Design of neural network PID controller based on E-FRIT

In this paper, an innovative design method of PID controller based on genetic algorithm tuning parameters, VHDL description and FPGA implementation is proposed. The parameters of the PID controller are optimized by using genetic algorithm. The intelligent PID controller based on genetic algorithm is realized by using FPGA. During the realization process, the functional modules of the system are divided based on top-down design method. Each module of the system is described with VHDL language. Finally, the closed loop tests of the intelligent PID controller based on FPGA and genetic algorithm are implemented by means of DSP Builder and Matlab/Simulink. Simulation results have shown that the design method of the intelligent PID controller based on FPGA and genetic algorithm is verified, and it has some advantages such as design flexible, self-tuning on line, high reliability, short cycle of technical development, high running speed and so on. This intelligent PID controller based on FPGA and genetic algorithm can be applied in high control precision digital control system with high speed responsibility and high stability.

A new design method for PID controller using user-specified gain margin and phase margin for delayed systems has been proposed. The proposed PID controller is particularly useful in auto-tuning. The controller parameters are calculated using several approximations and elimination processes. The robustness of the proposed controller is verified by different simulation examples.

Although the traditional PID controller is widely used in many fields, the system parameters varying and external disturbances existing in the DC servo system will cause large overshoot or poor stability. To improve the performance of the PID controller, a compound servo control system combining the conventional PID control and the fuzzy control is presented to meet the demand of a vehicular antenna servo system in this paper. Incorporating the fuzzy control and the conventional PID control, this paper presents a design method of the fuzzy PID controller that is based on the fuzzy tuning rules and formed by integrating two above control ideas. Simulation results are presented to show the efficiency of the proposed controller. The practical control effect shows that the control system that adopts the fuzzy PID controller has better performance than that of the traditional PID control system, and meets the performance requirements of the servo system.

In this paper, a frequency response based design method of PID controller is proposed for higher order (HO)/-plus delay time (HOPDT) systems. The HO/-PDT models are converted into real and imaginary part at a frequency where a criteria similar to amplitude and phase margin is utilized to get the constraints on the parameters of the controllers. To reject noise and load disturbances, the constraints given by sensitivity and complementary sensitivity function are also considered. The proposed method has superior features for systems with high order, time delays and monotonic or oscillatory responses. Three simulation examples are included to show the effectiveness, simplicity and applicability of the proposed method.

In the process of frozen food production, its refrigeration system is a typical large inertia, nonlinear, strongly coupled system, which is prone to large overshooting during operation, resulting in cost increase and quality control problems. This paper solves the problem of large overshoot in the digital PID control process by designing a fuzzy control, and designs a practical fuzzy controller so that the design method of the fuzzy PID controller can be realized in the simulation and actual control of the freezing system. A genetic algorithm optimization scheme is also designed for the fuzzy PID control system to optimize the fuzzy controller’s affiliation function, control rules, as well as proportionality and quantization factors, to enhance the model’s dynamic characteristics and robustness. Design of food quick-freezing simulation experiments and use the model in the refrigeration system control in practice, compare the performance of this model with the PID, fuzzy PID model, and found that the temperature difference set value of 4 ℃ and 3 ℃, respectively, the parameters of the traditional PID controller in the commissioning stage of the system after the end of the rectification to remain fixed, which makes the actual quality of the cooling water temperature difference control degradation. And the fuzzy genetic PID controller still has good control quality. In this paper, the model curve converges rapidly, the overshoot is only 3.9%, and the regulation time is only 450M seconds. The designed method of the study is feasible and has high industrial application value.

Based on the ant colony system (ACS) algorithm and fuzzy logic control, a new design method for optimal fuzzy PID controller was proposed. In this method, the ACS algorithm was used to optimize the input/output scaling factors of fuzzy PID controller to generate the optimal fuzzy control rules and optimal real-time control action on a given controlled object. The designed controller, called the Fuzzy-ACS PID controller, was used to control the CIP-I intelligent leg. The simulation experiments demonstrate that this controller has good control performance. Compared with other three optimal PID controllers designed respectively by using the differential evolution algorithm, the real-coded genetic algorithm, and the simulated annealing, it was verified that the Fuzzy-ACS PID controller has better control performance. Furthermore, the simulation results also verify that the proposed ACS algorithm has quick convergence speed, small solution variation, good dynamic convergence behavior, and high computation efficiency in searching for the optimal input/output scaling factors.

This paper proposes a fractional order PID analytical design framework for industrial robot based on digital twin. The effectiveness of the digital twin real-time optimization framework is verified by the simulation of PMSM motor speed control. The digital twin framework implemented in this paper consists of three following parts: multi-domain modeling, behavioral matching and control optimization. Firstly, the physical system is modeled to realize the omni-directional monitoring of robot and the kinematic simulation of single-axis motor. A model similar to the real physical system can be obtained in the step of behavioral matching, which is conducive to the next step of controller parameter optimization. The analytical design method of the five-parameter fractional order PID controller is then presented to optimize the parameters based on the accurate model after behavioral matching, so that the control system can meet the given frequency domain specifications and the required tracking performance. The simulation results show that the fractional order PID controller is robust to loop gain changes. In terms of time domain performance specifications like rise time, overshoot, and adjustment time, fractional order PID controller performs better than integer order PID controller.

This paper compares two approaches to tune PI controllers in a multivariable process. A modified Direct Synthesis method extends the technique to a multivariable system. Results show that the modified Direct Synthesis technique relies on model accuracy and suitability of the desired transfer function and process model. The effectiveness of this technique may break down when these conditions are not met. Genetic algorithms provide a method based on the evolution of species via natural selection. The GA method allows the use of a specific structure, such as PI controller, and then gets the best possible results for that structure. The concept has been adapted into an optimization technique that can be used to find the best controller parameters to meet user defined transient specifications for a complex process. Thus, the robustness of the method allows it to outperform the former technique. Simulation results are shown to illustrate the comparisons between these two methods and the effectiveness of Gas.

The photovoltaic (PV) system’s nonlinear nature, uncertainty in temperature, and light intensity in the environment are responsible for the fluctuation of the maximum power point (MPP). Many maximum power point tracking (MPPT) algorithms are proven effective in the literature, including use of artificial intelligence. Proportional integral derivative controller and fuzzy controllers found effective to track MPP. Using the combination of Fuzzy-PID controller, swiftly adapt to changes in the atmosphere to confirm the solar array is continually operating at its highest efficiency with MPP. Easy and effective oscillation elimination is possible with PID control. To regulate the maximum power of the PV system combination of Fuzzy-PID self-tuning controllers is found effective.The goal is to enhance a PV system’s dynamic response to in presence of irradiation and temperature fluctuations. For tuning Fuzzy logic controller (FLC), gain scheduler values of the proportional, integral, and derivative constants are fundamental. The design processes for non-linear conventional PID controllers, FLC, and series combination of Fuzzy-PID controller are discussed. It compares the traditional PID controller, FLC, and Fuzzy-PID controller. The comparison of outputs produced for PID, FLC, and series Fuzzy-PID controller confirms the effectiveness of the algorithm. Use of Fuzzy controller helps in self-tuning of PID controller constant value. MATLAB/Simulink based environment used to perform the simulations.

The technologic of PID control is very conventional. There is an extensive application in many fields at present. The PID controller is simple in structure, strong in robustness, and can be understood easily. Then neural networks have great capability in solving complex mathematical problems since they have been proven to approximate any continuous function as accurately as possible. Hence, it has received considerable attention in the field of process control. Due to the complication of modern industrial process and the increase of nonlinearity, time-varying and uncertainty of the practical production processes, the conventional PID controller can no longer meet our requirement. This paper introduces the theoretical foundation of the BP neural network and studying algorithm of the neural network briefly, and designs the PID temperature control system and simulation model based on BP neural network. Keywords-PID controller; BP neural network; temperature control system

The design and implementation of a multi-stage PID (MS-PID) controller for non-inertial referenced UAVs are highly complex. Symmetrical multirotor UAVs are unstable systems, and it is thought that the kinematics of the symmetrical UAV rotor, such as the quadrotor and hexacopter resembles the kinematics of an inverted pendulum. Several researchers have investigated the structure and design of PID controllers for high-order systems during the last decade. The designs were always concerned with the enhanced response, robustness, model reduction and performance of PID controllers. An accurate tuning process of such a controller depends on the engineer’s experience level. This is due to the number of variables and hyperparameters tuned during the process. An adaptive genetic algorithm (AGA) is utilized to optimize the MS-PID controllers for controlling the quadrotor in this study. The proposed method optimizes the offline-planned approach, providing several possibilities for adapting the controllers with various paths and or varying weather conditions. The MS-PID parameters are optimized in parallel, as every PID controller affects the other controller’s behavior and performance. Furthermore, the proposed AGA generates new chromosomes for “new solutions” by randomly developing new solutions close to the previous best values, which will prevent any local minima solution. This study intends to investigate the design and development of a highly tuned robust multi-stage PID controller for a symmetrical multirotor UAV. The work presents a model for a non-referenced inertial frame multirotor UAV (quadcopter). Once the model is defined, a robust multi-stage PID controller for the non-inertial referenced frame symmetrical multirotor UAV is designed, tuned, and tested. A genetic algorithm (GA) will be used to tune the MS-PID controller. Finally, the performance comparison between the proposed and conventional methods is presented. The results show that the proposed method provides stability improvement, better transient response, and power consumption.

Background/Objectives: The objective is to design decentralised PID Controller for 3x3 multivariable system using heuristic algorithms. Methods/Statistical analysis: It is very difficult to control Multivariable system is in the process industries due to the interaction between the process variables. It needs a decentrlised PID controller scheme for the effective control. Here, Ogunnaike and Ray distillation column is considered as an example for simulation. The PID controllers are tuned using using Firefly (FA) and Particle Swarm Optimization (PSO) algorithm for diagonal elements of the transfer function. Findings: From the Simulation it is observed that the performance of Firefly is better than Particle Swarm Optimization for the chosen process model in terms of Integral Square (ISE) and peak overshoot. Application/Improvements: This method of controller design can be used for other distillation column (Example: Wood-berry distillation column). Different optimization algorithm can be used for controller design (Ant Colony optimization algorithm).

To address slow convergence speed and low convergence accuracy in the inverse model (IM) learning and training of the controlled object during BP neural network (BP-NN) feedforward compensation control for tracking a step signal, which results in poor dynamic and steady-state performance of the control system. This study introduces an improved Adam algorithm (IADAM)-enabled BP-NN feedforward compensation control method. This method introduces a dynamic learning rate adjustment mechanism based on the Adam algorithm, which adjusts the learning rate in phases based on the NN training parameter gradients. When the gradient exceeds a preset threshold, the learning rate is dynamically increased via an adjustment function to accelerate IM convergence. When the gradient is below the threshold, the maximum historical second-order moment estimate of the gradient is retained to keep the learning rate at a lower value, thereby improving IM convergence accuracy. The control system is composed of a PID controller, a neural network controller (NNC), and an IADAM-based neural network identifier (IADNNI). The PID controller provides online learning samples. The IADNNI learns the IM of the controlled system online, while the NNC shares the same structure and parameters as the IADNNI, generates real-time compensation control signals. Experimental results from a DC servo motor position control platform reveal that the proposed control method achieves zero overshoot in tracking both continuous step signals and square signals, while the settling time is reduced by 60.05% and 39.53%, respectively, compared to the ADAM + BP + PID control method. resulting in significant improvements in both dynamic and steady-state performance.

This paper presents a method for PID controller design which can achieve dominant pole assignment using two of the controller parameters. The non-dominant poles are restricted on the left of the line s = σ̂, where σ̂ is the minimum feasible value, called as the feasibility border. It is obvious that a dominant pole assignment is not practical if σ̂ is close to the real parts of the required dominant poles. Hence, finding σ̂ for a given system is very important. The method, which parameterizes all such controllers in order to allow further design criteria, can be applied to other kinds of low-order compensators.

In this paper, model based PID controller design method is presented for single variable and multivariable processes with enhanced performance. Adequate responses can be anticipated for processes with various dynamics like high-order, large dead times, and oscillatory response. The design method is used first order plus dead time (FOPDT) modelling technique. For controller design higher order process models are approximated in the form of FOPDT and obtained controller is applied to original higher order process model. Simulation examples are given to show the efficacy and litheness of the controller in handling processes of different characteristics. Finally, decentralised PID controller is designed and simulated for two input two output Wood-Berry distillation process.