An event triggered control scheme for enhanced production of Escherichia coli and biomass concentration during fed-batch cultivation.

Control of a bioprocess is a challenging task mainly due to the nonlinearity of the process, the complex nature of microorganisms, and variations in critical parameters such as temperature, pH, and agitator speed. Generally, the optimum values chosen for critical parameters during Escherichia coli (E.coli) K-12fed-batch fermentation are37 ᵒC for temperature, 7 for pH, and 35 % for Dissolved Oxygen (DO). The objective of this research is to enhance biomass concentration while minimizing energy consumption. To achieve this, an Event-Triggered Control (ETC) scheme based on feedback-feed forward control is proposed. The ETC system dynamically adjusts the substrate feed rate in response to variations in critical parameters. We compare the performance of classical Proportional Integral (PI) controllers and advanced Model Predictive Control (MPC) controllers in terms of bioprocess yield. Initially, the data are collected from a laboratory-scaled 3L bioreactor setup under fed-batch operating conditions, and data-driven models are developed using system identification techniques. Then, classical Proportional Integral (PI) and advanced Model Predictive Control (MPC) based feedback controllers are developed for controlling the yield of bioprocess by manipulating substrate flow rate, and their performances are compared. PI and MPC-based Event Triggered Feed Forward Controllers are designed to increase the yield and to suppress the effect of known disturbances due to critical parameters. Whenever there is a variation in the value of a critical parameter, it is considered an event, and ETC initiates a control action by manipulating the substrate feed rate. PI and MPC-based ETC controllers are developed in simulation, and their closed-loop performances are compared. It is observed that the Integral Square Error (ISE) is notably minimized to 4.668 for MPC with disturbance and 4.742 for MPC with Feed Forward Control. Similarly, the Integral Absolute Error (IAE) reduces to 2.453 for MPC with disturbance and 0.8124 for MPC with Feed Forward Control. The simulation results reveal that the MPC-based ETC control scheme enhances the biomass yield by 7 %, and this result is verified experimentally. This system dynamically adjusts the substrate feed rate in response to variations in critical parameters, which is a novel approach in the field of bioprocess control. Also, the proposed control schemes help reduce the frequency of communication between controller and actuator, which reduces power consumption.

Active fault tolerant deep brain stimulator for epilepsy using deep neural network.

Millions of people around the world are affected by different kinds of epileptic seizures. A deep brain stimulator is now claimed to be one of the most promising tools to control severe epileptic seizures. The present study proposes Hodgkin-Huxley (HH) model-based Active Fault Tolerant Deep Brain Stimulator (AFTDBS) for brain neurons to suppress epileptic seizures against ion channel conductance variations using a Deep Neural Network (DNN). The AFTDBS contains the following three modules: (i) Detection of epileptic seizures using black box classifiers such as Support Vector Machine (SVM) and K-Nearest Neighbor (KNN), (ii) Prediction of ion channels conductance variations using Long Short-Term Memory (LSTM), and (iii) Development of Reconfigurable Deep Brain Stimulator (RDBS) to control epileptic spikes using Proportional Integral (PI) Controller and Model Predictive Controller (MPC). Initially, the synthetic data were collected from the HH model by varying ion channel conductance. Then, the seizure was classified into four groups namely, normal and epileptic due to variations in sodium ion-channel conductance, potassium ion-channel conductance, and both sodium and potassium ion-channel conductance. In the present work, current controlled deep brain stimulators were designed for epileptic suppression. Finally, the closed-loop performances and stability of the proposed control schemes were analyzed. The simulation results demonstrated the efficacy of the proposed DNN-based AFTDBS.

HH model based smart deep brain stimulator to detect, predict and control epilepsy using machine learning algorithm.

BACKGROUND: Epilepsy is the most common neurological disorder in the world. To control epilepsy, deep brain stimulation is one of the widely accepted treatment techniques. However, conventional deep brain stimulation technique provides continuous stimulation without optimizing the stimulation parameters, resulting in adverse side effects and unexpected death. Hence, understanding the dynamic behavior of brain neural networks at a cellular level is required for patient-specific epilepsy treatment. Considering the underlying mechanism of a single neuronal shift in the brain neural network, computational model-based techniques have a new face for healthcare, which aims to develop effective medical devices for preclinical investigations. NEW METHOD: This paper discusses the design of a Smart Deep Brain Stimulator (SDBS) using the Hodgkin-Huxley (HH) conductance-based cellular model of brain neurons to automatically detect, predict and regulate epilepsy against patient-specific conditions. Epileptic activity is simulated as a spike train of action potential due to sodium and potassium channel conductance variations in the single-neuron HH model. The proposed SDBS consists of three components:- i) seizure detection using bagging and boosting-based ensemble machine learning classifiers, ii) channel conductance prediction using Long Short Term Memory-Recurrent Neural Network (LSTM-RNN) based Deep Neural Network (DNN) for updating model parameters of brain neuron, and iii) model-based intelligent control of epileptic seizure with Nonlinear Autoregressive Moving Average-L2 (NARMA-L2) Controller and Nonlinear Model Predictive Controller (NMPC). RESULTS: For effective treatment, improving the overall accuracy and efficiency of SDBS is essential. For epilepsy detection, the ensemble bagging machine learning algorithm provides better accuracy of 92.7% compared to the ensemble boosting algorithm. LSTM-RNN deep neural network model with four layers predicts the variations in channel conductance with Root Mean Square Error (RMSE) of 0.00568 and 0.009081 for sodium and potassium channel conductance, respectively. From the closed-loop performances of SDBS with an intelligent control scheme, it is observed that SDBS with NMPC provides efficient and accurate stimulation with minimum energy consumption. From a stability point of view, SDBS with NMPC provides better stability than SDBS with NARMA-L2 Controller. COMPARISON WITH EXISTING METHOD: The proposed SDBS is designed to generate accurate stimulation pulses for epilepsy patients with specific conditions depending on the neuronal activity of a single neuron. Moreover, it will also adapt to the dynamic condition of epilepsy patients. The existing deep brain stimulator continuously provides stimulation pulses without adapting to the patient's conditions. CONCLUSION: The proposed SDBS could provide patient-specific treatment based on sodium/potassium channel conductance variations of brain neurons. It will help increase the use of deep brain stimulation techniques and reduce sudden death. Furthermore, the proposed technique will be extended to neural network models with larger neuronal populations to improve the practical feasibility.