Webinar: Digital Mission Engineering Part 4

0:00 good afternoon and welcome to the webinar today's webinar is called extending STK mission models with detailed ANSYS engineering simulation to track hypersonic vehicles from space i'm josh are gonna be doing some of the behind the scenes stuff from this webinar before i pass it over to the presenters if you have any questions we want this to be interactive so over in the GoToWebinar there is a questions panel please type any of your questions in there and at the end of the

0:24 presentation after all the speakers go I will read them to them and they will do our best to answer them in the allotted time some of the questions may be something we'll have to follow up with but we will get to we will get to all of them maybe not live during the webinar but at some point we'll be answering anything that you put in there and like I said that we have a bunch of presenters from several different

0:44 companies all around the country today and I'm gonna not interest them all right now they will be introduced at their time up presenting but I'm gonna start off with Jeff Baxter who heads up the engineers here at AGI and Jeff over to you to kick off this presentation all right Thank You Josh and thanks to everyone for joining us here today special thanks to everyone that has attended multiple in this series where in the last series finale here the

1:08 fourth webinar and digital Mission engineering I want to thank this webinars co-presenters from ANSYS Ming Shan and Valerio and also special thanks to Joshua Edwards from Phoenix integration who is joining us for our third installation here today and so this is really the grand finale of this digital mission engineering series and part one we introduced what digital mission engineering is and why it's important part two began to show how to apply digital mission engineering starting with early concept development

1:39 and trade space exploration in part three we showed how to integrate descriptive models with physics-based mission models and then today I'm really excited because we're gonna really focus on the detailed engineering design phase and show how to extend the mission model with pervasive engineering simulation will show several demonstrations with our partners throughout that go into detail on a variety of areas including thermal communications and hypersonics and how that all fits within the digital mission engineering ecosystem we have a

2:10 lot of really great content to cover a lot of great presenters not a lot of time so let's go ahead and get started as we discussed in previous webinars digital emission engineering is the use of computer-based modeling simulation and analysis to evaluate mission objectives and really the focus is all about the mission when programs are created they're meant to meet a need a gap a requirement that will help achieve the desired mission objective and as a commercial software

2:38 provider working with hundreds of organizations throughout aerospace and defense we get a pretty broad perspective of the industry and one of the trends we're seeing especially right now is that systems are getting more complex at the same time that development cycles are getting faster and less forgiving and you see the arrow representing the the complexity increasing the delivery time decreasing which is putting a lot of pressure on the industry to improve our engineering practice and deliver better systems

3:06 faster and traditional methods of developing tools can not really keep up with the pace here and so an AMD traditional document based requirements mean lots of time and effort go into building a custom incompatible and unconnected models and tools at each level of component subsystem or system level and so digital mission engineering addresses these by moving toward a model-based engineering approach where the model becomes the ultimate source of truth model can be persist throughout the lifecycle of this system and as the

3:39 mission changes or requirements change and evolve the model can also change and evolve so what we see today across the industry is as I mentioned a lot of isolated models reinvention of stovepipe tools and no common thread across the program lifecycle so if we look at this in this diagram starting at the concept development phase of a program there are typically some basic mission models that may loosely depend on physics to help understand the concept of the system and

4:07 as the program matures a new mission model is typically created to handle more detailed physics models and typically there will be multiple individual tools that all do a certain type of function for their specific special area and then these models are then used to define the requirements thread which is then used throughout design and implementation and then ultimately out onto the testing phase and and this design implementation is typically the the least infer that the furthest separated from the mission

4:37 model and the mission level and once the system is finished new mission models are then created for that system to evaluate how it's performing against that mission need typically a new mission model is used for training operations and sustainment and the cycle then starts again as the initial system is deployed and future improvements are made and so as you can see this leads to a lot of isolated tools reinventing the wheel and it makes it really hard to tie everything

5:06 together and if this is just looking at a single program so as new programs come online even similar programs the process usually starts over from scratch and so the vision of digital mission engineering is to connect all of the component subsystem and system models to the mission model that is rooted in a multi domain physics environment furthermore these models need to be pulled in all and integrated as early as possible in the process and then persist

5:34 throughout the life cycle so an important concept here is this concept of a digital thread which represents how each of these models are individual strands that are tied together like a thread so that when you change individual parts of one model it pulls through the rest of the models as well and in order to achieve this vision it really requires an ecosystem of commercial tools that all integrate together from day one so without that it makes it really difficult to identify

6:02 critical issues early on the process which can limit the overall delivered effectiveness so in previous webinars we started building a representative demonstration that that highlights the early phases of this vision applied to satellite constellation design we quickly built a mission model of a constellation covering select ground locations integrated that with a cost model ransom cost perverse performance architecture level of trade studies and then linked those mission models to a descriptive model in sysml and today's

6:34 webinar we're going to build off this demonstration by using the orbit information calculated from that satellite constellation design and we're really going to focus on an individual spacecraft and dig into a lot of the technical details at the subsystem and component level and and one of the key enablers of this is our recent partnership with ANSYS and and the reason we're so excited about this new partnership is that we're really providing a full digital mission engineering stack that connects

7:02 components subsystems systems and system of systems to the mission model so when you think about the engineering enterprise mission level problems are broken down into executable pieces component development subsystem integration and system integrations in use these complex systems come together and a system of systems architecture must operate in a coordinated way across multiple domains and achieve the desired mission outcomes so really that's what the mission and what we're talking about

7:33 here and you can do this at multiple levels you can do it at the local level at the regional level and all the way up at the global level so really that's what this partnership enables connects all of these digital models from the components to the system of system architectures to successfully evaluate mission objectives and so an important component in digital engineering is also this concept of the digital twin which is there's several definitions out there

8:02 in the industry actually multiple industries and this is a simple definition which is an adaption from the DoD definition which is simply an integrated multiphysics simulation of a physical system that pairs the virtual world to the physical world and there's an article in Forbes magazine that described this process of the pairing of the virtual and physical worlds that allows analysis of data and monitoring systems to head off problems before they occur prevent downtime develop new

8:33 opportunities and even plan for the future by using simulations and so the whole idea here is that lessons are learned and opportunities are uncovered within virtual environment that can be applied to the physical world and was they described ultimately transforming your business so what we want to show here is now that we're talked about the concepts of mission engineering including the digital thread in the digital twin we want to show all these in action and so

9:00 we're going to jump into some demonstrations now that start going down that path and I'd like to start off first of all by thanking our friends at Astra digital and ASI who provided the CAD model and digital spacecraft simulation model for this Corvis BC spacecraft 6u CubeSat which is actually running the same flight software on orbit that we used in this simulation with STK soleus so we'll be using this Corvis BC spacecraft as the example satellite we'll be using as our notional

9:31 satellite constellation and we're going to really dig into a lot of detail on this spacecraft and on the left you can see the physical layout of the system on the right you can see some of the images of the engineering and mission simulations we've we've run so the computer CubeSat has several subsystems and today we're really going to be focusing on the attitude control system multi spectral imager and ka-band radio and we've modeled each of these

9:56 subsystems in SDK and ANSYS and we'll evaluate their ability to meet mission requirements across a couple different use cases so real quick this is the the inputs that we used for modeling the attitude control system there are three reaction wheels and some magnetic tour curves that we've defined and we set up what's called a Target plan based on a deck of targets both ground target locations as well as scans and then down linking the data to a ground station and

10:27 this is a snapshot of what the the mission simulation looks like you can see the spacecraft slewing to this particular target the sensor field of view collecting that target and the images on image on the upper right shows the reaction wheel speed in order to slew the spacecraft to collect these targets and in the lower right hand corner we see the thermal loading from from the Sun and the earth albedo and Earth at planets eye are the yellow

10:57 vector represents the direction of this Sun vector in this case and so this is the interface that we can expose to report all of those elements out from the mission model there are thousands and thousands of these parameters so in this case we chose the the flux from albedo the planet's IR and the Sun as the thermal loads that we want to then run into ANSYS for the the dynamic thermal loading on the spacecraft and all of the components itself so here's a

11:30 quick video that highlights this is a typical imaging pass our satellite is required to collect these ground targets and scans and then downlink that data to the ground station and you can see the relative geometry each of the faces is exposing different areas to the Sun to the earth albedo and the IR throughout the entire throughout the entire pass so you can see here that the mission environment is playing a huge role in our in our thermal analysis starting

11:58 with the orbit altitude and inclination from the constellation from previous webinars and now we've used solace to model the attitude control system to point the spacecraft at the targets and then we're going to pass this over to ANSYS for higher fidelity thermal analysis so with that I'll turn things over to Ming from ANSYS everyone so so this is a ministry from Isis I'm happy location wind engineer with an sis mechanical in the next few slides I will

12:27 present the demo for trends in thermal analysis of CubeSat using NCC mechanical coupled with the AGI STK tool next slide please thanks the thermo analysis of the CubeSat is performed in as his mechanical transient thermal sober when the CubeSat is in the orbit it's constantly exposed to solar his Flags and Earth's albedo and we're also really the heat into the space and his mechanical will take the heat flux time history of the CubeSat calculated by a

13:00 gr STK and calculated the thermal performance within the satellite I also want to mention that and his workbench is a mounted physics platform that can integrate different solvers together for example using the coupling analysis of its high frequency electromagnetic views over HF SS and transients thermistor enhances mechanic and his customers can simulate thermal Buda in external antennas and the internal electronics to check if antennas are subjected to a potential

13:32 risk of loss of efficiency and drifting of desired tuned frequency then considering the thermal strain calculated by the thermal server we can perform structural analysis model analysis for natural frequency and the random vibration analysis and so on in the trends in thermal analysis each face of the cube set will have different time history of heat heat flux and it's mechanical will apply the heat flux as boundary conditions and also add the radiations of the external surfaces and

14:02 the internal components of the satellite thus over will calculate the temperature distributions versus time and the heat flux within the satellite calculated values are can be then be used in structural analysis considering as the thermal effect this slide shows the temperature distribution within the satellite changing with time as you can see a different time point on the temperature will be changed this slide shows the heat flux in the satellite

14:34 during the simulated period of time this is also a tree of the heat flux and different areas of the satellite were changing with time so this will conclude my demo for the transient thermo analysis of CubeSat in ences with a high fidelity thermal solution calculated appliances mechanical the model is ready for analysis of other physics next speaker up is my colleague John Carpenter from ANSYS he

15:04 will talk about the utilization of high-frequency electronic tools on the CubeSat great thank you very much Ming and hello everyone my name is Sean Carpenter and I'm a product manager for high-frequency electromagnetic analysis with answers as jeff has already indicated this technical partnership between AGI and answers has the potential to more effectively connect your systems and mission engineering groups to your subsystems and components engineering teams by bringing high

15:35 fidelity six models into the mission simulation domain now next slide I'll be talking about how this might apply to the RF telemetry system develop development and this validation in our digital mission engineering simulation for our candidate CubeSat now we have here again illustrated the CubeSat components and the subsystems next and we're going to focus in this section of the presentation on the ka-band radio system and you can see the specifications for the radio there on the right-hand side

16:05 now in particular we're going to be looking at the antenna system and this system is operating at at ka-band so it's in the neighborhood of 30 gigahertz and you'll see here in particular that it has a main beam focus of about 10.2 degrees that's the half Power Beam width and it's this whole system fits into a very nice compact size it's inside of a 1u volume next slide please now the land mapper sat a CubeSat has

16:32 three optical sensors which you can see is the white openings that are kind of on the upper side of the of the body and then the ka-band aperture is shown with a more yellowish opening that you can see there on the body of the satellite now for this demonstration we'll be leveraging a high fidelity full physics electromagnetic analysis by the ANZUS high frequency structure solver or HF SS which is a high frequency field solver used to assess to performance measures

16:59 of interest for this study first of all we'll be looking at how the antenna performs in isolation this is typically how we would normally measure a prototype we put it in an anechoic chamber on a test table all by itself and we'd measure its radiation characteristics but second and perhaps most important for integrators is how the antenna performs when it's installed on the CubeSat there will be electromagnetic coupling between the aperture and the platform body the solar

17:27 panels and the other structures which may serve to alter the antenna aperture performance but the question is how much now we'll use the installed performance data from this simulation in the overall satellite mission simulations so that this realistic antenna radiation pattern is used then in the overall telemetry link budget model employed in the SDK software as the mission evolves now our e/m analysis will provide installed radiation patterns for both the coal

17:57 pole radiation that's right hand circular polarization is defined from the transmitter point of view and the cross polarization patterns for the left hand curve left hand circular polarization looks like please and next now the process for modeling the antenna is pretty straightforward first we were provided a step model a CAD model containing the structure the 3d structure and the major components of the CubeSat now this was imported into the analysis suite to see the CubeSat structure and

18:29 to get at the various internal components next then we isolated the transmitter component and we captured the model from the circular waveguide opening upward into the antenna now we have the freedom to apply excitations of any polarization we wish and a right-hand circular polarization was selected to match the mode of operation for this device the ortho mode transducer was dropped out since the internal detail of the polarizer and the connection to the power amplifier and so

18:59 forth were not included in the CAD model but it's the end antenna here from its feed point that's a particular interest for this study now we were not provided with an explicit explicit description of the materials but we took an educated guess that the lens on the aperture opening which is used both to focus the e/m energy and to protect the interior of the horn and waveguides was a Rexha light material so our solver has a

19:23 fairly extensive data base for the dielectric and conductive properties of materials so we applied the material properties for Rex alight from that database now the lens also has and you can see the cross section there in that kind of grayish area it has some shape approximation that appeared to be sort of facet like that are probably an artifact of the simplification of the CAD model for model export and this is likely to affect some what the models

19:48 prediction of beam focus and probably side lobe level control next then finally we let the HF SS solver do the work now our solver employs automatic adaptive meshing in order to find the best bow between convergence of results and the minimization of computer and time resources required engineering using engineers who use this tool do not need to become experts in creating good solution measures the solver does that work the solver also provides

20:19 post-processing to display and to export quantities like the electromagnetic fields the fire field radiation pattern characteristics and the feed point characteristics to evaluate the power match to the power amplifier next slide the results for the simulation match the measured beam width of ten point two degrees nearly exactly now on the left you can see from a paper published by the author's or the designers in an AIA a conference paper and on the left hand

20:50 side so the copal of the right hand circular pole is on the left the left hand circular pulled across Paul on the bottom and then the H FSS simulation results are in the graphs in the middle of the slide now the side of the levels and the transitions on the side of the main beam are a little bit different from what the designers reported in the conference paper but it's probably due to the lens model facet ization and I do

21:13 believe a high fidelity NURBS model of the lens will likely pull those results in even more closely finally on the right you can see a time animation of the fields in this device at 30 gigahertz and this shows the march of the fields down the waveguide out the flared horn and through the lens and from this display you can kind of get a sense for how the fields are bending in the lens to present a nearly flat

21:37 wavefront as they exit the volume next slide and next the next step in our modeling process is to integrate our antenna model back into the CubeSat body so that we can assess the impact electromagnetic coupling to the platform and here you can see our iam fields from the model inside of the horn and exiting to a small you can kind of see a box like near field boundary just outside of and surrounding the RF aperture this

22:06 region is solved using our finite element simulation so that's what's going on inside of that simulation volume next please but the region outside of this box is using a different electromagnetic analysis technique called shooting and bouncing rays the full CubeSat body is a fair is really fairly large for a finite element volume meshing simulation an SBR shooting and bouncing race provides the ability to model electrically large scattering interactions and their effects on installed antennas now our

22:38 solvers funded element and SBR are able to operate within a hybrid fashion or a hybrid framework so the finite element results are actually used to excite the SBR simulation this is performed automatically by the solution process and next at the end we obtain a complete installed performance solution next slide please so here's our simulated installed radiation pattern and here we're showing the co polarization and the cross

23:08 polarization in a pattern cut that's in the XZ plane so that would be a slice that goes across the solar panels on the device next slide please it's also possible to visualize the COPE hole and cross poll results in a full 3d radiation pattern that's up superimposed as installed on the CubeSat platform now the complete yum simulation required about 40 minutes at about 24 gigabytes of RAM on a 32 core workstation a problem size for this is about 50 by 35

23:41 by 25 wavelengths or about 40,000 cubic wavelengths at 30 gigahertz this is not a very large problem for the SBR plus solver but it would be quite large for a full finite element volume meshing simulation if we only used a finite elements and that's kind of an illustration why you need a couple of different techniques operating in a hybridized fashion to accomplish something like this in an efficient and a time time conserving manner next slide please

24:11 finally it's useful to compare the isolated antenna radiation pattern of course that's what we would normally measure during the antenna design validation phase and compare it to the installed performance where the coupling to the CubeSat platform is considered now just to note the plots in the blue curve the isolated antenna pattern cut here for one of the pattern cuts this again is along that y-axis across the across the panel's the red curve shows us the installed antenna so that includes the

24:42 CubeSat body interaction effects and you can see here the main beam looks pretty good not much effect here until you're down about 10 DB or so there's a little bit of some structure due to diffraction and reflections on the kind of the right-hand side of that plot on the transition region but you do see the side lobes rise a bit now in some satellite applications and this will vary from antenna type to antenna type and from satellite body interaction

25:11 mechanism to you know it's different for different satellite bodies in some satellite applications these rise this rise in side lobe levels actually could be quite important particularly if we're trying to protect a secure Channel now this installed radiation pattern is what we would then move into the AGI miss submission simulation for the our RF telemetry downlink system evaluation in the context of a complete satellite in its mission profile and with that I will

25:42 turn it back to Jeff all right thank you very much Sean and thanks to me for that great explanation at the the detailed engineering design level so what we're seeing here is a an image that shows the results of bringing that same exact antenna pattern that Sean just talked about and putting that on our CubeSat model at the same location same geometry that we can then use within our link budget calculations so we just simply just browse to that file that that Sean

26:14 provided us and then we modeled the rest of our transmitter parameters such as frequency power and data rate and then we modeled the ground station antenna put some environmental models on there and computer the dynamic RF link parameter so very easy to set up very quick to do and appreciate you guys sharing that that great example of how we're now able to integrate mission analysis into the detailed engineering design process so so that was one of the

26:41 use case we wanted to look at an imaging path the next vignette is something a little bit more notional here and what we're gonna look at is combining SDK and ANSYS to analyze the performance of one of these satellites to track a hypersonic vehicle from space and this is a challenging topic for a variety of reasons so we're going to use these same existing systems and some notional data to explain the overall process and how it happened how it works so this is the

27:10 the first step here is to set up the overall mission geometry and we're still gonna use that same spacecraft simulation that we have used in the previous vignette that we again designed from that got the overall orbit geometry from the satellite constellation design and then what we did was we calculated when that site would fly over a hypersonic target as illustrated in this case you can see the the x-43a hypersonic vehicle in the upper right our CubeSat on the left and we've done

27:42 some basic geometry calculations such as the look angles the the azimuth elevation and range to target you can also see on the 2d graphics in the lower right as the satellite is passing over the hypersonic vehicle so that's kind of first step is setting up the overall geometry the next is to actually create the trajectory of the hypersonic vehicle and digging into that so so in this case we're just flying due north at a constant angle of attack the Mach number

28:08 and altitude vary throughout the flight and if you going motion as illustrated in this screenshot here and we have a data displays showing the time altitude angle of attack and Mach number which we've passed over to ANSYS for higher fidelity CFD analysis so with that i'll again turn things over to answers turn it over to a Valerio thanks so much Jeff and with that in mind that just basically as a difference from what we

28:40 just saw in the satellites we're coming down into the atmosphere and as Jeff has introduced we are looking at an example of a mock-up of geometry of Inasa x-43 and what we are doing from the CFD analyses is Adi tail analysis of the thermal signature that such a vehicle would generate when flying a Mach 9 at the specified altitude given by AGI so to give you an

29:11 example of what it will look like the analysis inside an C CFD fluent we've taken the geometry we specify boundary conditions and if you look over here we are flying as Jeff mentioned in mark 9 about 120,000 feet we're specifying the correct angle of attack and we're also including the effects of the engine or the scramjet the mounted on the x43 so that we can see the effects on the

29:43 infrared signature coming out from the plume of the scramjet we have created if you're familiar with CFD a computational mesh and we have put the vehicle inside what we would call the compositional domain in this case just surrounding the vehicle and a little bit downstream and we have created compositional mesh in this case we are using an sleaze polyhedral mesh elements and we also have inflation layers and just looking a

30:17 little bit more details or the master since it's always a big topic for CFD you can see at the level of detail that there is a very good quality mesh everywhere also in the problematic areas like on the fins and the nose and taking a different view we are looking at the inlet of the scramjet and the exhaust side of the scramjet so we set up an sis fluent to be running in steady state when making some assumption some

30:50 simplification to the problem such as we're using a DL gas for density therefore we don't have any chemical reactions but that would be very easy to include in terms of modeling and the next thing we run we let Fuu and do what it does best that run its simulations and these are some of the results we are looking at a cross plane on the plane of symmetry of the vehicle for the Mach number and we can see the plume coming

31:16 out from the scramjet at the back and we can see the Sharks in the front and these are just a different view in which we are plotting the simulated clearing picture so what I'm plotting over here is the density the gradient of the density so we can see the shock and we are coloring the surface of the vehicle with the temperature and this one is just a quick animation showing the streamlines coming out from the scramjet

31:46 and going into the intake of the scramjet it's just different views of what would be seen by an infrared sensor this one is the temperature distribution on the surface of the vehicle and if you look over here a temperature in Kelvin is extremely hot both because of the plume coming out from the outlet of the engine but also from the friction and the compressibility effects of this

32:18 vehicle flying in Mach 9 and I'll stop over here and I'll just say that we use the CFD to successfully create an infrared signature of the vehicle and we are passing the data to AGI all right so so thank you very much Valerius is another great example of integrating that mission analysis at the into the engineering design process in this case at the the CFD level and so now what we

32:48 want to do is take that that temperature profile and and run it through our Aoi our simulation to see if we're able to detect and track that that vehicle so we set up our scenario created our hypersonic trajectory run a CFD simulation and now we're near the EO our simulation and in order to do this again we're using this notional example with this Corvis BC CubeSat and and so what we'd have to do first and foremost is

33:18 a much better payload so what we have to do is swap out the swap out the sensor model first of all there's a thermal in or excuse me that broad broad-based multi area multi spectral sensor and then we have to put a new high-resolution thermal sensor and we'd probably also have to give it a deployable aperture to help increase the focal length to detect a target like

33:50 this so so yeah pretty pretty notional but when you do this with some of the parameters we started with the existing parameters from the Corvis VC platform and then played around with them inside of voi are made some assumptions for how well we could potentially improve the system and in the future and and this is the the values that we use the field of view the f-number the focal length the GSD and in this case we're gonna be

34:18 looking instead of at the visible we'll be looking at the the medium wave infrared spectrum here so this is the results of after we've modeled that put it all inside the mission simulation and what we see in the upper right now is a synthetic scene of what that uoy our sensor would actually be detecting and we can see that tiny blur at the center of the image representing our hypersonic vehicle there and just changing the view

34:49 a little bit more so instead of looking at the sensor output if you change that to just look at the radiometric input into the sensor we can actually and then zooming in a little bit a little bit more you can actually resolve the shape of that image and you can actually see the same temperature map that was used inside of and that was exported from ANSYS into sdk we can run that through the EOIR simulation to see to see those

35:15 results here as well so certainly a lot of detail here a lot more we could cover there's not enough time to go in this today but you know this is just one piece of the overall problem that sdk can take this this type of modeling integrate this into the engagement chain including scripting behavior with SDKs programming interface in that case SDK allows you to evaluate individual pieces of the chain such as the detection and tracking from

35:43 individual sensors like the one we showed here look at tipping and queuing Network Lake latency command and control logic and engagement performance such as probability of intercept given the location and number of interceptors countermeasures firing times load factors things like that and that can all be tied together inside the SDK analysis environment for a comprehensive kill chain assessment we just don't have time to go in that here today and we're running short on time already if you do

36:10 are interested more in this topic please put in the chat questions but now what we're going to do is we're going to turn switch gears a little bit and now that we've done all these detailed engineering design simulations we want to show how to then bring that back up to the descriptive models in sysml like we did discussed at previous webinars and we're not going to go into detail one the last webinar went a lot of

36:35 detail on this but I at least wanted to level set for the folks that didn't see that in the last webinar we showed a constant satellite constellation integrated that with descriptive models validated requirements for image resolution or visit times and so today's webinar we really want to focus on the ants SP so the antis integration at the subsystem and component level and so in this image you see the the high-level block definition diagram of our satellite constellation and what we've

37:03 done is we've extended this model for today to include each of the satellites and its subsystems so in particular we've added our Aoi our sensor model as well as our ka-band antenna which has both a gain and a gain pattern in temperature that we're that we showed an ancestor in the upper writing kind of see a snapshot of the component for the antenna there's a temperature requirement on that and you can also see the antenna component listed in the

37:33 lower right which was used for the the link performance so this the way we integrated that was again using those parametric diagrams and then once we have the parametric diagrams we can expose those value properties into model center which can then run directly through cameo to validate requirements or other system L tools so so with that I'm going to turn it over to Joshua to talk about how quickly we were able to add the ANSYS updates here to our overall

38:03 workflow so thank you Jeff a great presentation from anta's and from from egi so I just want to kind of close the loop and as Jeff said kind of focus in on the De Anza integration piece we've touched a lot on integrating SDK integrating with Cameo and so as a recap in webinar 2 we looked at the capabilities of Mauston or specifically integrate and explore to bring in SDK and a cost analysis doing some design

38:36 space exploration and looking at some trade-offs between access duration and total cost in the the third webinar just our last webinar we went in great depth on building up our system all diagrams building up those parametric diagrams and exposing those and the the model centered workflow is tied to them all within NBC analyzer running from here kicking off the workflow behind the scenes and validating requirements for that ground sample distance and time to

39:07 revisit so what I thought or what we thought we would work on today is a little bit of that but focusing on the integrating antis piece to really close the loop so we as we said we've already focused on SDK excel magic draw Cameo products so now we're going to focus on the antis piece first off before I do that I just want to kind of level set and talk about file i/o automation to file input output so this is a typical

39:41 pattern of running software either in a manual sense or from a programmatic sense where we would have some basic input file we see here we have deflection dot in has some input values and variables here with file i/o we use we have an input file that is usually consumed by an executable in this case we have an example for deflection Exe so the executable consumes the input file and then produces some output file with

40:12 variables usually this is run in batch mode as we see here this quick little command-line argument where we've specified the executable given some arguments for the inputs and outputs so the title of this slide is for automation prerequisites because anytime we're working with the file IO program we need to identify these key prereqs before we can actually automate and integrate those into model Center so the prereqs being the input file and input

40:44 variables the executable and any arguments to run it in the output file and any output variables so in the same way with file IO everyone knows that antis has a spectacular GUI a lot of people are used to running it from from that standpoint what we require though is the behind the scenes kind of file IO the execution of ANSYS and so the way we

41:15 do that is extending that file IO pattern over to ANSYS is we require an input file so we see a quick screenshot here where ANSYS allows us to record a journal file we can perform some activities some clicks in the GUI and the journal record a script force this will become our input file the executable being ANSYS in this specific case ANSYS workbench so we'll need to call the ANSYS executable and then for

41:46 our output file also in the ANSYS workbench GUI looking at our parameters we have the options to export our data out to a CSV so this CSV will become our output file that Model Center will parse and give us the output results and just as we said previously we need to know how to run this from a batch mode in this case we have a simple in here of how to run an says the run

42:17 workbench executable a couple commands pointing to the project and a couple arguments pointing to the journal file which was that input file so moving forward with that we're going to show how we're going to integrate antis into a model Center and the first thing we were going to bring up is model centers quick wrap plug-in so quick wrap can handle any file i/o program in this case we're going to focus on ANSYS and this is the GUI for quick wrap it's going to

42:48 be quite simple I'll even show a demo of bringing in ANSYS where we'll actually step through this little QuickStart ribbon here at the top to bring in all our prerequisites so the input file and its input variables the output file and its output variables the executable and any arguments and the commands to run and then we can even test run as if we were running from the command line so without further ado I'm gonna hop into a

43:13 very quick example this will move fast paced but this is bringing up or automating the thermal analysis that Ming described earlier so we're just going to focus on one of the use cases that the antis engineer is presented like I said in this case looking at the antenna temperature from the the cube set so to start off with we'll go ahead and bring in the antis plugin like I said this is quick wrap we'll start working down the Quick Start menu we'll

43:45 go into the input file specifying our journal file and then we'll need to expose the input variable so in this case is Z emissivity first off we need to be able to parse this text file which is kind of unevenly distributed so what we'll need to do is grab this single variable here and expose that as a variable so we have the emissivity wrapped we see over in the top left we've add that as a variable and now

44:16 we'll move into the output file so we're going to grab that CSV that ances output we're going to look at it just as a plain ASCII text file and we're going to you the same thing we just did with the input file and so we see here this is an output of all the parameters from the ANSYS workbench project for for that antenna temperature here in this second value we have the antenna temperature or the max antenna temperature I should say

44:45 and then we're also going to expose the the variable to the left of it which is going to be the overall spacecraft max temperature as well so a quick little appending of max to this always good to keep the detail straight and the next will need to specify that command-line argument that we showed earlier so in this case we see that run workbench we see the pointing to the journal file and the project doing a quick test run we

45:18 see we've spit out some outputs down here in the bottom jumping over into Model Center we can change up these emissivity values as needed run and fast forward through the calculation time and see that we actually get updated values for the antenna temperature and spacecraft temperature so once we've done that of course we've already touched on how to integrate SDK how to bring in other models and so forth so we're going to kind of skip to the end

45:48 and kind of show the final picture here bringing back those parametric diagrams so we see here the parametric diagram in this case we have a constraint block here in the middle that's going to point to the integrated workflow that we've constructed bringing in the value properties from the system model model running the calculation and then pushing those values back to the output value properties so in doing that if we run this from nvse analyzer we had some

46:18 requirement on the downloaded imagery and in this case we've actually we've run this beforehand taking a screenshot of it and presented so this is bringing back that imagery that Jeff and Ming we're talking about about pointing back down and making sure that that requirement can be met the second vignette or use case that Jeff was talking about was for the EO ir detection so we can go ahead and tie that integrated workflow into a second

46:51 parametric diagram here doing the same procedures that we've shown before in the previous webinars fast-forwarding to the very end we can see we have in a collection analysis on this one where we have a requirement on collected imagery and the image signal noise ratio in this case we've run our workflow beforehand once again and we get check marks that we've passed those two requirements and just to bring that back to what we were

47:21 referring to this is the capture of that hypersonic vehicle making sure that we can in fact get validation of our requirements for that use case so bringing that all in like I said in closing loops we can integrate workflows bringing in the SDK mission model and the detailed analysis of ANSYS we can import those into the Cameo sysml model and we can run that from M BSC analyzer validating the requirements and

47:52 executing that workflow behind the scenes and verifying that in all we're all working from a Thor tative source of truth doing detailed design and moving from left to right from conceptual down into detail Jeff how are we looking on time I'm gonna just talk to this very quickly as all my other appearances here with the AGI webinars i've touched on an industry example industry example with a quick webinar citation in this case

48:25 Northrop Grumman did a phased array antenna trade study I'm just going to flip through this very quickly of course they integrated a multitude of different tools including ANSYS Hsss but i think one of the key points here that really fits into the AGI DME philosophy is looking at this top piece right here we see that we can bring in all our systems models we can start at a low fidelity working

48:56 in 1d or 0d calculations kind of back of the envelope calculations with scripts or Excel but we can slowly move or move at our pace into higher fidelity which includes bringing in detailed analysis tools like ANSYS so in this case we're moving from left to right and then we're bringing in and as our as our design and our requirements mature we can move from left to right with the conceptual to preliminary to the detail and increasing

49:26 the fidelity of our tools and swapping those into our workflow at any time and just a quick little showcase of the results here of doing some trade-offs and some Pareto front optimizations between multiple objectives all right thank you very much Joshua so yeah I'll go quickly through a summary and trying to get as much time as I can for questions and we're approaching the end of the hour here all right so just wanted to circle back to what we were

49:53 when we really presented it since the start of this webinar series which is what the vision is of digital Mission engineering which is really to connect all of the engineering and systems models to the mission model that is rooted in a multi domain physics environment and what we showed in this webinar series is having all of these pull together as early in the possible having these commercial tools that are pre integrated from day one allows us to

50:19 do high fidelity simulations it allows us to identify critical issues or problems with the design early on that would impact our mission requirements and make changes at an early stage and then we can also use those in explore different mission vignettes at various phases of the lifecycle we discussed the concept of the digital thread which illustrates how each of these models are individual strands that are all tied together so when you pull on individual parts of one model it pulls through the

50:46 rest of the models as well we showed how using model Center is that common framework to do that linking all of the inputs and outputs from each of the tools so that is one changes from one tool it feeds to the next one and that's all integrated with the authoritative source of truth within the system ml descriptive model and we talked about really just having all those tools so that you don't have to integrate scripts as we showed in the the broken lines

51:15 diagram earlier with all of the isolated tools the time it takes to develop those custom tools at each phase a life cycle and then integrating them as you go it makes it really difficult to identify those critical issues early on the process which limits can limit your overall delivered effectiveness and how standardizing on these organization-wide allows you to transition from program to program maintaining that knowledge and not having to rebuild custom tools each time and really just alleviate

51:44 significant challenges of maintaining custom tools and then transferring knowledge as individuals change programs or leave the organization so a quick summary here of the the different tools that we highlighted we used SDK for the the mission model we showed how it integrates with other tools I through the program interface directly or through their model Center we talked about ANSYS we only covered these main three product lines the electronics mechanical and fluids in this webinar but talked about how you can integrate

52:16 SDK with those and then with model Center we showed how to integrate Explorer and then use MBS C to link up with the sysml tools so so again I wanted to thank all of our presenters today I want to thank everyone for attending today's presentation there's a bunch of next steps here there's the all of the these webinar videos will be available by going to a Jericho massage DME and if you want any more information I know we covered a lot

52:47 of ground today in a short amount of time so either visit our websites or send us an email as you see here and also I wanted to mention that we are planning a digital mission engineering roadshow where we're going to allow different presenters to present on these topics like these and then go into more detail on select topics so we're planning to roll that out in the early to mid November timeframe in the DC Denver and Los Angeles areas so stay

53:17 tuned on that more information coming soon so you ready for the questions Jeff yeah yeah okay I got a couple here that came in around the same time I'm gonna read them all the same time cuz they sound pretty related the question is about simulating plasma physics in terms of local and non-local thermodynamic equilibrium flow solving work in plasma engineering like RF ion thrusters and modeling the plasma pattern Larry oh can you take that one

53:50 I'll take that one and I think it Shawn if you want to pitch in as well from the H FSS society but what I can tell you is we actually are presenting next week a poster presentation at the hypersonics technology and systems symposium or conference down in Alexandria Virginia on communication blackout I know that's not plasma but that is a first step that we are doing towards the modeling of

54:21 these ionic fields and the interaction with radio frequencies and on the magnetohydrodynamics on the plasma pure plasma we are working on that we do not have a full-blown ready to use a solution just as yet but we can model we can compute the ion concentration and distribution around the vehicle I think

54:51 I'll just pick up from where valerio left off you know once you have that ionic distribution it is possible to compute a position dependent sort of a material property or conductivity sort of characteristic for the ion sheaf around the vehicle and we will be stealing or Thunder a little bit but we have a release of HF SS coming out your very soon that will have position dependent material property capabilities so that one could have a a position

55:22 dependent complex permittivity complex permeability or a conduct or or a conductivity values and that can be as I said position dependent within a volume so we anticipate that we will have on the electromagnetic side the ability to then compute how electromagnetic energy is affected by that sheath okay next question here is is designing for the flat wavefront out the exit important and that came in I think Shawn while you

55:53 are presenting well I think what's important is whatever you need to do to the design in order to maintain a coherent beam I think that the the the fact that they are fairly consistently flat as the wave fronts exit are sort of part of what's important to keeping a very focused shaped beam of course we want to conserve power the transmitter can use less power if we can concentrate more of the energy and a tighter beam so

56:20 we use as we're limited by physics in terms of how large the aperture can be unless we use a reflector or something else you know a little bit larger but that that would be one of the characteristics that we'd want to see that that lens does it takes the wave fronts and bends them down flat once once they exit that aperture in order to again create that nice shaped beam and keep it all the energy where we want it

56:45 great thank you all right now the question is how do we integrate this answers of SDK to implement it in our simulation it's I think that one might be a little bit too too in-depth to go through and this in this forum but we could follow up with you on on that question because I think this was just an example of doing it not the specifics of how to do it covered it in this webinar similar question are

57:10 similar answer rather no similar question can you go into the c2 chain as you mentioned in this webinar Jeff you were the one that mentioned that and I think there is an example of a case study on that on our website of the air force a nine but for more info we're going to follow up with you on that one because that's not something that we can go into a whole lot of detail in this forum here can your saw can your

57:34 software past the target data around the satellite network to the same fidelity to use by other satellites past the target data I'm guess I'm guessing this was the target data from the EOIR sensor same level of fidelity by other satellites I'm I guess I probably need a little more clarification I think one thing I'll mention is that once you have that data collected from your sensor you

58:05 can model the cross links from multiple satellites and then downlink it to to the ground stations and and you can model that to a high layer level of fidelity but I'm not sure I follow the follow the rest of the question so well we'll follow up that won't sound a little specific so I'd be better to have a more detailed more detailed discussion afterwards I think a question ISM says is for Phoenix how do two companies

58:28 collaborate on this project together I'm happy to answer this I think everyone could probably chime in on this one it takes you know a good amount of meeting and understanding and communication of course I know that's probably a cliche answer but I've gotten to work with Jeff now for a few weeks and then also the ANSYS engineers as well so making sure that you know things are defined upfront

58:59 we all have a common vision and then it's just a matter of getting the work done so to speak great thanks Joshua so another question about we had covered a lot of topics today and there's a question of all what are all the tools that you guys went over here and I know the specifics about what you're trying to do there's probably a follow-up question because there's various aspects of each one of the tools but Jeff or

59:27 anyone else can you go through sort of the laundry list of everything that you showed today the show the workflow front to back on this yeah I guess I can take a stab at dad so the the the overall products that were used were ANSYS products AGI products specifically the STK product line model Center as well as Cameo now I don't know all the details in terms of specific packages within each of those product lines

59:58 I talked briefly on the ANSYS product we're used on the AGI side we use variety of modules with an SDK we used the Aviator to do the hypersonic vehicle we used a oír for the EOIR simulation we used SAP Pro for the constellation design and analyzer for the trade studies and integration with model center we use solace for the the spacecraft simulation and attitude control system I'm probably forgetting a couple others but some of those are at

60:29 least some of the products that we used on the the AGI side you don't need to necessarily use that specific configuration to do this type of workflow within digital mission engineering there's lots of different than yets or use cases that can be applied whether it's an air mission a missile defense space design so yeah so that's that is probably more of a follow up on and see what specific problem areas and application errors are you looking to address and then we can

60:59 provide a more specific recommendation of specific modules for you for your use case I don't know if Phoenix or if ANSYS to add in on that I can do mine very quickly the the three products match the same three capabilities that Phoenix has that small Center integrate for integrating and automating models and explore for exploring or optimizing your design space and the third is model

61:29 Center M BSC for the bridging between the descriptive and the domain engineering analyses I'm an sis electromagnetics it's the HF SS high frequency structure simulator and the HF SS s BR + simulation option for it over on the fluid size its fluent for the hypersonics modeling and on the thermal modeling it's the ANSYS manacle suite great alright well that's all the time we have today for these questions if we didn't get to your question we will

61:59 follow up with you individually with the appropriate company answering your questions or companies thanks again to all of our presenters if you are looking for the slides and this recording later it is going to be available at AGI comm / DME the DME stands for digital Engineering so if you go to that website it's an easy way to stay in the loop about other upcoming digital mission engineering events that we will be having in the future and again thanks

62:24 again to all of our all of our presenters thanks Jeff Thank You Joshua thanks guys I you