Pastora Vega

State space neural network. Properties and application

In this paper, a specific neural network based model for the identification of non-linear systems is proposed. This neural network structure is able to identify a state space non-linear model of the plant. The use of the state space representation presents several advantages that must be taken into account. One of the most important advantages is that the resulting neural model can be easily linearized around different operating points, allowing application of classical stability theorems from the linear systems domain to this class of neural networks. In this way, some useful theoretical results for neural modelling and identification have been obtained and presented in the paper. In this paper, several stability theorems and practical implementation issues are addressed. Examples are also presented which show the training capability of the neural network and the validity of the theory presented.

Integrated design and control of chemical processes – Part I: Revision and classification

Abstract This work presents a comprehensive classification of the different methods and procedures for integrated synthesis, design and control of chemical processes, based on a wide revision of recent literature. This classification fundamentally differentiates between “projecting methods”, where controllability is monitored during the process design to predict the trade-offs between design and control, and the “integrated-optimization methods” which solve the process design and the control-systems design at once within an optimization framework. The latter are revised categorizing them according to the methods to evaluate controllability and other related properties, the scope of the design problem, the treatment of uncertainties and perturbations, and finally, the type the optimization problem formulation and the methods for its resolution.

Integration of set point optimization techniques into nonlinear MPC for improving the operation of WWTPs

Abstract Optimization and control strategies are necessary to keep wastewater treatment plants (WWTPs) operating in the best possible conditions, maximizing effluent quality with the minimum consumption of energy. In this work, a benchmarking of different hierarchical control structures for WWTPs that combines static and dynamic real time optimization (RTO) and nonlinear model predictive control (NMPC) is presented. The objective is to evaluate the enhancement of the operation in terms of economics and effluent quality that can be achieved when introducing NMPC technologies in the distinct levels of the multilayer structure. Three multilayer hierarchical structures are evaluated and compared for the N-Removal process considering the short term and long term operation in a rain weather scenario. A reduction in the operation costs of approximately 20% with a satisfactory compromise to effluent quality is achieved with the application of these control schemes.

A Transductive Neuro-Fuzzy Controller: Application to a Drilling Process

Recently, new neuro-fuzzy inference algorithms have been developed to deal with the time-varying behavior and uncertainty of many complex systems. This paper presents the design and application of a novel transductive neuro-fuzzy inference method to control force in a high-performance drilling process. The main goal is to study, analyze, and verify the behavior of a transductive neuro-fuzzy inference system for controlling this complex process, specifically addressing the dynamic modeling, computational efficiency, and viability of the real-time application of this algorithm as well as assessing the topology of the neuro-fuzzy system (e.g., number of clusters, number of rules). A transductive reasoning method is used to create local neuro-fuzzy models for each input/output data set in a case study. The direct and inverse dynamics of a complex process are modeled using this strategy. The synergies among fuzzy, neural, and transductive strategies are then exploited to deal with process complexity and uncertainty through the application of the neuro-fuzzy models within an internal model control (IMC) scheme. A comparative study is made of the adaptive neuro-fuzzy inference system (ANFIS) and the suggested method inspired in a transductive neuro-fuzzy inference strategy. The two neuro-fuzzy strategies are evaluated in a real drilling force control problem. The experimental results demonstrated that the transductive neuro-fuzzy control system provides a good transient response (without overshoot) and better error-based performance indices than the ANFIS-based control system. In particular, the IMC system based on a transductive neuro-fuzzy inference approach reduces the influence of the increase in cutting force that occurs as the drill depth increases, reducing the risk of rapid tool wear and catastrophic tool breakage.

State-space neural network for modelling, prediction and control

Abstract The state-space neural network paradigm is a neural model suitable for various applications in the field of control engineering. In this paper, it is shown how this neural model can be applied to three common tasks in control engineering: modelling of a diffusion section in a sugar industry, prediction in a wastewater plant, and neural model-based predictive control in a sugar factory. Results from these applications show the applicability and good performance of this neural model that, together with the theoretical results available for this type of neural model, gives an excellent alternative to classical linear models in cases where the non-linearity of the system requires it.

Metodología para la Obtención de Modelos Semifísicos de Base Fenomenológica Aplicada a una Sulfitadora de Jugo de Caña de Azúcar

This work presents the Phenomenological Based Semi-physical Model (PBSM) as a useful tool for the design, control and optimization of chemical and bio-technology processes. First, a detailed description of a methodology to obtain a PBSM based on the knowledge of the basic principles of the process: mass, energy and momentum conservation, as well as gradient principle for obtaining model constitutive equations. The described methodology is applied to the clarification stage of a sugar cane refining plant. Specifically, the sulfitation tower is modelled step by step. The model is simulated and validated by comparing its behaviour with data taken from a real sulfitation tower.

Integrated design and control of chemical processes : Part II: an illustrative example

Abstract In this paper, several methodologies of integrated design are proposed and applied to the design of wastewater treatment plants and their control system, focusing on the activated sludge process, within a novel multiobjective framework. The scope of the problem considers both fixed plant layout and plant structure selection by defining a simple superstructure. The control strategy chosen is a linear Model Predictive Controller (MPC) with terminal penalty. The evaluation of the controllability has been performed using norm based indexes, and the robustness conditions for different uncertainty sources have been considered, in the frequency and time domains. The optimization strategies used are based on the integration of stochastic and deterministic methods, as well as genetic algorithms. The presented methodologies and their application to wastewater treatment plants can be considered as an illustrative example in the universe of integrated design techniques presented in the Part I article of this series.

l1-Optimal regulation of a pH control plant

Abstract This paper presents an innovative approach to controller design, and applies it to the pH-control problem. The proposed approach uses a robust controller implemented with a full two-degree-of-freedom (2-DoF) structure. The first of “feedback” DoF was designed using a standard μ-synthesis approach. A procedure based on a novel variation of l1-optimal control theory is applied to design the second or “open loop” DoF. This design may be carried out completely independently of the first DoF, provided an appropriate factorization of the feedback controller is used. The approach is illustrated by the design and evaluation of a control system for a laboratory scale pH-control plant. The resulting controller was implemented and tested on the physical plant. It was found that it gave good performance over widely varying conditions. Specifically, the open loop properties of the μ-synthesis feedback controller acting on its own, were improved significantly by the use of a certain class of l1-optimal second DoF controller block. We conclude that the tradeoff between good command tracking and reasonable plant inputs given by a feedback controller can be improved by the use of the proposed approach, a novel variation on l1-optimization applied to the second DoF controller block.

Algorithmic synthesis and integrated design for activated sludge processes using genetic algorithms

Abstract This work presents an approach for the Synthesis and Integrated Design of an activated sludge process. The mathematical formulation translates a superstructure that contains all the design alternatives into a mixed-integer dynamical optimization problem with non-linear constraints. A real-coded genetic algorithm is proposed for the solution of such complex problem as an alternative to classical optimisation techniques. Thus, the process synthesis considering open-loop-dynamical-performance indexes and also the closed loop integrated design for the plant are carried out. The results are encouraging for future application of these techniques to solve process synthesis problems

Isolation of Multiplicative Faults in the Industrial Actuator Benchmark

Abstract This paper presents an approach to the problem of fault isolation applied to the actuator benchmark test based on multiplicative fault isolation with parity equations. One of the problems of this benchmark is relative to the isolation of a multiplicative fault from additive disturbances. The fact that a fault can be modelled as multiplicative is used to isolate it from additive disturbances. An optimization procedure gets the most suitable residuals for multiplicative fault isolation.