This article demonstrates how to create a reliable Computational Fluid Dynamic (CFD) model of a 250 kW Internal Permanent Magnet Synchronous Motor (IPMSM), used in a dynamometer bench for testing electric motors, to address its overheating problems. The purpose is to investigate the rotor's high temperature levels and to propose mitigation strategies. Firstly, a literature review is conducted to highlight key difficulties and identify overlooked factors. Therefore, various novel aspects of the developed numerical model are outlined, such as an alternative way to treat rotating boundaries, how to model the presence of resins in the actual motor and the importance of mesh resolution in the air gap. Additionally, a novel procedure to accurately account for friction losses is presented. Following the model validation against experimental data, CFD predictions lead to the identification of the main cooling problem, related to air flow losses through vents in the air conveyor and in other components of the motor. Two solutions are then proposed, which allow for operating temperature reductions ranging from 25 K to 120 K.

LEDs development has attracted attention over conventional mercury lamps for the tiny size, high efficiency, long lifetime, low operating temperature. The antimicrobial effectiveness of traditional UV-lamps radiation (wavelength of 254 nm) compared to UV-C LEDs (LED1 wavelength range 275–286 nm and LED2 range 260–270 nm) was carried out, for possible applications to automated sterile drug compounding. The UV lamp and the tested UV-LED devices remarkably reduced microbial load, following a time-dose response, but the best performance was evidenced by LED1, which guaranteed the complete inactivation of high concentrations of bacteria, yeasts, and spores at doses between 200 and 2000 J/m2.

In this paper, the design and implementation of a DC/DC converter for automotive component testing with state-of-the art performance is described. The converter is the core of a battery emulator for the characterization and development of automotive batteries, electronic chargers, traction inverters, DC-DC converters, E-motors and E-axles. Cutting edge performance, flexibility and compactness are obtained by exploiting 1200-V SiC modules, high switching frequency, planar transformer technology, suitable topology solutions and fast digital control strategies. The implemented system is a liquid-cooled, bidirectional converter with galvanic isolation capable of 350 kW continuous output power, output voltage range 48–1000 V, continuous output current up to 800 A (1600 A peak), voltage/current ramp-up time below 10/2 ms and 0.1% current/voltage accuracy. The entire instrument is implemented in a standard full-height 19-inch rack cabinet.

This work aims to improve the quality control of gasoline direct injection pumps through the application of artificial intelligence on a dedicated test bench. First, the forecasting capability of the proposed neural architectures are evaluated by comparing the predictions with the experimental measurement of the flow rate delivered by the pump. The second part of the herein study is focused on evaluating the possibility of replacing a physical sensor (i.e torque meter), since, due to resonance phenomenon, its breakage has already led the partner company to extensive damage to the test bench. Among the tested structures, a Nonlinear Autoregressive Exogenous (NARX) architecture and a Long Short-Term Memory (LSTM), respectively, proved to be valid alternative to the physical sensors. At the range of interest, the neural structures showed, with respect to the experimental data, percentage errors always lower than the limit imposed by the manufactures, i.e. equals to 10%. Preliminary activities allowed to evaluate the impact of the single measured quantities and to optimize the performance of the proposed artificial architectures. All this features potentially allow to reduce and/or eliminate unnecessary physical sensors thus reducing costs and operating times.

The feasibility of detecting PD in inverter-fed machines using acoustic emission sensor realized using fiber optics was investigated. The sensors, have a frequency response up to 500 kHz and are interrogated using a fiber-optic interferometric system. Three acoustic emission sensors were attached to the stack of the stator to explore the possibility of both PD detection and localization. Results were compared with those achieved using electrical detection and those provided by an ultrasonic camera. Preliminary tests using AC sinusoidal waveforms did show that the acoustic emission sensors and the ultrasonic camera could couple PD activity with the same sensitivity of a conventional electrical PD detection system having a sensitivity of 10 pC. Repetitive voltage impulses were found to cause a high disturbance level in both the acoustic emission sensor and the ultrasonic camera. The noise was ascribed to vibrations induced by the electrodynamic stress due to the impulsive charging current. However, working on the AE sensors in the frequency domain, it was possible to highlight frequency ranges where the PD signal exceeded the disturbance, thus enabling PD detection. The same was not possible for the ultrasonic camera.

Background and importanceA software module (APOTECAtrial) was introduced in clinical practice to manage clinical trials and investigational drugs, thereby minimising manual activities and ensuring maximum traceability (1). APOTECAtrial was developed in accordance with the Good Clinical Practice (GCP) guidelines, in particular with regard to subject safety, outcome reliability, characteristics of electronic systems/data, and quality management with a risk-based approach.Aim and objectivesThe objective of this study was to assess the risk associated with the pharmacy-based management of clinical trials before and after the implementation of the software module APOTECAtrial.Material and methodsThe conventional manual process and the improvements introduced after the implementation of APOTECAtrial were assessed through a comparative risk analysis. First, the process was divided into seven phases (delivery to the pharmacy, preparation/dispensing, returns management, disposal, storage, data management, monitoring). The activities related to each phase and the corresponding potential failures were identified. The risk was assessed by rating the severity (S), frequency (F) and detectability (D) of the potential effect of the failures. The risk index (S×F×D) was calculated for each activity (RI) and for the entire process (RItotal). The index of improvement (IR before implementation divided by IR after implementation) was calculated for each area (IM) and for the entire process (IMtotal).ResultsOverall, 37 activities were assessed. The RItotal decreased by 53%, from 449 (before implementation) to 207 (after implementation). The IMtotal amounted to 2.2. The highest IR reduction was found in the preparation/dispensing phase (from 152 to 42) with an IM equal to 3.6. IM values ranged between 1.7 and 4.5. Most of the improvements introduced (79%) referred to traceability and data integrity, while 21% impacted on the quality of the drug dispensed.Conclusion and relevanceThe risk analysis revealed that fully-automated management of clinical trials represents an important improvement of the clinical pharmacy practice in terms of safety. Since the potential risks are significantly reduced, the automated process guarantees high-quality standards and GCP-compliance. Several manual and repetitive activities were simplified, thereby allowing pharmacists to spend more time on clinical and patient-oriented tasks..References and/or acknowledgements1. Leoni S, et al. Integration of clinical trials management into a safe and fully-automated onco-haematology workflow. https://www.eahp.eu/sites/default/files/gpi_documents/pc10257.pdfConflict of interestNo conflict of interest

Background and importanceA robotic system for automated preparation of cytotoxic drugs, such as APOTECAchemo, ensures reduced occupational exposure to toxic substances and aseptic conditions. The most critical operations are performed by a robotic arm; the operator’s intervention is limited to loading/unloading materials in a rotating warehouse through an unloading/loading area enclosed within a laminar airflow barrier. In European hospital pharmacies which comply with good manufacturing practice (GMP), the performances of the robot are assessed by GMP qualification to confirm that the technology meets the set quality standards.Aim and objectivesThe aim of this study was to evaluate microbiological performances and environmental conditions during fully automated preparation with APOTECAchemo in a grade B cleanroom.Material and methodsEffectiveness of laminar airflow retention was checked by potassium iodide (KI) discus test in the unloading/loading area of the APOTECAchemo robot. Aseptic preparation of cytotoxic drugs was evaluated with media fill simulation tests on three consecutive days. In total, 240 products (180 infusion bags, 30 syringes, 30 elastomeric pumps) were automatically filled with single/double strength tryptic soy broth in lieu of drug products. Microbiological environmental controls were performed by passive air sampling (settle plates, four locations), surface sampling (contact plates/swabs, 14 locations), and active air sampling (three locations). Samples were taken for each shift. Media fill products were visually inspected for turbidity after 14 days of incubation. The number of colony forming units (CFU) per plate were counted and identified.ResultsThe results of the KI discus test were far below the acceptance limit, demonstrating the effectiveness of laminar airflow in preventing the escape of particles from the internal areas of the robot. None of 240 media fills showed turbidity after incubation, thereby indicating no contamination with microorganisms. The working area met the grade A limit (0 CFU/sample). In the warehouse, 1 CFU was found in two samples (mean 0.2 CFU/contact plate, 0.1 CFU/settle plate). Microbial contamination was slightly higher in the loading area (mean 0.3 CFU/contact plate, 0.8 CFU/settle plate). Most of the CFUs were identified as skin related microorganisms, such as Staphylococcus epidermidis and Staphylococcus hominis.Conclusion and relevanceExtensive media fill tests and environmental monitoring during automated preparation with the robot revealed well controlled aseptic procedures and adequate sterility levels, thereby complying with the quality standard set by the hospital pharmacy.References and/or acknowledgementsConflict of interestNo conflict of interest

Background and importanceThe introduction of the pharmacy based central intravenous additive service (CIVAS) enabled the batches production of ready-to-administer, non-hazardous intravenous drugs by using the robotic system APOTECAunit. The clinical benefits in terms of a higher quality aseptic process, reduced medication errors and increased quality control testing have been demonstrated.Aim and objectivesThe aim of the study was to evaluate the impact of the fully automatised CIVAS on the working efficiency of the wards by measuring the time saved in daily nursing practice.Material and methodsThe study was conducted over 3 months with data collected before and after introducing the CIVAS. Overall, three wards were analysed: inpatient haematology (IH), cardiac surgical (CS) and infectious diseases (ID). The nursing staff was observed daily for 4 weeks. Different tasks associated with the following intravenous drugs supplied by the CIVAS were recorded: ondansetron, palonosetron, cefazoline, piperacillin–tazobactam, ceftriaxone and dexamethasone. The average working time spent on managing intravenous drugs was calculated per full time equivalent (FTE), including direct activities (intravenous drug preparation) and indirect activities (procurement of drugs and medical devices, drugs inventory management, drug ordering). 1.0 FTE was equivalent to a nurse working 8.0 hours per day, 5 days per week.ResultsThe overall time spent on managing intravenous drugs decreased from 1.6 to 0.7 hours/day/FTE in the IH ward, from 0.75 to 0.15 hours/day/FTE in the CS ward and from 2.1 to 0.43 hours/day/FTE in the ID ward. Before implementing the CIVAS, on average 1.0 FTE spent 0.5 hours/day on preparing intravenous drugs, whereas the intravenous drug preparation time was reduced to zero afterwards in each of the three wards. The time spent on procurement decreased by 80% in the IH ward, by 60% in the CS ward and by 75% in the ID ward. The nursing staff shortened the time required for drug inventory management by 33% in the CS ward and by 50% in the IH and ID wards. In all wards, ordering of drugs was reduced by at least 50%.Conclusion and relevanceThe study showed that the centralised, automated preparation of intravenous drugs optimised the working efficiency of the wards, thereby allowing the nursing staff to dedicate more time to perform tasks directly involved with patient care.References and/or acknowledgementsConflict of interestNo conflict of interest