Davoud Soltani Sehat

Mr. Davoud Soltani Sehat

orcid

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 February 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
Davoud Soltani SehatDavoud Soltani Sehat
Davoud Soltani Sehat
Hydrogen is a versatile energy carrier essential for decarbonizing hard-to-abate sectors and long-duration storage. This study presents a unified techno-economic comparison of major production pathways—grey/blue steam methane reforming, biomass gasification, thermochemical cycles, biological methods, and solar-powered electrolysis—using 2025 benchmarks. Focus is on a 100 kW off-grid PV-electrolyzer system with realistic assumptions (PV performance ratio 0.85, electrolyzer efficiency 70% LHV). In Iran's high-insolation regions (PSH ≥ 5.15 kWh/kWp/day), annual yields reach 3.2–3.4 tonnes H₂—55–60% higher than northern Europe—with round-trip efficiency of 23.8%. Solar electrolysis offers zero direct emissions and 51–55 kWh/kg H₂ consumption. Scaling to multi-MW coastal hybrids with renewable desalination projects LCOH of 3.0–4.0 USD/kg by 2030, positioning Iran as a competitive exporter. A reproducible model and phased roadmap provide actionable insights.

17 February 2026 in TechRxiv
Verified
Preprint
Open access
Davoud Soltani SehatDavoud Soltani Sehat
Davoud Soltani Sehat
SSSina Soltani Sehat
Sina Soltani Sehat
This paper proposes a novel hybrid intelligent control framework integrating a fuzzy supervisory Proportional-Integral-Derivative (PID) controller with Soft Actor-Critic (SAC) reinforcement learning for pitch angle regulation in variable-speed wind turbines. The fuzzy supervisory layer employs seven triangular membership functions for both rotor speed error and its rate of change, resulting in a comprehensive 49-rule inference base that dynamically schedules PID gains to address aerodynamic nonlinearities. The SAC algorithm-operating as an off-policy actor-critic reinforcement learner with entropy maximization-performs online meta-tuning of the fuzzy-scaled gains, enabling continuous adaptation to turbulent wind conditions, parameter uncertainties, and unmodeled disturbances. Practical implementation challenges are mitigated through anti-windup back-calculation to prevent integral saturation and low-pass filtering to suppress derivative kick and measurement noise. Stability of the hybrid closed-loop system is analyzed using Lyapunov methods adapted for reinforcement learning frameworks, ensuring boundedness and convergence. Extensive simulations in MATLAB/Simulink on a 2 MW doubly-fed induction generator (DFIG) wind turbine under realistic IEC-compliant turbulent wind profiles demonstrate superior performance: up to 25% improvement in annual energy production (AEP), 22% reduction in fatigue equivalent loads, and markedly smoother rotor speed tracking compared to conventional PID, standalone fuzzy PID, and baseline actor-critic methods. This work extends the authors' prior SAC-based meta-tuning framework from robotic manipulator trajectory tracking [1] to renewable energy systems, offering a high-performance, learning-enabled solution for modern wind farms. achieving inference times under 5 ms on embedded platforms such as NVIDIA Jetson Nano with TensorRT optimization

6 February 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
Davoud Soltani SehatDavoud Soltani Sehat
Davoud Soltani Sehat
SSSina Soltani Sehat
Sina Soltani Sehat
This duplicate preprint is being withdrawn as multiple subsequent versions have been posted. Please refer to Version 3 for the most recent and complete version of the manuscript.

19 February 2026 in TechRxiv
Verified
Preprint
Open access
Davoud Soltani SehatDavoud Soltani Sehat
Davoud Soltani Sehat
This research introduces HAC-SAC², a novel hyper-adaptive chaos-aware controller based on twin reinforcement learning, for 100 kW dynamic wireless power transfer (DWPT) in electric vehicles. Fully implemented on the industrial Siemens S7-1516F-3 PN/DP PLC, the controller features three coordinated layers: (1) a dynamic Hyper Fuzzy-PID with 49 adaptive rules and real-time coefficient tuning, (2) a Twin SAC² reinforcement learning engine with parallel 5-layer neural networks and dual-entropy regularization for robust optimization in uncertain environments, and (3) a chaos-aware switcher driven by a novel Instantaneous Power Chaos Index for sub-2ms decision-making. An asymmetric compensation topology is proposed: LCC-S on the transmitter (roadside) side and LCL-P on the receiver (vehicle) side, operating at 87 kHz in alignment with emerging SAE J2954 guidelines for high-power dynamic charging. Initial simulations were performed in integrated MATLAB/Simulink 2025b and PLECS 4.9 using the virtual Iran-1408 highway model, followed by C++ implementation via ODK 1500S on the PLC and rigorous Hardware-in-the-Loop (HiL) validation. Key results demonstrate power transfer efficiency of 93.8–96.7% across all misalignment conditions (±35 cm), secondary current ripple below 3.9%, control settling time under 14ms, PLC cycle time up to 5.8ms, program memory usage at 68%, data at 42%, and full system stability at 140 km/h. Global asymptotic stability of the closed-loop system is formally proven using the Lyapunov–LaSalle theorem for the first time in this context. This work represents a pioneering deployment of advanced reinforcement learning-based control on an S7-1500 series industrial PLC for EV dynamic charging, offering a scalable pathway for national highway infrastructure integration by 2029 (1408 SH). Due to severe financial constraints as an independent researcher without institutional or grant funding, full-scale physical prototyping was not feasible; results are based on high-fidelity simulation and preliminary Hardware-in-the-Loop validation.