The study of neural estimator structure influence on the estimation quality of selected state variables of the complex mechanical part of electrical drive

2017 ◽  
Author(s):  
Dominik Luczak ◽  
Adrian Wojcik
Keyword(s):  
State Variables ◽  
Electrical Drive ◽  
Mechanical Part ◽  
Estimation Quality

scholarly journals Neural State Estimator for Complex Mechanical Part of Electrical Drive: Neural Network Size and Performance of State Estimation

2018 ◽  
Vol 3 (1) ◽  
pp. 205-216
Author(s):  
Dominik Łuczak ◽  
Adrian Wójcik
Keyword(s):  
Neural Network ◽  
Network Size ◽  
Feedforward Neural Network ◽  
State Variables ◽  
State Estimator ◽  
Electrical Drive ◽  
Mass System ◽  
Simulation Research ◽  
Mechanical Part ◽  
And Performance

Abstract This paper presents the results of simulation research of an off-line-trained, feedforward neural-network-based state estimator. The investigated system is the mechanical part of an electrical drive characterised by elastic coupling with a working machine, modelled as a dual-mass system. The aim of the research was to find a set of neural network structures giving useful and repeatable results of the estimation. The mechanical resonance frequency of the system has been adopted at the level of 9.3-10.3 Hz. The selected state variables of the mechanical system are load, speed and stiffness torque of the shaft.

Numerical Simulation Analysis of the Cold Bending Forming Technology Based on DEFORM

2012 ◽  
Vol 490-495 ◽  
pp. 3353-3357 ◽  
Author(s):  
Xiao Bing Xu ◽  
Hong Run Luo
Keyword(s):  
Simulation Analysis ◽  
Pipe Production ◽  
State Variables ◽  
Technical Parameters ◽  
Forming Technology ◽  
Cold Bending ◽  
Bent Pipe ◽  
Deformation Speed ◽  
3D Software

In this paper, the accurate numerical simulations using the DEFORM-3D software are carried out in terms of the cold bending pipe technology. Correspondingly, the changes of state variables are gained, such as the stress, the strain, the material deformation speed, the damage severity and so on. Meanwhile, the important technical parameters concerned with the quality of pipe bending formation like the speed of the punch press, the friction coefficient and the friction types are comprehensively analyzed. Thus, it can be realized in practice to control and optimize the quality of the bent pipe production

scholarly journals Hydraulic and transport parameter assessment using column infiltration experiments

2016 ◽  
Author(s):  
A. Younes ◽  
T. A. Mara ◽  
M. Fahs ◽  
O. Grunenberger ◽  
Ph. Ackerer
Keyword(s):  
Pressure Head ◽  
Solute Concentration ◽  
Richards Equation ◽  
Transport Parameter ◽  
State Variables ◽  
Transport Parameters ◽  
Uncertainty Range ◽  
Infiltration Experiment ◽  
Statistical Calibration

Abstract. In the present work, we study the quality of the statistical calibration of hydraulic and transport soil properties using an infiltration experiment in which, over a given period, tracer-contaminated water is injected into a laboratory column filled with a homogeneous soil. The numerical model is based on the Richards' equation for solving water flow and the advection-dispersion equation for solving solute transport. Several state variables (e.g., water content, solute concentration, pressure head) are measured during the experiment. Statistical calibration of the computer model is then carried out for different data sets and injection scenarios with the DREAM(ZS) Markov Chain Monte Carlo sampler. The results show that the injection period has a significant effect on the quality of the estimation, in particular, the posterior uncertainty range. The hydraulic and transport parameters of the investigated soil can be estimated from the infiltration experiment using the concentration and cumulative outflow, which are measured non-intrusively. A significant improvement of the identifiability of the parameters is observed when the pressure data from measurements taken inside the column are also considered in the inversion.

scholarly journals Graphical Convergence of Subgradients in Nonconvex Optimization and Learning

2021 ◽  
Author(s):  
Damek Davis ◽  
Dmitriy Drusvyatskiy
Keyword(s):  
Generalized Linear Model ◽  
Optimization Problem ◽  
Full Generality ◽  
Regression Problem ◽  
Smooth Maps ◽  
Sample Average Approximations ◽  
Sample Average ◽  
Stochastic Optimization Problem ◽  
Estimation Quality

We investigate the stochastic optimization problem of minimizing population risk, where the loss defining the risk is assumed to be weakly convex. Compositions of Lipschitz convex functions with smooth maps are the primary examples of such losses. We analyze the estimation quality of such nonsmooth and nonconvex problems by their sample average approximations. Our main results establish dimension-dependent rates on subgradient estimation in full generality and dimension-independent rates when the loss is a generalized linear model. As an application of the developed techniques, we analyze the nonsmooth landscape of a robust nonlinear regression problem.

Use the Differential Filters in the Control System of Pneumatic Muscle

2015 ◽  
Vol 816 ◽  
pp. 224-233
Author(s):  
Leszek Cedro
Keyword(s):  
Control Process ◽  
Quantization Noise ◽  
State Variables ◽  
Real Time Control ◽  
Pneumatic Muscle ◽  
Linear Dynamic Systems ◽  
Quality Of Control ◽  
Time Control ◽  
Derivatives Of

The paper discusses the use of differential filters in control algorithms. The filters are designed to determine the derivatives of the input signal and eliminate measuring and quantization noise. The differential filters improved the quality of control, with the results being better than those obtained with the classic Finite Difference Method (FDM). The primary purpose of the study was to employ the differential filters in a real-time control algorithm, which requires appropriate derivatives. The control process involved applying a method of aggregation of state variables, based on signal derivatives, which can be used for non-linear dynamic systems. The experiments were conducted on a test stand with a pneumatic muscle acting as the plant to be controlled.

scholarly journals A New Network for Particle Filtering of Multivariable Nonlinear Objects †

Energies ◽  
2020 ◽  
Vol 13 (6) ◽  
pp. 1355
Author(s):  
Piotr Kozierski ◽  
Jacek Michalski ◽  
Joanna Zietkiewicz ◽  
Marek Retinger and Wojciech Retinger ◽  
Wojciech Giernacki
Keyword(s):  
Sample Size ◽  
Particle Filtering ◽  
Mean Squared Error ◽  
Effective Sample Size ◽  
State Variables ◽  
Markovian Jump System ◽  
Transition Functions ◽  
Highly Nonlinear ◽  
Multivariable Nonlinear Systems ◽  
Estimation Quality

In this paper, a new object in the form of a theoretical network is presented, which is useful as a benchmark for particle filtering algorithms designed for multivariable nonlinear systems (potentially linear, nonlinear, and even semi-Markovian jump system). The main goal of the paper is to propose an object that potentially can have similar to the power system grid properties, but with the number of state variables reduced twice (only one state variable for each node, while there are two in the case of power systems). Transition and measurement functions are proposed in the paper, and two types of transition functions are considered: dependent on one or many state variables. In addition, 10 types of measurements are proposed both for branch and nodal cases. The experiments are performed for 14 different, four-dimensional systems. Plants are both linear and highly nonlinear. The results include information about the state estimation quality (based on the mean squared error indicator) and the values of the effective sample size. It is observed how the higher effective sample size resulted in the better estimation quality in subsequent cases. It is also concluded that the very low number of significant particles is the main problem in particle filtering of multivariable systems, and this should be countered. A few potential solutions for the latter are also presented.

scholarly journals Water quality assessment of Loktak Lake, Northeast India using 2-D hydrodynamic modelling

2021 ◽  
Vol 3 (4) ◽  
Author(s):  
Eliza Khwairakpam ◽  
Rakesh Khosa ◽  
Ashvani Gosain ◽  
Arvind Nema
Keyword(s):  
Water Quality ◽  
Environmental Concern ◽  
Oxygen Demand ◽  
Water Quality Assessment ◽  
Northeast India ◽  
State Variables ◽  
Lake Water Quality ◽  
Loktak Lake ◽  
Physically Based

AbstractThe paper comprises of an application of a multi-faceted physically based two-dimensional (2D) hydrodynamic model to simulate the transport phenomena of Loktak Lake, including the water quality of Loktak Lake, for which there is consensus that it is deteriorating due to river discharge from sub-catchments carrying sewage loads, soil sediments and agricultural fertilizers, and therefore, has emerged as a serious environmental concern. Accordingly, the study attempts to understand the overall environmental quality of the Loktak system and in particular simulate Loktak Lake water quality (state) variables by coupling through MIKE 21 ECO Lab. The model simulated dissolved oxygen and biochemical oxygen demand throughout the lake.

Sign in / Sign up

Export Citation Format

Share Document