Air Preheater Maintenance — Leakage Testing, Basket Optimization & AI Performance Monitoring

By Johnson on July 23, 2026

boiler-air-preheater-maintenance-leakage-basket-optimization-ai

Air preheater leakage in a coal-fired power plant is a silent efficiency drain that most process engineers can only measure during an outage. A Ljungstrom-type air preheater operating at 12% leakage instead of the design 6% wastes approximately $1.2 to $2.4 million per year in excess fuel costs on a 500MW unit. The data to detect rising leakage exists in your DCS every second, but without AI analytics connecting these signals to specific degradation pathways, you discover the problem when your next outage leakage test reveals a number that forces an emergency basket order. Book a 30-minute walkthrough to see how iFactory detects APH degradation in real time.

STEAM TURBINE · BOILER · AIR PREHEATER AI

Stop discovering air preheater leakage 18 months after it starts

iFactory's AI-powered APH monitoring detects leakage increases, basket degradation, and heat transfer loss in real time—using data you already have—so you can plan basket replacements and seal adjustments before your next outage.

3%
Max boiler efficiency lost to APH leakage at 15% leakage rate
$2.4M
Annual excess fuel cost from 12% vs 6% leakage on a 500MW unit
8–18 Mo
Blind operating period between outage leakage tests
35–50%
Leakage reduction achievable with AI-driven seal optimization
FOUR ZONES OF APH DEGRADATION

Every air preheater has four failure modes running simultaneously

Ljungstrom and Rothemuhle air preheaters operate in a harsh environment where gas-side corrosion, particulate erosion, chemical plugging, and mechanical seal wear all progress at different rates. Understanding which zone is driving your leakage increase is the difference between a targeted seal adjustment and an unnecessary full basket replacement.

COLD END

Acid Dew Point Corrosion

When flue gas temperature drops below the sulfuric acid dew point—typically 120 to 150°C depending on fuel sulfur content—H2SO4 condenses on basket elements. This acid aggressively corrodes both carbon steel and enameled elements, thinning the heating surface and creating holes that directly increase gas-to-air leakage. Cold end corrosion is the single largest driver of basket replacement in coal-fired units.

15–25%
Basket element mass loss per outage cycle in high-sulfur units
INTERMEDIATE ZONE

Ammonium Bisulfate Plugging

In units equipped with SCR systems for NOx control, unreacted ammonia combines with SO3 to form ammonium bisulfate. ABS is a sticky, hygroscopic deposit that adheres to basket surfaces in the 200 to 320°C temperature range. It progressively plugs basket passages, increasing gas-side differential pressure, reducing heat transfer area, and creating localized flow acceleration that erodes adjacent elements. Sootblowing has limited effectiveness against established ABS deposits.

40–60%
Flow area reduction from advanced ABS plugging in SCR-equipped units
HOT END

Fly Ash Erosion

High-velocity fly ash particles impact the hot end basket elements at gas inlet temperatures of 340 to 380°C. Over thousands of operating hours, this erosive wear thins the basket profile sheets and corrugations, reducing heat transfer surface area and structural rigidity. Eroded elements are more susceptible to mechanical damage from sootblower steam impingement and can break loose, causing downstream damage to economizer tubes or creating rotating debris that damages seals.

8–15%
Heat transfer surface loss from hot end erosion over a 5-year basket life
SEAL SYSTEM

Radial, Axial, and Circumferential Seal Wear

The sealing surfaces between the rotating rotor and stationary housing wear through continuous contact during thermal expansion and contraction cycles. Differential expansion between the rotor and casing causes the rotor to warp or "saddle" over time, increasing clearances non-uniformly around the circumference. Radial seals wear at different rates at the 12 o'clock position versus the 6 o'clock position, creating leakage pathways that cannot be corrected by uniform seal adjustments.

2–4x
Leakage increase factor from non-uniform rotor warping over 3-year seal life
LEAKAGE COST ESCALATION

Every percentage point of excess leakage has a dollar value

On a 500MW coal-fired unit operating at 85% capacity factor, the relationship between air preheater leakage and boiler efficiency loss is well-established. Each percentage point of leakage above design costs approximately $400,000 to $500,000 per year in additional fuel. Here is what that escalation looks like as degradation progresses between outages.

6–7%

Design range after outage maintenance
$0 excess
8–9%

Early seal wear, minor basket fouling beginning
$400–800K/yr
10–12%

Advanced seal wear, basket corrosion and plugging compounding
$1.2–2.4M/yr
13–15%+

Systemic degradation, rotor warping, emergency action required
$2.8–4.5M/yr
THE MONITORING BLIND SPOT

Outage testing shows you a snapshot. The damage happens in between.

Process engineers rely on outage leakage testing—typically conducted every 8 to 18 months—to assess air preheater health. This approach has a fundamental limitation: it measures the cumulative result of months of degradation but provides zero visibility into how or when that degradation occurred. The result is reactive maintenance driven by lagging indicators.

WHAT OUTAGE TESTING REVEALS
Total leakage rate at the single moment of testing
Basket visual condition after shutdown and cleaning
Seal clearance measurements at ambient temperature
Differential pressure at one operating point before shutdown
Rotor runout and warping measurements at standstill
8–18 MONTHS OF BLIND OPERATION
Degradation progresses continuously. Efficiency loss compounds daily. You see none of it.
WHAT HAPPENS BETWEEN OUTAGES THAT YOU CANNOT SEE
Leakage rate trajectory and rate of increase over time
Basket fouling rate and ABS accumulation progression in SCR units
Seal wear rate from daily thermal cycling and rotor warping
Heat transfer effectiveness decline as basket surface degrades
Sootblowing effectiveness drift as deposit patterns change
Correlation between leakage increase and boiler efficiency loss in real time

Your APH leakage is increasing right now. You just cannot see it.

iFactory's AI monitoring fills the blind spot between outages with continuous leakage detection, basket health tracking, and efficiency impact quantification. Book a 30-minute demo and see the analysis on your boiler data.

REAL-TIME APH MONITORING

Four data signals that reveal air preheater degradation between outages

iFactory analyzes data streams your DCS already produces—air and gas temperatures, differential pressures, oxygen levels, and mill outlet temperatures—to build a continuous picture of air preheater health. No new sensors, no modifications to the APH, deployed on an NVIDIA appliance inside your plant network.

1

Leakage Rate Estimation from Temperature Signals

iFactory calculates real-time APH leakage rate using air inlet and outlet temperatures, gas inlet and outlet temperatures, and air and gas flow measurements. The algorithm applies heat balance equations across the air preheater to estimate the fraction of gas that bypasses the heat transfer surface through leakage paths. This leakage estimate is updated every minute and trended over time to detect rate-of-change anomalies that indicate seal or basket degradation.

2

Differential Pressure Trending for Basket Health

Gas-side and air-side differential pressure across the APH are monitored and trended against baseline profiles established during clean-basket operation. Increasing gas-side DP indicates basket plugging from fly ash, ABS deposits, or corrosion products. The rate of DP increase reveals whether fouling is gradual—suggesting normal ash accumulation—or accelerating—indicating active ABS formation or cold-end corrosion debris. Sootblower activation events are correlated to DP response to measure cleaning effectiveness in real time.

3

Heat Transfer Effectiveness Tracking

iFactory calculates the actual heat transfer effectiveness of the air preheater by comparing measured air temperature rise against the theoretical maximum based on gas temperature drop and flow rates. Declining effectiveness—after correcting for load and ambient conditions—indicates basket surface degradation from erosion, corrosion, or fouling that reduces the active heat transfer area. This metric detects degradation that does not immediately show up in leakage or DP measurements.

4

Leakage-to-Efficiency Impact Quantification

Every leakage rate estimate is automatically correlated to boiler efficiency impact using unit-specific performance curves. Process engineers see not just that leakage is increasing, but exactly how much that increase is costing in terms of heat rate degradation, excess fuel consumption, and CO2 emissions per operating day. This quantified impact turns a technical parameter into a financial decision signal for maintenance prioritization and outage planning.

DEPLOYED CAPABILITIES

What iFactory delivers for your air preheater monitoring

These capabilities run on your plant network with zero cloud dependency. They connect to your existing DCS or historian and begin producing actionable insights within the first two weeks of data collection.


Continuous leakage rate calculation and trending

Real-time APH leakage percentage calculated from heat balance, updated every minute, with configurable alert thresholds for rate-of-change and absolute level. Trend visualization shows leakage trajectory since last outage with projected leakage at next scheduled outage date.


Basket condition scoring by zone

Separate health scores for cold end, intermediate, and hot end basket sections based on differential pressure trends, heat transfer effectiveness, and operating temperature profiles. Identifies which zone is driving degradation and whether the cause is corrosion, plugging, or erosion.


ABS plugging detection for SCR-equipped units

Pattern recognition on gas-side DP trends that distinguishes normal ash accumulation from ABS-related plugging. Detects the characteristic acceleration in DP rise rate that signals active ABS formation and recommends sootblowing strategy adjustments or water wash scheduling.


Sootblowing effectiveness monitoring

Measures the actual DP reduction achieved by each sootblower pass and tracks the declining effectiveness over time. When sootblowing returns diminish below a configurable threshold, iFactory alerts that mechanical cleaning or water wash is needed, preventing wasted steam and unnecessary sootblower wear.


Seal degradation pattern analysis

iFactory analyzes leakage rate changes during load ramps and temperature transients to infer seal behavior. Non-uniform leakage patterns—where leakage increases more at certain loads or temperatures—indicate rotor warping or localized seal wear that requires targeted adjustment rather than uniform seal repositioning.


Outage planning support with degradation forecasting

Projects basket condition and leakage rate to the next scheduled outage date based on current degradation trends. Provides maintenance recommendations—basket section replacement, seal adjustment, water wash—with estimated impact on post-outage leakage rate so you can evaluate the cost-benefit of different maintenance scopes.

DEPLOYMENT IN 8–10 WEEKS

From data connection to live APH monitoring

iFactory connects to your existing DCS or historian and delivers a working air preheater monitoring system without custom development, cloud migration, or new sensor installation.

Weeks 1–3
Data Integration

Connect to your historian or OPC UA source for APH temperatures, differential pressures, air and gas flows, oxygen levels, and sootblower status. Import historical outage leakage test results for baseline calibration.


Weeks 4–6
Baseline Learning

AI models learn the normal operating signatures for your APH including leakage baseline, DP profiles, heat transfer effectiveness, and sootblower response characteristics at different load points.


Weeks 7–10
Go-Live and Validation

Live monitoring begins with iFactory operations team support. Leakage estimates are validated against next outage test results. Models are refined and full handover to your engineering team with documentation.

35–50%
Leakage reduction from baseline through AI-driven seal optimization
From 12% to 6–8% leakage by targeting specific seal adjustment points identified by pattern analysis
0.8–1.5%
Boiler efficiency recovery from optimized basket maintenance timing
By replacing only the degraded basket sections rather than full baskets, reducing outage duration and cost
$1.2–2.4M
Annual fuel cost savings on a 500MW coal-fired unit
From reduced leakage and recovered heat transfer effectiveness maintained through continuous monitoring
97%
Faster detection of APH degradation vs. outage-only testing
From 8–18 month discovery lag to continuous real-time detection with daily trending and alerts
QUESTIONS PROCESS ENGINEERS ASK

Air preheater monitoring with AI, explained

How accurate is the leakage rate calculation without direct leakage measurement?
iFactory's leakage estimation uses a heat balance approach that calculates the enthalpy imbalance across the air preheater. The method has been validated against formal outage leakage tests across multiple coal-fired units and typically achieves accuracy within plus or minus 1 to 1.5 percentage points of the measured leakage rate. While this is not as precise as a formal tracer gas test conducted during an outage, it provides continuous trending capability that makes the absolute accuracy less important than the ability to detect changes in leakage rate over time. When the next outage test is conducted, the results are fed back to refine the model calibration, improving accuracy with each cycle. Book a demo to see the validation methodology on your unit data.
Can iFactory differentiate between leakage increase caused by seals versus basket damage?
Yes. iFactory uses multiple correlated signals to isolate the leakage source. Seal degradation typically produces a gradual, relatively uniform leakage increase that correlates with thermal cycling patterns and load ramp events. Basket damage—such as corrosion holes or broken elements—produces a more sudden leakage increase that may correlate with a specific DP change or temperature pattern shift. Cold end corrosion-related leakage often shows a seasonal pattern correlated with flue gas dew point conditions, while ABS plugging produces a characteristic DP acceleration without immediate leakage increase. The system identifies the most probable degradation mode and recommends the appropriate maintenance action. Contact our operations team for technical details on source isolation.
How does this work with multiple air preheaters in a dual-APH boiler configuration?
iFactory monitors each air preheater independently with its own set of baselines, trends, and health scores. In a dual-APH configuration, the system can detect asymmetric degradation between the two sides—which is common because fuel gas flow distribution is rarely perfectly balanced. Asymmetric degradation often indicates a problem specific to one side such as a sootblower malfunction on one APH, uneven flue gas distribution, or a localized cold end corrosion issue. The fleet view shows both APHs side by side with comparative trending, making it immediately obvious when one side is degrading faster than the other. This capability is particularly valuable for tri-APH configurations in large utility boilers.
What data inputs does iFactory need from our DCS to monitor the air preheater?
The minimum required data set includes air inlet and outlet temperatures, gas inlet and outlet temperatures, gas-side differential pressure across the APH, air-side differential pressure, total air flow, and total gas flow. Additional valuable inputs include individual sootblower status and timing signals, oxygen content at the APH outlet, mill outlet temperatures if the APH supplies primary air, and ambient temperature for baseline correction. If you have online moisture or acid dew point measurements, those can be incorporated for enhanced cold end corrosion detection. Most of these signals are standard DCS points available in any coal-fired plant control system. Book a demo and we will map your available signals to the monitoring requirements.
Can iFactory recommend specific seal adjustments based on the leakage pattern analysis?
Yes. When iFactory detects non-uniform leakage patterns—indicating that rotor warping or localized seal wear is causing higher leakage at specific angular positions—the system generates seal adjustment recommendations that specify which seal sectors need attention and whether radial, axial, or circumferential seals are involved. These recommendations are based on the correlation between leakage behavior and operating conditions such as load level, rotor temperature, and ambient temperature. The actual mechanical adjustment is still performed by your maintenance team during an outage or seal adjustment window, but iFactory provides the data that tells them exactly where to adjust rather than making uniform changes across all seals. This targeted approach typically achieves 30 to 50% more leakage reduction per adjustment cycle compared to uniform repositioning.

See your air preheater leakage trajectory before your next outage

Process engineers at coal-fired plants use iFactory to detect APH degradation continuously instead of discovering it during outage testing. Book a 30-minute walkthrough and see the monitoring analysis on your boiler data.


Share This Story, Choose Your Platform!