Design and Optimization of a Neuro-Fuzzy System for the Control of an Electromechanical Plant

One characteristic of neuro-fuzzy systems is the possibility of incorporating preliminary information in their structure as well as being able to establish an initial configuration to carry out the training. In this regard, the strategy to establish the configuration of the fuzzy system is a relevant aspect. This document displays the design and implementation of a neuro-fuzzy controller based on Boolean relations to regulate the angular position in an electromechanical plant, composed by a motor coupled to inertia with friction (a widely studied plant that serves to show the control system design process). The structure of fuzzy systems based on Boolean relations considers the operation of sensors and actuators present in the control system. In this way, the initial configuration of fuzzy controller can be determined. In order to perform the optimization of the neuro-fuzzy controller, the continuous plant model is converted to discrete time to be included in the closed-loop controller training equations. For the design process, first the optimization of a Proportional Integral (PI) linear controller is carried out. Thus, linear controller parameters are employed to establish the structure and initial configuration of the neuro-fuzzy controller. The optimization process also includes weighting factors for error and control action in such a way that allows having different system responses. Considering the structure of the control system, the optimization algorithm (training algorithm) employed is dynamic back propagation. The results via simulations show that optimization is achieved in the linear and neuro-fuzzy controllers using different weighting values for the error signal and control action. It is also observed that the proposed control strategy allows disturbance rejection.

An Application of the Z-Transform in Extracting the Density Regulation from the Periodic Output of an Age-Structured Population Model

The application of the Z-transform, a manipulation tool from the discrete signal processing (DSP) toolbox, on an ecological model was motivated by the mathematical similarities between an age-structured fish population model with a non linear density regulation and a linear time invariant (LTI) control system. Both models include a switching mechanism in regulating stock/signal throughput in accordance with a given density limitation/set value and both models can be expressed in terms of a negative feedback loop difference equations (Getz & Haight,1989; °Astr¨om & Murray, 2008). In the fish model, the switching mechanism is a density regulated stock-recruitment (SR) function which models the strategies implemented by the population in keeping the vulnerable egg-larvaejuvenile densities within an environmental limitation thereof (Subbey et al, 2014). A switching mechanism is also present in control engineering, for example, in the mechanism associated with cruise control in cars which keeps traveling speed close to a chosen set value midst varying weather and road conditions (Antsaklis and Gao, 2005). In both cases, the choosing of the control action and the tuning of its parameters requires careful consideration to avoid failures such as incorrectly timed switching actions in a control plant (see Kuphaldt (2019)) and errors in estimating total allowable catch (TAC) in the fishing industry (see Borlestean et al (2015), Skagen et al (2013) and Taboadai and R. Anadn (2016)). The Z-transform has proven itself useful in tuning LTI controlmodels for a desired control action (see Orfanidis, (2010) and Smith, (1999)) and it is on this account that its application was extended to the ecological model in pursuit of a more efficient way of estimating SR parameters to simulate an already existing output. It was however found that it could not be used for parameter tuning but rather for the extraction of the SR component hidden in the output together with components resulting from the age structure itself. Such an extraction can greatly assist in the mathematical identification of the SR, reducing the complexity of its choosing as there are many different types used in the fishing industry such as the classic Beverton-Holt model, the Ricker model and Shepherd model (Myers, 2001; Iles, 1994; Shepherd, 1982). It can also be used to monitor changes in the SR over time which can indicate the presence of strategy evolution (Apaloo et al, 2009; Br¨annstr¨om et al, 2013). In 1998 Schoombie and Getz investigated the latter by subjecting the Shepherd SR to strategy optimization with regards to a parameter associated with population interventions in regulating recruitment throughput and it is because of this versatility that the Shepherd SR is chosen for the intended extraction. In true control style, Simulink, a graphic environment for designing control simulations, is used to visualize the production of the output as well as the extraction of the SR from it. This paper showcases the versatility of the Z transform and the possibilities and unexpected finds when applied to similar systems designed to regulate signals or, in this case, recruitment densities.

Data-Driven Model-Free Adaptive Control of Z-Source Inverters

The universal paradigm shift towards green energy has accelerated the development of modern algorithms and technologies, among them converters such as Z-Source Inverters (ZSI) are playing an important role. ZSIs are single-stage inverters which are capable of performing both buck and boost operations through an impedance network that enables the shoot-through state. Despite all advantages, these inverters are associated with the non-minimum phase feature imposing heavy restrictions on their closed-loop response. Moreover, uncertainties such as parameter perturbation, unmodeled dynamics, and load disturbances may degrade their performance or even lead to instability, especially when model-based controllers are applied. To tackle these issues, a data-driven model-free adaptive controller is proposed in this paper which guarantees stability and the desired performance of the inverter in the presence of uncertainties. It performs the control action in two steps: First, a model of the system is updated using the current input and output signals of the system. Based on this updated model, the control action is re-tuned to achieve the desired performance. The convergence and stability of the proposed control system are proved in the Lyapunov sense. Experiments corroborate the effectiveness and superiority of the presented method over model-based controllers including PI, state feedback, and optimal robust linear quadratic integral controllers in terms of various metrics.

Decentralized Multi-Agent Control of a Manipulator in Continuous Task Learning

Many real-world tasks require multiple agents to work together. When talking about multiple agents in robotics, it is usually referenced to multiple manipulators in collaboration to solve a given task, where each one is controlled by a single agent. However, due to the increasing development of modular and re-configurable robots, it is also important to investigate the possibility of implementing multi-agent controllers that learn how to manage the manipulator’s degrees of freedom (DoF) in separated clusters for the execution of a given application (e.g., being able to face faults or, partially, new kinematics configurations). Within this context, this paper focuses on the decentralization of the robot control action learning and (re)execution considering a generic multi-DoF manipulator. Indeed, the proposed framework employs a multi-agent paradigm and investigates how such a framework impacts the control action learning process. Multiple variations of the multi-agent framework have been proposed and tested in this research, comparing the achieved performance w.r.t. a centralized (i.e., single-agent) control action learning framework, previously proposed by some of the authors. As a case study, a manipulation task (i.e., grasping and lifting) of an unknown object (to the robot controller) has been considered for validation, employing a Franka EMIKA panda robot. The MuJoCo environment has been employed to implement and test the proposed multi-agent framework. The achieved results show that the proposed decentralized approach is capable of accelerating the learning process at the beginning with respect to the single-agent framework while also reducing the computational effort. In fact, when decentralizing the controller, it is shown that the number of variables involved in the action space can be efficiently separated into several groups and several agents. This simplifies the original complex problem into multiple ones, efficiently improving the task learning process.

Calculation of the regulator for the object with a delay

The task of managing some systems is complicated due to the fact that real technical objects contain delay links. That is, there is a certain period of time when there is no reaction from the object of regulation to the control action. Usually, the presence of a delay link negatively affects the quality of management of such a system. There are various ways to synthesize a control system for such systems. These include: Smith predictors, specialized control tuning algorithms, the use of self-adjusting systems with active adaptation. However, they impose additional requirements on the dynamics of the system or are complex in technical implementation and configuration. Within the framework of this article, an attempt is made to calculate the regulator by the polynomial method for an object with a delay. The mathematical model of the delay is obtained by approximating the delay link next to the Pade. To ensure the necessary dynamics of the transition process from the system, we require the preservation of the poles of the delay link. Then the regulator, calculated for a system with a delay link in the form of a series of Pads, is applied to a system with an "ideal" delay. For clarity of the calculations carried out, an object in the form of a combination of aperiodic and integrating links connected in different ways is taken as an example. The integrating link is necessary to give the system astatic properties. As a delay, we will use the approximation of the range of different orders. The link of delay gives the system a non-stable character.

INCREASING AN ACCURACY OF CONTROL ACTION VOLUME CALCULATION IN CENTRALIZED EMERGENCY CONTROL SYSTEM FOR STATE ESTIMATION OF POWER SYSTEMS

The paper presents the results of the study on the effectiveness and advisability of voltage’s and current’s angles usage, collected from a wide-area monitoring system, to increase an accuracy of control actions volume calculation in case of power system’s state estimation. Centralized emergency control system architecture of a power pool system is shown to better understand the research core. We emphasize that the state estimation software module is the key module in a high level hardware and software package. Ways of telemetry and synchronized phasor measurements collection are outlined. For research practice, Gauss-Newton mathematical method is modified via measurement vector, vector-function, and scalar matrix of weight coefficients. Experiments are provided by IEEE 14-bus power system and 500–220 kV real backbone network. These power systems have several control areas, connected by interchanges. According to experiment results, we conclude that using not only voltage’s and current’s modules but also angles increases an accuracy of control actions volume calculation and effectiveness of a centralized emergency control system operation in the part of a control action formation. Therewith, the usage of current’s modules and angles raises the execution time of the state estimation software module. It is undesirable for real time systems operation. Therefore, it is reasonable to take into account current’s modules and angles only for those interchanges in emergency mode, when intensity factor, characterizing the limit of static stability, is more than 0.92. We also find out that control action volume calculation is sensitive to mistakes in current’s angles measurements. Thus, for reliable usage of current’s modules and angles as data for a state estimation and control action volume calculation, it is necessary to prevent timing errors of synchronized phasor measurement units and also develop a phase shift correction algorithm.