TECHNICAL FIELD

[0001] The present disclosure is directed, in general, to a system and method for automatically selecting design parameters during an engineering design process, and more specifically to such a system that monitors inputs provided by a designer to predict the needed design parameters.

BACKGROUND

[0002] Design engineers spend a significant portion of their time selecting the correct parameters for their designs. Typically, an iterative process is involved in which the engineer selects, designs, evaluates, and the adjusts the parameter. A design may encompass a part design, a system design, a manufacturing process design or other designs typically prepared by engineers. Usually it takes many iterations until the correct parameters are set. In the case of a system design, it takes even more time since higher complexity systems often include many different specialties (e.g., mechanical, electrical, materials, structural, etc.).

SUMMARY

[0003] A system for aiding in the design of an engineered product includes a computer including a processor and a readable storage media having computer-executable instructions including at least one engineering program. An observer application is associated with the engineering program and is operable to monitor inputs to the computer made by a user to identify an observed pattern, a knowledge graph includes data related to the engineered product, and an explorer application searches the knowledge graph to find a stored pattern that corresponds to the observed pattern, An insighter application is operable to identify and present a suggested parameter to the user based on the stored pattern.

[0004] In another construction, a method of aiding in the design of an engineered product includes monitoring with a computer, inputs to the computer made by a user, identifying an observed pattern to the inputs, applying the inputs to a knowledge graph to expand the content of the knowledge graph, and searching the knowledge graph to find a stored pattern that corresponds to the observed pattern. The method further includes applying the stored pattern to present a predicted parameter to the user, the predicted parameter corresponding to a subsequent step in the design process.

[0005] In another construction, a non-transitory computer readable storage media having computer-executable instructions, when executed by a processor in a computer, performs a method for the design of an engineered product, the instructions include monitoring inputs to the computer made by a user, identifying an observed pattern to the inputs, applying the inputs to a knowledge graph to expand or refine the content of the knowledge graph, searching the knowledge graph to find a stored pattern that corresponds to the observed pattern, and applying the stored pattern to present a predicted parameter to the user, the predicted parameter corresponding to a subsequent step in the design process.

[0006] The foregoing has outlined rather broadly the technical features of the present disclosure so that those skilled in the art may better understand the detailed description that follows.

Additional features and advantages of the disclosure will be described hereinafter that form the subject of the claims. Those skilled in the art will appreciate that they may readily use the conception and the specific embodiments disclosed as a basis for modifying or designing other structures for carrying out the same purposes of the present disclosure. Those skilled in the art will also realize that such equivalent constructions do not depart from the spirit and scope of the disclosure in its broadest form.

[0007] Also, before undertaking the Detailed Description below, it should be understood that various definitions for certain words and phrases are provided throughout this specification and those of ordinary skill in the art will understand that such definitions apply in many, if not most, instances to prior as well as future uses of such defined words and phrases. While some terms may include a wide variety of embodiments, the appended claims may expressly limit these terms to specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Fig. 1 is a schematic illustration of a design process for an engineered product.

[0009] Fig. 2 is a schematic illustration of potential inputs to be considered during the design process of Fig. 1.

[0010] Fig. 3 is a schematic illustration of a design process for an engineered product including a computer-implemented enhanced design system.

[0011] Fig. 4 is another more detailed schematic illustration of the computer-implemented enhanced design system of Fig. 3.

[0012] Fig. 5 is another more detailed schematic illustration of the computer-implemented enhanced design system of Fig. 3.

[0013] Fig. 6 is an enlarged and more detailed portion of schematic illustration of Fig. 5.

[0014] Fig. 7 is a detailed schematic of a distiller for use in the system illustrated in Fig. 4.

[0015] Fig. 8 is a detailed schematic of a discover phase of the design process of Fig. 3.

[0016] Fig. 9 is a schematic illustration of a portion of a knowledge graph.

[0017] Fig. 10 is an enlarged schematic illustration of a portion of the knowledge graph of Fig.

9.

[0018] Fig. 11 is a schematic illustration of another arrangement of the knowledge graph.

[0019] Fig. 12 is a schematic illustration of a prediction model for the enhanced design system of Fig. 3.

[0020] Fig. 13 is a circuit diagram for an example operational amplifier and filter.

[0021] Fig. 14 is a schematic illustration of the design process for the circuit of Fig. 13.

[0022] Fig. 15 illustrates a recommendation from an insighter made during the design of the circuit of Fig. 13. [0023] Before any embodiments of the invention are explained in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting.

DETAILED DESCRIPTION

[0024] Various technologies that pertain to systems and methods will now be described with reference to the drawings, where like reference numerals represent like elements throughout.

The drawings discussed below, and the various embodiments used to describe the principles of the present disclosure in this patent document are by way of illustration only and should not be construed in any way to limit the scope of the disclosure. Those skilled in the art will understand that the principles of the present disclosure may be implemented in any suitably arranged apparatus. It is to be understood that functionality that is described as being carried out by certain system elements may be performed by multiple elements. Similarly, for instance, an element may be configured to perform functionality that is described as being carried out by multiple elements. The numerous innovative teachings of the present application will be described with reference to exemplary non-limiting embodiments.

[0025] Also, it should be understood that the words or phrases used herein should be construed broadly, unless expressly limited in some examples. For example, the terms“including,” “having,” and“comprising,” as well as derivatives thereof, mean inclusion without limitation.

The singular forms“a”,“an” and“the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. Further, the term“and/or” as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items. The term“or” is inclusive, meaning and/or, unless the context clearly indicates otherwise. The phrases“associated with” and“associated therewith,” as well as derivatives thereof, may mean to include, be included within, interconnect with, contain, be contained within, connect to or with, couple to or with, be communicable with, cooperate with, interleave, juxtapose, be proximate to, be bound to or with, have, have a property of, or the like.

[0026] Also, although the terms "first", "second", "third" and so forth may be used herein to refer to various elements, information, functions, or acts, these elements, information, functions, or acts should not be limited by these terms. Rather these numeral adjectives are used to distinguish different elements, information, functions or acts from each other. For example, a first element, information, function, or act could be termed a second element, information, function, or act, and, similarly, a second element, information, function, or act could be termed a first element, information, function, or act, without departing from the scope of the present disclosure.

[0027] In addition, the term "adjacent to" may mean: that an element is relatively near to but not in contact with a further element; or that the element is in contact with the further portion, unless the context clearly indicates otherwise. Further, the phrase“based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise. Terms“about” or“substantially” or like terms are intended to cover variations in a value that are within normal industry manufacturing tolerances for that dimension. If no industry standard as available a variation of twenty percent would fall within the meaning of these terms unless otherwise stated.

[0028] The engineering design process 10 is often an iterative process that can be time- consuming and sometimes inefficient. Fig. 1 schematically illustrates the typical design process 10 as including a Define phase 15 in which the problem to be solved and the initial parameters of the design are defined, a Create phase 20 in which the design is refined, and additional parameters are defined, and an Evaluate phase 25 in which the completed design is evaluated and tested. During or after the Evaluate phase 25 there are often iterations 27 in which initial or subsequent parameters must be changed or adjusted. Often, this reverts the design process back to the Define phase 15 or the Create phase 20 This process continues until the designer arrives at a final design that satisfies the initial requirements.

[0029] Throughout this description reference is made to the following phases, programs, applications, functions, and pieces of code: Define Phase 15 - Portion of the overall design process in which the requirements, goals, operational characteristics and the like are defined.

Create Phase 20 - Portion of the overall design process in which the product design and any models are created.

Evaluate Phase 25 - Portion of the overall design process in which the design completed in the Create Phase 20 is evaluated against the parameters set in the Define Phase 15.

Capture Phase 30 - Process that runs in an enhanced design system 35 to capture inputs provided by the designer during the design process. The inputs are made using various software tools including but not limited to 3D modeling software, CAD software, spreadsheets, Internet searches, etc. as well as other inputs.

Explore Phase 40 - Process that runs in the enhanced design system 35 to analyze the various inputs to determine if suggestions such as design parameters 45 could be provided to the designer to enhance the process or design.

Discover Phase 50 - Process that runs in the enhanced design system 35 to make parameter or other recommendations to the designer during the design process.

Knowledge Graph, sometimes referred to as Digital Twin Graph 55 - Database including accumulated knowledge related to the particular design process to which the graph is associated. Each knowledge graph 55 includes nodes 60 and connections 65 between the nodes 60.

Neural Network 70 - An arrangement of data based on a collection of connected units or nodes 60 that are analogous to artificial neurons, which loosely model the neurons in a biological brain.

Nodes 60 - Bits of accumulated data stored in the knowledge graph 55.

Connections 65 - Links between different nodes 60. The connections 65 are logical links between related pieces of data. Digital Twin 75 - A fully operational model of a component or system that includes the component being designed.

Observer Programs 80 - Add-ins or plug-ins for pre-existing engineering tools that gather information from the designers as they work. Information such as keystrokes, drawing values, log files, engineering calculations, and the like can be gathered by observer programs 80.

Distiller Programs 85 - Review all the available data and reduce that data to useful pieces of information that can be provided to the knowledge graph 55. In addition, relationships between various pieces of data can be ascertained and provided to the knowledge graph 55 to assure that the knowledge graph 55 contains consistent, useful, and helpful information.

Alternative Generator Programs 90 - Deliver possible design parameters 45, alternative designs, or other choices to the designer during the design process.

Insighter Programs 95 - Programs that use patterns gathered from observer 80 and distiller programs 85 and artificial intelligence approaches to predict the next step or other options in the design process and to present those options to the designer.

Digital Twin Graph Algorithms 100 - Programs that explore past designs, build relations between the nodes 60 and improve knowledge over time.

Causal Explorer 105 - Records the evolution of the design workflows in the digital twin graph 55 and identifies causal links or connections 65 between nodes 60 to identify recurring design and engineering practices.

Past Designs Explorer 110 - Compares the design workflow with previous design workflows stored in the digital twin graph 55.

Recommendation Generator 115 - Program that delivers possible alternative designs or other choices to the designer during the design process. System Vision Server 120 - Server of a free cloud-based simulation tool to design and simulate complete analog, digital, mixed signal and electro-mechanical systems.

Mixture Density Network (MDN) 125 - A class of models obtained by combining a conventional neural network with a mixture density model. The mixture density model represents the conditional probability density function of the target variables conditioned on the input vector of the neural network.

[0030] Fig. 2 illustrates additional details that might be involved in the design process of a component or a part 130. The left side of the diagram illustrates design steps such as defining requirements 135 for the design (i.e., the Define phase) and the development of a system model 140 and the system architecture (i.e., the Create phase). Finally, a 3D design 145 is completed and evaluated with iteration steps possible back to any of the prior steps. Once the design is complete, assembly and component models can be created or completed. The right side of the schematic of Fig. 2 illustrates other design inputs or considerations that can complicate the design process. For example, some designs may require the designer to consider the supply chain 150 for the part or parts, the production schedule 155 for certain parts, and the

manufacturability 160 of the final design. All of these are considered by the designer. However, these considerations often arise late in the design process and require additional iterations that further slow the design.

[0031] Figs. 3-11 illustrate a computer-implemented enhanced design system 35 that utilizes advanced artificial intelligence (AI) to enhance the design process just described in an effort to reduce wasted time, increase engineering productivity, and produce superior quality designs.

[0032] The software aspects of the present invention could be stored on virtually any computer readable medium including a local disk drive system, a remote server, internet, or cloud-based storage location. In addition, aspects could be stored on portable devices or memory devices as may be required. The computer generally includes an input/output device that allows for access to the software regardless of where it is stored, one or more processors, memory devices, user input devices, and output devices such as monitors, printers, and the like. [0033] The processor could include a standard micro-processor or could include artificial intelligence accelerators or processors that are specifically designed to perform artificial intelligence applications such as artificial neural networks, machine vision, and machine learning. Typical applications include algorithms for robotics, internet of things, and other data- intensive or sensor-driven tasks. Often AI accelerators are multi-core designs and generally focus on low-precision arithmetic, novel dataflow architectures, or in-memory computing capability. In still other applications, the processor may include a graphics processing unit (GPU) designed for the manipulation of images and the calculation of local image properties.

The mathematical basis of neural networks and image manipulation are similar, leading GPUs to become increasingly used for machine learning tasks. Of course, other processors or

arrangements could be employed if desired. Other options include but are not limited to field- programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), and the like.

[0034] The computer also includes communication devices that may allow for communication between other computers or computer networks, as well as for communication with other devices such as machine tools, work stations, actuators, controllers, sensors, and the like.

[0035] Fig. 3 is a simplified schematic of a portion of the enhanced engineering design process 35 that is enhanced by the design system. In this portion of the process, the designer is providing inputs to the design using various software tools including but not limited to 3D modeling software, CAD software, spreadsheets, Internet searches, etc. This is illustrated as the capture phase 30. In the explore phase 40, the design system 35 analyzes the various inputs to determine if suggestions such as design parameters 45 could be provided to the designer to enhance the process. In addition, the explore phase 40 can use simulation tools to validate any parameters before they are suggested. Recommendations, provided in the discover phase 50, could include design parameters 45 or could include recommendations based on manufacturability, material selection, etc.

[0036] Figs. 4 and 5 illustrates the capture phase 30, the explore phase 40, and the discover phase 50 in additional detail and each interacting with the knowledge graph 55, sometimes referred to as a digital twin graph 55 and potentially one or more digital re-creations or digital twins 75 of devices or systems similar to that being designed by the designer. [0037] As noted, the capture phase 30 is used to gather data from the designer as well as from other available sources. As illustrated in Figs. 4 and 5, an engineer or designer at work uses multiple different software tools 165 such as NX, StarCCM, etc. An observer program 80, or multiple observer programs 80 monitor the inputs of the designer in these tools 165 and transmit that information to a distiller program 85. Typically, observer programs 80 are add-ins or plug ins for pre-existing engineering tools 165. Similarly, engineering data 170 from other programs such as EXCEL, MATHCAD, and the like is gathered and transmitted to the distiller program 85. Requirements of the design 175 may be stored in another location or program. These requirements 175 may include size limitations, cost limitations, performance requirements and the like and are also sent to the distiller program 85. In some cases, components or systems similar to those being designed are in use or operation and actual operating data 180 is available. If available, this data 180 can be provided to the distiller program 85 as well.

[0038] The distiller program 85, reviews all the available data and reduces that data to useful pieces of information that can be provided to the knowledge graph 55. In addition, relationships between various pieces of data can be provided to the knowledge graph 55 to assure that the knowledge graph 55 contains useful and helpful information. The knowledge graph 55 also contains links that might lead a designer to the next important piece of information needed in the design process.

[0039] In the explore phase 40, the computer interacts with the knowledge graph 55, the new design, and the designer to provide alternative designs 185 and validate and rank them through analysis and simulation. Specifically, the computer runs an alternative generator program 90 that delivers possible alternative designs 185 or other choices to the designer during the design process. The alternative generator 90 runs algorithms that generate new designs from seeds of previously generated design points and knowledge from the digital twin graph 55. In addition, the computer runs quantitative analysis 190 (quants) to evaluate the design as it develops. The quants 190 predict the performance of a given design, either through simulation or through learning from past experience stored in the digital twin graph 55. The results of these quants 190 are delivered to the designer during the design process to further advance the design. [0040] Fig. 12 schematically illustrates one possible technique used in the Explore phase to predict parameters. The system 35 includes forward prediction 195 in which the system 35 leams from past experience and provides suggestions for a given design parameter 45. Inverse prediction 200 is also employed. Inverse prediction 200 provides alternative suggestions based on the suggestions arrived at using forward prediction 195. Finally, the system 35 performs optimization analysis to arrive at the optimum suggestion for a design parameter 45.

[0041] In the discover phase 50, the system 35 includes an observer program 80 that observes the actions of the designer to discover the intention of the designer, design requirements, patterns, or features of the design and compares those discoveries to the knowledge graph 55.

An insighter program 95 uses these patterns to predict the next step or other options and presents them to the designer.

[0042] Figs. 4-6 also illustrate digital twin graph algorithms 100 that operate to explore past designs, build relations between the nodes 60, and improve knowledge over time. The digital twin graph algorithms 100 include a causal explorer 105 and a past designs explorer 110. The causal explorer 105 records the evolution of the design workflows in the digital twin graph 55 and identifies causal links to other digital twin graph nodes 60 to identify recurring design and engineering practices. The past designs explorer 110 compares the design workflow with previous design workflows stored in the digital twin graph 55. Insight on analyses results performed in the past on those cases is used to inform the designer and pre- validate the designs.

[0043] Fig. 6 illustrates a portion of Fig. 5 and includes additional data sources for the knowledge graph. The additional data includes databases or data sources 205 that may provide valuable information to the knowledge graph 55. In this example, team center data 210 (data in a team-based database) is available to the knowledge graph 55. Team center data 210 may include specifications and limitations that effect the design. Other databases or sources of data could include online engineering resources, materials databases, engineering tables and the like.

Additional databases may include operating data from prior similar designs or similar devices.

All this data can be incorporated into the knowledge graph 55 along with links or connections 65 between the data points or nodes 60. [0044] Fig. 7 better illustrates the operation of one of the distillers 85, specifically the engineer at work distiller 85. Engineers at work or designers often interact with engineering tools such as 3D modelers, CAD systems, CAM systems, and the like. The distiller 85 collects this information in the form of mouse events, keyboard events, screen captures, and the like and distills that collected data into useful knowledge. Additional useful knowledge can be transferred to the knowledge graph via engineer feedback 215. The useful knowledge is stored in the knowledge graph 55 along with links between related data. In addition, the distiller 85 determines a design intent from the processes knowledge that can be used to generate recommendations for the designer. In developing recommendations, a recommendation generator 115 may use knowledge from the knowledge graph and engineer feedback 215 to develop the recommendation. The recommendation generator 115 provides recommendations to the designer using a user interface, auto completion, a script player, or other communication means.

[0045] As discussed with regard to Figs. 4 and 5, in the discover phase 50, the system 35 observes the actions of the designer to discover patterns or features of the design and compare those patterns or features to the knowledge graph 55. Fig. 8 provides additional details of this process. As discussed, the observer program 80 generally provides data to the knowledge graph 55 while the insighter program 95 takes knowledge from the knowledge graph 55 and provides suggestions to the user. In the example illustrated in Fig. 8, the engineering tool is SIEMENS NX, a common CAD software provided by SIEMENS PLM and the insighter program 95 has provided a suggested design parameter 45 and asked if the designer would like to use that parameter 45. As discussed, the patterns and components used by the designer are searched in the knowledge graph 55 to find knowledge relating to similar patterns or components. The knowledge graph 55 includes links 65 that lead to additional knowledge 60 that may be helpful to the designer and which can be passed on to the designer via the insighter program 95.

[0046] Turning to Fig. 9 a schematic illustration of a portion of a knowledge graph 55 is provided. The knowledge graph 55 is essentially a neural network 70 including nodes 60 and connections 65. Each node 60 represents a piece of knowledge with each connection 65 representing a link between different pieces of knowledge. For example, the designer may be designing a turbine blade 220 as illustrated in Fig. 10. The insighter program 95, based in part on past similar products might suggest a particular material for the manufacture of the blade 220. Upon selecting the material, the central node 60 of Fig. 10 is identified. The connections 65 between the central node 60 and a first layer of secondary nodes 60a lead the insighter program 95 to possible suggestions for the designer. For example, the secondary nodes 60a could provide information regarding the cost of the selected material, manufacturing processes that are required for the selected material, material properties of the material, and other information that might be helpful. In addition, each secondary node 60a can lead to tertiary nodes 60b that include knowledge specific to the secondary nodes 60a and so on. For example, the secondary node 60a that includes the cost of the material may be connected to tertiary nodes 60b that contain information about possible material suppliers.

[0047] It should be clear that the nodes 60 can include virtually any type of knowledge including parts, available machines, manufacturing processes, machine parameters, process parameters, component parameters, relationships, sequence of events, etc. In addition, the connections 65 can define a sequence or order. For example, the selection of the material may provide a connection 65 that leads to a machining process, a connection 65 from the machining process might then lead to a node 60 including a polishing process, which includes a connection 65 that leads to a node 60 that includes a quality control process.

[0048] Fig. 11 illustrates a different arrangement of the data within the knowledge graph 55. In this arrangement, the nodes 60 are arranged or grouped by the type of data they contain. For example, nodes 60 that relate to operational issues may be categorized and/or stored in an operation region 225. Other categories could include design features 230, manufacturing features 235, process features 240, or process segment features 241, materials, suppliers, etc. By sorting the data in this manner, like nodes 60 can be compared to one another more easily.

Connections 65 extend between nodes 60 and between groups as described before.

[0049] Figs. 13-15 illustrate an example of how the system 35 aids an engineer during the design of a printed circuit board 245 including a low pass filter and amplifier in which a desired gain 250 and frequency 255 is known. Fig. 13 illustrates the basic circuit with the frequency 255 and gain 250 identified. Components including an amplifier 257, an input resistor 260, a ground resistor 265, an amplifier bypass resistor 270, and a capacitor 275. Each of these five components must be identified and optimized to arrive at the desired frequency 255 and gain 250.

[0050] Using a dedicated system vision server 120, multiple circuit parameterizations can be simulated for each of the four possible topologies in this example, or in general for a given set of known topologies for which simulation data can be acquired.

[0051] Given a desired gain 250 and frequency 255, a machine learning approach (Mixture Density Networks - MDN - in this example) 125 is used to choose the design parameters by sampling multiple alternative designs, predicting their gain 250 and frequency 255, and then choosing the design whose gain 250 and frequency 255 are closest to the desired values.

[0052] It is important to note that continued use of the system 35 will result in the collection of additional reusable engineering data, tacit knowledge, data, requirements, and product use data. This data becomes accessible and usable for future projects, designs, and decisions.

[0053] Although an exemplary embodiment of the present disclosure has been described in detail, those skilled in the art will understand that various changes, substitutions, variations, and improvements disclosed herein may be made without departing from the spirit and scope of the disclosure in its broadest form.

[0054] None of the description in the present application should be read as implying that any particular element, step, act, or function is an essential element, which must be included in the claim scope: the scope of patented subject matter is defined only by the allowed claims.

Moreover, none of these claims are intended to invoke a means plus function claim construction unless the exact words "means for" are followed by a participle.