1.

INTRODUCTION

The energy technology landscape is in a state of flux. There are those such as Rifkin¹, who wrote about the 3^rd Industrial Revolution, with five pillars in solar, wind, hydro, geothermal, waves, and biomass, who see a revolutionary transformation in the technologies employed to meet energy needs. Others are more conservative and see only a much slower evolution and an energy landscape which will maintain or transition with much of what is currently used, including natural gas and clean-coal, and to which can potentially be added new nuclear power for carbon free base-load electricity generation using small modular reactors (SMR”s). The situation is further complicated in the US by advanced oil/gas extraction technologies, commonly known as fracking, and global oil price drops, including the effects of the lower price relative to the production cost for some higher cost extraction technologies. In the energy landscape in the shorter and medium term there are a number of common threads which appear to emerge² and these do include some increased use of gas as a cleaner source of energy and power, but there are growing concerns about price instability and security of supplies, particularly in Europe. There is expected to be rapid development of at least some renewable energy sources, but market penetration remains unclear and in some scenarios there is renewed interest in nuclear power due to the long-term resource availability and the CO2-free nature of this power generation. There is expected to be increased investment in the development of clean coal technologies, including carbon dioxide capture and storage, designed to use the huge global resources of coal more cleanly. As new renewable technologies are deployed they will require new distributed power plants, new energy storage technologies, including potentially using hydrogen, leverage of the internet and symbiotic technologies, all potentially combined with a transition of transportation to electric and fuel cells. Such an industrial revolution would return (some) high technology manufacturing to the USA, include new (high) technology, new materials, new manufacturing process, new material fabrication, challenges from new stressor effects on systems, different degradation and aging phenomena, new needs for materials state awareness and life prediction, all of which bring opportunities and challenges in a globalized industry for NDT and research in advanced monitoring/diagnostics, and prognostics.

Under the new energy economy the Nation (USA) is faced with an enormous task if it is to make the least painful transition from its current energy sources (chiefly coal, petroleum and natural gas) to ones more abundant, and ultimately to those that are inexhaustible. This transition will require an unprecedentedly large and rapid shift to different, if not new, technologies. In this transition, the role of nondestructive evaluation (NDE) is certainly clear to its practitioners, but unappreciated or ignored by almost everyone else. It will not be possible to rely on accumulated wisdom to predict the safety, reliability and predicted lifetime of the new components that must be developed. Yet such knowledge can have a major effect on the cost of the new energy systems as they are deployed industrially³. This vision and need is not new: the statement was first made in 1978 by James Kane, then Associate Director, Basic Energy Sciences, U.S. Department of Energy. It remains equally, if not more true today as the global community seeks to provide sustainable energy and looks to move to low or no carbon energy sources.

As the energy economy goes through a transition one must ask what are the needs and opportunities for NDT/NDE, structural health monitoring and prognostics? A major challenge is found in the economics of operations and maintenance, including NDT, particularly when applied to renewable energy technologies. For example the wind energy operations and maintenance (O&M) costs are found to easily have an average share over the lifetime of the turbine of approximately 20%-25% of total levelized cost per kWh⁴. The deployment of monitoring/inspection is logistically clearly very different and can be more expensive, and more challenging, than when compared with those used for a large baseload generating facility. Increasingly NDT becomes part of a wider plant condition based maintenance (CBM) or structural health management (SHM) program. In looking at the technologies encountered in energy systems there are clearly some common elements which can be inspected using established NDT technologies, or variants of them: e.g. for pressure vessels, pipes and concrete^5,6. For some energy technologies, such as solar, there are specific materials, fabrication scales and geometries that are more closely related to challenges faced in the semi-conductor industry than for most current energy or at least electricity systems⁷.

There is a significant history which considers the development and application of NDT/NDE and prognostics to a wide range of the more energy technologies [e.g.,^8,9]. There is also extensive literature that considers and documents typical damage mechanisms. This paper considers aspects of materials, defects and allowables (section 2), the evolution of NDT/NDE (section 3) and then discusses some examples of advanced NDE for energy systems (section 4).

2.

MATERIALS, DEFECTS AND ALLOWABLES

For effective inspections it is necessary to understand the basic properties of materials, particularly in terms of the “fundamental interactions of the energy (i.e. ultrasound, electromagnetic field, thermal etc.)” used to nondestructively examine a medium. It is then necessary to understand the type, shape, size and significance of defects, including what features/anomalies are considered to be allowable at the time of fabrication, and where an allowable anomaly is a local non-uniformity that is insignificant in terms of its impact on performance. In materials assessments the determination of fitness for service or adequacy of quality requires understanding of the influence of damage on structural allowables (Figure 1) and quantification of “manufacturing acceptable features” with criteria for acceptable “anomalies.” Such assessments are becoming increasingly both important and challenging with new materials, e.g. composites, and new manufacturing, e.g. additive manufacturing. In an item it is typically not just bulk or surface damage and degradation, but assessments can include quality of joints, micro-cracks and pores and well as a diverse range of in-service phenomena, including impacts. In assessing any NDT technologies capabilities it is then the measurement of its performance, such as with a probability of detection (POD) for a discrete defect, and an assessment of material state or signature, such as grain size/shape in bulk materials, which is needed. However, the nondestructive determination of such signatures and mechanical properties are considered to be a holy grail in NDE¹⁰. Desired properties can include complex effects, such as combinations of texture, applied stress and temperature.

Figure 1

Concept of allowables for a composit material and imperfections effect in reducing design space.

There is a merging of what was considered to be materials characterization and more tradition NDE, particularly in the context of material state awareness and prognostics, in ideas that were formulated by Thompson¹¹. Traditional nondestructive evaluation (NDE) was most commonly applied to metals and it was a matter of detection of an anomaly, followed by determination of a size and location. Nondestructive materials characterization is seeking to characterize much more including microstructure (stress, strain, grain-size, moduli, fatigue, fracture toughness) and to evaluate/determine local maps of material properties prior to and after the formation of discernable flaw. The move to provide a more complete materials characterization is important to assessing infrastructure reliability and safety. This is because in some cases the first crack can be catastrophic, due to characteristics such as a zero tearing modulus. Identification of parts of a structure with anomalous or high load is required to guide more focused nondestructive testing, and SHM technologies, to regions with a high propensity to fail. Monitoring such areas can give data to provide an early warning of potential for loss of structural integrity prior to flaw formation and enable earlier and accurate lifetime prediction.

The detailed understanding of microstructure must be a key ingredient in development of state awareness strategies. The connection between NDE measurements and structural performance is then a goal and this is shown in schematic form as Figure 2. In progressing through the process to predict structural performance based on NDE data, each link has it challenges. There are issues on non-uniqueness in the inversion or linking a measured parameter to performance relationships. There can be inadequate sensitivity of NDE tools to key parameters, particularly with regard to measurement scales and early damage, which are all compounded by limitations of the theoretical base for such linkages. There are moves to use of novel NDE technologies (e.g. backscatter ultrasound) for microstructural characterization tools as well as for flaw detection. Adequate sensitivity to early damage or change is critical with the need to be able to assess the progression of damage before cracks form. Tools are needed for quantification of the initial state and to check for evolution of damage when possible, and to provide validation of prognostic calls. These limitations and challenges tend to force the use of a stochastic approach.

Figure 2

Schematic showing progression from NDE measurements to structural performance.

For in-service NDE it is critical to understand the stressors and how these relate to degradation mechanisms. In general terms aging degradation mechanisms are classified into: (a) Internal, that is changes to microstructure or chemical composition, a change in intrinsic properties (thermal aging, creep, irradiation damage, etc.), and (b) Imposed, physical damage on the component, which can be in the form of metal loss (corrosion, wear) or cracking or deformation (stress-corrosion, deformation, cracking).

In many cases a flaw is only detectable by traditional NDE relatively late in a components life. If there are high loads and risk of rapid growth of defects to failure, infrequent inspections can cause significant defects to be undetected. With distributed energy technologies, such as a wind turbine farm, where there can be many dozens, if not hundreds of turbines spread out of a large geographic area, limited numbers of inspections, particularly with items of relatively low capital value, is causing attention to move to embedded structural health monitoring. The phenomenon of aging degradation in materials are complex. They require sophisticated, state of the art science and technology procedures to effectively manage it and ensure safe and reliable operation. In this process it is not only technology which is involved, but an effective management system is needed in order to correctly implement mitigation or monitoring actions.

For aging management the necessary codes, standards and best practices are reasonably well established in high technology industries, such a nuclear power, where in the USA the Nuclear Regulatory Commission provides oversight and guidance. Nuclear power plants are considering prognostics and health management for both active and passive components, including on-line monitoring¹². In some energy technologies such as for oil/gas the American Petroleum Institute (API) provides standards and recommended practice. For example a comprehensive review of “Damag mechanisms Affecting Fixed equipment in the Refining Industry,” including suggested inspection approaches is given in API RP 571¹³. Both inspection and repair for in-service piping systems are also addressed^14,15.

In looking at renewable energy systems there are NDT challenges in providing technically adequate and economic testing methods for turbine blades, turbine towers and condition monitoring of gearboxes, rotor hubs and electrical equipment. Offshore wind, in the USA, wave and tidal power are all at an earlier stage of development. Providing inspection and monitoring for achievement of satisfactory availability will undoubtedly be a major challenge since repairs in the hostile marine environment will be exceptionally difficult.

3.

EVOLUTION OF ADVANCED NONDESTRUCTIVE TESTING AND EVALUATION

Nondestructive testing/evaluation (NDT/NDE) in its industrial implementation, is a mature technology, guided in large part by codes and standards. The test equipment market was estimated to be over $1.39B in 2011 and to reach $2.03B by 2016¹⁶. The costs and added value of inspection/quality services is clearly many times that for the equipment market, but reliable estimates are not generally agreed upon. It is however clear that the applications of NDT/NDE are growing to include; aging infrastructure, new materials and new manufacturing processes, together with the growing global equipment market. The range of technologies being deployed in NDT/NDE and SHM is also evolving.

The science and technology that is now employed in NDE, advanced quality assurance and component life management has been developed over a period of more than 40 years⁸. In the 1970s as more advanced technology systems were designed for use in “high-risk” technologies such as nuclear power and advanced aerospace, it was recognized that there was a need to better understand the effects of increasingly severe and hostile environments on materials and the significance of defects, in terms of potential for failure. A science base for the theory and measurement of equipment aging, including the use of accelerated aging programs, was established. In addition, it was seen that the capabilities of then available nondestructive testing (NDT) were limited, and there was a lack of an adequate science base for NDT to become a quantitative science. It was necessary to improve the reliability of inspection and relate size and types of defects to the effect that they have on performance or potential for loss of structural integrity. Several major programs were initiated to provide the required science base, including one sponsored by the United States Air Force-Defense Advanced Research Project Agency (USAF-DARPA), which considered the development of quantitative nondestructive evaluation (QNDE) to meet the needs of the aerospace community. The integration of materials, defects and inspection was achieved through the advent of fracture mechanics, which was greatly enhanced through the ever-improving capabilities of finite element analysis, which was in turn largely facilitated by the availability of ever-more-powerful computer systems. The philosophies of damage tolerance and retirement-for-cause were developed and applied in the 1970s and early 1980s for application to critical aircraft engine components, at all phases of the life-cycle design, manufacture and maintenance. At the same time, other groups of engineers and scientists were considering equally challenging problems of ensuring structural integrity in the nuclear power industry and in the oil and gas industries, in particular, for structures in the North Sea. During the 1970s and ’80s great progress was made in both materials science and QNDE in terms of providing a greatly enhanced science base for NDE, new sensors, instrumentation and data analysis tools for application at both the time of manufacture and during periodic inspection of some types of items in service.

The initial focus of much of the research within this emerging community was on metals. This is now expanding into advanced composites and ceramics, and being driven further with new fabrication approaches such as with additive manufacturing. The range of fields of application has also now expanded into civil engineering. Novel integrating design approaches such as unified life-cycle engineering (ULCE) were proposed and partially applied in various forms of concurrent engineering. The full power and potential of this approach was limited by then available materials science and in particular, the computation power needed to perform many of the design optimizations at a reasonable cost and within a reasonable time. UCLE was envisioned as a family of computer-based tools for complete life-cycle analysis that would enable optimizations and trade-offs to be performed at the time of initial design development. In the 1980s and early 1990s, it was increasingly recognized that structural assessment, including quantification and evaluation of defect and defect populations was not all that was required to evaluate the remaining safe-life for complex systems. It was necessary to identify and characterize discrete defects, cracks and corrosion, and determine a rate of growth, investigate the probability of occurrence and probability of detection (POD). It was also becoming clear that there was a need to provide measurements of materials state and changes in bulk material properties caused by the aging of the materials and the accumulation of “damage,” due to stressors.

The development of the science for damage mechanics and tools to quantify the properties of critical structures became a priority. Studies considered methods for the combination of damage and fracture mechanics, where the effects of damage are seen in micro-cracks and other physical-chemical changes, “short-crack” growth phenomena occur and macro-cracks described with linear elastic fracture mechanics interact under the influence of a multitude of both physical and chemical environmental factors. The complexity of the phenomena is further increased by inclusion of consideration of “random acts,” impacts, explosions and other short duration transient events, as well as longer term daily and seasonal thermal and chemical loading operational cycles.

What was initially called predictive engineering⁸ and was subsequently called “prognostics” was first applied to active components in CBM and is now being considered for passive structures subject to aging and potential life extension¹². It is becoming a family of models, science and measurement technology tools, for use in the integrated methodology, which seeks to provide the framework needed to assess a system, predict a remaining safe-life, formulate and test strategies for mitigation and minimization of the rate of degradation.

4.

EXAMPLES OF ADVANCED NDE FOR ENERGY SYSTEMS

In parallel with the development of new energy systems, particularly with regard to renewable energy there has been the development of advanced NDE techniques/tools which move beyond those currently employed and referenced in codes and standards. NDE was traditionally viewed as a quality assurance tool, a check on workmanship, an added cost, necessary to ensure fitness for service, with development being a drive to push for detection of small flaws (as stresses increase) and to give data to enable life extension (retirement for cause). Newer model-based approaches to NDE are providing simulations for inspection optimization, data interpretation, training, and are linked to materials characterization studies where understanding of physics enables interpretation of results. As technologies and applications emerge there are opportunities with guidance in the design of parts and inspection methods and interpretation of POD, including formal modal assisted probability of detection (MAPOD) and wider ranging performance reliability studies.

Some examples of the state of NDE in different energy industries are considered.

4.6

Large Scale Composites, Pipes and Concrete

In looking at the full spectrum of renewable technologies the interest in tidal and geothermal is growing and attention is starting to address the NDT needs. For geothermal systems, much is comprised of carbon and low alloy steel. Specific needs for improved instrumentation and NDT were identified by Allan⁴⁸ and more recently reviewed by Berndt⁴⁹. The NDT technologies are conventional, although the corrosion and specific applications are custom to geothermal, there is much in common with the needs and technologies employed in the petrochemical community [e.g.¹³]. For at least some tidal power applications large composite hydrofoils are being considered. In many ways these “blades” are not totally dissimilar to those used in wind turbines, and a number of the inspection methodologies used in construction and for monitoring can be employed. Some recent work is considering the use of guided waves to monitor critical regions⁵⁰.

Concrete is almost ubiquitous to civil engineering structures. For power systems it is a key element in dams, in nuclear, wind turbine base structures, and now also towers and in almost every other form of power plant. NDT for concrete using ultrasonic pulse velocity (UPV) dates back to the availability of relatively inexpensive instruments developed in the 1970’s. A variety of other technologies have followed and these have been extensively discussed in the literature^51,52. Most recently an assessment of the state of the art for concrete structures has been provided by the German, BAM⁶. This review targeted technologies for potential application in nuclear power plant structures, but it is much wider ranging. It includes tools for measurement of the covering and location of steel reinforcement, tendons and ducts, together with assessment of grout condition, and then methods for detecting cracking, voids, delamination and honeycombing in concrete structures. A final section addresses detection of inclusions as well as specific issues relating to assessment of materials condition next to containment liner, including corrosion. Table 4 presents a summary of testing methods for defect detection and degradation assessment in concrete structures.

Table 4

Summary of testing methods for defect detection in concrete structures.

5.

CONCLUSIONS

NDT/NDE for energy system applications can be expected to grow in its application: drivers include quality, safety, sustainability and cost of ownership. There will be better and more effective deployment of “Traditional and Advanced NDT” as more diverse energy systems are developed and deployed, new applications of current technologies will be developed. In addition new and emerging capabilities, such as those for solar cells and panels, will be needed. Some general challenges will include handling large data, providing permanent records for inspections, developing and utilizing better equipment, all in the context of increasingly employing automation and robotics. There can be expected to be some integration into manufacturing process monitoring measurement and control, and measurements of material state and mechanical properties, where full characterization of materials will merge activities into those traditional considered to be materials science. There is a trend to increasingly include an NDE or design for testability activity that includes increasing use of models in product development.

In looking across NDE the research and development problems are getting harder, due to design, manufacturing and materials combinations. NDT/NDE is increasingly seen as part of CBM/Prognostics and the total quality manufacturing process. To provide the information desired there is a general need for NDE to be more quantitative and more sensitive, such as for early degradation detection. These trends are requiring new sensors, and integration into manufacturing metrology. There is also a trend to move from NDT/NDE and structural health monitoring (SHM) to true prognostics, that provides remaining/safe life estimation at the system level. These NDT will be used to minimize ownership costs, but challenges will remain in terms of providing economic and effective NDE for decentralized energy generation (e.g. wind turbines) and systems that will be deployed in harsh environments (e.g. off-shore wind, tidal and wave systems). Fully Integration of NDE into engineering and product life cycle – design for inspectability and monitoring. Quantification of uncertainty – (ill-posed problems). Physics has not changed!! Heat, light and sound! (thermography, electromagnetics, ultrasound)