Fig 1 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
Source publication
![The nominal parameter values for the continuous fermentation process [16].](publication/354049840/figure/tbl1/AS:1059269197692931@1629560865828/The-nominal-parameter-values-for-the-continuous-fermentation-process-16_Q320.jpg)
Chemical and biochemical processes generally suffer from extreme nonlinearities with respect to internal states, manipulated variables, and also disturbances. These processes have always received special technical and scientific attention due to their importance as the means of large-scale production of chemicals, pharmaceuticals, and biologically...
Contexts in source publication
Context 1
... must be also noted that Fig. 1 refers to a specific structure of NARX networks that do not directly receive y output(s) as their inputs. In our case, y as the controller output (manipulated variable) is resolved in the closed-loop response of the process and does not directly appear as the network ...
Context 2
... preliminary NNARX network structure was constructed by the Neural Network ToolboxSimulink code generation facility [21]. The network was then modified in Simulink to suit the design required as a MIMO controller. The overall network structural parameters are given in Table 4. As shown in this table and earlier in Fig. 1, several delays were introduced at the input layer. It is worthy of attention that this network is fully connected, which means there is a connection between every input and every neuron in each layer. As mentioned earlier, the importance of these connections is controlled by the weight parameters, which are regulated by the ...
Citations
... However, NARX-NN can be used to design nonlinear neural network-based controller for extraction process owing to the flexibility of soft computational attribute of the model. NARX controller was developed for both neutralization and fermentation processes by Medi and Asadbeigi [13] and magnetic levitation system [14]. Therefore, the novelty in this study is to develop a special type of dynamic model for the prediction and control of oil yield from Hura crepitans seeds using NARX-NN technique, which has already been succsefully used to predict biogas production rate [15], solar radiation [16], energy production [17] and process behaviour [18]. ...
Vegetable oils are a crucial source of raw materials for many industries. In order to meet the rising demand for oil on global scale, it has become essential to investigate underutilized plant oilseeds. Hura crepitans seeds are one of the underused plant oilseeds from which oil can be produced via solvent-based extraction. For the purpose of predicting the oil yield from Hura crepitans seeds, the extraction process was modelled using a nonlinear autoregressive exogenous neural network (NARX-NN). The input variables to the model are seed/solvent ratio, extraction temperature and extraction time, while oil yield is the response output variable. NARX-NN model is based on 200 data samples, and model architecture comprises of 3 inputs, 1 hidden layer (with 15 neurons) and 1 output with 2 delay elements. The performance evaluation was carried out to examine the accuracy of the developed model in predicting oil yield from Hura crepitans using different statistical indices. The overall correlation coefficient, R (0.80829), mean square error, MSE (0.0120), root mean square error, RMSE (0.1080) and standard prediction error, SEP (0.1666) show that NARX-NN model can accurately be used for the prediction oil yield from Hura crepitans seeds.
... Based on a NARX model, two intelligent controllers to guarantee the desired growth conditions of crops, such as humidity and temperature, in a greenhouse were implemented in [63]. A general-purpose genetic algorithm (GA)-optimized NARX-ANN-based controller for control of nonlinear chemical and biochemical processes was introduced in [64]. A NARX-ANN-based predictive controller of personalized heating systems was designed in [65]. ...
The management of irrigation main canals are studied in this research. One way of improving this is designing an efficient automatic control system of the water that flows through the canal pools, which is usually carried out by PI controllers. However, since canal pools are systems with large time delays and nonlinear hydrodynamics, these PIs are tuned in a very conservative way so that the closed-loop instability that may appear depending on the chosen operation regime is avoided. These controllers are inefficient because they have slow time responses. In order to obtain faster responses that remain stable independently of the operation regime, a control system that combines a Smith predictor, which is appropriate to control linear systems with large time delays, with a NARX artificial neural network (ANN), that models the nonlinear dynamics of the
pools, is proposed. By applying system identification procedures, two nonlinear NARX-ANN-based models and a linear mathematical model of a real canal pool were obtained. These models were applied to implement a modified NARX-ANN-based SP controller and a conventional linear SP controller. Experimental results on our real canal pool showed that our modified NARX-ANNbased SP controller overcomes conventional linear SP controllers in both setpoint tracking and load disturbance rejection
The wind power generation system plays a significant role in the power sector as it is an environment-friendly green power system, increasing power demand, and technological development in wind power systems. Wind turbine systems are exposed to the harsh environment with continuous variation of wind speed with gusts causing damage and failure in system components along with the fluctuation of generated power. The hydrostatic transmission system has become one of the promising solutions over the gear transmission system for transmitting power from the turbine rotor to the generator. Further to reduce power generation costs in wind power systems, a suitable control system with parametric uncertainty and system fault plays a significant role. In this study, the 5 MW wind turbine model has been developed with the combination of blade element momentum theory and the electrohydraulic transmission system model. Moreover, the wind turbine system model has been imposed fault in the pump of electrohydraulic transmission system. The proposed wind turbine system model has been validated with the existing result. The blade element momentum theory has been used to estimate the optimum pump turbine couple rotational speed for maximum power tracking. Double loop controller has been used for wind turbine power transmission system control. The first controller loop has been used for pump and wind turbine system speed control for maximum power tracking, as a passive fault tolerance controller and the second control loop for motor and generator system speed control to regulate the frequency of the generated power. Interval type 2-fuzzy proportional–integral–derivative controller are suitable for high degree of uncertain system like wind power system due to their footprint of uncertainties. Proper choice of footprint of uncertainty provides robust performance against uncertainties and dynamic performance. Hence, the primary and secondary controller has been developed as interval type 2-fuzzy proportional–integral–derivative with inertial weight local search–based teaching–learning-based optimization controller. The inertial weight local search–based teaching–learning-based optimization interval type 2-fuzzy proportional–integral–derivative controller performance has been studied with benchmark sinusoidal test signals. The proposed inertial weight local search–based teaching–learning-based optimization interval type 2-fuzzy proportional–integral–derivative controller performance has been also compared with conventional proportional–integral–derivative and interval type 2-fuzzy proportional–integral–derivative controller. The proposed system performance has been compared with contemporary reported digital hydrostatic transmission wind turbine system and recently reported controller with consideration of fault in the pump. The proposed inertial weight local search–based teaching–learning-based optimization interval type 2-fuzzy proportional–integral–derivative controller performance has been compared through integral absolute error with interval type 2-fuzzy proportional–integral–derivative controller and recently reported proportional–integral–derivative sliding mode controller obtained as 0.0016, 0.0029, and 0.0031, respectively.
The learning process and hyper-parameter optimization of artificial neural networks (ANNs) and deep learning (DL) architectures is considered one of the most challenging machine learning problems. Several past studies have used gradient-based back propagation methods to train DL architectures. However, gradient-based methods have major drawbacks such as stucking at local minimums in multi-objective cost functions, expensive execution time due to calculating gradient information with thousands of iterations and needing the cost functions to be continuous. Since training the ANNs and DLs is an NP-hard optimization problem, their structure and parameters optimization using the meta-heuristic (MH) algorithms has been considerably raised. MH algorithms can accurately formulate the optimal estimation of DL components (such as hyper-parameter, weights, number of layers, number of neurons, learning rate, etc.). This paper provides a comprehensive review of the optimization of ANNs and DLs using MH algorithms. In this paper, we have reviewed the latest developments in the use of MH algorithms in the DL and ANN methods, presented their disadvantages and advantages, and pointed out some research directions to fill the gaps between MHs and DL methods. Moreover, it has been explained that the evolutionary hybrid architecture still has limited applicability in the literature. Also, this paper classifies the latest MH algorithms in the literature to demonstrate their effectiveness in DL and ANN training for various applications. Most researchers tend to extend novel hybrid algorithms by combining MHs to optimize the hyper-parameters of DLs and ANNs. The development of hybrid MHs helps improving algorithms performance and capable of solving complex optimization problems. In general, the optimal performance of the MHs should be able to achieve a suitable trade-off between exploration and exploitation features. Hence, this paper tries to summarize various MH algorithms in terms of the convergence trend, exploration, exploitation, and the ability to avoid local minima. The integration of MH with DLs is expected to accelerate the training process in the coming few years. However, relevant publications in this way are still rare.
