Self-Learning Fuzzy Logic Controller using Q-Learning

2000 ◽  
Vol 4 (5) ◽  
pp. 349-354 ◽  
Author(s):  
Min-Soeng Kim ◽  
◽  
Sun-Gi Hong ◽  
Ju-Jang Lee
Keyword(s):  
Fuzzy Logic ◽  
Fuzzy Logic Controller ◽  
Expert Knowledge ◽  
Learning Algorithm ◽  
A Priori ◽  
Fuzzy Rule ◽  
Fuzzy Rules ◽  
Q Learning ◽  
Continuous Output ◽  
Continuous Actions

Fuzzy logic controllers consist of if-then fuzzy rules generally adopted from a priori expert knowledge. However, it is not always easy or cheap to obtain expert knowledge. Q-learning can be used to acquire knowledge from experiences even without the model of the environment. The conventional Q-learning algorithm cannot deal with continuous states and continuous actions. However, the fuzzy logic controller can inherently receive continuous input values and generate continuous output values. Thus, in this paper, the Q-learning algorithm is incorporated into the fuzzy logic controller to compensate for each method’s disadvantages. Modified fuzzy rules are proposed in order to incorporate the Q-learning algorithm into the fuzzy logic controller. This combination results in the fuzzy logic controller that can learn through experience. Since Q-values in Q-learning are functional values of the state and the action, we cannot directly apply the conventional Q-learning algorithm to the proposed fuzzy logic controller. Interpolation is used in each modified fuzzy rule so that the Q-value is updatable.

scholarly journals PERFORMANCE IMPROVEMENTS OF SEPARATELY EXCITED DC MOTOR USING FUZZY LOGIC CONTROL

2020 ◽  
Vol 6 (2) ◽  
pp. 47-58
Author(s):  
R. Nagarajan ◽  
M. Gokulkannan ◽  
T. Dinesh ◽  
S. Murugesan ◽  
M. Naveenprasanth
Keyword(s):  
Fuzzy Logic ◽  
Fuzzy Logic Controller ◽  
Fuzzy Controller ◽  
Expert Knowledge ◽  
Speed Control ◽  
Dc Motor ◽  
Fuzzy Rules ◽  
Motor Speed ◽  
Performance Improvements ◽  
Speed Performance

This paper demonstrates the importance of a fuzzy logic controller over conventional method. The performance of the separately excited DC motor is analyzed by using fuzzy logic controller (FLC) in MATLAB/SIMULINK environment. The FLC speed controller is designed based on the expert knowledge of the fuzzy rules system. The proposed DC motor speed control fuzzy rules are designed for fuzzy logic controller. The output response of the system is obtained by using fuzzy logic controller. The designed fuzzy controller for speed control performance is investigated. Significantly reducing the overshoot and shortening the settling time of the speed response of the motor. They validate different control of approaches, the simulation results show improvement in motor efficiency and speed performance.

Design and Tuning of Distributed Importance-Based Fuzzy Logic Controller for Two-Link Rigid-Flexible Manipulator

2003 ◽  
Author(s):  
Linda Z. Shi ◽  
Mohamed B. Trabia
Keyword(s):  
Fuzzy Logic ◽  
Fuzzy Logic Control ◽  
Fuzzy Logic Controller ◽  
Expert Knowledge ◽  
Vertical Plane ◽  
Fuzzy Rules ◽  
Flexible Manipulator ◽  
Linear Controller ◽  
Logic Control ◽  
Single Link

Fuzzy logic control has been widely used in many industrial processes due to its computationally efficient and robust characteristics. In many applications, verbalization of expert-knowledge can be easily used to design a fuzzy logic controller (FLC). On the other hand, other applications with many variables and complex mathematical model offer challenges to fuzzy logic control. Multi-link flexible manipulators belong to this category. An earlier work, [1], presented a distributed importance-based FLC for a single-link flexible manipulator. This paper extends this idea to a two-link rigid-flexible manipulator that moves in a vertical plane where the gravity field is active. The structure of the proposed controller is based on evaluating the importance degrees of the variables of the system, over its range of operation, to consider the coupling effects between the rigid and the flexible links. Variables with higher importance degrees are grouped together while variables with lesser importance degrees may be deleted to simplify the design of the controller. After determining the importance degrees of the variables, a distributed controller composed of four two-input one-output FLC’s is created. Unlike the single-link flexible manipulator, the fuzzy rules of the distributed FLC for the two-link rigid-flexible manipulator cannot be written by an expert based on intuition and observation of the inertial system due to the complexity of the manipulator and the coupling effect of its variables. To solve this problem, an importance-based linear controller that has the same input-output structure as that of distributed importance-based FLC is constructed to help write the fuzzy rules of the distributed FLC. Fuzzy rules of the distributed FLC are then selected to mimic the performance of the corresponding linear controllers. To compare the performance of the distributed importance-based FLC with that of importance-based linear controller, these two controllers are tuned using nonlinear programming by varying the gains of the importance-based linear controller and the parameters of membership functions of the variables in the distributed importance-based FLC. Robustness of each of the controllers after tuning is tested by varying the payload of the manipulator. The two importance-based controllers are simulated and compared.

A Reinforcement Learning Scheme of Fuzzy Rules with Reduced Conditions

2000 ◽  
Vol 4 (2) ◽  
pp. 146-151 ◽  
Author(s):  
Hiroshi Kawakami ◽  
◽  
Osamu Katai ◽  
Tadataka Konishi ◽  
Keyword(s):  
Fuzzy Inference ◽  
Learning Algorithm ◽  
Fuzzy Rule ◽  
Fuzzy Rules ◽  
Q Value ◽  
Huge Number ◽  
Q Learning ◽  
Control Rules ◽  
A Value ◽  
Rule System

This paper proposes a new method of Q-learning for the case where the states (conditions) and actions of systems are assumed to be continuous. The components of Q-tables are interpolated by fuzzy inference. The initial set of fuzzy rules is made of all combinations of conditions and actions relevant to the problem. Each rule is then associated with a value by which the Q-values of condition/action pairs are estimated. The values are revised by the Q-learning algorithm so as to make the fuzzy rule system effective. Although this framework may require a huge number of the initial fuzzy rules, we will show that considerable reduction of the number can be done by adopting what we call Condition Reduced Fuzzy Rules (CRFR). The antecedent part of CRFR consists of all actions and the selected conditions, and its consequent is set to be its Q-value. Finally, experimental results show that controllers with CRFRs perform equally well to the system with the most detailed fuzzy control rules, while the total number of parameters that have to be revised through the whole learning process is considerably reduced, and the number of the revised parameters at each step of learning increased.

Performance Comparison of MANET, VANET and FANET

2018 ◽  
Vol 8 (5) ◽  
pp. 132
Author(s):  
Aarti Sahu ◽  
Laxmi Shrivastava
Keyword(s):  
Fuzzy Logic ◽  
Wireless Network ◽  
Packet Loss ◽  
Ad Hoc Network ◽  
Fuzzy Logic Controller ◽  
Ad Hoc ◽  
Mobile Ad Hoc Network ◽  
Performance Comparison ◽  
Fuzzy Rules ◽  
Mobile Ad Hoc

A wireless ad hoc network is a decentralized kind of wireless network. It is a kind of temporary Computer-to-Computer connection. It is a spontaneous network which includes mobile ad-hoc network (MANET), vehicular ad-hoc network (VANET) and Flying ad-hoc network (FANET). Mobile Ad Hoc Network (MANET) is a temporary network that can be dynamically formed to exchange information by wireless nodes or routers which may be mobile. A VANET is a sub form of MANET. It is an technology that uses vehicles as nodes in a network to make a mobile network. FANET is an ad-hoc network of flying nodes. They can fly independently or can be operated distantly. In this research paper Fuzzy based control approaches in wireless network detects & avoids congestion by developing the ad-hoc fuzzy rules as well as membership functions.In this concept, two parameters have been used as: a) Channel load b) The size of queue within intermediate nodes. These parameters constitute the input to Fuzzy logic controller. The output of Fuzzy logic control (sending rate) derives from the conjunction with Fuzzy Rules Base. The parameter used input channel load, queue length which are produce the sending rate output in fuzzy logic. This fuzzy value has been used to compare the MANET, FANET and VANET in terms of the parameters Throughput, packet loss ratio, end to end delay. The simulation results reveal that usage of Qual Net 6.1 simulator has reduced packet-loss in MANET with comparing of VANET and FANET.

scholarly journals Experts versus Algorithms? Optimized Fuzzy Logic Energy Management of Autonomous PV Hybrid Systems with Battery and H2 Storage

Energies ◽  
2021 ◽  
Vol 14 (6) ◽  
pp. 1777
Author(s):  
Lisa Gerlach ◽  
Thilo Bocklisch
Keyword(s):  
Fuzzy Logic ◽  
Energy Management ◽  
Fuzzy Logic Controller ◽  
Expert Knowledge ◽  
Large Power ◽  
Hybrid Energy ◽  
Knowledge Based ◽  
Long Term Storage ◽  
Fuzzy Logic Approach

Off-grid applications based on intermittent solar power benefit greatly from hybrid energy storage systems consisting of a battery short-term and a hydrogen long-term storage path. An intelligent energy management is required to balance short-, intermediate- and long-term fluctuations in electricity demand and supply, while maximizing system efficiency and minimizing component stress. An energy management was developed that combines the benefits of an expert-knowledge based fuzzy logic approach with a metaheuristic particle swarm optimization. Unlike in most existing work, interpretability of the optimized fuzzy logic controller is maintained, allowing the expert to evaluate and adjust it if deemed necessary. The energy management was tested with 65 1-year household load datasets. It was shown that the expert tuned controller is more robust to changes in load pattern then the optimized controller. However, simple readjustments restore robustness, while largely retaining the benefits achieved through optimization. Nevertheless, it was demonstrated that there is no one-size-fits-all tuning. Especially, large power peaks on the demand-side require overly conservative tunings. This is not desirable in situations where such peaks can be avoided through other means.

A FILTER PROPOSAL FOR INCLUDING FEATURE CONSTRUCTION IN A GENETIC LEARNING ALGORITHM

2012 ◽  
Vol 20 (supp02) ◽  
pp. 31-49 ◽  
Author(s):  
DAVID GARCIA ◽  
ANTONIO GONZALEZ ◽  
RAUL PEREZ
Keyword(s):  
Learning Algorithm ◽  
A Priori ◽  
Fuzzy Rule ◽  
Feature Construction ◽  
Rule Based ◽  
Filter Model ◽  
Information Measures ◽  
Construction Techniques ◽  
Basic Learning ◽  
Current Proposal

In system identification process often a predetermined set of features is used. However, in many cases it is difficult to know a priori whether the selected features were really the more appropriate ones. This is the reason why the feature construction techniques have been very interesting in many applications. Thus, the current proposal introduces the use of these techniques in order to improve the description of fuzzy rule-based systems. In particular, the idea is to include feature construction in a genetic learning algorithm. The construction of attributes in this study will be restricted to the inclusion of functions defined on the initial attributes of the system. Since the number of functions and the number of attributes can be very large, a filter model, based on the use of information measures, is introduced. In this way, the genetic algorithm only needs to explore the particular new features that may be of greater interest to the final identification of the system. In order to manage the knowledge provided by the new attributes based on the use of functions we propose a new model of rule by extending a basic learning fuzzy rule-based model. Finally, we show the experimental study associated with this work.

Data-Tuned Fuzzy Logic Controller Applied to a Horizontal Axis Wind System

2020 ◽  
Author(s):  
Lorenzo Dambrosio
Keyword(s):  
Fuzzy Logic ◽  
Wind Turbine ◽  
Control Algorithm ◽  
Fuzzy Logic Controller ◽  
Tracking Error ◽  
Synchronous Generator ◽  
Horizontal Axis ◽  
Fuzzy Rules ◽  
Wind System ◽  
Horizontal Axis Wind Turbine

Abstract This paper deals with the control problem concerning the output voltage frequency and amplitude regulation of a wind system power plant not connected to the supply grid. The wind system configuration includes a horizontal-axis wind-turbine which drives a synchronous generator. An appropriate modeling approach has been adopted for both the wind-turbine and the synchronous generator. The proposed controller makes use of the fuzzy logic environment in order to take advantage of the wind plant system informations integrated into a limited number of equilibrium condition points (input variable - output variable pairs). The fuzzy logic controller described in the present paper merges the most appropriate fuzzy rules clusters, based on the steady state working conditions. Then, thanks to a Least Square Estimator algorithm, the proposed control algorithm evaluates, for each sample time, the linear relation between control law correction and control tracking error levels. In order to demonstrate robustness of the suggested fuzzy control algorithm, two sets of results have been provided: the first one consider a fuzzy base with equally spaced rules, whereas, in the second set results, the number of fuzzy rules is reduced by a 25%.

Sign in / Sign up

Export Citation Format

Share Document