Neural Network Predictive Control Research Articles

Kinematic analysis and advanced control of a vectored thruster based on 3RRUR parallel manipulator for micro-size AUVs

Abstract Autonomous underwater vehicles (AUVs) have played a pivotal role in advancing ocean exploration and exploitation. However, traditional AUVs face limitations when executing missions at minimal or near-zero forward velocities due to the ineffectiveness of their control surfaces, considerably constraining their potential applications. To address this challenge, this paper introduces an innovative vectored thruster system based on a 3RRUR parallel manipulator tailored for micro-sized AUVs. The incorporation of a vectored thruster enhances the performance of micro-sized AUVs when operating at minimal and low forward speeds. A comprehensive exploration of the kinematics of the thrust-vectoring mechanism has been undertaken through theoretical analysis and experimental validation. The findings from theoretical analysis and experimental confirmation unequivocally affirm the feasibility of the devised thrust-vectoring mechanism. The precise control of the vector device is studied using Physics-informed Neural Network and Model Predictive Control (PINN-MPC). Through the adoption of this pioneering thrust-vectoring mechanism rooted in the 3RRUR parallel manipulator, AUVs can efficiently and effectively generate the requisite motion for thrust-vectoring propulsion, overcoming the limitations of traditional AUVs and expanding their potential applications across various domains.

Physics-Informed Neural Network Modeling and Predictive Control of District Heating Systems

Physics-Informed Neural Network Modeling and Predictive Control of District Heating Systems

Neural network predictive controller based on the improved TPA-LSTM model for ultra-supercritical units

To mitigate the impact of large-scale renewable energy power on the national grid in China, it is imperative to enhance the flexible peaking capability of coal-fired thermal power units. The coordinated control system, central to the load control of coal-fired units, faces challenges such as multivariable coupling, sluggish response, and uncertain coal quality parameters. This paper introduces a neural network predictive controller based on the improved TPA-LSTM model, aimed at addressing these issues. Initially, a data-driven control model is established to break through the limitations of traditional linear predictive control and effectively handle disturbance uncertainties. Then, a multivariable coordinated control strategy based on the neural network controller is designed, achieving effective decoupling of multiple parameters and ensuring high adaptability across all load conditions. Additionally, by integrating an automatic model updating mechanism, the system can recalibrate in real-time when model mismatches occur due to equipment aging, maintenance, or changes in coal quality, thereby enhancing overall control performance. Simulation results demonstrate that this strategy has excellent control effectiveness, meeting the flexible peaking demands of 1000 MW ultra-supercritical units. The calibration feature of the data-driven model significantly improves control performance following model mismatches.

Analysis of DC motor for process control application using neural network predictive controller

A DC motor is a critical actuator in process control systems. This study investigates the effectiveness of a deep learning (DL) based Neural Network Predictive Controller (NNPC) for precise DC motor speed control. The NNPC anticipates the motor’s future behaviour based on its current state and control inputs. The controller then optimally generates inputs to minimise tracking errors and enhance system performance. The NNPC demonstrated a remarkable reduction in Mean Squared Error (MSE), achieving a training MSE of 2.75 × 10−14 and the best validation MSE of 9.2023 × 10−14. These quantitative outcomes affirm the reliability and robustness of the proposed NNPC for speed control in DC motor systems across diverse applications.

Neural network predictive control for fault detection and identification in DFIG with SMES for low voltage ride-through requirements

The increasing reliance on renewable energy sources brings to light the operational challenges of doubly fed induction generators (DFIGs), particularly under grid disturbances. This study introduces an innovative approach employing Neural Network Predictive Control (NNPC) for fault detection and identification (FDI) in DFIG systems, integrated with Superconducting Magnetic Energy Storage (SMES) to meet Low Voltage Ride-Through (LVRT) requirements. Our method uniquely combines NNPC for dynamic wind speed prediction and precise control of rotor-side and grid-side converters with the stability enhancement offered by SMES. Simulation results demonstrate a notable improvement in DFIG performance under variable wind conditions and grid failures. The system-maintained voltage stability at the Point of Common Coupling (PCC) with less than 5% deviation during transients and ensured consistent power output, marking a significant advancement in LVRT compliance. However, the model assumes steady-state grid conditions and does not account for extreme weather scenarios, indicating areas for future research. This study's findings contribute valuable insights into the robust operation of DFIGs within modern renewable energy grids.

Neural network predictive control in renewable systems (HKT-PV) for delivered power smoothing

The reduction of power fluctuations from intermittent renewable sources is one of the most pressing challenges today. Recent research has shown that prediction and control mechanisms, when combined with energy storage systems, significantly contribute to improving these techniques. However, substantial research gaps still exist regarding the optimization of energy storage system operability. This article introduces an innovative power smoothing method based on neural network predictive control, in conjunction with the exponential moving average method. The proposed approach encompasses the ability to substantially reduce energy fluctuations, optimize battery state of charge, and mitigate ramp rates, thereby preventing deep discharges that shorten battery lifespan. Furthermore, the control system's primary objective is to optimize energy exchange with the grid, surpassing the performance offered by other conventional power smoothing methods. The control system excels in optimizing energy exchange within the network, surpassing conventional methods. Extensive testing on the University of Cuenca microgrid reveals a consistently more stable and higher battery charge compared to conventional methods. Numerical results for underscore the method's effectiveness with a fluctuation suppression rate of 30.78 % compared to 34.85 % (low pass filter) and 36.22 % (ramp rate) methods respectively. The enhanced voltage profiles at the common coupling point ensure the delivery of high-quality and stable power.

Neural network predictive control of converter inlet temperature based on event‐triggered mechanism in flue gas acid production

AbstractThe process of smelting non‐ferrous metals results in significant emissions of flue gas that contains sulfur dioxide (SO), which is very harmful to the environment. Through precise control of converter inlet temperature, it is feasible to enhance the conversion ratio of SO and simultaneously mitigate environmental pollution by generating acid from flue gas. Because of the high degree of uncertainty in smelting process, converter inlet temperature is challenging to regulate and controller frequently needs updating. To improve control performance and decrease controller update times, an event‐triggered neural network model predictive control (ETNMPC) strategy is proposed. First, long short‐term memory (LSTM) prediction model and model predictive controller are developed. Second, it is decided whether to update the existing controller by designing an event‐triggered mechanism. Finally, using real data from a copper facility in Jiangxi Province, the temperature control experiment of converter inlet is carried out. Simulation results demonstrate that the proposed ETNMPC outperforms conventional time‐triggered method in terms of control performance, greatly lowers the times of controller updates, and significantly lowers computation costs and communication burden.

Energy-Saving Testing System for a Coal Mine Emulsion Pump Using the Pressure Differential Flow Characteristics of Digital Relief Valves

Most energy-saving testing methods for plunger pumps use hydraulic motors. The loading test of coal mine emulsion pumps generally uses an overflow valve as the loading unit, which is characterized by high energy consumption. The coal mine emulsion pump uses emulsion as the transmission medium, and the viscosity and lubricity of the emulsion are much lower than those of hydraulic oil, which creates great difficulties in the development of high water-based hydraulic products. The nominal flow rate of the emulsion motor is much smaller than that of the emulsion pump, and there is no mature and reliable water-based flow control valve. Based on the above reasons, traditional energy-saving testing methods cannot be utilized for the testing process of emulsion pumps. The loading test of emulsion pumps generally uses an overflow valve as the loading unit, and during the testing process, all electrical energy is converted into internal energy, resulting in very high energy consumption. This article proposes an energy-saving testing system for emulsion pumps based on multiple emulsion motors in parallel. In order to solve the flow regulation problem of each parallel branch, a flow-intelligent control algorithm is proposed that utilizes the pressure difference flow characteristics of digital relief valves combined with artificial neural network predictive control. Firstly, the feasibility of the proposed system and method is theoretically verified through the analysis of the mathematical model of the digital relief valve. Secondly, further verification is carried out by establishing simulation and testing platforms. The simulation results show that the energy recovery efficiency of the system exceeds 53%. The experimental results show that the proposed testing system has a pressure control error of less than 1%, a flow control error of about 5%, and a maximum overshoot of about 9 L/min relative to the steady-state flow rate. The control accuracy and system stability are high.

A fuzzy based intelligent scheme for enhancing the performance of the optimal controllers by online weighting matrix selection in seismically excited nonlinear buildings

In the present work, an effective strategy is proposed to improve the efficiency of the optimal controllers that are based on minimizing a performance index by using a fuzzy controller. In the used fuzzy system, the weighting matrices of the optimal controller are tuned online according to the response of the structural system. In order to assess the effectiveness of the proposed control scheme, its function is investigated in two phases on a 3-story nonlinear benchmark building equipped with MR dampers. In the first phase, a linear quadratic Gaussian (LQG) controller is combined with a fuzzy logic system (Fuzzy-LQG) and the efficiency of the used technique is compared with three different cases of the conventional LQG controllers. The results show that the Fuzzy-LQG, in addition to using an optimal level of demand energy, has a better performance than the conventional LQG methods in controlling the maximum responses of the structure. In the second phase, the performance of a polynomial optimal controller that is a summation of polynomials in nonlinear states in combination with the fuzzy system is investigated. At this stage, the effectiveness of the designed control system (Fuzzy Polynomial) in terms of seismic performance criteria with the results of a conventional LQG active control system, an optimal adaptive neuro-fuzzy inference system (ANFIS) and a neural network predictive control (NNPC) system is evaluated. The results show that this method is successful in improving the structural performance of a nonlinear building subjected to earthquakes.

An AI framework integrating physics-informed neural network with predictive control for energy-efficient food production in the built environment

Relieving the stress from energy demand is critical for encouraging the application of built environment in agriculture, which is the most energy-intensive food-production sector. In this study, to enhance energy efficiency and crop growth optimization for the built environment, we present a novel artificial intelligence (AI)-based control framework which combines a physics-informed neural network and data-driven robust model predictive control. The physics-informed neural network is constructed for accurately predicting the built environment’s indoor climate and crop states based on control inputs, including controls of temperature, humidity, CO2, irrigation, and fertilization. Data-driven robust model predictive control uses machine learning techniques to account for unpredictable changes in weather conditions to make the best control decisions for actuators. Two examples of the implementation of the proposed AI-based control framework in tomato cultivation environments in Ithaca, New York and Tucson, Arizona are shown to showcase its effectiveness in varying climates. The results show our proposed AI-based control framework can maintain the ideal cultivation environment and, on average, lower cost by 46.4% compared to the conventional control approaches.

Neural network predictive control applied to automotive paint drying and curing processes

In the automotive painting process, maintaining optimal operating conditions in drying and curing ovens is crucial to ensure high-quality paint finishes, particularly for specific car body-in-white (BIW) parts' temperature profiles. This paper presents a methodology for developing and implementing a Neural Network Predictive Control (NNPC) system for painting drying and curing processes in an automotive oven used in the electrodeposition stage (Elpo oven). The objective is to enhance temperature control for BIW parts. To train the Artificial Neural Networks (ANN) in the various zones of the Elpo oven, a dataset was generated using a phenomenological model based on first principles. Random disturbances were applied to the input variables (conveyor speed and zone temperatures) to capture the dynamic response of the output variables (temperature in specific BIW parts). The dataset was split into training (80 %), validation (10 %), and test (10 %) sets, and the ANN training was performed using the backpropagation algorithm. The implemented NNPC utilizes the trained ANNs to predict future temperature values in specific BIW parts (controlled variables). To determine the optimal control signals (manipulated variables), an optimizer based on the Generalized Predictive Control (GPC) model was employed, and the objective function was minimized using the Ant Colony Optimization Algorithm (ACO). Through simulations of four operational scenarios with applied disturbances, the results demonstrate satisfactory performance of the NNPC, effectively maintaining controlled temperatures of BIW parts close to predefined setpoints. The utilization of NNPC offers improved temperature control for BIW parts, mitigating painting issues, reducing rework and operating costs, while ensuring painting quality is uncompromised.

Neural Network Predictive Control for Improved Reliability of Grid-Tied DFIG-Based Wind Energy System under the Three-Phase Fault Condition

This research explores a distinctive control methodology based on using an artificial neural predictive control network to augment the electrical power quality of the injection from a wind-driven turbine energy system, engaging a Doubly Fed Induction Generator (DFIG) into the grid. Because of this, the article focuses primarily on the grid-integrated wind turbine generation’s dependability and capacity to withstand disruptions brought on by three-phase circuit grid failures without disconnecting from the grid. The loading of the grid-integrated power inverter causes torque and power ripples in the DFIG, which feeds poor power quality into the power system. Additionally, the DC bus connection of the DFIG’s back-to-back converters transmits these ripples, which causes heat loss and distortion of the DFIG’s phase current. The authors developed a torque and power content ripple suppression mechanism based on an NNPC to improve the performance of a wind-driven turbine system under uncertainty. Through the DC bus linkage, it prevented ripples from being transmitted. The collected results are evaluated and compared to the existing control system to show the advancement made by the suggested control approach. The efficacy of the recommended control methodology for the under-investigation DFIG system is demonstrated through modelling and simulation using the MATLAB Simulink tool. The most effective control technique employed in this study’s simulations to check the accuracy of the suggested control methodology was the NNPC.

A Data-Driven Model Predictive Control for Quadruped Robot Steering on Slippery Surfaces

In this paper, the locomotion and steering control of a simulated Mini Cheetah quadruped robot was investigated in the presence of terrain characterised by low friction. Low-level locomotion and steering control were implemented via a central pattern generator approach, whereas high-level steering control manoeuvres were implemented by comparing a neural network and a linear model predictive controller in a dynamic simulation environment. A data-driven approach was adopted to identify the robot model using both a linear transfer function and a shallow artificial neural network. The results demonstrate that, whereas the linear approach showed good performance in high-friction terrain, in the presence of slippery conditions, the application of a neural network predictive controller improved trajectory accuracy and preserved robot safety with different steering manoeuvres. A comparative analysis was carried out using several performance indices.

A Comprehensive Overview on Performance of Cascaded Three Tank Level System using Neural Network Predictive Controller

A Neural Network Predictive Controller (NNPC) is a deep learning-based controller (DLC) that uses artificial neural networks (ANN) to predict the future behavior of a system and accordingly control its outputs. In this paper, an NNPC was used to predict the level of the three cascaded tank and then adjust the inputs as flow rate to maintain the desired level in the tank. A three-tank level system is a system consisting of three interconnected tanks used to store liquids. To achieve the desired level, the NNPC first collects data on system behavior, including inputs and outputs, and uses this data to train the neural network. The trained network was then used to make predictions about the future level of each tank and to generate control signals to adjust the inputs as needed. NNPC also incorporates feedback from the system to continuously refine its predictions and improve its control performance over time. The mean squared error (MSE) of different backpropagation training algorithms available in MATLAB deep learning toolbox were evaluated and presented. Based on the MSE and best validation, Levenberg Marquardt algorithm were used in NNPC controller for further step response tracking. Different performance metrics were evaluated and presented.

Design of autonomous vehicle trajectory tracking controller based on Neural Network Predictive Control

When the linear Model Predictive Control (MPC) algorithm is used for trajectory tracking control of autonomous vehicles in high-speed turning conditions, the tracking accuracy is usually decreased due to the low fidelity of the predictive model. To deal with this problem, a trajectory tracking controller based on Neural Network Predictive Control (NNPC) is designed in this paper. The data collected from driving simulation experiments is used for the Back Propagation Neural Network (BPNN) training process to obtain the vehicle state predictive models. To reduce the iteration times of the rolling optimization module of NNPC, Particle Swarm Optimization (PSO) with Fitness Allocating Inertia Weights (PSO-FAIW) algorithm is proposed, which can allocate inertia weights in light of the distance between each particle’s fitness and the best fitness of the swarm. Finally, the performance of the controller designed in this paper is verified based on MATLAB /Simulink. The simulation results demonstrate that the controller proposed in this paper can give consideration to tracking accuracy and real-time performance.

Neural-Network-Based Nonlinear Model Predictive Control of Multiscale Crystallization Process

The purpose of this study was to develop an integrated control strategy for multiscale crystallization processes. An image analysis method using a deep learning neural network is used to measure the fine-scale information of the crystallization process, and the mathematical statistical method is adopted to obtain the mean size of the crystal population. A feedforward neural network is subsequently trained and employed in a nonlinear model predictive control formulation to obtain the optimal profile of the manipulated variable. The effectiveness of the proposed nonlinear model predictive control method is evaluated using alum cooling crystallization experiments. Experimental results demonstrate benefits of the proposed combination of feedforward neural network and nonlinear model predictive control method for the multiscale crystallization process.

Physics-guided neural network and GPU-accelerated nonlinear model predictive control for quadcopter

Physics-guided neural network and GPU-accelerated nonlinear model predictive control for quadcopter

Auto tuned PID and neural network predictive controller for a flow loop pilot plant

This paper presents an advanced control strategy that uses the neural network (NN) predictive controller to govern the dynamics of a Flow loop pilot plant. The set point tracking using NN Predictive controller and conventional PID strategies are investigated. The control objective is to keep the outlet flow of the Single Input Single Output Flow loop at a reference value. An orifice meter as a flow measuring device and Variable Frequency Drive as a final control element is selected to collect Input, Output data. System Identification toolbox in MATLAB is used to construct the model of a flow loop from measured input–output data. Various models are estimated, validated and realized that the transfer function model gives the best fit. This model is used to build the PID and NN Predictive controller. NN predictive controller parameters are set, by the designer's expertise. The performance of a NN predictive controller is then compared with the conventional auto-tuned PID controller. Experimental results are presented, which emphasizes the superiority and effectiveness of the NN predictive controller.

Combustion timing control of HCCI engine based on NNPC and Elman‐NN model under complex conditions

AbstractHomogeneous Charge Compression Ignition (HCCI) combines the characteristics of gasoline engine and diesel engine with high thermal efficiency and low emissions. However, since there is no direct initiator of combustion, it is difficult to control the combustion timing in HCCI engines under complex working conditions. In this paper, Neural Network Predictive Control (NNPC) for combustion timing of the HCCI engine is designed and implemented. First, the black box model based on Elman neural network is designed and developed to estimate the combustion timing. The fuel equivalence ratio, intake valve closing timing, intake manifold temperature, intake manifold gas pressure, and engine speed are chosen as the system inputs. Then, a NNPC controller is designed to control combustion timing by controlling the intake valve closing timing. Simulation results show that the Elman neural network black box model is capable of estimating the HCCI engine combustion timing. In addition, regardless of whether the HCCI engine is in constant or complex condition, the designed NNPC controller is capable of keeping the combustion timing within the ideal range. In particular, under New European Driving Cycle (NEDC) working conditions, the maximum overshoot of the controller is 28.95% and the average error is 1.03 crank angle degree. It is concluded that the controller has good adaptability and robustness.

Theoretical and Experimental Analysis of a New Intelligent Charging Controller for Off-Board Electric Vehicles Using PV Standalone System Represented by a Small-Scale Lithium-Ion Battery

Electric vehicles are rapidly infiltrating the power grid worldwide, initiating an immediate need for a smart charging technique to maintain the stability and robustness of the charging process despite the generation type. Renewable energy sources (RESs), especially photovoltaic (PV), are becoming the essential source for electric vehicle charging points. The stochastic behavior of the PV output power affects the power conversion for regulating the battery charger voltage levels, which influences the battery to overheat and degrade. This study presents a PV standalone smart charging process for off-board plug-in electric vehicles, represented by a small-scale lithium-ion battery based on the multistage charging currents (MSCC) protocol. The charger comprises a DC–DC buck converter controlled by an artificial neural network predictive controller (NNPC), trained and supported by the long short-term memory recurrent neural network (LSTM). The LSTM network model was utilized in the offline forecasting of the PV output power, which was fed to the NNPC as the training data. Additionally, it was used as an alarm flag for any possible PV output shortage during the charging process in the long- and short-term prediction to be supported by any other electricity source. The NNPC–LSTM controller was compared with the fuzzy logic and the conventional PID controllers while varying the input voltage and implementing the MSCC protocol. The proposed charging controller perfectly ensured that the minimum battery terminal voltage ripple and charging current ripple reached 1 mV and 1 mA, respectively, with a very high-speed response of 1 ms in reaching the predetermined charging current stages. The present simulated and experimental results are in good agreement with the previous related work in the literature survey.