Alexander Schnerring

Concentrated solar thermal (CST) plants can provide dispatchable renewable energy but remain complex and costly to operate due to tightly coupled subsystems, dynamic environmental conditions, and high safety requirements. This work presents the Solar Energy Autopilot (SOLEA), a holistic four-layer architecture that transforms a traditional solar power tower plant into an autonomous cyber-physical system (CPS). The architecture integrates plant hardware and control systems, advanced optical and meteorological measurement systems, a CPS middleware platform with a standardized digital twin, and a multi-agent automation layer for operation planning and control. SOLEA is implemented and evaluated at an operational research facility, where intelligent agents autonomously plan operating days, start and shut down the plant, and regulate the receiver flux and outlet temperature under varying environmental conditions. A dual-sided evaluation quantifies benefits in terms of efficiency, robustness, and cost-related indicators and assesses the system’s autonomy using a multi-dimensional maturity model. The results show substantial reductions in operational and investment costs, highly robust operation despite challenging environmental conditions, and efficiency gains in plant operation enabled by an increased autonomy level of the facility. Overall, the study demonstrates that the layered modular SOLEA architecture can effectively bridge the gap between isolated automation solutions and plant-wide autonomous CST operation and provides a reusable blueprint for advancing autonomy in future CST plants and comparable industrial environments.

Quantum low-density parity-check (QLDPC) codes are promising candidates for error correction in quantum computers. One of the major challenges in implementing QLDPC codes in quantum computers is the lack of a universal decoder. In this work, we first propose to decode QLDPC codes with a belief propagation (BP) decoder operating on overcomplete check matrices. Then, we extend the neural BP (NBP) decoder, which was originally studied for suboptimal binary BP decoding of QLPDC codes, to quaternary BP decoders. Numerical simulation results demonstrate that both approaches as well as their combination yield a low-latency, high-performance decoder for several short to moderate length QLDPC codes.