Feedforward neural network emulation of a PID continuous-time controller for quadcopter attitude digital control

In this paper, the position of a pulse width modulation (PWM)-driven pneumatic actuator has been controlled using a dynamic neural network (DNN) and Proportional Integral Derivative (PID) controller. The harmony search algorithm (HSA) has been used to unravel the optimization problem. The DNN controller is optimally designed to control the position of the actuator. As to the performance of the PID controller, it can assist the DNN controller to give better results. Therefore, an optimal hybrid scheme with both DNN and PID controllers based on HSA is suggested. A pneumatic circuit containing a fast-switching valve is used to reduce the complexity of the PWM-driven servo pneumatic system along with its cost price.

The main objective of this paper is to implement the Fractional Order Proportional Integral Derivative (FOPID) controller for output voltage control of DC - DC Zeta converter. Implementing a distinct control technique for DC-DC converters has also been introduced with the commercial availability of integrated circuits (ICs) to carry out this task. The simulation of Classical Proportional Integral Derivative (PID) and Fractional Order PID (FOPID) control techniques for the control of Zeta converter and its characteristics of a dc-dc converter, where aspects such as the output voltage, required peak overshoot and rise time specifications are investigated using MATLAB/ Simulink model. Through simulation it is confirmed that the propounded FOPID controller gives better the transient response of the feedback system by using a significantly smaller control effort (i.e., less on-off switches), the Internal Absolute Error (IAE) of the variations of the output voltage is minimized and improves the strength of the DC - DC ZETA converter to variations in the input voltage. Artificial Bee Colony (ABC) algorithm offers fine tuning the factors of the FOPID controller.

The propulsion subsystem of multi-rotor Unmanned Aerial Vehicles (UAV) is one of the most complex due to the aerodynamic, aero-elastic and electromechanical elements it comprises. Therefore, accurate models of this subsystem can be difficult to work with. Therefore, simplified models are normally used for the design of control and navigation algorithms. Considering this, the effectiveness of these algorithms is heavily dependent on the identification process used for the estimation of the parameters of the simplified propulsion model. On the other hand, the novel method of fuzzy adaptive neurons (FANs) have interesting characteristics that make them attractive for applications in which a fast response and good precision are required. In this article, the identification of the parameters of the propulsion system and the trajectory tracking of a multi-rotor UAV using FANs is explored. The efficient learning algorithm of the FANs is applied to identify the parameters of a simplified model of the propulsion system and to the self-tuning proportional integral derivative (PID) controllers of the trajectory tracking system. The proposed simplified model with the identified parameters is tested with experimental data obtained with low speed wind tunnels. The proposed PID controllers with self-tuning gains defined by the algorithm of the FANs for trajectory tracking system, are verified with simulations in MATLAB/Simulink® environment. The results showed that the parameter identification and trajectory tracking with PID controllers with self-tuning gains defined by the algorithm of the FANs, are suitable for estimating the parameters of the simplified model and track the trajectory with better error reduction than a classical PID controller.

One of the key drivers in gaining the cost-effective, reliable and low-downtime Operation system for major Oil Fields is implementation of Water Injection System (WIS) with capability of online-automated regulation and control. This Paper outlines an application of Proportional, Integral, Derivative (PID) control loop algorithm for Decentralized WIS (with direct injection from Water Source wells to Injection wells by means of high capacity ESP's) used in West Qurna-2 oilfield located in Iraq, operated by LUKOIL. PID control loop algorithm has being used in the industry for the years, but not commonly applied as integral component of Water Injection System allowing to maintain optimal equipment modes. Such limited PID deployment is explained by complexity of operational parameters synchronizing for every Water Injection System component. The decentralized Water Injection System at the West Qurna 2 project is common and united pipeline system for every WellPad that provides water transfer from Water source wells (3-5 wells per WellPad) to injection wells (3-6 wells per WellPad) with line pressure of 70-130 Barg by means of high-capacity ESP equipped with VSD. Water injection process is regulated by controlling both the conventional ESP parameters (frequency, flow rate, motor current and load, Intake and discharge pressures), line pressure, required injection rate by means of flow control valves, and implemented PID control loop algorithm. This paper discusses the techniques of PID control loop algorithm application as the tool for: –Pipeline protection against high pressure without full shutdown of water injection process and–ESP automated regulation to ensure operation inside Recommended Operated Range (ROR) and smoothening of ESP mode when any changes in the system happen (i.e. shutdown of other water source well ESP's at the same WellPad or shutdown of injection well) with immediate feedback to Operator Control Room. Implemented approach allows to decrease number of ESP shutdowns and avoid high jumping of ESP loads that consequently leads to maximizing of ESP run life. Comparative examples and case studies of submersible equipment operation results before and after the application of PID regulation are specified. Mainly by implementing PID regulation and control stable and smooth operation of equipment and ESP's used for WIS (direct injection from Water Source Well to Injection Well) has been achieved. New approach for PID application as a tool for pipeline overpressure protection has been formed, which allowed to reduce Capital Expenses for the WIS construction works. Specified case can be used as example for further development and projects in the oil and gas industry.

This benchmark suite presents a detailed description of a series of closed-loop control systems with artificial neural network controllers. In many applications, feed-forward neural networks are heavily involved in the implementation of controllers by learning and representing control laws through several methods such as model predictive control (MPC) and reinforcement learning (RL). The type of networks that we consider in this manuscript are feed-forward neural networks consisting of multiple hidden layers with ReLU activation functions and a linear activation function in the output layer. While neural network con- trollers have been able to achieve desirable performance in many contexts, they also present a unique challenge in that it is difficult to provide any guarantees about the correctness of their behavior or reason about the stability a system that employs their use. Thus, from a controls perspective, it is necessary to verify them in conjunction with their corresponding plants in closed-loop. While there have been a handful of works proposed towards the verification of closed-loop systems with feed-forward neural network controllers, this area still lacks attention and a unified set of benchmark examples on which verification techniques can be evaluated and compared. Thus, to this end, we present a range of closed-loop control systems ranging from two to six state variables, and a range of controllers with sizes in the range of eleven neurons to a few hundred neurons in more complex systems.

In this paper, we present experimental validation of a novel method for controlling a robotic manipulator using an experience-based Neural Network (NN) controller derived from Deterministic Learning (DL) techniques. Studies are performed on simulated and real-world environments to quantify performance of this controller compared (as a baseline case) to traditional methods such as Proportional-Integral-Derivative (PID) control. Both of these controller types (NN and PID) are integrated into a 7 Degrees of Freedom (DOF) manipulator system, having the controller available to use on all 7 joints simultaneously. The NN controller utilizes adaptive learning strategies focusing on the state-feedback case. The learning algorithms train Radial Basis Function Neural Networks (RBFNN) to enhance tracking performance along a non-linear recurrent reference trajectory. With convergence of NN weights within the RBFNN, the unmodeled manipulator dynamics are guaranteed to be identified and stored for use with the experience-based algorithms without online adaptation. This paper details the process of integrating a NN controller with system state feedback and employing learning/experience based algorithms in a manner that host system computations are not affected beyond a pre-determined sampling rate.

Due to its simple structure and robustness, the traditional proportional-integral-derivative (PID) controller is commonly used in the field of industrial automation and process control, but it does not function well with nonlinear systems, time-delayed linear systems and time-varying systems. A new type of PID controller based on artificial neural networks and evolutionary algorithms is presented in this paper. An powerful instrument for a highly nonlinear system is the Artificial Neural Network. The interest in the study of the nonlinear system has increased through the implementation of a high-speed computer system,. In complex systems such as robotics and process control systems, the Neuro Control Algorithm is often applied. Systems of process management is also nonlinear and hard to control consistently.. This paper presents a comprehensive analysis in Which is offline trained by a multilayered feed forward back propagation neural network to act as a process control system controller, That is to say, a temperature control device without prior knowledge of its dynamics. Via the implementation of a range of input vectors to the neural network, the inverse dynamics model is developed. Based on these input vectors, the output of the neural network It is being studied by explicitly configuring it to monitor the operation. In this paper, based on set-point adjustment, impact of disturbances in load and variable dead time, compassion between the PID controller and ANN is conducted. The outcome shows that ANN outperforms the controller of the PID.

The objective of this paper is to show the results of the practical implementation of a neural network (NN) tracking controller on a single flexible link and compare its performance to that of proportional derivative (PD) and proportional integral derivative (PID) standard controllers. The NN controller is composed of an outer PD tracking loop, a singular perturbation inner loop for stabilization of the fast flexible-mode dynamics, and an NN inner loop used to feedback linearize the slow pointing dynamics. No off-line training or learning is needed for the NN. It is shown that the tracking performance of the NN controller is far better than that of the PD or PID standard controllers. An extra friction term was added in the tests to demonstrate the ability of the NN to learn unmodeled nonlinear dynamics.

Unmanned aerial vehicles (UAVs) have gained very much importance in both civilian and industrial areas. Especially, in military applications, there are growing demands on the UAVs’ usage for reconnaissance, surveillance, defence, and attack purposes. Therefore, autonomous control of the UAVs has great importance to achieve stable and successful missions. Most of the control techniques in the aircraft control systems use the classical PID (proportional + integral + derivative) control method but this method may have some insufficient response characteristics in terms of changing weather conditions and disturbances. At this point, because of the PID usage in most UAV systems, PID controller parameters must have been obtained for optimum control performance of any state dynamics of the UAV. In this study, longitudinal dynamics are obtained for pitch angle and altitude dynamics of a mini fixed-wing Cessna-182 UAV at 60 km/h flight speed. PID parameters are investigated for the best performance of evolutional genetic algorithms (GA) in terms of different performance indexes such as integrated square error (ISE), integrated absolute error (IAE), time multiply of square error and absolute errors (ITSE, ITAE) and linear quadratic regulator (LQR) cost functions. The simulation results indicate the best performance obtained for the LQR index in the point of overshoot and settling time. In addition, proportioned hybrid ITSE and ITAE cost also developed their performance either for overshoot or settling time. This paper contributes to sustainable development goal (SDG9) by investigating aerial vehicle innovation for a stable and reliable flight purpose in the industrial and civilian applications.KeywordsSDG9UAVGenetic algorithmPID controlLongitudinal dynamic

The increasing use of battery-powered electronic devices necessitates efficient and fast charging solutions. This research aims to optimize battery charging at a charging station using the Maximum Power Point Tracking (MPPT) method integrated with Proportional Integral Derivative (PID) control. This method is chosen to address overshoot issues and speed up the stabilization process. The study begins with a literature review to understand the basic theory of MPPT optimization and PID control. Subsequently, hardware and software for the battery charging station were designed and developed. An Arduino ATMega328 microcontroller was used as the main controller, with an LCD to display charging information. Tests were conducted on 12V batteries with capacities of 5Ah and 9Ah. The results showed that using PID control in MPPT optimization successfully increased battery charging efficiency. The fastest charging duration was achieved for a 12V 5Ah battery with parameters Kp=6, Ki=9, and Kd=4, where the battery was fully charged in 19 minutes, and for a 12V 9Ah battery with parameters Kp=5, Ki=3, Kd=1, where the battery was fully charged in 27 minutes. Conversely, the longest charging duration occurred for a 12V 5Ah battery with parameters Kp=5, Ki=3, Kd=1, where the battery was fully charged in 34 minutes, and for a 12V 9Ah battery with parameters Kp=2, Ki=5, Kd=1, with a charging time of 38 minutes. The designed system was able to reduce input current and increase input voltage to reach the 14.5V set point, ensuring that there was no excessive voltage and current that could prolong battery life. This research demonstrates that the PID method in MPPT optimization is effective in accelerating and optimizing battery charging. Future research is expected to further develop this method to improve battery charging efficiency and duration. Keywords: MPPT, PID, Optimization, Battery Charger, Optimal, PWM ABSTRAK Penggunaan perangkat elektronik yang bertenaga baterai semakin meningkat memerlukan solusi pengisian yang efisien dan cepat. Penelitian ini bertujuan untuk mengoptimalkan pengisian baterai pada stasiun pengisian dengan menggunakan metode Maximum Power Point Tracking (MPPT) yang diintegrasikan dengan kontrol Proportional Integral Derivative (PID). Metode ini dipilih untuk mengatasi masalah overshoot dan mempercepat waktu menuju keadaan stabil. Penelitian dimulai dengan studi literatur untuk memahami teori dasar optimasi MPPT dan kontrol PID. Selanjutnya dirancang dan dibuat perangkat keras serta perangkat lunak untuk stasiun pengisian baterai. Mikrokontroler Arduino ATMega328 digunakan sebagai pengendali utama dengan LCD untuk menampilkan informasi pengisian. Pengujian dilakukan pada baterai aki 12V dengan kapasitas 5Ah dan 9Ah. Hasil pengujian menunjukkan bahwa penggunaan kontrol PID pada optimasi MPPT berhasil meningkatkan efisiensi pengisian baterai. Durasi pengisian yang paling cepat dicapai baterai 12V 5ah dengan parameter Kp=6 Ki=9 dan Kd=4 dimana baterai terisi penuh dalam waktu 19 menit dan baterai 12V 9ah dengan parameter Kp=5, Ki=3, Kd=1 dimana baterai terisi penuh dalam waktu 27 menit. Sebaliknya durasi pengisian terlama terjadi pada baterai 12V 5ah dengan parameter Kp=5, Ki=3, Kd=1 baterai terisi penuh dalam waktu 34 menit dan baterai 12V 9ah dengan parameter Kp=2 Ki=5, Kd=1 dengan waktu pengisian 38 menit. Sistem yang dirancang mampu mengurangi arus input dan meningkatkan tegangan input untuk mencapai set point 14.5V menjaga agar tidak terjadi tegangan dan arus berlebih yang dapat memperpanjang umur baterai. Penelitian ini menunjukkan bahwa metode PID pada optimasi MPPT efektif dalam mempercepat dan mengoptimalkan pengisian baterai. Penelitian selanjutnya diharapkan dapat mengembangkan lebih lanjut metode ini untuk meningkatkan efisiensi dan durasi pengisian baterai. Kata kunci: MPPT, PID, Optimasi, Charger Baterai, Optimal, PWM

This paper presents a comparison of optimization for vehicle steering controls system simulation using several Artificial Intelligence (AI) for optimizing Proportional Integral Derivative (PID) control parameters to suppress errors on lateral motion and the yaw motion of vehicles. This paper compares five kinds of tuning methods of parameter for PID controller, among other are Firefly Algorithm (FA), Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Bat Algorithm (BA) and Imperialist Competitive Algorithm (ICA). The vehicles are represented in the model vehicle with 10 degrees of Freedom of vehicle dynamics system. The simulation results show that the PID control tuned by AI in the vehicle steering control system can adjust the plant output to the desired trajectory so that the stability of the vehicle is maintained. Vehicle yaw error and lateral error can be reduced by using ICA to determine PID parameter. The main advantage of proposed optimization is faster and more accurate compared with PID controller. So the error of the controller is reduced too. The results obtained are of vehicle motion can be maintained in accordance with the desired trajectory with smaller error and was able to achieve higher speeds than with the control system using optimized without parameters. This paper only deals with software simulation to proof the effect of AI optimization. The hardware implementation will be investigated in the next future.

Due to extensive research into DC motor speed regulation, many techniques have been created. The most popular compensating controller in nonlinear systems is the Proportional Integral Derivative (PID) controller. The wide range of control systems to which PID may be applied makes it valuable. PID parameter adjustment is essential since these parameters significantly impact the control system’s stability and effectiveness. The efficacy of PID, Fractional Order Proportional Integral Derivative (FOPID) and Neural Network Controllers (NNC) for controlling the speed of a DC motor is presented in this paper. However, in addition to the most famous traditional controller tuning methods like the Ziegler-Nichols and Cohen-Coon procedures, a Neural Network Tuner is created to optimize the processes by reducing the error between a process variable and its set point. The performance of three distinct controllers and tuning techniques have been examined using time response characteristics. To investigate the performance of DC motor with various controllers, a MATLAB/SIMULINK model is created.

BLDC motors require a high level of precision and efficiency in controlling speed, which is needed in a wide range of applications, such as electric vehicle, robotics, and industrial automation. The classical proportional-integral-derivative (PID) controllers are often characterized by overshoot, a slow rate of convergence, and have a low flexibility to changing operating conditions. As a counter to this we have suggested in this paper a hybrid artificial-intelligence-based control scheme, a combination of a neural-network controller and the standard PID control. A large dataset based on controlled simulations, using PID control, is then applied in training a feed-forward neural network to estimate optimal control behaviour using a supervised learning approach. The neural network that results is an adaptive controller that adjusts dynamically the control voltage in real time as a function of the difference between the current speed and its derivative. The experimental evidence shows that neural-network controller performs better than the traditional PID control in terms of eliminating overshoot (0% vs. 12.13%), the settling time by up to 88% (0.8826 s vs. undefined for PID) and the Integral Absolute Error (IAE) by over 40% (86.96 vs. 151.64). Besides, the AI-based system produces more fluent controlvoltage curves, reducing mechanical forces, and promoting energy savings. This paper highlights the potential of hybrid neural-PID control systems to the high-performance motor control of BLDC motors and outlines future research opportunities in this area to address real-time operation and computational bottlenecks within embedded computer systems.

This paper demonstrates the design and analysis of automatic generation control using intelligent genetic algorithm tuned fuzzy based controller. A two area thermal power system simulated for four different scenarios considers a reheat steam turbine in each area with Generator rate constraints. The Integral Time Squared Error (ITSE) employed to get an objective function for the optimization of controller gains. The simulation results compared with the conventional Proportional Integral Derivative (PID) controller, Genetic Algorithm (GA) tuned PID controller and GA tuned Fuzzy PID controller. The proposed GA tuned Fuzzy based PID Controller can generate the best performance for peak overshoot, undershoot and settling time with step load disturbances. Robustness of the performance of the proposed controller provided with system parametric uncertainties.

Linguistic PID Controllers