Nuclear power plants in France and Germany operate in a load-following mode to help stabilise the electricity grid on a minute-by-minute basis, and to balance daily and weekly shifts in supply and demand. In Germany, load-following has become important in recent years with the introduction of intermittent sources of electricity generation (for example, wind).
Generally speaking there are four operating modes currently used by the nuclear power plants in France: base-load generation mode (constant power), primary and secondary frequency control (grid balancing), and load-following.
In Figure 1 the history of nuclear generation in France in 2010 is presented, as well as its variation of the nuclear generation G (as percentage of the average nuclear generation for that day). It is defined as: (max G-min G)/Average. With this definition, the daily variation of the nuclear generation is typically less than 5-10% of the total nuclear generation in France. The average daily variation of nuclear generation in 2010 was about 6.7%. However, for some periods, the daily variation could be more than 20%. During warm periods of the year, nuclear generation is lower (because the temperature of the cooling water is higher), and thus the daily variation is also higher.
In base-load mode, nuclear power plants operate at constant power (usually at maximum rated power Pr) during almost the whole cycle. The first French NPPs of 900 MWe initially operated in this mode. Later, they were upgraded to improve their manoeuvrability capabilities. Still, today, a significant part of French nuclear power plants operate in the base-load mode.
Frequency control
Power demand can never be exactly evaluated in advance and thus there is a certain random variation of demand resulting in frequency fluctuations, typically of less than 20 mHz. Power plants have to monitor the frequency on the grid and immediately adapt their level of generation in order to keep the frequency stable at the desired value (primary control).
The variation of the frequency Δf would require a change of the power of the plant of:
where f0 is the target frequency (for example 50 Hz in France), P0 is the power level of the plant (as % of the rated power Pr) and S is the droop measured in %. Droop is the ratio of the steady state change of speed or in frequency to the steady state change in power output; in France the droop is close to 4% (for nuclear and thermal power stations). Thus, one has in France k≈50% Pr/Hz, and this means that if the frequency changes by Δf≈20 mHz, the power of the NPP would have to change by 1% Pr. The power modulations for the frequency regulation are performed in the interval of +/-2% Pr.
Primary frequency control allows short-term adjustment of electricity production and demand in the time frame of about 2 to 30 seconds after the deviation is observed. Another type of frequency regulation, secondary control, acts over longer timeframes (from several seconds to several minutes) and restores the exact frequency by calculating an average frequency deviation over a period of time. Secondary control is particularly important because of the interconnection of the French grid with other European grids, which can swing by up to 4000 MW in a single day. In order to adjust the frequency, taking into account the balance of electricity exchanges with other European grids, the grid operator sends a digital signal to the NPP to modify its power level by in the interval of +/- %Pr.
Load following
Nuclear power plants operating in the load-following mode follow a variable load programme with one or two power changes per period of 24 h (Figure 2). The load pattern is determined by the grid operator and the utilities, depending on the power demand and the manoeuvring capabilities of the plant.
Slow ramps of ≤1.5% Pr per minute are most often used in France and the typical low power level is about 50% Pr [1]. However, sometimes nuclear power plants operate at power levels below 50%. Some plants operate in a special operating mode (18 hours at rated power and 6 hours at low power) with steep ramps of 2-5% Pr per minute. In this mode the reactor is always capable of returning to the rated power level in a very short period, with a fast ramp of 5% of Pr per minute [2].
During the licensing process, an NPP’s mode of operation is defined, and all types of transients are analysed. In France and in Germany, load cycling is explicitly defined in the operating handbook of the NPPs. For example, in France the possibility of load-following is taken into account in the operating manual through a certain number of specific margins associated with operating in manoeuvring regime. Before a generic licence can be issued, experiments are performed on a selected unit to analyse operating experience and to validate the safety margins. Once the safety margins are established and the operating licence is issued, the utility commits itself to operate within these margins. In some situations, the regulator can ask to suspend manoeuvring, for example if the physico-chemical characteristics of the core indicate a leak in a fuel element or other malfunction. The operating licence also determines the maximum total number of load cycles based on the original design and the type of transient (magnitude and rate of power variation, etc.).
Utilities’ requirements
At the end of the 1980s, utilities from the United States, Europe and Asia united their efforts in preparing the set of requirements for advanced light water reactors. In 1990, the first edition of the advanced light water reactors’ utility requirements document (URD) was issued by the Electric Power Research Institute (EPRI) in the United States [3].
In 1991, five European utilities (British Energy/Nuclear Electric, EDF, Tractebel and groups of German and Spanish utilities) concluded that a more open specification would be needed to cover a wider range of designs, and thus the European Utilities’ Requirements (EUR) were created. EUR and URD specifications are similar.
However, in some countries such as the USA, there are explicit regulatory prohibitions on manoeuvring in the automatic mode (that is, responding to primary and secondary frequency control), although this does not prohibit power load variations controlled by the operator.
The EUR cover a broad range of conditions for a nuclear power plant to operate efficiently and safely. They include such areas as plant layout and specifications, systems, materials, components, probabilistic safety assessment methodology and availability assessment. Although still requiring regulatory design approval in each country, EUR compliance indicates that the reactor design meets the utilities’ requirements, and that could be proposed throughout Europe without any major design change. Plants certified as complying with EUR include: the Westinghouse AP1000, the Russian AES-92 (with VVER-1000/V-392), EDF/Areva EPR, Toshiba ABWR, Areva KERENA and ABB BWR 90.
The EUR explicitly state that modern reactors should implement significant manoeuvrability capability and, in particular, to be able to operate in the load-following mode. In simple terms, their requirements include [4]:
- The unit must be capable of continuous operation between 50 and 100% of rated power Pr (but not below the minimum power level).
- The standard plant design shall allow the implementation of scheduled and unscheduled load-following operation (that is, a drop in output followed by a plateau and an increase) during 90% of the whole fuel cycle. Restrictions are due to fuel conditions at the end of the cycle.
- The unit shall be capable of load-following operation in the range of output from 100% Pr down to the minimum load of the unit. The standard rate of change of electric output shall be 3% of Pr/min. The unit shall be expected to go through the following number of transients from full power to minimum load and back to full power: two per day, five per week, and cumulatively 200 per year.
- The load following shall be achieved in PWRs without adjusting soluble boron concentration during the manoeuvre. For evolutionary BWRs the load following shall be achieved by recirculation flow control as much as possible, that is, minimising control rod movements.
- The fuel shall be designed to avoid limitations on the rate of power increase for hot start-ups of the plant as well as for cold start-ups.
- The unit may be required to participate in emergency load variations (frequency: once every five years), based on an agreement between the grid operator and the operator of the unit.
- The unit shall be capable of taking part in the primary control of the grid. This is a prerequisite for connection to the grid. The primary control range shall be +/- 2% of the rated power Pr (mandatory), but higher values up to +/- 5% Pr may be agreed between system operators and plant operators.
- The unit shall be capable of activating, within 30 seconds, the total primary range of control requested at a quasi-steady frequency deviation of +/- 200 mHz, and maintaining supply for at least 15 minutes. This time is needed for the grid operator to completely activate the secondary control reserve and the minutes reserve in case of major disturbances.
- Participation in secondary control, which is optional, is based on an agreement between the grid operator and the electricity production company. The variation rate shall be 1% of Pr/min, although values up to 5% Pr/min may be agreed between system operator and plant operator.
An increased flexibility of all sources of electricity demanded by the grid operators is likely. New regulations in this field are currently being prepared in the European Union, for example, by ENTSO-E.
Manoeuvring capabilities
When the first series of 900 MWe PWR was built in France in the 1970, boron-based power regulation was widely used in transients. This is called mode “A”. Boron regulation is quite slow for daily load cycling, but has a considerable advantage because it does not affect the axial power distribution in the core. Boron regulation systems were upgraded to allow daily load cycling. The flexible mode A has been developed, but the amount of the boric acid used in transients was still significant.
Another power regulation mode “G” appeared in 1970. It is based on the use of control rods of variable efficiency, so-called black banks (the most efficient) and grey banks (several times less efficient than the black rods). The development of grey banks started in the middle of 1970. The first tests of assemblies with grey rods were tested in 1981 at the Tricastin NPP. The use of grey rods, combined with variable average temperature of the primary loop, allowed a significant increase of the manoeuvrability capabilities of the plant.
In 1980, development of mode “X” began with an objective to increase the manoeuvrability of the plants and to allow better control of axial power offset. Mode X uses grey and black banks, and boric acid to compensate for reactivity variations due to xenon poisoning and fuel burn-up.
Different operation modes for French NPPs and their manoeuvrability characteristics are summarised in Table 1. (This and following tables refer to the stage of the fuel cycle as a decimal figure, for example the half-way point is written 0.5 C). One may notice a significant improvement of manoeuvrability capability through time.
Russian pressurised water reactors used to be mainly operated in the baseload mode. However, some tests and experiments on the operating modes with variable load have been performed since 1980. The newer generation of large pressurised water reactors VVER-1000 (first series) had seen their manoeuvrability capabilities significantly increased [8]. This became possible because of considerable improvements of different systems: fast boric acid removal from the primary coolant, instrumentation and control (e.g. in-core measurements of the state of the active zone, better regulation of the power), automatic launching of the turbine, etc. The manoeuvring characteristics of the first serial VVER-1000 are summarised in Table 2. However, there are some limitations on the number of load manoeuvres possible, for example, step changes of +/-20% Pr were limited to 150; full-power reductions with speed up to 2% Pr/min to 5000, start-ups from hot conditions to 5000 and start-ups from cold conditions to 130. Manoeuvring capabilities of advanced light water reactors compliant with EUR are the EPR, Table 3, and the Russian AES-92 and AES-2006 with VVER-1000/1200 (versions V-392 and V-491), Table 4.
The economic consequences of load-following are mainly related to the reduction of the load factor. In the case of nuclear energy, fuel costs represent a small fraction of the electricity generating cost, especially compared to fossil sources. Thus, operating at higher load factors is profitable for nuclear power plants as they cannot make savings on fuel costs while not producing electricity. In France, the impact of load-following on the average unit capacity factor is sometimes estimated at about 1.2%.
Since most current nuclear power plants are designed with strong manoeuvrability capabilities (except for some very old NPPs), there is at most limited impact (within the design margins) of load-following on the acceleration of ageing of large equipment components. However, load-following does have some influence on the ageing of certain operational components (for example, valves), and thus one can expect an increase in maintenance costs. Moreover, for older plants some additional investment could be needed, especially in instrumentation and control, to become eligible for operation in load-following mode.
Dr Alexey Lokhov works in the NEA Nuclear Development Division.
This article was based on two recent OECD NEA articles, ‘Load-following with nuclear power plants,’ NEA News 2011- No 29.2, and ‘Technical and economic aspects of load following with nuclear power plants,’ June 2011.
This article was first published in the May 2012 issue of Nuclear Engineering International
References
[1] Cappolani, P, et al., (2004), "La chaudière des réacteurs à eau sous pression", EDP Sciences, 2004, France. [2] Kerkar, N. and P. Paulin (2008), "Exploitation des coeurs REP", EDP Sciences, 2008, France. [3] EPRI (2008), Utility Requirements Document, Revision 10. EPRI, United States, 2008. http://urd.epri.com [4] EUR (2001), European Utility Requirements, Vol. 2, revision C. EUR, France, 2001. [5] UK-EPR (2009), "Pre-Construction Safety Report", UKEPR-0002-012 Issue 01, 2009. Available at: www.epr-reactor.co.uk/ssmod/liblocal/docs/PCSR/Chapter 1 – Introduction and General Description/Sub-Chapter 1.2 – General Description of the Unit.pdf [6] Areva (2008), Presentation "Newbuid EPR Reactors", Ankara, Turkey, September 2008. Available at: http://www.nuke.hun.edu.tr/tr/webfiles/Activities/wnu/18 sept/Othman Salhi/4Newbuild EPR reactors.pdf. [7] Mokhov, V.A and M.A, Podshibiakin (2010) "Study of the NPP with VVER-1000/1200 reactors operation in manoeuvrable mode", Proceedings of the Energia-2010 conference (in Russian), pp. 103-106, 1-3 June 2010, Moscow, Russian Federation. Available at: www.energy2010.mpei.ru/_Files/Proceedings1.pdf [8] Aminov R.Z., et al., (1990), NPP with VVER: Regimes, Characteristics, Effitiency (in Russian), Energoatomizdat, 1990, Moscow, Russian Federation.
