Source
https://mro.massey.ac.nz/items/ea6a1399-0a5d-4b80-9a11-699836d3f4e7 — original source (opens in a new tab; the file is not redistributed)
Marshall-Tate (2017) Power Systems for Dairy Sheds (NZ dairy-shed electricity benchmark)
The core NZ dairy-shed electricity benchmark for the NI energy model (dairy-shed as an eco-village productive load)
Massey University MTech (Energy Management) thesis, 142 pp. Builds a mechanistic base-farm electricity model for a NZ dairy shed and cross-validates it against the NZ dairy-electricity survey literature. Load-bearing figures for Neobiome: NZ non-irrigated whole-shed demand 150–173 kWh/cow/yr (survey range); base-farm model total 177 (Manawatu) / 174 (Taranaki) / 172 (Bay of Plenty) kWh/cow/yr with a component breakdown; hot-water heat recovery cuts the model to 132 kWh/cow/yr (25% saving). All figures re-verified verbatim against the thesis PDF (pdftotext -layout). ⚠ North Island regions only (Manawatu / Taranaki / Bay of Plenty). A colder-groundwater region (Nelson-Tasman, Otago-Southland) would shift the hot-water and refrigeration components; transfer the per-cow envelope, not the exact regional split.
Summary
Hugh Marshall-Tate’s 2017 Massey University Master of Technology (Energy Management) thesis investigates stand-alone (off-grid) power systems for New Zealand dairy milking sheds. The work has two parts: (1) a mechanistic base-farm electricity load model for a typical NZ dairy shed, decomposed into refrigeration/milk cooling, vat heat loss, hot water, vacuum pump, water pump, effluent pump and other loads, with energy-efficiency and load-shifting variants; and (2) a HOMER Pro optimisation of solar-diesel and solar-diesel-biogas hybrid supply systems against that load. It is a NZ primary source: its base-farm demand model is built for the Manawatu region and re-run for Taranaki and Bay of Plenty, and it is cross-checked against the published NZ dairy-electricity surveys (Sims et al. 2004; Barber & Pellow 2005; Miller & Glenn 2011; Morison et al. 2007).
For Neobiome the value is the dairy-shed electricity demand benchmark: a per-cow annual load a community energy model can scale by herd size to size a community dairy’s electricity draw. The whole-shed figure is ~172–177 kWh/cow/yr in the base model (150–173 kWh/cow/yr across the NZ non-irrigated survey literature), it breaks down into named subsystems (so a partial dairy can be modelled), and it carries the efficiency levers (vat insulation, VSD vacuum pump, hot-water heat recovery) that move the number, with hot-water heat recovery from the generator being the single largest saving (to 132 kWh/cow/yr, 25% below the Manawatu base). The supply-side HOMER optimisation (PV + diesel +/- biogas) is background context that corroborates the project’s off-grid framing but is not the load-bearing figure ingested here.
Key claims
- claim: "NZ non-irrigated dairy-shed whole-shed electricity demand range. VERBATIM: 'Within the studies of dairy farms focused on New Zealand non-irrigated land, annual electricity use per cow ranged from 150 kWh/cow to 173 kWh/cow (Sims et al., 2004; Barber and Pellow, 2005; Miller and Glenn, 2011). The lower figure of 150 kWh/cow was taken from a small study of new farms in the Southland region (Morison, Gregory and Hooper, 2007).' This is the survey-literature envelope the thesis's model is validated against, and the headline per-cow demand benchmark for a NZ (non-irrigated) dairy shed. [verified verbatim, pdftotext -layout]"
source_location: "§2 'Electricity consumption', Figure 2 discussion (printed p.13; PDF p.15)"
- claim: "Base-farm model total electricity demand, three NZ regions (kWh/cow/year). VERBATIM (Table 9, 'Total' row): Manawatu = 177; Taranaki = 174; Bay Of Plenty = 172. The text: 'On an annual basis, this resulted in a variation of only 5 kWh/cow/year (Table 9).' Manawatu is the base region for the model; the three regions differ mainly by milking-time length and ground-water temperature. This is the modelled per-cow whole-shed load the NI engine can scale by herd size. [verified verbatim, pdftotext -layout]"
source_location: "Table 9 'Variation in the base farm model for three study regions (kWh/cow/year)' (printed p.52; PDF p.54; listed in ToC at p.52)"
- claim: "Component (subsystem) breakdown of the base-farm model, kWh/cow/year, for Manawatu / Taranaki / Bay Of Plenty. VERBATIM (Table 9): Vat heat loss 8 / 8 / 7; Milk cooling 25 / 26 / 26; Hot water 45 / 43 / 43; Vacuum pump 30 / 30 / 29; Water pump 34 / 33 / 33; Effluent pump 16 / 16 / 16; Other 19 / 18 / 18 (Total 177 / 174 / 172). Hot water is the single largest component, then water pump and vacuum pump, then milk cooling. This lets a community model build a partial dairy-shed load from named subsystems. [verified verbatim, pdftotext -layout]"
source_location: "Table 9 'Variation in the base farm model for three study regions (kWh/cow/year)' (printed p.52; PDF p.54)"
- claim: "Milk-vat insulation effect (refrigeration/cooling load). VERBATIM: actual NZ farms with 'uninsulated vats that were housed inside ... averaged 31.7 kWh/cow with a standard deviation of 3.12 kWh/cow'; the modelled insulated vat was '27 kWh/cow', higher than the measured insulated-vat average, 'for which the average of indoor and outdoor vats was 22 kWh/cow (standard deviation of 0.9 kWh/cow)'; and 'Insulation reduced the overall refrigeration electricity consumption by 13%.' The base-farm model refrigeration load was 33 kWh/cow. So vat insulation is a real but modest cooling-load lever (~22 insulated vs ~31.7 uninsulated kWh/cow on actual farms). [verified verbatim, pdftotext -layout]"
source_location: "§4.1 refrigeration discussion (printed pp.54-55; PDF pp.56-57)"
- claim: "Effect of five technologies on whole-shed demand vs the base model (Manawatu base = 177 kWh/cow/yr). VERBATIM (Table 10, 'per cow' kWh/year and 'Saving from base' %): Base scenario 177; Vat installation 171 (3%); Ice bank pre-cooling 204 (-15%, i.e. increases demand); Hot water heat pump 144 (14%); Variable speed drive vacuum pump 167 (6%); Hot water heat recovery 132 (25%). Text: 'The combined efficiency improvements by using three of the energy efficient technologies reduces electricity consumption to 134 kWh, which is a 24% reduction.' Hot-water heat recovery from the generator is the single largest efficiency lever (25% saving → 132 kWh/cow/yr). [verified verbatim, pdftotext -layout]"
source_location: "Table 10 'Effect of five technologies on the electricity consumption of the farm model' + following text (printed p.59; PDF p.61)"
- claim: "Surveyed NZ farm distribution of per-cow electricity consumption. VERBATIM: 'The two most common electricity consumptions were 190 to 199 kWh/cow in 13% of the samples and 170 to 179 kWh/cow in 12.6%. Between 130 and 209 kWh/cow represented 77% of farms surveyed. 13% of farms were above 210 kWh/cow and 10% below 129 kWh/cow.' The distribution is not normal around the mean; the model's 172-177 kWh/cow base sits inside the 130-209 band that covers 77% of surveyed farms. [verified verbatim, pdftotext -layout]"
source_location: "§2 herd-size / distribution discussion, Figure 5 (printed p.16; PDF p.18)"Neobiome Intelligence relevance
SUPPLIES the NZ dairy-shed electricity load item for the NI energy model (D01 demand side). A community with a small dairy (house cows supplying resident members on-site, the regulatory shape flagged at REG_031) needs an electricity load for its milking shed. This source gives it directly as a per-cow annual figure the engine scales by herd size: ~172-177 kWh/cow/yr modelled (150-173 kWh/cow/yr across the NZ non-irrigated survey range), so a 20-cow community dairy is on the order of ~3,500 kWh/yr of shed electricity before efficiency measures (20 x ~175 kWh/cow/yr). Because the figure is decomposed into named subsystems (hot water, cooling/vat, vacuum, water pump, effluent, other), a partial or manual dairy can be modelled by dropping the subsystems it does not have.
Efficiency levers are quantified, so the load is a range not a point. Hot-water heat recovery from the generator is the largest lever (to 132 kWh/cow/yr, 25% below the Manawatu base), vat insulation and a VSD vacuum pump are smaller (3% and 6%), and ice-bank pre-cooling actually increases whole-shed electricity (+15%) even though it shifts load off-peak. Three efficiency technologies combined reach ~134 kWh/cow/yr (24% reduction). For a self-sufficient community optimising its own generation, the design-relevant envelope is roughly ~130-180 kWh/cow/yr depending on efficiency measures.
Regional and scope caveats for transfer. The model is built for North Island regions (Manawatu / Taranaki / Bay of Plenty) whose main differences are milking-time length and ground-water temperature; the three totals span only 5 kWh/cow/yr. A colder-groundwater region (Nelson-Tasman, Otago-Southland) would raise the hot-water component and lower refrigeration, so transfer the per-cow envelope and the subsystem structure rather than the exact regional split. The figures are for a stand-alone (non-irrigated) milking shed and exclude irrigation load, which the same literature shows can push consumption above 250 kWh/cow.
Supply-side context (not the ingested figure). The thesis also optimises solar-diesel and solar-diesel-biogas off-grid supply for the shed (HOMER Pro), concluding vat insulation + VSD + generator heat recovery are good investments, ice banks are marginal below ~370 cows, and superheat heat pumps and biogas were poor investments in most cases. This corroborates the project’s off-grid framing and the diesel-genset-plus-battery backstop logic on D01, but the load-bearing contribution here is the demand benchmark, not the supply optimisation.
Research targets
No new research target opened, and none resolved. The corpus dairy backlog (RT_387 community-dairy CapEx/OpEx, RT_388 butter fat-recovery, RT_394 tallow/lard yield, RT_366 farm-scale solar payback) is about dairy production cost and food output, not shed electricity load; this source fills the electricity-load item cleanly and leaves no obvious quantitative gap requiring a new RT. If a later build needs a South Island / colder-groundwater dairy-shed load, that would be a recalibration note on this page, not a new RT.
Notes
- NZ primary,
data_quality: verified. The raw is the genuine Massey University MTech thesis (142 pp), not an AI synthesis; every figure applied here was re-verified verbatim against the thesis PDF (pdftotext -layout) at draft, and the model figures were cross-checked against the thesis’s own NZ survey-literature validation (the 150-173 kWh/cow range and the 77%-of-farms 130-209 band). Verified rather than medium because the source is the primary itself and its headline figure is corroborated by multiple independent NZ surveys it cites. - Page citations. Cited as the thesis’s own printed page numbers with the PDF/viewer page in parentheses (printed = PDF page − 2). Named tables/figures (Table 9, Table 10, Figure 2, Figure 5) make each claim unambiguous under either convention.
Retrieval provenance
- Upstream source: Marshall-Tate, H. (2017) Power Systems for Dairy Sheds, Master of Technology (Energy Management) thesis, Massey University, 142 pp — the authoritative thesis PDF.
- Prepared by: retrieval pass (dairy-shed and coolstore energy-primaries retrieval run) on 2026-07-25.
- Prompt / target: recorded target was to retrieve the NZ dairy-shed electricity benchmark primary (Marshall-Tate 2017), verifying the per-cow demand range, the Table 9 regional base-farm model and component breakdown, the vat-insulation figures, the hot-water heat-recovery saving, and the surveyed farm distribution against the thesis.
- Verification: the retrieved file is the genuine Massey thesis PDF (not an AI extraction), and every figure applied here was re-verified verbatim against that PDF (pdftotext -layout) at draft time, so this exceeds the “AI-extracted, unverified” baseline and is held at
data_quality: verified.
Connections
Links to
EDT domains (1): D01: Renewable Energy & Storage Systems
Sources (1): REG_031
Referenced by
EDT domains (1): D01: Renewable Energy & Storage Systems
Sources (1): CR_057