Search

30.0 The development of machine learning technologies are broadly changing how humans interact with their environments across all sectors. In industrial settings, this is referred to as the fourth industrial revolution, Industry 4.0, and encompasses several technologies that are pushing the boundaries of industrial automation. In this study, a general industrial process optimization (GIPO) methodology is formulated in the context of Industry 4.0 and tested on an industrial Injection Molding Machine (IMM). GIPO aims to encourage the practical inclusion of industrial artificial intelligence at all levels of the manufacturing process while enabling industrial equipment to adapt to a changing processing environment. Special attention is given to the generality of the methodology so that it can be extended to other applications. In the example case study presented here, GIPO combines nearest neighbors classification and nearest neighbors optimization methods to effectively optimize an Injection molding process. Practical implementation conducted on the IMM demonstrates a novel methodology to leverage data mining and machine learning methods in a real-world setting to improve the overall performance regarding production time, energy cost, and production quality.

Ball and beam balancing systems are commonly found in control systems laboratories because they are mechanically simple while presenting the challenge of controlling a nonlinear system. The main ideas of this paper are to model the ball and beam system and to control it using adaptive control. The model was formed by separating the system into two parts, and this was then controlled using a cascading control scheme. For the inner loop control, the motor was controlled using simple adaptive control (SAC) and compared to proportional derivative (PD) control. In the outer loop with the ball and beam system, ball position was controlled using PD control and was compared to linear quadratic regulator (LQR) control. Experimental results show that though PD control performed better when only controlling the motor, SAC performed better in the cascading control scheme when paired with either PD or LQR as outer loop control. When comparing PD and LQR, the results show that PD performs better.

In this paper, a neural network-based adaptive impedance control (NNAIC) scheme is proposed for controlling electrically-driven robotic systems. The neural network (NN) is used to compensate the robotic model uncertainties and the motor dynamic uncertainties, and a robust compensator term is derived to compensate the disturbances and approximation errors of the NN. In this way, the performances of the joint positions and force tracking can be then improved. Based on the Lyapunov stability theorem, it is proved that the control system is stable and all the signals in closed-loop system are bounded. Simulation tests on a 2-link electrically-driven robotic manipulator are presented to show the effectiveness of the proposed intelligent impedance control method.

The sensors being used in industrial settings are becoming increasingly complex and capable of acquiring significantly more data with higher functionality. One of these sensors is the thermal camera, which has the potential of capturing the spatial temperature profile of any surface. Through the extension of the conventional model-based control theory, a new model-based control scheme was formulated verify the novel concept of utilizing the spatial data from the thermal camera. To analyze this new model based control algorithm, a three zone heater barrel apparatus was designed and constructed for closed-loop testing to be performed. The experiments demonstrated the benefit of thermal camera feedback through the alterations in the control scheme. The availability of thermal profile of the system in comparison to single point temperature measurements resulted in the enhanced controller achieving better closed-loop transient responses, reduction in settling times. In addition, the use of spatial thermal images provides the possibility of controlling profiled temperature setpoints, showing the benefit and potential of using more advanced sensors in industry.

The industrial requirements for controllers able to perform tasks in the presence of plant nonlinearities are growing. In addition, an increase in industrial computation power is allowing the implementation of more complex control algorithms in the fast processing industry. In this investigation three different nonlinear model predictive control algorithms are tested and evaluated in simulation and experimentally. The methodologies are adaptive nonlinear model predictive control (nMPC), PID based nMPC (PIDnMPC), and a novel simplified nMPC (SnMPC). These are tested in simulation with an inverted pendulum, a Van der Pol oscillator, and a planar 2-link vertical robotic arm. The controllers are tested experimentally using a fabricated planar 2-link vertical robotic arm apparatus. A comparison of the different algorithms is made with special attention to trajectory tracking, computational complexity and transient response dynamics. HighlightsThree non-linear MPC Algorithms are developed to control fast systems, nMPC, PIDnMPC and SnMPC.The algorithms are tested in simulation using three non-linear models.The controller features are discussed and future directions of study are suggested.

Docking an unmanned underwater vehicle (UUV) with a submerged submarine in littoral waters in high sea states requires more dexterity than either the submarine or streamlined UUV possess. The proposed solution uses an automated active dock to correct for transverse relative motion between the vehicles. Acoustic, electromagnetic, and optical sensors provide position sensing redundancy in unpredictable conditions. The concept is being evaluated by building and testing individual components to characterize their performance, errors, and limitations, and then simulating the system to establish its viability at low cost.

An adaptive control architecture for uncertain, under-actuated, non-linear systems is proposed. The proposed methodology is tested in simulation in order to control a planar double inverted pendulum apparatus. The approach can be extended and applied to similar systems such mobile robots and exoskeletal assisted devices. Closed loop results show good performance of the adaptive control methodology.

This paper develops a unique approach that can be imposed on conventional predictive control schemes to provide tighter control when tracking complex setpoint trajectories. The method introduces a correction parameter η , independent of plant gain, evaluated online at each control timestep to drive the plant output to its reference more accurately than the original schemes. The strategy was applied to various systems that possess deadtime, higher order and multivariable characteristics with improved closed loop control. This is evidenced by significant reductions of mean square errors when using η . The strategy was applied on a nonlinear system and a practical setup, with reduced tracking error.

A model predictive controller (MPC) in the form of dynamic matrix control (DMC) is implemented for attenuating in-plane vibrations of a two flexible link planar manipulator. The rotation of the joints and inertia effect of both the joints and links induce vibration. Piezoelectric actuators, mounted in a bimorph configuration, provide the control actions to reduce vibrations. Implementation of this control scheme is shown to provide appreciable attenuation of vibration over the uncontrolled case, increasing the damping ratio for the first and second link by a factor of 5.99 and 3.40, respectively. DMC control is further shown to reduce the mean amplitude of dominant vibrations from the uncontrolled case by 90.0% and 87.4%, respectively, for the first and second links. Furthermore, for the two link setup, this control is shown to outperform the more conventional ProportionalIntegral-Derivative (PID) control and is sufficiently robust to handle an unknown payload.

This paper presents a unique approach for active vibration control of a one-link flexible manipulator. The method combines a finite element model of the manipulator and an advanced model predictive controller to suppress vibration at its tip. This hybrid methodology improves significantly over the standard application of a predictive controller for vibration control. The finite element model used in place of standard modelling in the control algorithm provides a more accurate prediction of dynamic behavior, resulting in enhanced control. Closed loop control experiments were performed using the flexible manipulator, instrumented with strain gauges and piezoelectric actuators. In all instances, experimental and simulation results demonstrate that the finite element based predictive controller provides improved active vibration suppression in comparison with using a standard predictive control strategy.

In this study, a Wiener-type neural network (WNN) is derived for identification and control of single-input and single-output (SISO) nonlinear systems. The nonlinear system is identified by the WNN, which consists of a linear dynamic block in cascade with a nonlinear static gain. The Lipschitz criteria for model order determination and back propagation for the adjustment of weights in the network are presented. Using the parameters of the Wiener model, the analytical expressions used in the controller, generalized predictive control (GPC) is modified every time step, to handle the nonlinear dynamics of the controlled variable. Finally, the proposed WNN-based GPC algorithm is tested in simulation on several nonlinear plants with different degrees of nonlinearity. Simulation results show that WNN identification approach has better accuracy, in comparison to other neural network identifiers. The WNN-based GPC has better control performance, in comparison to standard GPC.

This paper presents a unique method for improving the performance of the generalized predictive control (GPC) algorithm for controlling nonlinear systems which can be extended to other forms of predictive controllers. This method is termed adaptive generalized predictive control (AGPC) which uses a multidimensional workspace of the nonlinear plant to recalculate the controller parameters every sampling instant. This results in a more accurate process prediction and improved closed-loop performance over the original GPC algorithm. The AGPC controller was tested in simulation, and its control performance was compared to GPC on several nonlinear plants with different degrees of nonlinearity. Practical testing and comparisons were performed on a steel cylinder temperature control system. Simulation and experimental results show that the adaptive generalized predictive controller provided improved closed-loop performance over GPC. The formulation of the multidimensional workspace can be readily applied to othe...

This paper investigates the viscosity behavior of melt polymer during the plastication stage in an injection molding machine (IMM). The objective of this paper is to develop theoretical viscosity-based models for evaluating and predicting the bulk viscosity of the melt polymer near the tip of the screw during the plastication stage. The uniqueness of these models is that they include the effects of material properties and machine processing parameters, such as screw rotational speed, hydraulic back pressure, barrel pressure drop, melt pressure and melt temperature. Several open loop tests were conducted on an industrial IMM in order to quantify process parameter effects on the plastication cycle. The derived model was used to predict bulk viscosity using the open loop tests data. The viscosity-based models in this work will be used for developing future optimal controllers for controlling the polymer viscosity during the plastication stage of the injection molding cycle.

This paper presents a unique approach for designing a nonlinear regression model-based predictive controller (NRPC) for single-input-single-output (SISO) and multi-input-multi-output (MIMO) processes that are common in industrial applications. The innovation of this strategy is that the controller structure allows nonlinear open-loop modeling to be conducted while closed-loop control is executed every sampling instant. Consequently, the system matrix is regenerated every sampling instant using a continuous function providing a more accurate prediction of the plant. Computer simulations are carried out on nonlinear plants, demonstrating that the new approach is easily implemented and provides tight control. Also, the proposed algorithm is implemented on two real time SISO applications; a DC motor, a plastic injection molding machine and a nonlinear MIMO thermal system comprising three temperature zones to be controlled with interacting effects. The experimental closed-loop responses of the proposed algorithm were compared to a multi-model dynamic matrix controller (MPC) with improved results for various set point trajectories. Good disturbance rejection was attained, resulting in improved tracking of multi-set point profiles in comparison to multi-model MPC.

This article presents a new application of model-based predictive controller (MPC) for vibration suppression of a flexible one-link manipulator using piezoceramic actuators. Simulation and experimental studies were conducted to investigate the applicability of the MPC strategy to control vibration of the flexible structure having multiple inputs and multiple outputs (MIMO). The performance of the proposed technique was assessed in terms of level of vibration reduction. The results demonstrated that the proposed predictive control strategy is well suited for multivariable control of vibration suppression on flexible structures.

Part cooling parameters such as coolant flow rate and coolant temperature has been shown to have an important influence on the part quality in plastic injection molding. Molders commonly manipulate the coolant flow rate in an unclear manner to adjust part quality and reduce the overall cycle time. Manipulating such parameters effectively can indirectly control part quality when deviations in the injection processing variables occur if it is done in a controlled manner. In this investigation, coolant flow rate is selected to be the manipulated variable that affects an average temperature of the part surface as it cools within the mold cavity. The average part temperature is the controlled variable and its control represents the indirect control of part quality. Experiments were conducted to quantify the effects of coolant flow rate on the part surface temperature demonstrating that it is highly nonlinear and its control would require an advanced control scheme. An adaptive online modeling algorithm that uses information from temperature and flow rate sensors was developed to continuously retune a predictive controller for controlling the part surface temperature. Good control performance was obtained from the adaptive strategy for tracking setpoint changes in the part temperature and when the cooling cycle was reduced from 16 to 14 s.

Many model predictive control (MPC) algorithms have been proposed in the literature depending on the conditionality of the system matrix and the choice of its cost-function. This paper presents the newer MPC schemes such as extended predictive control (EPC) and shifted MPC as well as other well known forms. The control performance of these controllers are compared using two systems that are slow and fast reacting. The closed-loop responses are compared and the differences and similarities are explained on the basis of the structure of the control schemes. Disturbance rejection and the tracking of various setpoint trajectories are performed with good closed-loop results from all the controllers. It was found that the controllers that were specifically designed to reduce the system matrix ill-conditionality such as EPC and generalized predictive control provided better control performance when compared to other MPC methods.

Injection velocity is one of the key parameters in the injection molding process that has significant effect on part quality, influencing common problems, such as flashing and short shots. Different products require specific molding conditions, including mold and melt temperatures, velocity profiles, and polymers, making the injection velocity dynamics vary significantly and difficult for high precision velocity control. Two predictive controllers were developed to control the screw injection velocity. The first approach uses a position sensor as feedback and a simplified predictive controller to track an injection velocity setpoint profile. The controller was developed and implemented utilizing single-step change and multistep change open loop tests. The second strategy uses a multimodel dynamic matrix predictive controller to overcome the nonlinear characteristics of the injection velocity dynamics as the mold is being filled. Velocity feedback is provided by high speed processing of the position analog signal. This approach utilizes several open loop injection velocity profiles to generate corresponding dynamic matrices for the controller. As a result, this controller is modified or retuned automatically when setpoint changes in injection velocity occur, as well as when using different polymers and molds. The close loop results for both simulations and real time control have demonstrated that the two predictive controllers provided good setpoint tracking performance for wide ranging position and velocity profiles. POLYM. ENG. SCI., 47:390–399, 2007. © 2007 Society of Plastics Engineers.

A quality-controlled predictive control method, suitable for control of fast, remote systems subject to significant communication delays, is developed. Each move is quality controlled in that it independently satisfies a risk-based control performance criterion. The method is found to be capable of mitigating the ill effects of highly nonstationary delay distributions while providing good control performance for milder nonstationarity. It is demonstrated on simplified predictive control (SPC) of a single-input, single-output process. SPC is preferred here due to its simplicity and well-conditioned dependence of both the sampling time and its single parameter.

A unique method of coupling computational fluid dynamics (CFD) to model predictive control (MPC) for controlling melt temperature in plastic injection molding is presented. The methodology is based on using CFD to generate, via open-loop testing, a temperature and input dependent system model for multi-variable control of a three-heater barrel on an injection molding machine. Results clearly show the benefit of temperature and input dependent system models for MPC control, and that CFD can be used to dramatically reduce the time associated with open-loop testing through physical experiments.