Artificial Intelligence-based Control Research Articles

Sensing and Control Strategies Used in FES Systems Aimed at Assistance and Rehabilitation of Foot Drop: A Systematic Literature Review.

Functional electrical stimulation (FES) is a rehabilitation and assistive technique used for stroke survivors. FES systems mainly consist of sensors, a control algorithm, and a stimulation unit. However, there is a critical need to reassess sensing and control techniques in FES systems to enhance their efficiency. This SLR was carried out following the PRISMA 2020 statement. Four databases (PubMed, Scopus, Web of Science, Wiley Online Library) from 2010 to 2024 were searched using terms related to sensing and control strategies in FES systems. A total of 322 articles were chosen in the first stage, while only 60 of them remained after the final filtering stage. This systematic review mainly focused on sensor techniques and control strategies to deliver FES. The most commonly used sensors reported were inertial measurement units (IMUs), 45% (27); biopotential electrodes, 36.7% (22); vision-based systems, 18.3% (11); and switches, 18.3% (11). The control strategy most reported is closed-loop; however, most of the current commercial FES systems employ open-loop strategies due to their simplicity. Three main factors were identified that should be considered when choosing a sensor for gait-oriented FES systems: wearability, accuracy, and affordability. We believe that the combination of computer vision systems with artificial intelligence-based control algorithms can contribute to the development of minimally invasive and personalized FES systems for the gait rehabilitation of patients with FDS.

Artificial intelligence-based control for membrane bioreactor in sewage treatment

Artificial intelligence-based control for membrane bioreactor in sewage treatment

FRT capability improvement of a wind power system using support vector regression and deep neural network models

ABSTRACT The increasing energy demand and environmental conditions have led many countries to favor renewable energy,such as wind and solar,over conventional power plants. As per the Central Electricity Authority (CEA), India’s wind power capacity has grown substantially, reaching 45,154 MW in 2023, from 22,465 MW in 2013. However, integrating wind generators to an existing power system network can reduce transient stability during faults. To prevent wind turbine disconnection during faults, Indian grid code authorities mandate Fault Ride Through (FRT) or Low Voltage Ride Through (LVRT) capability for wind turbines. This paper proposes Dynamic Voltage Restorer (DVR) for enhancing the FRT capability of a constant speed wind turbine. Analysis and simulation are done on a fixed speed wind turbine employing Squirrel Cage Induction Generator (SCIG) under balanced and unbalanced voltage sags. Performance of DVR with two Artificial Intelligence-based controllers, Support Vector Machines(SVM) Regression Model Based Supervised Machine Learning algorithm, and Deep Neural Network (DNN) controller, is analysed and compared. The SVM Regression algorithm, achieves 100% voltage sag compensation under unbalanced voltage sags, reduces torque pulsations to 25% and maintains rotor speed at rated values, enabling the turbine to remain connected to the grid. Additionally it reduces Total Harmonic Distortion to 2%.

Seasonal threshold to reduce false positives for prediction-based outlier detection in building energy data

Recent studies have shown that artificial intelligence-based model predictive control systems have the potential to optimize building operations for energy and cost. But for these systems to function as designed, they require high-quality datasets, which can be challenging when working with actual sensor data that may contain missing or outlier values. Using contaminated data in these carefully tuned machine learning algorithms can cause errors in the prediction results, leading to unexpected building operations. There is a need to clean incoming data in real-time by detecting the outliers in incoming sensor data and correcting them to an appropriate value before incorporating them into the prediction algorithm. This research proposes using a seasonal threshold for prediction-based point outlier detection methods to accurately detect the outliers while maintaining the reliability of the data by reducing false positive detections. The approach combining the global and seasonal threshold showed improvement to the detection accuracy by 83 % and reduced the falsely positives by up to 78 % for the best improvement, making the method a more trustworthy way to detect outliers in building energy data. This method can utilize an already existing model to detect outliers is a promising approach, especially for buildings that have already been implemented or are in the process of implementing an artificial intelligence-based control systems. This will allow the system to predict and control building energy without interruption and with reduced risk of unexpected operation due to corrupted data.

Optimize the energy storage system with an artificial intelligence-based control strategy

Abstract Currently, energy storage systems adopt control strategies based on the crossover approach despite their limited generalization performance. To improve the control effect of the control strategy, the limitation of the SOC of the energy storage element is added. This adds smoothness to the system’s output power, enabling the energy storage element to distribute the power reasonably. Meanwhile, we propose and introduce a DPI control strategy based on PSO optimization into the control strategy of the second-order low-pass filtering method, thereby further enhancing and optimizing the original control strategy. Taking the investment cost of the energy storage system and the stable operation of the system as the objective functions, the constraint function is determined according to the parameters, the PSO algorithm is used to construct a two-layer optimization model of the energy storage system, and the FCEM is introduced to determine the objective weights. Based on the constructed model, an arithmetic example analysis of the energy storage system is carried out using artificial intelligence. Under the steady-state analysis of Buck mode, the low voltage side voltage adjustment time based on the PID control strategy is about 0.108s, and the low voltage side voltage adjustment time based on PSOPID is 0.032s, which is reduced by 0.070s compared with the PID control strategy. Also, when optimizing a system that generates electricity from renewable sources on its own, the two goals of minimizing the investment in the energy storage system and making sure it runs smoothly are met when the system’s fixed power is equal to 6.58 MW and its capacity is less than 267 MWh.

Two Feedback PID Controllers Tuned with Teaching–Learning-Based Optimization Algorithm for Ball and Beam System

The proportional integral derivative (PID) controller continues to be the most popular and widely used in the industry despite the development of various artificial intelligence-based controllers due to its simplicity and ease of use. Setting PID parameters, especially in non-linear systems, is still a significant problem. This study presents a cascade PID tuning technique based on the teaching–learning-based optimization (TLBO) algorithm. The proposed method is tested for the position control of a non-linear ball and beam system. A comparative analysis of the proposed method is done with the conventional tuning method and particle swarm optimization-tuned cascade PID controller. The optimization was carried out using integral time absolute error, integral time error, and integral square error as objective functions. It was observed that the evolutionary algorithm-based controller tuned by TLBO gives a much better response regarding rise time, settling time, and overshoot. To test the effectiveness and validity of the proposed controller robust analysis of the proposed controller is carried out with a step disturbance applied at t = 3 s. The comparative study proves that TLBO-tuned response is better than other controllers. Sensitivity analysis of the proposed controller is also performed by varying system parameters.

Deterministic reinforcement learning for optimized formation control of virtually-coupled trains via performance index monitor

This article presents a deterministic reinforcement learning based optimized formation control for virtually-coupled trains using a novel designed performance index monitor, forming a new paradigm of artificial intelligence-based control for virtually-coupled trains. First, a reinforcement learning-based optimized formation control is designed using backstepping design and neural networks-based actor-critic framework, achieving steady formation separation among multiple trains and providing key theoretical basis to ”moving block with soft wall” (or virtually-coupled) mode, by online approximating the solutions of Hamilton–Jacobi–Bellman equations in both virtual position adjusting and actual speed regulating control signals. Second, a hysteretic-lag performance index monitor is designed to determine the time moments that enough knowledge has been obtained and insufficient knowledge is available in dynamical environments and switch the previously-mentioned reinforcement learning-based optimized formation control to deterministic reinforcement learning manner and vice versa. The idea of deterministic reinforcement learning lies in that as long as the convergences of actor and critic neural networks are observed, speaking equivalently, sufficient knowledge has been obtained provisionally, inner weight vectors of neural networks can be set as deterministic constant ones, achieving the human-like learning and control and reducing the computational amount. Rigorous closed-loop stability using deterministic reinforcement learning is also given using Lyapunov stability theorem. Finally, by utilizing the designed performance index monitor, the switching behavior between reinforcement learning and deterministic reinforcement learning is guaranteed to be finite during finite time of running duration. Simulation examples are given to verify the effectiveness of proposed control.

Evolutionary artificial neural network for temperature control in a batch polymerization reactor

The integration of artificial intelligence techniques introduces fresh perspectives in the implementation of these methods. This paper presents the combination of neural networks and evolutionary strategies to create what is known as evolutionary artificial neural networks (EANNs). In the process, the excitation function of neurons was modified to allow asexual reproduction. As a result, neurons evolved and developed significantly. The technique of a batch polymerization reactor temperature controller to produce polymethylmethacrylate (PMMA) by free radicals was compared with two different controls, such as PID and GMC, demonstrating that artificial intelligence-based controllers can be applied. These controllers provide better results than conventional controllers without creating transfer functions to the control process represented.

Evaluation of greenhouse gas emissions from aerobic and anaerobic wastewater treatment plants in Southeast of Italy

An evaluation of the operative functioning data of 183 Wastewater Treatment Plants (WWTPs) in Apulia (Southeast of Italy) has been carried out aimed to assess their Green House Gases (GHGs) emissions and the level for which the use of anaerobic sludge treatment should be more convenient in terms of electricity consumption and of GHGs emissions. Out of the 183 studies WWTPs, 140 are practicing aerobic digestion of sludge, while the remaining 43 are practicing anaerobic digestion of sludge. WWTPs in Apulia are serving about 4,81 million PE (Population Equivalent), yielding approximately 600,000-ton equivalent CO2 per annum. The production of GHGs emissions has been estimated by evaluating the contribution of CO2 deriving from: a) electric energy consumption (fossil CO2), b) biogenic CO2, c) N2O and d) CH4 emissions. The present study investigates a number of technical measures for upgrading the existing WWTPs, so to reduce GHGs emissions through the amelioration of CH4 production and capture in the anaerobic step, and through reducing the production of biogenic N2O and CO2 emissions in the aerated basin. The methodology employees artificial intelligence-based control for upgrading the aerobic oxidation of the organic carbon and the nitrification-denitrification steps. As a result, GHGs emissions are expected to be reduced by approximately: 71% for CH4, 57% for N2O, 20% for biogenic CO2 and 15% for fossil derived CO2.

Deep Reinforcement Learning Algorithm based PMSM Motor Control for Energy Management of Hybrid Electric Vehicles

Hybrid electric vehicles (HEV) have great potential to reduce emissions and improve fuel economy. The application of artificial intelligence-based control algorithms for controlling the electric motor speed and torque yields excellent fuel economy by reducing the losses drastically. In this paper, a novel strategy to improve the performance of an electric motor-like control system for Permanent Magnet Synchronous Motor (PMSM) with the help of a sensorless vector control method where a trained reinforcement learning agent is used and provides accurate signals which will be added to the control signals. Control Signals referred to here are direct and quadrature voltage signals with reference quadrature current signals. The types of reinforcement learning used are the Deep Deterministic Policy Gradient (DDPG) and Deep Q Network (DQN) agents. Integration and implementation of these control systems are presented, and results are published in this paper. The advantages of the proposed method over the conventional vector control strategy are validated by numerical simulation results.

Artificial Intelligence-based Control Techniques for HVDC Systems

The electrical energy industry depends, among other things, on the ability of networks to deal with uncertainties from several directions. Smart-grid systems in high-voltage direct current (HVDC) networks, being an application of artificial intelligence (AI), are a reliable way to achieve this goal as they solve complex problems in power system engineering using AI algorithms. Due to their distinctive characteristics, they are usually effective approaches for optimization problems. They have been successfully applied to HVDC systems. This paper presents a number of issues in HVDC transmission systems. It reviews AI applications such as HVDC transmission system controllers and power flow control within DC grids in multi-terminal HVDC systems. Advancements in HVDC systems enable better performance under varying conditions to obtain the optimal dynamic response in practical settings. However, they also pose difficulties in mathematical modeling as they are non-linear and complex. ANN-based controllers have replaced traditional PI controllers in the rectifier of the HVDC link. Moreover, the combination of ANN and fuzzy logic has proven to be a powerful strategy for controlling excessively non-linear loads. Future research can focus on developing AI algorithms for an advanced control scheme for UPFC devices. Also, there is a need for a comprehensive analysis of power fluctuations or steady-state errors that can be eliminated by the quick response of this control scheme. This survey was informed by the need to develop adaptive AI controllers to enhance the performance of HVDC systems based on their promising results in the control of power systems. Doi: 10.28991/ESJ-2023-07-02-024 Full Text: PDF

Design MPPT with Anfis Method on Zeta Converter With DC Load

Maximum power point tracking (MPPT) for PV (Photovoltaic) systems is provided in this research using artificial intelligence-based control. The design of MPPT system with Anfis Method on the Zeta Converter with DC Load is used to optimize the work of the Photovoltaic which will be used for DC load sources. The MPPT process consists of four main stages, namely module training data, determining input and output data, determining the number and type of membership functions and ANFIS training data. Zeta converter works like a buck boost, which can increase or decrease the voltage which is an advantage in designing systems with very volatile Photovoltaic sources. Zeta Converter is used to get higher efficiency, smaller input and output current ripple values and smaller core losses in the inductor. To improve the efficiency of system performance, An MPPT algorithm for the adaptive neuro fuzzy inference system (ANFIS) that is programmed into a microcontroller controls the zeta converter. ANFIS control is used because the response is faster and more effective. The combined simulation's findings demonstrate that the ANFIS control was successful, and the system can now produce the best possible power from Photovoltaic ipanelsiiniMPPT mode by boosting efficiency by up to 19.96%.

Artificial intelligence-based control of continuous polymerization reactor with input dead-zone

Artificial intelligence-based control of continuous polymerization reactor with input dead-zone

Artificial Intelligence-Based Control and Coordination of Multiple PV Inverters for Reactive Power/Voltage Control of Power Distribution Networks

The integration of Renewable Energy Resources (RERs) into Power Distribution Networks (PDN) has great significance in addressing power deficiency, economics and environmental concerns. Photovoltaic (PV) technology is one of the most popular RERs, because it is simple to install and has a lot of potential. Moreover, the realization of net metering concepts further attracted consumers to benefit from PVs; however, due to ineffective coordination and control of multiple PV systems, power distribution networks face large voltage deviation. To highlight real-time control, decentralized and distributed control schemes are exploited. In the decentralized scheme, each zone (having multiple PVs) is considered an agent. These agents have zonal control and inter-zonal coordination among them. For the distributed scheme, each PV inverter is viewed as an agent. Each agent coordinates individually with other agents to control the reactive power of the system. Multi-agent actor-critic (MAAC) based framework is used for real-time coordination and control between agents. In the MAAC, an action is created by the actor network, and its value is evaluated by the critic network. The proposed scheme minimizes power losses while controlling the reactive power of PVs. The proposed scheme also maintains the voltage in a certain range of ±5%. MAAC framework is applied to the PV integrated IEEE-33 test bus system. Results are examined in light of seasonal variation in PV output and time-changing loads. The results clearly indicate that a controllable voltage ratio of 0.6850 and 0.6508 is achieved for the decentralized and distributed control schemes, respectively. As a result, voltage out of control ratio is reduced to 0.0275 for the decentralized scheme and 0.0523 for the distributed control scheme.

Artificial Intelligence-Based Control for Reactive Power of Electric Drive System for Pump

Abstract: This paper affords an impression device implementation found out for the electrical force device of a pump station supported analysis, this country of the control approach for the electrical force device of a pump station, and present challenges. The operation’s findings display that this type of control plan is feasible , which with the help of using incorporating computing , it'll successfully keep guide diligence while conjointly sleuthing pump station failures, thence growing the voltage pleasant of the power provide device. A nonlinear proportional operation approaches the transmission sign of the pumping station’s electric force device, and additionally the sign enters the nonlinear perform reckoner of the neural community controller, anyplace the neural community’s nonlinear mapping capacity is hired to understand computing control via the perform. Design questioning gives a framework that allows us. to spice up the proper query on the proper time while constructing partner AI solution. Moreover, it promotes a mind-set that accepts uncertainty, an massive component for varied superior AI comes Keywords: Electric force, Pump station, AI, Power supply system

Economic 6-DOF robotic manipulator hardware design for research and education

This work proposes economic design of robotic manipulator hardware for education and laboratory experiment purposes. This prototype can be used for conducting experiments and validation of various control strategies like robust control design, artificial intelligence-based control design, vision-based control, etc. Designed 6-DOF robotic manipulator is connected to the computer system through Arduino board, which acts as a communication device. High torque DC servo motors are used to control each joint of the manipulator. Frame of the manipulator is designed with mild steel brackets. Robot operating system (ROS), an open-source meta operating system is used to control and perform series of operations on a robotic manipulator. Robotic manipulator is configured and controlled using the MoveIt package. MoveIt will plan the trajectory using inverse kinematics for desired goal state and the same can be visualized through Rviz. Experiment results show that robotic manipulator prototype successfully follow the trajectory planned by the MoveIt and reached to the desired goal state. Because of open-source alternatives like ROS, cost of the prototype design is economical and hence, the prototype can be utilized for experimental validation in research and laboratory purpose.

Safe-visor architecture for sandboxing (AI-based) unverified controllers in stochastic cyber–physical systems

High performance but unverified controllers, e.g., artificial intelligence-based (a.k.a. AI-based) controllers, are widely employed in cyber–physical systems (CPSs) to accomplish complex control missions. However, guaranteeing the safety and reliability of CPSs with this kind of controllers is currently very challenging, which is of vital importance in many real-life safety-critical applications. To cope with this difficulty, we propose in this work a Safe-visor architecture for sandboxing unverified controllers in CPSs operating in noisy environments (a.k.a. stochastic CPSs). The proposed architecture contains a history-based supervisor, which checks inputs from the unverified controller and makes a compromise between functionality and safety of the system, and a safety advisor that provides fallback when the unverified controller endangers the safety of the system. Both the history-based supervisor and the safety advisor are designed based on an approximate probabilistic relation between the original system and its finite abstraction. By employing this architecture, we provide formal probabilistic guarantees on preserving the safety specifications expressed by accepting languages of deterministic finite automata (DFA). Meanwhile, the unverified controllers can still be employed in the control loop even though they are not reliable. We demonstrate the effectiveness of our proposed results by applying them to two (physical) case studies.

Design, Development, and Testing of an Intelligent Wearable Robotic Exoskeleton Prototype for Upper Limb Rehabilitation.

Neuromotor rehabilitation and recovery of upper limb functions are essential to improve the life quality of patients who have suffered injuries or have pathological sequels, where it is desirable to enhance the development of activities of daily living (ADLs). Modern approaches such as robotic-assisted rehabilitation provide decisive factors for effective motor recovery, such as objective assessment of the progress of the patient and the potential for the implementation of personalized training plans. This paper focuses on the design, development, and preliminary testing of a wearable robotic exoskeleton prototype with autonomous Artificial Intelligence-based control, processing, and safety algorithms that are fully embedded in the device. The proposed exoskeleton is a 1-DoF system that allows flexion-extension at the elbow joint, where the chosen materials render it compact. Different operation modes are supported by a hierarchical control strategy, allowing operation in autonomous mode, remote control mode, or in a leader-follower mode. Laboratory tests validate the proper operation of the integrated technologies, highlighting a low latency and reasonable accuracy. The experimental result shows that the device can be suitable for use in providing support for diagnostic and rehabilitation processes of neuromotor functions, although optimizations and rigorous clinical validation are required beforehand.

Artificial Intelligence-based Security Control Construction and Countermeasures

Artificial Intelligence-based Security Control Construction and Countermeasures

Pengaplikasian Kalman Filter sebagai Pengendali dalam Permainan The Open Racing Car Simulator (TORCS)

The Open Racing Car Simulator (TORCS) works as both a playable game and a framework to develop artificial intelligence-based controllers. As a platform for researchers, TORCS has become a platform in controller development with various approaches in artificial intelligence using sensors and actuators provided by the SCR Server. In this research, the author develops a controller using the Kalman Filter, an algorithm to predict and measure states based on previous measurements to determine future trajectory.