Why refrigeration changes the solar equation
Refrigeration dominates electricity demand in food retail. In Ukrainian supermarkets and convenience stores, 40-65% of annual kWh often comes from medium-temperature display cases, low-temperature freezers, walk-ins, and compressor racks. Unlike HVAC or lighting, these loads run year-round and follow ambient temperature and business hours. That profile creates both an opportunity and a risk: PV can shave a large, predictable daytime base load, but only if the design correctly models compressor cycling, defrost schedules, and seasonal performance of the refrigeration plant.
When we assess projects, we start with measured interval data. If you only have monthly bills, insist on installing temporary loggers on the main meter and on the refrigeration distribution board for at least two weeks in both summer and shoulder season. The goal is to identify the true daytime baseload and its variability. For a typical 1,000 m² supermarket in central Ukraine, daytime baseload might sit at 45-60 kW in summer and 30-40 kW in spring-autumn, with 24/7 minimum load rarely dropping below 20-25 kW due to cold rooms and controls.
This is why projects framed as supermarket solar for refrigeration load design and build require a different approach than generic commercial PV. You are not chasing a bill-credit story - you are stabilizing a mission-critical load with a generation asset that matches hours of operation, minimizes export, and respects food safety constraints.
What to measure before you model
Establish accurate electrical and thermal baselines
- Interval kWh and kW demand by end-use - main meter, refrigeration rack(s), HVAC, lighting, plug loads. Aim for 5-15 minute granularity.
- Rack COP/EER across ambient conditions - use manufacturer curves and validate with suction/discharge pressures and mass flow when available.
- Defrost types and timing - electric vs hot gas, staggered vs simultaneous, because defrost spikes distort PV self-consumption.
- Case door retrofits and night curtains - these change load shapes materially; model the “to-be” state, not just “as-is.”
Site and grid constraints to document
- Roof usable area after setbacks, shading, and fire aisles - count only net unobstructed m².
- Single-line diagram and protective device ratings - check available fault levels and reverse-power limits from the DSO.
- Net billing terms and export metering - confirm tariff structure and settlement window for your oblast.
- Structural checks - older precast roofs may need localized reinforcement under ballast paths.
Build a refrigeration-aware energy model
At minimum, your model should integrate four pieces: PV production, refrigeration thermal load, rack performance vs ambient, and store operating hours.
- Convert thermal load to electrical load across the year. If the medium-temp load is 70 kWth at +2 °C and the average effective COP is 2.2 in July and 2.8 in April, electrical power swings from ~32 kW to ~25 kW before defrost and auxiliaries. Add fan, pump, and control loads with their own schedules.
- Overlay defrost schedules. A 1.5-2.5 kW spike per case for 10-20 minutes can add 8-12% to the hourly peak if multiple cases defrost together. Staggering defrosts reduces coincidence with PV peaks.
- Use local irradiance and temperature data. In Ukraine, PV output and refrigeration load are positively correlated in summer - a design advantage. However, winter solar is low while refrigeration still runs, so annual self-consumption ratios depend on limiting export in summer.
- Apply coincidence and clipping logic. Start by sizing DC such that daytime self-consumption exceeds 85-90% in shoulder months, not only in July. Then test inverter loading ratio (ILR) between 1.2 and 1.35 to harvest early/late sun without increasing mid-day export too much.
Sizing logic that prevents surprises
A practical rule-of-thumb, then refine
For a supermarket with a stable 35-50 kW daytime refrigeration baseload, a first-pass array of 120-180 kWp often balances self-consumption and roof use. This is not a prescription - it’s a starting hypothesis to be validated against 15-minute data. After simulation, many stores land on inverter AC sizes of 80-120 kVA with ILR ~1.25-1.35, depending on roof geometry and shading.
The grid interface is a business decision
If the store participates in grid tied PV for retail net billing installation, your design target shifts from “zero export” to “economically optimal export.” You will maximize self-consumed kWh first, then accept limited export during high-irradiance hours if net billing credits make financial sense in your region. If export compensation is weak, prioritize demand shaving by adding modest battery capacity for intra-day shifting rather than chasing more DC.
Battery or no battery for refrigeration sites
You rarely need long-duration storage for supermarkets. What pays back is short-duration, high-cycle batteries sized for 0.5-1.5 hours at 20-40 kW to flatten midday peaks, catch defrost clusters, and ride through brief grid disturbances. Batteries also help the inverter ride through rapid cloud transients so case temperatures remain stable and compressors avoid short-cycling. Keep the BMS integrated with case controllers or rack PLCs to respect temperature setpoints during demand response events.
Standards, compliance, and food safety
Ukrainian retailers operate under food safety regimes aligned with HACCP principles. From an electrical and safety standpoint, reference IEC 60364-7-712 for PV installations on buildings, EN 378 for refrigeration systems safety and environmental requirements, and ISO 50001 for energy management systems to formalize continuous improvement. On the IT side, insist on secure monitoring - IEC 62443 concepts applied to the PV SCADA and refrigeration controllers reduce cyber risk for chain operators.
Design documentation should prove that critical cold rooms maintain temperature bands under credible contingencies: passing cloud cover, a PV inverter trip, or a short grid dip. Include a test plan in commissioning to log case temperatures, compressor status, and PV/battery behavior while you simulate step changes in irradiance and feeder voltage.
Financial model that CFOs trust
The winning business case in Ukraine blends three levers: avoided daytime purchases, demand charge reduction, and limited export value where available. Year 1 savings depend on how closely PV aligns with the refrigeration baseload. Sensitivity the model to:
- Summer vs winter solar spread - Kyiv vs Odesa will show different annual PV capacity factors.
- Product spoilage risk - quantify the avoided-loss value by ensuring PV-plus-battery improves ride-through during brief outages.
- Maintenance strategy - proactive cleaning and case gasket upkeep preserve both PV yield and refrigeration efficiency, compounding savings.
Capex benchmarks vary with roof complexity and crane logistics, but high-level ranges for quality equipment and EPC in Ukraine typically fall between 650-900 €/kWp for supermarket-scale rooftops with monitoring, commissioning, and training included. Battery adders for 30-60 kW / 30-60 kWh systems often land in the 300-450 €/kWh band, subject to brand and warranty terms. Always validate current prices with live quotes before committing.
Case patterns we see in the region
A 1,200 m² suburban food store near Lviv with a 50 kW refrigeration baseload and 25 kW other daytime loads installed 150 kWp DC with 100 kVA AC, ILR 1.5 constrained by roof layout. Self-consumption stabilized at ~88% annually without storage. A smaller urban convenience store with 18 kW refrigeration baseload selected 80 kWp DC, 60 kVA AC plus a 30 kW/30 kWh battery to shave peaks and smooth defrost-driven spikes. In both cases, alignment with store hours and staggered defrost schedules drove the economics more than panel efficiency deltas.
Common design pitfalls - and how to avoid them
Technical traps to watch
- Ignoring defrost clustering - leads to lower than expected self-consumption and nuisance high peak demand.
- Oversizing DC without policy support - creates summer export with weak revenue, stretching payback.
- Neglecting roof maintenance pathways - poor access increases O and M costs and reduces cleaning frequency.
- Failing to coordinate controls - PV curtailment and rack PLCs must “talk” so temperature bands are never compromised.
Procurement and EPC mistakes
- Lowest-price inverter or module choice without service footprint in Ukraine - future downtime risk rises.
- No performance guarantees tied to measured self-consumption ratio - the KPI most correlated with ROI.
- Weak commissioning plans - skipping thermologging during acceptance hides refrigeration stability issues.
How Dolya Solar Energy designs for refrigeration certainty
Our engineering workflow for grocery and food retail reflects global best practice and local constraints. We start with data acquisition, build a refrigeration-aware digital twin, and iterate array sizing to maximize on-site use. Where it improves ROI and resilience, we configure compact storage and verify performance against a food safety test plan. For distribution centers and ice-cream plants, we extend the same methodology to cold storage solar with refrigeration support "turnkey" solutions, integrating PV, short-duration batteries, and controls that prioritize temperature stability without sacrificing energy savings.
Quick checklist for your next project
- Do you have 15-minute load data for at least two seasons, split by refrigeration and non-refrigeration?
- Has the defrost schedule been staggered to reduce coincidence with PV peak?
- Does the simulation report self-consumption by month, not just annual averages?
- Are inverter protection settings coordinated with DSO requirements and food safety contingencies?
- Is the O and M plan aligned with ISO 50001 cycles and includes both PV cleaning and case gasket inspections?
Bottom line
Refrigeration loads make retail PV projects uniquely attractive - and uniquely sensitive to modeling details. If you capture the true baseload, respect defrost dynamics, and optimize for self-consumption within Ukrainian grid rules, your project will deliver stable savings and operational resilience. The stores that win treat PV not as a commodity purchase, but as an engineered asset tailored to the thermal reality of food retail.
