Catastrophes in Optimal Control

1999 ◽  
Vol 121 (4) ◽  
pp. 577-582 ◽  
Author(s):  
Osita D. I. Nwokah ◽  
E. Borzova ◽  
Gemunu S. Happawana ◽  
Dare´ Afolabi
Keyword(s):  
Closed Loop ◽  
Dynamic Instability ◽  
System Optimization ◽  
Characteristic Polynomials ◽  
Optimal Solutions ◽  
Quality Metric ◽  
Highly Sensitive ◽  
Parameter Variations ◽  
Catastrophic Loss ◽  
Differential Parameter

System optimization over a parameter space produces optimal solutions which lie on the bifurcation set of the ambient space. As such, the optimality (quality) metric (as a function of the parameters) is highly sensitive to the parameters, to the point of inducing instability for differential parameter variations. Singularities in this function diffeomorphically induce corresponding degenerate singularities in the optimal closed-loop characteristic polynomials, which serves as a signature for potential catastrophic loss of quality that is most easily exhibited by the resulting dynamic instability. In this paper, we examine the loss of quality in H∞ and related optimal systems via these diffeomorphic degenerate closed-loop poles.

LEXICOGRAPHIC PROBLEMS OF CONVEX OPTIMIZATION: SOLVABILITY AND OPTIMALITY CONDITIONS, CUTTING PLANE METHOD

2021 ◽  
Vol 1 ◽  
pp. 30-40
Author(s):  
Natalia V. Semenova ◽  
◽  
Maria M. Lomaga ◽  
Viktor V. Semenov ◽  
◽  
...  
Keyword(s):  
System Optimization ◽  
Cutting Plane ◽  
Recession Cone ◽  
Cutting Plane Method ◽  
Optimal Solutions ◽  
Feasible Set ◽  
Plane Method ◽  
Multicriteria Problems ◽  
Lexicographic Optimization ◽  
Feasible Solutions

The lexicographic approach for solving multicriteria problems consists in the strict ordering of criteria concerning relative importance and allows to obtain optimization of more important criterion due to any losses of all another, to the criteria of less importance. Hence, a lot of problems including the ones of com­plex system optimization, of stochastic programming under risk, of dynamic character, etc. may be presented in the form of lexicographic problems of opti­mization. We have revealed conditions of existence and optimality of solutions of multicriteria problems of lexicographic optimization with an unbounded convex set of feasible solutions on the basis of applying properties of a recession cone of a convex feasible set, the cone which puts in order lexicographically a feasible set with respect to optimization criteria and local tent built at the boundary points of the feasible set. The properties of lexicographic optimal solutions are described. Received conditions and properties may be successfully used while developing algorithms for finding optimal solutions of mentioned problems of lexicographic optimization. A method of finding lexicographic of optimal solutions of convex lexicographic problems is built and grounded on the basis of ideas of method of linearization and Kelley cutting-plane method.

Optimal Variability and Complexity

2013 ◽  
pp. 328-351 ◽  
Author(s):  
Joshua L Haworth ◽  
Srikant Vallabhajosula ◽  
George Tzetzis ◽  
Nicholas Stergiou
Keyword(s):  
Learning Strategies ◽  
Human Movement ◽  
System Optimization ◽  
Discovery Learning ◽  
Optimal Solutions ◽  
Systems Perspective ◽  
History Of ◽  
Self Discovery ◽  
Set Up ◽  
Management Staff

Management seeks to provoke system optimization throughout ever changing environmental and internal conditions. Typically, perturbations to stable organizations are unpredictable and difficult to define, except from within a chaos perspective. How should management staff set up their workforce to be best responsive to these changes? It is proposed that a dynamic systems theoretical approach to the organization of the management system would foster the ideal scenario. This approach lends well to the inclusion of discovery learning strategies that promote the valuable use of optimal variability in the exploration and self-discovery of optimal solutions to existent and novel problems. In this text, the authors walk the reader through a brief history of the development of the systems perspective on human movement optimization. Next, they extend the related discoveries to applications within management systems. It is hoped that a new appreciation for complexity and beneficial aspects of variability is conveyed.

Applying RBF Neural Networks and Genetic Algorithms to Nonlinear System Optimization

2012 ◽  
Vol 605-607 ◽  
pp. 2457-2460 ◽  
Author(s):  
Hong Fa Wang ◽  
Xin Ai Xu
Keyword(s):  
Neural Network ◽  
Nonlinear System ◽  
Rbf Neural Network ◽  
System Optimization ◽  
Rbf Neural Networks ◽  
Optimal Solutions ◽  
Base Function ◽  
The Neural Network ◽  
Engineering Practices ◽  
Radial Base Function

Nonlinear system optimization is always an issue that needs to be considered in engineering practices and management. In order to obtain optimal solutions without analysis formulas to nonlinear systems, we first construct a radial-base-function (RBF) neural network using the newrb() function in MALTAB 7.0, then train the neural network according to input and output, and finally obtain the solution using a genetic algorithm. Simulated experimental results show that the proposed algorithm is able to achieve optimal solutions with a relatively fast speed of convergence.

scholarly journals Circular RNAs in Hepatocellular Carcinoma: Emerging Functions to Clinical Significances

2021 ◽  
Vol 11 ◽  
Author(s):  
Yucheng Zhang ◽  
Yali Wang
Keyword(s):  
Hepatocellular Carcinoma ◽  
Closed Loop ◽  
Early Stage ◽  
Circular Rnas ◽  
High Morbidity ◽  
Future Perspectives ◽  
Diagnostic Biomarkers ◽  
Diagnosis And Prognosis ◽  
Highly Sensitive ◽  
Non Coding Rnas

Hepatocellular carcinoma (HCC) is the most common primary cancer of the liver and carries high morbidity and mortality. Diagnosing HCC at an early stage is challenging. Therefore, finding new, highly sensitive and specific diagnostic biomarkers for the diagnosis and prognosis of HCC patients is extremely important. Circular RNAs (circRNAs) are a class of non-coding RNAs with covalently closed loop structures. They are characterized by remarkable stability, long half-life, abundance and evolutionary conservation. Recent studies have shown that many circRNAs are expressed aberrantly in HCC tissues and have important regulatory roles during the development and progression of HCC. Hence, circRNAs are promising biomarkers for the diagnosis and prognosis of HCC. This review: (i) summarizes the biogenesis, categories, and functions of circRNAs; (ii) focuses on current progress of dysregulated expression of circRNAs in HCC with regard to regulation of the tumor hallmarks, “stemness” of cancer cells, and immunotherapy; (iii) highlights circRNAs as potential biomarkers and therapeutic targets for HCC; and (iv) discusses some of the challenges, questions and future perspectives of circRNAs research in HCC.

A Bulk Micromachined Si-on-Glass Tunneling Accelerometer

2007 ◽  
Author(s):  
Min Miao ◽  
Qifang Hu ◽  
Yilong Hao ◽  
Haifeng Dong ◽  
Haixia Zhang
Keyword(s):  
Closed Loop ◽  
Proof Mass ◽  
Anodic Bonding ◽  
Analysis And Design ◽  
Icp Etching ◽  
Highly Sensitive ◽  
Out Of Plane ◽  
Input Range ◽  
Pyrex 7740 ◽  
Si Wafer

A bulk micromachined tunneling accelerometer on Pyrex 7740 substrate is reported in this paper, which is intended for the applications in highly sensitive inertia measurements, such as those in microgravity environments and self-contained navigation. The tunneling tip is defined by an isotropic wet etching followed by a maskless wet thermal oxidation for the sharpening of the tip. Unlike the process ever reported by other facilities, an ICP etching on the side of the Si wafer with the tip is utilized to partially define the suspension and the proof mass before the anodic bonding of the Si wafer with the glass substrate, and an addition maskless ICP etching is used to release the whole movable structure after the bonding. Fabricated samples have displayed the effectiveness of the process proposed, which is relatively simple and may guarantee the yield of mass production. The theoretical analysis and design of the closed loop architecture of the device are demonstrated. Capable of sensing out-of-plane acceleration, this device has demonstrated a high resolution of 0.015mg/rtHz (@ 1∼100Hz) and a nonlinearity of less than 1% over ±1g input range.

Intelligent Fire-Fighting Robot

2014 ◽  
Vol 722 ◽  
pp. 204-208
Author(s):  
He Ming Cheng ◽  
Tian Cun Yang ◽  
Liang Li
Keyword(s):  
Robot Control ◽  
Closed Loop ◽  
System Optimization ◽  
Fire Fighting ◽  
Motor Speed ◽  
Loop Control ◽  
Precise Control ◽  
Object Based ◽  
And Control ◽  
Fire Fighting Robot

In this paper, intelligent fire-fighting robot as the research object, based on Mitsubishi FX2N series PLC, the establishment of intelligent fire-fighting robot control model and control system optimization, so that the fire-fighting robot among the various modules can be smooth and stable work together. Using L298 chip control two motors, the speed achieved by controlling the robot flexible steering, and using PLC produce different wave PWM stepless duty to ensure the smooth operation of the robot. Meanwhile, the detection of motor speed using encoder feedback to PLC will constitute a closed-loop control to achieve precise control and adjustment of robots routes. Intelligent fire-fighting robot each module to work together more quickly and safely find and reach the fire source and fight it.

scholarly journals Existence of solutions and solving method of lexicographic problem of convex optimization with the linear criteria functions

2020 ◽  
pp. 19-27
Author(s):  
N.V. Semenova ◽  
◽  
M.M. Lomaha ◽  
V.V. Semenov ◽  
◽  
...  
Keyword(s):  
Existence Of Solutions ◽  
Optimization Problems ◽  
System Optimization ◽  
Linear Functions ◽  
Recession Cone ◽  
Cutting Plane Method ◽  
Optimal Solutions ◽  
Lexicographic Optimization ◽  
Feasible Solutions ◽  
Significant Class

Among vector problems, the lexicographic ones constitute a broad significant class of problems of optimization. Lexicographic ordering is applied to establish rules of subordination and priority. Hence, a lot of problems including the ones of complex system optimization, of stochastic programming under a risk, of the dynamic character, etc. may be presented in the form of lexicographic problems of optimization. We have revealed the conditions of existence of solutions of multicriteria of lexicographic optimization problems with an unbounded set of feasible solutions on the basis of applying the properties of a recession cone of a con vex feasible set, the cone which puts it in order lexicographically with respect to optimization criteria. The obtained conditions may be successfully used while developing algorithms for finding the optimal solutions of the mentioned problems of lexicographic optimization. A method of finding the optimal solutions of convex lexico graphic problems with the linear functions of criteria is built and grounded on the basis of ideas of the method of linearization and the Kelley cutting plane method.

Sign in / Sign up

Export Citation Format

Share Document