Search results

Filters

  • Journals
  • Authors
  • Keywords
  • Date
  • Type

Search results

Number of results: 83
items per page: 25 50 75
Sort by:

Abstract

In this work nine non-linear regression models were compared for sub-pixel impervious surface area mapping from Landsat images. The comparison was done in three study areas both for accuracy of imperviousness coverage evaluation in individual points in time and accuracy of imperviousness change assessment. The performance of individual machine learning algorithms (Cubist, Random Forest, stochastic gradient boosting of regression trees, k-nearest neighbors regression, random k-nearest neighbors regression, Multivariate Adaptive Regression Splines, averaged neural networks, and support vector machines with polynomial and radial kernels) was also compared with the performance of heterogeneous model ensembles constructed from the best models trained using particular techniques. The results proved that in case of sub-pixel evaluation the most accurate prediction of change may not necessarily be based on the most accurate individual assessments. When single methods are considered, based on obtained results Cubist algorithm may be advised for Landsat based mapping of imperviousness for single dates. However, Random Forest may be endorsed when the most reliable evaluation of imperviousness change is the primary goal. It gave lower accuracies for individual assessments, but better prediction of change due to more correlated errors of individual predictions. Heterogeneous model ensembles performed for individual time points assessments at least as well as the best individual models. In case of imperviousness change assessment the ensembles always outperformed single model approaches. It means that it is possible to improve the accuracy of sub-pixel imperviousness change assessment using ensembles of heterogeneous non-linear regression models.
Go to article

Abstract

In the article problems related to human labor and factors affecting the increasing use of industrial robots are discussed. Since human factors affect the production processes stability, robots are preferred to apply. The application of robots is characterized by higher performance and reliability comparing to human labor. The problem is how to determine the real difference in work efficiency between human operator and robot. The aim of the study is to develop a method that allows clearly definition of productivity growth associated with the replacement of human labor by industrial robots. Another aim of the paper is how to model robotized and manual operated workstation in a computer simulation software. Analysis of the productivity and reliability of the hydraulic press workstation operated by the human operator or an industrial robot, are presented. Simulation models have been developed taking into account the availability and reliability of the machine, operator and robot. We apply OEE (Overall Equipment Effectiveness) indicator to present how availability and reliability parameters influence over performance of the workstation, in the longer time. Simplified financial analysis is presented considering different labor costs in EU countries.
Go to article

Abstract

Touch-trigger probes for CNC milling machines usually use wireless communication in the radio or IR band. Additionally they enable triggering signal filtering in order to avoid false triggers of the probe. These solutions cause a delay in trigger signal transmission from the probe to the machine tool controller. This delay creates an additional pre-travel component, which is directly proportional to the measurement speed and which is the cause of a previously observed but not explained increase of the pre-travel with the increase of the measurement speed. In the paper, a method of testing the delay time of triggering signal is described, an example of delay time testing results is presented and the previous, unexplained results of other researchers are analysed in terms of signal transmission delay.
Go to article

Abstract

This paper presents an alternative approach to the sequential data classification, based on traditional machine learning algorithms (neural networks, principal component analysis, multivariate Gaussian anomaly detector) and finding the shortest path in a directed acyclic graph, using A* algorithm with a regression-based heuristic. Palm gestures were used as an example of the sequential data and a quadrocopter was the controlled object. The study includes creation of a conceptual model and practical construction of a system using the GPU to ensure the realtime operation. The results present the classification accuracy of chosen gestures and comparison of the computation time between the CPU- and GPU-based solutions.
Go to article

Abstract

This paper presents the design process and the results of a novel fall detector designed and constructed at the Faculty of Electronics, Military University of Technology. High sensitivity and low false alarm rates were achieved by using four independent sensors of varying physical quantities and sophisticated methods of signal processing and data mining. The manuscript discusses the study background, hardware development, alternative algorithms used for the sensor data processing and fusion for identification of the most efficient solution and the final results from testing the Android application on smartphone. The test was performed in four 6-h sessions (two sessions with female participants at the age of 28 years, one session with male participants aged 28 years and one involving a man at the age of 49 years) and showed correct detection of all 40 simulated falls with only three false alarms. Our results confirmed the sensitivity of the proposed algorithm to be 100% with a nominal false alarm rate (one false alarm per 8 h).
Go to article

Abstract

This paper proposes a soft sensing method of least squares support vector machine (LS-SVM) using temperature time series for gas flow measurements. A heater unit has been installed on the external wall of a pipeline to generate heat pulses. Dynamic temperature signals have been collected upstream of the heater unit. The temperature time series are the main secondary variables of soft sensing technique for estimating the flow rate. A LS-SVM model is proposed to construct a non-linear relation between the flow rate and temperature time series. To select its inputs, parameters of the measurement system are divided into three categories: blind, invalid and secondary variables. Then the kernel function parameters are optimized to improve estimation accuracy. The experiments have been conducted both in the single-pulse and multiple-pulse heating modes. The results show that estimations are acceptable.
Go to article

Abstract

This paper presents a multivariate regression predictive model of drift on the Coordinate Measuring Machine (CMM) behaviour. Evaluation tests on a CMM with a multi-step gauge were carried out following an extended version of an ISO evaluation procedure with a periodicity of at least once a week and during more than five months. This test procedure consists in measuring the gauge for several range volumes, spatial locations, distances and repetitions. The procedure, environment conditions and even the gauge have been kept invariables, so a massive measurement dataset was collected over time under high repeatability conditions. A multivariate regression analysis has revealed the main parameters that could affect the CMM behaviour, and then detected a trend on the CMM performance drift. A performance model that considers both the size of the measured dimension and the elapsed time since the last CMM calibration has been developed. This model can predict the CMM performance and measurement reliability over time and also can estimate an optimized period between calibrations for a specific measurement length or accuracy level.
Go to article

Abstract

The following paper presents the solution to the problem of searching the best shape - structural form of the bottoms and optimal dimensions of the main cylinder of the carding machine with consideration to the criterion of minimal deflection amplitude. The ANSYS package of the Finite Element Method has been used for the analysis. Polak-Ribery conjugate gradient method has been applied for searching the optimal solution, basing on the parametric model of the cylinder written with the use of Ansys Parametric Design Language. As a result of the performed analyses, reduction of maximum deflection value at approximately 80 percent has been obtained. Optimal cylinder dimensions enable application of a new textile technology - microfibre carding and improvement in the quality of traditional carding technology of woollen and wool-like fibres.
Go to article

Abstract

The touch trigger probe plays an important role in modern metrology because of its robust and compact design with crash protection, long life and excellent repeatability. Aside from coordinate measuring machines (CMM), touch trigger probes are used for workpiece location on a machine tool and for the accuracy assessment of the machine tools. As a result, the accuracy of the measurement is a matter of interest to the users. The touch trigger probe itself as well as the measuring surface, the machine tool, measuring environment etc. contribute to measurement inaccuracies. The paper presents the effect of surface irregularities, surface wetness due to cutting fluid and probing direction on probing accuracy on a machine tool.
Go to article

Abstract

In this paper, the author presents the possibility of using phase trajectory for detecting damage in an axial piston pump. The wear on main part of pump elements, such as the rotor and the valve plate, was investigated, and phase trajectories were determined based on vibration signal measured in three directions on the pump's body. In order to obtain a quantitative measure of the analyzed trajectory, the At_{p,i} parameter was introduced, and the relation between this parameter and the wear on the pump's parts was determined.
Go to article

Abstract

In multi-stage wire drawing machines productivity growth can be achieved at higher drawing speeds by preventing wire breakage during the process. One disadvantage of high-speed wire drawing is the requirement imposed by machine dynamics in terms of its stability and reliability during operation. Tensile forces in the wire must maintained by fast synchronization of all capstans speed. In this process, the displacement sensors play the main role in providing the control system with feedback information about the wire condition. In this study, the influences between the sensors and actuator driven capstans have been studied, and tuner roll concept of a wire drawing machine was experimentally investigated. To this aim, measurements were carried out on two drawing stages at different drawing speeds and obtained results were presented. These results clearly show the fast changes of the capstans speed and the angular displacements of the rollers that tighten the wire, which only confirms the high dynamics of the wire drawing machine.
Go to article

Abstract

A fault diagnostics system of three-phase induction motors was implemented. The implemented system was based on acoustic signals of three-phase induction motors. A feature extraction step was performed using SMOFS-20-EXPANDED (shortened method of frequencies selection-20-Expanded). A classification step was performed using 3 classifiers: LDA (Linear Discriminant Analysis), NBC (Naive Bayes Classifier), CT (Classification Tree). An analysis was carried out for incipient states of three-phase induction motors measured under laboratory conditions. The author measured and analysed the following states of motors: healthy motor, motor with one faulty rotor bar, motor with two faulty rotor bars, motor with faulty ring of squirrel-cage. Measured and analysed states were caused by natural degradation of parts of the machine. The efficiency of recognition of the analysed states was good. The proposed method of fault diagnostics can find application in protection of three-phase induction motors.
Go to article

Abstract

Affective computing studies and develops systems capable of detecting humans affects. The search for universal well-performing features for speech-based emotion recognition is ongoing. In this paper, a small set of features with support vector machines as the classifier is evaluated on Surrey Audio-Visual Expressed Emotion database, Berlin Database of Emotional Speech, Polish Emotional Speech database and Serbian emotional speech database. It is shown that a set of 87 features can offer results on-par with state-of-the-art, yielding 80.21, 88.6, 75.42 and 93.41% average emotion recognition rate, respectively. In addition, an experiment is conducted to explore the significance of gender in emotion recognition using random forests. Two models, trained on the first and second database, respectively, and four speakers were used to determine the effects. It is seen that the feature set used in this work performs well for both male and female speakers, yielding approximately 27% average emotion recognition in both models. In addition, the emotions for female speakers were recognized 18% of the time in the first model and 29% in the second. A similar effect is seen with male speakers: the first model yields 36%, the second 28% a verage emotion recognition rate. This illustrates the relationship between the constitution of training data and emotion recognition accuracy.
Go to article

Abstract

The paper presents an algorithm for the construction of an approximation of a highly nonlinear current-flux characteristic of a synchronous reluctance machine. Such an approximation is required in a Hamiltonian model of an electric machine and the constructed approximation is suited to be used in such a model. It employs a simplicial approximation based on irregular points sets in the spaces of currents and fluxes. The sets are constructed by the iterative insertion of new points. Initially the sets contain an arbitrarily small number of elements. The insertion is based on an approximation error calculation. Based on the sets containing possibly small number of elements, the proposed procedure leads to smooth and precise approximation. Due to the nonlinearity of the approximated characteristics, ambiguities can occur. A method for the triangulation refinement of the sets of currents and fluxes that eliminates them is also presented. In the paper, a reluctance machine model using the constructed approximation is described and compared with a model using the approximation based on regular sets.
Go to article

Abstract

The paper presents results of studies on linear synchronous motors controlled in CNC feed axes through an intelligent digital servodrive. The research includes a conceptual design of an open servodrive control system and identification of dynamic models of a test stand with an open CNC system. Advantages of robust control over the classic one are discussed. A hybrid predictive approach to robust control of milling machine X-Y table velocity is proposed and results of simulation tests are presented. Was prepared during the work for the Ministry of Science and Higher Education grant number N N502 336936, (acronym for this project is M.A.R.I.N.E. multivariable hybryd ModulAR motIon coNtrollEr), while its main purpose is the development of new rob ust position/velocity model-based control system, as well as to introduce the measurement of the actual state into the switching algorithm between the locally synthesized controllers. Such switching increases the overall robustness of the machine tool feed-drive module. The paper is the extended version of material proposed in [10].
Go to article

Abstract

In this paper a comparison of numerically determined and measured electromagnetically exited noise of an induction motor is presented. The calculations are accomplished using FEM for an example motor, which is a 290 kW inverter-fed asynchronous machine. The approach starts with the electromagnetic and mechanical consideration. The focus is set on acoustic considerations, which contain the 3D-FE-model and measurement setup in the sound chamber.
Go to article

Abstract

The field of mechanical manufacturing is becoming more and more demanding on machining accuracy. It is essential to monitor and compensate the deformation of structural parts of a heavy-duty machine tool. The deformation of the base of a heavy-duty machine tool is an important factor that affects machining accuracy. The base is statically indeterminate and complex in load. It is difficult to reconstruct deformation by traditional methods. A reconstruction algorithm for determining bending deformation of the base of a heavy-duty machine tool using inverse Finite Element Method (iFEM) is presented. The base is equivalent to a multi-span beam which is divided into beam elements with support points as nodes. The deflection polynomial order of each element is analysed. According to the boundary conditions, the deformation compatibility conditions and the strain data measured by Fiber Bragg Grating (FBG), the deflection polynomial coefficients of a beam element are determined. Using the coordinate transformation, the deflection equation of the base is obtained. Both numerical verification and experiment were carried out. The deflection obtained by the reconstruction algorithm using iFEM and the actual deflection measured by laser displacement sensors were compared. The accuracy of the reconstruction algorithm is verified.
Go to article

Abstract

Land surveyors, photogrammetrists, remote sensing engineers and professionals in the Earth sciences are often faced with the task of transferring coordinates from one geodetic datum into another to serve their desired purpose. The essence is to create compatibility between data related to different geodetic reference frames for geospatial applications. Strictly speaking, conventional techniques of conformal, affine and projective transformation models are mostly used to accomplish such task. With developing countries like Ghana where there is no immediate plans to establish geocentric datum and still rely on the astro-geodetic datums as it national mapping reference surface, there is the urgent need to explore the suitability of other transformation methods. In this study, an effort has been made to explore the proficiency of the Extreme Learning Machine (ELM) as a novel alternative coordinate transformation method. The proposed ELM approach was applied to data found in the Ghana geodetic reference network. The ELM transformation result has been analysed and compared with benchmark methods of backpropagation neural network (BPNN), radial basis function neural network (RBFNN), two-dimensional (2D) affine and 2D conformal. The overall study results indicate that the ELM can produce comparable transformation results to the widely used BPNN and RBFNN, but better than the 2D affine and 2D conformal. The results produced by ELM has demonstrated it as a promising tool for coordinate transformation in Ghana.
Go to article

Abstract

The paper analyses the distorted data of an electronic nose in recognizing the gasoline bio-based additives. Different tools of data mining, such as the methods of data clustering, principal component analysis, wavelet transformation, support vector machine and random forest of decision trees are applied. A special stress is put on the robustness of signal processing systems to the noise distorting the registered sensor signals. A special denoising procedure based on application of discrete wavelet transformation has been proposed. This procedure enables to reduce the error rate of recognition in a significant way. The numerical results of experiments devoted to the recognition of different blends of gasoline have shown the superiority of support vector machine in a noisy environment of measurement.
Go to article

Abstract

There exist numerous modelling techniques and representation methods for digital control algorithms, aimed to achieve required system or process parameters, e.g. precision of process modelling, control quality, fulfilling the time constrains, optimisation of consumption of system resources, or achieving a trade-off between number of parameters. This work illustrates usage of Finite State Machines (FSM) modelling technique to solve a control problem with parameterized external variables. The structure of this work comprises six elements. The FSM is presented in brief and discrete control algorithm modelling is discussed. The modelled object and control problem is described and variables are identified. The FSM model is presented and control algorithm is described. The parameterization problem is identified and addressed, and the implementation in PLC programming LAD language is presented. Finally, the conclusion is given and future work areas are identified.
Go to article

Abstract

The paper presents the problem of assessing the accuracy of reconstructing free-form surfaces in the CMM/CAD/CAM/CNC systems. The system structure comprises a coordinate measuring machine (CMM) PMM 12106 equipped with a contact scanning probe, a 3-axis Arrow 500 Vertical Machining Center, QUINDOS software and Catia software. For the purpose of surface digitalization, a radius correction algorithm was developed. The surface reconstructing errors for the presented system were assessed and analysed with respect to offset points. The accuracy assessment exhibit error values in the reconstruction of a free-form surface in a range of ± 0.02 mm, which, as it is shown by the analysis, result from a systematic error.
Go to article

Abstract

This paper presents a comparison of different techniques to capture nominal data for its use in later verification and kinematic parameter identification procedures for articulated arm coordinate measuring machines (AACMM). By using four different probing systems (passive spherical probe, active spherical probe, self-centering passive probe and self-centering active probe) the accuracy and repeatability of captured points has been evaluated by comparing these points to nominal points materialized by a ball-bar gauge distributed in several positions of the measurement volume. Then, by comparing these systems it is possible to characterize the influence of the force over the final results for each of the gauge and probing system configurations. The results with each of the systems studied show the advantages and original accuracy obtained by active probes, and thus their suitability in verification (active probes) and kinematic parameter identification (self-centering active probes) procedures.
Go to article

Abstract

Power electronic circuits (PECs) are prone to various failures, whose classification is of paramount importance. This paper presents a data-driven based fault diagnosis technique, which employs a support vector data description (SVDD) method to perform fault classification of PECs. In the presented method, fault signals (e.g. currents, voltages, etc.) are collected from accessible nodes of circuits, and then signal processing techniques (e.g. Fourier analysis, wavelet transform, etc.) are adopted to extract feature samples, which are subsequently used to perform offline machine learning. Finally, the SVDD classifier is used to implement fault classification task. However, in some cases, the conventional SVDD cannot achieve good classification performance, because this classifier may generate some so-called refusal areas (RAs), and in our design these RAs are resolved with the one-against-one support vector machine (SVM) classifier. The obtained experiment results from simulated and actual circuits demonstrate that the improved SVDD has a classification performance close to the conventional one-against-one SVM, and can be applied to fault classification of PECs in practice.
Go to article

This page uses 'cookies'. Learn more