{"id":248703,"date":"2026-04-27T12:40:58","date_gmt":"2026-04-27T10:40:58","guid":{"rendered":"https:\/\/graphwise.ai\/?post_type=blog-post&#038;p=248703"},"modified":"2026-07-10T05:35:18","modified_gmt":"2026-07-10T03:35:18","slug":"advancing-power-system-analysis-with-ai-and-semantic-data","status":"publish","type":"blog-post","link":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/","title":{"rendered":"Advancing Power System Analysis with AI and Semantic Data"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">In the evolving landscape of the energy sector, the ability to efficiently interact with complex data models is becoming a necessity. This blog post explores the <a href=\"https:\/\/github.com\/statnett\/Talk2PowerSystem\" target=\"_blank\" rel=\"noreferrer noopener\">Talk2PowerSystem<\/a> project, a collaboration between <a href=\"https:\/\/www.statnett.no\/en\/\" target=\"_blank\" rel=\"noreferrer noopener\">Statnett<\/a> and Graphwise, that creates AI-based tools and methods\u00a0 to automate the analysis of complex power system models and deliver actionable insights for engineers.\u00a0<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Through the synergy of <a href=\"https:\/\/gws-sso-test.graphwise.ai\/fundamentals\/what-is-large-language-model\/\" target=\"_blank\" rel=\"noreferrer noopener\">large language models (LLMs)<\/a>, <a href=\"https:\/\/gws-sso-test.graphwise.ai\/fundamentals\/what-is-semantic-web-and-semantic-technology\/\" target=\"_blank\" rel=\"noreferrer noopener\">Semantic Web technologies<\/a>, and geospatial standards, <strong>Talk2PowerSystem<\/strong> aims to revolutionize the analysis of grid data. Today, power system engineers rely on the <a href=\"https:\/\/www.entsoe.eu\/data\/cim\/\" target=\"_blank\" rel=\"noreferrer noopener\">Common Information Model<\/a><strong> (CIM)<\/strong> \u2014 a robust but incredibly complex framework of standards (IEC 61970, 61968, 62325). While CIM is the gold standard for data, its sheer size makes crafting graph queries a daunting task for even the most seasoned engineers. Our vision is to let experts query these models using <strong>natural language<\/strong> instead.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Delivering on that vision of <a href=\"https:\/\/gws-sso-test.graphwise.ai\/fundamentals\/what-is-natural-language-querying\/\" target=\"_blank\" rel=\"noreferrer noopener\">natural language querying<\/a> is not straightforward: industry-standard <a href=\"https:\/\/gws-sso-test.graphwise.ai\/fundamentals\/what-is-ontology\/\" target=\"_blank\" rel=\"noreferrer noopener\">ontologies<\/a> are enormous, and while LLMs open up new ways to interact with <a href=\"https:\/\/gws-sso-test.graphwise.ai\/fundamentals\/what-is-a-knowledge-graph\/\" target=\"_blank\" rel=\"noreferrer noopener\">knowledge graphs<\/a>, their context windows are limited \u2014 feeding them verbose schemas <strong>can lead to confusion or lost instructions<\/strong>. In the rest of this post, we present three important aspects of our approach in <strong>Talk2PowerSystem<\/strong>:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>simplifying ontologies<\/strong> so LLMs can digest them efficiently,&nbsp;<\/li>\n\n\n\n<li><strong>applying semantic reasoning<\/strong> to collapse deep CIM hierarchies into query-friendly shortcuts, and&nbsp;<\/li>\n\n\n\n<li><strong>grounding the model in physical space with GeoSPARQL<\/strong> to unlock spatial querying and visualization.<\/li>\n<\/ul>\n\n\n\n<h2 id='simplifying-ontologies-for-llms-lessons-from-the-field'  id=\"boomdevs_1\" class=\"wp-block-heading\" id=\"h-simplifying-ontologies-for-llms-lessons-from-the-field\">Simplifying ontologies for LLMs: Lessons from the field<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">To enable a chatbot to reliably query complex ontologies, <strong>we needed to make them \u201cLLM-friendly\u201d<\/strong>. CIM for the electrical grid consists of over 20 parts (profiles) with a lot of duplication and complexity. Their union has over 900 classes and 5500 properties, but an actual grid data would use less than half of all ontology terms.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">It is not only in CIM. Other standard ontologies, such as the ERA Vocabulary, are also designed for completeness. They often contain:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Long descriptions of regulations<\/li>\n\n\n\n<li>Administrative <a href=\"https:\/\/gws-sso-test.graphwise.ai\/fundamentals\/what-is-metadata\/\" target=\"_blank\" rel=\"noreferrer noopener\">metadata<\/a> (creation dates, XML mappings)<\/li>\n\n\n\n<li>Translations in multiple languages<\/li>\n\n\n\n<li>Many unused classes and properties<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">When an LLM processes this raw data, it consumes many tokens and can lose focus on the relevant structure. We used <strong>a pipeline to clean and simplify the ontology before presenting it to the AI <\/strong>and employed the following key optimization strategies:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Pruning textual descriptions:<\/strong> An effective way to save tokens is shortening <code>rdfs:comment<\/code> fields by cutting descriptions at approximately 400 characters using <a href=\"https:\/\/gws-sso-test.graphwise.ai\/fundamentals\/what-is-sparql\/\" target=\"_blank\" rel=\"noreferrer noopener\">SPARQL<\/a> regex updates. This removes excessive legal citations or historical notes that do not help the LLM understand the current data structure, ensuring that the model focuses on the core definitions of entities.<\/li>\n\n\n\n<li><strong>Removing &#8220;bookkeeping&#8221; metadata:<\/strong> Ontologies are often cluttered with administrative data such as <code>dct:modified<\/code> dates, creator names, and contributor lists. While vital for governance, this information is &#8220;noise&#8221; for an LLM trying to generate a query. Deleting these properties significantly compacts the schema.<\/li>\n\n\n\n<li><strong>Enhancing presentation and formatting:<\/strong> The way Turtle (<a href=\"https:\/\/gws-sso-test.graphwise.ai\/fundamentals\/what-is-rdf\/\" target=\"_blank\" rel=\"noreferrer noopener\">RDF<\/a>) is serialized matters. Standard serializations often produce &#8220;ugly&#8221; output \u2014 scattered definitions, unordered terms, and confusing blank nodes for <a href=\"https:\/\/www.w3.org\/OWL\/\" target=\"_blank\" rel=\"noreferrer noopener\">OWL<\/a> restrictions. We recommend using tools like <code>turtle-formatter<\/code> to group classes, properties, and individuals logically and alphabetically. &#8220;Prettier&#8221; schemas lead to higher accuracy in query generation.<\/li>\n\n\n\n<li><strong>Ontology subsetting:<\/strong> Large-scale ontologies like CIM contain thousands of classes. However, specific datasets often use only a fraction of them. By subsetting the ontology &#8211; including only the classes and properties which are actually instantiated in the knowledge graph- developers can drastically reduce the token load.<\/li>\n<\/ol>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full is-resized\"><img decoding=\"async\" width=\"2048\" height=\"2048\" src=\"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/ontology-simplification-llm2.png\" alt=\"ontology simplification\" class=\"wp-image-248716\" style=\"width:600px\" srcset=\"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/ontology-simplification-llm2.png 2048w, https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/ontology-simplification-llm2-1280x1280.png 1280w, https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/ontology-simplification-llm2-980x980.png 980w, https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/ontology-simplification-llm2-480x480.png 480w\" sizes=\"(min-width: 0px) and (max-width: 480px) 480px, (min-width: 481px) and (max-width: 980px) 980px, (min-width: 981px) and (max-width: 1280px) 1280px, (min-width: 1281px) 2048px, 100vw\" \/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">By removing noise and reducing the size of the schema to 285 classes and 445 properties, we improved the LLM\u2019s ability to generate correct SPARQL queries. A cleaner schema also saves tokens and helps the model focus on the relevant data structure. Further details of our approach can be found in this detailed technical blog: <a href=\"https:\/\/github.com\/statnett\/Talk2PowerSystem\/wiki\/Blog-Ontology-Simplification-for-LLM\" target=\"_blank\" rel=\"noreferrer noopener\">Simplifying Ontologies for LLMs: Lessons from the Field<\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Beyond simplifying the schema, the project also had to address the logical complexity of how data is interconnected.<\/p>\n\n\n\n<h2 id='using-semantic-reasoning-to-simplify-llm-sparql-generation-for-electrical-cim'  id=\"boomdevs_2\" class=\"wp-block-heading\">Using semantic reasoning to simplify LLM SPARQL generation for electrical CIM<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">We apply custom semantic reasoning to CIM in order to simplify deep graph hierarchies into manageable &#8220;shortcut&#8221; relations, making the data model more manageable for querying by LLMs.<\/p>\n\n\n\n<h3 id='the-challenge-deep-nesting-and-long-traversals'  id=\"boomdevs_3\" class=\"wp-block-heading\">The challenge: Deep nesting and long traversals<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">As already discussed, CIM is a robust standard for representing the entire electrical enterprise. However, its granularity makes querying difficult. A simple request like <em>\u201cList all substations connected via an AC-line to substation X\u201d <\/em>requires navigating a deep hierarchy of objects:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Substations <\/strong>contain <strong>VoltageLevels<\/strong>, which contain <strong>Bays<\/strong>, which contain <strong>Equipments<\/strong><\/li>\n\n\n\n<li><strong>Equipments <\/strong>connect via <strong>Terminals<\/strong> to <strong>ConnectivityNodes<\/strong>, which connect to other <strong>Terminals <\/strong>of other <strong>Equipments<\/strong><\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large is-resized\"><img decoding=\"async\" width=\"1024\" height=\"751\" src=\"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/summary-reasoning-1024x751.jpg\" alt=\"\" class=\"wp-image-248717\" style=\"width:601px\" srcset=\"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/summary-reasoning-1024x751.jpg 1024w, https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/summary-reasoning-980x719.jpg 980w, https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/summary-reasoning-480x352.jpg 480w\" sizes=\"(min-width: 0px) and (max-width: 480px) 480px, (min-width: 481px) and (max-width: 980px) 980px, (min-width: 981px) 1024px, 100vw\" \/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">In order to answer the request, a raw SPARQL query needs to traverse from a <strong>Substation <\/strong>\u201cdown\u201d to its parts, to <strong>Terminals <\/strong>and <strong>ConnectivityNodes<\/strong>, then \u201cup\u201d to parts of <strong>Lines <\/strong>(called <code>ACLineSegment<\/code>), \u201cdown\u201d the other side of the <strong>Line<\/strong>, and \u201cup\u201d to the other <strong>Substation<\/strong>. This involves multiple unions, transitive closures, and property paths.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">While execution is fast (approximately 0.1s), the query structure is too complex for LLMs to generate reliably. LLMs often produce syntactically correct but inefficient or semantically incorrect queries when faced with such deep nesting.<\/p>\n\n\n\n<h3 id='the-solution-semantic-reasoning-and-inference'  id=\"boomdevs_4\" class=\"wp-block-heading\">The solution: Semantic reasoning and inference<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">To address this, we implemented standard OWL2 RL rules to introduce \u201cshortcut\u201d relations that simplify the graph structure for the LLM without altering the underlying data model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">We defined a new namespace <code>cimr:<\/code> (CIM Rules) and added the following inferred relations:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Super-properties (union properties):<\/strong> <code>cimr:hasPart<\/code> aggregates various containment relations (for example, <code>EquipmentContainer.Equipments<\/code>, <code>Substation.VoltageLevels<\/code>).<\/li>\n\n\n\n<li><strong>Transitive closure:<\/strong> <code>cimr:hasPartTransitive<\/code> allows querying parts at any depth.<\/li>\n\n\n\n<li><strong>Property paths:<\/strong> <code>cimr:connectedTo<\/code> shortcuts the Terminal-ConnectivityNode-Terminal path.<\/li>\n\n\n\n<li><strong>Composite paths:<\/strong> <code>cimr:connectedThroughPart<\/code> links containers directly if their internal components are connected.<\/li>\n<\/ol>\n\n\n\n<h3 id='outcomes'  id=\"boomdevs_5\" class=\"wp-block-heading\">Outcomes<\/h3>\n\n\n\n<h4 id='simplified-queries'  id=\"boomdevs_6\" class=\"wp-block-heading\"><strong>Simplified queries<\/strong><\/h4>\n\n\n\n<p class=\"wp-block-paragraph\">The complex SPARQL query was reduced to a few readable lines. The LLM can now simply ask for <code>cimr:connectedThroughPart<\/code> rather than constructing a multi-line property path.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Before (simplified snippet):<\/strong><\/p>\n\n\n\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\" data-enlighter-theme=\"\" data-enlighter-highlight=\"\" data-enlighter-linenumbers=\"\" data-enlighter-lineoffset=\"\" data-enlighter-title=\"\" data-enlighter-group=\"\">?sub (cim:EquipmentContainer.Equipments|\n      cim:Substation.VoltageLevels|\n      cim:VoltageLevel.Bays)+\n\/\n      cim:ConductingEquipment.Terminals\n\/\n      cim:Terminal.ConnectivityNode\n\/\n      cim:ConnectivityNode.Terminals\n\/\n      cim:Terminal.ConductingEquipment\n\/\n      cim:Equipment.EquipmentContainer\n?line .\n<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>After:<\/strong><\/p>\n\n\n\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\" data-enlighter-theme=\"\" data-enlighter-highlight=\"\" data-enlighter-linenumbers=\"\" data-enlighter-lineoffset=\"\" data-enlighter-title=\"\" data-enlighter-group=\"\">?sub cimr:connectedThroughPart ?line<\/pre>\n\n\n\n<h4 id='performance-and-efficiency'  id=\"boomdevs_7\" class=\"wp-block-heading\"><strong>Performance and efficiency<\/strong><\/h4>\n\n\n\n<p class=\"wp-block-paragraph\">We created a custom <code>cim.pie<\/code> ruleset with custom optimizations like <code>transitiveOver<\/code> and fixed-arity property chains to keep reasoning efficient. The inferred triples increased the knowledge graph size by approximately <strong>1.95x<\/strong>, a manageable expansion that enables significantly easier querying with negligible impact on query speed.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In summary, through semantic reasoning, we can bridge the gap between complex domain ontologies and LLM capabilities. Documenting these inferred properties allows the LLM to discover and use them, thus transforming a prone-to-error SPARQL generation task into a reliable one. If you are interested in getting further into the details of this process, please check out our post <a href=\"https:\/\/github.com\/statnett\/Talk2PowerSystem\/wiki\/Blog-Using-Semantic-Reasoning-to-Help-LLM-with-SPARQL-Generation-in-Electrical-CIM\" target=\"_blank\" rel=\"noreferrer noopener\">Using Semantic Reasoning to Help LLM with SPARQL Generation in Electrical CIM<\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Finally, to make Talk2PowerSystem insights actionable for field operations and planning, we had to ground the logical model in physical space.<\/p>\n\n\n\n<h2 id='mapping-electrical-resources-with-geosparql'  id=\"boomdevs_8\" class=\"wp-block-heading\">Mapping electrical resources with GeoSPARQL<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Power systems are inherently spatial: power lines traverse landscapes and substations occupy specific locations. By bridging CIM with geospatial standards, we unlock powerful visualization and analysis capabilities \u2014 from simple mapping to complex spatial querying and reasoning.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full is-resized\"><img decoding=\"async\" width=\"1200\" height=\"880\" src=\"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/summary-mapping-geosparql.jpg\" alt=\"\" class=\"wp-image-248718\" style=\"width:800px\" srcset=\"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/summary-mapping-geosparql.jpg 1200w, https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/summary-mapping-geosparql-980x719.jpg 980w, https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/summary-mapping-geosparql-480x352.jpg 480w\" sizes=\"(min-width: 0px) and (max-width: 480px) 480px, (min-width: 481px) and (max-width: 980px) 980px, (min-width: 981px) 1200px, 100vw\" \/><\/figure>\n\n\n\n<h3 id='the-standard-geosparql'  id=\"boomdevs_9\" class=\"wp-block-heading\">The Standard: GeoSPARQL<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">To represent geospatial objects in RDF, we use the <a href=\"https:\/\/www.ogc.org\/standards\/geosparql\/\" target=\"_blank\" rel=\"noreferrer noopener\">GeoSPARQL Ontology<\/a>. This allows us to map CIM structures directly to standard geospatial types:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Features:<\/strong> <code>cim:PowerSystemResource<\/code> becomes also geo:Feature<\/li>\n\n\n\n<li><strong>Geometries:<\/strong> <code>cim:Location<\/code> maps to <code>geo:Geometry<\/code>, storing coordinates as <a href=\"https:\/\/www.ogc.org\/standards\/wkt-crs\/\" target=\"_blank\" rel=\"noreferrer noopener\">Well-Known Text<\/a> (WKT) literals (for example, <code>POINT<\/code>, <code>LINESTRING<\/code>, <code>POLYGON<\/code>).<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">This standardization ensures compatibility with the broader ecosystem of geospatial tools and triple stores like <a href=\"https:\/\/gws-sso-test.graphwise.ai\/components\/graphdb\/\" target=\"_blank\" rel=\"noreferrer noopener\">Graphwise GraphDB<\/a>.<\/p>\n\n\n\n<h3 id='visualization-and-context'  id=\"boomdevs_10\" class=\"wp-block-heading\">Visualization and context<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">While raw data is useful, visualization provides immediate insight. Since GeoSPARQL is a commonly implemented OGC standard, we can use third-party tools like <a href=\"https:\/\/github.com\/TriplyDB\/YASGUI\" target=\"_blank\" rel=\"noreferrer noopener\">YasGUI<\/a> to render query results directly on a map.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Beyond simple plotting, SPARQL queries can dynamically style elements (for example, coloring power lines red or green based on their nominal voltage) and generate rich tooltips for better context.<\/p>\n\n\n\n<h3 id='spatial-querying-and-integration'  id=\"boomdevs_11\" class=\"wp-block-heading\">Spatial querying and integration<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The true power of GeoSPARQL lies in querying data based on spatial relationships, not just topological connections.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Geometric filters:<\/strong> We can query for objects that intersect specific bounding boxes or polygons (for example, <em>\u201cFind all assets within the Sandefjord industrial area\u201d<\/em>)<\/li>\n\n\n\n<li><strong>Distance analysis:<\/strong> We can order or filter objects by their distance from a specific point<\/li>\n\n\n\n<li><strong>Federated queries:<\/strong> We can combine our private grid data with public datasets like <a href=\"https:\/\/www.openstreetmap.org\/\" target=\"_blank\" rel=\"noreferrer noopener\">OpenStreetMap<\/a> (via SPARQL federation). This allows us to ask complex questions like \u201cWhich power lines cross this specific administrative boundary?\u201d without storing that boundary data ourselves.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">In brief, integrating GeoSPARQL with CIM smooths the transition between electrical engineering and GIS. It enables us to enrich power system analytics with geospatial context, leveraging existing standards to visualize, query, and reason about the grid in relation to the real world. If you would like to dig deeper, please check this detailed technical blog: <a href=\"https:\/\/github.com\/statnett\/Talk2PowerSystem\/wiki\/Blog-Mapping-Electrical-Resources-with-GeoSPARQL\" target=\"_blank\" rel=\"noreferrer noopener\">Mapping Electrical Resources with GeoSPARQL<\/a>.<\/p>\n\n\n\n<h2 id='conclusion'  id=\"boomdevs_12\" class=\"wp-block-heading\">Conclusion<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The <strong>Talk2PowerSystem<\/strong> project proves that the future of power systems isn&#8217;t just about more data \u2014 it&#8217;s about more <em>accessible<\/em> data. By simplifying the vocabulary, creating logical shortcuts through reasoning, and grounding everything in geospatial context, we are empowering the engineers to talk directly to the systems they manage and gather actionable insights.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Do you want to to learn more about the project and our approach to solving such challenges<\/strong><\/p>\n\n\n\n<div class=\"hs-cta-embed hs-cta-simple-placeholder hs-cta-embed-211771861846\"\n  style=\"max-width:100%; max-height:100%; width:700px;height:202.375px\" data-hubspot-wrapper-cta-id=\"211771861846\">\n  <a href=\"https:\/\/cta-service-cms2.hubspot.com\/web-interactives\/public\/v1\/track\/redirect?encryptedPayload=AVxigLIcOFgqs0L4NfnnDdVSMILgXfTGxMe69BK%2BfVlonCr%2FKfjLVe%2Be0tmqEUz5thGP38h76o3ZxpqfZySzxbq0OFl1TjDIwoMUy3nIomhvUJMF3Mkad1OHqSq0jI99PeqTCS66wgOFXX9Lgvniw1KYS%2F22KrTYZlKcoWcXyWWq8rlWdBC%2FTP6kkitzqV4nYmoPW5lJFzThbswLSM0qfIomeNnk48WX%2BVCEsNKxphqEWkkxDtPmmXGeOPmbvyzWcWgEwpRQj5kDkA%3D%3D&#038;webInteractiveContentId=211771861846&#038;portalId=5619976\" target=\"_blank\" rel=\"noopener\" crossorigin=\"anonymous\">\n    <img decoding=\"async\" alt=\"Join us for the webinar: Talk2PowerSystem: Democratizing Power System Analytics via Generative AI &nbsp;\" loading=\"lazy\" src=\"https:\/\/no-cache.hubspot.com\/cta\/default\/5619976\/interactive-211771861846.png\" style=\"height: 100%; width: 100%; object-fit: fill\"\n      onerror=\"this.style.display='none'\" \/>\n  <\/a>\n<\/div>\n","protected":false},"excerpt":{"rendered":"Read about how Talk2PowerSystem enables power system engineers to query complex grid data using natural language.","protected":false},"author":10,"featured_media":248712,"template":"","meta":{"_acf_changed":false,"_et_pb_use_builder":"off","_et_pb_old_content":"","_et_gb_content_width":"","content-type":"","_EventAllDay":false,"_EventTimezone":"","_EventStartDate":"","_EventEndDate":"","_EventStartDateUTC":"","_EventEndDateUTC":"","_EventShowMap":false,"_EventShowMapLink":false,"_EventURL":"","_EventCost":"","_EventCostDescription":"","_EventCurrencySymbol":"","_EventCurrencyCode":"","_EventCurrencyPosition":"","_EventDateTimeSeparator":"","_EventTimeRangeSeparator":"","_EventOrganizerID":[],"_EventVenueID":[],"_OrganizerEmail":"","_OrganizerPhone":"","_OrganizerWebsite":"","_VenueAddress":"","_VenueCity":"","_VenueCountry":"","_VenueProvince":"","_VenueState":"","_VenueZip":"","_VenuePhone":"","_VenueURL":"","_VenueStateProvince":"","_VenueLat":"","_VenueLng":"","_VenueShowMap":false,"_VenueShowMapLink":false},"categories":[152,23,21,463],"tags":[],"persona":[406],"resource-category":[17],"blog-category":[],"ppma_author":[202,402],"class_list":["post-248703","blog-post","type-blog-post","status-publish","has-post-thumbnail","hentry","category-ai-in-action","category-graph-database","category-graph-rag","category-semantic-digital-twin","persona-kevin","resource-category-blog-post"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v26.3 (Yoast SEO v27.8) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>Advancing Power System Analysis with AI and Semantic Data<\/title>\n<meta name=\"description\" content=\"Learn how Talk2PowerSystem enables power system engineers to query complex grid data using natural language.\" \/>\n<meta name=\"robots\" content=\"noindex, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<meta property=\"og:locale\" content=\"de_DE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Advancing Power System Analysis with AI and Semantic Data\" \/>\n<meta property=\"og:description\" content=\"Learn how Talk2PowerSystem enables power system engineers to query complex grid data using natural language.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/\" \/>\n<meta property=\"og:site_name\" content=\"Graphwise\" \/>\n<meta property=\"article:modified_time\" content=\"2026-07-10T03:35:18+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png\" \/>\n\t<meta property=\"og:image:width\" content=\"2000\" \/>\n\t<meta property=\"og:image:height\" content=\"1000\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Gesch\u00e4tzte Lesezeit\" \/>\n\t<meta name=\"twitter:data1\" content=\"8\u00a0Minuten\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/advancing-power-system-analysis-with-ai-and-semantic-data\\\/\",\"url\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/advancing-power-system-analysis-with-ai-and-semantic-data\\\/\",\"name\":\"Advancing Power System Analysis with AI and Semantic Data\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/advancing-power-system-analysis-with-ai-and-semantic-data\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/advancing-power-system-analysis-with-ai-and-semantic-data\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/wp-content\\\/uploads\\\/2026\\\/04\\\/AI-in-action-series.png\",\"datePublished\":\"2026-04-27T10:40:58+00:00\",\"dateModified\":\"2026-07-10T03:35:18+00:00\",\"description\":\"Learn how Talk2PowerSystem enables power system engineers to query complex grid data using natural language.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/advancing-power-system-analysis-with-ai-and-semantic-data\\\/#breadcrumb\"},\"inLanguage\":\"de\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/advancing-power-system-analysis-with-ai-and-semantic-data\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/advancing-power-system-analysis-with-ai-and-semantic-data\\\/#primaryimage\",\"url\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/wp-content\\\/uploads\\\/2026\\\/04\\\/AI-in-action-series.png\",\"contentUrl\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/wp-content\\\/uploads\\\/2026\\\/04\\\/AI-in-action-series.png\",\"width\":2000,\"height\":1000},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/advancing-power-system-analysis-with-ai-and-semantic-data\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Blog Posts\",\"item\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Advancing Power System Analysis with AI and Semantic Data\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/#website\",\"url\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/\",\"name\":\"Graphwise\",\"description\":\"AI THRIVES ON WHOLE DATA\",\"publisher\":{\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"de\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/#organization\",\"name\":\"Graphwise\",\"url\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/wp-content\\\/uploads\\\/2024\\\/10\\\/graphwise-logo-horizontal-slogan.svg\",\"contentUrl\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/wp-content\\\/uploads\\\/2024\\\/10\\\/graphwise-logo-horizontal-slogan.svg\",\"width\":\"1024\",\"height\":\"1024\",\"caption\":\"Graphwise\"},\"image\":{\"@id\":\"https:\\\/\\\/gws-sso-test.graphwise.ai\\\/de\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.linkedin.com\\\/company\\\/graphwise\\\/\"],\"description\":\"Graphwise enables organizations to unlock ROI for enterprise AI by delivering the most comprehensive and trusted industry solution in the field of knowledge graphs and semantic AI technologies. As enterprises pour millions into AI investment, Graphwise delivers the critical knowledge graph infrastructure to ensure enterprises are ready to realize the technology\u2019s full potential, is trusted, and can be implemented at scale. Graphwise, which is the result of the merger between tech visionaries Ontotext and Semantic Web Company, has over 200 employees worldwide, with offices located across North America, Europe and APAC.\",\"numberOfEmployees\":{\"@type\":\"QuantitativeValue\",\"minValue\":\"201\",\"maxValue\":\"500\"}}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Advancing Power System Analysis with AI and Semantic Data","description":"Learn how Talk2PowerSystem enables power system engineers to query complex grid data using natural language.","robots":{"index":"noindex","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"og_locale":"de_DE","og_type":"article","og_title":"Advancing Power System Analysis with AI and Semantic Data","og_description":"Learn how Talk2PowerSystem enables power system engineers to query complex grid data using natural language.","og_url":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/","og_site_name":"Graphwise","article_modified_time":"2026-07-10T03:35:18+00:00","og_image":[{"width":2000,"height":1000,"url":"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png","type":"image\/png"}],"twitter_card":"summary_large_image","twitter_misc":{"Gesch\u00e4tzte Lesezeit":"8\u00a0Minuten"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/","url":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/","name":"Advancing Power System Analysis with AI and Semantic Data","isPartOf":{"@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/#website"},"primaryImageOfPage":{"@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/#primaryimage"},"image":{"@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/#primaryimage"},"thumbnailUrl":"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png","datePublished":"2026-04-27T10:40:58+00:00","dateModified":"2026-07-10T03:35:18+00:00","description":"Learn how Talk2PowerSystem enables power system engineers to query complex grid data using natural language.","breadcrumb":{"@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/#breadcrumb"},"inLanguage":"de","potentialAction":[{"@type":"ReadAction","target":["https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/"]}]},{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/#primaryimage","url":"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png","contentUrl":"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png","width":2000,"height":1000},{"@type":"BreadcrumbList","@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/advancing-power-system-analysis-with-ai-and-semantic-data\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/gws-sso-test.graphwise.ai\/de\/"},{"@type":"ListItem","position":2,"name":"Blog Posts","item":"https:\/\/gws-sso-test.graphwise.ai\/de\/blog\/"},{"@type":"ListItem","position":3,"name":"Advancing Power System Analysis with AI and Semantic Data"}]},{"@type":"WebSite","@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/#website","url":"https:\/\/gws-sso-test.graphwise.ai\/de\/","name":"Graphwise","description":"AI THRIVES ON WHOLE DATA","publisher":{"@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/gws-sso-test.graphwise.ai\/de\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"de"},{"@type":"Organization","@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/#organization","name":"Graphwise","url":"https:\/\/gws-sso-test.graphwise.ai\/de\/","logo":{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/#\/schema\/logo\/image\/","url":"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2024\/10\/graphwise-logo-horizontal-slogan.svg","contentUrl":"https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2024\/10\/graphwise-logo-horizontal-slogan.svg","width":"1024","height":"1024","caption":"Graphwise"},"image":{"@id":"https:\/\/gws-sso-test.graphwise.ai\/de\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.linkedin.com\/company\/graphwise\/"],"description":"Graphwise enables organizations to unlock ROI for enterprise AI by delivering the most comprehensive and trusted industry solution in the field of knowledge graphs and semantic AI technologies. As enterprises pour millions into AI investment, Graphwise delivers the critical knowledge graph infrastructure to ensure enterprises are ready to realize the technology\u2019s full potential, is trusted, and can be implemented at scale. Graphwise, which is the result of the merger between tech visionaries Ontotext and Semantic Web Company, has over 200 employees worldwide, with offices located across North America, Europe and APAC.","numberOfEmployees":{"@type":"QuantitativeValue","minValue":"201","maxValue":"500"}}]}},"pbg_featured_image_src":{"full":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png",2000,1000,false],"thumbnail":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-150x150.png",150,150,true],"medium":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-300x150.png",300,150,true],"medium_large":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-768x384.png",768,384,true],"large":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-1024x512.png",1024,512,true],"1536x1536":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-1536x768.png",1536,768,true],"2048x2048":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png",2000,1000,false],"et-pb-post-main-image":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-400x250.png",400,250,true],"et-pb-post-main-image-fullwidth":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-1080x675.png",1080,675,true],"et-pb-portfolio-image":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-400x284.png",400,284,true],"et-pb-portfolio-module-image":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-510x382.png",510,382,true],"et-pb-portfolio-image-single":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-1080x540.png",1080,540,true],"et-pb-gallery-module-image-portrait":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-400x516.png",400,516,true],"et-pb-post-main-image-fullwidth-large":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png",2000,1000,false],"et-pb-image--responsive--desktop":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-1280x640.png",1280,640,true],"et-pb-image--responsive--tablet":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-980x490.png",980,490,true],"et-pb-image--responsive--phone":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series-480x240.png",480,240,true],"crp_thumbnail":["https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2026\/04\/AI-in-action-series.png",150,75,false]},"pbg_author_info":{"display_name":"Gergana Petkova","author_link":"https:\/\/gws-sso-test.graphwise.ai\/de\/author\/petkovag\/","author_img":"<img alt='Vladimir Alexiev' src='https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2025\/08\/Vadimir-Alexiev.jpeg' srcset='https:\/\/gws-sso-test.graphwise.ai\/wp-content\/uploads\/2025\/08\/Vadimir-Alexiev.jpeg 2x' class='avatar avatar-128 photo avatar-default' height='128' width='128' decoding='async'\/>"},"pbg_comment_info":" No Comments","pbg_excerpt":"Read about how Talk2PowerSystem enables power system engineers to query complex grid data using natural language.","_links":{"self":[{"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/blog-post\/248703","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/blog-post"}],"about":[{"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/types\/blog-post"}],"author":[{"embeddable":true,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/users\/10"}],"version-history":[{"count":17,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/blog-post\/248703\/revisions"}],"predecessor-version":[{"id":249390,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/blog-post\/248703\/revisions\/249390"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/media\/248712"}],"wp:attachment":[{"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/media?parent=248703"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/categories?post=248703"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/tags?post=248703"},{"taxonomy":"persona","embeddable":true,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/persona?post=248703"},{"taxonomy":"resource-category","embeddable":true,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/resource-category?post=248703"},{"taxonomy":"blog-category","embeddable":true,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/blog-category?post=248703"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/gws-sso-test.graphwise.ai\/de\/wp-json\/wp\/v2\/ppma_author?post=248703"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}