Local Model Networks For Modelling And Predictive Control Of Nonlinear Systems

Abstract

The paper deals with the problem of modelling and control using the Local Model Network (LMN). The idea is based on development of multiple local models for the whole operating range of the controlled process. The local models are then smoothly connected using the validity or weighting functions to provide a nonlinear global model of the plant. For saving the computational load, linear model is obtained by interpolating the parameters of local models at each sample instant and then used in Model Predictive Control (MPC) framework to calculate the future behaviour of the process. The supervisory program, based on a nonlinear global model, computes desired values of manipulated variables leading to minimum utility consumption. The approach is verified in a control of model of a heat exchanger.

Excised black hole spacetimes: Quasilocal horizon formalism applied to the Kerr example

“…We choose to encode the information in h ij T in the two scalar spectral potentials A and B presented in [52], and whose definitions are quickly recalled in the Appendix A. (A more extensive study shall be performed in [53]).…”

“…The differential systems require three boundary conditions on the excised surface, to be inverted (see [53]); we discuss them in Sec. IV, in a detailed description of the scheme.…”

“…Using the same method, we can write very similar expressions (that we do not explicitly give here) for the fields h rr and η, to be used as Robin boundary conditions applied to the gauge differential system (B6). The three boundary conditions are sufficient to invert the two gauge systems (B5-B6) [53], and reconstruct the whole h ij tensor from the tensor spherical harmonics components (see the Appendix B and [53] for details).…”

Control of a Simple Constrained MIMO System with Steady-state Optimization

“…The deviation decreases a steady-state control error and incorporating of supposed course of inputs u N,0 allows involving known unmanipulated inputs or disturbances (ambient temperature in our case). Supposed inputs can be also used for optimization of inputs values combination in case of MI+MO systems -similar way as in (Novák, 2009). The third part is the deviation of a desired and actual state at the end of horizon (terminal state).…”

A Modified Method for Radar Clutter Recognition Based on a Truncation Set

“…A typical airborne SAR system is capable of providing wide area search coverage (approximately 100 km2 area per minute) at medium 1.0m×1.0m resolution [1]. With the complexity of the radar backgrounds, different clutter models are required for different regions.…”