For existing and new plants to meet the requirements of tomorrow’s networks, their maintenance and operation must be realigned. Hans Christian Schröder and Achim Foos of TÜV SÜD Industrie Service GmbH explain what this will mean in practice.

Can there be a more exciting task than keeping existing large-scale fossil-fuelled power stations fit for future energy grids? As networks will increasingly include higher shares of intermittent renewables, new and existing power plants will have to operate more flexibly, with many changes to load, to ensure a secure power supply.

Power stations will run more often in load ranges for which they were not designed. Improvements can be made to plants to make this possible, but stakeholders tend to overlook just what this involves.

Strategies must be designed that ensure an optimum mode of operation, and associated maintenance programmes must be aligned. Economic considerations must also include the costs of the technical governance implemented to optimise the mode of operation and lifecycle costs.

So what will the future energy mix look like? Transmission network operators will increasingly have to control power plants to keep the grid stable. Closed-down plants will have to be reconnected to provide reserve capacity. What will be productive for future grid strategies is to ensure that base, medium and peak loads match demand? The focus will not be primarily on the thermal efficiency of plants but on their capacity utilisation.


Old-style power stations have already been amortised and often offer generous design margins that can accommodate variations in the mode of operation. There is still a lot of untapped potential to make existing power stations that run at base, medium and peak load more flexible – lowering minimum loads, improving the speed of load changes and improving startup and shutdown operations.

Further options include using different types of coal mills, reducing minimum loads without assistant fuel, changing load management and maximising use up to the design margins.

Another possibility is to raise the live steam temperature to within 1 K (272oC) of the design limit rather than the relevant standard’s 5 K requirement. Optimised sensors would help here – provided the experts can ensure that there have been no peculiar changes in the microstructure of the material.

The sensors must also permit measurement of the actual service parameters to ensure that service temperature can be reduced should changes in the material’s microstructure occur.

Special demands

Flexible grid-focused operation makes special demands on plants. The original designs and service parameters of plants will have to be reconsidered, and maintenance strategies aligned. As the new flexibility will reduce the estimated service life of components, accurate information will be needed about a plant’s current condition.

Experts must ask which incidents in the plant may cause special defects, and what assessments and measures are needed to allow reliable operation of the plant.

These questions mainly concern plant components. Some are vulnerable to unscheduled startups and shutdowns, and stresses caused by vibration, overloads or changes in the mode of operation.

The solution is a comprehensive analysis of the total system and its current operating conditions, based on years of experience. Such an analysis can also identify specific events to ensure that suitable inspection and calculation methods will be put to best use on selected components in the future. This prevents the need for cost-intensive general diagnosis. Isolated tests and inspections and individual random measures without reference to the total system are not very helpful, especially for plants with long service lives (Figure 1).

Figure 1
Figure 1

So qualified statements on the condition of a plant and its components depend on understanding its real mode of operation, knowing which defects have been detected in the system, and being aware of the condition identified in system analysis, while accounting for possible unidentified defects.

The only solution to determining whether a plant is operable is through an integrated system analysis that considers all possible acceptable faults or defects, or the fault status (Figure 2).

Figure 2
Figure 2

One fundamental criterion for such an analysis is to link the operational loads with their effects on individual areas. For this purpose, plant experts perform inspections and tests. Statements can then be made about the continued operation of a plant under changed framework conditions, and future service scenarios can be developed.

Process control

Improving flexibility can be a good starting-point for implementing retrofit measures at a plant or a site. However, greater flexibility requires many things to be considered, especially in low-load ranges. For coal-fired plants, ways must be found to reduce the required assistant fuel, and cut emissions through further primary measures. Modernised process control systems can play a big part.

More complex process control systems offer the opportunity for timely, integrated and event-focused monitoring of processes and procedures. Improved safety is ensured not least through the definition of uniform safety integrity levels for a plant. For cost reasons, experts are also considering centralising process control systems, to save manpower and thus costs. However, adapting control systems to the new operating conditions will not be enough on its own. There must also be a change in maintenance efforts, particularly at the periphery of the system.

Pooling operations, and monitoring and detecting events in individual plant areas becomes more difficult, which increases operational risks. The optimum registration of hazards and risks so far identified by staff on site may no longer be ensured. Additional technological and automated measures may become necessary.

Hardware in the system’s periphery would have to be modernised, necessary sensors installed and I&C systems added to provide for on-line transmission of their signals and information to control centres.

Whether existing standard requirements and available approvals cover these measures is under debate. The term ‘monitored operation’ has been reliably defined in the presently applicable codes, and must be differentiated from the term ‘permanently monitored operation’. The following question must be asked: is it possible to reliably determine potential ‘faults’, and the risks that they involve in terms of response time, in order to describe possible effects and risks in the system’s periphery?

Another issue concerns existing process control systems that have been operated for long periods, or for which suppliers will discontinue support.

Can these systems continue to operate against the backdrop of regulations such as EU Directives?

Can old equipment be used, and how can plant owners and operators evaluate whether third-party suppliers are acceptable to carry out repairs and provide services?

Do devices and systems already in operation require new certificates or do they fall under the grandfathering principle?

Standard EN 61511 provides a good basis for assessing this type of modernisation and for making necessary adjustments. It is applicable to the process industry, and defines quantitative or qualitative safety objectives for electrical, electronic and programmable electronic systems (E/E/PESs) in the form of safety integrity levels (SILs). The focus in power plants should be on assessing the overall function of the system, so the application of the standard requires the assessment and implementation of requirements related to procedure and process within a system context.

The process descriptions included in the standard make it imperative to adopt an integrated approach that covers all requirements related to procedures and safety in a plant. The standard also focuses on cost-efficient plant operation, which is intended to prevent over-engineering.

Further questions within the scope of such changes concern the necessary requirements, including SILs and performance levels (PLs) for proven-in-use existing plants and systems. A pragmatic approach and willingness on the part of all stakeholders to take on more responsibility are essential in maintenance.


The efficient operation of power stations with better flexibility must be ensured throughout their lifecycles. Here, lifecycle costs (LCCs) are among the key elements of the plant’s total cost of ownership. Reliable determination and optimisation of these is crucial to cost planning. This particularly covers warranty costs and calculations of subsequent service contracts or replacement investments.

Even though engineering and business requirements imposed on the conceptual design of a plant may differ when seen in terms of the entire service life, they must still be harmonised well. Decision-makers in business and engineering must co-operate from as early as the assessment process.

Assessment starts by analysing how much total energy in a plant can be cost-effectively converted into useable form. This requires an integrated system approach, optimised inspection schemes and collection of additional data, including measuring the quality of new components. The results permit an informative assessment to be made of the efficiency with which the set targets can be attained under the planned measures.

To ensure profitable energy generation, the procedural and operational concept must be aligned to the individual technical processes of the more flexible plant operation. This requires strategic integrated maintenance, which is regarded as value-adding rather than as a cost factor.

Strategic maintenance

Vital for flexible power station solutions is an integrated maintenance strategy, which only considers as relevant those measures that improve availability, reliability and safety, so the engineer’s striving for perfection is complemented by an entrepreneur’s business perspective.

In step one of four, strategic maintenance sees consultants, the client and the operator working together to identify what has to be assessed. Their focus is on components that will be most affected by the flexibility. Synthesis then links the strain imposed on the plant with its actual condition. This produces an overall picture of the system that presents the current operability of the plant and the derived risks. Optimum operation can only be ensured when process, plant, operational and control solutions are interconnected.

Step two involves fracture mechanics testing and components analysis. Knowing the interactions between materials, loads and fault status is critical to ensure safety and maximum availability.

In step three, probabilistic methods assess the acceptability of faults and risks by assessing the results of non-destructive or other test methods. First, there is only a certain probability that faults that are basically identifiable will be detected. Second, once a fault has been identified, the resulting risk and its consequences for the operation of the component must be responded to.

The first aspect can be addressed by applying known methods of mathematical statistics and probabilistic models. The second requires a more differentiated approach. The plant or component can be taken out of service and repaired, or operation can be continued at the same or reduced load until the next scheduled turnaround, until a defined limit risk is reached.

To weigh all these options, all risks and costs must be determined and their interrelationships determined. Here the focus is on the safe and profitable continued operation of the overall system. If component failure involves little risk, the above issue is reduced to its economic aspects. Otherwise the two aspects cannot be regarded as separate, and may lead to conflicts of interest.

In each case, quantitative assessment criteria, such as acceptable risks, must be available. The risk resulting from the failure of a component is the frequency of occurrence of an event multiplied by its possible consequences (Figures 3 and 4).

Figure 3
Figure 3
Figure 4
Figure 4

In step four, experts develop measures to ensure flexible plant management, which is largely influenced by the type, structure, function and condition of a plant. Assessment is restricted to non-conformities that can result in operational or process failure.

The objective is to supply operators, investors or manufacturers with the information they need for decision-making, including cost-benefit analyses, computer simulations and forecasts, and tailored proposals for maintenance strategies.

Limits of maintenance

For existing power plants, maintenance tools have been tried and tested over the years. But recent years have seen a trend towards incident-oriented maintenance. Component failure can be due to the end of service life or to obsolete practices.

Spontaneous failure of old components can be due to failure to make use of probabilistic methods to determine possible failure scenarios and calculate the expected service life.

Plausibility checks – such as in a piping system – are important, to assess geometrical data such as wall thickness. This calls for visual inspection and verification by measurement. Maintenance staff may wrongly assume that degradations occur primarily on complex geometries such as fittings.

Figure 5 illustrates the initial assessment of a straight pipe section, showing noticeable differences in wall thickness depending on pipe circumference. This example shows that different localised reductions in service life must be anticipated depending on the nominal wall thickness on which the calculation of service life has been based. A power station in which a straight section of a pipe made from 14 MoV 6 3 failed spontaneously after around 30 years of operation provides a dramatic example.

Experts found that the failure was caused by a unique feature of the material. Our analyses revealed the same differences in wall thickness as those in Figure 5. Component failure occurred at the end of the expected service life, which amounted to roughly 200,000 hours.

Figure 5
Figure 5

Efficiency and flexibility in energy generation will come with an optimum plant scheme and an aligned operational strategy. An integrated profitability analysis done in the run-up, and a maintenance programme customised to specific plants make good economic sense. This also includes optimum planning of maintenance costs, which includes possible investments with foresight.

It may be necessary to whether plants can be operated with permissible imperfections. This, requires a high level of know-how to assess risks and realise safe, profitable operation while ensuring availability.

Hans Christian Schröder is head of Power Station and Energy Services and Achim Foos is part of his team at TÜV SÜD Industrie Service GmbH, Mannheim, Germany. For more information, visit

More Power Engineering International Issue Articles
Power Engineering International Archives
View Power Generation Articles on