Engineers Clinic
AdvancedEngineering trackFP-EE-05

District-cooling plant: electrical demand, transformer checks and detuned power-factor correction

Work out the real electrical demand of a district-cooling plant through a Gulf summer day, check its transformers, and design a detuned power-factor correction bank that will not blow up the way the last one did.

Core EngineeringIndustrial Automation
6 Guided Steps36h Internship TierSubmit: Load list + hourly demand model + transformer study + PFC sizing + resonance check script + APFC specification and plant report (PDF or Markdown)Verified CertificateAI Rubric Graded
Program Overview

Industry project · for Al Qudra District Cooling, operator of a 16,000 TR chilled-water plant serving 40 towers in Dubai

Al Qudra's Plant 3 has eight 2,000 TR centrifugal chillers on an 11 kV bus and a 400 V auxiliary system on four 2,500 kVA transformers. The 400 V side feeds condenser-water pumps, primary and secondary chilled-water pumps, 16 cooling-tower fans and plant building services. Many of the pumps and all the tower fans run on 6-pulse VFDs.

Two problems landed on the electrical engineer's desk this summer: - The supply agreement asks for a power factor of at least 0.95 lagging, but the 400 V buses run at about 0.86 at peak. The plant is paying a penalty. - A fixed 600 kvar capacitor bank fitted three years ago failed twice. Its capacitors bulged and its fuses blew. The maintenance team suspects harmonics.

Before buying anything, the plant manager wants a proper study. On a design day of 46 °C dry bulb, the cooling load climbs from about 9,000 TR at 04:00 to 15,200 TR at 15:00. In January it barely reaches 5,000 TR.

You will build the load list, model hourly electrical demand from the cooling load and chiller staging, check the auxiliary transformers including an N-1 case, size automatic power-factor correction for each 400 V bus, and check the resonance risk with the VFD harmonics before choosing a detuning ratio. Everything is calculation and simulation in Python or a spreadsheet; no plant access is needed.

Done looks like a report the plant manager can use to procure a new APFC panel: demand profiles for summer and winter, transformer loading, kvar per bus with step sizes, a resonance check that explains why the old bank failed, and a specification for the new detuned bank. If PF correction at 400 V cannot reach 0.95 at the utility meter on its own, say what else is needed.

Load list and maximum demandHourly demand modellingTransformer loading and N-1 checksPower-factor correction sizingHarmonic resonance analysisTechnical specification writing
What you submit

Load list + hourly demand model + transformer study + PFC sizing + resonance check script + APFC specification and plant report (PDF or Markdown)

Stack: Python (numpy, pandas, matplotlib) or a spreadsheet, Chiller and VFD datasheets, draw.io or any CAD tool, Markdown or Word

How it is graded
  • Credible demand model25

    The load list separates running and standby duty, and hourly demand for summer and winter follows from the cooling load and chiller staging with kW/TR stated.

  • Correct transformer checks15

    Loading for each 400 V transformer is calculated at peak, including an N-1 case, with the result compared against rating and ambient.

  • Right PF correction25

    kvar per bus is calculated from the real reactive demand, the steps suit the load swing, and the target PF is met or the gap is explained.

  • Resonance understood25

    The parallel resonance order is calculated for the old bank and the new options, the cause of the old failure is explained, and the detuning ratio is justified.

  • Procurement-ready report10

    The APFC specification is complete enough for vendors to quote, and the summary is clear to the plant manager.

Pass 60% · Merit 75% · Distinction 90%

Learning Outcomes
Build a plant load list with running and standby duty
model hourly electrical demand from cooling load
check transformer loading under normal and N-1 conditions
size automatic PF correction with sensible steps
predict parallel resonance and pick a detuning ratio
Target Audience

Engineering Students

Diploma, B.Tech, B.E., BCA, MCA, or M.Tech students seeking practical experience.

Aspiring Developers

Developers who want to build proof-of-work capstone projects for tech resumes.

Job Seekers

Graduates looking for AI-evaluated code review & verified industry credentials.