http://lutpub.lut.fi:80/handle/10024/158303 Wed, 19 Aug 2026 23:05:09 GMT 2026-08-19T23:05:09Z http://lutpub.lut.fi:80/handle/10024/172773 Synchronization dynamics between rider and saddle motion across gaits revealed by IMU analysis Louis, Archana Treesa; Treesa Louis, Anjana; Mikkola, Aki Background: Effective rider–horse coordination is essential for performance and welfare, yet objective, field-based quantification of rider–saddle synchronization across gaits remains limited. Aims/objectives: To quantify dynamic synchronization between rider motion and saddle motion (as a proxy for horse trunk oscillations) across walk, trot, and canter using a minimal inertial measurement unit (IMU) setup, and to determine whether synchronization differs between rider body segments. Methods: Two experienced riders (8–20 years riding experience) rode their own or familiar warmblood horses (n = 2; age 6–14 years) at walk, trot, and canter under collected, medium, and extended riding frame (collection level) conditions. Four wireless inertial measurement units (pelvis, trunk, head, saddle) recorded tri-axial acceleration at 60 Hz. Vertical acceleration of rider segments was analyzed relative to saddle motion. Synchronization was quantified using cross-correlation coefficients, phase lag, and phase-locking values (PLV). Mean values were calculated across repeated trials. Results: Synchronization strength increased with gait speed. Mean cross-correlation coefficients between pelvis and saddle were 0.64 at walk, 0.92 at trot, and 0.96 at canter, with corresponding mean phase lags of 22 ms, 0 ms, and 0 ms, respectively. Trunk and head segments showed lower synchronization at walk (r = 0.62–0.65) but high coupling at trot (r = 0.88–0.93) and canter (r = 0.91–0.95). Mean PLV values increased from 0.52 to 0.55 at walk to 0.91–0.95 at trot and 0.92–0.94 at canter. Riding frame condition influenced synchronization primarily at walk, with lower mean correlation and PLV in the extended frame. Conclusion: Within this population, rider–saddle synchronization increased with gait speed and was strongest at the pelvis, supporting its role as the primary mechanical interface transmitting horse trunk motion to the rider. A minimal IMU-based approach captured repeatable, segment-specific synchronization patterns under field conditions. Mon, 18 May 2026 00:00:00 GMT http://lutpub.lut.fi:80/handle/10024/172773 2026-05-18T00:00:00Z http://lutpub.lut.fi:80/handle/10024/172772 Comparative life cycle assessment of second- and third-life repurposing versus recycling of electric vehicle batteries Hayati Soloot, Hesam Edin; Hupponen, Mari; Horttanainen, Mika Rapid electric vehicle uptake has elevated the importance of end-of-life (EoL) management of electric vehicle batteries (EVBs). Due to the high environmental burden of EVB production, the focus should shift from EVB production to EoL management over time. Existing life cycle assessment (LCA) studies on EoL management of EVBs typically treat repurposing and preparation for repurposing as a single generic unit process, limit repurposing to one second-life application, and adopt “ready-to-repurpose” assumptions that overlook inhomogeneity among EVB packs and the performance variability of lithium-ion battery (LIB) modules. These simplifications introduce substantial uncertainty into environmental results and may compromise the technical reliability of repurposed EVBs. To address these gaps, this study develops a system-wide Preparation for Repurposing (P4REP) approach that integrates quality screening, performance compatibility, and module-level disassembly within a consequential LCA of EoL strategies for NMC 622 EVBs. Three scenarios—hydrometallurgical recycling, second-life repurposing in an energy storage system (ESS), and combined second- and third-life repurposing in ESS and uninterruptible power supply (UPS) units—are compared using the basket-of-products/functions approach within an expanded multifunctional system boundary. Sensitivity analyses cover the LIB module conversion rate (80–100%) and the 2023–2050 marginal energy transition. The results indicate that any scenario that applies the P4REP approach for at least one additional life has a consistently lower environmental impact than the recycling scenario. Moreover, the scenario combined second- and third-life repurposing reduced the global warming potential relative to the recycling scenario from 13.85 kg CO2 eq.kWh−1 to −6.83 kg CO2 eq.kWh−1. This lower total GWP impact is mainly associated with the reduced contribution of newly manufactured ESS and UPS provision within the expanded multifunctional system boundary, and is conditional on the basket-of-functions construction, a relatively high LIB modules conversion rate, and the projected 2050 European marginal electricity and heat mixes. Mon, 17 Aug 2026 00:00:00 GMT http://lutpub.lut.fi:80/handle/10024/172772 2026-08-17T00:00:00Z http://lutpub.lut.fi:80/handle/10024/172761 Topology optimization of liquid-cooled heat exchangers for data centers: Methods, tools and evaluation Li, Zhendong; Wang, Yu; Jamil, Kainaat; Yuan, Xiaolei With the continuous rise in energy consumption within data centers, optimizing liquid-cooled heat exchangers has become crucial for effective thermal management. Topology optimization has emerged as a groundbreaking design method that transcends the limitations of traditional design performances. This review systematically examines recent advancements in the topology optimization of liquid-cooled heat exchangers, focusing on key methodological aspects, including design parameterization methods, numerical simulation tools, and optimization algorithms. Through comparative analysis, current research trends and key gaps are identified. Density-based methods remain dominant due to their high computational efficiency. However, these methods still suffer from blurred boundaries. These methods also rely heavily on post-processing techniques. In contrast, explicit methods can generate clear geometric boundaries. However, they still face many significant challenges. Furthermore, a persistent problem has been identified in the research. Most studies are still limited to single-objective, two-dimensional, and steady-state simulations. Whether these designs can be manufactured and fulfill their intended performance requirements is another concern. Experimental validation for these designs is limited. This has led to significant challenges in translating research findings into manufacturable, high-performance designs applicable to real-world data centers. However, recent years have brought new perspectives, driven by upgrades in computer hardware, enhanced computational capabilities, and the integration of new technologies. This review provides a structured framework to help bridge the gap between research and application, with the ultimate goal of guiding the field toward more efficient, compact, and sustainable heat exchanger solutions. Thu, 06 Aug 2026 00:00:00 GMT http://lutpub.lut.fi:80/handle/10024/172761 2026-08-06T00:00:00Z http://lutpub.lut.fi:80/handle/10024/172734 Edge AI System for EV Trip Analytics: On-Device WLTP Class 3B Processing and Battery Temperature Prediction Hasan, Nabeel; Narayanan, Arun; Nardelli, Pedro EV analytics are currently reliant on cloud computing; this introduces latency and does not allow drivers to receive real-time feedback. Current systems can hardly estimate interpretable 1 Hz measurements or do short-term thermal predictions on small embedded environments. This study aims to demonstrate that these tasks can be completely performed on-device. In this study, we describe a low-cost edge platform, based on transforming EV telemetry into real-time analytics on an ESP32-S3 microcontroller. The device is capable of calculating energy consumption, coasting performance, regenerative efficiency, battery-flow efficiency, and pack temperature at 1 Hz. It also executes a TinyML model prediction algorithm of the battery temperature 120 s in advance. To our best knowledge, this is one of the first demonstration of 1 Hz thermal forecasting based on a 1 Hz TinyML model running on the resource-constrained microcontroller. It provides a fully edge native system, and sends reduced size summaries over MQTT to a Node-RED dashboard. WLTP Class 3B cycle experiments indicate that it has stable 1 Hz throughput (1801/1801 messages), fast inference (6 ms), low memory consumption (1 MB flash), and highly accurate thermal prediction (MAE of 0.09 C). These findings demonstrate that real-time EV analytics and compact horizon thermal prediction is possible without cloud computing, which provides a viable alternative to standard EV monitoring architectures. Thu, 02 Jul 2026 00:00:00 GMT http://lutpub.lut.fi:80/handle/10024/172734 2026-07-02T00:00:00Z