Combination of Elman Neural Network and Kalman Network for Modeling of Batch Distillation Process

Putri, Adi Novitarini; Machbub, Carmadi; Hidayat, Egi

doi:10.23919/ascc56756.2022.9828080

Cited by 7 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reflux ratio is developed with a range of 0 − 1 which represents the 0% until 100% PWM. The identification system for the second case study was already published on [38].The state, input, and output matrices is define in Eq. ( 49).…”

Section: ) 2nd Case Study : Batch Distillation Systemmentioning

confidence: 99%

Output Feedback Control for Deterministic Unknown Dynamics Discrete-Time System Using Deep Recurrent Q-Networks

Putri,

Hidayat,

Mahayana

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

The current application of control theory is commonly carried out in systems with a model or known system dynamics. However, in practice this is a formidable task to achieve as not all state information can be known. The use of the Output Feedback (OPFB) scheme in the field of control systems also possesses a weakness because it requires the use of an observer. This appears rather contradictory as the use of an observer requires system dynamics information. This research proposes an optimal control scheme using Deep Recurrent Q-Networks (DRQN) to generate an optimal control signal trajectory based on a collection of input and output data from the system itself. The approach proposed in this study is based on the Q-Learning method from the Reinforcement Learning (RL) scheme. The Long-Short Term Memory (LSTM) is used to approximate the Q-function and determine the control signals for a system without a known model. The method that we proposed in this study has been tested on four case studies. The control signal trajectory generated from our proposed algorithm, is much smaller than the control signal that generated from classical Q-Learning scheme. The results of this research are certainly relevant to the aim of OPFB, namely that the controller is designed to be able to regulate (bring the state trajectory to zero) and minimize control signal energy. It is empirically discovered that the same result is proven by the norm values resulting from the Q-function trajectory. The norm of Q-function trajectory for our proposed algorithm on the 1st, 2nd, 3rd, and 4th case studies are 2.11E-08, 3.15E-06, 3.79E-09, and 1.59E-13, respectively.

show abstract

Section: ) 2nd Case Study : Batch Distillation Systemmentioning

confidence: 99%

Output Feedback Control for Deterministic Unknown Dynamics Discrete-Time System Using Deep Recurrent Q-Networks

Putri,

Hidayat,

Mahayana

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…Treating the mixture of hydrocarbons, they showed in two study cases for a high concentration of the first component, (>95%), for double period reflux ratios to a difference of 1.54% between, it seems that the gap between the first and the second-period reflux ratio, increase for more than 2.5 times. Most recently, Putri et al, 11 showed from normalized product concentration output from a particular neural network approach, observed the "quasi zero-bang" sequence in the very beginning, fluctuating further around a mean value (of 0.8), subsequently dropped for "quasi-zero" almost instantaneously to switch to the instantaneous increase in the next moment, however, the combined control made up from neural network and Kalman filter, ("smoothened") resulted in lowering the "quasibang-zero" sequence (from previous maximum 1 or 0.99) to less than 0.95. Mahida et al 12 studied a pressure swing distillation process for separating ternary azeotropic mixtures of acidic aqueous solutions, having examined three compositions of feed mixture: the first component in excess (ie.…”

Section: Introductionmentioning

confidence: 99%

Pressure-Driven Batch Distillation Optimal Control for Ethanol-Water Separation

Stojkovic¹

2023

Fine Chemical Engineering

View full text Add to dashboard Cite

Ethanol-water separation by pressure-swing batch distillation in a double-batch rectifier system is a process under investigation. In this work, a complete global solution to the optimal control problem in the form of a sequential synthesis of controlled trajectories is derived. During this study, the optimal reflux strategy through cyclic operation was extended to the separation of a second non-ideal minimum boiling azeotrope of industrial importance. The optimal cyclic reflux ratio operating policy, brought significant reductions in time duration, going up to even 50%. Consequently even more significant reductions in energy expenditure, since the derived discontinuous energy function tends to show cyclic “behavior” as well. Lastly, the proposed strategy provides the possibility to highly recover the desired product even within a single period, and finally maybe even the most importantly, flexibility in operating different units within the time frame.

show abstract

“…ERNN is used for the prediction of vehicle interior noise [14]. ERNN describes powerful learning techniques such as uncertainty estimation [15], environmental adaptability [16], increasing the amount and accuracy of forecast data [17], improving the accuracy of time series forecasting [18], [19], early detection of circuit failures by combining with cuckoo search [20], improvement of accuracy of identification processing by combining with the Kalman filter [21], face recognition possible by connecting with principal component analysis (PCA) [22].…”

Section: Introductionmentioning

confidence: 99%

Eligibility of village fund direct cash assistance recipients using artificial neural network

Midyanti

Bahri

Suhardi

et al. 2023

IJ-AI

View full text Add to dashboard Cite

<p>Bantuan Langsung Tunai Dana Desa (BLT-DD), or known as Village Fund Direct Cash Assistance is assistance from the Indonesian government which causes problems and conflicts in the community when the assistance is not on target. The classification algorithm is proven to use in determining BLT-DD recipients. In this study, the radial basis function (RBF) and elman recurrent neural network (ERNN) models compare to classify the eligibility of BLTDD recipients. In the experiment, the optimal performance of the RBF and ERNN compare in determining the eligibility of BLT-DD recipients. Also, it’s compared with the classification algorithm that implements the same data, namely BLT-DD data for Kubu Raya District. The experimental results show the effectiveness of the RBF model in recognizing test data, while the ERNN model is effective in identifying test data. The RBF and ERNN models can achieve the same total accuracy of 98.10%.</p>

show abstract

Combination of Elman Neural Network and Kalman Network for Modeling of Batch Distillation Process

Cited by 7 publications

References 17 publications

Output Feedback Control for Deterministic Unknown Dynamics Discrete-Time System Using Deep Recurrent Q-Networks

Output Feedback Control for Deterministic Unknown Dynamics Discrete-Time System Using Deep Recurrent Q-Networks

Pressure-Driven Batch Distillation Optimal Control for Ethanol-Water Separation

Eligibility of village fund direct cash assistance recipients using artificial neural network

Contact Info

Product

Resources

About