Adaptive neurofuzzy controller to regulate UTSG water level in nuclear power plants

Munasinghe, S.R.; Kim, Min-Soeng; Lee, Ju-Jang

doi:10.1109/tns.2004.842723

Cited by 62 publications

(23 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neuro-fuzzy can approximate certain types of nonlinear functions well in nature. Therefore, neuro-fuzzy models have been applied in designing control systems, such as the temperature control system for greenhouse [33], an antilock braking system of motor vehicle [34], a water-level control of Utube steam generators in nuclear power plants [35], and so on [3,7,8]. This study have exhibited that proposed methods have better properties than the conventional counter methods in function approximations and realworld benchmark problems.…”

Section: Discussionmentioning

confidence: 99%

Simplified Real-, Complex-, and Quaternion-Valued Neuro-Fuzzy Learning Algorithms

Hata¹,

Akhand²,

Islam³

et al. 2018

IJISA

View full text Add to dashboard Cite

Abstract-The conventional real-valued neuro-fuzzy method (RNF) is based on classic fuzzy systems with antecedent membership functions and consequent singletons. Rules in RNF are made by all the combinations of membership functions; thus, the number of rules as well as total parameters increase rapidly with the number of inputs. Although network parameters are relatively less in the recently developed complex-valued neuro-fuzzy (CVNF) and quaternion neuro-fuzzy (QNF), parameters increase with number of inputs. This study investigates simplified fuzzy rules that constrain rapid increment of rules with inputs; and proposed simplified RNF (SRNF), simplified CVNF (SCVNF) and simplified QNF (SQNF) employing the proposed simplified fuzzy rules in conventional methods. The proposed simplified neuro-fuzzy learning methods differ from the conventional methods in their fuzzy rule structures. The methods tune fuzzy rules based on the gradient descent method. The number of rules in these methods are equal to the number of divisions of input space; and hence they require significantly less number of parameters to be tuned. The proposed methods are tested on function approximations and classification problems. They exhibit much less execution time than the conventional counterparts with equivalent accuracy. Due to less number of parameters, the proposed methods can be utilized for the problems (e.g., real-time control of large systems) where the conventional methods are difficult to apply due to time constrain.

show abstract

Section: Discussionmentioning

confidence: 99%

Simplified Real-, Complex-, and Quaternion-Valued Neuro-Fuzzy Learning Algorithms

Hata¹,

Akhand²,

Islam³

et al. 2018

IJISA

View full text Add to dashboard Cite

show abstract

“…By substituting virtual control law Equation (A12) for inequality Equation (A8), and by considering inequalities Equations (23) From inequalities Equations (23)- (25), it is clear that σ d11 , σ p24 and σ p35 are positive scalars. Thus, the closed-loop constituted by subsystem Equation (A1), virtual control Equation (A12) and control laws Equations (17) and (18) is certainly globally asymptotically stable. Moreover, from Energies 2016, 9, 37 13 of 14 inequalities Equation (A13) and Equations (A9)-(A11), feedback gains k d11 , k p24 and k p35 are higher, the closed-loop system is more stable.…”

Section: Appendix: Proof Of Theoremmentioning

confidence: 99%

“…The physics-based control (PBC) method is also an effective way to design nonlinear reactor control laws by retaining or strengthening stable subdynamics and by cancelling or suppressing unstable subdynamics, which has been applied to the load-following control design for the PWRs (Pressurized Water Reactors) [11][12][13] and MHTGRs [14,15] recently. From the aspect of SG controller design, some advanced control approaches such as the model predictive control (MPC) [16], neurofuzzy [17] and feedback-dissipation [18] methods have all been applied to improve the regulation performance of SGs.…”

Section: Introductionmentioning

confidence: 99%

“…However, these control laws have the drawback of heavy model-reliance. Some control laws needs the state prediction or observation directly based upon the reactor or SG models [9][10][11][12][14][15][16][17], and the control parameters also vary with the reactor or SG physical and thermal-hydraulic parameters [9][10][11][12][13][14][15][16][17][18], which leads to the difficulty in practically tuning and implementation. Therefore, it is necessary to develop model-free control methods to relieve the heavy model reliance of the designed control strategies.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Model-Free Coordinated Control for MHTGR-Based Nuclear Steam Supply Systems

Dong

2016

Energies

View full text Add to dashboard Cite

Abstract:The modular high temperature gas-cooled reactor (MHTGR) is a typical small modular reactor (SMR) that offers simpler, standardized and safer modular design by being factory built, requiring smaller initial capital investment, and having a shorter construction period. Thanks to its small size, the MHTGRs could be beneficial in providing electric power to remote areas that are deficient in transmission or distribution and in generating local power for large population centers. Based on the multi-modular operation scheme, the inherent safety feature of the MHTGRs can be applicable to large nuclear plants of any desired power rating. The MHTGR-based nuclear steam supplying system (NSSS) is constituted by an MHTGR, a side-by-side arranged helical-coil once-through steam generator (OTSG) and some connecting pipes. Due to the side-by-side arrangement, there is a tight coupling effect between the MHTGR and OTSG. Moreover, there always exists the parameter perturbation of the NSSSs. Thus, it is meaningful to study the model-free coordinated control of MHTGR-based NSSSs for safe, stable, robust and efficient operation. In this paper, a new model-free coordinated control strategy that regulates the nuclear power, MHTGR outlet helium temperature and OTSG outlet overheated steam temperature by properly adjusting the control rod position, helium flowrate and feed-water flowrate is established for the MHTGR-based NSSSs. Sufficient conditions for the globally asymptotic closed-loop stability is given. Finally, numerical simulation results in the cases of large range power decrease and increase illustrate the satisfactory performance of this newly-developed model-free coordinated NSSS control law.

show abstract

“…The research to date into advanced techniques of water level control in steam generators comes to algorithms based on gain scheduling control [1,6,12] artificial neural networks [4,6,11], fuzzy logic [5,6,11,13], adaptive control [8,11], Model Predictive Control [9,10] and combinations of aforementioned.…”

Section: Introductionmentioning

confidence: 99%

Multi-region fuzzy logic controller with local PID controllers for U-tube steam generator in nuclear power plant

Puchalski

Duzinkiewicz

2015

Archives of Control Sciences

View full text Add to dashboard Cite

In the paper, analysis of multi-region fuzzy logic controller with local PID controllers for steam generator of pressurized water reactor (PWR) working in wide range of thermal power changes is presented. The U-tube steam generator has a nonlinear dynamics depending on thermal power transferred from coolant of the primary loop of the PWR plant. Control of water level in the steam generator conducted by a traditional PID controller which is designed for nominal power level of the nuclear reactor operates insufficiently well in wide range of operational conditions, especially at the low thermal power level. Thus the steam generator is often controlled manually by operators. Incorrect water level in the steam generator may lead to accidental shutdown of the nuclear reactor and consequently financial losses. In the paper a comparison of proposed multi region fuzzy logic controller and traditional PID controllers designed only for nominal condition is presented. The gains of the local PID controllers have been derived by solving appropriate optimization tasks with the cost function in a form of integrated squared error (ISE) criterion. In both cases, a model of steam generator which is readily available in literature was used for control algorithms synthesis purposes. The proposed multi-region fuzzy logic controller and traditional PID controller were subjected to broad-based simulation tests in rapid prototyping software -Matlab/Simulink. These tests proved the advantage of multi-region fuzzy logic controller with local PID controllers over its traditional counterpart.

show abstract

Adaptive neurofuzzy controller to regulate UTSG water level in nuclear power plants

Cited by 62 publications

References 18 publications

Simplified Real-, Complex-, and Quaternion-Valued Neuro-Fuzzy Learning Algorithms

Simplified Real-, Complex-, and Quaternion-Valued Neuro-Fuzzy Learning Algorithms

Model-Free Coordinated Control for MHTGR-Based Nuclear Steam Supply Systems

Multi-region fuzzy logic controller with local PID controllers for U-tube steam generator in nuclear power plant

Contact Info

Product

Resources

About