What a Power Plant Energy Audit Actually Finds (and Why Heat Rate Is Where the Money Is)

Every plant knows what it spends on fuel. Very few know what it wastes.

The distinction matters because waste in a thermal plant is almost entirely invisible from the control room. Nothing alarms when a condenser fouls. No indicator lights when turbine seal clearances open by a millimetre. The steam trap that has been blowing live steam into the atmosphere for two years does not appear on any screen. The plant produces its megawatts, the fuel bill arrives, and the number looks like the cost of doing business.

A power plant energy audit is the exercise of finding out how much of that bill is buying output and how much is buying nothing at all. Done properly, it is one of the highest-return pieces of engineering work available to an operating plant, because the savings recur every year and most of the fixes have short paybacks.

This article covers what an audit actually measures, why heat rate dominates the economics, the findings that recur across almost every plant, and how to make sure the report becomes work rather than a document in a drawer.

Heat Rate: The Number That Governs Everything

For any thermal plant, heat rate is the master efficiency metric. It expresses how much fuel energy is consumed per unit of electrical energy produced, and a rising heat rate means the plant is buying more fuel to deliver the same output.

Its economic leverage is easy to underestimate until the arithmetic is done. Analysis published in POWER magazine puts the annual fuel saving from a 1% heat rate reduction at roughly $700,000 for a 500 MW coal unit running at 80% capacity factor on $2.00 per MMBtu fuel, and roughly $580,000 for a 500 MW combined cycle plant at 60% capacity factor on $3.00 per MMBtu gas.

One percent. That is a change small enough to hide inside normal operational variation, invisible without deliberate measurement, and worth more per year than most maintenance budgets. And heat rate degradation is not hypothetical: fossil plants worldwide typically lose several percent of design efficiency over their operating lives through exactly the accumulation of small, unaddressed losses that an audit exists to find.

For industrial plants that are not generating power, the equivalent framing is specific energy consumption: energy per tonne of product, per thousand litres, per unit of whatever the site makes. The principle is identical. What gets measured against output gets managed. What gets measured only as a monthly bill does not.

What an Audit Actually Measures

A credible audit is a measurement exercise, not a walk round with a clipboard. The distinction shows up immediately in the deliverable: an audit that quantifies produces a ranked list of interventions with costs and paybacks, while an audit that observes produces a list of recommendations nobody can fund because nobody can size the benefit.

Energy balance first. Everything entering the site as fuel, steam, compressed air or electricity is accounted for against everything leaving as product, power or loss. The gaps in that balance are where the money is, and building it usually exposes that a meaningful share of site energy is unaccounted for, which is itself the finding.

Performance testing against design. Each major energy conversion is tested against its design point: turbine stage efficiencies and heat rate, boiler efficiency by the losses method, condenser performance and terminal temperature difference, heat exchanger approach temperatures, pump and fan efficiencies at their actual duty points. Formal performance testing of steam turbines follows established codes such as ASME PTC 6, which matters when the results will inform a warranty claim or a capital decision.

Thermal survey. Infrared imaging across insulation, steam lines, valves, furnace casings, electrical connections and switchgear finds heat going where it should not, and finds electrical connections heading toward failure.

Steam system survey. Every trap tested, every leak logged and quantified, condensate return measured, flash steam recovery assessed, and the steam pressure levels reviewed against what the process actually requires.

Compressed air survey. Leak detection by ultrasound, pressure profile through the distribution system, and an audit of what compressed air is being used for. Compressed air is consistently the most expensive utility per unit of useful work on an industrial site and the most casually wasted.

Electrical assessment. Motor loading profiles, power factor, harmonic distortion, transformer loading and losses, and identification of motors running continuously that need not.

Operating practice review. How the plant is actually run, as distinct from how the procedures say it is run. Load dispatch between units, start-up and shutdown practice, soot blowing frequency, cleaning intervals, and the setpoints operators have quietly adjusted over the years for reasons nobody remembers.

Running a plant with no measured baseline against design? Our industrial engineering services cover plant audits, performance testing and efficiency assessment.

Identified savings from a comprehensive industrial energy audit examining compressed air, motors, HVAC, lighting and process heating as one connected system. Identified savings and realised savings are different numbers. Source: MIMAH analysis of industrial energy audit practice.

The Findings That Recur Almost Everywhere

Audit enough plants and the same items appear with a consistency that becomes almost predictable.

Failed steam traps. In a system that has not been surveyed for several years, a substantial proportion of traps will typically be found failed, and the ones that fail open blow live steam continuously. Each is a small, invisible, permanent loss. Collectively they are frequently among the largest single findings in an audit, and trap replacement is cheap with paybacks often measured in months.

Condenser fouling and air ingress. On a steam plant the condenser has enormous leverage on heat rate, because it sets the back pressure the turbine exhausts against. Fouled tubes or air leaking into the vacuum both raise back pressure, and raised back pressure directly degrades output and efficiency. Both are correctable, and both are easy to overlook because the plant continues to run.

Turbine internal losses. Seal and gland clearance degradation, blade fouling from carryover, and solid particle erosion all reduce the work extracted from each kilogram of steam. These are found by performance testing and confirmed when the machine is opened, which is why an audit and an overhaul programme belong together rather than in separate budget lines. Our guide on steam turbine overhaul covers what those findings look like once the casing is off.

Excess combustion air. Boilers commonly run with more air than combustion requires, because operators are managing a safety margin against incomplete combustion and because oxygen trim control has drifted or been switched to manual. Every unit of excess air is heated to stack temperature and thrown away.

Compressed air leaks and misuse. Leak rates of a quarter to a third of total compressed air production are common in plants that have never surveyed for them, and generating compressed air is expensive. The related finding is compressed air used for jobs a blower or a broom could do.

Motors oversized and unmanaged. Motors sized for a worst case that never occurs, running at low load and therefore at poor efficiency and power factor. Motors running continuously that could be controlled. Pumps throttled at a control valve to reduce flow, which is the hydraulic equivalent of driving with the handbrake on, where a variable speed drive would deliver the same flow for a fraction of the energy.

Insulation gaps. Insulation removed for maintenance and never replaced, damaged lagging, and uninsulated valves and flanges. Individually small, cumulatively significant, and visible instantly on a thermal survey.

Poor load dispatch. Where a site has multiple boilers or generating units, running several at part load when fewer at higher load would be more efficient. This one costs nothing to fix and is therefore the finding operators like best.

Why 15% to 25% Is a Realistic Expectation

The figure most commonly cited for a well-run industrial energy audit is that it identifies savings somewhere in the range of 15% to 25% of site energy consumption, provided the audit examines compressed air, motors, process heating and the steam system together as one connected system rather than as separate surveys.

That range surprises people, and the scepticism is fair. Two things explain it.

The first is that energy waste compounds quietly. No single item on the list above is large. A failed trap, a fouled exchanger, a throttled pump, a motor at 40% load. Each is a fraction of a percent. It is the accumulation across a plant that has never systematically looked for them that produces a double-digit total.

The second is that most plants have never actually measured. Energy is treated as a fixed cost of operating, monitored at the meter and the invoice, and nobody has ever built the balance that shows where it goes. The International Energy Agency's Energy Efficiency 2024 assessment makes the same point at a national scale: the largest efficiency opportunities are consistently in industrial systems that have not been measured rather than in technologies that have not been invented.

The honest caveat is that identified savings and realised savings are different numbers. An audit that identifies 20% and results in 6% realised has still paid for itself many times over, but the gap between the two is where most audit programmes disappoint, and it is worth understanding why.

What a single percentage point of heat rate is worth in annual fuel cost. Source: Coal-Fired Power Plant Heat Rate Improvement Options, POWER Magazine, 2014.

Turning a Report Into Realised Savings

The most common outcome of an energy audit is a well-produced document that changes nothing. Avoiding that is mostly about how the audit is structured before it starts.

Rank by payback, and be honest about capital. Findings should be sorted into no-cost operational changes, low-cost quick wins and capital projects, each with an estimated saving, an implementation cost and a payback. The no-cost and low-cost tiers usually account for a large share of the total and can proceed immediately. Presenting a single undifferentiated list guarantees that the whole thing waits for a capital cycle.

Assign every finding an owner and a date. A recommendation without a name attached is a recommendation that will not happen. This sounds trivial and it is the single biggest determinant of whether an audit produces savings.

Establish measurement and verification up front. Decide before implementation how each saving will be verified, against what baseline, and adjusted for what variables such as production rate and ambient conditions. Without this, the plant cannot demonstrate the saving, the finance function stops believing the numbers, and the next round of proposals goes unfunded. This is the mechanism by which credible energy programmes lose their credibility.

Embed it in a management system rather than repeating it as a project. ISO 50001 exists to convert episodic auditing into a continuous cycle with defined energy performance indicators, targets and management review. The value is not the certificate. It is that energy performance becomes something reviewed on a schedule by people with authority, rather than something examined once every five years by a consultant.

Connect it to the maintenance programme. Many audit findings are maintenance findings wearing different clothes. A fouled exchanger, a degraded seal, a failing bearing that has raised a motor's power draw: these belong in the same work management system as everything else, and they benefit from the same condition monitoring evidence discussed in our guide to vibration analysis for rotating equipment.

Audit findings sitting in a report nobody has actioned? Our maintenance and asset management team converts audit outputs into scoped, costed and sequenced work packages.

Where On-Site Generation Enters the Picture

An energy audit frequently ends up examining not just how efficiently energy is used but where it comes from. For industrial sites in Nigeria, Egypt and Sudan, this is often the largest finding of all, because the cost per kilowatt-hour of self-generated diesel power is in a different league from the cost of grid or solar energy.

The sequence matters, though, and it is routinely got backwards. Reducing demand comes first, because every kilowatt-hour eliminated is a kilowatt-hour that does not need generating by any source. Sizing a solar installation to a plant's current, unoptimised consumption means paying for capacity that a compressed air leak survey would have made unnecessary. Efficiency first, then generation, is the correct order and it usually reduces the capital cost of the generation project substantially. We cover the generation side of that analysis in solar for factories in Africa.

MIMAH's own audit work follows this pattern. The plant audit conducted at KRPC in Nigeria in 2025 led directly to four subsequent turbine and controls contracts, which is the shape a useful audit takes: a measurement exercise that defines a scope of work, rather than a report that describes a condition.

The eight findings that recur across almost every industrial and power plant energy audit. Source: MIMAH engineering analysis.

Frequently Asked Questions

How long does a power plant energy audit take? A focused audit of a specific system, such as steam or compressed air, can be completed in days. A comprehensive plant-wide audit with performance testing typically takes several weeks on site plus analysis time, and requires access to operating data covering a representative period rather than a single week.

What data should we have ready before an audit starts? At minimum: fuel and electricity consumption records with production output over the same periods, design heat balance and equipment datasheets, operating logs, previous performance test results if any exist, maintenance history for major plant, and single line diagrams. Auditors will measure what is missing, but every gap costs time and widens the uncertainty band on the findings.

Is an audit worth doing if we already know our equipment is old? Usually yes, and often more so. Knowing equipment is old tells you nothing about which of it is costing the most, and replacement is rarely the highest-return action available. Audits on older plant frequently find that operational and maintenance fixes deliver most of the achievable saving at a small fraction of replacement cost.

How is heat rate degradation different from normal ageing? Some degradation is genuinely irreversible, principally material changes in high-temperature components. Much of what plants attribute to ageing is not: fouling, seal clearance loss, air ingress, excess combustion air and instrument drift are all recoverable. Distinguishing the recoverable from the permanent is one of the main purposes of performance testing.

What is the difference between an energy audit and an efficiency guarantee? An audit identifies and quantifies opportunities. A guarantee is a contractual commitment, usually by an energy services provider, to deliver a specified saving with payment tied to the result. Guarantees only work where measurement and verification is rigorously defined beforehand, because the entire contract turns on how the saving is calculated.

Should the audit be done by the equipment supplier? There is an obvious conflict where the recommendation is likely to be new equipment. Independent auditing avoids that, and independence tends to produce more findings in the no-cost and low-cost tiers, which is where the best paybacks usually sit.

The Fuel You Do Not Burn

Efficiency work has an image problem. It has none of the visibility of a new installation, no ribbon to cut, and no photograph for the annual report. A replaced steam trap looks exactly like the one it replaced.

But the arithmetic is hard to argue with. A percentage point of heat rate on a large unit is a seven-figure annual sum in many fuel markets, recurring every year, requiring no new capacity and no new fuel supply. Efficiency measures are typically among the cheapest sources of effective energy available to any plant, and unlike a fuel contract, they do not reprice.

The plants that capture it are not the ones with the newest equipment. They are the ones that measure against design, that treat a rising heat rate as a fault to be investigated rather than a fact of ageing, and that give every finding an owner and a date.

Want to know what your plant is actually losing? Talk to our engineering team. We will measure against design, quantify the gap, and give you a ranked list of what to fix first with the arithmetic behind each number.