{"id":{"repo_id":"clemson","oai_identifier":"oai:open.clemson.edu:all_theses-5729"},"canonical_url":"https://search.ndltd.org/etd/clemson/oai:open.clemson.edu:all_theses-5729","repository":{"repo_id":"clemson","name":"Clemson University","base_url":"https://open.clemson.edu/do/oai/"},"display":{"title":"Exploration Of Real-Time Power Electronic Simulation on AMD NPU","abstract":"The rapid electrification of the automotive and data center sectors has created a critical demand for high-fidelity, real-time simulation of complex power electronic systems. While Hardware- in-the-Loop (HIL) simulation on Field-Programmable Gate Arrays (FPGAs) is the current industry standard, it presents significant challenges regarding development complexity and memory limita- tions. This dissertation investigates the feasibility of utilizing Neural Processing Units (NPUs)— specifically the AMD AI Engine (AIE) XDNA2 architecture—as a novel platform for real-time, deterministic circuit simulation. This study establishes a comprehensive automated modeling framework based on graph theory and Massarini’s method to derive linear state-space representations for arbitrary circuit topologies, capable of handling ideal switching events and Dirac impulses. To evaluate the hardware’s capabilities, a Half-Bridge LLC Resonant Converter was simulated using two distinct memory management architectures: a Matrix Cached implementation and a Runtime Matrix Fetching implementation. Experimental results conducted on an AMD Ryzen AI 9 HX 370 processor demonstrate that the Matrix Cached implementation achieves a deterministic iteration time of approximately 0.5 µs, meeting the stringent real-time requirements for high-frequency power electronics. Conversely, the Runtime Matrix Fetching approach revealed significant bottlenecks due to driver latency and cache misses. The study concludes that while the AIE architecture is currently compute-bound due to emulated floating-point arithmetic, it offers a viable, flexible alternative to FPGAs for simulations that fit within localized memory.","abstract_html":"The rapid electrification of the automotive and data center sectors has created a critical demand for high-fidelity, real-time simulation of complex power electronic systems. While Hardware- in-the-Loop (HIL) simulation on Field-Programmable Gate Arrays (FPGAs) is the current industry standard, it presents significant challenges regarding development complexity and memory limita- tions. This dissertation investigates the feasibility of utilizing Neural Processing Units (NPUs)— specifically the AMD AI Engine (AIE) XDNA2 architecture—as a novel platform for real-time, deterministic circuit simulation. This study establishes a comprehensive automated modeling framework based on graph theory and Massarini’s method to derive linear state-space representations for arbitrary circuit topologies, capable of handling ideal switching events and Dirac impulses. To evaluate the hardware’s capabilities, a Half-Bridge LLC Resonant Converter was simulated using two distinct memory management architectures: a Matrix Cached implementation and a Runtime Matrix Fetching implementation. Experimental results conducted on an AMD Ryzen AI 9 HX 370 processor demonstrate that the Matrix Cached implementation achieves a deterministic iteration time of approximately 0.5 µs, meeting the stringent real-time requirements for high-frequency power electronics. Conversely, the Runtime Matrix Fetching approach revealed significant bottlenecks due to driver latency and cache misses. The study concludes that while the AIE architecture is currently compute-bound due to emulated floating-point arithmetic, it offers a viable, flexible alternative to FPGAs for simulations that fit within localized memory.","abstract_has_math":false,"creators":["Du, Shouyu"],"institution":null,"degree_name":"Master of Science in Engineering (MSE)","degree_level":"Thesis","degree_discipline":"Electrical and Computer Engineering (Holcomb Dept. of)","degree_department":null,"school":null,"contributors":["Tao Wei","Fatemeh Afghah","Jon Calhoun"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T07:34:14Z","subjects":["NPU","Hardware-in-the-Loop","Power Electronic Simulation","Computer and Systems Architecture","Digital Circuits","Hardware Systems"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://open.clemson.edu/all_theses/4790","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://open.clemson.edu/all_theses/4790","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.clemson.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Aopen.clemson.edu%3Aall_theses-5729","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Tao Wei","Fatemeh Afghah","Jon Calhoun"]},{"key":"dc:creator","label":"Author","values":["Du, Shouyu"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2026-03-30T07:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2026-05-01T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical and Computer Engineering (Holcomb Dept. of)"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science in Engineering (MSE)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["NPU","Hardware-in-the-Loop","Power Electronic Simulation","Computer and Systems Architecture","Digital Circuits","Hardware Systems"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://open.clemson.edu/all_theses/4790"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The rapid electrification of the automotive and data center sectors has created a critical demand for high-fidelity, real-time simulation of complex power electronic systems. While Hardware- in-the-Loop (HIL) simulation on Field-Programmable Gate Arrays (FPGAs) is the current industry standard, it presents significant challenges regarding development complexity and memory limita- tions. This dissertation investigates the feasibility of utilizing Neural Processing Units (NPUs)— specifically the AMD AI Engine (AIE) XDNA2 architecture—as a novel platform for real-time, deterministic circuit simulation. This study establishes a comprehensive automated modeling framework based on graph theory and Massarini’s method to derive linear state-space representations for arbitrary circuit topologies, capable of handling ideal switching events and Dirac impulses. To evaluate the hardware’s capabilities, a Half-Bridge LLC Resonant Converter was simulated using two distinct memory management architectures: a Matrix Cached implementation and a Runtime Matrix Fetching implementation. Experimental results conducted on an AMD Ryzen AI 9 HX 370 processor demonstrate that the Matrix Cached implementation achieves a deterministic iteration time of approximately 0.5 µs, meeting the stringent real-time requirements for high-frequency power electronics. Conversely, the Runtime Matrix Fetching approach revealed significant bottlenecks due to driver latency and cache misses. The study concludes that while the AIE architecture is currently compute-bound due to emulated floating-point arithmetic, it offers a viable, flexible alternative to FPGAs for simulations that fit within localized memory."]},{"key":"dc:title","label":"Title","values":["Exploration Of Real-Time Power Electronic Simulation on AMD NPU"]}]}],"canonical_facts":{"dc:contributor":["Tao Wei","Fatemeh Afghah","Jon Calhoun"],"dc:creator":["Du, Shouyu"],"dc:date.available":["2026-03-30T07:00:00Z"],"dc:date.created":["2026-05-01T07:00:00Z"],"dc:description.abstract":["The rapid electrification of the automotive and data center sectors has created a critical demand for high-fidelity, real-time simulation of complex power electronic systems. While Hardware- in-the-Loop (HIL) simulation on Field-Programmable Gate Arrays (FPGAs) is the current industry standard, it presents significant challenges regarding development complexity and memory limita- tions. This dissertation investigates the feasibility of utilizing Neural Processing Units (NPUs)— specifically the AMD AI Engine (AIE) XDNA2 architecture—as a novel platform for real-time, deterministic circuit simulation. This study establishes a comprehensive automated modeling framework based on graph theory and Massarini’s method to derive linear state-space representations for arbitrary circuit topologies, capable of handling ideal switching events and Dirac impulses. To evaluate the hardware’s capabilities, a Half-Bridge LLC Resonant Converter was simulated using two distinct memory management architectures: a Matrix Cached implementation and a Runtime Matrix Fetching implementation. Experimental results conducted on an AMD Ryzen AI 9 HX 370 processor demonstrate that the Matrix Cached implementation achieves a deterministic iteration time of approximately 0.5 µs, meeting the stringent real-time requirements for high-frequency power electronics. Conversely, the Runtime Matrix Fetching approach revealed significant bottlenecks due to driver latency and cache misses. The study concludes that while the AIE architecture is currently compute-bound due to emulated floating-point arithmetic, it offers a viable, flexible alternative to FPGAs for simulations that fit within localized memory."],"dc:identifier":["https://open.clemson.edu/all_theses/4790"],"dc:subject":["NPU","Hardware-in-the-Loop","Power Electronic Simulation","Computer and Systems Architecture","Digital Circuits","Hardware Systems"],"dc:title":["Exploration Of Real-Time Power Electronic Simulation on AMD NPU"],"thesis:degree_discipline":["Electrical and Computer Engineering (Holcomb Dept. of)"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Master of Science in Engineering (MSE)"]},"updated_at":"2026-11-19T07:34:14Z"}