Search NASA⌕ Search

SEARCH · Search NASA

Results for “hybrid algorithm”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,603 records · Page 89

Ground Vibration Test Planning and Pre-Test Analysis for the X-33 Vehicle

This paper describes the results of the modal test planning and the pre-test analysis for the X-33 vehicle. The pre-test analysis included the selection of the target modes, selection of the sensor and shaker locations and the development of an accurate Test Analysis Model (TAM). For target mode selection, four techniques were considered, one based on the Modal Cost technique, one based on Balanced Singular Value technique, a technique known as the Root Sum Squared (RSS) method, and a Modal Kinetic Energy (MKE) approach. For selecting sensor locations, four techniques were also considered; one based on the Weighted Average Kinetic Energy (WAKE), one based on Guyan Reduction (GR), one emphasizing engineering judgment, and one based on an optimum sensor selection technique using Genetic Algorithm (GA) search technique combined with a criteria based on Hankel Singular Values (HSV's). For selecting shaker locations, four techniques were also considered; one based on the Weighted Average Driving Point Residue (WADPR), one based on engineering judgment and accessibility considerations, a frequency response method, and an optimum shaker location selection based on a GA search technique combined with a criteria based on HSV's. To evaluate the effectiveness of the proposed sensor and shaker locations for exciting the target modes, extensive numerical simulations were performed. Multivariate Mode Indicator Function (MMIF) was used to evaluate the effectiveness of each sensor & shaker set with respect to modal parameter identification. Several TAM reduction techniques were considered including, Guyan, IRS, Modal, and Hybrid. Based on a pre-test cross-orthogonality checks using various reduction techniques, a Hybrid TAM reduction technique was selected and was used for all three vehicle fuel level configurations.

Bedrossian, Herand↗

Verification and benchmarking relativistic electron beam transport through a background gas

It is necessary to establish confidence in high-consequence codes containing an extensive suite of physics algorithms in the regimes of interest. Verification problems allow code developers to assess numerical accuracy and increase confidence that specific sets of model physics were implemented correctly in the code. The two main verification techniques are code verification and solution verification. In this work, we present verification problems that can be used in other codes to increase confidence in simulations of relativistic beam transport. Specifically, we use the general plasma code EMPIRE to model and compare with the analytical solution to the evolution of the outer radial envelope of a relativistic charged particle beam. Additionally, we also outline a benchmark test of a relativistic beam propagating through a vacuum and pressurized gas cell, and present the results between EMPIRE and the hybrid code GAZEL. Further, we discuss the subtle errors that were caught with these problems and detail lessons learned.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A Viscoelastic Hybrid Shell Finite Element

An elastic large displacement thick-shell hybrid finite element is modified to allow for the calculation of viscoelastic stresses. Internal strain variables are introduced at he element's stress nodes and are employed to construct a viscous material model. First order ordinary differential equations relate the internal strain variables to the corresponding elastic strains at the stress nodes. The viscous stresses are computed from the internal strain variables using viscous moduli which are a fraction of the elastic moduli. The energy dissipated by the action of the viscous stresses in included in the mixed variational functional. Nonlinear quasi-static viscous equilibrium equations are then obtained. Previously developed Taylor expansions of the equilibrium equations are modified to include the viscous terms. A predictor-corrector time marching solution algorithm is employed to solve the algebraic-differential equations. The viscous shell element is employed to numerically simulate a stair-step loading and unloading of an aircraft tire in contact with a frictionless surface.

Johnson, Arthur↗

SU-E-CAMPUS-I-02: Estimation of the Dosimetric Error Caused by the Voxelization of Hybrid Computational Phantoms Using Triangle Mesh-Based Monte Carlo Transport

Purpose: Computational voxel phantom provides realistic anatomy but the voxel structure may result in dosimetric error compared to real anatomy composed of perfect surface. We analyzed the dosimetric error caused from the voxel structure in hybrid computational phantoms by comparing the voxel-based doses at different resolutions with triangle mesh-based doses. Methods: We incorporated the existing adult male UF/NCI hybrid phantom in mesh format into a Monte Carlo transport code, penMesh that supports triangle meshes. We calculated energy deposition to selected organs of interest for parallel photon beams with three mono energies (0.1, 1, and 10 MeV) in antero-posterior geometry. We also calculated organ energy deposition using three voxel phantoms with different voxel resolutions (1, 5, and 10 mm) using MCNPX2.7. Results: Comparison of organ energy deposition between the two methods showed that agreement overall improved for higher voxel resolution, but for many organs the differences were small. Difference in the energy deposition for 1 MeV, for example, decreased from 11.5% to 1.7% in muscle but only from 0.6% to 0.3% in liver as voxel resolution increased from 10 mm to 1 mm. The differences were smaller at higher energies. The number of photon histories processed per second in voxels were 6.4×10{sup 4}, 3.3×10{sup 4}, and 1.3×10{sup 4}, for 10, 5, and 1 mm resolutions at 10 MeV, respectively, while meshes ran at 4.0×10{sup 4} histories/sec. Conclusion: The combination of hybrid mesh phantom and penMesh was proved to be accurate and of similar speed compared to the voxel phantom and MCNPX. The lowest voxel resolution caused a maximum dosimetric error of 12.6% at 0.1 MeV and 6.8% at 10 MeV but the error was insignificant in some organs. We will apply the tool to calculate dose to very thin layer tissues (e.g., radiosensitive layer in gastro intestines) which cannot be modeled by voxel phantoms.

60 APPLIED LIFE SCIENCES↗

Effects of Ordering Strategies and Programming Paradigms on Sparse Matrix Computations

The Conjugate Gradient (CG) algorithm is perhaps the best-known iterative technique to solve sparse linear systems that are symmetric and positive definite. For systems that are ill-conditioned, it is often necessary to use a preconditioning technique. In this paper, we investigate the effects of various ordering and partitioning strategies on the performance of parallel CG and ILU(O) preconditioned CG (PCG) using different programming paradigms and architectures. Results show that for this class of applications: ordering significantly improves overall performance on both distributed and distributed shared-memory systems, that cache reuse may be more important than reducing communication, that it is possible to achieve message-passing performance using shared-memory constructs through careful data ordering and distribution, and that a hybrid MPI+OpenMP paradigm increases programming complexity with little performance gains. A implementation of CG on the Cray MTA does not require special ordering or partitioning to obtain high efficiency and scalability, giving it a distinct advantage for adaptive applications; however, it shows limited scalability for PCG due to a lack of thread level parallelism.

Oliker, Leonid↗

Data-Driven Nonlinear Subspace Modeling for Prediction and Control of Molten Iron Quality Indices in Blast Furnace Ironmaking

Blast furnace (BF) in ironmaking is a nonlinear dynamic process with complicated physical-chemical reactions, where multi-phase and multi-field coupling and large time delay occur during its operation. In BF operation, the molten iron temperature (MIT) as well as Si, P and S contents of molten iron are the most essential molten iron quality (MIQ) indices, whose measurement, modeling and control have always been important issues in metallurgic engineering and automation field. This paper develops a novel data-driven nonlinear state space modeling for the prediction and control of multivariate MIQ indices by integrating hybrid modeling and control techniques. First, to improve modeling efficiency, a data-driven hybrid method combining canonical correlation analysis and correlation analysis is proposed to identify the most influential controllable variables as the modeling inputs from multitudinous factors would affect the MIQ indices. Then, a Hammerstein model for the prediction of MIQ indices is established using the LS-SVM based nonlinear subspace identification method. Such a model is further simplified by using piecewise cubic Hermite interpolating polynomial method to fit the complex nonlinear kernel function. Compared to the original Hammerstein model, this simplified model can not only significantly reduce the computational complexity, but also has almost the same reliability and accuracy for a stable prediction of MIQ indices. Last, in order to verify the practicability of the developed model, it is applied in designing a genetic algorithm based nonlinear predictive controller for multivariate MIQ indices by directly taking the established model as a predictor. Industrial experiments show the advantages and effectiveness of the proposed approach.

Blast furnace (BF)↗

State-space self-tuner for on-line adaptive control

Dynamic systems, such as flight vehicles, satellites and space stations, operating in real environments, constantly face parameter and/or structural variations owing to nonlinear behavior of actuators, failure of sensors, changes in operating conditions, disturbances acting on the system, etc. In the past three decades, adaptive control has been shown to be effective in dealing with dynamic systems in the presence of parameter uncertainties, structural perturbations, random disturbances and environmental variations. Among the existing adaptive control methodologies, the state-space self-tuning control methods, initially proposed by us, are shown to be effective in designing advanced adaptive controllers for multivariable systems. In our approaches, we have embedded the standard Kalman state-estimation algorithm into an online parameter estimation algorithm. Thus, the advanced state-feedback controllers can be easily established for digital adaptive control of continuous-time stochastic multivariable systems. A state-space self-tuner for a general multivariable stochastic system has been developed and successfully applied to the space station for on-line adaptive control. Also, a technique for multistage design of an optimal momentum management controller for the space station has been developed and reported in. Moreover, we have successfully developed various digital redesign techniques which can convert a continuous-time controller to an equivalent digital controller. As a result, the expensive and unreliable continuous-time controller can be implemented using low-cost and high performance microprocessors. Recently, we have developed a new hybrid state-space self tuner using a new dual-rate sampling scheme for on-line adaptive control of continuous-time uncertain systems.

Shieh, L. S.↗

Shuttle Payload Ground Command and Control: An Experiment Implementation Combustion Module-2 Software Development, STS-107

This presentation covers the design of a command and control architecture developed by the author for the Combustion Module-2 microgravity experiment, which flew aboard the STS-107 Shuttle mission, The design was implemented to satisfy a hybrid network that utilized TCP/IP for both the onboard segment and ground segment, with an intermediary unreliable transport for the space to ground segment. With the infusion of Internet networking technologies into Space Shuttle, Space Station, and spacecraft avionics systems, comes the need for robust methodologies for ground command and control. Considerations of high bit error links, and unreliable transport over intermittent links must be considered in such systems. Internet protocols applied to these systems, coupled with the appropriate application layer protections, can provide adequate communication architectures for command and control. However, there are inherent limitations and additional complexities added by the use of Internet protocols that must be considered during the design. This presentation will discuss the rationale for the: framework and protocol algorithms developed by the author. A summary of design considerations, implantation issues, and learned lessons will be will be presented. A summary of mission results using this communications architecture will be presented. Additionally, areas of further needed investigation will be identified.

Carek, David Andrew↗

Statistical shape model-based reconstruction of a scaled, patient-specific surface model of the pelvis from a single standard AP x-ray radiograph

Purpose: The aim of this article is to investigate the feasibility of using a statistical shape model (SSM)-based reconstruction technique to derive a scaled, patient-specific surface model of the pelvis from a single standard anteroposterior (AP) x-ray radiograph and the feasibility of estimating the scale of the reconstructed surface model by performing a surface-based 3D/3D matching. Methods: Data sets of 14 pelvises (one plastic bone, 12 cadavers, and one patient) were used to validate the single-image based reconstruction technique. This reconstruction technique is based on a hybrid 2D/3D deformable registration process combining a landmark-to-ray registration with a SSM-based 2D/3D reconstruction. The landmark-to-ray registration was used to find an initial scale and an initial rigid transformation between the x-ray image and the SSM. The estimated scale and rigid transformation were used to initialize the SSM-based 2D/3D reconstruction. The optimal reconstruction was then achieved in three stages by iteratively matching the projections of the apparent contours extracted from a 3D model derived from the SSM to the image contours extracted from the x-ray radiograph: Iterative affine registration, statistical instantiation, and iterative regularized shape deformation. The image contours are first detected by using a semiautomatic segmentation tool based on the Livewire algorithm and then approximated by a set of sparse dominant points that are adaptively sampled from the detected contours. The unknown scales of the reconstructed models were estimated by performing a surface-based 3D/3D matching between the reconstructed models and the associated ground truth models that were derived from a CT-based reconstruction method. Such a matching also allowed for computing the errors between the reconstructed models and the associated ground truth models. Results: The technique could reconstruct the surface models of all 14 pelvises directly from the landmark-based initialization. Depending on the surface-based matching techniques, the reconstruction errors were slightly different. When a surface-based iterative affine registration was used, an average reconstruction error of 1.6 mm was observed. This error was increased to 1.9 mm, when a surface-based iterative scaled rigid registration was used. Conclusions: It is feasible to reconstruct a scaled, patient-specific surface model of the pelvis from single standard AP x-ray radiograph using the present approach. The unknown scale of the reconstructed model can be estimated by performing a surface-based 3D/3D matching.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Tracking Performance of Upgraded "Polished Panel" Optical Receiver on NASA's 34 Meter Research Antenna

There has been considerable interest in developing and demonstrating a hybrid "polished panel" optical receiver concept that would replace the microwave panels on the Deep Space Network's (DSN) 34 meter antennas with highly polished aluminum panels, thus enabling simultaneous opticaland microwave reception. A test setup has been installed on the 34 meter research antenna at DSS-13 (Deep Space Station 13) at NASA's Goldstone Deep Space Communications Complex in California in order to assess the feasibility of this concept. Here we describe the results of a recent effort todramatically reduce the dimensions of the point-spread function (PSF) generated by a custom polished panel, thus enabling improved optical communications performance. The latest results are compared to the previous configuration in terms of quantifiable PSF improvement. In addition, the performance of acquisition and tracking algorithms designed specifically for the polished panel PSF are evaluated and compared, based on data obtained from real-time tracking of planets and bright stars with the 34 meter research antenna at DSS-13.

optima↗

Aero-Engines AI - A Machine-Learning App for Aircraft Engine Concepts Assessment

Effective deployment of machine-learning (ML) models could drive a high level of efficiency in aircraft engine conceptual design. Aero-Engines AI is a user-friendly app that has been created to deploy trained machine-learning (ML) models to assess aircraft engine concepts. It was created using tkinter, a GUI (graphical user interface) module that is built into the standard Python library. Employing tkinter greatly facilitates the sharing of ML application as an executable file which can be run on Windows machines (without the need to have Python or any library installed). The app gets user input for a turbofan design, preprocesses the input data, and deploys trained ML models to predict turbofan thrust specific fuel consumption (TSFC), engine weight, core size, and turbomachinery stage-counts. The ML predictive models were built by employing supervised deep-learning and K-nearest neighbor regression algorithms to study patterns in an existing open-source database of production and research turbofan engines. They were trained, cross-validated, and tested in Keras, an open-source neural networks API (application programming interface) written in Python, with TensorFlow (Google open-source artificial intelligence library) serving as the backend engine. The smooth deployment of these ML models using the app shows that Aero-Engines AI is an easy-touse and a time-saving tool for aircraft engine design-space exploration during the conceptual design stage. Current version of the app focuses on the performance prediction of conventional turbofans. However, the scope of the app can easily be expanded to include other engine types (such as turboshaft and hybrid-electric systems) after their ML models are developed. Overall, the use of a machine-learning app for aircraft engine concept assessment represents a promising area of development in aircraft engine conceptual design.

machine learning↗

A Hybrid Reinforcement Learning-MPC Approach for Distribution System Critical Load Restoration: Preprint

This paper proposes a hybrid control approach for distribution system critical load restoration, combining deep reinforcement learning (RL) and model predictive control (MPC) aiming at maximizing total restored load following an extreme event. RL determines a policy for quantifying operating reserve requirements, thereby hedging against uncertainty, while MPC models grid operations incorporating RL policy actions, i.e., the reserve requirement, renewable (wind and solar) power predictions, and load demand forecasts. We formulate the reserve requirement determination problem as a sequential decision making problem based on the Markov Decision Process (MDP) and design an RL learning environment based on the OpenAI Gym framework and MPC. The RL agent reward and MPC objective function aim to maximize and monotonically increase total restored load and minimize load shedding and renewable power curtailment. The RL algorithm is trained off-line using historical forecast of renewable generation and load demand. The method is tested using a modified IEEE 13-bus distribution test feeder containing wind turbine, photovoltaic, microturbine and battery. Case studies demonstrated that the proposed method outperforms other operating reserve determination methods.

distribution system↗

Nano-enhanced solid-state hydrogen storage: Balancing discovery and pragmatism for future energy solutions

Nanomaterials have revolutionized the battery industry by enhancing energy storage capacities and charging speeds, and their application in hydrogen (H 2 ) storage likewise holds strong potential, though with distinct challenges and mechanisms. H 2 is a crucial future zero-carbon energy vector given its high gravimetric energy density, which far exceeds that of liquid hydrocarbons. However, its low volumetric energy density in gaseous form currently requires storage under high pressure or at low temperature. This review critically examines the current and prospective landscapes of solid-state H 2 storage technologies, with a focus on pragmatic integration of advanced materials such as metal-organic frameworks (MOFs), magnesium-based hybrids, and novel sorbents into future energy networks. These materials, enhanced by nanotechnology, could significantly improve the efficiency and capacity of H 2 storage systems by optimizing H 2 adsorption at the nanoscale and improving the kinetics of H 2 uptake and release. We discuss various H 2 storage mechanisms—physisorption, chemisorption, and the Kubas interaction—analyzing their impact on the energy efficiency and scalability of storage solutions. The review also addresses the potential of “smart MOFs”, single-atom catalyst-doped metal hydrides, MXenes and entropy-driven alloys to enhance the performance and broaden the application range of H 2 storage systems, stressing the need for innovative materials and system integration to satisfy future energy demands. High-throughput screening, combined with machine learning algorithms, is noted as a promising approach to identify patterns and predict the behavior of novel materials under various conditions, significantly reducing the time and cost associated with experimental trials. In closing, we discuss the increasing involvement of various companies in solid-state H 2 storage, particularly in prototype vehicles, from a techno-economic perspective. In conclusion, this forward-looking perspective underscores the necessity for ongoing material innovation and system optimization to meet the stringent energy demands and ambitious sustainability targets increasingly in demand.

25 ENERGY STORAGE↗

Aero-Engines AI - A Machine-Learning App for Aircraft Engine Concepts Assessment

Effective deployment of machine-learning (ML) models could drive a high level of efficiency in aircraft engine conceptual design. Aero-Engines AI is a user-friendly app that has been created to deploy trained machine-learning (ML) models to assess aircraft engine concepts. It was created using tkinter, a GUI (graphical user interface) module that is built into the standard Python library. Employing tkinter greatly facilitates the sharing of ML application as an executable file which can be run on Windows machines (without the need to have Python or any library installed). The app gets user input for a turbofan design, preprocesses the input data, and deploys trained ML models to predict turbofan thrust specific fuel consumption (TSFC), engine weight, core size, and turbomachinery stage-counts. The ML predictive models were built by employing supervised deep-learning and K-nearest neighbor regression algorithms to study patterns in an existing open-source database of production and research turbofan engines. They were trained, cross-validated, and tested in Keras, an open-source neural networks API (application programming interface) written in Python, with TensorFlow (Google open-source artificial intelligence library) serving as the backend engine. The smooth deployment of these ML models using the app shows that Aero-Engines AI is an easy-touse and a time-saving tool for aircraft engine design-space exploration during the conceptual design stage. Current version of the app focuses on the performance prediction of conventional turbofans. However, the scope of the app can easily be easily expanded to include other engine types (such as turboshaft and hybrid-electric systems) after their ML models are developed. Overall, the use of a machine-learning app for aircraft engine concept assessment represents a promising area of development in aircraft engine conceptual design.

machine learning↗

A Hybrid Reinforcement Learning-MPC Approach for Distribution System Critical Load Restoration

This paper proposes a hybrid control approach for distribution system critical load restoration, combining deep reinforcement learning (RL) and model predictive control (MPC) aiming at maximizing total restored load following an extreme event. RL determines a policy for quantifying operating reserve requirements, thereby hedging against uncertainty, while MPC models grid operations incorporating RL policy actions (i.e., reserve requirements), renewable (wind and solar) power predictions, and load demand forecasts. We formulate the reserve requirement determination problem as a sequential decision-making problem based on the Markov Decision Process (MDP) and design an RL learning environment based on the OpenAI Gym framework and MPC simulation. The RL agent reward and MPC objective function aim to maximize and monotonically increase total restored load and minimize load shedding and renewable power curtailment. The RL algorithm is trained offline using a historical forecast of renewable generation and load demand. The method is tested using a modified IEEE 13-bus distribution test feeder containing wind turbine, photovoltaic, microturbine, and battery. Case studies demonstrated that the proposed method outperforms other policies with static operating reserves.

distribution system↗

Proactive Intrusion Detection and Mitigation System

SAND2023-05661O The proactive intrusion detection and mitigation system (PIDMS) provides grid-edge situational awareness for cybersecurity defense by capturing real-time distributed energy resource (DER) network traffic and performance data with a novel approach that improves the detection and prevention of cyber-physical attacks. The PIDMS addresses the grid-edge security gap with real-time analysis of both network traffic and photovoltaic performance data to deliver a novel, cyber-physical intrusion detection system (IDS) approach that increases the accuracy and effectiveness of detection and mitigation. This hybrid IDS analysis enables dual monitoring that increases the workload of the adversary; both cyber and physical data would have to be simultaneously spoofed to evade detection. Furthermore, monitoring and analyzing cyber data are insufficient in some cases. For example, in an insider threat aimed at disrupting inverter grid-support functions where proper credentials and authentication are achieved, only the altered PV performance would indicate abnormal behavior. All in all, the PIDMS provides novel capabilities for: • Distributed, real-time cyber-physical detection and mitigation analysis • Cybersecurity defense for grid-edge systems • Analysis framework that can provide situational awareness across the transmission, distribution, and DER systems The PIDMS sensor is designed to collect cyber-physical data, process the data using machine-learning algorithms, detect abnormal events, and deploy mitigations. With these goals, the main functional PIDMS objectives are: • Capability to collect cyber-physical data • Onboard storage of cyber-physical data • Peer-to-peer communication • Computationally efficient machine-learning algorithms • Online cyber-physical data analysis • Alerting/visualization capabilities • Mitigation deployment capability with bump-in-the-wire (BITW) implementation Each of these functional objectives enable PIDMS to perform effective cyber-physical intrusion detection and mitigation. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Jones, Christian↗

O(N) Quantum Molecular Dynamics (QMD) Algorithms, Code Requirements, Release Notes, and Library Deployment

The Exascale Computing Project (ECP) Co-design Center for Particle Applications (CoPA) is addressing the challenges for particle applications to run on upcoming exascale computing architectures. This includes short- and long-range molecular dynamics (MD), particle-in-cell (PIC), and quantum molecular dynamics (QMD). This document addresses $\mathcal{O}(N)$ QMD. The goal of this CoPA team is to develop $\mathcal{O}(N)$ linear scaling methods for QMD optimized for new and next generation architectures. This will provide the capability to simulate larger systems and run for longer timeframes than is possible today. Focus is initially on the density matrix construction. This requires efficient matrix formats and linear algebra operations for distributed architectures. A significant reduction in the wallclock time for an MD timestep is necessary to achieve practical large scale QMD. Equally challenging is to develop scalable distributed implementations. This requires the exploration of efficient data formats (dense, sparse, etc.), data decomposition and distribution across compute nodes using block- and graph-based methods. Traditional programming models such as MPI/OpenMP, hybrid MPI/OpenMP/Accelerator, as well as asynchronous task-based models will be explored to achieve maximum parallelism. Parallelism at the level of a matrix operation, solver, and/or application needs to be explored to gain the most benefit on new and emerging architectures. This report serves as a “living document” of the additions and enhancements throughout the ECP CoPA project for $\mathcal{O}(N)$ QMD. The following sections address the applications we are working with, the quantum chemistry libraries (BML and PROGRESS), and our co-design exploration through the use of a proxy application.

36 MATERIALS SCIENCE↗

Automated Generation of Integrated Digital and Spiking Neuromorphic Machine Learning Accelerators

The growing numbers of application areas for artificial intelligence (AI) methods have led to an explosion of domain-specific accelerators that could support every new machine learning (ML) algorithm advancement, clearly highlighting the need for a capability to quickly and automatically transition from algorithm definition to hardware implementation and explore design space along a variety of SWaP (size, weight and Power). The software defined architectures (SODA) synthesizer implements a compiler-based modular infrastructure for the end-to-end generation of machine learning accelerators from high-level frameworks to hardware description language. At the same time, neuromorphic computing, by mimicking how the brain operates, promises to perform artificial intelligence tasks at efficiencies orders of magnitude higher than the current conventional tensor-processing based accelerators, as demonstrated by a variety of specialized designs leveraging Spiking Neural Networks (SNNs). Nevertheless, the mapping of an artificial neural network (ANN) to solutions supporting SNNs is still a non-trivial and very device-specific task, and completely lack the possibility to design hybrid systems that integrate conventional and spiking neural models. In this paper we discuss the support for such an integrated generation leveraging the SODA Synthesizer framework and its modular structure. In particular, we present a new MLIR dialect (part of the SODA frontend) that allows expressing spiking neural network features (e.g., available resources, spiking sequences, analog signal reading, etc.) and illustrate how it enables mapping to Spiking Neurons and deployment to the related specialized hardware (which, in the digital domain, could be generated through the other existing layers of the SODA Synthesizer). We then discuss the opportunities for even deeper integration afforded by the hardware compilation infrastructure, providing a path towards the generation of complex heterogeneous artificial intelligence systems.

Curzel, Serena↗