Data Augmentation and Deep Learning Methods for Pressure Prediction in Ignition Process of Solid Rocket Motors

Polymeric composites are widely used in aerospace. One such important use of polymeric composites is the filament-wound composite case for the Solid Rocket Motor (SRM). Environmental conditions expected during flight and other phases of the SRM’s life cycle affect its performance. The performance of SRM, in turn, depends primarily on the composite case and propellant grain. In this study, the performance evaluation of SRM with the composite case is carried out for different environmental conditions. These cases are designed to sustain very high pressure and temperature environments generated due to the combustion of propellant grains and are intended to perform adequately at any time during their service life. The performance evaluation test matrix is formulated considering various expected dynamic and thermal conditions. The small SRM with a composite case is subjected to planned tests as per the matrix, followed by static tests. The pressure versus time data is plotted for all test conditions. The ballistic parameters result of the virgin (without subjecting it to any test) small SRM is considered as a reference for comparative studies with the rest of all other test conditions results. All the ballistic performance test results are also analysed with respect to the pressure-time curve for mission bounds. The analysis of experimental test results reveals that various environmental dynamic conditions like random vibration, acceleration and shock, do not affect the composite case and the ballistic performance of a small SRM. The effect of grain temperature on ballistic performance is also found as expected.

A number of options for Solid Rocket Motor (SRM) performance improvement were studied and evaluated by NASA with respect to payload improvement potential, technical risk, schedule, cost, and impact on Shuttle system performance. It was found that the SRM delivered specific impulse could be improved by increasing the nozzle expansion ratio from 7.16 to 7.72. This was achieved by decreasing the throat diameter of the nozzle exit cone, while staying within tooling, facility, and ICD limitations. A simple modification of the grain inhibiting pattern was made to reshape the thrust-time history and thereby obtain more delivered impulse during the early portion of motor operation. These performance options, initiated in October 1980, provide a 3,000 lb payload increase. Longer range options are the use of a filament wound composite case, a HTPB propellant, and further increases in the nozzle expansion ratio. These improvements have a potential of 10,000 to 11,000 lb payload increase.

Solid rocket motor (SRM) is a core component of rockets. Considering the high cost of solid rocket motors and the high demand for transportation and storage, there is an urgent need to predict the performance of solid rocket motors before the ground ignition test to lower the cost of tests. Using data mining technology to predict various parameters of rocket motors with historical data has become a new development direction. In order to improve the data fitting effect and prediction accuracy, a novel prediction model was proposed, which tries to combine the Whale Optimization Algorithm (WOA) with the Error Correction Gated Recurrent Unit neural network (ECGRU) to correct the basic prediction performance of the Long Short-Term Memory neural network model (LSTM). Compared with traditional methods, experiments showed that the novel model was beneficial in improving the reliability of the prediction of the thrust and specific impulse.

Curing activity in the preparation of solid composite propellants determines the performance of solid rocket motors in operation. Limited by the lack of effective monitoring tools, the complete curing behavior and thermal-induced curing kinetics are rarely disclosed. It is still a challenge to monitor in situ and in real-time the physical and chemical cross-linking reaction during the curing of propellant. Herein, we demonstrate a promising approach based on optical fiber capable of being implanted inside the propellant to monitor the internal stress evolution during the curing process, by taking hydroxyl-terminated polybutadiene propellant as an example. Attributed to the strain and temperature sensitivity of a pair of optical fiber gratings, the thermal-assisted physico-chemical cross-linking states of curing process have been demonstrated in detail. By tracking the stress-induced wavelength shifts of fiber gratings and calculating the curing mechanism function, the complete curing roadmap, including the viscous flow stage, gel stage, hardening stage can be clearly revealed, and the curing completion times are obtained as 154, 81, and 40 h, at the curing temperatures of 60, 70, and 80 °C, respectively. The apparent activation energy of this curing system obtained by calculation is 73.88 kJ/mol. This flexible fiber-based sensor provides an effective tool for unraveling the cure kinetic mechanism, and paves a universal pathway to guide the preparation and applications of versatile composite materials for solid rocket motors.

A lightweight rubber foaming insulation reinforced by carbon nanotubes and carbon fibers for solid rocket motors

Nitrate ester plasticized polyether (NEPE) propellants, are high energy solid propellants that are used widely in solid rocket motor. It is necessary to study the mechanical properties of NEPE propellants in order to ensure an accurate prediction of the performance of solid rocket motors and ensure the safety of launch platforms.We studied the viscoelastic hyperelastic constitutive model of NEPE propellants. The speckle defocusing method and digital image correlation(DIC) method are used to obtain the strain of solid propellant NEPE in three directions during unidirectional stretching, and the volume change rate, Poisson's ratio and true strain curves of NEPE under different stretching rates are obtained. With the increase of loading rate, the maximum stress that NEPE can encounter increases, and the modulus at the initial elastic stage also increases. At the initial stage of stretching, the volume change rate is close to 1, the Poisson's ratio fluctuates within a range close to 0.5. When the engineering principal strain in the tensile direction reaches a certain value, the volume change rate begins to increase and the Poisson's ratio begins to decrease. The strain of solid propellant NEPE in three directions during uniaxial tension was successfully measured by speckle defocus method and digital image correlation (DIC) method. NEPE is close to being a super elastic material at the initial stage of stretching. The change trend of the volume change rate curve of different loading rates is consistent, and the stretching rate has no effect on the volume change rate of NEPE.

Due to the influence of internal factors and external environmental factors, the internal ballistic performance of solid rocket motors (SRM) shows certain deviation during its production and working. In order to describe the internal ballistic performance more accurately, a Monte Carlo method-based internal ballistic prediction and analysis method is proposed, and compared with the traditional limit deviation method. First, a simulation model of internal ballistics was established. Then according to the probability distribution law of the input factors, the simulation was carried out. The results show that under the premise of the same input deviation boundary, the probabilistic method can effectively reduce the dispersion of the internal ballistic performance parameters. Finally, according to the established internal ballistic probabilistic model, the sensitivity of the specific impulse to the influencing parameters was analyzed, and the degree of influence of each parameter to the specific impulse was obtained.

High-speed motion analysis system could record images up to 12,000fps and analyzed with the image processing system. The system stored data and images directly in electronic memory convenient for managing and analyzing. The high-speed motion analysis system and the X-ray radiography system were established the high-speed real-time X-ray radiography system, which could diagnose and measure the dynamic and high-speed process in opaque. The image processing software was developed for improve quality of the original image for acquiring more precise information. The typical applications of high-speed motion analysis system on solid rocket motor (SRM) were introduced in the paper. The research of anomalous combustion of solid propellant grain with defects, real-time measurement experiment of insulator eroding, explosion incision process of motor, structure and wave character of plume during the process of ignition and flameout, measurement of end burning of solid propellant, measurement of flame front and compatibility between airplane and missile during the missile launching were carried out using high-speed motion analysis system. The significative results were achieved through the research. Aim at application of high-speed motion analysis system on solid rocket motor, the key problem, such as motor vibrancy, electrical source instability, geometry aberrance, and yawp disturbance, which damaged the image quality, was solved. The image processing software was developed which improved the capability of measuring the characteristic of image. The experimental results showed that the system was a powerful facility to study instantaneous and high-speed process in solid rocket motor. With the development of the image processing technique, the capability of high-speed motion analysis system was enhanced.

The performance of solid rocket motors is limited by finite-rate chemical reactions occurring in the nozzle expansion. In the present work, a numerical approach is adopted in order to investigate the properties of the reactive flow and their impact on the motor specific impulse. Numerical simulations are performed on two different test cases, the second stage of the Minuteman III and the third stage of the Vega launcher, comparing the impact that different chemical mechanisms have on flow evolution and specific impulse estimate. The outcomes of the analysis suggest that CFD simulations represent a viable and robust method to provide a satisfactory prediction of the chemical non-equilibrium efficiency, outperforming semi-empirical laws when it comes to dealing with new solid motors of latest generation.

: This report describes progress on a collaborative research program combining the expertise of individuals from several universities to develop a new ability to predict the propulsion performance of solid rocket motors. The focus of the research on nonsteady behavior is unique and the overall project is not possible at any one of the institutions participating in this coordinated research. The individual tasks which we are studying will pursue solid propellant decomposition under unsteady conditions, nonsteady aspects of gas phase flame structure measurements, numerical modeling of multidimensional flame structure, propellant/flame interactions and overall nonsteady propellant combustion characteristics in realistic rocket motor environments. Our goal has been to develop general models of fundamental mechanisms of combustion instability that can be applied to a variety of new energetic materials.

In the captive testing of conventional solid rocket motors, a large number of fundamental physical measurements are normally implemented. These include the determination of instantaneous motor force, pressure, temperature, deflection, acceleration and electrical power data associated with the particular motor configuration. In the majority of applications, however, the motor performance indicators of most general interest are not measured directly, but are calculated from these basic data. Thus, for example, motor specific impulse is computed from given instantaneous values of measured thrust, pressure and weight parameters. All physical measurements are subject to inaccuracies, i.e., even the most sophisticated data acquisition system will necessarily inject discrepancies into the measured data. Hence, the mechanisms through which errors inherent in the measured quantities affect the accuracy of the calculated ballistic parameters are of extreme practical importance and must be quantitatively defined. At the Aerojet-General Corporation, an intensive and continuing analysis of overall measurement system errors (through rigorous analytical and experimental techniques) has permitted the statistical characterization of the error parameters within the framewcrk of each current large scale motor development program. Based upon these data and for a specific ballistic parameter computational routine, an extensive error analysis is required to deduce the errors associated with the computed performance parameters. Because of the nonlinearities and interactions inherent within the ballistic computations, a Monte Carlo technique is invoked.

Data augmentation (DA) increases the diversity of training data to improve the model generalization ability. Most of the DA methods focus on image classification or object detection. Directly using the existing DA methods on change detection tasks not only falls short of fully exploring the specificity of the change image pairs but also leads to longer training times. In this article, we first propose a mask-guided mixing (MGM) DA for change detection, which mixes the change regions of the current training sample based on prediction results and labels to generate high-quality samples with more positive samples. We then propose a new reinforcement learning (RL)-based Adaptive DA method, AdaAug+, to adaptively select the optimal DA policy for the training samples. An actor selects the best augmentation operation from the operation set according to the image pair. The augmented image pairs make it easier for the change detector to learn the optimal parameters and improve the final detection performance. To reduce the training time, we identify and remove the redundant training samples during the training process by our redundancy searching policy. We have conducted various experiments on four remote sensing change detection datasets with different change detectors. The experimental results demonstrate that AdaAug+ achieves promising performance compared to the state-of-the-art DA methods and requires less training time.

Fluid–solid coupled simulation of the ignition transient of solid rocket motor

Effects of particle size on two-phase flow loss in aluminized solid rocket motors

Launch vehicles experience peak vibro-acoustics stress during its captive firing or liftoff. The starting pressure rise transient in solid rocket motor chamber and subsequent flow of supersonic jet exhaust from the rocket nozzle generate high-intensity pressure waves. These pressure waves with different wavelengths are capable of exciting various structural elements of the launch vehicle. Especially the panel type of structural components located in very close vicinity to the supersonic jet and coming directly in the path of high-intensity pressure waves experience significant vibration. Hence, it is imperative to assess the vibration response of typical panel structures to high sound-level environment near the supersonic jet. Particularly, in the near field, these pressure waves behave in a nonlinear manner. Hence, developing a transfer function between vibrations to acoustics is essential for the proper design of structures for dynamic environment. In the present investigation, a typical panel structure instrumented with microphones and accelerometers are mounted near to a scaled down solid rocket motor supersonic jet exhaust. The plate is analyzed for its modes and frequency response function (FRF) by both experimental modal analysis (EMA) and finite element analysis (FEA). Also, the critical frequency of the plate is found out theoretically. The inclination of the plate with respect to the jet axis is varied for investigating the interaction of impinging and grazing waves. Further, the microphone and accelerometer measurements for different inclinations of the panel are analyzed by deriving power spectra and correlation functions. The ignition overpressure (IOP) wave interaction with the panel during the startup of the solid rocket motor is corroborated with the accelerometer response in the time domain. The high-intensity nonlinear Mach wave’s interaction with plate is analyzed from microphone and accelerometer time series data. The skewness of microphone and accelerometer data are compared. Finally, the frequency-dependent transfer function of vibration response to acoustic power input and vibrational efficiency factor is derived.