ABB Group

Deep models deployed on edge devices frequently encounter resource variability, which arises from fluctuating energy levels, timing constraints, or prioritization of other critical tasks within the system. State-of-the-art machine learning pipelines generate resource-agnostic models, not capable to adapt at runtime. In this work we introduce Resource-Efficient Deep Subnetworks (REDS) to tackle model adaptation to variable resources. In contrast to the state-of-the-art, REDS use structured sparsity constructively by exploiting permutation invariance of neurons, which allows for hardware-specific optimizations. Specifically, REDS achieve computational efficiency by (1) skipping sequential computational blocks identified by a novel iterative knapsack optimizer, and (2) leveraging simple math to re-arrange the order of operations in REDS computational graph to take advantage of the data cache. REDS support conventional deep networks frequently deployed on the edge and provide computational benefits even for small and simple networks. We evaluate REDS on seven benchmark architectures trained on the Visual Wake Words, Google Speech Commands, Fashion-MNIST and CIFAR10 datasets, and test on four off-the-shelf mobile and embedded hardware platforms. We provide a theoretical result and empirical evidence for REDS outstanding performance in terms of submodels' test set accuracy, and demonstrate an adaptation time in response to dynamic resource constraints of under 40$\mu$s, utilizing a 2-layer fully-connected network on Arduino Nano 33 BLE.

Effective thermal management is one of the primary design bottlenecks for high-power transformers. While rectangular windings offer good fill factor and current capacity, a characterization of their thermal performance is limited. This paper presents a comparative analysis of edgewise and flatwise rectangular windings for transformers through FEM. Six 1-kW transformer designs with different conductor thickness, which results in "high-profile" and "low-profile" transformers, were analyzed in terms of thermal performance. The analysis revealed that the flatwise structure provides a parallel, simultaneous thermal path that is consistently superior, achieving lower peak temperatures and more uniform heat distribution across all geometries in both natural and forced-air cooling. The edgewise structure, conversely, presents a series thermal path that impedes heat dissipation from the inner turns, leading to localized hotspots that worsen as the profile becomes higher. A representative edgewise and flatwise transformer with a square window area was implemented in a 100 V:100 V, 1-kW CLLC bidirectional resonant converter at 28.5 kHz switching frequency for experimental comparison under natural and forced-air cooling conditions. The experimental results in both validated the thermal superiority of the flatwise structure, demonstrating lower peak winding temperatures and reduced thermal gradients between the winding and the outer surfaces of the transformer.

Grid design and battery integration for industrial microgrids pose significant challenges due to fluctuating loads and the intermittency of renewable energy sources. Preliminary frontend engineering and design (preFEED) enables cost-effective assessments based on high-level system modelling. This study introduces a techno-economic preFEED framework tailored to industrial microgrids, with a focus on battery energy storage systems (BESS) that balance energy shifting and grid stability under uncertainty. The framework emphasizes cost-optimal operation while incorporating grid-stability considerations, drawing insights from a comprehensive meta-study on microgrid design. We apply our framework to a mining use case, in which electrification by means of renewables and battery energy systems shall reduce the carbon footprint and contribute to more sustainable mining. Our results show that complex and costly FEED design can be streamlined through an interactive and user-friendly preFEED approach, enabling early-stage techno-economic evaluations. This methodology serves as a practical entry point for site managers and facilitates informed decision-making for a transition to low-carbon industrial site operation.

ABSTRACT Corrosive gas tests are commonly used to study dendritic corrosion growth in the isolation trenches of power semiconductor modules. However, conventional tests typically apply static DC bias conditions and therefore do not reproduce the time‐varying backplate‐referenced electric‐field components present in frequency converter operation, where the module backplate is commonly coupled to protective earth through the heat sink. In this paper, a square‐wave bias voltage was applied to the backplates of standard alumina substrate power modules during mixed flowing gas exposure to study the effect of a time‐varying electric field between the backplate and the top‐side conductors on dendrite growth. The dynamic backplate bias was found to modify dendrite morphology in DC‐biased trenches and to promote dendritic growth in trenches without an externally applied top‐side conductor DC bias. The results indicate that converter‐operation‐related backplate‐referenced field components can affect corrosion growth mechanisms and should be considered when developing field‐representative corrosive gas tests for power semiconductor modules.

Mine dewatering is a safety-critical and energy-intensive process in underground mining, commonly implemented as a multilevel chain of reservoirs and pumping stations. Many installations still rely on level-threshold, rule-based control that maintains safe reservoir bounds but does not exploit flexible electricity prices. This paper investigates price-aware control for a multilevel mine dewatering use case using a physics-based MATLAB/Simulink model parameterized from industrial specifications. We propose model predictive control (MPC) strategies designed to minimize electricity costs while ensuring reservoir safety constraints and adhering to discrete pump actuation limits. Our formulation models the non-linear behavior of pumps and on-off decisions as a mixed-integer non-linear program (MINLP). As a more computationally efficient alternative, we propose a linear approximation of the non-linear flow mapping, resulting in a linear programming (LP) based MPC variant. We also model and assess the use of variable-frequency drives with the pumps. The results indicate that price-aware optimization can reduce dewatering energy cost almost 30%.