Wyniki wyszukiwania - BazTech

1

Machine Learning-Driven RUL Prediction and Uncertainty Quantification for Ball Screw Drives in a Cloud-Ready Maintenance Framework

Bott Alexander, Liu Bolin, Puchta Alexander, Fleischer Juergen

Journal of Machine Engineering

|

2024

|

Vol. 24, No. 3

17--31

EN

In today's rapidly evolving industrial landscape, efficient predictive maintenance solutions are essential for minimizing downtime and enhancing productivity. This research introduces an adaptive cloud-based model pipeline for predicting the Remaining Useful Life (RUL) of machine components, specifically ball screws. The pipeline integrates local pre-processing, edge computing, and cloud-based adaptive model training, ensuring data privacy and reducing data transmission volumes. The system classifies wear states using various machine learning models and predicts RUL through regression analysis, incorporating uncertainty quantification for robust maintenance scheduling. The experimental setup includes accelerated degradation of ball screws, with data collected via a three-dimensional accelerometer. Feature extraction and data augmentation techniques are employed to enhance prediction accuracy. Random Forest and Gradient Boosting models demonstrate superior performance, with Random Forest selected for its robustness and uncertainty quantification capabilities. Empirical results indicate high prediction accuracy, with Random Forest achieving up to 91% accuracy in Phase 2. This cloud-ready predictive maintenance framework leverages scalable cloud infrastructure for efficient data processing and real-time updates, offering a practical solution for industrial applications. The proposed approach significantly advances the adoption of digital business models within the manufacturing industry, providing a reliable and efficient tool for predictive maintenance.

2

Model‐free sliding mode control for a nonlinear teleoperation system with actuator dynamics

Abdelwaheb Henni Mansour, Abderrahmane Kacimi, AEK Belaidi

Journal of Automation Mobile Robotics and Intelligent Systems

|

2023

|

Vol. 17, No. 1

69--77

EN

Teleoperation robotic systems control, which enables humans to perform activities in remote situations, has become an extremely challenging field in recent decades. In this paper, a Model Free Proportional‐Derivative Slid‐ ing Mode Controller (MFPDSMC) is devoted to the syn‐ chronization problem of teleoperation systems subject to actuator dynamics, time‐varying delay, model uncer‐ tainty, and input interaction forces. For the first time, the teleoperation model used in this study combines actuator dynamics and manipulator models into a single equation, which improves model accuracy and brings it closer to the actual system than in prior studies. Further, the proposed control approach, called Free, involves the simple mea‐ surement of inputs and outputs to enhance the system’s performance without relying on any knowledge from the mathematical model. In addition, our strategy includes a Sliding Mode term with the MFPD term to increase system stability and attain excellent performance against external disturbances. Finally, using the Lyapunov func‐ tion under specified conditions, asymptotic stability is established, and simulation results are compared and provided to demonstrate the efficacy of the proposed strategy.

3

Hands-on MPC tuning for industrial applications

Ionescu C., Copot D.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2019

|

Vol. 67, nr 5

925--945

EN

This paper proposes a practical tuning of closed loops with model based predictive control. The data assumed to be known from the process is the result of the bump test commonly applied in industry and known in engineering as step response data. A simplified context is assumed such that no prior know-how is required from the plant operator. The relevance of this assumption is very realistic in the context of first time users, both for industrial operators and as educational competence of first hand student training. A first order plus dead time is approximated and the controller parameters immediately follow by heuristic rules. Analysis has been performed in simulation on representative dynamics with guidelines for the various types of processes. Three single-input-single-output experimental setups have been used with no expert users available in different locations – both educational and industrial – these setups are representative for practical cases: a variable time delay dominant system, a non-minimum phase system and an open loop unstable system. Furthermore, in a multivariable control context, a train of separation columns has been tested for control in simulation, followed by experimental tests on a laboratory system with similar dynamics, i.e. a sextuple coupled water tank system. The results indicate the proposed methodology is suitable for hands-on tuning of predictive control loops with some limitations on performance and multivariable process control.

4

Ocena niepewności modeli nośności elementów żelbetowych

Woliński S.

Budownictwo i Architektura

|

2013

|

Vol. 12, nr 1

203--210

PL

W pracy przedstawiono analizę i sposoby oceny niepewności modelowej nośności elementów żelbetowych. Sposób pierwszy opiera się na normowej procedurze kalibracji modeli nośności. Sposób drugi jest autorską propozycją oszacowania niepewności modelowej z wykorzystaniem dostępnych wyników badań doświadczalnych dotyczących zmienności nośności. Zgodnie z koncepcją metody częściowych współczynników jako miarę niepewności modelowej przyjęto wartość odpowiedniego współczynnika częściowego. Rozważania i propozycje przedstawione w pracy zilustrowano przykładem oceny niepewności modelu nośności na ścinanie strefy przypodporowej belki żelbetowej.

EN

The paper presents the analysis and evaluation methods of uncertainty in the resistance models of reinforced concrete members. The first method is based on a standardized procedure for the calibration of resistance models given in the Eurocode. The second method is an original proposal for the evaluation of uncertainty in the resistance models using the available test data on resistance variation. According to the concept of semi-probabilistic partial factors design, the partial factor resulting from model uncertainty, separated from the factor of resistance, represents a measure of model uncertainty. The approach to model uncertainty evaluation is demonstrated with the use two different sample models for shear capacity assessment of reinforced concrete beams.

5

Robust control of depth of anesthesia based on H∞ design

Caiado D. V., Lemos J. M., Costa B. A.

Archives of Control Sciences

|

2013

|

Vol. 23, no. 1

41--59

EN

This paper presents a case study on the design of a robust controller for the depth of anesthesia (DoA) induced by the drug propofol. This process is represented by a linear model together with a non-parametric uncertainty description that is evaluated using a patient model bank with 20 patients undergoing sedation. By using H∞ methods, the controller is aimed to comply with robust stability and performance specifications for the class of patient models considered. A minimization problem of sensitivity and complementary sensitivity is made to design the controller. The controller that results from this procedure is approximated by a controller with a lower order, that in turn is redesigned in discrete time for computer control application. The resulting controller is evaluated in simulations using a realistic nonlinear model of DoA.

6

Odporna detekcja uszkodzeń przy zastosowaniu addytywnego modelu regresji

Łabęda-Grudziak Z.

Pomiary Automatyka Kontrola

|

2011

|

R. 57, nr 7

774-778

PL

W artykule przedstawiono wykorzystanie addytywnego modelu regresji oraz statystycznych technik eksploracji danych do konstrukcji układu detekcji uszkodzeń odpornej na zakłócenia i niepewność modelu, a następnie do oceny wrażliwości modelu na występowanie poszczególnych uszkodzeń. Do uzyskania właściwości odporności, niepewność otrzymanego modelu wyznaczana jest poprzez zastosowanie techniki modelowania błędu modelu addytywnego. Przedstawione rozwiązanie zostało przetestowane dla przykładowego zaworu regulacyjnego na podstawie danych laboratoryjnych próbkowanych na stanowisku regulacji poziomu wody w zbiorniku walczakowym

EN

The detection of faults in engineering systems is of great practical significance. The detection performance of the diagnostic technique is characterized by important and quantifiable benchmarks, like the fault sensitivity and the reaction speed. Also its robustness, i.e., the ability of the technique to operate in the presence of noise, disturbances and modelling errors, is affected by the design of a detection algorithm. This paper develops a new approach to the design of robust fault detection systems via an additive model and knowledge discovery data. To achieve robustness, an uncertainty associated with the additive model is also taken into account. The model error modelling is used to deal with noise corrupting the data and unmodelled dynamics. The backfitting algorithm with nonparametric smoothing techniques has been used for estimation of the additive model. The modelling results as well as the fault detection procedures are presented. The proposed approach is tested on an example of a control valve for measurement tracks in the boiler laboratory setup in order to demonstrate the sensitivity of faults.

7

Design of centralized PI/PID controllers for processes with parameter uncertainty and interactions

Malwatkar G. M., Waghmare L. M.

Systems Science

|

2010

|

Vol. 36, no 2

39-50

EN

In this paper, design of centralized PI/PID controller for multi-variable, highly interactive processes is proposed. Interactive processes are always difficult to control due to coupling in inputs-outputs, and inaccurate choice of input-output pairing affects the performance, complexity, and cost of the control system. Practical difficulty in the tuning of centralized/decentralized PI/PID controller is raised due to methods based on trial and error. In this work, design rule and tuning procedure of the controller for multi-variable process with delay time are discussed without decomposing process model. The success of controller does not fully depend on the input-output pairing due to off-diagonal elements. The resultant controller can work with satisfactory performance even if the primary process parameters change significantly. Multi-variable processes with major interaction characteristics and time delays are employed to demonstrate the simplicity and strength of the method proposed.

8

Supervisory predictive control and on-line set-point optimization

Tatjewski P.

International Journal of Applied Mathematics and Computer Science

|

2010

|

Vol. 20, no 3

483-495

EN

The subject of this paper is to discuss selected effective known and novel structures for advanced process control and optimization. The role and techniques of model-based predictive control (MPC) in a supervisory (advanced) control layer are first shortly discussed. The emphasis is put on algorithm efficiency for nonlinear processes and on treating uncertainty in process models, with two solutions presented: the structure of nonlinear prediction and successive linearizations for nonlinear control, and a novel algorithm based on fast model selection to cope with process uncertainty. Issues of cooperation between MPC algorithms and on-line steady-state set-point optimization are next discussed, including integrated approaches. Finally, a recently developed two-purpose supervisory predictive set-point optimizer is discussed, designed to perform simultaneously two goals: economic optimization and constraints handling for the underlying unconstrained direct controllers.

9

Projektowanie odpornych układów diagnostyki w oparciu o modelowanie niepewnosci modelu

Patan K.

Pomiary Automatyka Kontrola

|

2005

|

R. 51, nr 9 bis

106--108

PL

Artykuł przedstawia proces projektowania odpornego układu diagnostyki dla reaktora krakingu katalitycznego, przy zastosowaniu sztucznych sieci neuronowych. Identyfikacja rozważanego procesu jest przeprowadzana przy użyciu rekurencyjnych sieci neuronowych. Do uzyskania właściwości odporności, niepewność otrzymanego modelu wyznaczana jest poprzez zastosowanie techniki modelowania błędu modelu. W artykule zaproponowano neuronową wersję tej metody. Przedstawione rozwiązanie zostało przetestowane na przykładzie procesu krakingu katalitycznego.

EN

The paper presents designing a robust fault diagnosis system for a catalytic cracking process using aitificial neural networks. Identification of the considered process is carried out by using recurrent neural networks. To achieve a robust fault diagnosis system, an uncertainty associated with the model is also taken into account. Neural version of the Model Error Modelling is used to deal with two main uncertainty sources: unmodelled dynamics and noise corrupting the data. The proposed approach is tested on the example of catalytic cracking converter at the nominal operation conditions as well as in the case of faults.

10

Projektowanie odpornych obserwatorów stanu dla dyskretnych w czasie systemów nieliniowych

Witczak M.

Pomiary Automatyka Kontrola

|

2005

|

R. 51, nr 9 bis

60--62

PL

Przedmiotem niniejszej pracy jest projektowanie obserwatorów stanu dla dyskretnych w czasie systemów nieliniowych. W szczególności, stosując metodę Lapunowa wyznacza się warunek zbieżności obserwatora. Bazując na otrzymanych rezultatach proponuje się procedurę projektowania obserwatora, a następnie przedstawia się jej programową implementację w środowisku MATLAB®. W pracy pokazuje się również jak zastosować powyższe rozwiązania do projektowania odpornych obserwatorów stanu o nieznanym wejściu.

EN

The paper deals with the problem of designing observers for discrete-time non-linear systems. In particular, with the use of the Lyapunov method, the convergence criterion of the observer is developed. Based on the achieved results a design procedure is proposed and implemented with the MATLAB® programming environment. The paper also shows how to use the above approach to design robust unknown input observers.

11

Detekcja uszkodzeń odporna na niepewność modelu neuronowego

Mrugalski M., Korbicz J.

Pomiary Automatyka Kontrola

|

2005

|

R. 51, nr 9 bis

103--105

PL

W pracy przedstawiony zosta) problem detekcji uszkodzeń odpornej na niepewność modelu neuronowego. Na przykładzie sieci neuronowej GMDH przedstawiono przyczyny powstawania niepewności modelu otrzymywanego podczas identyfikacji. Zaprezentowana metoda wyznaczania niepewności modelu w postaci przedziału ufności wyjścia systemu umożliwiła opracowanie odpornego układu detekcji uszkodzeń w oparciu o technikę adaptacyjnych progów decyzyjnych.

EN

In the paper the problem of the robust fault detection under the neural model uncertainty was presented and widely discussal. In particular, the causes of forming the GMDH neural model uncertainty obtained via system identification were shown. The presented method of confidence estimation of GMDH neural networks in (he form of the system output uncertainty interval enables development of the robust fault detection scheme on the basis of the adaptive threshold technique.

12

Truss models of RC corbels verified by experimental tests

Nagrodzka-Godycka K.

Archives of Civil Engineering

|

2005

|

Vol. 51, nr 4

515-531

EN

The work presents selected truss models used in design of RC corbels, which served as grounds for the calculating truss model proposals adopted in the European and the Polish standards. The work provides a detailed discussion of individual theoretical models, along with the standard recommendations. It contains a comparison of corbel load capacities obtained from experimental tests — carried out by various researchers — with the capacities calculated by means of the models in question.

PL

Przedstawiono wybrane modele kratownicowe krótkich wsporników żelbetowych oraz obliczone na podstawie tych modeli nośności graniczne wsporników na tle wyników uzyskanych w badaniach eksperymentalnych. Przedstawiono także weryfikację metody ścinania-tarcia wg wytycznych normy amerykańskiej. Dokonano obliczeń częściowych współczynników bezpieczeństwa uwzględniających niepewność rozważanych modeli. Stwierdzono, że modele kratownicowe uwzględniające zmienną wysokość strefy ściskanej betonu i zmienne ramię sił wewnętrznych dobrze odwzorowują stan nośności krótkich wsporników. Powyższy wniosek został potwierdzony uzyskanymi wartościami częściowego współczynnika bezpieczeństwa.

13

Variable- and Fixed-Structure Augmented Interacting Multiple-Model Algorithms for Manoeuvring Ship Tracting Based on New Ship Models

Semerdjiev E., Mihaylova L.

International Journal of Applied Mathematics and Computer Science

|

2000

|

Vol. 10, no 3

591-604

EN

Real-world tracking applications are related to a number of difficulties caused by the presence of different kinds of uncertainty, e.g. unknown or incompletely known system models and statistics of random processes or abrupt changes in the system modes of functioning. These problems are especially complicated in the marine navigation practice, where the commonly-used simple models of rectilinear or curvilinear target motions are not adequate for highly non-linear dynamics of the manoeuvring ship motion. A solution to these problems is to derive more suitable descriptions of real ship dynamics and to design adaptive estimation algorithms. After an analysis of basic hydrodynamic models, new ship models are derived in the paper. They are implemented in two versions of the Interacting Multiple Model (IMM) algorithm which has become very popular recently. The first one is a standard IMM version based on fixed model structures (FS's). They represent various modes of ship motion, distinguished by their rates of turns. The same rate of turn is additionally adjusted in the proposed new augmented versions of the IMM (AIMM) algorithm by using FS's and variable structures (VS's) of adaptive models estimating the current change in the system control parameters. Monte Carlo simulation experiments indicate that the VS AIMM algorithm outperforms the FS AIMM and FS IMM ones with respect to both accuracy and adaptability.

14

Robust Control of Nonlinear Time-Delay Systems

Antoniades C., Christofides P. D.

International Journal of Applied Mathematics and Computer Science

|

1999

|

Vol. 9, no 4

811-837

EN

This paper focuses on single-input single-output nonlinear differential difference equation (DDE) systems with uncertain variables. For such systems, a general methodology is developed for the synthesis of robust nonlinear state feedback controllers that guarantee boundedness of the states and ensure that the ultimate discrepancy between the output and the external reference input in the closed-loop system can be made arbitrarily small by an appropriate choice of controller parameters. The controllers are synthesized by using a novel combination of geometric and Lyapunov-based techniques and enforce the above properties in the closed-loop system independently of the size of the state delay. The proposed control method is successfully applied to a fluidized catalytic cracking unit with a time-varying uncertain variable and is shown to outperform a proportional integral (PI) controller, a nonlinear controller that does not account for the uncertainty, and anonlinear controller that does not account for the state delays.