Artificial Neural Network-based Neurocontroller for Hydropower Plant Control (original) (raw)
Related papers
New Approach in Hydropower Plant Control Based on Neural Networks
Energija, ekonomija, ekologija
A new approach to efficient, faster, and intelligent hydropower plant (HPP) control, where constituent equipment is described with highly non-linear mathematical models based on the recommendation from the working group of IEEE on prime movers, is represented in this paper. HPP stability and high efficiency are important factors dependent on the dynamic changes in the energy system demands and the starting time of the plant because the obtained energy is very flexible to those changes in the energy system. This paper is shown and analysed the implementation of the artificial neural network-based controller with PID as an auxiliary controller which helped achieve better behaviour, faster plant stabilization, and operation. The benefits of new technologies and possibilities led to improvements in HPP control and faster system operation. This is achieved by using MATLAB® – Deep Learning Toolbox whereas the simulations are prepared in Simulink. Artificial Neural Networks (ANN) as a tech...
PID Control for Micro-Hydro Power Plants based on Neural Network
Modelling, Identification and Control / 770: Advances in Computer Science and Engineering, 2012
Micro-hydro power plants are power plants with small capacity, which is built in specific locations. The main problem of micro-hydro is the voltage generated is not stable at 220 VA and frequency of 50 Hz. A microhydro that was constructed by Lie Jasa et al. in Gambuk village at Pupuan sub-district, Tabanan district of Bali province, Indonesia in 2010 is still an open loop system in which spin turbine is stable when it is set from the high water level in reservoirs. This will be problematic when the generator load changes. This study will overcome the problem by proposing to build a closed loop system from the change in output frequency for the control circuit. The control circuit is a circuit constructed neural networkbased PID control by using the Brandt-Lin algorithm to control the governor. The governor function is to regulate the amount volume of water running into turbine. By applying Matlab simulation, the result shows that the best output is obtained when the the change in frequency will stabilize at about 40 seconds and using the value of Kp = 0.0637533, Ki=0.00021801 and Kd=0.00301846.
Neural Control System and Performance Characterization with Pid Controller for Water Level Control
2011
ABSTRACT:-The objective of this thesis is to investigate and find a solution by designing the intelligent controller for controlling water level system, such as neural network. The controller also can be specifically run under the circumstance of system disturbances. To achieve these objectives, a prototype of water level control system has been built and implementations of both PID and neural network control algorithms are performed. In PID control, Ziegler Nichols tuning method is used to control the system. In neural network control, the approach of Model Reference Adaptive Neural Network (ANN) Control based on the back propagation algorithm is applied on training the system. Both control algorithms are developed to embed into a standalone DSP-based micro-controller and their performances are compared. Fig.1: Configuration of water level control system. The system consists of servo motor, valves, pump, infrared sensor and a DSP controller.
Control of a hydrolyzer using neural-network based controller
2009
Hydrolyzer is a commonly found unit operation in oleochemical industry. Control of hydrolyzer has to be done carefully since efficiency in the control of this unit will affect the yield of the process. At present conventional controllers such as PI and PID have been used to achieve the setpoint especially under presence of disturbances. In this study, neural network have been applied as an alternative to cope with the dynamics behavior of the hydrolyzer. Two types of control strategies namely, direct inverse controller (DIC) and internal model controller (IMC) were implemented in the control system. Two sets of data were used to develop the DIC and IMC. The controllers were evaluated on the ability to track set-points, load disturbance and noise disturbance test and the IMC was found to be the most versatile controller.
IRJET-Modeling of Micro-Hydro Power Plant and Its Control Based On Neural Network
Micro hydro power plants are hydro plants with small capacity. In the present scenario, the main problem is that, the voltage generated and its frequency is not stable when there is a change in load demand. Hence, we propose to build a closed loop system with change in output frequency as the control variable which can be fed into the PID controller and necessary actions can be taken so as to maintain constant parameters. The control circuit will employ a neural network based PID control which can effectively control the governor which regulates the amount volume of water running into turbine. The neural network block is constructed using Brandt-Lin Algorithm, which enables the controller to adapt changes of plant efficiently. In this paper, it was observed that the most accurate and precise result was given by neural network based controller in minimum stipulated time which effectively improved the plant performance.
Simulation of Hydropower Systems Operation using Artificial Neural Network
The complexity and challenging issues arisen in the management of water resources have called for interdisciplinary collaboration of experts and development of hydro models. Hydro power plants play a key role in electric power systems, due to their low operating costs and their flexibility in real time operation. In addition, sustainability and environmental concerns support their use in current power systems, jointly with other renewable sources of energy, like wind and solar energy. Descriptive simulation models, due to their computational advantages, are able to consider more details of real systems than optimization models. However, they require that the model builder specifies an operating policy. Use of rule curves [5], heuristic rules such as space rule [2], New York City (NYC) rule [3], [4], hedging rules [7] are common ways of defining operating policies required in simulation models. [6] discuss the operating rules for hydropower systems in series and parallel. Reservoir simulations normally require large computational effort and considerable time consumption so that the activities connected with reservoir simulators suffer severe limitations that make it difficult with the vigorous development. Recently some techniques such as Spline, Kriging, Artificial Neural Networks, Experimental Design [8], [1], have been proposed to minimize these problems. The successful applications of Artificial Neural Networks in several research fields suggest the investigation of appropriated architectures to be used as proxies to reservoir simulator. In this article RBS model do not use the iterative process which is so time consuming.
In this paper, a neural Pseudoderivative control (PDF) is applied to a nonlinear mathematical model of the Dinorwig pumped-storage hydroelectric power station. The response of the system with this auto-tuning controller is compared with that of a classic controller, currently implemented on the system. The results show how the application of PDF control to a hydroelectric pumped-storage station improves the dynamic response of the power plant, even when multivariable effects are taken into account.
Adaptive PID Controller Design with Application to Nonlinear Water Level in NEKA Power Plant
In this paper, two novel adaptive PID-Iike controllers capable of controlling multi-variable, non-linear multi-input multiple-output (MIMO) systems are proposed. The proposed controllers are based on neural networks techniques and the learning algorithms are derived according to minimization of the error between the output of the system and the desired output. At first, two kinds of PID-Iike neural network controller named neural network PID and Neural network PID with internal dynamic feedbacks are introduced both of which can be used for controlling multivariable systems. The difference between these two controllers is mainly in the structure of their hidden layers that leads to their different performance. These controllers are applied to different kinds of black box, linear or nonlinear and time variant or time invariant systems. The stability of the proposed algorithm is also proven mathematically. Compared to conventional methods, more satisfactory results are achieved using the proposed methods. The simulation results show the quality performance of the proposed adaptive controllers and algorithms. Finally to show the performance of the proposed method, it is applied to the water level of tanks in water refinement process in NEKA Power Plant that is generally a very nonlinear system. Simulation results in this paper show the satisfying performance of the proposed adaptive controllers. I.
Modeling of Hydropower Plant Production using Artificial Neural Network
Jour of Adv Research in Dynamical & Control Systems , 2018
A new prediction modelling approach for hydropower generation in the presence of input random variables is investigated in this paper. Firstly, they decomposed the annual input variables like the net head of the turbine, and flow rate of water into a certain number of detail signals through artificial neural network (ANN) transform. ANN model used to predict the annual power generation. After the stationary simulation prediction model is obtained, the prediction results were superposed. Finally, simulation results have demonstrated the robustness and effectiveness of this new approach when compared with the existing one.
Controlling of non-minimum phase micro hydro power plant based on adaptive B-Spline neural network1
2015
Hydro power plant is a power generation system that have non-minimum phase model showing initial inverse response characteristic. For span of broad electrical load regulation, conventional non adaptive control techniques, such as PI and PID control would degrade the performance of this power generation system. To ensure the stability of Hydro power plant for severe load variations, we need a kind of controller that has adaptive capability. On the other hand, the utilization of conventional adaptive techniques such as Self Tuning Regulator and Model Reference Adaptive Controller will be diverge to control plants showing non-minimum phase mode. In this paper, the implementation of adaptive intelligence control based on B Spline neural network along with fo rward controller for controlling micro hydro power plant will be presented. Based on its characteristic, this adaptive control technique could be implemented directly without any prior training phase. From the simulation studies, the proposed scheme results fast transient response to load variations compared to traditional PI control and also very stable in responding to severe disturbance.