M.A. Brdyś

Adaptive Prediction of Stock Exchange Indices by State Space Wavelet Networks

Adaptive Prediction of Stock Exchange Indices by State Space Wavelet Networks The paper considers the forecasting of the Warsaw Stock Exchange price index WIG20 by applying a state space wavelet network model of the index price. The approach can be applied to the development of tools for predicting changes of other economic indicators, especially stock exchange indices. The paper presents a general state space wavelet network model and the underlying principles. The model is applied to produce one session ahead and five sessions ahead adaptive predictors of the WIG20 index prices. The predictors are validated based on real data records to produce promising results. The state space wavelet network model may also be used as a forecasting tool for a wide range of economic and non-economic indicators, such as goods and row materials prices, electricity/fuel consumption or currency exchange rates.

An Algorithm for Steady-State Optimizing Dual Control of Uncertain Plants

Abstract A new optimizing control algorithm is developed within the framework of the Integrated System Optimization and Parameter Estimation (ISOPE) technique. The ISOPE is an on-line technique for controller set-point optimization (steady-state control) taking into account the uncertainty in the steady-state process model. The new algorithm is the first ISOPE algorithm with dual control effect, i.e., the current control signal is generated to satisfy the main control goal and, simultaneously, to provide sufficient information for future identification action. This leads to substantially increased efficiency of the technique, confirmed by the simulation results.

IMPLEMENTATION OF INTEGRATED CONTROL IN DRINKING WATER DISTRIBUTION SYSTEMS–IT SYSTEM PROPOSAL

Abstract Implementation of integrated control algorithms requires suitable hardware and software platforms. Proposed solution must allow realizing control and monitoring tasks, while ensuring high reliability and security of processed data. Additionally, the software and hardware solutions must be immune to outside influence. This paper presents an approach to implementation of control systems in drinking water distribution system. Presented approach takes into consideration safety and reliability of the control core.

Improving Optimality in Multilayer Control Systems by Tighter Constraint Control and Supervision

Abstract A multilayer control of industrial processes is revisited in the paper. The multilayer decomposition of complex control problem into a sequence of simpler control tasks leads to appealing and reliable control system. However, the price to be paid versus centralised dynamic solution (if such is achievable) is a loss on optimality. A problem of reducing the loss of optimality is undertaken in the paper. Several ways of approaching the problem are proposed and investigated. They consist in integrating operations of the control layers, developing new class of predictive controllers for the follow-up control together with their fuzzy-logic supervision and in supervisory actions against abrupt faults and failures of the process operation by employing artificial intelligence techniques. Ability to maintain robustly the process operating point closer to constraint boundaries, thus reducing the constraint margins needed to guarantee operational feasibility is a key property of the new predictive controllers.

Recurrent Networks for Nonlinear Adaptive Control

Abstract An adaptive control technique for nonlinear plants with unmeasurable state is presented. It is based on a recurrent neural network employed as a dynamical model of the plant. Using this dynamical model a feedback linearizing control is computed and applied to the plant. Parameters of the model are updated on line to allow for partially unknown and time varying plant. Stability of the algorithm is proved fo~ the case of constant reference output and some further insights into convergence Issues for the general case of tracking problem are provided. Performance of the proposed control method is illustrated in simulations.

EVENT DRIVEN MPC FOR NETWORKED CONTROL SYSTEMS

Abstract Because of variable delays and stochastic data packets loss networked control systems require suitable algorithms to ensure stability of the control system and guarantee desired control performance. This paper presents the idea of an event driven approach with MPC controller. In opposition to network compensation, where standard regulators are used, the presented solution integrates network with plant. A MPC based controller is applied to extended plant that consists of plant and communication network. In order to reduce network influence on control quality, control signal computation and control signal application are done, when a data receive event occurs.

CONDITION MONITORING USING PCA BASED METHOD AND APPLICATION TO WASTEWATER TREATMENT PLANT OPERATION

Abstract Monitoring of multidimensional, sophisticated processes still can be a challenge. The paper presents tools to handle such cases, which are based on the Principal Component Analysis with the adaptive enhancement. A novel PCA algorithm is derived with the adaptive supervisor. It is applied to monitoring operation of wastewater treatment plant and tested by simulation based on real data records.

Optimizing Control of Uncertain Plants with Feedback Controlled Output Constraints

Abstract Optimizing control of technological processes within a multilayer structre is considered, for the case with important constraints on certain process outputs. A multilayer structure with an additional feedback follow-up controller responsible for keeping the output constraints satisfied both in steady states and during transient processes, in spite of inaccurate models used by optimizing control algorithms, is introduced. A new formulation of the Integrated System Optimization and Parameter Estimation (ISOPE) method, suitable for the structure, is derived. Implementation aspects are then discussed and simulation results for a distillation column example process are given

SSWN learning algorithms with application to WWTP

Abstract State Space Wavelet Network is a specific neural network with non trivial structure. Such a structure implies problems with the network training. In SSWN, during the training process, weights and wavelons parameters are adapted. This paper presents algorithms and methods for optimising the SSWN parameters. During researches Hybrid Distributed Evolutionary Algorithm has been developed and Extended initialisation strategy was applied. Finally it was found that usage of Hybrid Distributed Evolutionary Algorithm with extended initialisation significantly reduce the time of SSWN learning.

MODELLING OF WASTEWATER TREATMENT PLANT FOR MONITORING AND CONTROL PURPOSES BY STATE - SPACE WAVELET NETWORKS

Most of industrial processes are nonlinear, not stationary, and dynamical with at least few different time scales in their internal dynamics and hardly measured states. A biological wastewater treatment plant falls into this category. The paper considers modelling such processes for monitorning and control purposes by using State - Space Wavelet Neural Networks (SSWN). The modelling method is illustrated based on bioreactors of the wastewater treatment plant. The learning algorithms and basis function (multidimensional wavelets) are also proposed. The simulation results based on real data record are presented.