT — surpassed IEEE 519 limits many times over |
| Voltage Unbalance Ratio | NON-COMPLIANT — exceeded permitted levels for periods longer than the standard allows |
| Power factor | LAGGING — indicating reactive power losses and reduced distribution efficiency |
| Root cause | Uncontrolled non-linear loads (SMPS, UPS, VFDs, imaging equipment) on internal distribution |
| Key finding | Utility supply was clean — all PQ problems originated inside the hospital’s own distribution system |
01 Context and Background
This case study presents the findings of an extensive, multi-parameter power quality assessment conducted at the University of Lampung Hospital in Indonesia — a large teaching hospital providing both clinical and academic services. The study by Nama, Despa, Tugiyono, and Bangsawan (2025) represents one of the first rigorous, data-driven PQ evaluations at a major Indonesian healthcare facility, filling a gap in regional literature where most prior PQ studies addressed only a single parameter rather than the full spectrum of disturbances.[1]
Modern hospitals are among the most demanding power quality environments in any sector. The load mix is simultaneously highly non-linear — Switch-Mode Power Supplies (SMPS) in computers and monitors, Variable Frequency Drives (VFDs) in HVAC systems, UPS systems, diagnostic imaging equipment including CT and MRI scanners — and highly sensitive, with patient monitoring, life-support, and diagnostic instruments that are vulnerable to waveform distortion, voltage unbalance, and power interruptions.[1]
The most demanding non-linear loads — diagnostic imaging, UPS systems, electronic ballasts — are the same equipment that generates the harmonic distortion that threatens the sensitive clinical instruments connected to the same distribution system. The hospital is simultaneously its own most significant internal PQ disturbance source and its own most vulnerable victim.
The study is particularly valuable because it applies internationally recognised standards throughout: IEEE 1159 for definitions, IEC 61000-4-30 for measurement methods, and IEEE 519 for compliance assessment. This makes the findings directly comparable to PQ studies in other jurisdictions and relevant to North American and European engineering practice, not only to the Indonesian context.
02 Measurement Methodology
Measurement point and instrumentation
Measurements were taken at the Main Distribution Panel (MDP) — the primary point of supply distribution inside the hospital, downstream of the utility service entrance and upstream of the individual load feeders. Four datasets were acquired using high-sampling-rate electrical measurement equipment. This measurement point captures the aggregate behaviour of all hospital loads as seen from the main supply point, which is the most representative location for assessing the overall internal PQ environment.[1]
Parameters measured
The following power quality parameters were systematically computed from the recorded waveform data:
- Three-phase voltage and current — RMS magnitudes and waveforms for all three phases
- Frequency — deviation from nominal 50 Hz
- Power factor — ratio of real power to apparent power, with leading/lagging classification
- Voltage Unbalance Ratio (VUR) — ratio of negative-sequence to positive-sequence voltage components per IEC symmetrical components definition
- Total Harmonic Distortion of voltage (THD-V) and current (THD-I)
- Total Demand Distortion (TDD) — the IEEE 519 compliance metric for current distortion, referenced to peak demand load current
THD-I is the ratio of harmonic current to the instantaneous fundamental current — it can appear extremely high under light load conditions when the fundamental is small. TDD normalises harmonic current to the system’s peak demand load current (IL), giving a stable metric that reflects the actual burden on the network regardless of load level. IEEE 519 specifies TDD limits, not THD-I limits, precisely because TDD is the quantity that determines the voltage distortion seen by all customers on the same supply.[2]
Operational pattern observed
The dataset revealed a strong positive correlation between electrical load and building operational schedule. Peak current loads consistently occurred on weekdays between 06:30 and 17:30 (Monday to Friday), with a marked decrease on weekends. This pattern is important for PQ assessment: harmonic distortion, voltage unbalance, and power factor all vary with load composition, and a single snapshot measurement would not capture the full range of conditions the distribution system experiences.[1]
03 Key Findings
Summary of compliance assessment
| Parameter | Standard / Limit | Observed result | Compliance |
|---|---|---|---|
| Supply voltage — magnitude | Within nominal limits | Stable — mean within nominal range | COMPLIANT |
| Frequency | 50 Hz ± tolerance | Stable — consistent with good utility supply | COMPLIANT |
| Voltage Unbalance Ratio (VUR) | EN 50160: ≤ 2% for 95% of week | Exceeded permitted level for periods exceeding the standard’s allowance | NON-COMPLIANT |
| Current TDD | IEEE 519: limit depends on ISC/IL ratio | Surpassed IEEE 519 limits many times over | NON-COMPLIANT |
| Power factor | Ideally ≥ 0.90 lagging | Somewhat lagging — indicating reactive power losses | MARGINAL |
| Measurement point: Main Distribution Panel (MDP). Standards: IEEE 1159 / IEC 61000-4-30 / IEEE 519. Source: Nama et al. (2025).[1] | |||
Harmonic distortion — the dominant problem
Current harmonic distortion was the most significant finding. TDD at the MDP surpassed IEEE 519 recommended values by a large margin. This is consistent with the load mix in a modern hospital: SMPS in computers, monitors, and LED lighting; UPS systems; VFDs in HVAC; and high-power diagnostic imaging equipment — all are non-linear loads that inject harmonic currents into the internal distribution system. The literature cites one reported case where a radiography machine alone produced a current THD exceeding 100%.[1]
Third, ninth, and fifteenth harmonic currents (triplen — odd multiples of 3) are zero-sequence quantities. In a three-phase system they add arithmetically in the neutral conductor rather than cancelling. A hospital with a high density of single-phase SMPS loads — computers, monitors, LED power supplies — can produce neutral currents significantly exceeding the phase conductor current. A neutral conductor sized at 100% of phase ampacity — the legacy default — is undersized for this condition and will overheat silently without tripping any overcurrent device. This is a fire risk as well as a PQ problem.
Voltage unbalance
Voltage unbalance exceeded permitted levels for durations beyond what the standard allows. In a hospital, this is particularly consequential because three-phase motor loads — HVAC compressors, fans, pumps — are sensitive to negative-sequence voltage. A voltage unbalance of 2% can produce rotor current unbalance of 6–10 times the voltage unbalance factor, with corresponding additional heating and accelerated insulation ageing. HVAC reliability is directly linked to patient comfort and infection control — a consequence that extends well beyond the electrical engineering domain.
Power factor
The lagging power factor indicates that the distribution system is supplying reactive power to inductive loads — predominantly motor loads and UPS systems — without local reactive compensation. A lagging power factor increases the apparent current in distribution conductors and transformers for a given real power demand, increasing I²R losses and reducing the effective capacity of the distribution system.
04 Root Cause Analysis
The utility was not the problem
The voltage and frequency measurements at the MDP were stable and within nominal limits — consistent with a well-regulated utility supply. The PQ problems observed were entirely of internal origin: generated by the hospital’s own non-linear loads, circulating through the hospital’s own internal distribution impedances, and affecting the hospital’s own sensitive equipment. The utility delivered a clean supply. The hospital’s internal loads degraded it.
This is the central finding, and it is consistent with the Fluke field statistic cited in IPQDF Case Study 01: the majority of PQ problems in healthcare facilities originate inside the facility. The utility meter compliance boundary is the wrong place to look for the source of internal equipment problems.
Non-linear load concentration
Modern hospitals have an exceptionally high density of non-linear loads per unit floor area compared to other building types. Every patient monitor, every infusion pump controller, every computer workstation, every LED luminaire, and every UPS system is a harmonic current source. Unlike industrial facilities where non-linear loads are concentrated in defined production areas, hospital non-linear loads are distributed throughout every ward, every corridor, every administrative office, and every diagnostic room — connected to the same distribution system as the most sensitive clinical equipment.
The strong correlation between PQ problems and operational hours (weekday peak 06:30–17:30) tells the engineer exactly what to look for: the harmonic sources are the equipment that is switched on during clinical hours — diagnostic imaging, patient monitoring, surgical suite loads. The weekend reduction confirms that the baseline harmonic environment from always-on loads (refrigeration, emergency lighting, security systems) is manageable; it is the clinical load that drives the MDP over the IEEE 519 TDD limit.
05 Recommendations
The study authors identified the following mitigation measures as priorities:[1]
- Active harmonic filtering (AHF) — adaptive cancellation of harmonic currents at the MDP or at individual load feeders. AHF adjusts to changing load composition throughout the clinical day, making it well suited to the variable harmonic environment of a hospital
- Load equalization across phases — systematic redistribution of single-phase loads among the three phases to reduce voltage unbalance at the MDP
- Reactive power compensation — local capacitor bank or active reactive compensation to improve power factor and reduce conductor losses
- Neutral conductor sizing review — assessment of triplen harmonic neutral current loading across the distribution system, with upsizing where required
- IoT-based continuous monitoring — real-time PQ monitoring system at the MDP and at key sub-distribution panels, providing early warning of developing harmonic problems before equipment failure occurs
A one-time PQ survey captures a snapshot. A hospital’s PQ environment changes with every shift, every season, and every equipment addition. The correlation between operational schedule and harmonic loading demonstrated in this study argues strongly for permanent monitoring at the MDP — not a periodic survey. The cost of a monitoring system is a fraction of the cost of one diagnostic equipment failure caused by harmonic-induced control malfunction.
06 Power Quality Perspective
This study is a textbook demonstration of the compliance paradox described in the IPQDF technical overview on power quality. The utility supply was compliant. IEEE 519 at the PCC would have shown nothing wrong. Yet inside the hospital, TDD was exceeding IEEE 519 limits by a large margin, voltage unbalance was out of specification, and the power factor was lagging — all conditions that directly threaten the reliability of clinical equipment and the safety of the distribution system.
From a utility engineering background, the finding is not surprising. Utility engineers know that the PCC is a contractual and metrological boundary, not a protection boundary for the customer’s internal equipment. A clean supply at the meter becomes a distorted supply inside the facility the moment the facility’s own non-linear loads are energised. The degree of distortion depends on the internal impedance of the distribution system — which, unlike the utility network, is not designed to absorb large harmonic currents without voltage distortion.
This study from Indonesia is representative of a finding that repeats across every healthcare facility PQ assessment in the IPQDF case study series: the utility delivers a clean supply; the hospital degrades it internally. The engineering response is not to demand better utility supply quality — it is to conduct an internal EMC audit, measure at equipment terminals rather than at the service entrance, and address the harmonic sources and distribution system inadequacies that the utility standards were never designed to control. EMC audits inside the facility are valuable. The payback is fast — particularly in healthcare, where the cost of a diagnostic instrument failure or a neutral conductor fire can dwarf the cost of the audit and mitigation combined.
References
- Gigih Forda Nama, Dikpride Despa, Tugiyono, Satria Bangsawan. “Data-Driven Evaluation of Power Quality in Hospital Electrical Systems: Case Study of University of Lampung, Indonesia.” International Journal of Electrical and Electronics Engineering, vol. 12, no. 12, pp. 104–116, 2025. DOI: 10.14445/23488379/IJEEE-V12I12P108. Open access under CC BY-NC-ND 4.0.
- IEEE Std 519-2022. IEEE Standard for Harmonic Control in Electric Power Systems. IEEE, New York, NY, 2022.
- IEEE Std 1159-2019. IEEE Recommended Practice for Monitoring Electric Power Quality. IEEE, New York, NY, 2019.
- IEC 61000-4-30:2015+AMD1:2021. Electromagnetic compatibility (EMC) — Part 4-30: Power quality measurement methods. IEC, Geneva.
This case study is based on an open-access research article published under CC BY-NC-ND 4.0:
Nama GF, Despa D, Tugiyono, Bangsawan S. “Data-Driven Evaluation of Power Quality in Hospital Electrical Systems: Case Study of University of Lampung, Indonesia.” International Journal of Electrical and Electronics Engineering, 12(12), 104–116, 2025.
DOI: 10.14445/23488379/IJEEE-V12I12P108 · Read the original article →
This case study is presented in summary and commentary form for educational purposes under the open-access terms of the original publication (CC BY-NC-ND 4.0). The PQ Perspective section (Section 6) represents IPQDF editorial commentary by Denis Ruest, M.Sc. (Applied), P.Eng. (ret.). IPQDF does not claim authorship of the original research.