Solutions / Engineering guide
Wind Turbine Annual Energy: Inputs, Calculation & Uncertainty
A useful annual energy estimate explains how site wind becomes electricity at a defined connection point. Build the calculation from a documented resource, an applicable turbine power curve and explicit operating assumptions, then show which parts of the result remain uncertain.
Calculate the energy your project can use.
Rated power identifies equipment capacity. Annual energy requires a site-specific calculation; useful energy also depends on when the load, grid or other equipment can accept it.
01 / Energy boundary
State exactly which kWh the estimate represents.
Write the measurement point beside every energy result: generator terminals, converter output, site connection or energy accepted by the load.
Write the measurement point beside every energy result: generator terminals, converter output, site connection or energy accepted by the load. Also state whether the figure describes one turbine or a complete installation, and identify the assessment year. This prevents a generator-side estimate being presented as delivered site energy.
Choose the question before calculating. A grid-connected project may need export and self-consumption separated; an independent supply may need energy used by essential loads. Keep the applicable equipment arrangement with the power curve, including its control version and measurement location.
02 / Wind inputs
Use the wind distribution or an aligned time series.
Annual mean wind speed alone does not identify how often each wind speed occurs.
Annual mean wind speed alone does not identify how often each wind speed occurs. A distribution records the frequency of speed ranges; a time series retains their sequence. SAM resource guidance distinguishes annual distribution methods from time-series methods needed for time-dependent analysis.
For your calculation file, record coordinates, measurement or model height, period, units and any hub-height adjustment. Keep the original resource file and a separate processed version. If only a short campaign exists, label the result for that period unless the assessor provides a justified method for annual representation.
03 / Calculation
Multiply electrical power by the hours represented.
Interval method: E = Σ(P_t × Δt), where P is electrical power in kW and Δt is hours.
Interval method: E = Σ(P_t × Δt), where P is electrical power in kW and Δt is hours. A frequency-bin calculation similarly sums power assigned to each wind-speed bin multiplied by that bin’s hours. State the interpolation, out-of-range treatment and control assumptions used to obtain P.
Hypothetical arithmetic example: an invented profile supplies 2 kW for 100 hours and 5 kW for 60 hours: 2 × 100 + 5 × 60 = 500 kWh. This covers only those 160 hours. It is neither an annual estimate nor a MARSTON power curve or yield. Remaining periods must be accounted for separately.
04 / Operating deductions
Record deductions once, at the correct boundary.
Create a ledger for assumptions such as downtime, downstream cable loss, auxiliary demand or restricted export.
Create a ledger for assumptions such as downtime, downstream cable loss, auxiliary demand or restricted export. Each row should name the affected equipment, time period, basis and whether the power curve already includes it. A converter-output curve should not receive the same converter loss deduction again.
Separate curtailment from conversion loss. Curtailment concerns generation that cannot be accepted or is intentionally restricted in the proposed operating case. Derive it from acceptance limits and time alignment where those data exist. An arbitrary reduction factor is a scenario assumption, not evidence of how the installation will operate.
05 / Confidence
Show uncertainty separately from expected deductions.
List the unresolved inputs beside the result: wind representation, height adjustment, curve applicability and planned availability, for example.
List the unresolved inputs beside the result: wind representation, height adjustment, curve applicability and planned availability, for example. Produce clearly labelled sensitivity cases by changing one input or a defined set of assumptions. Preserve each case’s inputs so a reviewer can reproduce why the number changed.
SAM’s uncertainty documentation treats uncertainty as an analysis of distributions and input factors. A low-case sensitivity is not automatically a P90 estimate. Do not attach an exceedance probability unless the supporting statistical method and data justify that designation.
06 / Project decision
Compare generated, accepted and useful energy.
For loads or export restrictions, retain an interval table with generation, demand, accepted wind energy and rejected surplus.
For loads or export restrictions, retain an interval table with generation, demand, accepted wind energy and rejected surplus. Where storage or backup is proposed, the project model must include the actual equipment and dispatch constraints. Equal annual generation and demand do not establish uninterrupted supply.
MARSTON’s current range covers 5, 10, 20, 30 and 50 kW electric-yaw turbines. Request the evidence applicable to the candidate configuration before modelling output. Combine this calculation with the project cost assessment; no fixed annual yield is promised by the model name.
Continue the review with Wind Measurement Data Quality: A Project Handover Guide.
Continue the review with Wind vs Solar for Remote Power: Compare the Project.
Continue the review with Small Wind Project Cost: Build a Complete Comparison.
Further project guides: Load Profiles for Small Wind Power: kW, kWh & Timing · One Wind Turbine vs Multiple Small Turbines.
Further project guides: Wind Turbine Start-Up, Cut-In & Operating Wind Speeds · Average Wind Speed & Wind-Speed Distributions · Wind Turbine Measurement Calibration & Uncertainty Review · Wind Turbine Availability, Downtime & Missing Records.
Further project guides: Wind Power Self-Consumption & Self-Sufficiency Metrics · Electricity Bills: Inputs for a Wind Project Assessment · Wind vs Diesel Power: Compare the Same Remote Service · Wind Power vs Grid Extension for Remote Sites.
Review seasonal wind and load matching, long-term resource and MCP review and P50/P90 evidence and probability labels alongside the energy calculation and its assumptions.
General background: NLR System Advisor Model: Wind Resource · NLR System Advisor Model: Wind Uncertainties. These references explain industry concepts; they are separate from the applicable MARSTON product and project documents.
Prepare the information for this review.
Use the checklist to organise the project information. Mark unavailable items and retain document references.
| Review item | Information to provide | Purpose |
|---|---|---|
| Output point | One defined meter or connection boundary | Keep every kWh result comparable. |
| Resource file | Coordinates, height, timestamps and coverage | Identify the wind represented. |
| Power mapping | Curve revision, units and interpolation method | Reproduce the electrical-power calculation. |
| Deduction ledger | Loss, downtime and acceptance assumptions | Avoid double-counting reductions. |
| Result cases | Generated, accepted and useful energy with sensitivities | Connect estimates to the project decision. |
Engineering questions.
Can I multiply rated kW by 8,760?
That arithmetic assumes continuous rated output through a non-leap year. It is not a site yield estimate; use the applicable curve, resource and operating assumptions.
Is annual mean wind speed enough?
It does not specify the speed distribution or timing. Identify the additional resource assumptions needed for the calculation and the decision being assessed.
Can a short measurement campaign predict a whole year?
Only through an appropriate, explained assessment method. A period-total calculation alone must retain its actual period label.
What is the difference between gross and useful energy?
Gross energy follows the chosen generation boundary. Useful energy is the portion accepted for the stated purpose after the relevant restrictions and downstream effects.
Does a conservative case equal P90?
No. A probability label needs a justified uncertainty model; a sensitivity case describes altered assumptions without establishing that probability.
What should accompany an energy estimate?
The resource file, curve identity, output boundary, processing method, deduction ledger and result cases, with unresolved assumptions clearly identified.
Technical enquiry
Bring your project information to the discussion.
A useful annual energy estimate explains how site wind becomes electricity at a defined connection point. Share the available information and the equipment scope you want reviewed in a technical quotation.