Methods for optimizing the technical design and operation of an aircraft / vehicle with regard to target parameters of noise development

Abstract

comprising methods for the design, optimization of the technical design and / or operation of a machine, aircraft or vehicle - Input (S1) of parameter values ​​of the machine, aircraft or vehicle by at least one test subject on an end device, wherein the parameter values ​​are design parameter values, technical design parameter values ​​and / or operational parameter values ​​of the machine, aircraft or vehicle, - Playback of audio signals (S2) on the terminal device, which reproduce the sound of the machine, aircraft or vehicle based on the entered parameter values, - Evaluating the auditory impression (S3) of the audio signals by the test subject and obtaining a subjective evaluation result from the test subject, - Input (S4) of the subjective evaluation result by the test subject on the terminal device, - Determining (S5) a relationship between the parameter values, or sets of parameter values, and the subjective evaluation result, - Design (S6), optimization of the technical design and / or operation of the machine, aircraft or vehicle, taking into account the relationship of the parameter values, or sets of parameter values, with the subjective evaluation result, wherein in the design, optimization of the technical design and / or operation parameter values ​​are used which, in the operation of the machine, aircraft or vehicle, result in noises of the machine, aircraft or vehicle that correlate with a subjective evaluation result corresponding to a predetermined evaluation result, preferably a positive evaluation result.

Description

[0001] The present invention relates to a method for the design, optimization of the technical design and / or operation of an aircraft or vehicle. The method is particularly useful for the design, optimization of the technical design and / or operation of an aircraft / vehicle with regard to target parameters of noise generation and perception.

[0002] To predict and evaluate engine noise, analytical and numerical methods as well as experimental procedures are typically used. Rapid analytical methods are employed in the preliminary design phase to explore the range of relevant design parameters and define the optimal area considering specific acoustic parameters (e.g., sound pressure level). Based on these results, virtual flyovers can be performed. Acoustic flyover simulations virtually simulate individual flight segments. In this simulation, sound is propagated through the atmosphere from the aircraft's sound source to defined positions on the ground. The resulting sound levels on the ground can then be converted into time series and stored as audio files. These results are used, among other things, to provide input for psychoacoustic studies, such as listening tests.Known tools for the described process chain, consisting of (1) calculating the sound source, (2) performing flyover simulations, and (3) making the noise audible, include PropNoise, described in doi:10.14279 / depositonce-5935 and doi:10.14279 / depositonce-7057, by Violin and Coral. Further literature on this topic can be found in A. Moreau, A. Prescher, S. Schade, M. Dang, R. Jaron, and S. Guérin, "A framework to simulate and to auralize the sound emitted by aircraft engines," InterNoise conference 2023 conference proceedings (2023).

[0003] Until now, the assessment of psychoacoustic properties and noise quality of conventional aircraft as well as modern flight systems, such as drones, has been carried out primarily using physical or psychoacoustic parameters (for an overview, see Lotinga, Ramos-Ramero, Green & Torija, 2023, https: / / doi.org / 10.1007 / s40726-023-00285-4). Furthermore, some studies compared original recordings of aircraft noise and / or synthetically altered noise recordings with listeners' evaluations (e.g., regarding annoyance or unpleasantness) from listening experiments (e.g., Janssens, Vecchio & Van der Auweraer, 2008, https: / / doi.org / 10.1016 / j.ast.2007.10.002; Gwak, Han & Lee, 2020, https: / / doi.org / 10.1016 / j.jsv.2020.115651; Christian & Cabell, 2017, D01:10.2514 / 6.2017-4051). Additionally, the sound quality was examined, for example, through associations with specific attributions.However, such investigations have so far been carried out with a limited number of test participants and, as described above, using sounds from existing aircraft or synthetic modifications thereof.

[0004] Therefore, an evaluation can currently only take place once at least one prototype exists and a field or laboratory study can be conducted. These evaluations are carried out as costly and time-consuming individual studies, for example, in a laboratory environment with specially recruited participants. Only small sample sizes can be tested in this way, making a systematic testing of virtually all changes to the design parameters impossible. Furthermore, the production of aircraft prototypes is extremely expensive, especially with a high number of possible iterations. For these reasons, the number of possible design variations is often limited, leading to suboptimal solutions.

[0005] Another technical problem is that aircraft have problematic psychoacoustic properties that cannot be captured by purely physical parameters, such as maximum sound pressure level or extended metrics (e.g. EPNL).

[0006] The purpose of the invention was to provide a solution to these problems.

[0007] A method according to claim 1 is specified. Specific embodiments are specified in dependent claims and this description.

[0008] A method for the design, optimization of the technical design and / or operation of a machine, aircraft or vehicle is specified. - Input of parameter values ​​of the machine, aircraft or vehicle by at least one test subject, preferably a plurality of test subjects, on an end device, preferably on an end device assigned to each test subject in the case of a plurality of test subjects, wherein the parameter values ​​are design parameter values, technical design parameter values ​​and / or operational parameter values ​​of the machine, aircraft or vehicle, - Playback of audio signals on the terminal device, preferably on a terminal device assigned to each test subject when dealing with a large number of test subjects, which reproduce the sound of the machine, aircraft or vehicle based on the entered parameter values, - Evaluating the auditory impression of the audio signals by the test subject, preferably a large number of test subjects, and obtaining a subjective evaluation result from the test subject(s), - Input of the subjective, preferably qualitative, evaluation result by the test subject or each of the test subjects, using the terminal device; in the case of a large number of test subjects, preferably on the terminal device assigned to each test subject. - Determining a relationship between the parameter values, or sets of parameter values, and the subjective evaluation result(s), - Designing, optimizing the technical design and / or operating the machine, aircraft or vehicle, taking into account the relationship of the parameter values, or sets of parameter values, with the subjective evaluation result or results (with a large number of test subjects), wherein parameter values ​​are used in the design, optimization of the technical design and / or operation which, in the operation of the machine, aircraft or vehicle, lead to noises of the aircraft or vehicle that correlate with a subjective evaluation result corresponding to a predetermined evaluation result, preferably a positive evaluation result.

[0009] The invention, in its most general form or in one of its specific embodiments described below, achieves one or more of the following advantages: • The method allows for multiple regression analysis to establish the relationship between noise ratings and derived psychoacoustic parameters with the design parameters of the drive. • The procedure allows the identification of the most important predictors of noise assessment and - with a large number of test subjects - the collection of the variance that can be explained by the tested predictors. • The acoustic development of new machines, aircraft, or vehicles can be improved from a psychoacoustic perspective. In particular, psychoacoustic properties can be incorporated into the acoustic design of new, i.e., not yet realized, configurations of aircraft or vehicles. Furthermore, evaluations from large-scale listening tests can be integrated into the development of new aircraft or vehicle designs. • the coupling of the auralization (making audible) of machine, aircraft or vehicle noises based on a specific configuration, the evaluation by psychoacoustic indicators and assessments from broad-based listening tests and the derivation of active noise control measures on the machine, aircraft or vehicle, in particular with regard to the reduction of annoyance and disturbance perception, reduction of sleep disturbances during nighttime operation and thus the preservation of the health and well-being of those affected by noise. • the coupling of the auralization of different movement routes with the evaluation by psychoacoustic indicators and assessments from listening tests on a broad basis with the aim of deriving noise-optimized movement routes, for example flight routes, as well as reducing noise by adjusting parameters during movement (especially in urban areas). • The invention is particularly interesting for aircraft manufacturers. During the aircraft development phase, it allows for the continuous, rapid, and cost-effective recording of the effects of design decisions. • Targeted control of a drive during a movement, for example a flyover, to achieve a desired, pleasant and not very disturbing auditory impression (especially during takeoff and landing of an aircraft and consequently at low altitudes). • Diverse listening tests can involve test subjects in their home environment and are not limited to the laboratory. Implementation can be carried out using a mobile device. Contextual factors of this home environment can be captured by the device's sensors and statistically considered as additional influencing factors when evaluating the sounds being assessed. • Noise-optimized design of drives, not only with regard to minimizing sound pressure levels, but also with regard to a pleasant perception of overhead and passing noises or drive-by noises. • Development of new mathematical models for the acoustic certification of new aircraft and vehicle concepts. • Forming a ranking of drive configurations with regard to the auditory sensation they elicit as pleasant / unpleasant or annoying To avoid unpleasant and annoying noises in new machine, aircraft, or vehicle concepts, the invention provides a new and rapid method for making the sound of future designs audible and psychoacoustically evaluating it. In particular, design rules for low-noise construction of novel drive systems can be researched and developed. • A large number of people can be actively involved in the development process of machines, aircraft, or vehicles at an early stage. This allows for the establishment of a database from which it can be analyzed which design parameters, or technical design parameters in the case of aircraft, lead to pleasant, subjective sound perceptions and, for example, in the case of an aircraft, reduce the long-term disturbance to residents affected by overflights. This enables a link between design parameters and psychoacoustic parameters. At the same time, the invention can also increase identification with and acceptance of the innovative product in line with the acceptance-by-design approach. The invention allows users to adjust configurations based on design parameters and to directly evaluate the auditory experience.In this evaluation of, for example, aircraft engines, with regard to their acoustic impression, currently unknown concepts, for example for drive systems (= unknown noises), can be evaluated. • The invention enables a link between (a) design parameters, (b) noise, (c) sound perception and evaluation by test subjects, and (d) psychoacoustic parameters such as roughness, sharpness, fluctuation strength, tonality, etc. This link also allows feedback from the psychoacoustic parameters back to the design parameters.

[0010] The invention is applicable to machines, aircraft, and vehicles. The vehicle can be a rail or road vehicle, in particular a motor vehicle or rail vehicle.

[0011] For large groups of test subjects, quantitative, analysis of variance (ANOVA) statistical evaluation methods are preferably used to determine the relationship between parameter values, or sets of parameter values, and subjective evaluation results. The subjective evaluation results exhibit a variance that can be determined. For example, a maximum value for a standard deviation can be defined for a large group of test subjects to uniquely assign a specific subjective evaluation result.

[0012] Parameter values ​​in an embodiment can be selected from design parameter values, engineering design parameter values ​​and / or operational parameter values.

[0013] The noise of the aircraft or vehicle can be a noise as it appears inside a person's aircraft or vehicle, or as it appears outside a person's aircraft or vehicle.

[0014] The end device can be a mobile device, such as a smartphone.

[0015] The terminal device preferably has software that allows at least the following procedural measures: - Entering parameter values - Playback of audio signals on the terminal device, - Evaluation of the auditory impression of the audio signals by the test subject - Saving the evaluation result or transmitting the evaluation result to an external storage device can be carried out.

[0016] Audio signals, which reproduce the sound of the machine, aircraft, or vehicle based on the entered parameter values, can be retrieved from external storage that is not the device's own storage. The audio signals can either be generated based on the entered parameter values ​​or have already been previously generated based on identical or similar parameter values ​​and stored in external storage.

[0017] During or after the test subject enters their subjective rating, the rating and preferably also the associated audio signals can be saved on the device. Alternatively or additionally, the rating can be transmitted to an external storage device, where it is preferably linked to the audio signal that served as the basis for the rating.

[0018] Audio signals that reproduce the sound of the machine, aircraft or vehicle based on the entered parameter values ​​can be obtained or have been obtained as follows: a. Calculation of the noise emission of at least one sound source of the machine, aircraft or vehicle based on design or operating parameter values b. Simulation of noise immission, preferably virtual simulation c. Auralization and preferably the creation of an audio file.

[0019] Steps a., bc are hereinafter also referred to as the analytical process chain.

[0020] A well-known tool for the described analytical process chain is, for example, PropNoise, described in doi:10.14279 / depositonce-5935 and doi:10.14279 / depositonce-7057, VIOLIN. Further literature on this topic can be found in A. Moreau, A. Prescher, S. Schade, M. Dang, R. Jaron, and S. Guérin, "A framework to simulate and to auralize the sound emitted by aircraft engines," InterNoise conference 2023 conference proceedings (2023).

[0021] For the virtual simulation of noise emissions, the route of a vehicle or aircraft, such as an aircraft's flight path, can be used as additional operating parameters. These operating parameters, along with the operating parameters of a sound source, are also relevant to the perceived sound.

[0022] The analytical process chain described above is preferably applied to a variety of parameter configurations.

[0023] The generated audio files are preferably stored once in an audio database. Each combination of design or operating parameters is then linked to an audio file that can be played back on the end device.

[0024] By entering parameter values, the test subject can configure a machine, aircraft, or vehicle based on these values. For example, the test subject can configure a drive system. When playing audio signals on the device, the linked audio file can be played, preferably retrieved from an audio database. The audio file can reproduce a calculated noise emission. The test subject can then evaluate the auditory impression.

[0025] Determining a correlation between parameter values, or sets of parameter values, and the subjective evaluation result can reveal: The generation of multiple datasets, each containing the following linked together: - Design parameter values ​​and / or operating parameter values - Audio file containing the audio signals resulting from the selected design and / or operation - Subjective evaluation result of the audio signals of the audio file

[0026] The subjective evaluation result preferably includes how disturbing / unpleasant or pleasant the generated audio signal is perceived to be.

[0027] A predetermined evaluation result can be defined according to a target specification. A predetermined evaluation result is preferably one that is desirable for the operation of the machine, aircraft, or vehicle. A positive evaluation result according to a selected rating scale is preferred.

[0028] When operating the machine, aircraft or vehicle, parameter values ​​can be used in such a way that these parameter values ​​serve as a specification for a control element which adjusts the operating parameters of the machine, aircraft or vehicle during operation in such a way as to minimize the impairments perceived as negative, in particular psychoacoustic effects.

[0029] By determining the relationship between parameter values, or sets of parameter values, and the subjective evaluation result, a connection can be established between the parameter values ​​and the positively / negatively evaluated sounds. This allows for tracing which parameters are linked to positively or negatively evaluated sounds. Furthermore, it is possible to analyze which parameters show a correlation with positively or negatively evaluated sounds. A relationship between positively / negatively evaluated sounds and parameters can be obtained, so that the parameters are linked not only to the audio files but also to the subjective sound evaluations. The design and / or operation of the machine, aircraft, or vehicle is based on this information.The invention thus enables the obtaining of design guidelines and specific constructions for future aircraft engines, which result in the least disturbing noise possible.

[0030] One example of aircraft operation is a change in flight altitude, especially at night and over densely populated areas. Conversely, under conditions of lower population exposure, other parameters, such as thrust and efficiency, can be maximized.

[0031] Examples of design parameters in the case of an aircraft include, without limitation: the type of propulsors (propellers or shrouded fans), their number, their position on or under the wing, the number of propeller blades, and, in the case of a fan, the number of rotor and stator blades or the fan diameter.

[0032] The term "design" may preferably refer to the development of a new aircraft or vehicle, while the term "optimization of the technical design" may preferably refer to an existing aircraft or vehicle. Similarly, the terms "design parameters" and "technical design parameters" may be differentiated. It is possible that "design parameters" and "technical design parameters" are synonymous. The term "technical design" preferably means a design or configuration that has an impact on technical characteristics.

[0033] Examples of technical design parameter values ​​include, without limitation: the type of propulsors (propellers or shrouded fans), their number, their position on or under the wing, the number of propeller blades, and, in the case of a fan, the number of rotor and stator blades or the fan diameter.

[0034] Examples of operating parameters of an aircraft, especially of the propulsors, are the rotational speed - which can also differ between individual propulsors - or the mass flow rate.

[0035] Other examples of operating parameters, using an aircraft as an example, are takeoff trajectories, landing trajectories, overflight trajectories, and flight altitudes.

[0036] In one embodiment of the method, the parameter values ​​are parameter values ​​of a drive system of the machine, aircraft, or vehicle, and the parameter values ​​are design parameter values, engineering design parameter values, and / or operating parameter values ​​of the drive system. The audio signals on the terminal device reproduce the noise of the drive based on the entered parameter values ​​of the drive, and The design, optimization of the technical design and / or operation is the design, optimization of the technical design and / or operation of the drive, and in the design, optimization of the technical design and / or operation, parameter values ​​are used that lead to noises of the drive in operation, which correlate with a subjective evaluation result that corresponds to a predetermined evaluation result, preferably a positive evaluation result.

[0037] In one embodiment of the method, the method has: - Measuring ambient noise with the terminal device during playback of the audio signals and / or during evaluation of the listening experience of the audio signals - Analyzing ambient noise with regard to type and / or physical characteristics, - Determining a relationship between the subjective assessment result and the type and / or physical characteristics of the ambient noise.

[0038] In this embodiment, the sensor technology of the terminal device is used, for example a microphone present on the terminal device, to open up the usage context as a data source.

[0039] Physical characteristics of ambient noise include, for example, volume or frequency spectrum.

[0040] This embodiment offers an additional advantage over conventional questionnaire methods: In conventional questionnaire methods, user input is often insufficiently contextualized. This lack of contextualization can lead to misclassifications in data processing using machine learning methods. To avoid missing contextualization, this embodiment offers the following technical solution: The end device's sensors are used to capture the usage context. For this purpose, ambient noise is recorded during use. It has been shown that the evaluation of an auditory impression of the audio signals, which reproduce the noise of the machine, aircraft, or vehicle based on the entered parameter values, also depends on the ambient noise. For example, the sound of an aircraft flying overhead in an otherwise silent environment, e.g.,In quiet nature, it is perceived as more disturbing / negative than in a noisy environment, for example an urban environment that already has traffic noise.

[0041] In this embodiment, the type and / or physical characteristics of the ambient noise can be used to modify the subjective evaluation result obtained for this type and / or these physical characteristics of the ambient noise for the further course of the process. A correlation between the parameter values, or sets of parameter values, and the modified subjective evaluation result can be obtained. The modification can take the form of a weighting.

[0042] In a preferred embodiment, determining the relationship between the subjective evaluation result and the type and / or physical characteristics of the ambient noise is achieved by weighting the subjective evaluation result with the type and / or physical characteristics of the ambient noise. Using the previous example of an aircraft in a "quiet natural environment" or an "urban environment," this can be explained as follows: The same overhead noise is perceived as more disturbing in a quiet natural environment than in an urban environment. Although the audio signals played back on the terminal device, which reproduce the noise of the machine, aircraft, or vehicle based on the entered parameter values, are the same, the subjective evaluation results therefore differ depending on the ambient noise. By weighting the subjective evaluation result, this deviation can be reduced or eliminated.The specifications for the design or operation of an aircraft may be such that, during operation, parameter values ​​are used that result in aircraft or vehicle noise levels that correlate with a positive rating in urban environments. In this case, the rating in quiet natural settings, i.e., with little to no ambient noise, can be weighted less. Alternatively, the reverse scenario is conceivable.

[0043] In a further variant of this embodiment, determining a relationship between the subjective evaluation result and the type and / or physical characteristics of the ambient noise is an analysis of the influence that the type and / or physical characteristics of the ambient noise have on the subjective evaluation result obtained with this type and / or these physical characteristics of the ambient noise. In this variant of this embodiment, the type and / or physical characteristics of the ambient noise can also be used to analyze their influence on the subjective evaluation result obtained with this type and / or these physical characteristics of the ambient noise.

[0044] After the aforementioned analysis of the influence of the type and / or physical characteristics of the ambient noise on the subjective evaluation result obtained with this type and / or these physical characteristics of the ambient noise, the following further step can be taken: - Changing an operating state of the machine, aircraft or vehicle to achieve, under the given ambient noise, a subjective evaluation result that corresponds to a modified predetermined evaluation result, preferably a positive or more positive evaluation result.

[0045] Beyond purely statistical analysis of the data, it is possible to deduce whether and to what extent a change in the operating states, e.g., a change in the rotational speed, can contribute to a change / improvement in the subjective evaluation of an aircraft configuration in a given ambient noise scenario (e.g., an urban scenario with acoustic energy distribution patterns in frequency ranges typical for road traffic noise pollution).

[0046] In one embodiment of the method, when analyzing the type of ambient noise, the ambient noise is classified according to its source, whereby the reference, in particular a weighting, of the subjective evaluation result differs depending on the classification of the source.

[0047] Classification can be done, for example, using an algorithm.

[0048] Sources or classifications include, but are not limited to: voices, road traffic, music.

[0049] Captured ambient noises can be automatically classified.

[0050] The classification of the usage environment can be done alternatively or cumulatively as follows: 1) Classification of the usage environment on the terminal device, preferably by numerical methods.

[0051] When recording ambient sounds, the surrounding noise can be classified, for example, into quiet and busy environments. Additionally, the geolocated location (via device interfaces) can be added as further information to support this categorization. Geolocation can indicate whether the test subject is in a busy or quiet environment. This allows the mobile hearing test data to be enriched for statistical analysis.

[0052] 2) Categorization of the usage environment on an external server, preferably using numerical methods. Recorded external noises can be classified using powerful methods on an external server and the results added to the classification on the terminal device in order to enrich the mobile hearing test data for statistical evaluation.

[0053] In one embodiment of the method, when analyzing the ambient noise with regard to physical characteristics, an analysis of at least the loudness, sound pressure level, frequency spectrum, periodicity, tonality, temporal patterns and / or frequency of the ambient noise is performed.

[0054] In one embodiment of the method, the method has: - Measuring the test subject's vital parameters with the terminal device or with a recording device coupled to the terminal device during playback of the audio signals and / or during evaluation of the auditory impression of the audio signals, - Determining the relationship between the subjective assessment result and the vital parameters.

[0055] A recording device paired with the terminal device is, for example, a smartwatch or a fitness tracker.

[0056] Sensors already present in the terminal device or the coupled recording device, such as cameras, motion sensors, microphones, etc., can be used to measure vital parameters.

[0057] In this embodiment, the vital parameters can be used to modify the subjective assessment result obtained for these vital parameters for the further course of the procedure. A correlation between the parameter values, or sets of parameter values, and the modified subjective assessment result can be obtained. The modification can take the form of weighting.

[0058] In a preferred embodiment, determining the relationship between the subjective evaluation result and the vital parameters involves weighting the subjective evaluation result with the vital parameters. This can be illustrated using the previous example of an aircraft with different stress levels as vital parameters. The stress level can be determined by heart rate and blood pressure. The same overhead noise is perceived as more disturbing at a high stress level than at a low stress level. Although the audio signals played back on the terminal device, which reproduce the noise of the machine, aircraft, or vehicle based on the entered parameter values, are the same, the subjective evaluation results therefore differ depending on the test subject's stress level. By weighting the subjective evaluation result, this deviation can be reduced or eliminated.The specifications for the design or operation of an aircraft may be such that, in the case of operation, parameter values ​​are used that result in aircraft or vehicle noises which correlate with a subjective assessment result corresponding to a predetermined assessment result at a low stress level, preferably a positive assessment result. In this case, the assessment result at a high stress level may be weighted less. Alternatively, the reverse case is conceivable.

[0059] In a further preferred embodiment, determining the relationship between the subjective assessment result and the vital parameters includes checking the plausibility of the subjective assessment result based on the vital parameters.

[0060] In this process, vital parameters are used to verify the test subject's subjective assessment of the noise or audio signals (pleasant / unpleasant rating). For example, this could be done as follows: a person rates a noise as pleasant, but their vital parameters (e.g., blood oxygen and heart rate) indicate an elevated stress level. In this example, the aforementioned weighting can be applied such that the supposedly positive subjective rating is given less weight due to vital parameters that contradict this assessment.

[0061] In one embodiment of the method, the method has: - Measuring the test subject's vital parameters with the terminal device or with a recording device coupled to the terminal device during playback of the audio signals and / or during evaluation of the auditory impression of the audio signals, - Determining a relationship between parameter values, or sets of parameter values, and vital parameters, - Designing, optimizing the technical design and / or operating the machine, aircraft or vehicle, taking into account the relationship of the parameter values, or sets of parameter values, with the vital parameters, whereby parameter values ​​are used in the design, optimization of the technical design and / or operation which, in the operation of the machine, aircraft or vehicle, lead to noises of the machine, aircraft or vehicle which correlate with vital parameters which correspond to specified vital parameters.

[0062] Predefined parameter values ​​are preferably those that can be associated with a healthy state of the person, such as blood pressure within a defined normal range, stress level within a defined normal range, or heart rate within a defined normal range.

[0063] In this embodiment, the vital parameters or changes in the vital parameters are considered a criterion variable themselves, alongside the subjective evaluation result, which may be a consciously verbalized assessment. This means that the results and possible relationships between the vital parameters and changes in the configurations and operating states, in addition to the subjective evaluation result, serve to identify advantageous configurations and operating states.

[0064] Vital signs can be measured in various ways. The following are some, but not exhaustive, examples. - Motion sensors of the device: The evaluation of motion data allows conclusions to be drawn about the activities of test subjects during the execution of the procedure [GM Harari, SR Müller, MS Aung, and PJ Rentfrow, “Smartphone sensing methods for studying behavior in everyday life”, Current Opinion in Behavioral Sciences, Vol. 18, pp. 83-90, Dec. 2017, doi: 10.1016 / j.cobeha.2017.07.018; W. Sousa Lima, E. Souto, K. El-Khatib, R. Jalali, and J. Gama, “Human Activity Recognition Using Inertial Sensors in a Smartphone: An Overview”, Sensors, Vol. 19, No. 14, p. 3213, July 2019, doi: 10.3390 / s19143213]. - Device camera, e.g., smartphone camera: eye movements and pupil dilation. In psychology, eye movements serve as indicators of a subject's attentional distribution. Recent studies combine the availability of smartphone cameras with algorithms that can compensate for the high variance in screen distance, camera position, shaking, etc. (compared to stationary eye-tracking devices) [B. Brousseau, J. Rose, and M. Eizenman, “SmartEye: An Accurate Infrared Eye Tracking System for Smartphones”, in 2018 9th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), New York City, NY, USA: IEEE, Nov. 2018, pp. 951–959. doi: 10.1109 / UEMCON.2018.8796799; C. Barry, J. De Souza, Y. Xuan, J. Holden, E. Granholm, and EJ Wang, "At-Home Pupillometry using Smartphone Facial Identification Cameras," in CHI Conference on Human Factors in Computing Systems, New Orleans LA USA: ACM, Apr. 2022, pp. 1-12. doi: 10.1145 / 3491102.3502493; N. Valliappan et al., “Accelerating eye movement research via accurate and affordable smartphone eye tracking”, Nat Commun, Vol. 11, No. 1, p. 4553, Sep. 2020, doi: 10.1038 / s41467-020-18360-5]. Changes in pupil size have been studied as involuntary responses to surprise, rewards, and other stimuli. For example, a particularly unpleasant airplane noise could cause a stronger gaze response. - Microphone: Respiratory rate. Initial studies achieved relatively accurate measurements of users' respiratory rates using microphone recordings [KK Phokela and V. Naik, “Use of Smartphone's Headset Microphone to Estimate the Rate of Respiration”, in 2020 International Conference on COMmunication Systems & NETworkS (COMSNETS), Bengaluru, India: IEEE, Jan. 2020, pp. 64-69. doi: 10.1109 / COMSNETS48256.2020.9027297; M. Alkhodari and AH Khandoker, “Detection of COVID-19 in smartphone-based breathing recordings: A pre-screening deep learning tool”, PLoS ONE, Vol. 17, No. 1, p. e0262448, Jan. 2022, doi: 10.1371 / journal.pone.0262448]. This can increase or decrease in response to stimuli. It therefore represents another implicit measure of users' response to the noise of machines, vehicles, or aircraft. - Smartwatch: Current research investigates the relationship between heart rate variability (HRV) signals derived from smartwatches and acute stress perception or stress vs. resting state [T. Chalmers, BA Hickey, P. Newton, C.-T. Lin, D. Sibbrit, CS McLachlan, R. Clifton-Bligh, . Morley, S. Lal. Stress Watch: The Use of Heart Rate and Heart Rate Variability to Detect Stress: A Pilot Study Using Smart Watch Wearables. Sensors 2022, 22, 151. https: / / doi.org / 10.3390 / s22010151; PE Velmovitsky, M. Lotto, P. Alencar, ST Leatherdale, D. Cowan and PP Morita. Can heart rate variability data from the Apple Watch electrocardiogram quantify stress? Front. Public Health. 2023 11:1178491. doi: 10.3389 / fpubh.2023.1178491]. While the results are not yet consistent with regard to the quality of predicting stress experience, the data obtained in this way appear to be able to distinguish between stress and resting states.

[0065] In one embodiment of the method, the vital parameters are selected from one or more of the following: heart rate, heart rate variability, blood pressure, body movement, eye movement, pupil dilation, sleep duration, respiratory rate of the test subject.

[0066] In one embodiment of the method, the method has: - Determining a relationship between the subjective evaluation result and the sound pressure level of the audio signal playback on the terminal device.

[0067] In this embodiment, the sound pressure level of the audio signal playback on the terminal device can be used to modify the subjective evaluation result obtained at that sound pressure level for the subsequent procedure. A correlation between the parameter values, or sets of parameter values, and the modified subjective evaluation result can be established. The modification can involve weighting.

[0068] In a preferred embodiment, determining the relationship between the subjective evaluation result and the sound pressure level involves weighting the subjective evaluation result by the sound pressure level. This can be illustrated using the example of an aircraft with varying sound pressure levels during playback. The same overhead noise is perceived as more disturbing at a high sound pressure level than at a low sound pressure level. Although the audio signals played back on the terminal device, which reproduce the noise of the machine, aircraft, or vehicle based on the entered parameter values, are identical, the subjective evaluation results therefore differ depending on the sound pressure level of playback. By weighting the subjective evaluation result, this deviation can be reduced or eliminated.The weighting can be done in such a way that an average sound pressure level of the playback is used as the standard condition.

[0069] In one embodiment of the method, the method has: - Relating the subjective evaluation result to a technical class of the terminal device.

[0070] Examples of technical classes of the terminal device, especially the playback device, are: Regarding connectivity, Bluetooth or wired; regarding wearing style, in-ear, on-ear or bone conduction; regarding design, open, semi-open or closed; regarding frequency response (roughly divided), bass-heavy, treble-heavy or neutral.

[0071] In this embodiment, the technical class of the terminal device can be used to modify the subjective evaluation result obtained with this technical class of the terminal device for the further course of the procedure. A correlation between the parameter values, or sets of parameter values, and the modified subjective evaluation result can be obtained. The modification can take the form of a weighting.

[0072] In one embodiment of the method, the method has: - Analyzing audio signals with regard to psychoacoustic characteristics, - Determining a correlation between the psychoacoustic characteristics of the audio signals and the subjective evaluation result, and furthermore with the parameter values, where the design, optimization of the technical design and / or operation of the machine, aircraft or vehicle is additionally carried out taking into account the relationship between the psychoacoustic characteristics of the audio signals and the subjective evaluation result, wherein parameter values ​​are used in the design, optimization of the technical design and / or operation which lead to psychoacoustic characteristics of noises caused by the operation of the machine, aircraft or vehicle, which correlate with a subjective evaluation result that corresponds to a predetermined evaluation result, preferably a positive evaluation result.

[0073] In this embodiment, for example, a connection is established between positively / negatively evaluated audio signals and psychoacoustic characteristics (e.g. tonality, sharpness, fluctuation strength) of the audio signals.

[0074] The psychoacoustic parameters, such as loudness, sharpness, fluctuation strength, roughness, tonality, and psychoacoustic annoyance, quantify the subjective perception of sound by the human ear, in contrast to the sound pressure level, which quantifies sound based on purely physical pressure fluctuations.

[0075] The subjective evaluation result is then additionally linked to the psychoacoustic parameters.

[0076] In particular, the preferred audio signals are correlated with the design parameters of the target product and the psychoacoustic characteristics obtained from the audio signals in order to obtain information on how the design should be carried out.

[0077] This embodiment allows for the determination of how psychoacoustic characteristics are linked to design parameters, which design parameters lead to pleasant or unpleasant listening experiences, which psychoacoustic characteristics influence this perception, and whether sounds perceived as pleasant elicit fewer physiological reactions than sounds perceived as annoying, unpleasant, or threatening. Answering these questions enables the derivation of recommendations for noise-optimized design, for example, of engines. In particular, design guidelines and specific designs for future aircraft engines can be derived, ensuring that the listening experience is perceived as pleasant and as minimally disturbing as possible (according to the design-to-perceived-noise approach).

[0078] Analyzing audio signals with regard to psychoacoustic characteristics, especially those that are crucial for a positive / negative noise rating, can be done using established commercial software (e.g., HEAD Artemis Suite) or open-source programs (e.g., SQAT).

[0079] In one embodiment of the method, the psychoacoustic parameters are selected from one or more of the following: loudness, frequency, tonality, roughness, fluctuation strength, sound pressure level, Effective Perceived Noise Level (EPNL), sound exposure level (SEL) and / or sharpness.

[0080] The audio signals are evaluated by the test subjects. Based on these evaluations from different test subjects, sounds can be grouped, for example, within the framework of cluster analysis. The audio signal files can be labeled and later grouped according to sound perception (e.g., pleasant vs. unpleasant).

[0081] When analyzing audio signals, the focus is primarily on identifying noise components that are responsible for sounds perceived as unpleasant. These can be individual frequencies, frequency sequences, frequency combinations, etc.

[0082] For the analysis and determination of the aforementioned relationship between the psychoacoustic characteristics of the audio signals and the subjective evaluation result, two methods are preferred, which are described in the following embodiments: frequency decomposition methods and machine learning methods.

[0083] In one embodiment of the method, the determination of a relationship between the physical characteristics of the audio signals and the subjective evaluation result is carried out by a frequency decomposition method, wherein frequencies, frequency sequences and / or frequency combinations are determined that correlate with a subjective evaluation result corresponding to a predetermined evaluation result, preferably a positive evaluation result.

[0084] An example of a frequency decomposition method is Empirical Mode Decomposition (EMD). This method is used to analyze nonlinear and non-stationary signals. Individual frequencies can be isolated and analyzed from the complete signal. For example, it makes it possible to examine audio signals with regard to specific frequencies and to identify commonalities between sounds perceived as pleasant or unpleasant.

[0085] In one embodiment of the method, the correlation between the physical characteristics of the audio signals and the subjective evaluation result is determined by means of a spectrogram analysis. A spectrogram shows the intensity (volume) of different frequencies over time. The horizontal axis represents time, the vertical axis the frequency, and the color intensity represents the amplitude or volume of the respective frequency.

[0086] Preferably, one or more of the following analyses are performed in spectrogram analysis: - Determining the dominant frequency range of a signal. - Detection of characteristic frequency patterns. - Detection of temporal changes in the audio signal. - Event detection through recurring patterns. - Detecting and isolating unwanted noise outside the desired frequency range. - Pattern recognition for automated analysis. - Band-wise energy measurement to identify dominant frequencies.

[0087] Accordingly, a dominant frequency range of a signal, a characteristic frequency pattern, a temporal change in the audio signal, recurring patterns, noise outside a desired frequency range, patterns for automated analysis and / or dominant frequencies can be determined that correlate with a subjective evaluation result corresponding to a predetermined evaluation result, preferably a positive evaluation result.

[0088] In one embodiment of the method, the determination of a relationship between the physical characteristics of the audio signals and the subjective evaluation result is carried out by a machine learning method, whereby frequencies, frequency sequences and / or frequency combinations are determined that correlate with a subjective evaluation result that corresponds to a predetermined evaluation result, preferably a positive evaluation result.

[0089] This embodiment makes it possible to identify noise components perceived as pleasant or unpleasant, as well as their sources, in order to take them into account during the development process, e.g., of technology in the field of aviation.

[0090] The machine learning process is preferably based on artificial intelligence.

[0091] In particular, the relationship between the parameter values, or sets of parameter values, obtained by the method according to the invention, and the subjective evaluation result, or several such relationships, is used to train an artificial intelligence. This enables the evaluation of an auditory impression of the audio signals by a machine process instead of by a test subject.

[0092] Artificial intelligence can learn, in particular, the relationships between a noise signature and perceived annoyance. Based on the trained data, it is then possible to predict perceived noise annoyance even for construction parameter values ​​(new configurations) that have not yet been evaluated by test subjects. Predictions about the psychoacoustics of these new configurations are also possible. This could be achieved, among other things, by integrating XAI approaches to explain the features in the noise signature that contribute to perceived annoyance.

[0093] Audio signals can be represented as spectrograms, and an artificial intelligence learns to infer subjective evaluations from these spectrograms. The trained system is then able to evaluate new sounds from audio signals within the spectrum of the training data in terms of a person's or test subject's perception of sound. Furthermore, it is possible to explain the AI ​​algorithm's decision-making process. The regions in the spectrogram that led to the AI ​​algorithm's decision can be highlighted. These regions represent specific frequency groups but can also be interpreted in terms of their origin.

[0094] The invention is described below with reference to exemplary embodiments. The figures shown are: Fig. 1 a basic procedural sequence of a method according to the invention; Fig. 2. A detail from the procedural sequence.

[0095] Fig. 1 illustrates the procedure using the specific example of the construction of a [unit of text]. The [unit of text] Fig. The procedure shown in step 1 consists of: S1: the input of parameter values ​​of an aircraft by at least one test subject on a terminal device, where the parameter values ​​are design parameter values ​​of the aircraft engine. S2: the playback of audio signals (S2) on the terminal device, which reproduce the sound of the aircraft based on the entered parameter values. S3 involves the evaluation of an auditory impression (S3) of the audio signals by the test subject and obtaining a subjective evaluation result from the test subject. S4 the entry of the subjective rating result by the test subject on the terminal device, S5 the determination of a relationship between the parameter values, or sets of parameter values, and the subjective evaluation result, S6 the design of the aircraft taking into account the relationship of the parameter values, or sets of parameter values, with the subjective evaluation result, wherein parameter values ​​are used in the design and / or operation which in operation lead to noises of the aircraft which correlate with a subjective evaluation result that corresponds to a predetermined evaluation result, preferably a positive evaluation result.

[0096] The Fig. Figure 2 shows steps S1-S4 in more detail using a smartphone with the corresponding app as the end device.

[0097] In Fig. Figure 2 on the left shows how the test subject (not shown), who owns the smartphone, selects the design and operating parameter values: number of drives with the value 24, number of rotor blades of a drive with the value 2, number of stator blades of a drive with the value 16, rotational speed of the drives with 3680 rpm and mass flow with 600 kg / s.

[0098] Then, as in the middle of the Fig. Figure 2 shows an audio file being retrieved and played back from an external storage device (database) in the form of audio signals that reproduce the sound of the aircraft based on these design parameter values.

[0099] In Fig. Figure 2 on the right shows that the test subject has evaluated the auditory impression of the audio signals and entered a subjective rating. On a scale of 0 to 10, the test subject indicates how unpleasant the sound of the audio signals from the audio file is perceived to be, with the value given here being 4, in the middle range from "not unpleasant at all" to "extremely unpleasant." "Not unpleasant at all" can be considered a positive rating, although depending on the definition, a value up to 2 or 3 could also be considered positive.

[0100] In another query in Fig. In the lower right corner, an additional assessment is requested under certain boundary conditions, in this case, the assumption that the person is sitting in a garden, which could represent a quiet environment or a feeling of low stress. However, according to the invention, it is preferably provided that ambient noise and vital parameters are determined using a smartphone or a connected smartwatch worn by the test subject, and that the assessment result in the upper right corner is weighted accordingly, so that these influencing factors can be taken into account for test subjects with different vital parameters or in different environments.

[0101] If a positive evaluation result is received, for example a rating of 0, 1 or 2 in the scale at the top right in Fig. 2. The underlying design parameters, which are in Fig.The two links associated with this evaluation result serve as the basis for the design of the specific aircraft. It is preferred to have a large number of test subjects complete the procedure and statistically analyze their results. A positive result can be assumed, for example, if the average rating is no greater than 1 or 2 and the standard deviation is less than or equal to a predefined threshold. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited non-patent literature

[0000] doi:10.14279 / depositonce-5935, doi: 10. 14279 / depositonce-7057, VIOLIN, CORA

[0002] A. Moreau, A. Prescher, S. Schade, M. Dang, R. Jaron, S. Guérin, A framework to simulate and to auralize the sound emitted by aircraft engines, InterNoise conference 2023 conference proceedings (2023 [0002, 0020] Lotinga, Ramos-Ramero, Green & Torija, 2023, https: / / doi.org / 10.1007 / s40726-023-00285-4

[0003] Janssens, Vecchio & Van der Auweraer, 2008, https: / / doi.org / 10.1016 / j.ast.2007.10.002; Gwak, Han & Lee, 2020, https: / / doi.org / 10.1016 / j.jsv.2020.115651, Christian & Cabell, 2017, D01:10.2514 / 6.2017-4051

[0003] PropNoise, beschrieben in doi:10.14279 / depositonce-5935, doi: 10. 14279 / depositonce-7057, VIOLIN

[0020] G. M. Harari, S. R. Müller, M. S. Aung, und P. J. Rentfrow, „Smartphone sensing methods for studying behavior in everyday life“, Current Opinion in Behavioral Sciences, Bd. 18, S. 83-90, Dez. 2017, doi: 10.1016 / j.cobeha.2017.07.018; W. Sousa Lima, E. Souto, K. El-Khatib, R. Jalali, und J. Gama, „Human Activity Recognition Using Inertial Sensors in a Smartphone: An Overview“, Sensors, Bd. 19, Nr. 14, S. 3213, Juli 2019, doi: 10.3390 / s19143213

[0064] B. Brousseau, J. Rose, und M. Eizenman, „SmartEye: An Accurate Infrared Eye Tracking System for Smartphones“, in 2018 9th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), New York City, NY, USA: IEEE, Nov. 2018, S. 951-959. doi: 10.1109 / UEMCON.2018.8796799; C. Barry, J. De Souza, Y. Xuan, J. Holden, E. Granholm, und E. J. Wang, „At-Home Pupillometry using Smartphone Facial Identification Cameras“, in CHI Conference on Human Factors in Computing Systems, New Orleans LA USA: ACM, Apr. 2022, S. 1-12. doi: 10.1145 / 3491102.3502493; N. Valliappan u. a., „Accelerating eye movement research via accurate and affordable smartphone eye tracking“, Nat Commun, Bd. 11, Nr. 1, S. 4553, Sep. 2020, doi: 10.1038 / s41467-020-18360-5

[0064] K. K. Phokela und V. Naik, „Use of Smartphone's Headset Microphone to Estimate the Rate of Respiration“, in 2020 International Conference on COMmunication Systems & NETworkS (COMSNETS), Bengaluru, India: IEEE, Jan. 2020, S. 64-69. doi: 10.1109 / COMSNETS48256.2020.9027297; M. Alkhodari und A. H. Khandoker, „Detection of COVID-19 in smartphone-based breathing recordings: A pre-screening deep learning tool“, PLoS ONE, Bd. 17, Nr. 1, S. e0262448, Jan. 2022, doi: 10.1371 / journal.pone.0262448

[0064] T. Chalmers,B.A. Hickey, P. Newton,C.-T. Lin, D. Sibbrit,C.S. McLachlan, R. Clifton-Bligh, . Morley, S. Lal. Stress Watch: The Use of Heart Rate and Heart Rate Variability to Detect Stress: A Pilot Study Using Smart Watch Wearables. Sensors 2022, 22, 151. https: / / doi.org / 10.3390 / s22010151; P.E. Velmovitsky, M. Lotto, P. Alencar, S.T. Leatherdale, D. Cowan and P.P Morita. Can heart rate variability data from the Apple Watch electrocardiogram quantify stress? Front. Public Health. 2023 11:1178491. doi: 10.3389 / fpubh.2023.1178491

[0064]

Claims

[1] comprising a method for the design, optimization of the technical design and / or operation of a machine, aircraft or vehicle - Input (S1) of parameter values ​​of the machine, aircraft or vehicle by at least one test subject on an end device, wherein the parameter values ​​are design parameter values, technical design parameter values ​​and / or operational parameter values ​​of the machine, aircraft or vehicle, - Playback of audio signals (S2) on the terminal device, which reproduce the sound of the machine, aircraft or vehicle based on the entered parameter values, - Evaluating the auditory impression (S3) of the audio signals by the test subject and obtaining a subjective evaluation result from the test subject, - Input (S4) of the subjective evaluation result by the test subject on the terminal device, - Determining (S5) a relationship between the parameter values, or sets of parameter values, and the subjective evaluation result, - Design (S6), optimization of the technical design and / or operation of the machine, aircraft or vehicle, taking into account the relationship of the parameter values, or sets of parameter values, with the subjective evaluation result, wherein in the design, optimization of the technical design and / or operation parameter values ​​are used which, in the operation of the machine, aircraft or vehicle, result in noises of the machine, aircraft or vehicle that correlate with a subjective evaluation result corresponding to a predetermined evaluation result, preferably a positive evaluation result. [2] The method of claim 1, wherein the parameter values ​​are parameter values ​​of a drive of the machine, aircraft or vehicle and the parameter values ​​are design parameter values, technical design parameter values ​​and / or operating parameter values ​​of the drive, and the audio signals on the terminal device reproduce the noise of the drive based on the entered parameter values ​​of the drive, and the design, optimization of the technical design and / or operation of the drive is and In the design, optimization of the technical design and / or operation, parameter values ​​are used that lead to noises from the drive during operation, which correlate with a subjective evaluation result that corresponds to a predetermined evaluation result, preferably a positive evaluation result. [3] Method according to claim 1 or 2, comprising - Measuring ambient noise with the terminal device during playback of the audio signals and / or during evaluation of the listening experience of the audio signals - Analyzing ambient noise with regard to type and / or physical characteristics, - Determining a relationship between the subjective assessment result and the type and / or physical characteristics of the ambient noise. [4] Method according to claim 3, wherein when analyzing the nature of the ambient noise, the ambient noise is classified according to its source, the reference of the subjective evaluation result differs depending on the classification of the source. [5] Method according to claim 3 or 4, wherein the analysis with regard to physical characteristics includes an analysis of at least loudness, sound pressure level, frequency spectrum, periodicity, tonality, temporal patterns and / or frequency of the ambient noise. [6] Method according to one of claims 3-5, wherein the type and / or physical characteristics of the ambient noise are used to modify the subjective evaluation result obtained for this type and / or these physical characteristics of the ambient noise for the further course of the method, wherein a relationship of the parameter values, or sets of parameter values, with the modified subjective evaluation result is obtained, wherein the modification is preferably a weighting of the subjective evaluation result. [7] Method according to any of the preceding claims, comprising - Measuring the test subject's vital parameters with the terminal device or with a recording device coupled to the terminal device during playback of the audio signals and / or during evaluation of the auditory impression of the audio signals, - Determining the relationship between the subjective assessment result and the vital parameters. [8] Method according to claim 7, wherein the vital parameters are selected from one or more of the following: heart rate, heart rate variability, blood pressure, body movement, eye movement, pupil dilation, sleep duration, respiratory rate of the test subject. [9] Method according to any one of the preceding claims, comprising - Determining a relationship between the subjective evaluation result and the sound pressure level of the audio signal playback on the terminal device. [10] Method according to claim 9, further comprising - Relating the subjective evaluation result to a technical class of the terminal device. [11] Method according to any of the preceding claims, comprising - Analyzing audio signals with regard to psychoacoustic characteristics, - Determining a correlation between the psychoacoustic characteristics of the audio signals and the subjective evaluation result, where the design, optimization of the technical design and / or operation of the machine, aircraft or vehicle is additionally carried out taking into account the relationship between the psychoacoustic characteristics of the audio signals and the subjective evaluation result, wherein parameter values ​​are used in the design, optimization of the technical design and / or operation which lead to psychoacoustic characteristics of noise caused by the operation of the machine, aircraft or vehicle, and which correlate with a subjective evaluation result that corresponds to a predetermined evaluation result, preferably a positive evaluation result. [12] Method according to claim 11, wherein the psychoacoustic parameters are selected from one or more of the following: loudness, frequency, tonality, roughness, fluctuation strength, sound pressure level, Effective Perceived Noise Level (EPNL), sound exposure level (SEL) and / or sharpness. [13] Method according to claim 11 or 12, wherein the determination of a relationship between the physical characteristics of the audio signals and the subjective evaluation result is carried out by a frequency decomposition method, wherein frequencies, frequency sequences and / or frequency combinations are determined and correlate with a subjective evaluation result that corresponds to a predetermined evaluation result, preferably a positive evaluation result. [14] Method according to one of claims 11-13, wherein the determination of a relationship between the physical characteristics of the audio signals and the subjective evaluation result is carried out by a machine learning method, wherein frequencies, frequency sequences and / or frequency combinations are determined which correlate with a subjective evaluation result corresponding to a predetermined evaluation result, preferably a positive evaluation result. [15] Method according to one of claims 11-14, wherein the determination of a relationship between the physical characteristics of the audio signals and the subjective evaluation result is carried out by a spectrogram analysis, wherein a dominant frequency range of a signal, a characteristic frequency pattern, a temporal change in the audio signal, recurring patterns, noise outside a desired frequency range, patterns for automated analysis and / or dominant frequencies are determined which correlate with a subjective evaluation result that corresponds to a predetermined evaluation result, preferably a positive evaluation result.

Images