Mandd/kg improvs

src/dackar/knowledge_graph/py2neo.py

@@ -107,20 +107,20 @@ def _create_relation(tx, l1, p1, l2, p2, lr, pr):
             lr (str): relationship label
             pr (dict, optional): attributes for relationship. Defaults to None.
         """
+        # Keep the MERGE parts, check node existance prior _create_relation
         if pr is not None:
             query = f"""
-                MERGE (l1:{l1} {{ {', '.join([f'{k}:"{v}"' for k, v in p1.items()])} }})
-                MERGE (l2:{l2} {{ {', '.join([f'{k}:"{v}"' for k, v in p2.items()])} }})
+                MATCH (l1:{l1} {{ {', '.join([f'{k}:"{v}"' for k, v in p1.items()])} }})
+                MATCH (l2:{l2} {{ {', '.join([f'{k}:"{v}"' for k, v in p2.items()])} }})
                 MERGE (l1)-[r:{lr} {{ {', '.join([f'{k}: ${k}' for k in pr.keys()])} }} ]->(l2)
             """
             tx.run(query, **pr)
         else:
             query = f"""
-                MERGE (l1:{l1} {{ {', '.join([f'{k}:"{v}"' for k, v in p1.items()])} }})
-                MERGE (l2:{l2} {{ {', '.join([f'{k}:"{v}"' for k, v in p2.items()])} }})
+                MATCH (l1:{l1} {{ {', '.join([f'{k}:"{v}"' for k, v in p1.items()])} }})
+                MATCH (l2:{l2} {{ {', '.join([f'{k}:"{v}"' for k, v in p2.items()])} }})
                 MERGE (l1)-[r:{lr}]->(l2)
             """
-            # print(query)
             tx.run(query)
 
     def find_nodes(self, label, properties=None):
@@ -278,7 +278,7 @@ def load_dataframe_for_nodes(self, df, labels, properties):
         """
         assert set(properties).issubset(set(df.columns))
         for _, row in df.iterrows():
-            self.create_node(labels, row.to_dict())
+            self.create_node(labels, row[properties].to_dict())
 
     # Load csv function to create relations
     def load_dataframe_for_relations(self, df, l1='sourceLabel', p1='sourceNodeId', l2='targetLabel', p2='targetNodeId', lr='relationshipType', pr=None):
@@ -297,17 +297,19 @@ def load_dataframe_for_relations(self, df, l1='sourceLabel', p1='sourceNodeId',
         # for nodes. Future development need to be performed.
         # label (l1/l2), properties (p1/p2), and relation label (lr), relation properties (pr)
         valid = []
+
         valid.extend([l1, l2, lr, p1, p2])
         if pr is not None:
             valid.extend(pr)
+
         assert set(valid).issubset(set(df.columns))
 
         with self.__driver.session() as session:
             for _, row in df.iterrows():
                 l1_ = row[l1]
-                p1_ = {'nodeId': row[p1]}
+                p1_ = {p1: row[p1]}
                 l2_ = row[l2]
-                p2_ = {'nodeId': row[p2]}
+                p2_ = {p2: row[p2]}
                 lr_ = row[lr]
                 if pr is not None:
                     pr_ = row[pr].to_dict()

src/dackar/knowledge_graph/schemas/conditionReportSchema.toml

@@ -0,0 +1,35 @@
+title = "Condition Report Graph Schema"
+version = 1.0
+
+# ====================
+# Node 
+# ====================
+
+[node.condition_report]
+node_description = """Represents a structured report documenting an observed abnormal event or condition within 
+                      a plant or operational environment. This entity captures descriptive information provided 
+                      by plant staff, including the nature, context, and potential implications of the anomaly."""
+node_properties = [{name = "date", type = "str", optional = false},
+                   {name = "ID",   type = "str", optional = false}]
+
+# ====================
+# Relations
+# ====================
+
+[relation.refers]
+relation_description = """Indicates that a generalized nuclear-relevant entity such as materials, chemical elements 
+                          and compounds, chemical reactions, failure modes, degradation mechanisms, and components 
+                          across electrical, hydraulic, and mechanical systems has been explicitly referenced within 
+                          a condition report"""
+from_entity = "condition_report"
+to_entity = "nuclear_entity"
+relation_properties = [{name = "date", type = "string", optional = false}]
+
+
+[relation.mentions]
+relation_description = """Indicates that a specific system, equipment, or component contained in the provided MBSE model 
+                          has been explicitly referenced within a condition report, establishing a contextual link between 
+                          the reported abnormal event or condition and the physical or functional asset involved."""
+from_entity = "condition_report"
+to_entity = "mbse_entity"
+

src/dackar/knowledge_graph/schemas/customMbseSchema.toml

@@ -0,0 +1,44 @@
+title = "Custom MBSE Model Graph Schema"
+version = "1.0"
+
+# ====================
+# Node Entities
+# ====================
+
+[entities.mbse_entity]
+description = "Represents a specific MBSE component, equipment, or system"
+properties = [
+    { name = "label", type = "string"},
+    { name = "ID",    type = "string", optional = true},
+]
+
+# ====================
+# Relationships
+# ====================
+
+[relationships.link]
+description = "Represents a functional relationship between two entities—such as components equipment 
+               or systems—where one entity influences, supports, or enables the operation, performance or 
+               behavior of the other within a defined operational or engineering context."
+from_entity = "mbse_entity"
+to_entity = "mbse_entity"
+
+
+[relationships.composition]
+description = "Denotes a hierarchical or structural relationship in which one entity—such as a component, 
+               equipment, or system—is a constituent part of another, indicating that the entities are 
+               physically or logically assembled together to form a larger, integrated whole. 
+               This link captures part-whole dependencies essential for understanding system architecture, 
+               configuration, and functional decomposition."
+from_entity = "mbse_entity"
+to_entity = "mbse_entity"
+
+
+[relationships.support]
+description = "Indicates a non-hierarchical relationship in which one entity—such as a component, equipment, 
+               or system—provides auxiliary functionality, infrastructure, or services that enable, enhance, 
+               or maintain the operation of another entity. This link captures dependencies where the supporting 
+               entity is not a direct part of the other, but is essential for its sustained performance, reliability, 
+               or availability within a broader system context."
+from_entity = "mbse_entity"
+to_entity = "mbse_entity"

src/dackar/knowledge_graph/schemas/monitoringSystemSchema.toml

@@ -0,0 +1,31 @@
+title = "Custom MBSE Model Graph Schema"
+version = "1.0"
+
+# ====================
+# Node Entities
+# ====================
+
+[entities.monitoring_variable]
+description = "Represents a measurable physical variable that is continuously or periodically recorded by monitoring 
+               sensors from a specific component, equipment, or system. This entity captures dynamic operational 
+               data that reflects the real-time or historical state of the asset, enabling monitoring, diagnostics, 
+               performance analysis, and integration with predictive models or digital twins within engineering and 
+               plant operations contexts."
+properties = [
+    { name = "label", type = "string", optional = true },
+    { name = "ID",    type = "string"},
+]
+
+# ====================
+# Relationships
+# ====================
+
+[relationships.monitors]
+description = "Defines a semantic relationship in which a monitoring variable is associated with the performance 
+               monitoring of a specific component, equipment, or system. This relation captures the functional 
+               linkage between sensor-derived data and the operational behavior of 
+               the asset, enabling performance assessment, anomaly detection, and predictive maintenance. 
+               It supports traceability between real-time or historical measurements and the physical or logical entities 
+               they characterize within the system model."
+from_entity = "monitoring_variable"
+to_entity = "mbse_entity"

src/dackar/knowledge_graph/schemas/nuclearEntitySchema.toml

@@ -0,0 +1,20 @@
+title = "Nuclear Entity Graph Schema"
+version = "1.0"
+
+# ====================
+# Node Entities
+# ====================
+
+[entities.nuclear_entity]
+description = "Represents a generalized nuclear-relevant entity that may be referenced in unstructured 
+               text authored by plant personnel, encompassing a broad spectrum of technical and 
+               operational subjects. These include materials, chemical elements and compounds, chemical 
+               reactions, failure modes, degradation mechanisms, and components across electrical, 
+               hydraulic, and mechanical systems."
+properties = [
+    { name = "class", type = "string"},
+]
+
+
+
+

src/dackar/knowledge_graph/schemas/numericPerfomanceSchema.toml

@@ -0,0 +1,31 @@
+title = "Custom MBSE Model Graph Schema"
+version = "1.0"
+
+# ====================
+# Node Entities
+# ====================
+
+[entities.anomaly]
+description = " Represents a specific occurrence of an anomaly (such as threshold 
+                violations, signal deviations, or unexpected patterns) associated 
+                with a particular component, equipment, or system that have been 
+                detected by monitoring sensors coupled with an anomaly detection 
+                algorithm. " 
+
+properties = [
+    { name = "t_initial", type = "datetime"},
+    { name = "t_final",   type = "datetime", optional = true },
+]
+
+# ====================
+# Relationships
+# ====================
+
+[relationships.detected_by]
+description = " Defines a semantic relationship indicating that a detected anomaly 
+                (such as a deviation from expected operational behavior) was identified 
+                through the observation or analysis of a specific monitoring variable. 
+                This link establishes traceability between the anomaly instance and the 
+                physical variable that triggered its detection, "
+from_entity = "anomaly"
+to_entity = "monitoring_variable"