Artificial intelligence for interconnect queue management

Abstract

Methods, systems, and apparatus, including non-transitory computer readable media, for managing interconnection requests.

Description

[0001] Attorney Docket No. 43374-0925WO1

[0002] ARTIFICIAL INTELLIGENCE FOR INTERCONNECT QUEUE MANAGEMENT

[0003] CROSS-REFERENCE TO RELATED APPLICATIONS

[0004] [1] This application claims the benefit of priority to U.S. Provisional App. No.

[0005] 63 / 777,483, filed on March 25th, 2025, to U.S. Provisional App. No. 63 / 777,497, filed on March 25th, 2025, to U.S. Provisional App. No. 63 / 934,132, filed on December 8th, 2025, and to U.S. Provisional App. No. 63 / 934,267, filed on December 8th, 2025, the contents of which are all hereby incorporated by reference.

[0006] TECHNICAL FIELD

[0007] [2] This specification relates to electrical power grids, and specifically management of interconnect queues to validate proposed power grid interconnections.

[0008] BACKGROUND

[0009] [3] Electrical power grids transmit electrical power to loads such as residential and commercial buildings. An interconnection to an electrical power grid can be a generation resource, and may be a renewable energy source. An interconnection to an electrical power grid can also be a new load, such as a new building.

[0010] [4] Adding interconnections to the electrical power grid can affect conditions of the power grid, for example, by adding capacity, including clean and renewable energy sources, such as solar power systems. In addition, since electrical demands evolve, an electrical grid can undergo additions and changes on a continual basis. New buildings, renewable power plants, stationary storage, mobile storage, and expansions to existing buildings, facilities, and loads are some examples of potential changes that can be proposed and made to existing electrical distribution feeders.

[0011] [5] Planned and / or requested additions to a power grid are referred to as an interconnection queue. More succinctly, an interconnection queue is a collection of power generation and transmission projects requesting to connect to the power grid.

[0012] [6] When power developers propose new proj ects to add to a grid, they submit an interconnection request to the independent service operator (ISO) or regional transmission organization (RTO) that maintains the electrical grid to join the interconnection queue. An interconnection queue request would include data and documentation surrounding the project. This would include information such as project type, fuel type, net megawatt, location,Attorney Docket No. 43374-0925WO1

[0013] operating data, synchronization date, point of interconnection, system impact study, builders, suppliers and more. Using this information, the ISO or RTO performs a series of impact studies to assess changes necessary to facilitate the addition to the electrical grid.

[0014] [7] The management of an interconnection queue is a complex task requiring multiple rounds of human labor, review and evaluation. The complexity of the task is also time intensive, which increases costs as projects are delayed. Finally, there is no one standard for evaluating an interconnection request for a particular interconnection queue. This, in turn, increases the complexity and time required for a particular developer that deals with multiple ISO’s and RTO’s.

[0015] SUMMARY

[0016] [8] This specification describes an interconnection queue management system implemented as computer programs on one or more computers in one or more locations that can generate an assessment of project data associated with an interconnection request for an electrical grid. In this specification, an interconnection request refers to a request to modify an existing electrical grid by connecting a new energy project to the electrical grid. In particular, the system can determine whether the interconnection request satisfies electrical criteria, e.g., by validating compatibility with the existing electrical grid, and property rights criteria, e.g., by validating land rights associated with the set of parcels identified in the interconnection request. This can include, for example, verifying whether the connection of the new energy project to the existing electrical grid satisfies compatibility with the existing electrical grid and property rights criteria.

[0017] [9] For example, the system can evaluate the geographic boundaries of parcels of land associated with the proposed energy project, verify land access and right of way from the location of a proposed grid connection point of the existing electrical grid to the specified parcels of the new energy project. The system can generate and provide a geospatial visualization of the proposed energy project that highlights any geographic or legal inconsistencies with the proposed energy project and validation data, thereby facilitating the validation and review of proposed energy projects.

[0018]

[0010] As another example, the system can evaluate one or more electrical parameters specifying connection configurations included in the modification to the electrical grid proposed by the interconnection request using a grid knowledge graph. The grid knowledge graph can encode data regarding the specification and operation of grid elements, e.g., power plants, transmission lines, distribution lines, transformers, circuit breakers, switches, andAttorney Docket No. 43374-0925WO1

[0019] substations, to name a few examples. The system can validate that the proposed modification is not likely to cause operation of the electrical distribution feeder to violate any limits or metrics that exist to ensure safe and reliable operation of the electrical power grid. As a related example, the system can determine a measure of impact of adding interconnections specified by energy project proposals using one or more electrical grid interconnection simulations, and can select a subset of the interconnection requests for review based on the measure of impact.

[0020] [1] The system can determine whether the interconnection request satisfies the electrical criteria and the property rights criteria using a set of one or more interconnection assessment tools. In some cases, the system can rely on a multi-agent based approach managing an interconnection queue to analyze the interconnection request. In this case, the system can use one or more interconnection assessment agents corresponding with the interconnection assessment tools to generate the assessment of the interconnection request. Each interconnection assessment agent can be an artificial intelligence (Al) agent that is specialized to perform a particular task that is associated with an aspect of processing geolocation data, land access, right of way data, electrical data, grid knowledge graph data, etc. to determine the assessment of the interconnection request.

[0021] [2] Other implementations of this aspect include corresponding systems, apparatus, and computer programs, configured to perform the actions of the methods, encoded on computer storage devices.

[0022] [3] Particular embodiments of the subject matter described in this specification can be implemented so as to realize one or more of the following advantages.

[0023] [4] The system provides for a streamlined robust analysis of a large cohort of energy¬ projects by an energy project developer. In particular, the system allows an energy project developer to evaluate multiple energy project proposals in a cohort, or multiple versions of the same energy project across cohorts. Instead of spending significant time reviewing geographical and legal documents, the system can process the project data associated with the energy project proposal using a set of interconnection assessment tools to identify any discrepancies between interconnection request data and ground-truth data and to generate and provide a user interface that can facilitate further review of the energy project.

[0024] [5] More specifically, the system can distribute the complex task of determining the interconnection request assessment to highly-specialized interconnection assessment tools. In particular, the system can validate the parcel data and the electrical data included in an interconnection request using one or more interconnection assessment tools. For example, theAttorney Docket No. 43374-0925WO1

[0025] system can use data ingestion tools, electrical tools, and geospatial tools to validate the interconnection request. The system also provides improvements to the clustering studies performed by grid operators by allowing for identification of groups of interconnection requests based on a quantification of their impact to the grid. In this case, the system can perform a first, less complex electrical simulation to validate the electrical data, before performing a second, more complex and computationally expensive simulation as part of an impact study. Thus, the system can reduce the use of computational resources by filtering out invalid interconnection requests from the impact study.

[0026] [6] The system can orchestrate the generation of different outputs using different interconnection assessment tools that can be combined to yield the project assessment, thereby leveraging the deep and narrow expertise of the specialized assessment agents to more efficiently generate an interconnection request assessment. Moreover, the multi-agent architecture promotes parallelism, e.g., in processing separate tasks that relate to the project assessment in parallel using different assessment agents, or in processing multiple projects in parallel using different assessment agents.

[0027] [7] The system can orchestrate the generation of different outputs using different interconnection assessment tools that can be combined to yield the interconnection request assessment, thereby leveraging the deep and narrow expertise of the specialized interconnection assessment tools to generate a robust interconnection assessment more efficiently. By orchestrating of the interconnection assessment tools to perform highly-specialized tasks in an organized sequence of tasks, the system can avoid redundant processing through a minimal use of the tools in an organized manner, thereby reducing the use of computational resources necessary7for generating the interconnection request assessment relative to an uncoordinated system of geospatial tools or manually -coordinated interconnection assessment tools. Furthermore, since the interconnection assessment tools are designed specifically for a particular task and not to perform many tasks, the tools can implement each task in accordance with an approximate computational load of efficiently performing the task, e.g., less computationally resource-intensive tasks using less-resource intensive solutions and more computationally resource-intensive tasks using more-resource intensive solutions, thereby precluding the excess use of computational resources by using a resource-intensive tool for every7task.

[0028] [8] In some implementations, the system can use one or more interconnection assessment agents that are associated with and are experts in querying one or more interconnection assessment tools in the suite of interconnection assessment tools to generateAttorney Docket No. 43374-0925WO1

[0029] the interconnection request assessment. More specifically, the system can be implemented as an agentic system that includes a dynamically configurable selection of interconnection assessment tools responsive to an agentic query in which an interconnection assessment agent is specialized to determine an ordered selection of a set of tools according to the agentic query' and to generate customized instructions to generate the output. In this case, the system can orchestrate the generation of the different outputs using the interconnection assessment agents. Using different interconnection assessment tools to generate the interconnection assessment promotes parallelism in processing separate tasks that relate to the interconnection assessment in parallel. This can reduce the overall processing required by a single entity, such as a single processing-unit, and promotes parallelism in orchestrating separate and / or concurrent tasks to respond to the request.

[0030] [9] Furthermore, the distributed multi-agent architecture is modular, scalable, and provides for enhanced data security. The multi-agent architecture supports the adding of new interconnection assessment tools / agents and the modification of tools / agents without disrupting the generation of the project assessment. Additionally, the domain-specific specialized tools can be sandboxed from each other to limit access to sensitive data based on domain, which can reduce risk of cross-domain data leakage. For example, different interconnection assessment tools can have respective permissions which can be applied per interconnection assessment tool, or in the case of a multi-agent framework, per agent.

[0031]

[0010] By relying on the interconnection assessment agents to perform highly-specialized tasks using respective interconnection assessment tools, the system can reduce the use of computational resources necessary' for generating the project assessment relative to a single agent with broad, but shallow knowledge of a variety of project assessment tasks.

[0032] Furthermore, by combining different outputs of respective interconnection assessment tools to yield the project assessment the system can provide a more robust and accurate project assessment than would be generated using a single interconnection assessment tool.

[0033]

[0011] Additionally, the assessment agent can access interconnection assessment tools that leverage dozens or hundreds of data layers of the large scale areas. The data of the data layer can be updated on a more regular basis, e.g.. weekly, biweekly, daily, etc., for efficient integration of newly available data, which in turn improves the accuracy and sensitivity of the outputs of the system. Moreover, the system can use the electrical data of the data layer to generate a grid knowledge graph encoding domain knowledge of a power systems engineer as a structured representation of each of the grid elements in the electrical grid and the relationships among the grid elements. In particular, by aggregating the electrical data into aAttorney Docket No. 43374-0925WO1

[0034] grid knowledge graph, the system can enhance the standardization and repeatability' of generating an interconnection request assessment for an interconnection request.

[0035]

[0012] Furthermore, a user of the platform is abstracted away from the interconnection assessment tools, and does not need to know which tools are available, e.g., does not need to have domain-specific knowledge of the interconnection assessment tools, their individual capacity. This can reduce the bar of entry for tool usage. Moreover, the system promotes deep interaction with the tool with minimal understanding of the underlying tools by generating and providing a user interface that a user can use to interact wi th the interconnection assessment tools by way of the assessment agents with respect to the current assessment. For example, the user interface provides a visualization of the project assessment, but can include an input portion that allows a user to update the project assessment.

[0036]

[0013] The details of one or more embodiments of the subject matter described in this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the invention will become apparent from the description, the drawings, and the claims.

[0037] BRIEF DESCRIPTION OF THE DRAWINGS

[0038]

[0014] FIG. 1 is a system diagram of an interconnection queue management system.

[0039]

[0015] FIG. 2 illustrates an example of an agentic platform that can implement an interconnection queue management system.

[0040]

[0016] FIG. 3 depicts an example impact similarity engine that can determine a respective measure of impact to the electrical grid for one or more interconnection requests and can select interconnection requests for prioritization based on the measure of impact.

[0041]

[0017] FIG. 4 is an example user interface of an example geospatial visualization that includes a depiction of an aggregated geographic boundary of the set of parcels included in an interconnection request.

[0042]

[0018] FIG. 5 is a flow diagram of an example process for generating an impact assessment for an interconnection request.

[0043]

[0019] FIG. 6 is a flow diagram of an example process for generating a regulatory assessment of the interconnection request.

[0044]

[0020] FIG. 7 is a flow diagram of an example process for selecting one or more interconnection requests for prioritization based on the respective measures of impact for each interconnection request.Attorney Docket No. 43374-0925WO1

[0045]

[0021] Like reference numbers and designations in the various drawings indicate like elements.

[0046] DETAILED DESCRIPTION

[0047]

[0022] FIG. 1 is a system diagram of an interconnection queue management system. The interconnection queue management system 100 is an example of a system implemented as computer programs on one or more computers in one or more locations in which the systems, components, and techniques described below are implemented.

[0048]

[0023] The system 100 can generate an interconnection request assessment 170 of an interconnection request 110 for an electrical grid. For example, the sy stem 100 can receive an interconnection request 110 that seeks to modify the electrical grid to connect with a new energy project, e g., from a submitting user at client device 104. As an example, the submitting user can submit an interconnection request 110 to connect a new clean energy generator, e.g., a solar, wind, or gas project, to the electrical grid. As another example, the submitting user can submit an interconnection request 110 to expand the transmission network of the electrical grid.

[0049]

[0024] The interconnection request 110 can include parcel data 112 and electrical data 114. The parcel data 112 can include an identification of the set of parcels of land that require site control to actuate the interconnection request 110. In this context, site control can refer to land use rights, including exclusive land use or land access rights, associated with the parcels. The electrical data 114 can specify' the modification to the electrical grid, e g., as one or more connection configurations that each specify a respective modification to a specific element of the electrical grid. For example, the grid elements can include power plants, transmission lines, distribution lines, transformers, circuit breakers, switches, and substations.

[0050]

[0025] More specifically, the system 100 can identify any missing, incomplete or inconsistent information provided in the interconnection request 110 relative to ground truth parcel data and electrical specifications of the electric grid. In particular, the system 100 can validate the land rights associated with the set of parcels in the parcel data 112 and can validate compatibility of the electrical data 114 with the electrical grid using one or more interconnection assessment tools. Each interconnection assessment tool can be tailored to perform a highly -specialized interconnection assessment task. For example, the interconnection assessment tools can include data ingestion tools 130, geospatial tools, electrical tools 150, and planning tools 160, which are described in more detail below.Attorney Docket No. 43374-0925WO1

[0051]

[0026] The system 100 can coordinate the use of the different interconnection assessment tools to generate the interconnection request assessment 170. In the particular example depicted, the interconnection assessment tools correspond with one or more interconnection assessment agent(s) 120. In this case, the interconnection assessment agents 120 can be interconnection reasoning agents, e.g., an artificial intelligence agent that can understand and analyze parcel data 112 and electrical data 114 of an interconnection request 110 to generate an interconnection request assessment 170.

[0052]

[0027] For example, the interconnection assessment agents 120 can be implemented using machine learning models, such as generative neural networks or other machine learning models, that can be trained or finetuned to perform the operations described below. In the case that one or more of the interconnection assessment agents 120 are implemented as generative neural networks, the generative neural network can have a recurrent neural network architecture that is configured to sequentially process the contents of an input question to generate an output. For example, the generative neural network can be a recurrent neural network (RNN), long short-term memory (LSTM). or gated-recurrent unit (GRU). As another example, the generative neural network can be transformer-based, e.g., an encoderdecoder transformer, an encoder-only transformer, or a decoder-only transformer, configured to perform parallel processing of the contents of an input using a multi-headed attention mechanism and trained to perform next element prediction, e.g., to define a likelihood score distribution over a set of next elements.

[0053]

[0028] As a particular example, one or more of the interconnection assessment agents 120 can be a language processing neural networks, e.g., a foundation model such as a transformer large language model. Large language models have been demonstrated to achieve state of the art performance in semantic understanding, e.g.. their ability to effectively capture semantic information from inputs.

[0054]

[0029] In some cases, the system 100 can use one or more interconnection assessment agents 120 that correspond with each tool, e.g., in a one agent-to-one tool mapping or a one agent-to-many tool mapping, to generate the interconnection request assessment 170. For example, each interconnection assessment tool can be associated with one or more agents that are specialized in querying the geospatial tool that each agent corresponds with.

[0055]

[0030] While described below in terms of the system 100 performing (i) a parcel validation task and (ii) an electrical data validation task to generate the interconnection request assessment 170 using a variety of different interconnection assessment tools, it is understood that the system 100 can perform each task described below in parallel orAttorney Docket No. 43374-0925WO1

[0056] sequentially. It is also understood that the system 100 can coordinate the use of the interconnection assessment tools directly or use the interconnection assessment agents 120 to coordinate use of the different interconnection assessment tools. As apparent from the interconnection assessment management platform implementation described in more detail with respect to FIG. 2, the system 100 can use the interconnection assessment agents 120 to (i) provide inputs to the one or more interconnection assessment tools, (i) generate instructions that cause execution of the one or more interconnection assessment tool to perform a sub-task, and (iii) provide outputs of the one or more interconnection assessment tools to downstream interconnection tools for further processing.

[0057]

[0031] The description below relates to performing a parcel validation task as part of generating an interconnection request assessment 170.

[0058]

[0032] For example, the system 100 can apply one or more data ingestion tools 130 to the interconnection request 110 to extract the identification of the set of parcels from the parcel data 112. Generally, the parcel data 112 can include a set of legal documents, e.g., leases, amendments, extensions, deeds, etc., that are associated with different parcels specified in the interconnection request 110. Generally, the set of documents include word-processing file, PDF documents, and, in some cases, comma-separated value files. In particular, the system 100 can use one or more document extraction tool(s) 132 to extract text-based location identifiers, e.g., Parcel IDs, addresses, latitude-longitude coordinates, and property right identifiers, e.g., exclusive use, lease, and term, from the set of documents included in the parcel data 112. The system 100 can also use the document extraction tool(s) 132 to extract metadata, including a unique identifier, from the interconnection request 110, the type of document the data was extracted from, and submitter information.

[0059]

[0033] For example, the data ingestion tools 130 can implement PDF and wordprocessing optical character recognition techniques to extract the text-based location and property right identifiers from the set of legal documents. As another example, the data ingestion tools 130 can process the set of documents using a language processing neural network with instructions to extract the text-based location and property' right identifiers. In this case, the system 100 can prompt the language processing neural network to provide an output constrained by a schema with expected fields, e g., as a structured data file, e.g., JSON, XML, or CSV file. As an example, the expected fields can be formatted as key-value pairs, where the value is empty when processed by the language processing neural network. For example, the expected fields can include a parcel identifier field or an address field that specifies a unique identifier for each parcel, a conveyance agreement identifier field thatAttorney Docket No. 43374-0925WO1

[0060] specifies whether the parcel data 112 indicates that a property rights conveyance has been made for the parcel, and one or more term date fields that specify the length of time associated with the property rights conveyed for the parcel.

[0061]

[0034] The system 100 can use the text-based location identifiers to access one or more external datasources, e.g., aparcel database 115 and a property rights database 118, to obtain ground truth data using one or more dataset access tools 134. For example, the sy stem 100 can use an interconnection assessment agent 170 to generate instructions to cause the dataset access tool 134 to obtain ground-truth parcel data from the parcel database 115 and property rights data from the property rights database 118. In the case that the location-based identifier is an address instead of a parcel identifier, the dataset access tools 134 can identify the parcel identifiers corresponding with the address using the parcel database 115.

[0062]

[0035] The parcel database 115 can maintain parcel data that specifies (i) an identifier of a parcel, and (ii) parcel metadata including one or more latitude-longitude coordinates and size, e.g., boundary and acreage, corresponding with the parcel. For example, the one or more latitude-longitude coordinates can specify the defined boundaries of the parcel, e g., as Geographic Information System Specialist (G1SS) data. In some cases, the parcel metadata can also include an indicator of whether the parcel is a public parcel, e.g., a publicly-owned parcel, or private parcel, e.g., a privately-owned parcel. In some cases, the parcel data can include an indication of whether any buildings or infrastructure, e.g., schools or hospitals, are located on the parcel.

[0063]

[0036] The property rights database 118 can include any legal documents, e.g., deeds, leases, easements, restrictions, etc. for a number of parcels. In some cases, the property rights database 118 can be sourced from a county survey or’s office. In other cases, the property rights database can be sourced from the grid operator, or a partner entity of the grid operator.

[0064]

[0037] The system 100 can use the dataset access tools 134 to obtain ground truth data for validating the parcel data 112 included in the interconnection request 110. In particular, the dataset access tools 134 can obtain ground-truth parcel data from the parcel database 115 and ground-truth property rights data from the property rights database 118 corresponding with the extracted text-based location identifiers, e.g.. the parcel identifiers, identified by the document extraction tools 132 for the parcel data 112. For example, the dataset access tools 134 can use the extracted text-based location identifiers can be used as a primary key in a table-based database to identify corresponding ground-truth parcel data and ground-truth property rights data for each parcel.Attorney Docket No. 43374-0925WO1

[0065]

[0038] The system 100 can determine whether there are any discrepancies between the parcel data 112 and the ground-truth parcel data and ground-truth property rights obtained from the databases 115, 118 data using one or more geospatial tools 140. For example, the system 100 can use an interconnection assessment agent 170 to generate instructions to cause the geospatial tools 140 to validate the parcel data 112.

[0066]

[0039] In the particular example depicted the geospatial tools 140 include parcel and boundary aggregator tool(s) 142 and land use validation tool(s) 144. More specifically, the system 100 can use the parcel and boundary aggregator tool(s) to determine whether the locations, boundaries, and acreage of the parcels specified in the parcel data 112 are accurate relative to the ground-truth parcel data, and the system 100 can use the land use validation tool(s) 144 to determine whether the legal rights to the parcels specified in the parcel data 112 convey the authorized use of the parcels for the interconnection request 110 based on the ground-truth property rights data.

[0067]

[0040] For example, the parcel and boundary7aggregator tool(s) 142 can determine whether the identification of the set of parcels in the parcel data 112 aligns with the identification of the set of parcels in the ground-truth parcel data. In particular, the parcel and boundary aggregator tools(s) 142 can, based on the extracted parcel identifiers, identify parcels specified by the parcel data 112 that do not match the location, boundaries, or acreage of parcels specified by the ground parcel data. In some cases, the dataset access tool 134 can generate the boundary outline of the parcel from the specified latitude-longitude coordinates of the parcel data 112 and the ground-truth parcel data.

[0068]

[0041] As another example, the parcel and boundary aggregator tools 142 can also aggregate the boundaries of the set of parcels specified by the parcel data 112 and the set of parcels specified by the ground-truth parcel data, and can determine whether a discrepancy exists between the aggregated boundaries. In some cases, the parcel and boundary aggregator tools 142 can generate polygon data that represents the aggregated boundaries of the set of parcels specified by the parcel data 112 and the ground-truth parcel data. In particular, the parcel and boundary aggregator tools 142 can aggregate the boundaries of the set of parcels in both the parcel data 112 and the ground-truth parcel data using an iterative union of parcel boundaries to generate the polygon data. For example, the parcel and boundary aggregator tools 142 can iterate through pairs of parcels from the set of parcels to determine whether the pair of parcels is contiguous, and, in the case that the pair of parcels is contiguous, can merge the boundaries of the pair of parcels as the polygon data. In this case, the parcel and boundaryAttorney Docket No. 43374-0925WO1

[0069] aggregator tools 142 can perform a geometric union to identify the gaps between the polygon data of the parcel data 112 and the ground- truth polygon data.

[0070]

[0042] As yet another example, the land use validation tool(s) 144 can determine whether the term, exclusivity, and conveyance specified by the parcel data 112 for the interconnection request 110 aligns with the ground-truth term, exclusivity7, and conveyance specified by the ground-truth parcel data and the ground-truth property rights data. In this context, term refers to the length of time of property rights, exclusivity refers to the strength of the land rights, e.g., rights to exclusive use, partial use through an easement, or lack of exclusive use through a restriction, and conveyance refers to whether the property7rights have been conveyed to the submitting user for the parcels specified by the parcel data. For example, the land use validation tool(s) 144 can identify that an interconnection request includes parcels that have right of way land use rights, but do not have exclusive control.

[0071]

[0043] In more detail, the land use validation tool(s) 144 can determine whether the property7rights specified for the parcels in the parcel data 112 comply with the land use rights specified by the ground-truth property data. In particular, the land use validation tool(s) 144 can process the property rights data extracted from the set of documents included in the parcel data 112 and the ground-truth property rights data to generate an output that includes an identification of any missing or inconsistent property7rights for each parcel based on the term, exclusivity, and conveyance. In some cases, the land use validation tool(s) 144 can produce natural language descriptions that explain the reasoning of the missing or inconsistent property right determination and provide a citation identify ing the source of the ground-truth data that the parcel data 112 was inconsistent with.

[0072]

[0044] For example, the land use validation tool(s) 144 can identify groupings of one or more documents from the set of documents included in the parcel data 112 that relate to each parcel. The land use validation tool(s) 144 can associate the ground-truth property rights data corresponding with the one or more documents related to each parcel in a single data structure, e.g., an agreements object, to simplify the property7rights validation for each parcel. The land use validation tool(s) 144 can then process the groupings based on the term, exclusivity, and conveyance, e.g., serially. In particular, the tool(s) 144 can first validate the term, then the exclusivity, and finally the conveyance, or any other order. In some cases, by validating the term, exclusivity7, and conveyance serially, the system 100 can filter out parcels that do not satisfy the previously evaluated criteria from further processing, thereby reducing the use of computational resources relative to evaluating the criteria even for parcels that do not meet one or more of term, exclusivity, or conveyance requirements.Attorney Docket No. 43374-0925WO1

[0073]

[0045] In the particular example depicted, the land use validation tool(s) 144 can additionally process requirements of the grid operator 135 to validate the property rights of the parcel data 112 with respect to the required term, exclusivity, and conveyance rights of the grid operator. For example, the land use validation tool(s) 144 can determine whether strict land use conveyance is required to validate the property' rights data of interconnection request 110, or whether the grid operator only expects land use rights obtained for N%, e.g., 65%. 75%. 80%. of the acreage specified by the set of parcels in the interconnection request 1 10. In particular, the minimum acreage can be specified by operator based on the type of power generation associated with the interconnection request 110. As another example, the land use validation tool(s) 144 can determine whether the term for each of the parcels satisfies the minimum term required by the grid operator. As yet another example, the land use validation tool(s) 146 can determine whether the parcels satisfy an exclusive use requirement or meet a specification that only N%, e.g., 5-20%, of the acreage specified by the set of parcels is associated with an easement for a right of way. As a further example, the land use validation tool(s) 146 can determine whether a threshold percentage of acreage associated with the parcels satisfies two or more of the required term, exclusivity, and conveyance rights of the grid operator.

[0074]

[0046] In some implementations, the operations of the data ingestion tools 130 relative to processing the parcel data 112 and the geospatial tools 140 can be implemented using chain-of-thought prompting of a language processing neural network. In this case, the geospatial tools 140 can decompose the complex task of extracting the location-based and propertyrights identifiers from the parcel data, and validating the location, boundary', and land use rights of the parcel data 112 included in the interconnection request 110 into a sequence of related sub-tasks that the language processing neural network can consecutively perform to complete the parcel validation. As described above, the system 100 can prompt the language processing neural netw ork to generate outputs in accordance with a defined schema of a structured data file, e.g., to allow' for the system 100 to iteratively prompt the language processing neural network using previously generated outputs.

[0075]

[0047] For example, the language processing neural network can be implemented using Gemini, e g., as is described in “Gemini: A Family of Highly Capable Multimodal Models’’ (arXiv:2312.11805v5) and “Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities’’

[0076] (arXiv:2507.06261). In some cases, the language processing neural network can have beenAttorney Docket No. 43374-0925WO1

[0077] trained or finetuned from a foundational model using a set of training data for a number of interconnection requests.

[0078]

[0048] In particular, the system, or a training system, can obtain an input and a target output for each of a number of steps in the sequence of related sub-tasks that are involved in completing parcel validation. In this case, the system 100 can process each input using the language processing neural network, and can determine the discrepancies between the generated outputs and the corresponding target output for each step in the sequence. The system 100 can then update one or more of the set of parameters of the language processing neural network based on the discrepancies, e.g., using the update rule of any appropriate gradient descent optimization algorithm, e.g., RMSprop, or Adam. Furthermore, in this case, the system can be further tailored using online learning based on user provided feedback regarding the interconnection request assessment 170.

[0079]

[0049] For example, the system 100 can prompt the language processing neural network to: (i) extract the location-based and property -rights identifiers from the parcel data 112, (ii) identify corresponding ground truth data from the parcel database 115 and the property rights database 118, (iii) determine whether the locations, boundaries, and acreage of the parcels specified in the parcel data 112 are accurate relative to the ground-truth parcel data, (iv) to determine whether a discrepancy exists between an aggregated boundary7of the parcel data 112 and an aggregated boundary of the ground-truth parcel data, (v) to determine whether the terms of the property rights to the parcels specified in the parcel data 112 match the groundtruth terms and satisfy the requirements of the grid operator 134, (vi) to determine whether the exclusivity of the property rights to the parcels specified in the parcel data 112 match the ground-truth exclusivity and satisfy the requirements of the grid operator 134, and (vii) to determine whether the conveyance of the property rights to the parcels specified in the parcel data 112 match the ground-truth conveyance and satisfy the requirements of the grid operator 134.

[0080]

[0050] As an example, when determining whether the legal rights to the parcels specified in the parcel data 112 convey the authorized use of the parcels for the interconnection request 110 based on the ground-truth property rights data, the system 100 can further prompt the language processing neural network to identify and group legal documents by parcel in both the parcel data 112 and the ground-truth property data and to determine, using the grouped documents, whether the term, exclusivity7, and conveyance requirements are met for each parcel. In this case, the language processing neural network can use context from the set ofAttorney Docket No. 43374-0925WO1

[0081] documents to group documents into agreement obj ects for each parcel based on a set of guidelines.

[0082]

[0051] As a further example, in the case that a parcel has a structure on it, the land use validation tool(s) 144 can detect whether the structure falls in between the generation site and the point of interconnection. For example, the land use validation tool(s) 144 can determine a generator tie distance vector between an energy generation site and the point of interconnection, and can determine whether the distance vector intersects the location of an existing structure.

[0083]

[0052] In the particular example depicted, the geospatial tools 140 can provide the aggregated boundary data, the identification of whether the parcels match the location or boundaries of the ground-truth parcels, and the indication of whether use of each parcel is authorized based on the ground-truth property rights to a user interface (UI) generator 145 to generate a geospatial visualization UI 108 that can be provided to a client device 104 of a user 102. For example, the geospatial visualization UI 108 can overlay a visualization of the submitted parcel data 112 with the ground-truth data to highlight any discrepancies, e.g., for further evaluation by a user 102. An example geospatial visualization UI is depicted in FIG.

[0084] 4.

[0085]

[0053] The system 100 can provide a parcel assessment 174 to a client device 104 of a user 102 based on the validation. In the case that the geospatial tools 120 validate that the parcel data 112, e.g.. the location, boundaries, and aggregated boundaries and the property rights, the system 100 can provide a positive parcel assessment 174 for the interconnection request 110 in the interconnection request assessment 170. In the case that the geospatial tools do not validate the parcel data 112, e.g., because of discrepancies present in one or more of the location, boundaries, aggregated boundaries, or property rights specified by the parcel data 112 relative to the ground-truth parcel data and the ground-truth property rights data, the system 100 can provide a negative parcel assessment 174 for the interconnection request 110 in the interconnection request assessment 170. Generally, the system 100 can identify the nature of the discrepancy in the negative parcel assessment, e.g., to provide an identification of one or more requested corrections that can be transmitted to the submitter of the interconnection request 110 with a request to rectify the interconnection request.

[0086]

[0054] The description below relates to performing an electrical validation task as part of generating an interconnection request assessment 170.

[0087]

[0055] For example, the system 100 can apply one or more of data ingestion tools 130 to the interconnection request 110 to extract one or more electrical parameters from theAttorney Docket No. 43374-0925WO1

[0088] electrical data 114. Generally, the electrical data 114 can be included in a set of technical documents included in the interconnection request 110 that specify proposed points of interconnection, operating behavior, and results of any preliminary simulations run for the electrical data. In particular, the document extraction tool(s) 132 can process the electrical data 114 to extract one or more electrical parameters for each connection configuration specified in the interconnection request that correspond with one or more electrical parameters of the grid knowledge graph 155 to facilitate comparison. Additionally, the data ingestion tools 130 can preprocess the data to ensure that the format of the one or more electrical parameters aligns with the expected format of input parameters to one or more electrical simulations.

[0089]

[0056] The system 100 can process the electrical parameters extracted from the electrical data 114 and a grid knowledge graph 155 using one or more electrical tools 150 to determine whether the electrical data 114 satisfies one or more criteria in accordance with the grid knowledge graph 155. For example, the sy stem 100 can use an interconnection assessment agent 170 to generate instructions to cause the electrical tools 150 to validate the electrical data 114. In particular, the grid knowledge graph 155 includes a repository of information about the electrical grid including interconnection-specific grid requirements such as technical specifications, grid modeling parameters, electrical diagrams, equipment certifications, applicable regulations, site control documentation, and other information the electrical data 114 is expected to comply with, e.g., based on the one or more criteria, to be a feasible interconnection request.

[0090]

[0057] The grid knowledge graph 155 is a structured representation of grid element entities, attributes, and relationships within the electricity grid infrastructure, e.g., an electricity sector-specific application of a knowledge graph. That is, the grid knowledge graph can encode domain knowledge of a power systems engineer relative to each of the elements of the electrical grid as well as the connections betw een the elements. More specifically, the grid knowledge graph 155 can include (i) a number of nodes representing each element of the electrical grid, e.g., power plants, transmission lines, distribution lines, transformers, circuit breakers, switches, and substations of the electrical grid , and (ii) a number of edges connecting a pair of nodes of the number of nodes and representing a connection between the elements of the electrical grid represented by the pair of nodes.

[0091]

[0058] In some cases, the system 100 can generate the grid knowledge graph 155, e.g., using one or more interconnection assessment agents to convert grid requirements for interconnection and other relevant information into standard formats, a considerableAttorney Docket No. 43374-0925WO1

[0092] challenge given the heterogeneous, siloed, and at-times unstructured data that characterizes the electrical grid. In other cases, the system 100 can obtain the grid knowledge graph, e.g., as an input, or from a grid knowledge graph database (not depicted).

[0093]

[0059] For example, the electrical validation tool(s) 150 can validate the electrical data 11 4 by processing the one or more electrical parameters using the grid knowledge graph 155 to determine whether the one or more criteria specified by the grid knowledge graph 155 are met. In more detail, the electrical validation tool(s) 150 can validate that the proposed changes specified in the interconnection request are not likely to cause operation of the electrical distribution feeder to violate any limits or metrics that exist to ensure safe and reliable operation of the electrical power grid.

[0094]

[0060] In particular, the electrical validation tool(s) 150 can process the one or more electrical parameters against the structured relationships between grid elements represented by the grid knowledge graph 155. For example, the electrical validation tool(s) 150 can evaluate the electrical parameters for consistency within the interconnection request 110, e.g., the electrical data provided within a single-line diagram included in the interconnection request 110 should be consistent with the connection configurations of the interconnection request 110. As another example, the electrical validation tool(s) 150 can validate the electrical parameters in accordance with expected ranges defined by the grid operator for the project scale, e.g., based on the relative position of the proposed point of interconnection in the grid topology. In some cases, the expected ranges can be provided as an input to the system. As another example, the electrical validation tool(s) 150 can identify any missing data fields based on the electrical parameters.

[0095]

[0061] In some implementations , the system 100 can traverse the structured relationships of the grid knowledge graph 155 to validate each connection configuration specified by the electrical data 114 against each impacted grid element based on the topology of the grid knowledge graph 155. In particular, the system 100 can rely on the grid knowledge graph 155 as a multi-dimensional model that encodes the topology of the electrical grid by defining by the relational dependencies among grid elements . The system 100 can process the one or more electrical parameters by mapping them onto the grid knowledge graph 155 to verify that specific connection configurations adhere or follow the electrical constraints defined by one or more neighboring grid elements. As a result, the system 100 can ensure that each parameter is compatible with the structured relationships of the upstream and downstream grid elements.Attorney Docket No. 43374-0925WO1

[0096]

[0062] As another example, the electrical simulation tool(s) 152 can execute one or more simulations for the electrical data 114 by processing the one or more electrical parameters to simulate electrical grid operation with the proposed interconnection. Since the electrical simulation tool(s) 152 are generally computationally expensive, in some implementations, the system 100 can reduce the use of computational resources by validating the electrical data using the grid knowledge graph 155 prior to using the electrical simulation tool(s) 152. More specifically, the electrical validation tools 150 can verify that the one or more parameters include the input fields for executing the simulation.

[0097]

[0063] For example, the one or more electrical simulation tools 152 can include a load flow or power flow simulation tool to evaluate the steady -state performance of the modified electrical grid after integrating the connection configurations specified by the electrical data 114. As another example, the electrical simulation tools 152 can include an electrical overload simulation tool to evaluate stress to the modified electrical grid in the event that the connection configurations specified by the electrical data 114 draw more current than the new grid elements are designed to handle. As yet another example, the one or more grid simulation tools 152 can include a dynamic stability simulation tool to ensure that the voltage or power across each grid element remains stable, e g., within a tolerance range specified for each grid element, after integrating the connection configurations specified by the electrical data 114 in the modified electrical grid. As a further example, the one or more grid simulation tools 152 can include a generator fault and circuit breaker response simulation or a short circuit response simulation to ensure that protections remain coordinated after integrating the connection configurations specified by the electrical data 114.

[0098]

[0064] The system 100 can generate a measure of impact for the interconnection request 110 based on the simulation results using one or more impact assessment tool(s) 156. In addition to evaluating each individual proposed connection configuration, the order in which electrical power grid alterations occur, including adding connection configurations, can substantially impact grid operation. For example, an interconnection request that includes both a solar power connection configuration and an upgrade to a feeder connection configuration can enable subsequent connections of additional clean energy’ sources, while adding only the clean energy' sources without performing the feeder upgrade could impact grid reliability.

[0099]

[0065] More specifically, the system can determine the measure of impact caused by the modification to the electrical grid specified by the interconnection request as a set of data that describes the impact to the grid. Electrical power grids are quite complex, and adding anAttorney Docket No. 43374-0925WO1

[0100] interconnection can cause unexpected or unintended results, e.g., to parts of the electrical grid that are removed from the modifications. In this context, the measure of impact caused by the modification to the electrical grid can quantify whether operations of the electrical grid given the interconnection request satisfy safety and operational constraints over the entire electrical grid.

[0101]

[0066] In more detail, the impact assessment tool(s) 156 can determine the impact of adding interconnections using one or more simulations to reduce the likelihood that adverse situations arise. Generally, the simulations involved in impact assessment are more comprehensive than the simulations involved in validating the electrical data. In some cases, the set of data of the impact assessment includes an evaluation of the electrical data 114 relative to multiple impact metrics over a range of conditions using an influence graph that characterizes contingencies among grid elements of the electrical grid, e.g., the likelihood that one grid element of the electrical grid will affect another component.

[0102]

[0067] The system 100 can evaluate the impact of the interconnection request 110 to each part, e.g.. grid element, of the electrical grid using the contingencies. Based on the simulation results for each of the contingencies, the impact assessment tool(s) 156 can evaluate a number of metrics, e g., voltage constraint violations, voltage variability, voltage transients, thermal limits, backfeed constraints, capacity constraints, etc., as the set of data. As an example, when outputting a pass output, the impact assessment tool(s) 156 can provide margins to operating limits. As another example, when outputting a fail output, the impact assessment tool(s) 156 can provide the specific failing factors, the timing, frequency, and duration of the failing conditions, and the locations of the failure or failures. The impact assessment tool(s) 156 can then compare the simulation results to determine which portions of the electrical grid would be most impacted by the interconnection request 110. An example for using an influence graph to determine the measure of impact for the interconnection request 110 will be described in more detail with respect to FIG. 3.

[0103]

[0068] In some implementations, the impact assessment tool(s) 156 can additionally determine aspects of the electrical grid that should be upgraded to support one or more new interconnections. The recommended changes identified by the impact assessment tool(s) 156 can include, for example, curtailment, rebuild of electrical assets, addition of storage, voltage controls, amendment of operational parameters, equipment sizing, protection schemes, etc.

[0104]

[0069] The system 100 can provide an impact assessment 172 to a client device 104 of a user 102 based on the validation. In the case that the electrical tools 150 validate that the electrical data 114 satisfies the one or more criteria of the grid knowledge graph 155 and thatAttorney Docket No. 43374-0925WO1

[0105] results of electrical simulations for the electrical data 114 satisfy safety and regulatory- constraints, the system 100 can provide a positive impact assessment 174 for the interconnection request 110 in the interconnection request assessment 170. In the case that the electrical tools 150 do not validate the electrical data 114, e.g., because of discrepancies in the electrical data relative to the specification of the electrical grid represented in the grid knowledge graph 156 or simulation results that indicate an unsafe interconnection request, the system 100 can provide a negative impact assessment 172 for the interconnection request 110 in the interconnection request assessment 170. Generally, the system 100 can identify the reason that the electrical data 114 was not validated in the negative impact assessment 172, e.g., to provide an identification of one or more requested corrections that can be transmitted to the submitter of the interconnection request 110 with a request to rectify the interconnection request.

[0106]

[0070] In response to validating the parcel data 112 and the electrical data 114, the system 100 can use one or more planning tools 160 to store the interconnection request 110 in an interconnection request project database 165. In particular, the system 100 can use the interconnection request project database 165 to maintain validated interconnection requests, e.g., for further analysis. In some cases, the interconnection request project database 165 stores the parcel data 112 and the electrical data 114 of interconnection requests received in the same cohort in respective tables that correspond with different calls for energy projects from the grid operator. In other cases, the interconnection request project database 165 stores the parcel data 112 and the electrical data 114 of interconnection requests in respective tables that correspond with a submitter identifier, e.g., to track multiple versions of interconnection requests within the same cohort.

[0107]

[0071] More specifically, the system 100 can further evaluate the interconnection request 110 relative to other validated interconnection requests obtained for the electrical grid from the database 165 using one or more planning tool(s) 1 0. For example, the system 100 can use an interconnection assessment agent 170 to generate instructions to cause the planning tool(s) 160 to obtain data from the interconnection request project database 165 to facilitate the analysis of interconnection requests received for the same cohort or from the same submitter.

[0108]

[0072] For example, the system 100 can use one or more cohort analysis tool(s) 162 to review the valid interconnection requests obtained in the same cohort. In particular, the cohort analysis tool(s) 162 can provide data from the interconnection request project database 165 to the UI generator 145 to allow a user to visually compare multiple interconnectionAttorney Docket No. 43374-0925WO1

[0109] requests on the geospatial visualization UI 108, thereby facilitating review. An example for comparing multiple interconnection requests in the same cohort using a geospatial visualization UI 108 will be described in more detail with respect to FIG. 4.

[0110]

[0073] As another example, the system 100 can obtain one or more validated interconnection requests that were received from the same submitter to compare the one or more validated interconnection requests using one or more cohort analysis tool(s) 162. In particular, in the case that the system 100 receives the interconnection request 110 as a second, updated submission relative to one or more previous interconnection requests. After submitters make any necessary' changes to their applications to remedy missing, incomplete, or incorrect information, the system can validate the updated interconnection request.

[0111]

[0074] For example, the submitter can have made a revision to a lease or include a document that corresponds with an additional easement in the updated interconnection request. In this case, the system 100 can obtain the other interconnection requests from the submitter and can identify the discrepancy between the parcel data 112 and the ground-truth parcel data, the electrical data 114 and the ground-truth electrical data, or both that require reanalysis using the interconnection assessment tools, thereby precluding the need to re-process and re-analyze the entire interconnection request.

[0112]

[0075] As another example, the system 100 can use one or more impact similarity engine tool(s) 166 to cluster the respective measures of impact determined for each of the valid interconnection requests, e.g., in a cohort, to identify groups of interconnection requests based on their impact to different parts of the electrical grid. In particular, the impact similarity' engine tool(s) 166 can cluster the respective sets of data representing the measure of impact among each valid interconnection request based on the sets of data satisfying a threshold similarity’ distance. In this case, the impact similarity engine tool(s) 166 are associated with an impact similarity engine that can improve cluster studies for gnd operators by clustering interconnection requests based on similarity' of impact, rather than clustering by application date or by some other clustering proxy. An example impact similarity engine will be described in more detail with respect to FIG. 3.

[0113]

[0076] While described as being used to generate an interconnection request assessment 170 for a particular electrical grid managed by a particular electrical grid operator, the system 100 can generate an interconnection request assessment 170 for any electrical grid managed by any electrical grid operator. In particular, the system 100 can be tailored to each electrical grid managed by an electrical grid operator by obtaining the requirements of the grid operator 135 and a grid knowledge graph 155.Attorney Docket No. 43374-0925WO1

[0114]

[0077] FIG. 2 illustrates an example operating environment 200 of an agentic interconnection queue management platform 203 that can implement an interconnection queue assessment system.

[0115]

[0078] The interconnection queue management platform 203 is depicted as a layered system that transforms raw data into interconnection queue management decision support using a suite of interconnection assessment tools. More specifically, the interconnection queue management platform 203 includes a dynamically configurable selection of interconnection assessment tools 220 responsive to an agentic query in which one or more interconnection assessment agents 210 are specialized to determine an ordered selection of a set of tools according to the agentic query and to generate customized instructions to generate the interconnection request assessment as output.

[0116]

[0079] As depicted, the interconnection queue management platform 203 includes multiple layers: an agentic layer 210, an interconnection assessment tool layer 220, a grid knowledge graph layer 230, and a data layer 240. Although depicted as an agentic layer 210, an interconnection assessment tool layer 220, a grid knowledge graph layer 230, and a data layer 208, the operations described with reference to the interconnection queue management platform 203 can be performed by more or fewer layers. Additionally, although operations are described with reference to respective ones of the layers, the operations can be performed by two or more layers, e.g., can be divided between layers.

[0117]

[0080] Interconnection queue management platform 203 is implemented on one or more computers, e.g., one or more cloud-based servers 212, in data communication through a network, e.g., a local area network (LAN), a wide area network (WAN), the Internet, or a combination thereof. The interconnection queue management platform 203 is in data communication with client devices, e.g.. user devices, through which users can interact with the platform 203, e.g., interact with the platform 203 through a geospatial visualization UI provided by the interface agent 212. The network connects the one or more servers, user devices, third-party content servers, and so on. For example, one or more layers of the georeasoning platform can be hosted on respective cloud-based servers.

[0118]

[0081] The agentic interconnection queue management platform 203 can provide developers an Al-based preliminary response to interconnection requests that identifies missing, incomplete, or inconsistent information in interconnection requests, e.g., as described with respect to FIG. 1, and can also allow for planning using validated interconnection requests. In more detail, the platform 203 can facilitate the review of an interconnection request using one or more interconnection assessment agents in the agenticAttorney Docket No. 43374-0925WO1

[0119] layer 210 that can coordinate the use of one or more interconnection assessment tools in the tool layer 220. Generally, the interconnection assessment tools 220 can be implemented as Al tools that are configured interpret and analyze complex sets of structured and unstructured information.

[0120]

[0082] Generally, the artificial intelligence agents 210 are generative machine learning model agents. In some cases, the agents can be language processing model agents, e.g., large-language model agents. In some implementations, the agents of the agentic layer 210 are domain-specific agents, e g., vertically-specialized agents configured to have deep knowledge within a specific, respective domain. An agent of the agentic layer is evaluated using an evaluation set including domain-specific prompts in which feedback is provided back to refine the agent responsive to the comprehensiveness of the agent to respond to the evaluation set. For example, the interconnection assessment tools can include large language model tools, computer vision tools, and retrieval augmented generation (RAG) tools.

[0121]

[0083] As used herein, RAG is a technique that enhances large language models (LLMs) by providing them with external knowledge sources during the generation process. This allows LLMs to access and incorporate information beyond their pre-trained knowledge, improving their accuracy and factual grounding. In some implementations, there is a multimodal RAG in which the RAG tool handles not only with unstructured text data, but also structured text, numeric data, maps (e.g., of the project site), electrical data (e.g., SLD of the project, say a solar farm, or the interconnection being proposed), images (e.g., of specific components) or video.

[0122]

[0084] Generally, the RAG tool can handle requests for different types of facilities, including different generation types (e.g., wind, solar), or storage (e.g., batteries, pumped hydro). Moreover, multiple RAG tools can be specialized for different use cases, e.g., different RAG tools can be tailored to a wind, solar, or energy storage task, each of which may have different and overlapping regulation and interconnection requirements that may differ by city, county, or state geography

[0123]

[0085] In particular, interconnection queue management platform 203 receives user prompts 205. e.g., an interconnection request including a set of files, from one or more client device(s) 214. The one or more client device(s) 214 are an electronic device that are capable of requesting and receiving resources, e.g., virtual environment applications, over the network. Example user devices include personal computers, smartphones, electronic tables, mobile communication devices, and other devices that can send and receive data over the network. A user device typically includes a user application to facilitate the sending andAttorney Docket No. 43374-0925WO1

[0124] receiving of data over the network, for example, a web browser or a native platform-specific application.

[0125]

[0086] The interconnection queue management platform 230 can process the interconnection request using one or more of the agents 210. In more detail, the agents 210 can use a standardized agentic language to communicate with each of the interconnection assessment tools 220 and can communicate with the different tools using a cloud-based framework. For example, the agents 210 can generate executable-code methods responsive to receiving the user prompt 250 and to engage with the one or more interconnection assessment tools 220. For example, the agents 210 can generate, for the dynamic configuration and in response to the user prompt, an ordered set of input arguments for the selected interconnection assessment tools 220 to respond to the user prompt, e.g., method calls, and passes the method call(s) to the interconnection assessment tools 220.

[0126]

[0087] In some implementations, the agents 210 generate ordered set of arguments to access each of the one or more interconnection assessment tools 220 as well as a set of parameters for each interconnection assessment tool of the selected one or more tools. The set of parameters can include configuration parameters for the interconnection assessment tool, e.g., based on the requirements of the tool. In some instances, the set of parameters are detailed in the tool information accessible by the agent for each of the geospatial tools.

[0127]

[0088] In some implementations, the ordered set of arguments correspond to the order of sequential operations of one or more of the interconnection assessment tools, e.g., two or more interconnection assessment tools where output of a first tool is input to a second tool. In some implementations, the ordered set of arguments correspond to the order of parallel operations of two or more of the interconnection assessment tools, e.g., where parallel processing of data is implemented to increase a speed of providing a response 207, e.g., the interconnection request assessment, to the user prompt 205.

[0128]

[0089] In some implementations, one or more of the interconnection assessment tools of the layer 220 generate output that is used as input for one or more other interconnection assessment tools. In such instances, the agents 210 can engage with an ordered set of interconnection assessment tools 220 in response to a user prompt 205. The ordered set of interconnection assessment tools can perform a sequential or parallel operations to perform a full-stack of operations, e.g., from raw data through a visualization responsive to the user prompt 205.

[0129]

[0090] Each of interconnection assessment tools 220 have corresponding tool information defining tool parameters, operation, expected inputs, outputs, and the like. The toolAttorney Docket No. 43374-0925WO1

[0130] information can be stored, for example, in a readable format accessible to the tool layer 220 where one or more of the agents 210 can access the tool information to determine whether to engage with the tool to respond to a user prompt 205. For example, the agent can select a dynamic configuration of the interconnection assessment tool layer 220 in response to the user prompt 205 and using the tool-specific information.

[0131]

[0091] For example, the interconnection queue management platform 203 can include an Interface Agent 212 that can accept documents associated with the energy project proposal through a user interface. In particular, the Interface Agent 212 can accept graphical user interface (GUI) as well as API-based interconnection requests from one or more submitters, and can register the interconnection requests in a repository. In some implementations, the Interface Agent evaluates the application according to first evaluation criteria and facilitates queries and responses about applications in natural language to elicit any further necessary data to prepare the interconnection application for evaluation by a downstream agent. The Interface Agent 212 can also provide one or more user interfaces to a grid operator user, e.g., to facilitate review and comparison of multiple interconnection requests.

[0132]

[0092] After the Interface Agent 212 has completed processing a received interconnection request, the Interface Agent 212 can provide the interconnection request to a Review Agent 214 that can check the interconnection request for logical correctness and evaluate whether the parcel data included in the interconnection request satisfy the regulatory' requirements for submission, e.g., accuracy of location and boundary, term, exclusivity, conveyance, and that the electrical data of technical specifications included in the interconnection request satisfy compliance requirements for submission.. In some cases, the Review Agent 214 is implemented as multiple Review Agents 214, e.g., a Parcel Review Agent and an Electrical Revieyv Agent.

[0133]

[0093] In this example implementation, the Review Agent 214 can access one or more of the interconnection assessment tools as part of evaluating the energy project for approval. As described with respect to FIG. 1, the Review Agent 214 can generate instructions that cause one or more of data ingestion tools 130, geospatial tools 140, or electrical tools 150 to obtain and process data to validate the parcel data and the electrical data of interconnection request. In particular, the Review Agent 214 can access the data layer 240 to obtain external data, e.g., from a parcel database or property rights database, or internal data, e.g., predefined term, exclusivity, and conveyance requirements of the grid operator to validate the interconnection request.Attorney Docket No. 43374-0925WO1

[0134]

[0094] In particular, data layer 140 can include multiple sources of data stored in databases and in data communication with the interconnection assessment platform 203 through the network. Data sources in the data layer 208 can include, for example, satellite images, parcel data, grid operator requirement data, electrical data, or a combination of any of these. In some cases, the data layer 140 can include outputs from models, e.g., synthetic data. For example, synthetic data can include synthetic data generated by one or more of the interconnection assessment tools to fill in or otherwise supplement missing or incomplete data.

[0135]

[0095] In some implementations, the Review Agent 214 can consult a Grid Knowledge Graph (GKG), e.g., a repository of information about the electrical grid including interconnection-specific grid requirements such as technical specifications, grid modeling parameters, electrical diagrams, equipment certifications, that is generated using electrical data and specifications of the grid operator, e.g., from the data layer 240.

[0136]

[0096] More specifically, the GKG can map relationships between entities and information and concepts relevant to the interconnection process for a particular electrical grid. This rich representation of the electrical grid allows robust and explainable inferences to be made, providing a uniquely strong single source of truth to power the interconnection application intake agents. In particular, the GKG can convert grid requirements for interconnection and other relevant information into standard formats, a considerable challenge given the heterogeneous, siloed, and at-times unstructured data that characterizes the electrical grid, thereby promoting integration and accessibility of data for the agents 210.

[0137]

[0097] In some implementations, the Review Agent 214 uses large language models (LLMs) to regularly consult and update internal as well as external sources of information required for this process, such as equipment specifications, property rights databases, and parcel databases. For example, a list of relevant data sources can be provided to the platform 203, e.g., as an input, and can be updated by utilities / ISOs and other stakeholders. In some cases, the Review Agent 214 can maintain a log of all sources consulted and verifications completed as part of validating the parcel data or the electrical data, e.g., for an audit.

[0138] Moreover, the platform 203 supports the review of the outputs of the Review Agent 214. e.g., by a human-in-the-loop (HITL).

[0139]

[0098] In particular, the Review Agent 214 can use LLMs to perform causal reasoning to understand the causal relationships between events or variables, e.g., the connection configurations included in the interconnection request, rather than just identifying correlations. This involves determining which factors cause which outcomes, and howAttorney Docket No. 43374-0925WO1

[0140] interventions might affect those outcomes. In this case, the Review Agent 214 can use Causal RAG tools. Causal RAG tools also allow for analyzing any causal links between multiple sources of data and documentation relevant to the application. This allows an understanding of the reasoning of the system and narrowing down areas probing further. Optionally, it can help provide the review er with responses to the applicant, and select causes that can automatically generate responses. Examples of Causal RAG techniques are: (a) Causal Graph RAG, where causal graphs are used as external sources for establishing causal relationship, and these external causal graphs are accessed by agents in the multi-agent system; (b) autonomous LLM-based causal discovery; (c) graph-augmented LLM where large knowledge graphs are filtered to emphasize cause-effect edges, and the knowledge graph is a Grid Knowledge Graph.

[0141]

[0099] In some implementations, the Review Agent 214 also has access to a global developer repository, which is a database that includes information about the proposer, including previous applications, active submissions, areas of expertise, and development partners. The global developer repository is. in some implementations, accessible to other utilities and ISOs under appropriate privacy and security controls. Such access can provide context for reviews, helping to identify areas likely to require more attention based on previous proposals as well as potential areas for improvements to the application process.

[0142]

[0100] The Review- Agent 214 can transmit the interconnection request assessment as response 207 to the Interface Agent 212, which will, in turn, present the results of the validation to the submitter, the grid operator, or both. In some implementations, the Review Agent 214 can also maintain the interconnection request assessment, and, in some cases, associated metadata to preserve an audit trail, in a Global Application Repository or another database of submitted interconnection requests.

[0143]

[0101] In the case that the interconnection request assessment is negative, e.g., due to missing, incomplete, or inaccurate information, the platform 203 can provide a description of remedies to the missing, incomplete, or inaccurate information deficiencies, thereby allowing the submitter to make targeted revisions before resubmitting. If changes are required, the Review Agent 214 can identify deficiencies where the proposal does not meet requirements or could otherwise be improved. For example, the platform 203 can alert submitters if the project does not meet setback requirements, if parcel data indicates site control issues, or other incorrect information is found in the response 207. Generally, the process of generating an interconnection request assessment for each submission can continue until the Review- Agent 214 determines the interconnection request is validated.Attorney Docket No. 43374-0925WO1

[0144]

[0102] In some implementations, the Review Agent 214 can transmit the initial interconnection request assessment to a Verification Agent to perform a verification of the validation of the parcel data and the electrical data. In this case, the Verification Agent can provide a final interconnection request assessment to the Interface Agent 212 as response 207, e.g., by verifying the initial interconnection request assessment, or by modifying the initial interconnection request assessment .

[0145]

[0103] Following the validation by the Review Agent 214 or the verification by the Verification Agent 216, a Planning Agent 218 can facilitate schedule scoping calls to refine the application, after which it can be approved for interconnection studies. In some implementations, the Planning Agent utilizes a cloud-based simulation engine and an AI-based scenario generation and analysis, the outputs of which allow for an accurate, efficient end-to-end process for a complete interconnect request response.

[0146]

[0104] For example, the Planning Agent 218 can use a Causal RAG tool to determine whether a particular interconnection request will have immediate impact on the connected feeder and substation or other directly connected facilities, and whether a latent impact will cause further effects deeper in the power grid. In particular, the Planning Agent 218 can explore these using transitive exploration of the electrical GKG or separately derived Influence Graphs, e.g., as is described in more detail with respect to FIG. 3.

[0147]

[0105] As another example, the Planning Agent 218 can learn patterns from prior validated interconnection requests submitted, and generate an alert if an application deviates significantly from these patterns. In this case, the Planning Agent 218 can indicate that the validated interconnection request uses too small or too large of a plot of land to construct a generating unit with a certain type and output.

[0148]

[0106] As described above, the platform 203 provides for an autonomous multi-step validation workflow that uses interconnection assessment agents for perception (ingesting and analyzing all provided documentation), reasoning (identifying key information and causal relationships and action (e.g., assisting human operators review by providing appropriate validation information).

[0149]

[0107] In some implementations, the interconnection assessment agents 210. instead of simply following a linear or two-way chain, can form a more general graphical association, with decision points and loops, forming the multi-agent system. The system 100 can be fully automated, or, alternatively, implemented as a decision support system, which assists a human to make determination as to whether the application is valid.Attorney Docket No. 43374-0925WO1

[0150]

[0108] In the case that one or more of the interconnection assessment agents 210 are language processing neural networks, e.g., large language models, a training system can have trained the agents 210 using supervised fine-tuning. For example, the training system can have trained or finetuned a foundational LLM underlying LLM to incorporate electrical grid specific information. This can be general grid information as well as that contained in the Grid Knowledge Graph. In particular, the training system can update the foundational LLM using a database of historical applications and grid operator validation decisions as ground truth. In some implementations, training data can also be synthetically generated by taking the parameters of the application and (exhaustively or by sampling) enumerating different applications that would be designed to succeed or fail the validation process.

[0151]

[0109] FIG. 3 depicts an example impact similarity engine that can determine a respective measure of impact to the electrical grid for one or more validated interconnection requests and can select interconnection requests for prioritization based on the measure of impact.

[0152]

[0110] Traditionally, transmission operators study the interconnection request queue in a 'Tirst-come, first-serve?’ manner. Transmission operators have been unable to keep up with the grid grow th, leading to years long interconnection queue delays and backlogs in the thousands. However, the impact determines what upgrades must be made to presen e reliable grid operation. Therefore, the impact is the primary determinant of the cost and feasibility of an interconnection request. Since grid operators often consider multiple proposed projects that may each have multiple proposed grid alterations for approval, it can be advantageous to evaluate the impact of each proposed grid alteration as well as the order in which the proposed grid alterations occur over the proposed project and among other proposed projects.

[0153] [Hl] The impact similarity engine 300 can determine the measure of impact of each validated interconnection request. In particular, the impact can be quantified as a set of data that describes the impact to the grid for the interconnection request based on contingencies among the grid elements of the grid, e.g., the likelihood that one grid element of the electrical grid will affect another component. For example, the grid elements can include power plants, transmission lines, distribution lines, transformers, circuit breakers, switches, and substations of the electrical grid.

[0154]

[0112] As described with respect to FIG. 1, the system 100 can use an influence graph 330 that represents the likelihood that one grid element of an electrical grid will affect another component. The influence graph 330 can, for example, predict increased thermalAttorney Docket No. 43374-0925WO1

[0155] violations, under / over voltage, likelihood of cascading failures, and other impacts in a grid change.

[0156]

[0113] In the particular example depicted, the impact similarity engine 300 can generate the influence graph 330 using an existing model of the electrical grid, e.g., the grid knowledge graph 305. As described with respect to FIG. 1, the grid knowledge graph 330 can represent each element of the electrical grid as a node and represents a connection betw een the elements of the electrical grid using an edge. For example, the impact similarity engine 300 can generate the influence graph from the grid knowledge graph 305 by executing prior simulations of the grid knowledge graph 305 based on one or more contingencies.

[0157]

[0114] In more detail, the impact similarity engine 300 can process the grid knowledge graph 305 using an influence graph generator tool 320 to generate the influence graph 310 using a suitable statistical method, e.g., Markovian graphs, graph neural networks, etc. In the case the influence graph generator tool 320 uses a graph neural network (GNN), the GNN can include an encoder block, a sequence of one or more message passing neural network layers, and a decoder block. Each message passing layer is configured to process a set of node embeddings, e.g., representing gnd elements, and a set of edge embeddings, e.g., representing relationships between grid elements, by neural network operations that are parametrized by a set of neural netw ork parameters of the message passing layer and that are conditioned on the topology of the graph representing the electrical grid, to update the node embeddings and the edge embeddings. The decoder block can process the node and edge embeddings generated by the final message passing layer to generate the GNN output.

[0158]

[0115] For example, the influence graph generator tool 320 can simulate potential outages or overloads of one or more various grid elements across the grid, e.g., N-l or N-2, as contingencies. In particular, the influence graph generator tool 320 can iteratively remove one grid element at a time and execute one or more simulations for the N-l contingency. In some implementations, the impact similarity engine can augment the simulation results of the contingencies with additional contingencies based on hypothetical or historical interconnection requests.

[0159]

[0116] By analyzing the impact of each contingency across the grid elements, the influence graph generator tool 320 can generate data specifying the relationships among the grid elements that are encoded by the influence graph 330. For example, the influence graph generator tool 320 can implement techniques described in Hines, P, Dobson, I, and Rezaei, P. “Cascading Power Outages Propagate Locally in an Influence Graph that is not the Actual Grid Topology’’ (arXiv:1508.01775v2, June 2016) to generate the influence graph 330 fromAttorney Docket No. 43374-0925WO1

[0160] the simulated contingencies. As another example, the influence graph generator tool 320 can implement techniques described in Marot. A., et. al. 'Guided Machine Learning for power grid segmentation” (arXiv: 1711.09715v3, Mar 2018) to generate the influence graph 330 from the simulated contingencies.

[0161]

[0117] While described above with respect to generating the influence graph 330, in some implementations, the influence graph 330 can be obtained, e.g., as an input to the impact similarity engine 300. In this case, the influence graph 330 can be created by an upstream system.

[0162]

[0118] The impact similarity engine 300 can use the influence graph 330 to determine the impacts to the grid 370 for one or more interconnection requests 350. In particular, the engine 350 can obtain a number of validated interconnection requests 350 from an interconnection request project database 360, e.g., the interconnection request project database 360 of FIG. 1. Each interconnection request can include electrical data 355 with one or more connection configurations 353.

[0163]

[0119] The engine 300 can process the electrical data 355 and the influence graph using an influence graph calculator tool 340 to determine the impact to the grid 370 specified by the connection configurations 352. The influence graph calculator tool 340 can determine the impact of the interconnection request by determining how the addition of the connection configurations 352 modify the influence graph 330. For example, the tool 340 can use causal analysis, e.g., Bayesian approximation, to determine how the connection configurations propagate through each grid element of the influence graph 330. As another example, the tool 340 can use flow-based methods to simulate a random walk across the influence graph 330 to determine the impact of the interconnection request to each grid element.

[0164]

[0120] In more detail, the contingencies, which describe a loss of a grid asset, are symmetrical to the addition of a grid asset, which makes the influence graph a useful model for evaluating the impact of new additions to the grid specified in interconnection requests. For example, the addition of a generator can be represented as a negative load. Due to the symmetry' between the removal of existing components and the addition of new components, the influence graph calculator tool 340 can use the relationships derived from the contingencies to infer which components will be impacted by a proposed modification to the electrical grid.

[0165]

[0121] The engine 300 can then process the respective impacts to the grid 370 for each of the interconnection requests using a clustering engine 380. More specifically, the clustering tool 380 can cluster the interconnection requests 350 using the respective sets of data thatAttorney Docket No. 43374-0925WO1

[0166] specify the impacts to the grid 370 to identify interconnection requests that affect similar grid elements of the electrical grid. For example, the clustering engine 380 can identify interconnection requests that require upgrades to similar grid elements or result in impacts to different geographical regions.

[0167]

[0122] The clustering tool 380 can use a clustering technique, e.g., a multivariate clustering algorithm, to identify impact groups 385 of one or more interconnection requests by clustering the interconnection requests 350. For example, the clustering tool 380 can cluster the interconnection requests 350 using K-means clustering, density-based clustering, hierarchical clustering, or subspace clustering, to identify groups of interconnection requests as respective clusters based on the measures of impact. In particular, the clustering tool 380 can cluster interconnection requests 350 within a threshold measure of similarity as an impact group. For example, the respective measure of similarity can be a vector distance, e.g., cosine similarity, Euclidean distance, dot product distance, or Manhattan distance, in an embedding space that parameterizes the respective sets of data generated for each interconnection request.

[0168]

[0123] In some implementations, the clustering tool 380 can augment the respective sets of data of the impacts 370 that are used to cluster the interconnection requests 350 using auxiliary data 375. For example, the auxiliary data 375 can include application date, geographic location signals, generation type or size, or any other data from the interconnection request to further inform the clustering.

[0169]

[0124] The impact similarity engine 300 can use the impact groups 385 to select one or more interconnection requests 390 for prioritization. For example, the impact similarity engine 300 can select the interconnection requests 390 based on impact to particular grid elements or a particular geographic region. As another example, the impact similarity engine 300 can select the interconnection requests that have a low impact to the electrical grid, e.g., relative to a predefined impact threshold, for expedited review.

[0170]

[0125] By taking the existing queue and using the influence graph 330 to predict which interconnection requests 350 will affect similar grid elements, geographical regions, or both, the impact similarity engine 300 can generate clusters for review that improve upon a grid operator’s existing clusters, which are generally grouped by their respective application date. The impact clusters can expedite the review process by providing a more meaningful metric than application date to review the interconnection requests 350. Furthermore, the impact similarity clusters can also provide useful information for identifying cost-sharing mechanisms across interconnection requests 350, e.g., as interconnection requests that impactAttorney Docket No. 43374-0925WO1

[0171] similar grid elements, geographic regions, or both are more likely to benefit from upgrades specified by other interconnection requests in the same cluster.

[0172]

[0126] While described above in terms of using an influence graph 330 to determine the impact, in other implementations, the impact similarity engine 300 can process the interconnection requests 350 using a trained impact similarity' machine learning model to generate impact groups 385. In particular, the impact similarity engine 300 can generate a ground-truth training set of interconnection request batches and clusters that can be used to train the impact similarity machine learning model to generate impact groups for a particular electrical grid. For example, this ground-truth training set can be sourced from historical interconnection requests and clusters, or by means of simulations. In this case, a supervised learning approach can be used to train the impact similarity machine learning (ML) model to generate clusters of interconnection requests for an electrical grid, e.g., by updating the respective values of parameters of the model, e.g., using the update rule of any appropriate gradient descent optimization algorithm, e.g., RMSprop, or Adam. In another implementation, the ground-truth training set can be used to finetune a base impact similarity’ machine learning model for a particular electrical grid.

[0173]

[0127] FIG. 4 is an example user interface of an example geospatial visualization that includes a depiction of an aggregated geographic boundary of the set of parcels included in an interconnection request. For example, the UI generator 145 of the interconnection queue management system 100 or the interface agent 212 of the agentic interconnection queue management platform 203 can provide the geospatial visualization UI 400 to a computing device 405.

[0174]

[0128] In particular, the computing device 405 can present the geospatial visualization UI 400 on a display that allows a user to interact with a provided data source. For example, the data source can be provided over a network, such as the internet or an internally available network. In the particular example illustrated, the provided data source includes geospatial data, which has been received and rendered by the computing device 405.

[0175]

[0129] Generally, the geospatial visualization UI 400 is associated with an applied programming interface (API) that enables a user, e.g.. the user of computing device 405, to interact with the geospatial data provided by the geospatial visualization UI 400. As an example, the API can be provided over the same network as the user uses to receive and render the data source.

[0176]

[0130] While described here with respect to the geospatial visualization UI 400, it is understood that an interconnection queue management system or an agentic queueAttorney Docket No. 43374-0925WO1

[0177] management platform can provide data sources associated with other UI to the computing device for rendering. For example, the system or platform can provide an additional UI for uploading an interconnection request for validation and interacting with the interconnection request assessment. As another example, the system or platform an provide an additional UI for viewing the results of impact clusters.

[0178]

[0131] More specifically, the geospatial visualization UI 400 depicts a ‘‘Site Map” 410 for a particular interconnection request. For example, the UI 400 can display the Site Map in response to a selection of a particular interconnection request by a user of the computing device 405. In the particular example depicted, the Site Map 410 includes a polygon 420 that represents the aggregated boundary' of the parcels specified by the parcel data in the interconnection request overlay ed with a satellite image of the geographic area including and surrounding the parcels. As described with respect to FIG. 1, the system or the platform can perform an iterative union on each pair of parcels in the set of parcels to generate the aggregated boundary of the set of parcels.

[0179]

[0132] In particular, the UI 400 can provide a satellite view of the parcels and the geographic region surrounding the parcel to provide a more detailed and realistic site visualization than boundary data alone. The UI 400 can also display the interconnection point 430 to the existing transmission line 435 and the gen-tie or generation interconnection line 440 between the generator facility and the interconnection point on the satellite image. In some cases, the UI 400 can highlight any structures, including hospitals, schools, bridges, etc., on the set of parcels included in the interconnection request, e.g., by providing point-of-interest location icons on the satellite image. In a related case, the UI 400 can highlight any geographic features, e.g., rivers, lakes, mountains, etc., that have to be considered to actuate the interconnection request.

[0180]

[0133] In some implementations, the UI 400 allows a user to specify which data is displayed in the Site Map 410, e.g., by providing one or more visualization layers 425 to the user. In particular, the UI 400 can provide for customization of the geospatial visualization by enabling a user to select one or more layers 425 from a set of layers. In the particular example depicted, the polygon 420 is overlay ed with the satellite map.

[0181]

[0134] In another example, the UI 400 can display the polygon 420 for the interconnection request overlay ed with a ground-truth polygon for the set of parcels identified in the parcel data of the interconnection request (not depicted). In this case, the UI 400 can provide a geospatial view of the discrepancies between the ground-truth parcel data and the parcel data included in the interconnection request, e.g., by overlaying the aggregatedAttorney Docket No. 43374-0925WO1

[0182] boundaries of the interconnection data and the ground-truth data using a ground-truth layer. In some implementations, the UI 400 can display a recommendation for how to rectify the missing, inconsistent, or inaccurate data that resulted in the discrepancy relative to the ground-truth data.

[0183]

[0135] The UI 400 can illustrate whether each of the parcels in the set of parcels satisfies the property rights requirements of the grid operator, e.g., the term, exclusivity, and conveyance of the grid operator. In the particular example depicted, the UI 400 outlines each parcel 420 within the polygon and labeled as either compliant or deficient in accordance with the parcel assessment of the interconnection request assessment. As another example, the UI 400 can indicate whether the parcel is compliant or deficient based on the color of the border of the parcel within the polygon 420 or the hatching of the border of the parcel within the polygon 420.

[0184]

[0136] In some cases, in response to a selection by a user of the computing device 405, the UI 400 can identify one or more of the term, exclusivity, and conveyance criteria that were not met for the deficient polygons. In particular, the UI 400 can update the visualization displayed based on a click or hover-over of the parcel with a drop-box or pop-up window that includes non-spatial data related to the parcel. For example, the non-spatial data can include term, exclusivity, and conveyance data and metadata for the parcel, e.g., a list of extracted location identifiers, the data source of the property rights information for the parcel, or how7many other proposals in the same cohort include the parcel.

[0185]

[0137] While not depicted in FIG. 4, the UI 400 can illustrate easement or right of way data, e.g., by including a visualization of a right of way path through a parcel that is not exclusively controlled by the submitter. As an example, the UI 400 can highlight the nonexclusive control by overlaying an easement layer that represents any easements using a repeated pattern of icons aligned along the right of way path or another stylized line with different shading than the parcel lines in the polygon 420. As another example, the drop-box or pop-up window7can additionally include data specifying the easement or right of w ay through the selected parcel.

[0186]

[0138] In some implementations, the geospatial visualization UI 400 can be configured to display the polygon 420, e.g., the aggregated boundary of the parcels, for each of a number of interconnection requests in the same cohort. In this case, the UI 400 can display a “Cohort Map” of the polygons for all interconnection requests in a cohort in response to a selection of a cohort by a user of the computing device 405, or in response to a selection of two or more interconnection requests in the same cohort by the user of the computing device 405.Attorney Docket No. 43374-0925WO1

[0187]

[0139] In another implementation, the geospatial visualization UI 400 can be configured to display the polygon 420 for a number of interconnection requests received from the same submitter. For example, the UI 400 can allow for comparison of the polygons of different versions of the same project from the same submitter. As another example, the geospatial visualization UI 400 can be configured to display the polygon 420 for distinct interconnection requests, e.g., for different projects, received from the same submitter. In this case, the UI can aid the user of the computing device 405 to identify double claiming of a parcel for distinct interconnection requests.

[0188]

[0140] FIG. 5 is a flow diagram of an example process for generating an impact assessment for an interconnection request. For convenience, the process 500 will be described as being performed by a system of one or more computers located in one or more locations. For example, an interconnection queue management system, e.g., the interconnection queue management system 100 of FIG. 1, appropriately programmed in accordance with this specification, can perform the process 500.

[0189]

[0141] The system can obtain project data associated with an interconnection request specifying a modification to an electrical grid (step 510). In particular, the project data can include parcel data specifying a first identification of a set of parcels of land and first electrical data specifying a modification to the electrical grid. Generally, the first identification of the set of parcels and the first electrical data are specified by the project data. However, the project data is not guaranteed to include meaningful project data. For example, the first identification of the set of parcels, the first electrical data, or both can include incorrect, legally inconsistent, or incompatible data that specifies a modification to the electrical grid that cannot feasibly be completed.

[0190]

[0142] The system can obtain a knowledge graph for the electrical grid (step 520). In this context, a knowledge graph can refer to a data structure that connects data corresponding with different entities based on relationships among the entities. In particular, the knowledge graph for the electrical grid can include (i) a number of nodes representing each element of the electrical grid, and (ii) a number of edges connecting a pair of nodes of the number of nodes and representing a connection between the elements of the electrical grid represented by the pair of nodes. For example, the nodes can correspond with power plants, transmission lines, distribution lines, transformers, circuit breakers, switches, and substations of the electrical grid, and the knowledge graph can encode domain knowledge of a power systems engineer relative to each of the elements of the electrical grid as well as the connections between the elements.Attorney Docket No. 43374-0925WO1

[0191]

[0143] In some cases, the system can obtain the knowledge graph for the electrical grid from a data repository of knowledge graphs that have been previously generated for different electrical grids. In other cases, the system can generate the knowledge graph, e.g., in response to determining that the knowledge graph is not included in the data repository of knowledge graphs that have been previously generated. In this case, the system can process data including grid element specification parameters for each of the one or more grid elements and operational parameters to generate the knowledge graph for the electrical grid. For example, the system can process the manuals of each of the electrical grid elements as well as a diagram that specifies the relationship between the electrical grid elements to generate the knowledge graph.

[0192]

[0144] The system can process the project data and the knowledge graph by one or more interconnection assessment tool(s) corresponding with one or more interconnection assessment agent(s) (step 530). More specifically, the system can use one or more interconnection tool(s) to validate the project data, e.g., the first identification of a set of parcels of land and first electrical data specifying a modification to the electrical grid, to ensure that the project data specifies a feasible interconnection request. In some cases, the system can process the project data against the nodes of the knowledge graph, e g., the system can process the project data against the structured relationships between grid elements represented by the grid knowledge graph. For example, the system can traverse the structured relationships represented by the topology of the knowledge graph to validate the project data.

[0193]

[0145] In more detail, the system can validate, by one or more first tools of the interconnection assessment tools, the first identification of the set of parcels using a second identification of the set of parcels. Generally, the first tools can be parcel tools that relate to parcel assessment. For example, the system can obtain the second identification of the set of parcels from one or more databases that include ground truth geolocation data for the set of parcels. In some cases, the system can obtain ground truth geolocation data from a parcel database that maintains land assessment data, e.g., recorded for taxation purposes. In some cases, the first identification of the set of parcels can include first geographic boundary data for each parcel of the set of parcels, as identified in the project data, and the second identification of the set of parcels can include second geographic boundary data for each parcel of the set of parcels, as identified by the ground truth geolocation data.

[0194]

[0146] For example, the system can validate the first identification of the set of parcels using the second identification of the set of parcels by identifying a discrepancy that characterizes conflicting land use between the first identification of the set of parcels and theAttorney Docket No. 43374-0925WO1

[0195] second identification of the set of parcels. In this context, conflicting land use can refer to either conflicting geographic boundaries of legally owned parcels or conflicting ownership or right of way access to parcels with consistent geographic boundaries. An example for validating the first identification of the set of parcels based on a second identification of the set of parcels will be described in more detail with respect to FIG. 6.

[0196]

[0147] The system can also validate, by one or more second tools of the interconnection assessment tools, that the first electrical data satisfies one or more criteria in accordance with second electrical grid data of the knowledge graph. Generally, the second tools can be electric tools that relate to assessment of the electrical grid. In particular, the first electrical data of the interconnection request can include one or more connection configurations, e.g., that each specify a respective modification to a specific element of the electrical grid.

[0197]

[0148] For example, the system can extract, by the one or more second tools, a first number of electrical parameters corresponding with the one or more connection configurations of the interconnection request. In some cases, the system can extract the first number of electrical parameters using one or more document extraction tools. The system can determine a second number of electrical parameters of the electrical grid that correspond with the one or more connection configurations of the interconnection request from the knowledge graph, and can determine whether the first number of electrical parameters satisfy a set of operational criteria in accordance with the second number of electrical parameters.

[0198]

[0149] Generally, the system can determine whether the first number of electrical parameters satisfy the set of operational criteria by performing a non-exhaustive validation of the first number of electrical parameters associated with the interconnection request. In particular, the system can verify that the first number of electrical parameters align with the second number of electrical parameters of the knowledge graph, e.g., to ensure that connections to different grid elements are feasible.

[0199]

[0150] The system can provide an interconnection request assessment generated for the interconnection request to a user (step 540). For example, the system can provide the impact assessment by at least one interconnection agent to the user. In particular, the interconnection request assessment can include the impact assessment for the interconnection request. In some cases, the interconnection request assessment can further include results of validating the first identification of the set of parcels based on a second identification of the set of parcels.

[0200]

[0151] More specifically, in response to validating the first identification of the set of parcels and the first electrical grid data, the system can generate an impact assessment thatAttorney Docket No. 43374-0925WO1

[0201] reflects one or more effects of modification to the electrical grid specified by the interconnection request. For example, the system can determine a measure of impact caused by the modification to the electrical grid specified by the interconnection request.

[0202]

[0152] In some implementations, determining the measure of impact can involve, generating a respective set of data that characterizes the measure of impact caused by the modification to the electrical grid specified by the interconnection request by processing one or more electrical parameters corresponding with the interconnection request using one or more grid simulation tools of the number of interconnection assessment tools. As an example, the one or more grid simulation tools can include an electrical overload simulation tool and an instability simulation tool. Since executing a grid simulation using one or more grid simulation tools can be computationally expensive, the system can reduce the use of computational resources by generating the impact assessment after validation both the first identification of the set of parcels and the first electrical grid data.

[0203]

[0153] In some implementations, the system can generate respective impact assessments for a number of interconnection requests for the electrical grid. In this case, the system can determine a respective measure of similarity for each of the interconnection request between the respective impact assessment of the interconnection request and one or more impact assessments for other interconnection requests, e.g., based on the respective sets of data generated by the one or more grid simulation tools.

[0204]

[0154] As a related example, the system can determine whether a discrepancy between an impact assessment of a particular interconnection request and each of the one or more other impact assessments for other interconnection requests exceeds a threshold tolerance criterion. In response to determining that the discrepancy exceeds the threshold tolerance criterion, the system can provide a notification to the user. An example for evaluating discrepancies between impact assessments of a number of interconnection requests for the electrical grid will be described in more detail with respect to FIG. 7.

[0205]

[0155] In the case that the system does not validate the first identification of the set of parcels, nor the first electrical grid data, the system can provide an error message or alert to a user of the system that indicates that the project data was not validated. In the case that the system does not validate the first identification of the set of parcels, but validates the first electrical grid data, or does not validate the first electrical grid data but validates the first identification of the set of parcels, the system can provide an error message or alert that details which portion of the project data was not validated. For example, the system canAttorney Docket No. 43374-0925WO1

[0206] provide the error message or alert to the grid operator, the submitter of the interconnection request, or both.

[0207]

[0156] FIG. 6 is a flow diagram of an example process for generating a regulatory assessment of the interconnection request. For convenience, the process 600 will be described as being performed by a system of one or more computers located in one or more locations. For example, an interconnection queue management system, e.g., the interconnection queue management system 100 of FIG. 1, appropriately programmed in accordance with this specification, can perform the process 600. In some cases, step 530 of process 500 can be performed using process 600.

[0208]

[0157] The system can obtain parcel data including a first identification of a set of parcels of land associated with an interconnection request (step 610). In some cases, the system can obtain the parcel data as a portion of the project data obtained in step 510 of FIG. 5. For example, the system can obtain a number of documents associated with the interconnection request, and can extract, by one or more geospatial tools, one or more identifiers from the number of documents as the first identification of the set of parcels. As described previously, the first identification of the set of parcels is not guaranteed to include correct, legally consistent, or compatible data that specifies a modification to the electrical grid that cannot feasibly be completed, e.g., based on property right constraints.

[0209]

[0158] The system can generate a parcel assessment for the interconnection request using one or more geospatial tools (step 620). In particular, the system can generate the parcel assessment by coordinating the execution of one or more highly-specialized tools that are tailored to perform particular geospatial tasks. For example, the system can process the first identification of the set of parcels using the one or more geospatial tools to determine whether the parcel data is consistent with ground truth parcel data and compliant with data that specifies legal property rights to the set of parcels of land. In some implementations, the number of geospatial tools can correspond with one or more interconnection assessment agents.

[0210]

[0159] In more detail, the system can, by one or more geospatial tools of a number of geospatial tools, obtain a second identification of the set of parcels and can validate the first identification of the set of parcels using the second identification of the set of parcels. For example, the system can obtain the second identification of the set of parcels from one or more ground truth geolocation databases. In some cases, the first identification of the set of parcels includes first geographic boundary data for each parcel of the set of parcels, and the second identification of the set of parcels includes second target geographic boundary dataAttorney Docket No. 43374-0925WO1

[0211] for each parcel of the set of parcels. In this case, the system can validate the first identification of the set of parcels by identifying a discrepancy between the first geographic boundary data and the second target geographic boundary data for each parcel of the set of parcels. The system can generate the parcel assessment based at least in part on the discrepancy.

[0212]

[0160] The system can also obtain, by one or more geospatial tools of the number of geospatial tools, data that specifies legal property rights to the set of parcels of land, and can determine whether the project data is compliant with data that specifies legal property rights to the set of parcels of land. For example, the system can obtain, by one or more geospatial tools of the number of geospatial tools, one or more legal documents corresponding with the first identification of the set of parcels of land from one or more property record databases. The system can extract, by one or more geospatial tools of the number of geospatial tools, the data that specifies legal property rights to the set of parcels of land from the one or more legal documents, and can validate that ownership of each parcel in the set of parcels permits the interconnection request. In some cases, extracting the data that specifies legal property rights can involve generating the data that specifies legal property rights by processing the one or more legal documents using a generative neural network.

[0213]

[0161] In some implementations, the set of parcels is a first set of parcels, and the parcel data further includes an identification of a second set of parcels of land corresponding with the first set of parcels of land and associated with a right of way to an interconnection location of the interconnection request. In this case, the system can validate that the right of way to the interconnection location of the interconnection request is authorized from the data that specifies legal property rights to the second set of parcels of land.

[0214]

[0162] The system can generate the parcel assessment based at least in part on the validation. For example, in response to validating that the ownership of each parcel in the set of parcels permits the interconnection request, the system can generate a positive parcel assessment. In contrast, in response to not validating that the ownership of each parcel in the set of parcels permits the interconnection request, the system can generate a negative parcel assessment. As another example, in response to validating that the right of way to the interconnection location is authorized, the system can generate a positive parcel assessment. In contrast, in response to not validating that the right of way to the interconnection location is authorized, the system can generate a negative parcel assessment.

[0215]

[0163] The system can provide the parcel assessment for the interconnection request to a user (step 630). For example, the system can generate a user interface for displaying aAttorney Docket No. 43374-0925WO1

[0216] geospatial visualization of the assessment for the interconnection request and can provide the user interface to a user, e.g., by at least one geospatial tool. In particular, the user interface can (i) associate the first identification of the set of parcels with the second identification of the set of parcels, and (ii) provide an approval indicator determined based at least on the outputs of the one or more geospatial tools. In some implementations, generating the user interface can involve overlaying a first aggregated geographic boundary generated from the first identification of the set of parcels on a second aggregated geographic boundary generated from the second identification of the set of parcels.

[0217]

[0164] Generally, the approval indicator can indicate whether or not the parcel data is compliant with the data that specifies legal property rights to the set of parcels of land. In some cases, the user interface can include an input portion, e.g., for a reviewing user to modify the approval indicator. In this case, the system can receive input data for modifying the approval indicator determined based at least on the outputs of the one or more geospatial tools.

[0218]

[0165] FIG. 7 is a flow diagram of an example process for selecting one or more interconnection requests for prioritization based on the respective measures of impact for each interconnection request. For convenience, the process 700 will be described as being performed by a system of one or more computers located in one or more locations. For example, an impact similarity engine, e.g., the impact similarity engine 300 of FIG. 3, appropriately programmed in accordance with this specification, can perform the process 700.

[0219]

[0166] The system can obtain a number of interconnection requests specifying a modification to an electrical grid (step 710). Each interconnection request can be associated with a set of parcels of land and electrical data specifying the modification to the electrical grid, e.g., as one or more connection configurations. In this context, each connection configuration can detail a respective modification to a specific element of the electrical grid. In particular, the respective sets of parcels of land and the respective electrical data of the number of interconnection requests can have been validated using a number of interconnection assessment tools. For example, the system can perform process 500 to validate some or all of the interconnection requests, and can provide the validated interconnection requests for further processing using process 700.

[0220]

[0167] The system can determine a measure of impact for each interconnection request caused by the modification to the electrical grid specified by the interconnection request (step 720). In particular, the system can generate a set of data that describes the impact to theAttorney Docket No. 43374-0925WO1

[0221] electrical grid for the interconnection request. For example, the system can extract, by one or more interconnection assessment tools, one or more electrical parameters corresponding with each connection configuration specified by the interconnection request. The system can generate the set of data by processing the one or more electrical parameters using an electrical simulation tool.

[0222]

[0168] In some implementations, the system can obtain an influence graph characterizing a number of contingencies among grid elements of the electrical grid. In this context, a contingency can refer to the likelihood that one grid element of the electrical grid will affect another component. The system can process the influence graph and the electrical data of the interconnection request to determine the measure of impact caused by the modification to the electrical grid.

[0223]

[0169] The system can determine a respective measure of similarity of the measure of impact for each interconnection request and other interconnection requests obtained for the electrical grid (step 730). For example, the system can determine a vector distance, e.g., cosine similarity. Euclidean distance, dot product distance, or Manhattan distance, between each interconnection request in an embedding space that parameterizes the respective sets of data generated for each interconnection request.

[0224]

[0170] The system can select one or more interconnection request for prioritization based on the measure of similarity (step 740). In particular, the system can select one or more interconnection requests by clustering the interconnection requests into a number of clusters based on the determined measures of impact, e.g., using any appropriate clustering technique. The system can select the one or more interconnection requests from a first grouping of interconnection requests. In some cases, the first grouping of interconnection requests can include interconnection requests that have an impact on a particular portion of the grid.

[0225]

[0171] Embodiments of the subject matter and the functional operations described in this specification can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Embodiments of the subject matter described in this specification can be implemented using one or more modules of computer program instructions encoded on a computer-readable medium for execution by, or to control the operation of, data processing apparatus. The computer-readable medium can be a manufactured product, such as a hard drive in a computer system or an optical disc sold through retail channels, or an embedded system. The computer-readable medium can be acquired separately and later encoded with the one or more modules of computer programAttorney Docket No. 43374-0925WO1

[0226] instructions, such as by delivery of the one or more modules of computer program instructions over a wired or wireless network. The computer-readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, or a combination of one or more of them.

[0227]

[0172] The term “data processing apparatus” encompasses all apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. The apparatus can include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, a runtime environment, or a combination of one or more of them. In addition, the apparatus can employ various different computing model infrastructures, such as web services, distributed computing and grid computing infrastructures.

[0228]

[0173] A computer program (also know n as a program, software, software application, script, or code) can be written in any suitable form of programming language, including compiled or interpreted languages, declarative or procedural languages, and it can be deployed in any suitable form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub-programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

[0229]

[0174] The processes and logic Hows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as. special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit).

[0230]

[0175] Processors suitable for the execution of a computer program include, by w ay of example, special purpose microprocessors. Generally, a processor will receive instructions and data from a read-only memory or a random access memory or both. The essentialAttorney Docket No. 43374-0925WO1

[0231] elements of a computer are a processor for performing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto-optical disks, or optical disks. However, a computer need not have such devices. Moreover, a computer can be embedded in another device, e.g., a mobile telephone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a Global Positioning System (GPS) receiver, or a portable storage device (e.g., a universal serial bus (USB) flash drive), to name just a few. Devices suitable for storing computer program instructions and data include all forms of non-volatile memory', media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

[0232]

[0176] In this specification the term "engine" is used broadly to refer to a software-based system, subsystem, or process that is programmed to perform one or more specific functions. Generally, an engine will be implemented as one or more software modules or components, installed on one or more computers in one or more locations. In some cases, one or more computers will be dedicated to a particular engine; in other cases, multiple engines can be installed and running on the same computer or computers.

[0233]

[0177] To provide for interaction with a user, embodiments of the subject matter described in this specification can be implemented on a computing device capable of providing information to a user. The information can be provided to a user in any form of sensory format, including visual, auditory, tactile or a combination thereof. The computing device can be coupled to a display device, e.g., an LCD (liquid crystal display) display device, an OLED (organic light emitting diode) display device, another monitor, a head mounted display device, and the like, for displaying information to the user. The computing device can be coupled to an input device. The input device can include a touch screen, keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computing device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any suitable form of sensory feedback, e.g.. visual feedback, auditor)’ feedback, or tactile feedback; and inputAttorney Docket No. 43374-0925WO1

[0234] from the user can be received in any suitable form, including acoustic, speech, or tactile input.

[0235]

[0178] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. Embodiments of the subject matter described in this specification can be implemented in a computing system that includes a back-end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front-end component, e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the subject matter described is this specification, or any combination of one or more such back-end, middleware, or front-end components. The components of the system can be interconnected by any suitable form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (‘“LAN”) and a wide area network (‘“WAN”), an inter-network (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks).

[0236]

[0179] In addition to the embodiments described above, the following embodiments are also innovative:

[0237]

[0180] Embodiment 1 is a computer-implemented method comprising:

[0238] obtaining parcel data comprising a first identification of a set of parcels associated with an interconnection request;

[0239] generating, by one or more geospatial tools of a plurality of geospatial tools, a parcel assessment for the interconnection request by processing the first identification of one or more parcels, wherein the generating comprises:

[0240] obtaining, by one or more geospatial tools of the plurality of geospatial tools, a second identification of the set of parcels;

[0241] validating, by one or more geospatial tools of the plurality of geospatial tools, the first identification of the set of parcels using the second identification of the set of parcels;

[0242] obtaining, by one or more geospatial tools of the plurality of geospatial tools, data that specifies legal property rights to the set of parcels of land;

[0243] determining, by one or more geospatial tools of the plurality of geospatial tools, whether the parcel data is compliant with data that specifies legal property rights to the set of parcels of land; andAttorney Docket No. 43374-0925WO1

[0244] providing, by at least one geospatial tool, the parcel assessment for the interconnection request to a user.

[0245]

[0181] Embodiment 2 is the method of embodiment 1 , further comprising:

[0246] generating, using outputs of the one or more geospatial tools, a user interface for displaying a geospatial visualization of the assessment for the interconnection request that (i) associates the first identification of the set of parcels with the second identification of the set of parcels, and (ii) provides an approval indicator determined based at least on the outputs of the one or more geospatial tools; and

[0247] providing, by at least one geospatial tool, the user interface to a user.

[0248]

[0182] Embodiment 3 is the method of embodiment claim 2, wherein the user interface comprises an input portion, and further comprising receiving, via the input portion of the user interface, input data for modifying the approval indicator determined based at least on the outputs of the one or more geospatial tools.

[0249]

[0183] Embodiment 4 is the method of embodiment of claim 2, wherein generating the user interface that associates the first identification of the set of parcels with the second identification of the set of parcels comprises overlaying, on the user interface, a first aggregated geographic boundary generated from the first identification of the set of parcels on a second aggregated geographic boundary generated from the second identification of the set of parcels.

[0250]

[0184] Embodiment 5 is the method of any one of embodiments 1 -4, wherein obtaining parcel data comprising a first identification of a set of parcels of land associated with an interconnection request comprises:

[0251] obtaining a plurality' of documents associated with the interconnection request; and extracting, by one or more geospatial tools, one or more identifiers from the plurality of documents as the first identification of the set of parcels.

[0252]

[0185] Embodiment 6 is the method of any one of embodiments 1-5, wherein obtaining the second identification of the set of parcels comprises obtaining the second identification of the set of parcels from one or more ground truth geolocation databases.

[0253]

[0186] Embodiment 7 is the method of any one of embodiments 1 -6, wherein the first identification of the set of parcels comprises first geographic boundary data for each parcel of the set of parcels, and wherein the second identification of the set of parcels comprises second target geographic boundary' data for each parcel of the set of parcels.Attorney Docket No. 43374-0925WO1

[0254]

[0187] Embodiment 8 is the method of embodiment 7, wherein validating the first identification of the set of parcels using a second identification of the set of parcels comprises:

[0255] identifying a discrepancy between the first geographic boundary data and the second target geographic boundary' data for each parcel of the set of parcels; and

[0256] generating the parcel assessment based at least in part on the discrepancy.

[0257]

[0188] Embodiment 9 is the method of any one of embodiments 1-8, wherein obtaining data that specifies legal property rights to the set of parcels of land comprises:

[0258] obtaining, by one or more geospatial tools of the plurality of geospatial tools, one or more legal documents corresponding with the first identification of the set of parcels of land from one or more property record databases; and

[0259] extracting, by one or more geospatial tools of the plurality of geospatial tools and from the one or more legal documents, the data that specifies legal property' rights to the set of parcels of land.

[0260]

[0189] Embodiment 10 is the method of embodiment 9. wherein determining whether the parcel data is compliant with the data that specifies legal property rights to the set of parcels of land comprises:

[0261] validating, from the data that specifies legal property' rights to the set of parcels of land, that ownership of each parcel in the set of parcels permits the interconnection request; and

[0262] generating the parcel assessment based at least in part on the validation.

[0263]

[0190] Embodiment 11 is the method of embodiment 10, wherein generating the parcel assessment based at least in part on the validation comprises:

[0264] in response to validating that the ownership of each parcel in the set of parcels permits the interconnection request, generating a positive parcel assessment.

[0265]

[0191] Embodiment 12 is the method of embodiment 10, wherein generating a parcel assessment based on the validation comprises:

[0266] in response to not validating that the ownership of each parcel in the set of parcels permits the interconnection request, generate a negative parcel assessment.

[0267]

[0192] Embodiment 13 is the method of any one of embodiments 1-12, wherein the set of parcels is a first set of parcels, and wherein the parcel data further comprises an identification of a second set of parcels of land corresponding with the first set of parcels of land and associated with a right of way to an interconnection location of the interconnection request.Attorney Docket No. 43374-0925WO1

[0268]

[0193] Embodiment 14 is the method of embodiment 13, wherein determining whether the parcel data is compliant with the data that specifies legal property rights to the set of parcels comprises:

[0269] validating, from the data that specifies legal property rights to the second set of parcels of land, that the right of way to the interconnection location of the interconnection request is authorized; and

[0270] generating the parcel assessment based at least in part on the validation.

[0271]

[0194] Embodiment 15 is the method of embodiment 14, wherein generating a parcel assessment based on the validation comprises:

[0272] in response to validating that the right of way to the interconnection location is authorized, generating a positive parcel assessment.

[0273]

[0195] Embodiment 16 is the method of embodiment claim 14, wherein generating a parcel assessment based on the validation comprises:

[0274] in response to not validating that the right of way to the interconnection location is authorized, generating a negative parcel assessment.

[0275]

[0196] Embodiment 17 is the method of embodiment 9, wherein extracting the data that specifies legal property rights to the set of parcels of land comprises:

[0276] generating the data that specifies legal property7rights by processing the one or more legal documents using a generative neural network.

[0277]

[0197] Embodiment 18 is the method of embodiment method of any one of embodiments 1-17, wherein the plurality of geospatial tools correspond with one or more interconnection assessment agents.

[0278]

[0198] Embodiment 19 is a computer-implemented method comprising:

[0279] obtaining project data associated with an interconnection request, the project data comprising a (i) parcel data specifying first identification of a set of parcels of land and (ii) first electrical data specifying a modification to the electrical grid;

[0280] obtaining a knowledge graph for the electrical grid comprising (i) a plurality of nodes representing each element of the electrical grid, and (ii) a plurality' of edges, each edge connecting a pair of nodes of the plurality of nodes and representing a connection between the elements of the electrical grid represented by the pair of nodes;

[0281] processing, by one or more interconnection assessment tools of a plurality of interconnection assessment tools corresponding with one or more interconnection assessment agents, the project data and the knowledge graph, wherein processing comprises:Attorney Docket No. 43374-0925WO1

[0282] validating, by one or more first tools of the plurality of interconnection assessment tools, the first identification of the set of parcels using a second identification of the set of parcels;

[0283] validating, by one or more second tools of the plurality of interconnection assessment tools, that the first electrical data satisfies one or more criteria in accordance with second electrical grid data of the knowledge graph;

[0284] in response to validating the first identification of the set of parcels and the first electrical data:

[0285] generating, by one or more third tools of the plurality7of interconnection assessment tools, an impact assessment that reflects one or more effects of modification to the electrical grid specified by the interconnection request; and

[0286] providing, by at least one interconnection assessment agent, an interconnection request assessment comprising the impact assessment for the interconnection request and to a user.

[0287]

[0199] Embodiment 20 is the method of embodiment 19, wherein the first tools relate to parcel assessment.

[0288]

[0200] Embodiment 21 is the method of any one of embodiments 19-20, wherein the second tools relate to assessment of the electrical grid.

[0289]

[0201] Embodiment 22 is the method of any one of embodiments 19-21, wherein obtaining the knowledge graph comprises:

[0290] processing data comprising grid element specification parameters for each of one or more grid elements and operational parameters to generate the knowledge graph for the electrical grid.

[0291]

[0202] Embodiment 23 is the method of any one of embodiments 19-22, further comprising obtaining the second identification of the set of parcels from one or more databases that comprise ground truth geolocation data.

[0292]

[0203] Embodiment 24 is the method of any one of embodiments 19-23, wherein the first identification of the set of parcels comprises first geographic boundary data for each parcel of the set of parcels, and wherein the second identification of the set of parcels comprises second target geographic boundary7data for each parcel of the set of parcels.

[0293]

[0204] Embodiment 25 is the method of any one of embodiments 19-24, wherein validating the first identification of the set of parcels using the second identification of the set of parcels comprises:Attorney Docket No. 43374-0925WO1

[0294] identifying a discrepancy that characterizes conflicting land use between the first identification of the set of parcels and the second identification of the set of parcels.

[0295]

[0205] Embodiment 26 is the method of any one of embodiments 19-25, wherein processing comprises processing the project data against the nodes of the knowledge graph.

[0296]

[0206] Embodiment 27 is the method of any one of embodiments 19-26, wherein the first electrical data of the interconnection request comprises one or more connection configurations, and wherein validating that the first electrical data satisfies one or more criteria in accordance with second electrical grid data of the knowledge graph comprises: extracting, by one or more second tools of the plurality of interconnection assessment tools, a first plurality of electrical parameters corresponding w ith the one or more connection configurations of the interconnection request;

[0297] determining, from the knowledge graph, a second plurality of electric parameters of the electrical grid that correspond with the one or more connection configurations of the interconnection request; and

[0298] determining whether the first plurality of electrical parameters satisfy a set of operational criteria in accordance with the second plurality of electrical parameters.

[0299]

[0207] Embodiment 28 is the method of any one of embodiments 19-27, wherein generating, by one or more third tools of the plurality of interconnection assessment tools, the impact assessment of the interconnection request comprises:

[0300] determining a measure of impact caused by the modification to the electrical grid specified by the interconnection request.

[0301]

[0208] Embodiment 29 is the method of embodiment 28, wherein determining the measure of impact comprises:

[0302] generating, using one or more grid simulation tools of the plurality of interconnection assessment tools, a respective set of data that characterizes the measure of impact caused by the modification to the electrical grid specified by the interconnection request by processing one or more electrical parameters corresponding with the interconnection request.

[0303]

[0209] Embodiment 30 is the method of embodiment 29, wherein the one or more grid simulation tools comprise a first grid simulation tool that corresponds with an electrical overload simulation and a second grid simulation tool that corresponds with an instability simulation.

[0304]

[0210] Embodiment 31 is the method of any one of embodiments 19-30, further comprising:

[0305] generating respective impact assessments for a plurality of interconnection requests for the electrical grid; andAttorney Docket No. 43374-0925WO1

[0306] determining, for each of the interconnection requests, a respective measure of similarity between the respective impact assessment of the interconnection request and one or more impact assessments for other interconnection requests.

[0307] [2H] Embodiment 32 is the method of embodiment 31, further comprising:

[0308] determining whether a discrepancy between a first impact assessment of a first interconnection request and each of the one or more other impact assessments for other interconnection requests exceeds a threshold tolerance criterion; and

[0309] in response to determining that the discrepancy exceeds the threshold tolerance criterion, providing a notification to a user.

[0310]

[0212] Embodiment 33 is a computer-implemented method comprising:

[0311] obtaining a plurality of interconnection requests comprising a modification to an electrical grid, each associated with (i) a set of parcels and (ii) electrical data specifying the modification to an electrical grid;

[0312] determining, for each interconnection request, a measure of impact caused by the modification to the electrical grid specified by the interconnection request;

[0313] determining, for each of the interconnection requests, a respective similarity of the measure of impact and the measures of impact of other interconnection requests obtained for the electrical grid; and

[0314] selecting one or more interconnection requests for prioritization based on the measure of similarity.

[0315]

[0213] Embodiment 34 is the method of embodiment 33, wherein determining the measure of impact caused by the modification to the electrical grid specified by the interconnection request comprises generating, for each interconnection request, a set of data that describes the impact to the electrical grid for the interconnection request.

[0316]

[0214] Embodiment 35 is the method of embodiment 34, wherein the interconnection request comprises one or more connection configurations, wherein generating the set of data comprises:

[0317] extracting, by one or more interconnection assessment tools of a plurality of interconnection assessment tools, one or more electrical parameters corresponding with each connection configuration specified by the interconnection request; and

[0318] generating the set of data by processing the one or more electrical parameters using an electrical simulation tool.

[0319]

[0215] Embodiment 36 is the method of any one of embodiments 33-35, wherein selecting one or more interconnection requests comprises:Attorney Docket No. 43374-0925WO1

[0320] clustering the interconnection requests into a plurality of clusters based on the determined measures of impact; and

[0321] selecting the one or more interconnection requests from a first grouping of interconnection requests.

[0322]

[0216] Embodiment 37 is the method of embodiment 36, wherein the first grouping of interconnection requests comprises interconnection requests that have an impact on a particular portion of the electrical grid.

[0323]

[0217] Embodiment 38 is the method of any one of embodiments 33-37, wherein the respective sets of parcels of land and the respective electrical data of the plurality of interconnection requests have been validated using a plurality of interconnection assessment tools.

[0324]

[0218] Embodiment 39 is the method of any one of embodiments 33-38, wherein determining, for each interconnection request, a measure of impact caused by the modification to the electrical grid specified by the interconnection request comprises:

[0325] obtaining an influence graph characterizing a plurality of contingencies among grid elements of the electrical grid; and

[0326] processing the influence graph and the electrical data of the interconnection request to determine the measure of impact caused by the modification to the electrical grid.

[0327]

[0219] Embodiment 40 is the method of any one of embodiments 33-39, wherein determining the respective similarity of the measure of impact and the measures of impact of other interconnection requests obtained for the electrical grid comprises:

[0328] determining a vector distance in an embedding space that parameterizes the respective sets of data generated for each interconnection request.

[0329]

[0220] While this specification contains many implementation details, these should not be construed as limitations on the scope of what is being or may be claimed, but rather as descriptions of features specific to particular embodiments of the disclosed subject matter. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.

[0330] Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination. Thus, unless explicitly stated otherwise, orAttorney Docket No. 43374-0925WO1

[0331] unless the knowledge of one of ordinary skill in the art clearly indicates otherwise, any of the features of the embodiments described above can be combined with any of the other features of the embodiments described above.

[0332]

[0221] Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and / or parallel processing may be advantageous. Moreover, the separation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0333]

[0222] Thus, particular embodiments of the invention have been described. Other embodiments are within the scope of the following claims. For example, the actions recited in the claims can be performed in a different order and still achieve desirable results.

Claims

Attorney Docket No. 43374-0925WO1CLAIMSWhat is claimed is:

1. A computer-implemented method comprising:obtaining project data associated with an interconnection request, the project data comprising a (i) parcel data specifying first identification of a set of parcels of land and (ii) first electrical data specifying a modification to the electrical grid;obtaining a knowledge graph for the electrical grid comprising (i) a plurality of nodes representing each element of the electrical grid, and (ii) a plurality of edges, each edge connecting a pair of nodes of the plurality of nodes and representing a connection betw een the elements of the electrical grid represented by the pair of nodes;processing, by one or more interconnection assessment tools of a plurality of interconnection assessment tools corresponding with one or more interconnection assessment agents, the project data and the knowledge graph, wherein processing comprises:validating, by one or more first tools of the plurality of interconnection assessment tools, the first identification of the set of parcels using a second identification of the set of parcels;validating, by one or more second tools of the plurality of interconnection assessment tools, that the first electrical data satisfies one or more criteria in accordance with second electrical grid data of the know ledge graph;in response to validating the first identification of the set of parcels and the first electrical data:generating, by one or more third tools of the plurality of interconnection assessment tools, an impact assessment that reflects one or more effects of modification to the electrical grid specified by the interconnection request; andproviding, by at least one interconnection assessment agent, an interconnection request assessment comprising the impact assessment for the interconnection request and to a user.

2. The computer-implemented method of claim 1, wherein the first tools relate to parcel assessment.Attorney Docket No. 43374-0925WO13. The computer-implemented method of any one of claims 1-2, wherein the second tools relate to assessment of the electrical grid.

4. The computer-implemented method of any one of claims 1-3, wherein obtaining the knowledge graph comprises:processing data comprising grid element specification parameters for each of one or more grid elements and operational parameters to generate the knowledge graph for the electrical grid.

5. The computer-implemented method of any one of claims 1-4, further comprising obtaining the second identification of the set of parcels from one or more databases that comprise ground truth geolocation data.

6. The computer-implemented method of any one of claims 1-5, wherein the first identification of the set of parcels comprises first geographic boundary data for each parcel of the set of parcels, and wherein the second identification of the set of parcels comprises second target geographic boundary data for each parcel of the set of parcels.

7. The computer-implemented method of any one of claims 1-6, wherein validating the first identification of the set of parcels using the second identification of the set of parcels comprises:identifying a discrepancy that characterizes conflicting land use between the first identification of the set of parcels and the second identification of the set of parcels.

8. The computer-implemented method of any one of claims 1-7, wherein processing comprises processing the project data against the nodes of the knowledge graph.

9. The computer-implemented method of any one of claims 1-8, wherein the first electrical data of the interconnection request comprises one or more connection configurations, and wherein validating that the first electrical data satisfies one or more criteria in accordance with second electrical grid data of the knowledge graph comprises: extracting, by one or more second tools of the plurality of interconnection assessment tools, a first plurality of electrical parameters corresponding with the one or more connection configurations of the interconnection request;Attorney Docket No. 43374-0925WO1determining, from the knowledge graph, a second plurality of electric parameters of the electrical grid that correspond with the one or more connection configurations of the interconnection request; anddetermining whether the first plurality of electrical parameters satisfy a set of operational criteria in accordance with the second plurality of electrical parameters.

10. The computer-implemented method of any one of claims 1-9. wherein generating, by one or more third tools of the plurality of interconnection assessment tools, the impact assessment of the interconnection request comprises:determining a measure of impact caused by the modification to the electrical grid specified by the interconnection request.

11. The computer-implemented method of claim 10, wherein determining the measure of impact comprises:generating, using one or more grid simulation tools of the plurality of interconnection assessment tools, a respective set of data that characterizes the measure of impact caused by the modification to the electrical grid specified by the interconnection request by processing one or more electrical parameters corresponding with the interconnection request.

12. The computer-implemented method of claim 11, wherein the one or more grid simulation tools comprise a first grid simulation tool that corresponds with an electrical overload simulation and a second grid simulation tool that corresponds with an instability simulation.

13. The computer-implemented method of any one of claims 1-12, further comprising: generating respective impact assessments for a plurality of interconnection requests for the electrical grid; anddetermining, for each of the interconnection requests, a respective measure of similarity between the respective impact assessment of the interconnection request and one or more impact assessments for other interconnection requests.

14. The computer-implemented method of claim 13, further comprising:Attorney Docket No. 43374-0925WO1determining whether a discrepancy between a first impact assessment of a first interconnection request and each of the one or more other impact assessments for other interconnection requests exceeds a threshold tolerance criterion; andin response to determining that the discrepancy exceeds the threshold tolerance criterion, providing a notification to a user.

15. A system comprising one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:obtaining project data associated with an interconnection request, the project data comprising a (i) parcel data specifying first identification of a set of parcels of land and (ii) first electrical data specifying a modification to the electrical grid;obtaining a knowledge graph for the electrical grid comprising (i) a plurality of nodes representing each element of the electrical grid, and (ii) a plurality' of edges, each edge connecting a pair of nodes of the plurality of nodes and representing a connection between the elements of the electrical grid represented by the pair of nodes;processing, by one or more interconnection assessment tools of a plurality of interconnection assessment tools corresponding with one or more interconnection assessment agents, the project data and the knowledge graph, wherein processing comprises;validating, by one or more first tools of the plurality of interconnection assessment tools, the first identification of the set of parcels using a second identification of the set of parcels;validating, by one or more second tools of the plurality of interconnection assessment tools, that the first electrical data satisfies one or more criteria in accordance with second electrical gnd data of the knowledge graph;in response to validating the first identification of the set of parcels and the first electrical data:generating, by one or more third tools of the plurality of interconnection assessment tools, an impact assessment that reflects one or more effects of modification to the electrical grid specified by the interconnection request; andproviding, by at least one interconnection assessment agent, an interconnection request assessment comprising the impact assessment for the interconnection request and to a user.Attorney Docket No. 43374-0925WO116. The system of claim 15, wherein the first electrical data of the interconnection request comprises one or more connection configurations, and wherein validating that the first electrical data satisfies one or more criteria in accordance with second electrical grid data of the knowledge graph comprises:extracting, by one or more second tools of the plurality7of interconnection assessment tools, a first plurality of electrical parameters corresponding with the one or more connection configurations of the interconnection request;determining, from the knowledge graph, a second plurality of electric parameters of the electrical grid that correspond w ith the one or more connection configurations of the interconnection request; anddetermining whether the first plurality of electrical parameters satisfy a set of operational criteria in accordance with the second plurality of electrical parameters.

17. The system of claim 1, wherein generating, by one or more third tools of the plurality of interconnection assessment tools, the impact assessment of the interconnection request comprises:determining a measure of impact caused by the modification to the electrical grid specified by the interconnection request.

18. One or more non-transitory computer storage media storing instructions that when executed by one or more computers cause the one or more computers to perform operations comprising:obtaining project data associated with an interconnection request, the project data comprising a (i) parcel data specifying first identification of a set of parcels of land and (ii) first electrical data specifying a modification to the electrical grid;obtaining a knowledge graph for the electrical grid comprising (i) a plurality of nodes representing each element of the electrical grid, and (ii) a plurality7of edges, each edge connecting a pair of nodes of the plurality of nodes and representing a connection between the elements of the electrical grid represented by the pair of nodes;processing, by one or more interconnection assessment tools of a plurality of interconnection assessment tools corresponding yvith one or more interconnection assessment agents, the project data and the knowledge graph, wherein processing comprises:Attorney Docket No. 43374-0925WO1validating, by one or more first tools of the plurality of interconnection assessment tools, the first identification of the set of parcels using a second identification of the set of parcels;validating, by one or more second tools of the plurality of interconnection assessment tools, that the first electrical data satisfies one or more criteria in accordance with second electrical grid data of the knowledge graph;in response to validating the first identification of the set of parcels and the first electrical data:generating, by one or more third tools of the plurality7of interconnection assessment tools, an impact assessment that reflects one or more effects of modification to the electrical grid specified by the interconnection request; andproviding, by at least one interconnection assessment agent, an interconnection request assessment comprising the impact assessment for the interconnection request and to a user.

19. The non-transitory computer storage media of claim 18, wherein the first electrical data of the interconnection request comprises one or more connection configurations, and wherein validating that the first electrical data satisfies one or more criteria in accordance with second electrical grid data of the knowledge graph comprises:extracting, by one or more second tools of the plurality of interconnection assessment tools, a first plurality of electrical parameters corresponding with the one or more connection configurations of the interconnection request;determining, from the knowledge graph, a second plurality' of electric parameters of the electrical grid that correspond with the one or more connection configurations of the interconnection request; anddetermining whether the first plurality of electrical parameters satisfy a set of operational criteria in accordance with the second plurality' of electrical parameters.

20. The non-transitory computer storage media of claim 18. wherein generating, by one or more third tools of the plurality of interconnection assessment tools, the impact assessment of the interconnection request comprises:determining a measure of impact caused by the modification to the electrical grid specified by the interconnection request.

Images