Pub Date : 2023-11-06DOI: 10.24425/bpasts.2020.134655
Most of the basic control methods of the grid-connected converter (GCC) are defined to work with a sine wave grid voltage. In that case if the grid voltage is distorted by higher harmonics, the grid current may be distorted too, which, in consequence, may increase the value of the THD of the grid voltage. The paper deals with an improved finite control set model predictive control (FCS-MPC) method of an LCL-filtered GCC operating under distorted grid conditions. The proposed method utilizes supplementary grid current feedback to calculate the reference converter current. The introduced signal allows to effectively improve the operation when the grid is subject to harmonic distortion. The paper shows a simulation analysis of the proposed control scheme operating with and without additional feedback under grid distortions. To validate the practical feasibility of the proposed method an algorithm was implemented on a 32-bit microcontroller STM32F7 with a floating point unit to control a 10 kW GCC. The laboratory test setup provided experimental results showing properties of the introduced control scheme.
{"title":"Finite control set MPC of LCL-filtered grid-connected power converter operating under grid distortions","authors":"","doi":"10.24425/bpasts.2020.134655","DOIUrl":"https://doi.org/10.24425/bpasts.2020.134655","url":null,"abstract":"Most of the basic control methods of the grid-connected converter (GCC) are defined to work with a sine wave grid voltage. In that case if the grid voltage is distorted by higher harmonics, the grid current may be distorted too, which, in consequence, may increase the value of the THD of the grid voltage. The paper deals with an improved finite control set model predictive control (FCS-MPC) method of an LCL-filtered GCC operating under distorted grid conditions. The proposed method utilizes supplementary grid current feedback to calculate the reference converter current. The introduced signal allows to effectively improve the operation when the grid is subject to harmonic distortion. The paper shows a simulation analysis of the proposed control scheme operating with and without additional feedback under grid distortions. To validate the practical feasibility of the proposed method an algorithm was implemented on a 32-bit microcontroller STM32F7 with a floating point unit to control a 10 kW GCC. The laboratory test setup provided experimental results showing properties of the introduced control scheme.","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"371 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135636167","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2021.137726
Time invariant linear operators are the building blocks of signal processing. Weighted circular convolution and signal processing framework in a generalized Fourier domain are introduced by Jorge Martinez. In this paper, we prove that under this new signal processing framework, weighted circular convolution also has a generalized time invariant property. We also give an application of this property to algorithm of continuous wavelet transform (CWT). Specifically, we have previously studied the algorithm of CWT based on generalized Fourier transform with parameter 1. In this paper, we prove that the parameter can take any complex number. Numerical experiments are presented to further demonstrate our analyses.
{"title":"Time invariant property of weighted circular convolution and its application to continuous wavelet transform","authors":"","doi":"10.24425/bpasts.2021.137726","DOIUrl":"https://doi.org/10.24425/bpasts.2021.137726","url":null,"abstract":"Time invariant linear operators are the building blocks of signal processing. Weighted circular convolution and signal processing framework in a generalized Fourier domain are introduced by Jorge Martinez. In this paper, we prove that under this new signal processing framework, weighted circular convolution also has a generalized time invariant property. We also give an application of this property to algorithm of continuous wavelet transform (CWT). Specifically, we have previously studied the algorithm of CWT based on generalized Fourier transform with parameter 1. In this paper, we prove that the parameter can take any complex number. Numerical experiments are presented to further demonstrate our analyses.","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"366 1‐2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135636172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2021.139391
Numerous examples of physically unjustified neural networks, despite satisfactory performance, generate contradictions with logic and lead to many inaccuracies in the final applications. One of the methods to justify the typical black-box model already at the training stage involves extending its cost function by a relationship directly inspired by the physical formula. This publication explains the concept of physicsguided neural network (PGNN), makes an overview of already proposed solutions in the field and describes possibilities of implementing physics-based loss functions for spatial analysis. Our approach shows that the model predictions are not only optimal but also scientifically consistent with domain specific equations. Furthermore, we present two applications of PGNNs and illustrate their advantages in theory by solving Poisson’s and Burger’s partial differential equations. The proposed formulas describe various real-world processes and have numerous applications in the area of applied mathematics. Eventually, the usage of scientific knowledge contained in the tailored cost functions shows that our methods guarantee physics-consistent results as well as better generalizability of the model compared to classical, artificial neural networks.
{"title":"Physics-guided neural networks (PGNNs) to solve differential equations for spatial analysis","authors":"","doi":"10.24425/bpasts.2021.139391","DOIUrl":"https://doi.org/10.24425/bpasts.2021.139391","url":null,"abstract":"Numerous examples of physically unjustified neural networks, despite satisfactory performance, generate contradictions with logic and lead to many inaccuracies in the final applications. One of the methods to justify the typical black-box model already at the training stage involves extending its cost function by a relationship directly inspired by the physical formula. This publication explains the concept of physicsguided neural network (PGNN), makes an overview of already proposed solutions in the field and describes possibilities of implementing physics-based loss functions for spatial analysis. Our approach shows that the model predictions are not only optimal but also scientifically consistent with domain specific equations. Furthermore, we present two applications of PGNNs and illustrate their advantages in theory by solving Poisson’s and Burger’s partial differential equations. The proposed formulas describe various real-world processes and have numerous applications in the area of applied mathematics. Eventually, the usage of scientific knowledge contained in the tailored cost functions shows that our methods guarantee physics-consistent results as well as better generalizability of the model compared to classical, artificial neural networks.","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"364 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135636175","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2020.134175
This paper describes practical issues related to control of the active power buffer (APB) developed for a 2 kVA single-phase inverter. The buffer is designed using the latest GaN HEMTs controlled with triangular current mode to reduce switching losses, however, the switching frequency should be limited to 1 MHz. In the case of the presented analogue-digital controller, frequency is influenced by a reference current of the APB and circuit. Therefore, the operation at start-up and shut-down is especially challenging. A modified control algorithm that also includes pre-charging and discharging process of the energy buffer is presented and experimentally verified by series of tests of the 2 kVA GaN based inverter with the APB.
{"title":"GaN-based soft-switched active power buffer operating at ZCS – problems of start-up and shut-down","authors":"","doi":"10.24425/bpasts.2020.134175","DOIUrl":"https://doi.org/10.24425/bpasts.2020.134175","url":null,"abstract":"This paper describes practical issues related to control of the active power buffer (APB) developed for a 2 kVA single-phase inverter. The buffer is designed using the latest GaN HEMTs controlled with triangular current mode to reduce switching losses, however, the switching frequency should be limited to 1 MHz. In the case of the presented analogue-digital controller, frequency is influenced by a reference current of the APB and circuit. Therefore, the operation at start-up and shut-down is especially challenging. A modified control algorithm that also includes pre-charging and discharging process of the energy buffer is presented and experimentally verified by series of tests of the 2 kVA GaN based inverter with the APB.","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"382 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135636360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2021.137347
. Voice acoustic analysis can be a valuable and objective tool supporting the diagnosis of many neurodegenerative diseases, especially in times of distant medical examination during the pandemic. The article compares the application of selected signal processing methods and machine learning algorithms for the taxonomy of acquired speech signals representing the vowel a with prolonged phonation in patients with Parkinson’s disease and healthy subjects. The study was conducted using three different feature engineering techniques for the generation of speech signal features as well as the deep learning approach based on the processing of images involving spectrograms of different time and frequency resolutions. The research utilized real recordings acquired in the Department of Neurology at the Medical University of Warsaw, Poland. The discriminatory ability of feature vectors was evaluated using the SVM technique. The spectrograms were processed by the popular AlexNet convolutional neural network adopted to the binary classification task according to the strategy of transfer learning. The results of numerical experiments have shown different efficiencies of the examined approaches; however, the sensitivity of the best test based on the selected features proposed with respect to biological grounds of voice articulation reached the value of 97% with the specificity no worse than 93%. The results could be further slightly improved thanks to the combination of the selected deep learning and feature engineering algorithms in one stacked ensemble model.
{"title":"Deep learning vs feature engineering in the assessment of voice signals for diagnosis in Parkinson’s disease","authors":"","doi":"10.24425/bpasts.2021.137347","DOIUrl":"https://doi.org/10.24425/bpasts.2021.137347","url":null,"abstract":". Voice acoustic analysis can be a valuable and objective tool supporting the diagnosis of many neurodegenerative diseases, especially in times of distant medical examination during the pandemic. The article compares the application of selected signal processing methods and machine learning algorithms for the taxonomy of acquired speech signals representing the vowel a with prolonged phonation in patients with Parkinson’s disease and healthy subjects. The study was conducted using three different feature engineering techniques for the generation of speech signal features as well as the deep learning approach based on the processing of images involving spectrograms of different time and frequency resolutions. The research utilized real recordings acquired in the Department of Neurology at the Medical University of Warsaw, Poland. The discriminatory ability of feature vectors was evaluated using the SVM technique. The spectrograms were processed by the popular AlexNet convolutional neural network adopted to the binary classification task according to the strategy of transfer learning. The results of numerical experiments have shown different efficiencies of the examined approaches; however, the sensitivity of the best test based on the selected features proposed with respect to biological grounds of voice articulation reached the value of 97% with the specificity no worse than 93%. The results could be further slightly improved thanks to the combination of the selected deep learning and feature engineering algorithms in one stacked ensemble model.","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"256 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135678736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2022.140694
{"title":"Arriving air traffic separations generalized model identification","authors":"","doi":"10.24425/bpasts.2022.140694","DOIUrl":"https://doi.org/10.24425/bpasts.2022.140694","url":null,"abstract":"","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"241 2‐3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135678753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2022.143832
{"title":"Sustainability in production in the aspect of Industry 4.0","authors":"","doi":"10.24425/bpasts.2022.143832","DOIUrl":"https://doi.org/10.24425/bpasts.2022.143832","url":null,"abstract":"","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"298 2‐3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135678889","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2021.139961
. Object tracking based on Siamese networks has achieved great success in recent years, but increasingly advanced trackers are also becoming cumbersome, which will severely limit deployment on resource-constrained devices. To solve the above problems, we designed a network with the same or higher tracking performance as other lightweight models based on the SiamFC lightweight tracking model. At the same time, for the problems that the SiamFC tracking network is poor in processing similar semantic information, deformation, illumination change, and scale change, we propose a global attention module and different scale training and testing strategies to solve them. To verify the effectiveness of the proposed algorithm, this paper has done comparative experiments on the ILSVRC, OTB100, VOT2018 datasets. The experimental results show that the method proposed in this paper can significantly improve the performance of the benchmark algorithm.
{"title":"Single target tracking algorithm for lightweight Siamese networks based on global attention","authors":"","doi":"10.24425/bpasts.2021.139961","DOIUrl":"https://doi.org/10.24425/bpasts.2021.139961","url":null,"abstract":". Object tracking based on Siamese networks has achieved great success in recent years, but increasingly advanced trackers are also becoming cumbersome, which will severely limit deployment on resource-constrained devices. To solve the above problems, we designed a network with the same or higher tracking performance as other lightweight models based on the SiamFC lightweight tracking model. At the same time, for the problems that the SiamFC tracking network is poor in processing similar semantic information, deformation, illumination change, and scale change, we propose a global attention module and different scale training and testing strategies to solve them. To verify the effectiveness of the proposed algorithm, this paper has done comparative experiments on the ILSVRC, OTB100, VOT2018 datasets. The experimental results show that the method proposed in this paper can significantly improve the performance of the benchmark algorithm.","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"289 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135678900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2020.135383
. Several recent earthquakes have indicated that the design and construction of bridges based on former seismic design provisions are susceptible to fatal collapse triggered by the failure of reinforced concrete columns. This paper incorporates an experimental investigation into the seismic response of nonductile bridge piers strengthened with low-cost glass fiber reinforced polymers (LC-GFRP). Three full-scale bridge piers were tested under lateral cyclic loading. A control bridge pier was tested in the as-built condition and the other two bridge piers were experimentally tested after strengthening them with LC-GFRP jacketing. The LC-GFRP strengthening was performed using two different configurations. The control bridge pier showed poor seismic response with the progress of significant cracks at very low drift levels. Test results indicated the efficiency of the tested strengthening configurations to improve the performance of the strengthened bridge piers including crack pattern, yield, and ultimate cyclic load capacities, ductility ratio, dissipated energy capacity, initial stiffness degradation, and fracture mode.
{"title":"Seismic strengthening of nonductile bridge piers using low-cost glass fiber polymers","authors":"","doi":"10.24425/bpasts.2020.135383","DOIUrl":"https://doi.org/10.24425/bpasts.2020.135383","url":null,"abstract":". Several recent earthquakes have indicated that the design and construction of bridges based on former seismic design provisions are susceptible to fatal collapse triggered by the failure of reinforced concrete columns. This paper incorporates an experimental investigation into the seismic response of nonductile bridge piers strengthened with low-cost glass fiber reinforced polymers (LC-GFRP). Three full-scale bridge piers were tested under lateral cyclic loading. A control bridge pier was tested in the as-built condition and the other two bridge piers were experimentally tested after strengthening them with LC-GFRP jacketing. The LC-GFRP strengthening was performed using two different configurations. The control bridge pier showed poor seismic response with the progress of significant cracks at very low drift levels. Test results indicated the efficiency of the tested strengthening configurations to improve the performance of the strengthened bridge piers including crack pattern, yield, and ultimate cyclic load capacities, ductility ratio, dissipated energy capacity, initial stiffness degradation, and fracture mode.","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"284 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135678907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.24425/bpasts.2022.143102
. In this paper the controllability properties of the convex linear combination of fractional, linear, discrete-time systems are characterized and investigated. The notions of linear convex combination and controllability in the context of fractional-order systems are recalled. Then, the controllability property of such a linear combination of discrete-time, linear fractional systems is proven. Further, the reduction of an infinite problem of transition matrix derivation is reduced to a finite one, which greatly simplifies the numerical burden of the controllability issue. Examples of controllable and uncontrollable, single-input, linear systems are presented. The possibility of extension of the considerations to multi-input systems is shown.
{"title":"Controllability of linear convex combination of linear discrete-time fractional systems","authors":"","doi":"10.24425/bpasts.2022.143102","DOIUrl":"https://doi.org/10.24425/bpasts.2022.143102","url":null,"abstract":". In this paper the controllability properties of the convex linear combination of fractional, linear, discrete-time systems are characterized and investigated. The notions of linear convex combination and controllability in the context of fractional-order systems are recalled. Then, the controllability property of such a linear combination of discrete-time, linear fractional systems is proven. Further, the reduction of an infinite problem of transition matrix derivation is reduced to a finite one, which greatly simplifies the numerical burden of the controllability issue. Examples of controllable and uncontrollable, single-input, linear systems are presented. The possibility of extension of the considerations to multi-input systems is shown.","PeriodicalId":55299,"journal":{"name":"Bulletin of the Polish Academy of Sciences-Technical Sciences","volume":"278 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135678916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：