Machine learning for predicting the outcome of terminal ballistics events

Machine learning(ML) is well suited for the prediction of high-complexity,high-dimensional problems such as those encountered in terminal ballistics.We evaluate the performance of four popular ML-based regression models,extreme gradient boosting(XGBoost),artificial neural network(ANN),support vector regression(SVR),and Gaussian process regression(GP),on two common terminal ballistics’ problems:(a)predicting the V50ballistic limit of monolithic metallic armour impacted by small and medium calibre projectiles and fragments,and(b) predicting the depth to which a projectile will penetrate a target of semi-infinite thickness.To achieve this we utilise two datasets,each consisting of approximately 1000samples,collated from public release sources.We demonstrate that all four model types provide similarly excellent agreement when interpolating within the training data and diverge when extrapolating outside this range.Although extrapolation is not advisable for ML-based regression models,for applications such as lethality/survivability analysis,such capability is required.To circumvent this,we implement expert knowledge and physics-based models via enforced monotonicity,as a Gaussian prior mean,and through a modified loss function.The physics-informed models demonstrate improved performance over both classical physics-based models and the basic ML regression models,providing an ability to accurately fit experimental data when it is available and then revert to the physics-based model when not.The resulting models demonstrate high levels of predictive accuracy over a very wide range of projectile types,target materials and thicknesses,and impact conditions significantly more diverse than that achievable from any existing analytical approach.Compared with numerical analysis tools such as finite element solvers the ML models run orders of magnitude faster.We provide some general guidelines throughout for the development,application,and reporting of ML models in terminal ballistics problems.

Adaptive optimisation of explosive reactive armour for protection against kinetic energy and shaped charge threats

We evaluate an adaptive optimisation methodology,Bayesian optimisation(BO),for designing a minimum weight explosive reactive armour(ERA)for protection against a surrogate medium calibre kinetic energy(KE)long rod projectile and surrogate shaped charge(SC)warhead.We perform the optimisation using a conventional BO methodology and compare it with a conventional trial-and-error approach from a human expert.A third approach,utilising a novel human-machine teaming framework for BO is also evaluated.Data for the optimisation is generated using numerical simulations that are demonstrated to provide reasonable qualitative agreement with reference experiments.The human-machine teaming methodology is shown to identify the optimum ERA design in the fewest number of evaluations,outperforming both the stand-alone human and stand-alone BO methodologies.From a design space of almost 1800 configurations the human-machine teaming approach identifies the minimum weight ERA design in 10 samples.

A bayesian optimisation methodology for the inverse derivation of viscoplasticity model constants in high strain-rate simulations

We present an inverse methodology for deriving viscoplasticity constitutive model parameters for use in explicit finite element simulations of dynamic processes using functional experiments, i.e., those which provide value beyond that of constitutive model development. The developed methodology utilises Bayesian optimisation to minimise the error between experimental measurements and numerical simulations performed in LS-DYNA. We demonstrate the optimisation methodology using high hardness armour steels across three types of experiments that induce a wide range of loading conditions: ballistic penetration, rod-on-anvil, and near-field blast deformation. By utilising such a broad range of conditions for the optimisation, the resulting constitutive model parameters are generalised, i.e., applicable across the range of loading conditions encompassed the by those experiments(e.g., stress states, plastic strain magnitudes, strain rates, etc.). Model constants identified using this methodology are demonstrated to provide a generalisable model with superior predictive accuracy than those derived from conventional mechanical characterisation experiments or optimised from a single experimental condition.

Machine learning-based discovery of vibrationally stable materials

The identification of the ground state phases of a chemical space in the convex hull analysis is a key determinant of the synthesizability of materials.Online material databases have been instrumental in exploring one aspect of the synthesizability of many materials,namely thermodynamic stability.However,the vibrational stability,which is another aspect of synthesizability,of new materials is not known.Applying first principles approaches to calculate the vibrational spectra of materials in online material databases is computationally intractable.Here,a dataset of vibrational stability for~3100 materials is used to train a machine learning classifier that can accurately distinguish between vibrationally stable and unstable materials.This classifier has the potential to be further developed as an essential filtering tool for online material databases that can inform the material science community of the vibrational stability or instability of the materials queried in convex hulls.

Towards understanding structure–property relations in materials with interpretable deep learning

Deep learning(DL)models currently employed in materials research exhibit certain limitations in delivering meaningful information for interpreting predictions and comprehending the relationships between structure and material properties.To address these limitations,we propose an interpretable DL architecture that incorporates the attention mechanism to predict material properties and gain insights into their structure–property relationships.The proposed architecture is evaluated using two well-known datasets(the QM9 and the Materials Project datasets),and three in-house-developed computational materials datasets.Train–test–split validations confirm that the models derived using the proposed DL architecture exhibit strong predictive capabilities,which are comparable to those of current state-of-the-art models.Furthermore,comparative validations,based on first-principles calculations,indicate that the degree of attention of the atoms’local structures to the representation of the material structure is critical when interpreting structure–property relationships with respect to physical properties.These properties encompass molecular orbital energies and the formation energies of crystals.The proposed architecture shows great potential in accelerating material design by predicting material properties and explicitly identifying crucial features within the corresponding structures.