Commercial Space Operations Noise and Sonic Boom Modeling and Analysis (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4-1 CHAPTER 4. Model Validation Plan 4.1. Introduction As part of the development of the rocket noise and sonic boom models, the ACRP recognized the need to conduct comprehensive and systematic model validation studies. The purpose of these model validation studies is to check the accuracy of the model’s representation of the real system and to identify parts of the models that may need further improvements. Data collection should be conducted for commercial spacecraft that are currently in service or in the development/testing stage. The noise and sonic boom generated by these spacecraft should be measured for a range of operations including launch, reentry/landing, and static operations. Validation of the rocket noise and sonic boom models requires checking the accuracy of the models’ acoustic predictions compared to field measured acoustic data for commercial spacecraft operations. This model validation plan is a general purpose plan that can be used to conduct rocket noise and sonic boom measurements at any space launch facility. Field acoustic measurements and model validation tests are not part of the current project. Any future field measurement programs will need to incorporate site specific details into the planning for such tests. Field tests of this nature involve significant advance planning and coordination with facility operators, vehicle operators, airfield and airspace managers, and safety and environmental representatives. Site specific planning at this level should occur once a new field measurement effort has started and facility options for conducting the measurements have been reviewed. The following sections of this plan include standard validation procedures used for computer models, a summary of modeling requirements for commercial spacecraft operations, and the data collection and validation test requirements needed to validate the rocket noise and sonic boom models. 4.1.1. Validation of Computer Models Validation of a computer model is conducted during software development to ensure that an accurate and credible model is produced. Computer models, including simulation models, are approximate representations of real world systems. As such, models should be validated to the degree needed for the intended application. Many approaches can be used to validate a computer model. Naylor and Finger [Naylor et al., 1967] defined a three-step approach to model validation that has been widely followed: 1. Develop Model with High Face Validity - A model that has face validity appears to reasonably predict the real world system to people, who are familiar with the real world system. Face validity is tested by having users knowledgeable about the system examine model output for reasonableness and in the process identify deficiencies. The rocket noise and sonic boom models have high face validity since the outputs from both models have been compared against measurement data and appear to reasonably predict accurate noise levels. However, the rocket noise and sonic boom models have not been thoroughly validated for spacecraft flight operations, which is the subject of this plan. 2. Validate Model Assumptions - To validate model assumptions, one must check the structural assumptions (i.e., algorithms) used and ensure that sufficient data are used from reliable sources. If a model does not have all of the necessary algorithms or sufficient, high quality data, a model validation test is likely to fail. The rocket noise and sonic boom models are based on mature research. However, part of the model validation plan is to check the algorithms and data assumptions being used to assess noise and sonic boom specifically from new commercial space flight operations, which have limited data. 3. Validation Tests - Validation tests are used to compare the input-output transformations of each model to that of the real system. In this case, acoustic predictions from the rocket noise and sonic boom models will be compared with actual field measured acoustic data for the same set of input

4-2 conditions from a commercial space vehicle flight operation including vehicle type, flight trajectory, and atmospheric conditions. Launch platform configuration would also be included as an input consideration when this modeling capability has been fully developed in the rocket noise model, RUMBLE. 4.2. Spaceport Facilities and Types of Commercial Space Flight Operations Before discussing methods to validate the rocket noise and sonic boom models, it is useful to review the spaceport facilities and operators of commercial space flight operations. These items are described in detail elsewhere in this report; therefore, only a brief summary is presented here. 4.2.1. Spaceport Facilities and Operators The spaceports that currently support or plan to support commercial space flight operations are described in Section 1. During the planning of a model validation exercise, coordination is required with the mission operator and the spaceport facility. Those conducting the model validation will need to request information from the operator and the facility as well as obtain permission to conduct field measurements during a flight operation. Examples of operations include SpaceX Falcon 9 launches from Cape Canaveral Air Force Station or Vandenberg Air Force Base and Virgin Galactic air-launched suborbital flights from the Mojave Air and Space Port. 4.2.2. Launch and Reentry Operations The types of launch and reentry operations that currently exist are described in Section 1 and summarized below:  Conventional expendable and reusable vertical launch vehicles;  Air-launched spacecraft, like Virgin Galactic’s SpaceShipTwo;  Horizontal-takeoff, horizontal-landing spacecraft;  Ballistic entry vehicles, similar to the Apollo or Dragon modules or the Falcon 9 First Stage; and  Lifting body reentry vehicles like the Dream Chaser. Each combination of vehicle type and flight operation may require specific modeling data requirements. In addition, modeling these different operational modes will exercise different parts of the rocket noise and sonic boom models. For example, a ballistic entry vehicle would specifically exercise the hypersonic blunt body reentry algorithm in the sonic boom model, but it would not involve other algorithms of the rocket noise and sonic boom models. To validate this particular algorithm, sonic boom measurements would need to be made of a ballistic reentry operation. To ensure that both models generate accurate predictions for the different types of flight operations, it is recommended to validate both models for a variety of vehicle types and operations and not limit the measurements to vertical launch operations. 4.3. Validation of RUMBLE RUMBLE should enable accurate prediction of the acoustic environments from existing and new launch vehicles. Validating RUMBLE to ensure accurate predictions involves several steps. The first step is to identify RUMBLE’s structural and data assumptions to be validated; these elements include the rocket noise source properties such as source power spectrum and directivity, the propagation of the noise source, and the received noise. As mentioned above, RUMBLE is based on mature research, such that it’s structural and data assumptions have been validated to a reasonable degree. But, since RUMBLE has had limited validation for actual space operations, it is important to continue to examine the model’s assumptions for any new measurement/modeling comparisons made. Next, a review of existing rocket acoustic and supporting data should be conducted for possible inclusion in the validation study. Existing data gaps should be filled by planning and conducting validation tests specifically designed to fill the identified gaps. The data collection should target rocket launch, landing, and static firing noise data for a

4-3 variety of commercial space flight vehicles. Finally, to quantify and assess RUMBLE’s accuracy, validation tests should be performed to compare the input-output transformations of RUMBLE to the noise measurement results from the actual tests. 4.3.1. Review of Existing Acoustic Measurement and Supporting Data As the validation efforts objective is to validate RUMBLE for a range of commercial space vehicles and operations, it is important to determine if existing acoustic and supporting data are available from prior tests. Existing data sets with all the information and fidelity required to validate RUMBLE may be limited and difficult to obtain. However, acquiring an existing, complete data set would represent significant cost savings compared with acquiring similar data by conducting a new field measurement program. 4.3.2. Conduct Data Collection and Validation Tests Validation tests will be conducted for various commercial space operations to verify RUMBLE’s performance and to identify any modeling gaps. The general requirements to conduct a validation test include test planning, coordination with vehicle and site operators, acoustic and supporting data collection, and documentation of data collection. A summary of the test flight, atmospheric, and noise measurement data requirements are provided below. Flight Data Requirements Flight data that describes the vehicle trajectory and operating state are required inputs for RUMBLE. These data include the vehicle’s coordinates, velocity vector, flight path angle, heading, attitude, and thrust for every second during the entire flight operation. Actual flight data are required for model validation; therefore prior arrangements should be made with the operator to obtain these data. Atmospheric Data Requirements An atmospheric profile, similar in format to the U.S. Standard Atmosphere 1976 [NASA, 1976], is required. For accuracy, an atmospheric profile should be obtained from GPSsonde data from a weather balloon launch or similar sources that are local to the operating facility and conducted near the time of the flight operation. Parameters measured include altitude, pressure, temperature, relative humidity, wind speed (and direction) and geographic location. The standard atmosphere extension can be used to extend GPSsonde profiles as necessary. Noise Measurement Requirements Noise measurements will be conducted during the commercial space flight operation to characterize the environmental noise exposure due to the flight (or static) operation and to obtain noise data for the validation of RUMBLE. Microphone arrays, instrumentation requirements, and setup and test procedures should be designed for the purpose of model validation. Because no specific arrangements have been made with operators or facilities at the time of this writing, array placement is described in general. Figures 4-1 and 4-2 show example noise contours for two different launch vehicle operations. Figure 4-1 shows A-weighted overall sound pressure level (OASPL) contours predicted for a vertical launch vehicle. Using Figure 4-1 to illustrate array placement, a linear microphone array (red circles) should extend from the launch pad vicinity, radially outward, a sufficient distance to be able to validate OASPL (or other metrics) at the lowest levels of interest. Due to the long propagation distances for low frequency rocket noise, portable noise monitors may need to be placed up to eight miles from the launch pad. Depending on resources, one or more linear arrays, semi-circular arrays, or individual microphones may be placed at locations within the noise exposure area to collect noise data for model validation. Eight to ten microphones are a reasonable number to adequately cover the noise exposure area. Microphones should be located to test RUMBLE’s overall performance (i.e. how well it predicts the far field noise environment in the areas around the launch pad and extending into the surrounding communities); measurements can be taken out to distances where the vehicle noise is predicted to be above the ambient.

Measurem such as th A in Figure operation of microp range of should al collection track and far out the predicted ents should e rocket engi FIGUR n example of 4-2. In this e to the south. hones (red ci far field noi so be includ may be perf perpendicula noise conto to be above t also be taken ne noise direc E 4-1 Exam the noise exp xample, the S Using the co rcles) and in se levels for ed in plannin ormed using r to the flight urs extend. M he ambient. to validate th tivity algorit ple vertical osure from a ound Exposu ntours in Fig dividual mic model valid g. If this ty arrays of mi track, exten easurements e performan hm. launch vehic conceptual h re Level (SE ure 4-2 to ill rophones sho ation. Measu pe of launch crophones po ding up to fiv can be taken ce of specific le A-weighte orizontal lau L) contours ustrate array uld be place rements of operation o sitioned und e miles from out to dista elements of d OASPL c nch vehicle are shown fo placement, o d around the rocket engin ccurs over l er and/or off the runway nces where t the RUMBLE ontours. operation is s r a runway la ne or more a airfield to co e noise direc and, acoustic set from the depending on he vehicle no 4-4 tool, hown unch rrays ver a tivity data flight how ise is

In micropho model val T rocket sho frequency the Nation associated operated r mile from requires m sensitivity recorders least 48,0 portable r M to the rock the above ne positions idation effort  A-wei  Maxim  A-wei  Unwei he data recor uld acquire d range of les al Instrumen with close emotely. Low the rocket, icrophones t microphone or noise mon 00 samples p ecorder that m icrophones s et as follow FI discussion, should yield should be de ghted and Un um A-weigh ghted Sound ghted 1/3-oct ding system( ata at a rate s than 10 Hz ts PXI chass in microphon sensitivity where levels hat have a fr s include th itors used at er second an eets these re hould be mo s. Pressure m GURE 4-2 OASPL and calibrated ac signed to tes weighted Ov ted and Unw Exposure Lev ave Band Sp s) for primar of at least 10 to above 40 is with NI-44 es will need microphones may exceed equency rang e Bruel & K the more dis d have a dyn quirements i unted on trip icrophones sh Example hor SEL are u oustic pressu t the primary erall Sound P eighted Soun els ectra y microphon 2.4 kHz, usin kHz. Exampl 62 or NI-447 to be config should be us 150 decibels e of less tha jaer 4938 an tant array loc amic range s the Larson D ods five feet ould be orie izontal laun sed as exam re time histo noise metric ressure Leve d Pressure L es located wi g 24-bit ana es of system 2 data acqui ured to auto ed for all arr (dB). The f n 10 Hz to a d the GRAS ations should in excess of avis LD 83 above the gr nted at grazin ch vehicle SE ples. Record ries. The arr s of concern ls versus tim evels thin about a log to digital s with this pe sition cards. R matically sta ay positions w requency ran bove 40 kHz 40 BF (or sample the 120 decibels 1 sound level ound and ori g incidence L contours ings made ay placemen including: e half mile fro converters, o rformance in ecording sy rt recording ithin about ge of rocket . Examples o 40 BH). Po data at a rate . An example meter. ented with re with respect . 4-5 at all t and m the ver a clude stems or be a half noise f low rtable of at of a spect to the

4-6 rocket. Free field microphones should be oriented (pointed) at the rocket. Because the rocket will be moving during a flight operation, microphone orientation, with respect to the source, will not be constant and should be accounted for. Microphones should be located away from reflecting surfaces other than the ground. All recording channels should be calibrated before and after the flight test. System response checks should be conducted before and after the test and any corrections applied in post processing. 4.3.3. RUMBLE Validation Tests The validation tests consist of comparing the output of the real system (i.e. measured noise levels from a test) to the predicted noise levels generated by RUMBLE for the same set of input conditions. A proper test of the model requires that all critical inputs are accurate. Therefore, the actual vehicle configuration, flight data, and atmospheric data collected during the test will be used to generate the model inputs. During the course of the model validation study several tests should be conducted, but it is expected that each test will target a different type of flight operation or vehicle. Each flight test would likely involve only one vehicle operation, either a launch, landing, or static test, although it is now possible to have a launch and subsequent reentry and landing of the same vehicle. RUMBLE’s primary output should be used as the measure of performance. Direct comparison of the measured and modeled noise metrics, at each measurement location, may be used to identify discrepancies in the model related to a particular noise metric or to a more fundamental element of the model, such as the propagation algorithm. The comparison of modeled and measured values for all of the measurement locations should indicate an overall measure of RUMBLE’s performance. Comparisons made specifically to validate a particular element of the model, such as the source directivity algorithm, may reveal that improvements need to be made to that specific element. 4.4. Validation of PCBoom PCBoom should enable accurate prediction of the acoustic environments from existing and new launch vehicles. Validating PCBoom to ensure accurate predictions involves several steps. The first step is to identify PCBoom’s structural and data assumptions to be validated; these elements include the space vehicle’s sonic boom source or F-function, the propagation of the sonic boom source, and the received noise. PCBoom is based on mature research, therefore it’s structural and data assumptions have been validated to a reasonable degree. But, like the rocket noise model, PCBoom has had limited validation tests for actual space operations, therefore it is important to continue to examine the model’s assumptions for any new measurement/modeling comparisons made. Next, a review of existing sonic boom measurements and supporting data should be conducted for possible inclusion in the validation study. Existing data gaps should be filled by planning and conducting validation tests specifically designed to fill the identified gaps. The data collection should target rocket launch and reentry/landing operations for a variety of commercial space flight vehicles. Finally, to quantify and assess the accuracy of PCBoom, validation tests should be performed to compare the input-output transformations of PCBoom to the boom measurement results from the actual tests. Before discussing validation guidelines for the sonic boom model, it is necessary to mention that for most of the flight operations described above, the rocket propulsion acoustic exposure at the ground and the sonic boom exposure at the ground do not occur at the same location. For a vertical launch, the acoustic exposure at the ground is typically greatest during the initial portion of the trajectory when the vehicle is sub-sonic. For the same vertical launch operation, sonic boom is not generated until the vehicle is supersonic and typically does not reach the ground until the vehicle pitches over. For this operation, sonic boom exposure at the ground would occur many miles away from the launch pad and over water for coastal launches. SpaceX’s Falcon 9 first stage landing operation is the only known case where the rocket noise exposure and sonic boom exposure occur at about the same location on the ground, in the vicinity of the landing pad or offshore barge.

4-7 Because rocket noise and sonic boom typically do not occur at the same ground location, this should be factored into the data collection effort. Although it is possible to measure the noise exposure and sonic boom exposure from the same flight operation, in two completely different geographic areas, it would require two separate measurement teams. 4.4.1. Review of Existing Sonic Boom Measurement and Supporting Data Since the objective is to validate PCBoom for a variety of commercial space vehicles and operations, it is important to check if existing sonic boom measurement and supporting data are available from prior flight operations. Existing data sets that have all of the information required to validate the sonic boom model may be limited and difficult to obtain. However, acquiring an existing, complete data set would represent significant cost savings compared with acquiring similar data by conducting a new field measurement program. 4.4.2. Conduct Data Collection and Validation Tests Validation tests will be conducted for various commercial space operations to verify PCBoom’s performance and to identify any modeling gaps. The general requirements to conduct a validation test include test planning, coordination with vehicle and site operators, sonic boom and supporting data collection, and documentation of data collection. A summary of the test flight, atmospheric, and sonic boom measurement data requirements are provided below. Flight Data Requirements Flight data describes the vehicle’s trajectory and operating state. From this data, the vehicle’s coordinates, Mach number, flight path angle, and heading can be determined. In further post processing, the first and second derivatives of Mach number, flight path angle, and heading are computed as part of the trajectory input to PCBoom. The vehicle’s F-function and the engine exhaust plume F-function (if needed) are also model inputs that must be developed for the test vehicle. Actual flight data are required for model validation therefore prior arrangements should be made with the operator to obtain these data. Atmospheric Data Requirements An atmospheric profile, similar in format to the U.S. Standard Atmosphere 1976, is required. For accuracy, an atmospheric profile should be obtained from GPSsonde data from a weather balloon launch or similar sources that are local to the operating facility and conducted near the time of the flight operation. Parameters measured include altitude, pressure, temperature, relative humidity, wind speed (and direction) and geographic location. The standard atmosphere extension can be used to extend GPSsonde profiles as necessary. Sonic Boom Measurement Requirements Sonic boom measurements will be conducted during the commercial space flight operation to characterize the sonic boom exposure due to the flight operation and to obtain data for the validation of PCBoom. Microphone arrays, instrumentation requirements, and setup and test procedures should be designed for the purpose of model validation. Because no specific arrangements have been made with operators or facilities at the time of this writing, array placement is described in general. Figures 4-3 and 4-4 show example sonic boom contours for two different launch vehicle operations. Figure 4-3 shows overpressure contours in units of pounds per square foot (psf) predicted for a launch vehicle (trajectory shown as a black line). Using Figure 4-3 to illustrate array placement, microphones (red circles) should be located within the expected sonic boom footprint at locations necessary to validate a range of overpressure levels down to 0.1 psf. Due to the large size of a typical boom footprint, measurements will need to be made at multiple locations, not necessarily in the same vicinity. The coastal launch depicted in Figure 4-3 would require that boom measurements be recorded at locations in the Atlantic Ocean, although in many other cases, sonic boom from commercial space operations can be recorded on land. Microphones should be located to test the overall performance of PCBoom, i.e., how well it predicts the

sonic boo vehicle so FI A operation 0.3 psf (g (trajectory boom foo or individ validation focal regi also be in in the ce micropho regions w Hobbs, C validation micropho 0.1 psf c northern l large size PCBoom need to be m footprint a urce (F-funct GURE 4-3 L n example o is shown in F reen), 0.5 ps shown as a tprint. Using ual micropho (0.1 psf up ons of the fo cluded; in Fig nter of the nes in the co ill occur. PC .M., and Plot for comme ne array inclu ontour perim ocation.) At t sonic boom for a range conducted in nd signature ion) or predi aunch vehic f sonic boom igure 4-4. In f (red) and 1 black line). T this contour nes should b to the highe otprint, i.e., a ure 4-4, the figure. Foca rrect location Boom’s foc kin, K.J., 20 rcial space des four pos eter, and on he widest po footprints th of overpressu several diff s, and the pe ction of focal le (ascent to contours for this case, pe .0 psf (magen he peak ove shape as an e e placed with st peak level reas where i focal region l regions ar to measure us prediction 10] and spac flight opera itions under e microphon int, the 0.1 ps at are genera re levels (e. erent areas of rformance o zones. orbital traj a horizontal ak overpressu ta) are show rpressure con xample, one in the footpr s expected.) sopemps cro is indicated b e only abou a focus requi capability h ecraft (e.g. T tions is rec the flight tra e position o f contour is a ted by space g. 0.1 psf to the footprint f specific ele ectory) sonic ly launched a re contours o n for a desc tours outline or more arra int to cover a Measuremen ss and boom y the small r t 300 feet w res estimatin as been val itan IV) flig ommended. ck, three oth utside of the pproximatel craft flight o 1.0 psf), son . ments of PC boom footp nd horizonta f 0.1 psf (blu ent back tow the general ys of microp range of pea ts of peak o energy is co ed and mage ide, such t g, prior to th idated for ai ht operation In Figure er micropho 0.1 psf con y 100 miles w perations. In ic boom me Boom such a rints. lly landed ve e), 0.2 psf (c ards the spac shape of the hones (red ci k levels for m verpressure i ncentrated, s nta contour r hat placing e test, where rcraft [Page, s, although fu 4-4, the exa ne positions tour (at the ide, typical order to va asurements w 4-8 s the hicle yan), eport sonic rcles) odel n the hould egion boom these J.A., rther mple at the most of the lidate ould

FIG In all microp validation weighted T 51,200 sa to 20 kHz acquisitio operations L of sonic b micropho 20 kHz. respect to ground b response processing 4.4.3. P T boom lev conditions vehicle co the model is expecte likely inv URE 4-4 H the above d hone positio effort shoul SEL, C-weig he data recor mples per sec . Example sy n cards. Rec . ow frequency oom is conc ne is the Brue Microphone the incomin oard. All rec checks shoul . CBoom Valid he validation els from a fl . A proper t nfiguration, inputs. Durin d that each t olve only one orizontally iscussion, pe ns will yield d be designed hted SEL and ding system( ond, using 2 stems includ ording syste microphone entrated in th l & Kjaer 41 s should be g wave front ording chan d be conduct ation Tests tests consis ight test, to t est of the mo flight data, a g the course est will targe vehicle oper launched an ak overpressu calibrated p to test PCB perceived lo s) for primary 4-bit analog e the Nationa ms may nee s should be u e 0.1 to 100 93, a pressur mounted on and should nels should ed before and ts of compar he predicted del requires nd atmospher of the model t a different t ation, either d horizontal re contours a ressure time oom’s prima udness (PLd microphone to digital con l Instrument d to be oper sed for all so Hz frequen e-field micro ground boar be located aw be calibrated after the fli ing the outp levels gener that all critic ic data colle validation st ype of flight a launch or re ly landed ve re used as an histories. T ry output me B). s should acq verters, over s PXI chassis ated automa nic boom rec cy range. On phone with a ds and orien ay from ref before and ght test and ut of the rea ated by PCB al inputs are cted during t udy several t operation or entry/landing hicle sonic b example. R he array plac trics includi uire the data a frequency with NI-446 tically or rem ordings sinc e example o frequency r ted at grazi lecting surfa after the f any correctio l system, i.e oom for the accurate. Th he test will b ests should b vehicle. Eac . oom footprin ecordings ma ement and m ng peak leve at a rate of at range from 0 2 or NI-4472 otely for c e the energy f a low frequ ange of 0.07 ng incidence ces other tha light test. Sy ns applied in ., measured same set of erefore, the a e used to gen e conducted, h flight test w 4-9 ts de at odel ls, A- least .1 Hz data ertain range ency Hz to with n the stem post sonic input ctual erate but it ould

4-10 The primary output of PCBoom should be used as the measure of performance. Direct comparison of the measured and modeled sonic boom metrics, at each measurement location, may be used to identify discrepancies in the model related to a particular boom metric or to a more fundamental element of the model, such as the propagation algorithm. The comparison of modeled and measured values for all of the measurement locations should indicate an overall measure of performance of PCBoom. Comparisons made specifically to validate a particular element of the model, such as the vehicle source definition, may reveal that improvements need to be made to that specific element. 4.5. Recommendations for Model Improvement Validation studies conducted for RUMBLE and PCBoom should quantify and assess the accuracy of these models compared to measurements of actual commercial space flight operations. In the process, deficiencies in these models may be identified that warrant further investigation and improvement. Any deficiencies found should be identified along with recommendations for model improvement. Model validation studies and any potential model improvements are recommended future research efforts (Section 6) and are not part of the current project.