POSC Specifications: Epicentre Version 3.0 | Logical Data Model |
These changes represent mechanical fixes which implement the intent of the model.
In Version 2.2, two new attributes were added to Well_test_recovery with a mandatory select rule on these two attributes plus attribute reservoir_fluid_system. The new attributes were not included in the SI rule for its subtypes.
Pty_dip_angle has a non-abstract supertye with an SI. The SI rule in Pty_dip_angle conflicts with this inherited rule because it is attempting to broaden the uniqueness. Move the fracture relationship to its supertype so that it can be included in the SI.
Make Topological Object a subtype of Business Object. Move some spatial related behavior up to Spatial Object.
Eliminate the subtypes of Grahical Element as subtypes of Spatial Object. Replace spatial attributes with 2D geometrical characteristics.
Eliminate the many-to-many relationships between spatial subtypes.
Add the capability to assert geometrical relationships independent of topological assertions. For example, objects AA and BB might have a topological assertion that AA is "disjoint" from BB and might also have multiple geometrical assertions such as AA is "above" BB and AA has a 20ft "separation" from BB.
Delete Earth Model Object and all of its subtypes. Move any required locate/represent behavior to Spatial Objects.
Eliminate Other_spatial_object and its subtypes since they are now redundant.
Drop other spatial object from the uniqueness rule of all subtypes of property.
Add Earth_feature as the supertype of the Earth related objects.
More changes related to Earth_feature.
Some changes related to Facility. Move common behavior up to Facility.
Alter the Material hierarchy to include Document and to drop Rock_material_part.
Insert entity Bulk_material between Material and its current subtypes. Move all somewhat specialized behavior from Material to Bulk_material so that Document and Equipment_item will not inherit them.
Eliminate Material_sample.
Eliminate Inventory_object.
Make equipment a subtype of material. Resolve the variations of composition. The change to allow a facility to be classified as to its material nature allows many of the associations to "typical" to be eliminated.
Since Catalog_equipment is non-abstract (and has an SI rule), we need to either eliminate its subtypes or we need to eliminate the SI rule from the subtypes. This change deletes them.
Add an association entity to the many-to-many relationship between Data_collection and E_and_p_data.
Make Ref_data a subtype of E_and_p_data. Create Ref_aliasable_data as the supertype of all ref entities which are aliasable. Move Ref_code to be a subtype of Ref_aliasable_data. Move all non-code entities from Ref_code to Ref_aliasable_data.
Add attribute identifier to Ref_code. Eliminate description and name/kind/identifier from all subtypes. Change the SI rules accordingly.
Move some entities which have the behavior of "codes" to be subtypes of Ref_code. Do not eliminate the inverse relationships under the assumption that all inverse relationships will be explicitly modeled.
The subtypes of Typical_facility have a uniqueness clause even though Typical_facility also has a uniqueness clause. Delete the redundant SI from the subtypes.
Create a new supertype of all subtypes of Technical_object which are also subtypes of Reference_behavior. Move common behavior to the supertype (i.e., identifier, description).
Eliminate Technical_object as a supertype of the subtypes of Subject_of_guideline_or_privilege and then drop it.
Create a new supertype of all subtypes of Association which are also currently subtypes of Reference_behavior. Note that a few subtypes of Technical_object have been included when they have two relationships (and no identifier) as part of their SI rule.
Eliminate the distinction between classification system and naming system. Add a parent/child relationship to the naming system in order to support the semantics of ref_classification_system. Add a many-to-one relationship to a new entity REF_ENTITY to allow an assertion that the identifiers in a naming system should be constrained to use with one entity. Add naming system to most aliasable entities (including ref_code). Adding naming system to reference entities will allow standard instance identifiers to be substituted by an alias identifier. The standard meaning is still provided by the standard instance but the 'name' can be interchanged.
Relate all aliasable entities to Object_alias. The semantics that something is a 'well alias' is provided by the fact that an instance of Well is being aliased.
Move Process_data underneath E_and_data and make it the supertype of most things which are "created by" an activity.
Replicate the Material properties into Rock_feature.
Because Rock_feature is no longer a subtype of Rock_material, replicate the Rock_material properties into Rock_feature.
Hole does not refer to the "usage" of the void. Move it to Rock_material.
Consolidate the classification of all of the subtypes of Earth_feature.
Add new subtype of aliasable_technical_reference for those "classes" which qualify an association as opposed to an object. Note that this adds naming_system to object_abundance_class.
Delete Object_ranking, Ranking_system and Relative_rank as being far too general to be semantically useful. If something is needed then it needs to be more narrowly defined at a lower level.
Change attribute "name" to "identifier" in order to be consistent with other non-ref entities. At the same time, change the type to ndt_identifier.
Most entities in Epicentre have a description attribute. About half of the subtypes of Technical_object and Association do not have this attribute. Note however that Transient_association does have it. Only a few subtypes of Property have the attribute. This change insures that ALL non-abstract entities will have the attribute. Instead of adding this attribute to E_and_p_data, it is being "replicated" at a lower level.
This change creates a supertype of material and facility classes. Since a Facility is the "use" of a material in some role, a facility must be able to be classified based on the material nature of the usage.
Add the capability for a class to be a specialization of one other class. Note that this is a specific variation of the existing class classification. The difference is that specialization asserts a "narrowing" of definition while class classification can now be constrained to assert a "conformance" to a definition.
The changes to the material model have greatly reduced the significance of "typical" materials. The primary semantics that they might now provide will be any required "point" behavior as opposed to the "range" behavior available to classes. Delete all subtypes of typical material and move the semantics to material class. The new "specialization" capability in class classification will allow, for example, the class "isotype" to be the generalization of the class "C12". That is, classes will need to be created to represent the existing subtypes of typical.
: Replicate the Reference_behavior attributes instead of inheriting them. Drop most former subtypes of Reference_behavior and add the new the new entities which consolodate the former subtypes.
Add missing inverse attributes in order to make them visible and queryable. For REF entities, use the verb phrase to generate a generic description (i.e., similar to inverse relationships to properties).
Delete the forward side of attributes which were incompletely deleted by change #1051 in Epicentre V2.2.
Allow a Local_spatial_coordinate_system to be identified within a context. This is needed for parameterized coordinate systems such as measured depth systems within a wellbore and binset coordinate systems.
Add a GIS oriented geometry property which simply uses a list of real. Unless we are going to add a suite of manipulation functions to the DAE then this is just about as useful as adding specialized data type.
Allow most classifiable objects to have the fundamental nature of one class. Each leaf subtype (where the subtypes are considered to be incomplete) may specify that the "kind" is identifying on the instance. This replaces existing one to many relationships to typical_activity and to some REFs.
Eliminate the concept of a Wellbore_reference. Instead, retain Wellbore_point and Wellbore_interval as subtypes of Wellbore_component_facility to represent interesting "named" point/interval facilities in the wellbore. Since the functionality has changed radically, this is logically equivalent to a delete/add.
Make all identifier attributes the same length (80) and use the same NDT for all identifiers. Change the type of any remaining identifiers to ndt_identifier. Any renamed attributes were changed at the time that they were renamed. Any added attributes were properly defined when added. Do not alter non-identifying string attributes.
Fix parsing problems in where rules.
Fix rules which are badly formed.
Create interpreted_feature as the supertype of all interpreted objects.
Convert all remaining many-to-many relationships to an association entity. This will more easily allow local extensions to qualify the relationship. Where appropriate, the new entity will have a name derived from the relational projection intersection table.
Utilize class instead of typical for object semantics. Utilize typical as an archetype for class.
This change implements the recommendations of the EPSG.
Eliminate all subtypes of Activity and move any required behavior up to Activity. In order to capture most of the existing behavior, create Object_activity_involvement as a transient association between Activity and Object_of_interest.
Eliminate all subtypes of Temporal_period and Temporal_event in order to bring this area in line with the geologic changes made in V2.2. Move Temporal_object from Topological_object to Aliasable_technical_object. This is because there does not appear to be any topological behavior which benefits the time concepts.
Rename "typical" geologic to "class" in order to be more consistent with Activity terminology.
Add "common" composition.
Rename Pty_equipment_facility to describe its function rather than who it is "a property of". Moved the uniqueness to the supertype so that the identity of the property is independent of the representation.
Remove many-to-many which contained void. This change represents an adjustment to change ERP_eliminate_void_inverse.
Add missing descriptions in ndt where rules.
Add subtypes of material composition to allow for composition of types. This change represents and adjustment to change ERP_equipment.
DROP facility_material_class and replace it with explicit relationships with MSE rules. Allow an earth_feature to also be classified based on its material nature. This change represents adjustments to changes: ERP_typical_material_subtypes ERP_facility_classification ERP_adjustments
Add rock_material to Interpreted_feature as an alternative to geologic_feature. Allow a wellbore_pick to be alternatively located by wellbore_point or wellbore_interval. Change Seismic_feature_classification to Seismic_pick_classification (i.e., seismic_pick replaced part of its semantics). These changes represent an adjustment to change ERP_interpreted_feature.
Make the data collection identifier and type attributes mandatory.
This modification eliminates the subtypes of field to allow the lifecycle description of a field from prospect to abandonment. This is accomplished by eliminating the subtypes of field and making field non-abstract.
This combines several miscellaneous changes to the ndt specifications.
This change proposes a general remark entity for business objects. Each remark is dated as an optional part, which is included in the uniqueness clause (therefore, only 1 undated remark allowed). No capability is defined to order these remarks except for the date information. Proposed instances for ref_remark_type include: 'source' = 'A comment describing the source of the remark.' 'quality' = 'A remark that relates to the quality of the business object being described.' 'history' = 'Unstructured information concerning the history of the object.'
This change allows the instantiation of a contract amendment. Without this extension, the amendment has to be instantiated as a distinct kind of contract.
This extension increases the scope of objects for which ownership interest may be described with derived interest makeup. Now technical objects may be owned.
This change adds a date for each document_specification, giving the creation date of the document_specification (not the date loaded into the data store).
This change provides the capability to define roles for business associates with respect to document specifications. This replaces the capability to define a business associate in a recorded document with a more general structure. The roles are provided in a new reference entity, in which instances like author, publisher, officer-in-charge, etc. should be defined. This should be an _open_ reference entity. This proposal specifically solves a problem for regulatory agencies who need to know which regulatory officer is assigned which license.
This change creates a subtype of GENERAL FACILITY for the storage of hardcopy documents.
This change adds an attribute to PTY_LOCATION_ACCURACY that provides a qualitative indication of location accuracy. The population of this indicator will have to be further refined.
This change adds an attribute to PTY_LOCATION that provides a status of the location property.
This change adds explicit diameters to pipeline.
This change provides additional attributes to selected property entities for recording probability distribution curves.
This change introduces several facilities that are useful in describing production facilities (compressor, lact_unit ,stock_tank, tank_battery, vapor_recovery_unit).
These modifications describe the changes proposed to improve the description of fluid samples (deleting pressurized sample and adding ability to describe new sample created from existing samples).
This change allows each instance of ref_generic_fluid_component to be described as normally existing in a gaseous, liquid or solid state.
This modification replaces the text form of postal codes with a validated set of postal codes.
This change provides the measurement datum for the water depth information.
This change adds additional descriptive attributes for a seismic source.
This change describes the missing intervals for a subsurface rock segment. It also provides indicator of ordering for repeating intervals.
This change adds surface_cable as a new subtype to general_facility.
This change defines a flag that indicates that one instance of a spatial object is the preferred spatial representation of a locatable object.
This modification removes the redundant entities that duplicate the behavior of facility classification. These entities were not removed when facility classification was introduced.
This removes a redundant uniqueness rule given for well pattern participation.
This change adds the ability to describe the curvature of a wellbore at a point or over an interval, such as the description of dogleg severity. To describe a wellbore as a whole, an interval over the entire wellbore is created.
This extension provides an address subtype for the world wide web. The length of the URL is limited to 255 characters.
This change adds the concept of effective tubing volume. This characteristic is used in transient pressure test analysis to estimate and correct for the effects of after-flow during shut-in tests and unloading during flow tests.
This change adds a new characteristic to fluid flow streams that describes the presence of contaminant in a fluid flow stream. The amount of contaminant is described as a property of association.
This change adds specific entities to describe the borders and boundaries for geopolitical features.
This extension adds an attribute to store the binary contents of an electronic document (i.e., the file itself).
This change defines a new characteristic for a component material type. This property a specific isotope ratio for a type of material within a sample.
This change improves the description of the rock material in a rock fluid feature. The extension moves the relationship from rock fluid feature and rock feature to the supertype of rock feature (rock material). The definition of this relationship is also improved to indicate a parts and assembly relationship. The relationship from rock fluid feature to reservoir fluid system is improved to explicitly describe the concept of the fluid filling the rock material.
This extension defines the orientation point for a noncurved text element. When annotating a graphical representation, this defines the point in the text string at which the location is specified.
This change allows the capture of the trace mnemonic for a well log trace. These are currently loaded into Epicentre as well_log_trace_aliases after a change proposed by POSC for v222. This proposal refines that change.
These are several changes to simplify and improve the description of points and intervals within a wellbore.
This extension provides a strong typing for an earth model. This attribute is intended to describe the primary purpose for which the earth model was created or is to be used.
This change adds the forchheimer coefficient to the other forms describing non-Darcy flow.
These are a series of new properties that are needed for geochemical analysis.
This change defines an optional uncertainty value for geologic age in pty_geologic_age.
This change allows any geologic feature to be the archetype for a material_class.
This change adds a general description of existence probability for geologic features.
This change adds an attribute to PTY_LOCATION_ACCURACY that describes the confidence level used to evaluate the confidence level of the accuracy description. Also, a default confidence level is added to PTY_EDGE_GEOMETRY_UNCERTAINTY to suggest how uncertainty realizations should be generated.
Fix typos found using a spell checker.
Apply additional EPSG recomendations.
Add the Field Bank Potential Field model.
Drop attribute parent_class from the uniqueness of classes.
Shorten entity and attribute names from a maximum of 40 characters to a maximum of 30 characters. This will bring the logical model more in line with available physical engines.
Add Objective as a subtype of Guideline_or_privilege and Objective_set as a subtype of Document_specification. Create Ref_guideline_or_priviledge as the supertype of all "types" of guidelines and privilege and move the "subject of" relationship from Guideline_or_privilege to Ref_guideline_or_priviledge. That is, classes may be related to types rather than specifics. Rename Ref_guideline_compliance_stat to Ref_guideline_compliance so that compliance characterizations other than status may be defined.
Remove references to activity from the relationship between material_installation and material.
Move seismic_data_set to be a subtype of data_trace. Rename well_log_trace_class, well_log_trace_classification, well_log_trace_class_clsn and well_log_trace_alias to refer to data trace instead of well log trace. Add kind to data_trace and add it to the uniqueness of the subtypes. Add entity Common_data_trace_property in order to capture the ELEMENT properties which are associated with each class. Drop Synthetic_log_trace and move the optional "feature" relationship to Well_log_trace.
Add NDT documentation which was inadverdently eliminated by the parser in the alpha version.
Move attribute description up to E_and_p_data.
This deletes ref_wellbore as this duplicates wellbore.kind.
This removes an attribute that is redundant to facility classification.
This generalizes the description of fluid saturations in rock materials and reservoirs.
This removes the alternative production report.
Add missing inverse relationships. Eliminate SELF redeclarations which were obviated by conversion from m-m relationships. Add missing verb phrases. Change the "be a property of" relationship from properties to rock_feature to be optional instead of mandatory.
Make the material, facility and activity composition behavior consistent. Add more composition behavior.
This change reworks the ref property type/kind part of the model as shown on diagram 'MTP: Reference Properties'. The purpose is to simplify the type/kind specification and to add the capability of capturing the information which is contained in the NDT specifications. The kind of property will be independent of any possible representations. Replace many-to-many relationships with intersection entity. Add an optional mechanism to specialize/classify kinds.
Add new subtypes of pty_generic_property. Add a reciprocal relationship on Property which asserts that one property is an equivalent representation of another property. Move the existing relationships up to material and facility.
These changes provide additional 'meta' information on the instantiation state of the data store. This includes indicators for the current valid instance for transient associations, creation and last update information, the content provider for data, and the geographic distribution of naming systems.
This extension adds an attribute to PTY_LOCATION that provides the content source of the location information.
This change alters the current PVT model for fluid phase properties. The existing model is very difficult for both reading and writing due to: * separation of point and curve data into separate instances * disconnected instances for properties that are measured together * excessive coordinate system requirements (e.g., must have temperatures) This proposal uses the concepts developed within the SAVE project to combine multiple, concurrently defined properties in the same property instance. A similar concept was also proposed during the PPDM-POSC data model merger investigation (i.e., aggregate properties). The resulting PVT model is much easier to implement, populate and query. Also, this satisfies POSC change requests 1086 and 1224.
This extension defines grid convergence at a point on the Earth. This is the angular deviation between the local grid north and true north at a point.
This allows the description of the fluid samples that are extracted from each fluid system. This may be used to describe the samples that are extracted from other fluid samples, as well.
This extension adds the recording time delay for a seismograph.
The concept of a cargo describes a discrete amount of material which is stored and/or transported, probably being the subject of a contract.
This modification reorganizes the subtypes of earth_feature. The goals of this change is to eliminate the multiple inheritance created as part of the license_and_lease change proposal. This also adds an exploration area. It also insures that all subtypes of earth_feature have another feature as part of its identity.
This modification merges the surface and mineral (subsurface) concepts for agreements and rights. This is done to accomodate business practices in which subsurface parts of the earth are referenced as if they were surface parts. Extensions included in this change include classification of license rights and license right payments.
This change clarifies the representation of offshore blocks, and adds offshore areas and quadrants.
This proposal adds the capability to describe the content and structure of document specifications. This provides a description of the structure of a document based upon a subset of features in XML Schema.
This applies miscellaneous cleanup changes to the beta version. Move relationship from the subtypes of pty_azimuth up to pty_azimuth so that the relationship can participate in the uniqueness. Make data_value mandatory in properties when that is the only attibute. Add the description back to ndt_acceleration_linear. Drop ref_element_kind since it is redundant to the new ndt_element_type. Add missing rule types. Drop unused NDTs.
Add verb phrases for relationships to REF entities. These originally did not have verb phrases because of the way they were treated in the Oracle case tool.
Fix missing inverse relationships.
Insure that associations between reference data is also reference data.
Convert SET[1:?] to SET[0:?] and add rule which makes the attribute mandatory. This affects only inverse attributes which were formerly mandatory many-to-many.
Properly set the population codes (i.e., local, open, fixed) for some entities. Eliminate population code for entities for which standard instances are questionable.
This change adds some omitted spatial predicates to the enumeration. It also updates the documentation to refer to the new attributes.
This change attempts to do a better job of handling well log trace mnemonics which were stored in Well_log_trace_alias in V2.2.
This change eliminates all remaining LISTs.
Change the naming system from mandatory to optional in Facility. This will make it consistent with all other objects.
Drop the well completion status entities since they are redundant to facility class and classification. Drop the application status and contract status entities since they are redundant to document specification class and classification.
This set of changes finalizes the convergence with the EPSG v5 data.
Make some final cleanup as identified during loading and testing.
Alter ndt_azimuth to use QUANTITY instead of POINT. Knowledge of the point is generally carried by the "object".
This applies final cleanup changes to the beta version. Fix SI rule for a class entity. Drop or fix some rules. Fix ndt_de9im. Change BAG[0:1] to SET[0:1]. Change the population status of some reference entities.
This change sets the default entity projection types for entities. The default is for entites with an SI rule to become tables. Supertypes of those entities will be replicated and subtypes of those entites will be consolidated. Note that this change will not appear as a change to entities in the logical model documentation.
This change sets the default projection name to match the logical name. This necessary in order to override any residual names from V2.2. Note that this change will not show up as a change to the entities or attributes in the logical model documentation.
This change overides the "default" projection names. Attribute names longer than 28 characters that are relationships will have a shorter projected name (to allow for a "_s" suffix). Duplicate attribute names with different data types that are consolidated up to a table will have a different projected name. Attribute names of type quantity that are longer than 28 characters will have a shorter projected name (to allow for "_u" on the units column).
This change defines candidate entities for using natural keys instead of surrogate keys for foreign keys to the specified tables. The setting is set to "yes" for all entities which are a subtype of ref_code (i.e., have a single component in its uniqueness rule that has a type of STRING and is a mandatory attribute). All other entities are set to "no" with the exception of entity naming_system which is set to "yes". Note that a natural key will only be used if the relationship is directly to the specified entity (i.e., not inherited) and the entity becomes a table (i.e., not consolidated or replicated). Otherwise, a surrogate key will be used. Note that this change will not be listed as a change to the entities in the documentation.
This change alters the previously defined "default" point in the hierarchy where a table is created. All of its subtypes below the new point will now consolidate to the supertype while all of the subtypes above the new point will be replicated. This change attempts to replicate the projection points which were implemented in V2.2.
This defines more changes to the previously defined "default" point in the hierarchy where a table is created. All of its subtypes below the new point will now consolidate to the supertype while all of the subtypes above the new point will be replicated. This change adds points which are new for V3.0.
This change alters the model to conform to the new V2.3 methodology. These are the minimal changes required. There is no attempt to take advantage of new capatilities.