Water Chemistry

The Water Chemistry Data Use Case

Water flowing through the primary piping in an electrical generation unit is converted to steam in the steam generators, and that steam subsequently drives the turbines. The water is monitored for impurities which, when left untreated, can lead to corrosion and other undesirable consequences. Meticulous control of the water chemistry is required to minimize these adverse effects and improve operational efficiency Corrosion and other water chemistry phenomena can affect the structural integrity of the piping, other pressure boundary components, and internal structures within the reactor vessel. Thinning walls, cracking and embrittlement are all possible. These problems are common to both nuclear and non-nuclear power generation units. However, the high levels of radiation in nuclear generating units can further contribute to structural degradation.

In the case of nuclear power generation units, corrosion products can get caught in the fuel channels and plate out on fuel channels and fuel pins. This may affect local flow rates and thereby heat transfer which in turn may cause overheating and damage to the fuel pins. This can result in the escape of highly radioactive fission products into the reactor coolant.

A further problem is that corrosion products can absorb radiation and thereby themselves become radioactive. This can be hazardous to personnel working in the vicinity if these corrosion products then settle in the pipes outside the reactor chamber.

To alleviate these problems, it is essential to maintain water quality. This implies monitoring the levels and composition of impurities. It also means adding chemicals to the water to reduce corrosion and limit other material degradation.

The CMA Database

EPRI maintains the Chemistry Monitoring and Assessment (CMA) database which is used to store water chemistry data received from nuclear power plants both across and outside the US. This data collection process is a manual activity which relies on the filling-out of Excel spreadsheets and their subsequent processing so that they can be loaded into the CMA database. The data are comprehensive, and include:

• identifications of the power plants and units;
• measurement locations;
• measurement types – for example, temperature or concentration;
• if a concentration measurement then the name of the chemical being sampled;
• the unit-of-measure;
• measurement metadata, such as sensor or personnel identifier;
• the actual measurement.

Even within the same utility, plants often do not share the same data definitions, and this makes it difficult to express precise requirements for the data collection process. This renders the data collection process both inefficient and error prone. For each nuclear plant concerned, data analysts from both EPRI and the plant are required to work together to identify how the measurements made at the plant map to the CMA database and, where necessary, how to convert these from the local to the CMA format.

To remedy this situation, the ES-CIM project has extended the CIM to define a common vocabulary for the water chemistry data from which an interoperable messaging format has been developed.

The first step in extending the CIM has been to create a common and consistent set of definitions for all water chemistry measurements. These extensions are described in the paragraphs below.

The details of how these definitions subsequently map to changes in the CIM and hence to an interoperable messaging format are described in “Extending the CIM”.

Standardizing Water Chemistry Measurements

Each water chemistry measurement is qualified by defining what is being measured in the sample, the date and time at which the sample was taken and the location of where the sample was taken.

The notion of what is being measured in a sample is captured by the concept of measurement type. A measurement type is associated with a specific unit of measure (for example, “μCi/ml”). In some cases, a multiple may also be associated with a measurement type such as “ppm” or “ppb”. (For data exchange purposes, special characters are removed to make messages easier to generate and process. For example, “oF” becomes “degF”. Some measurements have no units at all in which case they are given the indicator “none”).

A few measurement types – for example, temperature – are straightforward in that they describe the sample with no need for further qualification. However, many measurement types require an additional parameter termed the measurement name. For example, concentration is a measurement type, but this is useful only when the concentration of the substance involved is also specified – for example, the concentration of chlorine or the concentration of sulphate. The current water chemistry extensions define 37 frequently used concentrations (such as “calcium” or “ammonia”) and 67 specific activities (such as “strontium-91” or “strontium-92”).

Some measurement types have their own variants. For example, pressure may be given as “pressure absolute” or “pressure gauge”.

The current set of measurement types and their corresponding units of measure is provided in the table below:

Measurement Type	Units of Measure	Units of Measure (implemented as)
Concentration	ppb, ppm, cm3/kg	ppb, ppm, cm3/kg
Specific Activity	μCi/ml	μCI/ml
Temperature	oF	degF
Pressure	psi	psi
Mass	g	g
pH	-	none
Conductivity	μS/cm	μS/cm
Volume Flow Rate	gal/min, gal/day	gal/min, gal/day
Per Cent	%	none
Normalized Level	-	none
Hardness	mg/l	mg/l
Alkalinity	mg/l	mg/l
Bacterial Count	planktonicCount, sessileCount and speciesCount
Corrosion Rate	mils/year	mils/years
Scaling Index	-	none
Turbidity	NTU	NTU

The notion of when a sample is taken is defined by the timestamp of the measurement. Non-chemistry measurements such as voltage or pressure can generally be measured instantaneously. However, for water chemistry, the timestamp must be assigned based on the time at which the sample is taken and not when the sample is processed.

The concept of where a measurement is made is defined by a measurement point. A given component within a nuclear power plant – a pump, for example – may have several measurement points. Such a pump might have a pressure measurement at the inlet and another at the outlet. The electrical load of the pump might also be taken at the electrical connection and the measurement of current which thus constitutes a third measurement point.

The ability to specify the location of a measurement point to which the component is associated is tied into the logical structure of a power plant. This concept is useful for correlating measurements of many different varieties within a single generation station as well as for reliably comparing measurements across different generating stations.

For the purposes of standardizing the water chemistry definitions, the ES-CIM project has provided the following generic definitions for the systems and components within a generation station:

• A system is defined as a collection of equipment that work together to perform a function. A system may contain other (sub-)systems.
• A component is a physical device which is generally serviced and possibly replaced as a single device. A component always belongs to at least one system and some components may appear in multiple systems. For example, a coolant pump is a component of the reactant coolant system whereas a steam generator is component of both the reactor coolant system and the main steam system.

The way in which components relate to systems and of how systems relate to one another to form a hierarchy is shown in the figure below.

The structures of a PWR plant hierarchy in terms of systems and components, the details of a primary system and the details of a secondary system are depicted in the figures below.

The table below is an inventory of measurement points currently defined for the CMA database:

Measurement Point	Component / System	Notes
RCS	Reactor Coolant System	Usually either at the hot leg or the CVCS letdown
BAST	Boric acid storage tank	Usually from a recirculation loop
RCSMW	Pure water storage tank	Usually from a recirculation loop
CVCS	Charging Pump	Sampled from a letdown line (downstream of the heat exchangers and upstream of the demineralizer), downstream of the demineralizers or downstream of the filter
VCT	Volume control tank	Dedicated sampling points in the “gas space” of the volume control tank
SFP	Spent fuel pool	Usually from the recirculation loop but may also be via dip sample from the spent fuel pool surface
SGBD	Steam generator(s)	Bulk water at the steam generator blow down. Also known as “boiler blowdown” or “steam generator downcomer”
MSR	Moisture separator reheater	Usually at the heater drain pump sampling location
Hotwell	Condenser	Using vacuum pumps at multiple locations around the hotwell
CPD	Condensate pumps	Withdrawn from the condensate piping downstream of the condensate pump and upstream of the first feedwater heater
FW	High-pressure feedwater heaters	Downstream of the last feedwater heater but upstream of the steam generators. Also called “main feedwater”
RWST	Refuelling water storage tank	Usually conducted from a recirculation loop
Circulating water	Circulating water system	Usually from a dedicated sample point at the inlet of the main condenser
Service water	Service water system

The ES-CIM project has defined details for systems and components to meet the water chemistry requirements. Other business domains will require additional modelling.

The table below illustrates two complete sample measurements:

		Measurement 1	Measurement 2
What	Measurement Type	Specific Activity	Concentration
	Measurement Name	Xenon-131	Chlorine
	Unit Of Measure	μCi/ml	ppb
Location	Measurement Point	RCS	BAST
Timestamp	Date/Time	2021-04-15 09:04	2021-02-24 16:43
Sample Value	0.0123	190

Extending the CIM for the Water Chemistry Data Use Case

The steps that were carried out to build a proof-of-concept implementation for the water chemistry use case were as follows:

1. Analyze the CMA data requirements to identify the kinds of objects whose identifiers or values need to be transferred in the various data messages.
2. For each one of these CMA objects, examine the CIM to determine whether it already has a class counterpart in the CIM.
3. If such a class exists in the CIM, then determine whether its existing definition needs to be extended to match its usage in the CMA database.
4. If such a class does not exist in the CIM, then determine how to incorporate a suitable new class definition into the CIM.
5. Define the individual use cases in terms of the kinds of message that need to be transferred in the various message exchanges.
6. From the newly extended CIM, generate JSD artifacts that correspond to these use cases. A listing of these JSD files is provided in the annex.
7. From these JSD files, generate sample JSON messages and populate them with data taken from the CMA database. A listing of these JSD files is provided in the annex.
8. Using the same JSD files, generate Java or C# stubs with which to build a web server that is capable of accepting such messages and storing the received data in the CMA database.
9. Using this web server, demonstrate message exchange functionality by which utility operators are able to populate the CMA database with new water chemistry data.

The classes that were added to the CIM to capture the CMA objects are as follows:

Class	Definition
ApplicationDataCategory	Provides a means for conveying application-specific data and metadata. A Names.name value of the ApplicationDataCategory (inherited from IdentifiedObject) provides a means for the data to be associated with a database table.
ApplicationDataParameter	Provides a means for conveying application-specific data and metadata.
AssetOperator	The operator of the given asset.
EquipmentOutage	The expected or unexpected outage of an item of equipment
GenerationUnit	A generation unit is a collection of equipment within a generation station that converts a fuel source into electrical power.
GenerationUnitOutage	A generation unit outage occurs when a generation unit is offline and not producing power
NuclearGenerationUnit	A GenerationUnit that generates electricity using nuclear energy
NuclearGenerationUnitOutage	A nuclear generating plant outage occurs when a nuclear generating plant is offline and not producing power
NuclearOperatingCycle	The period of time between refueling outages that a nuclear generating plant is operational
PWRNuclearGenerationUnit	A pressurized water reactor nuclear generation unit
PWRSteamGenerator	A steam generator in a pressurised water reactor generating plant.
PWRSteamGeneratorAsset	The asset fulfilling the role of a steam generator in a pressurised water reactor nuclear generating unit.
PWRSteamGeneratorAssetModel	A pressurised water reactor asset model by a specific manufacturer.
PWRWaterChemistryDataSet	A set of data and measurements resulting from laboratory analysis of water samples taken from a pressurised water reactor nuclear generating plant

Besides these classes, additional classes were also added to the CIM to hold values of specific analog measurement kinds. These are the AlkalinityAnalog, BacterialCountAnalog, ConcentrationAnalog, ConductivityAnalog, CorrosionRateAnalog, HardnessAnalog, MassAnalog, NormalizedLevelAnalog, pHAnalog, PercentAnalog, PressureAnalog, ScalingIndexAnalog, SpecificActivityAnalog, TemperatureAnalog, TurbidityAnalog and VolumeRateAnalog classes. This approach provides a quick and easy way to add new units of measure to the model and is a key feature to making the API specification much faster for subsequent iterations.

The following diagrams illustrate the inheritance and association relationships among these newly defined CIM classes. In these diagrams, a beige box represents classes already present in the CIM. A blue box represents a class that was added during the course of the ES-CIM project. A green box denotes an enumeration set; that is, a set of pre-defined values.

(These diagrams are all from a single CIM UML model. For the sake of clarity, this has been broken up into multiple diagrams. Not all the relationships with other classes are shown).