Search

Nowadays, with the rise of drone and satellite technology, there is a possibility for its application in sea and coastal surveillance. An advantage of this type of application is the automated recognition of marine objects, among which the most important are vessels. This paper presents the principle of vessel recognition based on the extraction of satellite image features of the vessel and the application of a multilayer perceptron (MLP). Dataset used in this research contains the total of 2750 images, where 2112 images are used as training set while the remaining 638 images are used for testing purposes. The SIFT and SURF algorithms were used to extract image features, which were later used as the input vector for MLP.The best results are achieved if a model with four hidden layers is used. These layers are constructed with 32, 128, 32, 128 neurons with ReLU activation function, respectively. Regarding the application of feature extraction, it can be observed that better results are achieved if the SIFT algorithm is used. The ROC AUC value achieved with the combination of SIFT and MLP reaches 0.99.

The importance of error detection is high, especially in modern manufacturing processes where assembly lines operate without direct supervision. Stopping the faulty operation in time can prevent damage to the assembly line. Public dataset is used, containing 15 classes, 2 types of faultless operation and 13 types of faults, with 463 force and torsion datapoints. Four different methods are used: Multilayer Perceptron (MLP) selected due to high classification performance, Support Vector Machines (SVM) commonly used for a low number of datapoints, Convolutional Neural Network (CNN) known for high performance in classification with matrix inputs and Siamese Neural Network (SNN) novel method with high performance in small datasets. Two classification tasks are performed-error detection and classification. Grid search is used for hyperparameter variation and F1 score as a metric, with a 10 fold cross-validation. Authors propose a hybrid system consisting of SNN for detection and CNN for fault classification.

Fish population monitoring systems based on underwater video recording are becoming more popular nowadays, however, manual processing and analysis of such data can be time-consuming. Therefore, by utilizing machine learning algorithms, the data can be processed more efficiently. In this research, authors investigate the possibility of convolutional neural network (CNN) implementation for fish species classification. The dataset used in this research consists of four fish species (Plectroglyphidodon dickii, Chromis chrysura, Amphiprion clarkii, and Chaetodon lunulatus), which gives a total of 12859 fish images. For the aforementioned classification algorithm, different combinations of hyperparameters were examined as well as the impact of different activation functions on the classification performance. As a result, the best CNN classification performance was achieved when Identity activation function is applied to hidden layers, RMSprop is used as a solver with a learning rate of 0.001, and a learning rate decay of 1e-5. Accordingly, the proposed CNN model is capable of performing high-quality fish species classifications.

An artificial neural network (ANN) approach is proposed to the problem of estimating the propeller torques of a frigate using combined diesel, electric and gas (CODLAG) propulsion system. The authors use a multilayer perceptron (MLP) feed-forward ANN trained with data from a dataset which describes the decay state coefficients as outputs and system parameters as inputs – with a goal of determining the propeller torques, removing the decay state coefficients and using the torque values of the starboard and port propellers as outputs. A total of 53760 ANNs are trained – 26880 for each of the propellers, with a total 8960 parameter combinations. The results are evaluated using mean absolute error (MAE) and coefficient of determination (R2). Best results for the starboard propeller are MAE of 2.68 [Nm], and MAE of 2.58 [Nm] for the port propeller with following ANN configurations respectively: 2 hidden layers with 32 neurons and identity activation and 3 hidden layers with 16, 32 and 16 neurons and identity activation function. Both configurations achieve R2 value higher than 0.99.

An artificial neural network (ANN) approach is proposed to the problem of estimating the propeller torques of a frigate using combined diesel, electric and gas (CODLAG) propulsion system. The authors use a multilayer perceptron (MLP) feed-forward ANN trained with data from a dataset which describes the decay state coefficients as outputs and system parameters as inputs – with a goal of determining the propeller torques, removing the decay state coefficients and using the torque values of the starboard and port propellers as outputs. A total of 53760 ANNs are trained – 26880 for each of the propellers, with a total 8960 parameter combinations. The results are evaluated using mean absolute error (MAE) and coefficient of determination (R2). Best results for the starboard propeller are MAE of 2.68 [Nm], and MAE of 2.58 [Nm] for the port propeller with following ANN configurations respectively: 2 hidden layers with 32 neurons and identity activation and 3 hidden layers with 16, 32 and 16 neurons and identity activation function. Both configurations achieve R2 value higher than 0.99.

Fish population monitoring systems based on underwater video recording are becoming more popular nowadays, however, manual processing and analysis of such data can be time-consuming. Therefore, by utilizing machine learning algorithms, the data can be processed more efficiently. In this research, authors investigate the possibility of convolutional neural network (CNN) implementation for fish species classification. The dataset used in this research consists of four fish species (Plectroglyphidodon dickii, Chromis chrysura, Amphiprion clarkii, and Chaetodon lunulatus), which gives a total of 12859 fish images. For the aforementioned classification algorithm, different combinations of hyperparameters were examined as well as the impact of different activation functions on the classification performance. As a result, the best CNN classification performance was achieved when Identity activation function is applied to hidden layers, RMSprop is used as a solver with a learning rate of 0.001, and a learning rate decay of 1e-5. Accordingly, the proposed CNN model is capable of performing high-quality fish species classifications.

In this paper, the publicly available dataset of condition based maintenance of combined diesel-electric and gas (CODLAG) propulsion system for ships has been utilized to obtain symbolic expressions which could estimate gas turbine shaft torque and fuel flow using genetic programming (GP) algorithm. The entire dataset consists of 11934 samples that was divided into training and testing portions of dataset in an 80:20 ratio. The training dataset used to train the GP algorithm to obtain symbolic expressions for gas turbine shaft torque and fuel flow estimation consisted of 9548 samples. The best symbolic expressions obtained for gas turbine shaft torque and fuel flow estimation were obtained based on their R2 score generated as a result of the application of the testing portion of the dataset on the aforementioned symbolic expressions. The testing portion of the dataset consisted of 2386 samples. The three best symbolic expressions obtained for gas turbine shaft torque estimation generated R2 scores of 0.999201, 0.999296, and 0.999374, respectively. The three best symbolic expressions obtained for fuel flow estimation generated R2 scores of 0.995495, 0.996465, and 0.996487, respectively.

Authors present a Multilayer Perceptron (MLP) artificial neural network (ANN) method for the purpose of estimating a speed of a frigate using a combined diesel-electric and gas (CODLAG) propulsion system. Dataset used is publicly available, as condition-based maintenance of naval propulsion plants dataset, out of which GT Compressor decay state coefficient and GT Turbine decay state coefficient are unused, while 15 features are used as input and ship speed is used as dataset output. Data set consists of 11934 data points out of which 8950 (75%) are used as a training set and 2984 (25%) are used as a testing set. 26880 MLPs, with 8960 different parameter combinations are trained using a grid search algorithm, quality of each solution being estimated with coefficient of determination (R2) and mean absolute error (MAE). Results show that a high-quality estimation can be made using an MLP, with best result having an error of just 3.4485x10-5 knots (absolute error of 0.00014% of the range). This result was achieved with a MLP with three hidden layers containing 100 neurons each, logistic activation function, LBFGS solver, constant learning rate of 0.1 and no L2 regularization.

The Traveling salesman problem (TSP) defines the problem of finding the optimal path between multiple points, connected by paths of a certain cost. This paper applies that problem formulation in the maritime environment, specifically a path planning problem for a tour boat visiting popular tourist locations in Medulin, Croatia. The problem is solved using two evolutionary computing methods – the genetic algorithm (GA) and the simulated annealing (SA) - and comparing the results (are compared) by an extensive search of the solution space. The results show that evolutionary computing algorithms provide comparable results to an extensive search in a shorter amount of time, with SA providing better results of the two.

This paper presents an analysis of the gas turbine real process (with all losses included) before and after a major maintenance. The analysis of both gas turbine operating regimes is based on data measured during its exploitation. Contrary to authors’ expectations, the major maintenance process did not result either in any decrease in losses or increase in efficiencies for the majority of the gas turbine components. However, the major maintenance influenced positively the gas turbine combustion chambers (reduction in losses and increase in the combustion chambers efficiency). After the major maintenance, the overall process efficiency decreased from 43.796% to 41.319% due to a significant decrease in the air mass flow rate and to an increase in the fuel mass flow rate in combustion chambers. A decrease in the gas turbine produced cumulative and useful power after a major maintenance also increased the specific fuel consumption.

The Traveling salesman problem (TSP) defines the problem of finding the optimal path between multiple points, connected by paths of a certain cost. This paper applies that problem formulation in the maritime environment, specifically a path planning problem for a tour boat visiting popular tourist locations in Medulin, Croatia. The problem is solved using two evolutionary computing methods – the genetic algorithm (GA) and the simulated annealing (SA) - and comparing the results (are compared) by an extensive search of the solution space. The results show that evolutionary computing algorithms provide comparable results to an extensive search in a shorter amount of time, with SA providing better results of the two.

This paper presents an analysis of the gas turbine real process (with all losses included) before and after a major maintenance. The analysis of both gas turbine operating regimes is based on data measured during its exploitation. Contrary to authors’ expectations, the major maintenance process did not result either in any decrease in losses or increase in efficiencies for the majority of the gas turbine components. However, the major maintenance influenced positively the gas turbine combustion chambers (reduction in losses and increase in the combustion chambers efficiency). After the major maintenance, the overall process efficiency decreased from 43.796% to 41.319% due to a significant decrease in the air mass flow rate and to an increase in the fuel mass flow rate in combustion chambers. A decrease in the gas turbine produced cumulative and useful power after a major maintenance also increased the specific fuel consumption.

In this paper multilayer perceptron (MLP) approach to condition-based maintenance of combined diesel-electric and gas (CODLAG) marine propulsion system is presented. By using data available in UCI, online machine learning repository, MLPs for prediction of gas turbine (GT) and GT compressor decay state coefficients are designed. Aforementioned MLPs are trained and tested by using 11 934 samples, of which 9 548 samples are used for training and 2 386 samples are used testing. In the case of GT decay state coefficient prediction, the lowest mean relative error of 0.622 % is achieved if MLP with one hidden layer of 50 artificial neurons (AN) designed with Tanh activation function is utilized. This configuration achieves the best results if it is trained by using L-BFGS solver. In the case of GT compressor decay state coefficient, the best results are achieved if MLP is designed with four hidden layers of 100, 50, 50 and 20 ANs, respectively. This configuration is designed by using Logistic sigmoid activation function. The lowest mean relative error of 1.094 % is achieved if MLP is trained by using L-BFGS solver.

One of the challenges of maritime affairs is automatic object recognition from aerial imagery. This can be achieved by utilizing a Convolutional Neural Network (CNN) based algorithm. For purposes of these research a dataset of 5608 marine object images is collected by using Google satellite imagery and Google Image Search. The dataset is divided in two main classes ("Vessels" and "Other objects") and each class is divided into four sub-classes ("Vessels" sub-classes are "Cargo ships", "Cruise ships", "War ships" and "Boats", while "Other objects" sub-classes are "Waves", "Marine animals", “Garbage patches" and "Oil spills"). For recognition of marine objects, an algorithm constructed with three CNNs is proposed. The first CNN for classification on the main classes achieves accuracy of 92.37 %. The CNN used for vessels recognition achieves accuracies of 94.12 % for cargo ships recognition, 98.82 % for cruise ships recognition, 97.64 % for war ships recognition and 95.29 % for boats recognition. The CNN used for recognition of other objects achieves accuracies of 88.56 % for waves and marine animals recognition, 96.92 % for garbage patches recognition and 89.21 % for oil spills recognition. This research has shown that CNN is appropriate artificial intelligence (AI) method for marine object recognition from aerial imagery.

The paper presents an exergy analysis of marine steam turbine labyrinth (gland) seals - an inevitable component of any marine steam turbine cylinder, in three different operating regimes. Throughout labyrinth seals, steam specific enthalpy can be considered as a constant because the results obtained by this assumption do not deviate significantly from the results of complex numerical models. Changes in labyrinth seals exergy efficiency and specific exergy destruction are reverse proportional. The analyzed labyrinth seals have high exergy efficiencies in each observed operating regime at the ambient temperature of 298 K (above 92%), what indicates seals proper operation. An increase in the ambient temperature resulted with a decrease in labyrinth seals exergy efficiency, but even at the highest observed ambient temperature of 318 K, seals exergy efficiency did not fall below 90.5% in each observed operating regime.

This paper describes the implementation of magnetic bearings and their active control system. The working principle of the special combined axial and radial magnetic bearings and their implementation is explained. The overall laboratory system consists of two magnetic bearings with their housings, the suspended shaft and control system with sensors and power electronics. Sensor electronics are used to measure the position of the shaft within a bearing and to determine the appropriate bearing current with outer position control loops and nested current control loops. The motor control board featuring the TMS320F28335 DSP and LDC1000 proximity sensors is used for control and power electronics. In order to implement the control system the X2C tool is used. This is an open source model based development and code generation tool embedded in the Scilab/Xcos environment. The sensor data acquisition time is optimized to increase the speed of the control system.

Research on the implementation of inductive proximity sensors in active magnetic bearings is performed in this paper. LDC1000 digital sensors manufactured by Texas Instruments are used since they claim to provide precision measured in microns and simple acquisition of the output data. The Sensor chip is connected with the sensing coil which is placed inside the magnetic bearing, close to electromagnets. This newly developed system created a platform for research in signal quality and its resistance to electrical noise. The controller used in this application was Texas Instruments TMS320F28335 digital signal processor (DSP). Software for data acquisition from LDC1000 was written and implemented in DSP using Code Composer Studio development environment. Two different output data are acquired and processed: proximity data and frequency count. Graphs presented in this paper show different resistance to electrical noise. A conclusion derived from this research can be applied in an industry where electromagnetic noise exists, together with the need for precise distance measurement.

Zadnjih je godina vidljiv ulazak High-Performance Computing u klaster tipove robnih servera sustava. U isto vrijeme svjedočimo računarstvu u oblacima kao dokazanoj, stabilnoj i pouzdanoj IT paradigmi koja rabi virtualizirane klastere robnih servera kao ključne komponente. U radu se daje analiza svojstava kod primjene direktne prirodne u odnosu na virtualnu HPC MPI aplikaciju na području HPC aplikacija opće uporabe virtualizacijom InfiniBanda via IpoIB na nivou virtualnog stroja na VMware ESXi kao i KVM hypervisora uz podršku aktivne migracije spremne na proizvodnju. Aplikacije za razne male i srednje HPC klastere usmjerene su na korisnike, imajući u vidu potpunu virtualizaciju IT infrastrukture povezujući infrastrukturu opće namjene s HPC IT infrastrukturom. Aktivna migracija je ključna komponenta u okviru virtualiziranog okruženja budući da omogućuje dinamičku virtualnu IT infrastrukturu. Stoga se djelovanje svojstava aktivne migracije procjenjuje kroz arhitekturu s više platformi na slučaju proaktivne uporabe greška-tolerancija., Over recent years we have seen a constant evolution of High-Performance Computing moving into commodity servers cluster types of systems. At the same time we have seen Cloud Computing being a proven, stable and reliable IT paradigm, utilising virtualized clusters of commodity servers as key components. This paper presents a face-to-face native-to-virtual HPC MPI application performance analysis on a range of general-use HPC applications by virtualizing InfiniBand via IPoIB at the guest virtual machine level on VMware ESXi as well as KVM hypervisor with production ready live migration support. Applications running on various small and medium HPC clusters are targeted to address users, considering full IT infrastructure virtualization by merging general purpose with HPC IT infrastructure. Live migration is a key component within a virtualized environment as it enables dynamic virtual IT infrastructure, therefore live migration performance impact is evaluated through a pluggable multi-platform architecture on a proactive fault-tolerance use case.

U posljednje je vrijeme zamjetan veliki interes HPC (High Performance Computing) zajednice prema Cloud Computing paradigmi, odnosno računalstvu u oblacima. Mnoge su prednosti izvođenja HPC aplikacija u računalnom oblaku, od kojih su najvažnije bolje korištenje računalnih resursa, učinkovita naplata njihovog korištenja, te dinamička i bez prekida u radu aplikacije preraspodjela računalnih resursa. Bez obzira na očite prednosti Cloud Computing okruženja za HPC, trenutni omjer korištenja HPC aplikacija u računalnim oblacima u odnosu na one koje se izvode u tradicionalnom HPC okruženju još je uvijek vrlo malen. Neki od razloga za takvo stanje su očiti, dok kod drugih nije tako. Tako na primjer tradicionalni proizvođači HPC opreme i aplikacija još uvijek pokušavaju iskoristiti njihova ulaganja u postojeća rješenja na način koji pogoduje tradicionalnom načinu rada HPC-a. Nadalje, iako je prisutan snažan razvoj virtualizacijske tehnologije, još uvijek postoje otvorena pitanja, poput skaliranja HPC aplikacija u virtualnom okruženju, te suradnje između fizičkih i virtualnih resursa. Tu se također nameće i pitanje podrške standardnih HPC alata za uspostavu virtualiziranih okolina. Na kraju, ali ne i manje važno, pojava heterogenog HPC računalstva koje se zasniva na kombinaciji standardnih procesora (CPU) i onih specijalizirane namjene (GPU procesori i akceleratori) nameće izazovna pitanja, poput skalabilnosti, učinkovitog Linpack mjerenja performansi, te razvoja HPC aplikacija u heterogenom CPU/GPU okolišu. Uz sve to, unatoč tome što mnogi proizvođači HPC opreme i aplikacija govore o tome kako imaju potpuno funkcionalna HPC rješenja u računalnim oblacima, postoji potreba da se osigura odgovore i na dodatna pitanja: Jesu li potpuno ispunjena obećanja mogućnosti virtualizacije svih komponenti sustava, ne samo onih virtualnih, nego i fizičkih? Koje vrste računalnih oblaka su podržane: privatne, javne ili njihova kombinacija, tj. hibridni oblaci? Kako dobro pojedina HPC aplikacij, In the last years a strong interest of the HPC (High Performance Computing) community raised towards cloud computing. There are many apparent benefits of doing HPC in a cloud, the most important of them being better utilization of computational resources, efficient charge back of used resources and applications, on-demand and dynamic reallocation of computational resources between users and HPC applications, and automatic bursting of additional resources when needed. Surprisingly, the amount of HPC cloud solutions related to standard solutions is still negligible. Some of the reasons for the current situation are evident, some not so. For example, traditional HPC vendors are still trying to exploit their current investment as much as possible, thus favoring traditional way of doing HPC. The next, although virtualization techniques are developing in ever increasing rate, there are still open questions, like scaling of HPC applications in virtual environment, collaboration between physical and virtual resources, and support of the standard HPC tools for virtualized environments. At last but not the least, the advent of GPU computing has raised difficult questions even in the traditional and well developed HPC segment, like scalability, development of HPC GPU based applications. In addition, because many HPC vendors are speaking about already having a fully functional HPC cloud solution, there is a need to provide answers to the following questions: Are they fulfilling virtualization promises, both physical and virtual? Which type of clouds do they support: private, public or both, i.e. hybrid clouds? How well HPC applications scale on their cloud solutions? This paper is an overview of the current HPC cloud solutions, for sure not complete and solely a view of authors, however intended to be a helpful compass for someone trying to shift from standard HPC to large computations in cloud environments.

Članak se bavi eksperimentalnim ispitivanjem rezne sposobnosti izmjenjivih reznih umetaka. Testirano je 11 tipova izmjenjivih reznih umetaka od pet različitih proizvođača. Ispitivani su rezni umetci bili istog oblika a razlikovali su se naročito po materijalu i vrsti površinskog sloja. Glavni je cilj bio izabrati odgovarajući test za određivanje rezne sposobnosti zamjenjivih reznih umetaka te kreirati takav postupak ispitivanja koji bi omogućio usporedbu i ocjenu rezne sposobnosti odabranih reznih umetaka. Nakon ispitivanja izvršeni su potrebni proračuni za usporedbu rezne sposobnosti i bilježenje trošenja alata korištenjem mikroskopa. Uspoređena je i prodiskutirana rezna sposobnost., The article deals with experimental testing of the cutting ability of exchangeable cutting inserts. Eleven types of exchangeable cutting inserts from five different manufacturers were tested. The tested cutting inserts were of the same shape and were different especially in material and coating types. The main aim was both to select a suitable test for determination of the cutting ability of exchangeable cutting inserts and to design such testing procedure that could make it possible to compare and evaluate the cutting ability of the selected cutting inserts. After the testing the necessary calculations were performed for comparison of the cutting abilities and recording the tool-wear using a microscope. The cutting ability was compared and discussed.

U usporedbi s konvencionalnim metodama obrade, proizvodnja složenih dijelova hladnim istiskivanjem često nudi tehničke i ekonomske prednosti. U ranoj fazi industrijske primjene hladnim istiskivanjem su se uglavnom izgrađivali aksijalno simetrični dijelovi. U toku daljnjeg razvoja, tehnologija hladnog istiskivanja primijenjuje se i za izradu složenijih ne-aksijalno simetričnih oblika. Samim time povećan je značaj i proširena područja primjene protusmjernog deformiranja i u drugim oblastima industrije, posebice u automobilskoj industriji. U radu su prikazana istraživanja tehnologije protusmjernog deformiranja ne-aksijalno simetričnih dijelova, s akcentom na izradu dijelova pomoću žigova kvadratnog poprečnog presjeka. Ispitivane su vrijednosti deformacione sile i pritisaka na čelo žiga, kao i naponsko-deformacijsko stanje u dva karakteristična presjeka radnog komada. Proces je analiziran metodom konačnih elemenata (MKE) i eksperimentalno. Za eksperimentalna istraživanja konstruirani su i izrađeni specijalni alati., Manufacturing of complex parts by cold extrusion often offers technical and economical advantages when compared with the conventional production methods. In the initial stage of industrial application of cold extrusion mainly axisymmetric components were produced. In the course of further development, application of cold extrusion has been extended to more complex and non-axisymmetric shapes. These have increased the significance and broaden the application fields of cold extrusion within other production technologies, especially in the automobile industry. Current paper elaborates non axisymmetric cold extrusion. Focus of the work is placed on the backward extrusion by the punch with square cross section. In this process load – stroke has been determined as well as punch pressure, and strain and stress distribution in two characteristic different workpiece cross sections. The process has been analyzed by FE method and experimentally. For the experimental investigation special tooling has been designed and made.

Trenje je neizbježna pojava u svim postupciima plastičnog deformiranja metala i ima negativan utjecaj na parametre procesa, radni vijek alata, kao i kvalitetu radnog komada. U hladnom prostornom deformiranju veliki pritisci djeluju između alata i radnog komada. Ovi pritisci mogu biti limitirajući faktor prilikom primjene tehnologija deformiranja. Smanjenjem kontaktnog trenja, smanjuju se i kontaktni pritisci. Poznavanje parametara trenja (faktora trenja µ i/ili omjera trenja m) je od velikog značaja prilikom proračuna parametara procesa (npr. sila ili energija) i odabira odgovarajućeg sredstva za podmazivanje, kao i numeričku simulaciju obrade. Postoje mnogobrojne metode za određivanje parametara trenja u obradama deformiranjem. Neke od ovih metoda opisane su u ovom radu. Također je predstavljen i novi model za određivanje parametara trenja u obradama deformiranjem. U radu je izvršeno i mjerenje sile trenja u procesu protusmjernog istiskivanja za različite vrste maziva., Friction occurs in all metal forming operations and, in general, it has a negative impact on process parameters, die life, as well as workpiece quality. In cold metal forming processes high interface pressures between die and material take place. These pressures can be the limiting factor of application of cold forming. By reducing interfacial friction, contact pressures can be reduced too. Knowledge of friction amount (factor of friction µ and/or friction ratio m) is essential for calculation of the main process parameters (load, energy), for choosing a proper lubricant but also for numerical modelling of forming operations. There exist a number of experimental methods to determine friction in metal forming processes. Current paper deals with the possibilities to evaluate friction. It analyses and assesses a number of existing friction models in cold metal operations. Proposal of a new friction model for cold metal forming operations is presented and discussed. Measurement of friction force in backward extrusion for different lubrication conditions was also preformed.

Metal forming technology has been applied to effective production of metallic components in various fields. One of the newer metal forming techniques is so called „joining by forming“. With this method two or more separate parts are joined together using plastic deformation. Self-piercing riveting, clinching and coin and medal production are some of the joining by forming operations, which have been successfully applied in industry. The present study is bound to a process of joining of two bimetallic axisymmetric components from various materials by upsetting bimetallic components in a closed die. The external component is a ring and the central part is a cylinder – like element. By upsetting those two elements in closed die, inseparable workpiece has been obtained. Therefore, two different cases (geometries) have been investigated. In the first case the simple ring and profiled inner cylinder are combined while in the second case vice versa combination is applied. Also, forming load, material flow and filling of joint section were analyzed. The process has been investigated experimentally. In short, current investigation has yielded considerable insight into the process of joining two metallic components.

The purpose of this research is to focus on the solution of the resource sizing problem for the real production system. The resource sizing problem is tackled by usage of the mathematical makespan minimization method and comparative simulations of the two different production models. Only two production model simulations were realized due to specificities of the handled production system. Simulation results were presented, compared and discussed. This paper presents an approach which allows studying the heavy machine sizing problem realistically, through the use of simulation tools. The approach used would be of great help in making a decision on whether two new heavy machines should be introduced into an existing production system or not.

The purpose of this research is to focus on the solution of the resource sizing problem for the real production system. The resource sizing problem is tackled by usage of the mathematical makespan minimization method and comparative simulations of the two different production models. Only two production model simulations were realized due to specificities of the handled production system. Simulation results were presented, compared and discussed. This paper presents an approach which allows studying the heavy machine sizing problem realistically, through the use of simulation tools. The approach used would be of great help in making a decision on whether two new heavy machines should be introduced into an existing production system or not.

Metal forming technology has been applied to effective production of metallic components in various fields. One of the newer metal forming techniques is so called „joining by forming“. With this method two or more separate parts are joined together using plastic deformation. Self-piercing riveting, clinching and coin and medal production are some of the joining by forming operations, which have been successfully applied in industry. The present study is bound to a process of joining of two bimetallic axisymmetric components from various materials by upsetting bimetallic components in a closed die. The external component is a ring and the central part is a cylinder – like element. By upsetting those two elements in closed die, inseparable workpiece has been obtained. Therefore, two different cases (geometries) have been investigated. In the first case the simple ring and profiled inner cylinder are combined while in the second case vice versa combination is applied. Also, forming load, material flow and filling of joint section were analyzed. The process has been investigated experimentally. In short, current investigation has yielded considerable insight into the process of joining two metallic components.

Kompleksnost proizvodnog okruženja, kao i složenost rješavanih zadataka unutar njega, kontinuirano i eksponencijalno raste. U mnogim industrijskim scenarijima tradicionalni centralizirani i hijerarhijski pristup nije adekvatan kada se primijeni za upravljanje složenim proizvodnim procesima, te može dovesti do neadekvatnog i sporog odaziva modeliranog sustava prema sustavima s visokim stupnjem kompleksnosti i praktičnih zahtjeva za robusnosti i rekonfigurabilnosti. Stoga smo u ovom radu primijenili moderne metode modeliranja kompleksnih proizvodnih sustava zasnovane na heterarhijskom modelu (biološki proizvodni sustavi – BMS), s primjenom multiagentske tehnologije za opis pojedinih komponenti sustava. Poslije definirane metodologije razvijen je simulacijski model fleksibilne obradne ćelije i simuliran je njezin rad u stvarnim proizvodnim uvjetima., The complexity of the production environment and the sophistication of solved tasks from within the same, continuously and exponentially increases. In today’s industrial scenarios, the traditional centralized and hierarchical approach is not adequate for the control and scheduling of the complex production processes. This can lead to an inadequate and slow response of the system to systems with a high level of complexity and practical demands for robustness and reconfiguration. Due to that reason, in this paper we have introduced modern methods for the modelling of complex production systems that are based on the heterachical model (Biological Manufacturing Systems–BMS), with the implementation of multi-agent technology for description of some of the system entities. After this methodology has been defined, a simulation model of a flexible manufacturing cell was developed and its work was simulated in real production conditions.

Kompleksnost proizvodnog okruženja, kao i složenost rješavanih zadataka unutar njega, kontinuirano i eksponencijalno raste. U mnogim industrijskim scenarijima tradicionalni centralizirani i hijerarhijski pristup nije adekvatan kada se primijeni za upravljanje složenim proizvodnim procesima, te može dovesti do neadekvatnog i sporog odaziva modeliranog sustava prema sustavima s visokim stupnjem kompleksnosti i praktičnih zahtjeva za robusnosti i rekonfigurabilnosti. Stoga smo u ovom radu primijenili moderne metode modeliranja kompleksnih proizvodnih sustava zasnovane na heterarhijskom modelu (biološki proizvodni sustavi – BMS), s primjenom multiagentske tehnologije za opis pojedinih komponenti sustava. Poslije definirane metodologije razvijen je simulacijski model fleksibilne obradne ćelije i simuliran je njezin rad u stvarnim proizvodnim uvjetima., The complexity of the production environment and the sophistication of solved tasks from within the same, continuously and exponentially increases. In today’s industrial scenarios, the traditional centralized and hierarchical approach is not adequate for the control and scheduling of the complex production processes. This can lead to an inadequate and slow response of the system to systems with a high level of complexity and practical demands for robustness and reconfiguration. Due to that reason, in this paper we have introduced modern methods for the modelling of complex production systems that are based on the heterachical model (Biological Manufacturing Systems–BMS), with the implementation of multi-agent technology for description of some of the system entities. After this methodology has been defined, a simulation model of a flexible manufacturing cell was developed and its work was simulated in real production conditions.

U članku je dana analiza između nekolicine modernih inteligentnih proizvodnih sustava poput: biološkog, holonskog (HMS), fraktalnog (FrMS) te ubiquitous proizvodne koncepcije. Na brzorastućim tržištima ključna je fleksibilnost proizvodnih sustava kako bi se što učinkovitije moglo odgovoriti na promjene nastale u tržišnom okruženju. Dan je detaljan prikaz karakteristika navedenih sustava, te su navedene njihove komparativne prednosti i nedostaci. Prikazani su utjecajni čimbenici i dane su smjernice za daljnji nastavak istraživanja., This article gives an analysis of a few modern intelligent production systems: biological, holon (HMS), fractal (FrMS) and ubiquitous production concepts. The key to success in modern fast growing systems is a flexibility of the production system in order to respond quickly to changes within the market. A detail representation of the characteristics as well as the advantages and disadvantages for these systems is given. Influential factors have been listed as well as guidelines for further research.

U članku je dana analiza između nekolicine modernih inteligentnih proizvodnih sustava poput: biološkog, holonskog (HMS), fraktalnog (FrMS) te ubiquitous proizvodne koncepcije. Na brzorastućim tržištima ključna je fleksibilnost proizvodnih sustava kako bi se što učinkovitije moglo odgovoriti na promjene nastale u tržišnom okruženju. Dan je detaljan prikaz karakteristika navedenih sustava, te su navedene njihove komparativne prednosti i nedostaci. Prikazani su utjecajni čimbenici i dane su smjernice za daljnji nastavak istraživanja., This article gives an analysis of a few modern intelligent production systems: biological, holon (HMS), fractal (FrMS) and ubiquitous production concepts. The key to success in modern fast growing systems is a flexibility of the production system in order to respond quickly to changes within the market. A detail representation of the characteristics as well as the advantages and disadvantages for these systems is given. Influential factors have been listed as well as guidelines for further research.