Emerging digital transformation in industry is noticeable among others in Supply Chain Management (SCM). For instance, applying new-generation digitalized technologies in the Dairy Supply Chain (DSC) enables an increase of manufacturing productivity, improves planning and forecasting, and also enhances competitive capabilities according to Industry 4.0 assumptions. It is worth mentioning, that in modern DSC, high visibility of raw materials, components, products, and processes by all contributors on all stages of DSC is crucial. This article focuses on the transparency aspect of the DSC supported by IoT-based technologies enabling interoperability among all DSC participants. The paper addresses the problem of effective integration of heterogeneous data sources, i.e., deployed new technological IoT solutions with traditional SCM systems and a third-party software component. The main objective of this report is to propose the IoT-based DSC model comprising four chain stages: milk production, milk transportation, milk processing, and dairy products distribution. Moreover, the comprehensive DSC domain ontology as a knowledge model is formulated and described. The ontology aims on improvement of the DSC management efficiency by facilitating interoperability within DSC. The applicability of the proposed ontological model is verified using a sustainable-oriented case study, which estimates the environmental footprint at the milk transportation stage of the DSC. © 2021 Kauno Technologijos Universitetas. All rights reserved.

Cardiovascular pathologies cause 23.5% of human deaths, worldwide. An auto-diagnostic system monitoring heart activity, which can identify the early symptoms of cardiac illnesses, might reduce the death rate caused by these problems. Phonocardiography (PCG) is one of the possible techniques able to detect heart problems. Nevertheless, acoustic signal enhancement is required since it is exposed to various disturbances coming from different sources. The most common denoising enhancement is based on the Wavelet Transform (WT). However, the WT is highly susceptible to variations in the noise frequency distribution. This paper proposes a new adaptive denoising algorithm, which combines WT and Time Delay Neural Networks (TDNN). The acquired signal is decomposed by means of the WT using the coif five-wavelet basis at the tenth decomposition level and then provided as input to the TDNN. Besides the advantage of adaptive thresholding, the reason for using TDNNs is their capacity of estimating the Inverse Wavelet Transform (IWT). The best parameters of the TDNN were found for a NN consisting of 25 neurons in the first and 15 in the second layer and the delay block of 12 samples. The method was evaluated on several pathological heart sounds and on signals recorded in a noisy environment. The performance of the developed system with respect to other wavelet-based denoising approaches was validated by the online questionnaire.

The manufacturing industry are dependent of engineering expertise. Currently the ability to supply the industry with engineering graduates and staff that have an up-to-date and relevant competences might be considered as a challenge for the society. In this paper an education approach is presented where academia - industry - research institutes cooperate around the development and implementation of master level courses. The methods applied to reach the educational goals, concerning expert competence within remote diagnostics, have been on site and remote lectures given by engineering, medical and metrology experts. The pedagogical approach utilized has been flipped classroom. The main results show that academic courses developed in cooperation with industry requires flexibility, time and effort from the involved partners. The evaluation interviews indicate that student are satisfied with the courses and pedagogical approach but suggests more reconciliation meetings for course development. Labs early in the course was considered good, and division of labs at the system and the component level. However further long-term studies of evaluation of impact is necessary.

The remotely controlled laboratory setup for Active Noise Control (ANC) developed by Blekin-ge Institute of Technology, Sweden provides an efficient learning platform for the students to implement and learn ANC algorithms with real world physical system, hardware and signals. The initial laboratory prototype based on a Digital Signal Processor (DSP) TMS320C6713 from Texas Instruments (TI) was successfully tested with Filtered-x Least Mean Square (F-XLMS) algorithm applied to control noise in a ventilation duct. The resources of the DSP platform used in the remote laboratory setup enable testing and investigating substantially more challenging and computationally demanding algorithms. In this paper, we expand the horizon of the laboratory setup by testing more advanced and complicated single channel feed forward ANC algorithms. Filtered-x versions of algorithms such as the normalized least mean square (N-LMS), leaky least mean square (L-LMS), Filtered-U recursive least mean square (FURLMS) and recursive least square (RLS) algorithm etc. have been implemented utilizing the remote web based client provided in the remote laboratory. A comprehensive performance comparison of the aforementioned algorithms for the remote laboratory setup is presented to demonstrate the viability of the remote laboratory.

Remotely controlled laboratories in educational institutions are gaining popularity at an exponential rate due to the multidimensional benefits they provide. The Virtual Instrument Systems in Reality (VISIR) project by Blekinge Institute of Technology (BTH) Sweden has successfully implemented remotely controlled laboratories, with remotely controlled real instruments and experimental setups. Currently these laboratories provide students the opportunity to conduct experiments in the field of electronics, antenna theory and mechanical vibration measurements. In this paper a prototype system of a remotely controlled laboratory for active noise control (ANC) is introduced. The proposed lab will focus on addressing the problem of a ventilation duct noise. The laboratory is informative and to a great extent introduces a student to the general steps in ANC when it is suggested as a plausible solution for a noise problem. The student can perform an investigation concerning feasibility of active control, design, configuration and implementation of an active control system. The laboratory is based on a modern and relevant DSP platform with the corresponding software development environment controlled remotely. In addition, it may be utilized remotely both for lab assignments in acoustics courses and digital signal processing courses.

In a narrow duct, a relatively simple single-channel feedforward ANC system may be used to attenuate noise propagating as plane waves. However, for ducts with larger dimensions the cut-on frequencies for one or several higher-order acoustic modes may be within the frequency range where ANC is applied. In such situations it is generally necessary to use a multiplechannel feedforward ANC system with several secondary sources, error sensors, and perhaps reference sensors. Such a system has a significantly higher complexity than a single-channel ANC system. In this paper another approach is described. Instead of using a multiple-channel feedforward ANC system on a duct of large dimension, the idea is to divide the duct into several more narrow parallel ducts. In this way the complexity of the ANC system may be reduced. In the experiments conducted for this paper, a duct was divided into two more narrow ducts. The noise propagating in each duct was controlled by a feedforward ANC system based on the leaky filtered-x LMS algorithm, where different reference- and error microphone configurations were used. The different configurations were compared to a configuration where the noise in respective narrow duct was controlled using a basic single-channel ANC system per duct. The results preliminary show that high attenuation of low-frequency noise in a duct of large dimension may be achieved using this approach.

This paper presents a personal voice analyzer and trainer that allow the user to perform four daily exercises to improve the voice capacity. The system grades how well the user is performing the exercises by analyzing the duration, the intensity and the pitch of the user’s voice.

In active noise control (ANC) applications the sensor signals feeding the controller may be contaminated by signal components not linearly related to the actual noise to be controlled, e.g. measurement noise. An example is the noise in the microphone signals generated by airflow turbulence when applying ANC to ducts. The measurement noise may significantly degrade the performance of an ANC system. This paper is concerned with theoretical investigations of the influence of measurement noise on the performance and convergence of the filtered-x LMS algorithm. Further, computer simulations have been performed to verify the theoretical results. In the theoretical investigations, the convergence of the filtered-x LMS algorithm is derived for different cases with measurement noise affecting the different sensor signals. These cases are compared to the ideal case with no measurement noise present at neither sensor signal. The results from both the theoretical investigations and the simulations show that measurement noise can, depending on the SNR of the sensor signals, degrade the performance of the filtered-x LMS algorithm regarding both the filter coefficients the algorithm converges to in mean, and the convergence rate of the algorithm.

Today mechanical ventilation systems are installed in many buildings to handle the ventilation. Such ventilation systems constitute a potential source of unwanted background noise in the buildings. In order to reduce the noise propagating through the ducts passive silencers are normally installed as a part of the ventilation system. However, the passive silencers are relatively ineffective in the low frequency range. A solution can be to use a combination of active noise control (ANC) and passive techniques where the ANC system extends the noise attenuation to include the low frequency noise. The construction of the ventilation systems can vary substantially between different buildings and may contain many different duct parts in different compositions etc. Hence, it may not be trivial to find a cost- and performance efficient installation of an ANC system in a ventilation system, e.g. to find a proper installation position of it and suitable passive silencers to combine it with, finding out if one or several ANC systems should be used, and so on. In order to make the design of a complete ventilation system (including one or several ANC systems) cost- and time efficient, an adequate mathematical model of the duct is required. The purpose of the model is to describe sound propagation in the duct. Such model can be built e.g. based on the two-port theory or by finite element method. In this paper initial investigations concerning modelling of standard duct parts were performed. Simulations were carried out to find parameters such as Noise Reduction etc., using both plane wave two-port theory and finite element modelling. The results of the simulations were compared with measurement results.