System for automated generative design synthesis using data from design tools and knowledge from a digital twin

{
  "publication_number": "US11423189B2",
  "country": "US",
  "kind": "B2",
  "title": "System for automated generative design synthesis using data from design tools and knowledge from a digital twin",
  "abstract": "A system for autonomous generative design in a system having a digital twin graph a requirements distillation tool for receiving requirements documents of a system in human-readable format and importing useful information contained in the requirements documents into the digital twin graph, and a synthesis and analysis tool in communication with the digital twin graph, wherein the synthesis and analysis tool generates a set of design alternatives based on the captured interactions of the user with the design tool and the imported useful information from the requirements documents. The system may include includes a design tool with an observer for capturing interactions of a user with the design tool, In addition to the observer, an insighter in communication with the design tool and with the digital twin graph receives design alternatives from the digital twin graph and present the receive design alternatives to a user via design tool.",
  "claims": [
    "1. A system for autonomous generative design in a system having a digital twin graph, the system comprising: a requirements distillation tool for receiving requirements documents of a system in human-readable format, extracting and importing information relating to a requirement contained in the requirements documents into the digital twin graph; a synthesis and analysis tool in communication with the digital twin graph, wherein the synthesis and analysis tool generates and evaluates a set of design alternatives based on information relevant to a design problem contained in the digital twin graph; and an observer in communication with a design tool for capturing interactions of a user with the design tool relating to design of the system, wherein the observer captures time series of user interactions, identifies patterns of user interactions as an action or reaction of the user and includes the action or reaction of the user to the digital twin graph.",
    "2. The system of claim 1, further comprising: an insighter in communication with the design tool and with the digital twin graph, the insighter configured to receive design alternatives from the digital twin graph and present the received design alternatives to a user via the design tool.",
    "3. The system of claim 1 further comprising: a requirements distiller in communication with the digital twin graph, the requirements distiller receiving requirements documents in human-readable format and extracting information relating to a requirement from the requirements documents and importing the information into the digital twin graph.",
    "4. The system of claim 3, wherein the requirements distiller performs the steps of: receiving requirements documents in human-readable form; converting the human-readable documents into a digital form; analyzing the digital documents to identify data contained in the documents; compressing the data contained in the documents to extract information related to a requirement from the data in the requirements documents; and importing the information into the digital twin graph.",
    "5. The system of claim 1, the synthesis and analysis tool comprising: an alternatives generator for generating a set of design alternatives based on requirements and relationships between components of the system represented by the digital twin graph; a quantification verifier for verifying each design alternative based on the requirements and design constraints contained in the requirements documents, and computing certain metrics to quantify a quality of a design alternative; and a causal explorer for identifying design workflows to identify design and engineering practices.",
    "6. The system of claim 5, wherein the causal explorer creates a recommended sequence of operations for a user based on the identified design workflows.",
    "7. The system of claim 1, further comprising: a product in use (PIU) data distiller in communication with the digital twin graph, the PIU data distiller receiving operations data from a physical system represented in the digital twin graph and extracting information for importation to the digital twin graph.",
    "8. The system of claim 7, wherein the PIU data distiller performs the steps of: receiving product in use data from components of the system; converting the product in use data into a digital form; analyzing the digital data to identify information contained in the data; compressing the information from the PIU data to extract the information from the PIU data; and importing the information into the digital twin graph.",
    "9. The system of claim 1, further comprising: an engineering history distiller in communication with the digital twin graph, the engineering history distiller receiving data relating to past engineering design choices and extracting information from the past engineering design choices and importing the information into the digital twin graph.",
    "10. A method of autonomous generative design in a system including a digital twin graph, the method comprising: distilling requirements documents to extract information relating to at least one requirement and importing the information into the digital twin graph; capturing interactions between a user as a time-series of user actions and a design tool related to design of the system and importing the interactions to the digital twin graph; identifying a pattern from the time-series of user actions as an action or reaction of the user and including the action or reaction in the digital twin graph; synthesizing a set of design alternatives based on the information of the requirements documents and the captured interactions between the user and the design tool; verifying and quantifying each design alternative in the set of design alternatives against the information of the requirements documents; and suggesting to the user, a selected design alternative of the set of design alternatives.",
    "11. The method of claim 10, further comprising: analyzing each design alternative in the set of design alternatives by performing a simulation for each design alternative in the set of design alternatives; and choosing the selected design alternative based on the analysis and quantified metrics.",
    "12. The method of claim 10, further comprising: in a design tool, receiving commands from a user representative of a design operation; in a synthesis and analysis tool, generating a suggested design operation based on an analysis of the set of design alternatives and providing the suggested design operation to the user; and alternatively receiving commands from the user and generating a suggested design operation and presenting the suggested design operation to the user and repeating this step until the design process is complete.",
    "13. The method of claim 10, wherein distilling the requirements documents comprises: receiving requirements documents in human-readable form; converting the human-readable documents into a digital form; analyzing the digital documents to identify data contained in the documents; compressing the data contained in the documents to extract information relating to at least one requirement from the data in the requirements documents; and importing the information into the digital twin graph.",
    "14. The method of claim 10, further comprising: importing product in use data relating to operations of a physical system represented in the digital twin graph, to the digital twin graph for use in generating the set of design alternatives.",
    "15. A method for autonomous generative design of a system having a digital twin graph, comprising: importing information from requirements documents into the digital twin; capturing and importing user interactions as a time-series of actions of the user with a software design tool related to design of the system and importing the interactions into the digital twin graph; identifying a pattern from the time-series of user actions as an action or reaction of the user and including the action or reaction in the digital twin graph; synthesizing a set of design alternatives based on the imported user interactions and the imported requirements information; simulating the system in the digital twin graph for each of the set of design alternatives; and selecting an optimal design alternative based on the simulations.",
    "16. The method of claim 15, further comprising: distilling data from the requirements documents by performing the steps of: receiving requirements documents in human-readable form; converting the human-readable documents into a digital form; analyzing the digital documents to identify data contained in the documents; compressing the data contained in the documents to extract information relating to at least one requirement from the data in the requirements documents; and importing the information into the digital twin graph.",
    "17. The method of claim 15 further comprising: in a synthesis and analysis tool: generating a set of design alternatives based on semantic information represented in the digital twin graph; verifying each of the design alternatives against the requirements information; identifying design operations corresponding to captured interactions between a human user and a software design tool; and running a simulation in the digital twin graph for each of the design alternatives.",
    "18. The method of claim 17 further comprising: selecting a design operation for performing an optimal design alternative based on the simulations.",
    "19. The method of 17, further comprising: importing product in use information from a physical system represented by the digital twin graph into the digital twin graph for use in the simulating of the system."
  ],
  "description_excerpt": "This application relates to design of systems. More particularly, the application relates to a system of generative design aided by a digital twin.\n\nDesigning a complex system, such as a gas turbine, automobile or aircraft requires a substantial amount of manual effort. Computer-aided design (CAD) software has provided some relief providing computer-based efficiencies. For example, CAD systems have made it possible to create and analyze detailed virtual models of systems and their constituent parts and assemblies. CAD systems may also digitally simulate the performance and behavior of the parts and assemblies. Computer aided systems may also generate manufacturing plans and instructions for the production of parts and systems. CAD systems may further assist in managing design related information in digital form over the lifecycle of the produce and provide information to aid in collaborative product development.\n\nDespite the advances of computer aided processes, many engineering tasks still require manual intervention and are tedious and time-consuming. Moreover, domain expertise is tied up in human capital, requiring experts with many years of experience to be present and available to generate design variants, setup analysis models, interpret results, fine tune design parameters to meet design requirements and create suitable manufacturing process plans.",
  "cpc": [
    "G06F 30/12",
    "G06F 2111/04",
    "G06F 30/00"
  ],
  "ipc": [
    "G06F 30/12"
  ],
  "assignees": [
    "Siemens AG"
  ],
  "inventors": [
    "Livio Dalloro",
    "Edward Slavin, III",
    "Sanjeev Srivastava",
    "Lucia Mirabella",
    "Suraj Ravi Musuvathy",
    "Arquimedes Martinez Canedo",
    "Erhan Arisoy"
  ],
  "filing_date": "2018-03-27",
  "publication_date": "2022-08-23",
  "grant_date": "2022-08-23",
  "priority_date": "2017-03-27",
  "application_number": "US-201816496678-A",
  "family_id": "62028108",
  "cited_by_count": 3,
  "citations": [
    "US5822206A",
    "JP2002197122A",
    "US9424528B2",
    "JP2013518449A",
    "US20160188675A1",
    "US20160247129A1",
    "US20180137219A1"
  ]
}