Checking up on Chashma9 June 2017
Modern surveillance and monitoring techniques have been employed at Pakistan’s Chashma nuclear power plant for safety and ageing management. By Muhammad Haroon, Zafar Kazmi, Saleem A. Ansari
An extensive surveillance programme has been developed for Chashma, with in-service condition monitoring, along with ageing management of critical plant components, reactor core, internals and coolant pressure boundary. The work is part of the reactor safety enhancement programme, in the Fukushima Response Action Plan.
The systems, structures, and components of a commercial nuclear power plant are subjected to time-dependent ageing degradation during operations due to normal operational wear, mechanical deterioration, corrosion or fatigue. If they are not mitigated, these effects will eventually adversely affect plant safety and performance.
Nuclear power plants strive to improve safety, maintain availability and reduce operation and maintenance costs. At the same time, the nuclear industry is striving to increase the reactor power through power uprates, and to extend plant operating life, which increases the probability of material degradation and structures or component failure due to ageing and increased loads. To operate the plant under optimum conditions without compromising reactor safety, in-service monitoring is essential. Specialised and advanced techniques using high-resolution data processing are used to assess the status of reactor structures, systems and components.
Monitoring at Chashma
There are four PWR plants (three operational and one in the final commissioning stage) at the Chashma site. All are two-loop plants with capacities ranging from 300MWe to 340MWe. The Pakistan Atomic Energy Commission (PAEC) has initiated a comprehensive ageing management and surveillance programme for these reactors. In addition, extensive R&D has been done in the Centre for Vibration Analysis & Condition Monitoring (CVCM).
The surveillance and ageing management programme has five monitoring regimes:
- In-situ monitoring and analysis of the flow-induced vibrations (FIV) of reactor internals, to assess the structural integrity of the core support structure;
- Loose parts monitoring system to prevent degradation of reactor pressure boundary;
- Condition monitoring and fault diagnostics of plants’ critical rotating machines;
- Thermal fatigue monitoring of the pressuriser surge line and other sections of primary piping;
- Surveillance testing of reactor pressure vessel materials.
Some of the above-mentioned techniques have been employed in all reactors, whereas others have been implemented in selected plants. The RPV materials surveillance programme was described in detail in NEI October 2016 pp.36-38 and will not be discussed here.
Codes & standards for surveillance monitoring
Table 1 shows the list of ASME operation & maintenance codes, standards and guides related to plant surveillance and ageing. The reactor core and the core support structure are the most important components of nuclear power plants. As the high-velocity, high-pressure coolant enters the reactor pressure vessel of a PWR through inlet nozzles, it imparts tremendous force on the reactor core barrel, which holds and supports the core and reactor internals. The core barrel is fixed at the top on the ledge of the reactor vessel and pressed down in position by a hold-down spring; at the bottom the barrel is hanging freely.
The forces generated by the fluid-structure interaction produce flow-induced vibrations in the core barrel, and it oscillates at its natural frequency like a cantilever beam. This results in dynamic displacement of the reactor core and core support structure.
The perpetual core barrell vibration leads to high-cycle fatigue of the core support structure and can cause degradation in the structural integrity of the reactor internals. It is therefore necessary to measure in-situ the vibrations of core barrel and the integrity of the core support structure throughout the plant life, and international standards have been devised for such monitoring. Such monitoring is also very valuable for the plant life-extension programme.
For an effective in-core surveillance programme, the CVCM developed the Reactor Internals Vibration Monitoring System (IVMS) to monitor the dynamic displacement of the core barrel motion (swing) due to flow-induced vibrations. The system also computes the mode shapes and modal frequencies of vibrations in reactor core components.
Design and installation of IVMS
The functional diagram of the IVMS is shown in Figure 1. The methodology, hardware and software of the IVMS meet the requirements of ASME-OM-S/G-2013, Part 5 guide and the US Nuclear Regulatory Commission (NRC) Regulatory Guide 1.20 (2006). The in-core surveillance system IVMS was installed at Chashma 1, which is a 300MWe PWR.
The signals from all four ex-core neutron flux detectors were interfaced with IVMS. Double signal isolation is provided in the signal protection module of IVMS to protect the class 1E signals of plant safety channels. The signal conditioning module provides frequency filtering and signal amplification of the neutron flux signals before interfacing with the high-resolution analogue-to-digital converter at the input to the IVMS.
The IVMS is capable of performing the following surveillance functions:
- Collection of reactor noise signatures during plant operation;
- Determination of possible contact between core barrel and pressure vessel (detection of free-swing and swing-with-contact modes.)
- Detection of relaxation or reduction in strength of the hold-down spring by monitoring the shift in beam mode frequency with core lifetime.
- Determination of natural frequencies of shell and beam modes of core barrel motion.
- Detection of fuel assembly vibration. (The first two modes of vibration of the fuel assemblies closest to the neutron detectors.)
There are three distinct functional phases of IVMS utilisation: baseline; surveillance for detection of deviations from acceptable values; and diagnostics.
Measurement of vibrations in reactor core support structure
The power spectral density (PSD) and phase plots of the fluctuations in neutron flux signals obtained from A1 & B1 channels of power range instrumentation for Chashma 1 were obtained.
The PSD plot shows a predominant peak in the frequency range 4.0Hz-5.5Hz. High signal coherence (~ 50%) existed between detector pair A1-B1 in this frequency range. At this frequency the phase difference between the A1 and B1 ex-core detectors is ~1800. All these analyses point to a beam-mode vibration of the reactor core barrel at 4.2Hz, induced by the interaction of coolant flow with core support structure.
With the help of reactor core surveillance measurements using IVMS, the damping factor and the stiffness of the hold-down spring were determined. Any reduction in the holding strength and stiffness of the spring would result in a reduction in the frequency of vibration of the core barrel from the present value. Also, the barrel displacement is expected to increase. These ageing-induced mechanisms would manifest in the neutron noise analysis and PSD plots obtained from IVMS.
Calculation of flow-induced vibrations
The measurements of vibrations in reactor internals were verified with a rigorous flow-induced vibration (FIV) analysis programme. A detailed algorithm was developed for FIV analysis, combining four major analyses: structural modal analysis; computational fluid dynamic analysis; power spectral density calculation of fluid forces; and dynamic response analysis of structure to the gross fluid force. The power spectral density of random turbulence force was calculated using the pressure fluctuation model of Wambganss, Chen and Paidoussis. The modal frequencies, displacement, loads and stresses at key locations of reactor internals’ structure due to random turbulence force were calculated. These values are compared with the measurement results in Table 2. There was close agreement between calculation and measurement.
Loose parts monitoring system
A loose parts monitoring system (LPMS) is an essential device for integrity monitoring of nuclear steam supply system in PWRs to protect against catastrophic failures in the reactor. CVCM has developed design and analytical software of an indigenous LPMS for detection of loose parts at the reactor pressure boundary and determination of their mass and energy. The LPMS methodology is quite complex, since the acoustic waves produced by metal-metal impact of loose parts are emitted at high frequencies up to 20kHz, and they travel through the reactor structure in a complex pattern at speeds up to 5km/s. The indigenous LPMS design employs piezoelectric accelerometers to detect acoustic bursts. Specialised LPMS software has been developed for ultra-fast data acquisition and processing of multiple burst signals. The LPMS design, analytical software and criteria for sensor locations are compatible with the requirements of international standards.
Loose parts localisation (r,θ) is determined by employing circle intersection technique. Loose parts mass and energy is determined with the help of the Hertz Impact Theory, Lamb Diagram of S0 & A0 wave mode analysis, waveform propagation/attenuation analysis, and frequency spectrum analysis.
The LPMS was extensively tested in rigs and coolant loops before installation in the reactor coolant system at Chashma 1. More than a thousand acoustic bursts generated by loose parts impacts were analysed. The results show that the software can locate loose parts with an accuracy of 20cm over a 2m distance.
The LPMS design employs 16 high-temperature piezoelectric accelerometers for detection of acoustic bursts generated by the impact of loose parts. The signal conditioning hardware performs charge-to-voltage conversion, amplification, frequency filtering, and video/audio alarm generation. LPMS software controls ultra- fast data acquisition and processing of multiple burst signals. The software makes use of special functions of digital filtering, and time & frequency domain analyses of burst signals.
The indigenised LPMS and sensors were retro-fitted at Chashma 1 during the RFO-6 fuelling outage in 2010. A total of 16 accelerometers were installed on critical components of the pressure vessel and other main steam system equipment (see Table 3). The accelerometers were mounted using specially designed clamps and fixtures. To mount the clamps and accelerometers piping insulation was removed so that the clamps were flush with the piping surface. Due to the very high radiation dose rates and high temperature in the area, extreme care and careful planning was made to avoid over- exposure to working personnel.
Thermal fatigue monitoring
Thermal fatigue monitoring (TFMS) of piping associated with a reactor coolant circuit is an important regulatory requirement for the safe operation of a nuclear power plant. To achieve this objective at Chashma, the TFMS was designed for on-line temperature monitoring of different layers of coolant flow along the circumferential sections of piping. The TFMS measured the temperature difference (ΔT) between the hot and cold layers and calculated the cumulative usage factor as well as the fatigue life. The system was successfully installed on three reactors. In-situ testing, calibration and fine-tuning of the system were made at the plant with the help of IR thermometer and direct contact thermocouples.
The TFMS design (Figure 3) conforms to the EPRI Guides MRP-25, MRP-29 and MRP-32 and ASME Section III. The system employs 58 K-type (NiCr-NiAl) thermocouples for temperature measurement. The hardware performs data acquisition, signal amplification, voltage to temperature conversion, open circuit detection and cold junction compensation of a maximum of 64 thermocouples.
Thermal fatigue is monitored at the reactor pressure boundary in all four units at Chashma. The typical locations of the thermocouples of TFMS are shown in Table 4. The thermocouples were not installed directly on the piping surface, to maintain the integrity of the safety piping. Since drilling or welding on the reactor pressure boundary was not possible, an innovative installation design based on specially designed clamps was adopted.
Steam turbine condition monitoring
CVCM has evolved a comprehensive plan to fulfill the mandatory requirements of vibration monitoring in components of the nuclear steam supply system. An on-line turbine fault diagnostic system (TFDS) has been designed for advanced condition monitoring of steam turbines, with particular emphasis on vibration analysis of turbine blades.
The diagnostic system employs existing relative shaft vibration sensors (proximity probes) in the turbine protection system (see diagram). Additional absolute vibration sensors (accelerometers) are mounted on the bearing casings for high-frequency vibration analysis. The system can be used for detection of turbine blade vibrations, blade cracks and identification of resonance conditions.
The system design and methodology was validated at the Chashma 1 low-pressure steam turbine. It made it possible to determine the root cause of the erratic vibrations at bearings thee, six and seven of one of the steam turbines, observed during one fuel cycle. The excitation of turbine blades were monitored during normal operation, coast-down and run-up. It was confirmed that there was no resonance excitation of the blades’ natural frequencies that could lead to destructive vibrations. The vibrations at blade-pass frequency had relatively small magnitude with weak side bands. Based on detailed vibration analysis & diagnostics it was concluded that the turbine was fit for continuous operation up to the next refuelling outage.