Feedforward Control Algorithm Research Articles

On inversion-based approaches for feedforward and ILC

System inversion is at the basis of many feedforward and learning control algorithms. The aim of this paper is to analyze several of these approaches in view of their subsequent use, showing inappropriate use that is previously overlooked. This leads to different insights and new approaches for both feedforward and learning. The methods are compared in various aspects, including finite vs. infinite preview, exact vs. approximate, and quality of inversion in various norms which directly relates to their use. In addition, extensions to (non-square) multivariable and time-varying systems are presented. The results are validated on a nonminimum-phase benchmark system.

A novel feed‐forward segmented digital automatic gain control algorithm for long‐term evolution digital radio‐over‐fibre systems

© 2017, The Institution of Engineering and Technology. This study describes a novel feed-forward segmented digital automatic gain control (FSD-AGC) algorithm for long term evolution (LTE) digital radio-over-fibre (DRoF) systems. It experimentally demonstrates the algorithm using a DRoF platform based on real-time field programmable gate arrays. The FSD-AGC algorithm is able to update power status accurately with a short delay (about 100 samples) and adjust the gain precisely by only one step. It can make full use of analogue-to-digital converter quantisation bits width without oversaturation and maintain a stable peak-to-average ratio to reduce the overall cost and energy consumption of digital links and to minimise the deterioration of error vector magnitude (EVM) in LTE systems. Compared with conventional solutions, the proposed algorithm shows significant improvement in the dynamic range (a 20 dB improvement is demonstrated experimentally) of 4G-LTE radio frequency signals in a DRoF system. This is a major step forward for the future commercialisation of DRoF systems for 4G and 5G network deployments. Meanwhile, FSD-AGC can be applied in both LTE time division duplexing and frequency division duplexing signals. Experimental results have shown a mean EVM of the DRoF system below 4.5%, which is well below the 8% required by 3rd generation partnership project.

Inverting Nonminimum-Phase Systems from the Perspectives of Feedforward and ILC

System inversion is at the basis of many feedforward and learning control algorithms. This paper aims to analyze several of these approaches in view of their subsequent use, showing inappropriate use that is previously overlooked. This leads to new insights and approaches for both feedforward and learning. The methods are compared in various aspects, including finite vs. infinite preview, exact vs. approximate, and quality of inversion in various norms which directly relates to their use. The results are validated on a nonminimum-phase benchmark system.

Time-domain versus frequency-domain effort weighting in active noise control design

Active noise control is often designed by minimizing an error index which includes a penalty on the actuator inputs. In this way, an effort-weighting parameter allows one to monitor the maximum voltages applied to the control sources. Two effort-weighting techniques have been widely implemented in previous studies, Tikhonov regularization for frequency-domain simulations and the leaky filtered-reference least mean square algorithm for real-time feedforward control. This paper introduces the formal connection between the weighting parameters of these two techniques. Simulations of a simple set-up show that, with these two connected parameters, the control performances are indeed very close.

Mold Oscillation Feedforward Control Algorithm for Sinusoidal Oscillation of Various Asymmetries

Mold Oscillation Feedforward Control Algorithm for Sinusoidal Oscillation of Various Asymmetries

Analysis and experiments of adaptive feedforward and combined vibration control system with variable step size and reference filter

An adaptive feedforward and combined vibration control system with variable step size and reference filter is analyzed and implemented to suppress the vibration of a kind of thin-walled structure. The theoretical analysis and experiments of adaptive feedforward and combined vibration control algorithms are presented. The experimental setup of a flexible plate bonded PZT (Lead Zirconate Titanate) sensors/actuators is constructed. The control algorithm combines an adaptive feedforward controller and an adaptive feedback controller together, which has shown a superior control performance in the experiments. Considering the disadvantages of fixed step size used in the conventional filtered-U least mean square (FULMS) control algorithm, a variable step-size (VSS) method is applied. Also, a reference filter is used to extract the desired signal from the positive feedback and measurement noise. A proportional derivative (PD) feedback control algorithm is also applied as comparison. The experimental results demonstrate that the adaptive combined control algorithm has better control performance than the adaptive feedforward and PD feedback control algorithms.

Comparison of adaptive algorithms for the control of tonal disturbances in mechanical systems

This paper presents a study on the performance of adaptive control algorithms designed to reduce the vibration of mechanical systems excited by a harmonic disturbance. The mechanical system consists of a mass suspended on a spring and a damper. The system is equipped with a force actuator in parallel with the suspension. The control signal driving the actuator is generated by adjusting the amplitude and phase of a sinusoidal reference signal at the same frequency as the excitation. An adaptive feedforward control algorithm is used to adapt the amplitude and phase of the control signal, to minimise the mean square velocity of the mass. Two adaptation strategies are considered in which the control signal is either updated after each period of the oscillation or at every time sample. The first strategy is traditionally used in vibration control in helicopters for example; the second strategy is normally referred to as the filtered-x least mean square algorithm and is often used to control engine noise in cars. The two adaptation strategies are compared through a parametric study, which investigates the influence of the properties of both the mechanical system and the control system on the convergence speed of the two algorithms.

Back-Propagation Algorithm-Based Controller for Autonomous Wind–DG Microgrid

This paper deals with a wind–diesel generator hybrid configuration of the microgrid using a voltage source converter (VSC) as a voltage and frequency (VF) controller. The wind power generated by permanent magnet brushless dc generator, and the maximum power is captured by a maximum power point technique using a boost converter with an incremental conductance approach. This power is supplied to the consumer loads and surplus power is stored into battery system (BS). BS is attached at dc link of VSC, which provides load leveling during less or no wind conditions. With such combination of energy resources, a reduced rating, diesel engine-driven squirrel cage induction generator feeds loads, and VSC at point of common coupling (PCC) supports the system when the wind generation is unable to meet out the load demand. Back-propagation feed-forward control algorithm is used for VF control of VSC. This controller provides harmonics elimination, load leveling, and reactive power compensation, and also regulates the voltage at PCC. This microgrid is modeled in MATLAB Sim power tools, and simulation results are produced to verify the appropriate working of both the converters and the overall system. A prototype of the microgrid is also developed to validate the design and model through test results. A comparative study of simulated and experimental work is given in detail.

Adaptive positioning control of an ultrasonic linear motor system

A 3PRR parallel precision positioning system, driven by three ultrasonic linear motors, was designed for use as the object stage of a scanning electron microscope (SEM). To improve the tracking accuracy of the parallel platform, the positioning control algorithms for the drive joints needed to be studied. The dead-zone phenomenon caused by static friction reduces the trajectory tracking accuracy significantly. Linear control algorithms such as PID (Proportion Integration Differentiation) are unable to compensate effectively for the dead-zone nonlinearity. To address this problem, two types of feedforward compensation control algorithms have been investigated. One is constant feedforward with the integral separation PID; the other is adaptive feedback and feedforward based on the model reference adaptive control (MRAC). Simulations and experiments were conducted using these two control algorithms. The results demonstrated that the constant feedforward with integral separation PID algorithm can compensate for the time-invariant system after identifying the dead-zone depth, while the adaptive feedback and feedforward algorithm is more suitable for the time-varying system. The experimental results show good agreement with the simulation results for these two control algorithms. For the dead-zone nonlinearity caused by the static friction, the adaptive feedback and feedforward algorithm can effectively improve the trajectory tracking accuracy. A positioning system has been built for the driven joints of the 3-PRR in SEM.The dead-zone of the system has been identified and applied to compensation.The proposed control algorithm improves the tracking precision significantly.

Improved feedforward inverse control with adaptive refinement for acceleration tracking of electro-hydraulic shake table

Shake tables are essential facilities in the laboratory for evaluating structural performance subject to vibration excitation. In this paper, to improve the time waveform replication accuracy of electro-hydraulic shake tables (EHSTs), an improved feedforward inverse control algorithm with adaptive refinement is proposed. The EHST system transfer function and stable inverse model is firstly estimated and designed by multi-step recursive extended least squares algorithm and zero magnitude error tracking controller technology, respectively. To reduce the side effect of model identification errors between the estimated and actual system, a system model corrector is further identified based on the previous estimated EHST model and an inverse corrector is obtained so as to constitute the improved feedforward inverse controller (iFIC) by cascading the inverse corrector to the previous designed inverse model. Then, an adaptive refinement controller using a least mean square algorithm is applied to the iFIC controlled system to cope with system uncertainties and nonlinear effects. Therefore, the proposed algorithm combines the merits of feedforward inverse control and adaptive control. Finally, with the help of xPC rapid prototyping technology, experiments are conducted on a real uniaxial EHST system and the experimental results demonstrate that the proposed algorithm exhibits a better tracking accuracy than the conventional controllers for shake tables.

DC‐link current estimation for load‐side converter of brushless doubly‐fed generator in the current feed‐forward control

In the stand-alone brushless doubly-fed generator (BDFG) control system, there are two back-to-back converters composed of the machine-side converter (MSC) and the load-side converter (LSC). The LSC is a three-phase AC-DC pulse width modulation converter which usually adopts the double-loop control strategy with outer voltage loop and inner current loop. The introduction of the DC-link current feed-forward for improving the dynamic performance of the LSC can reduce the effect of the load disturbance to the system. However, the DC-link current has a complex harmonic spectrum with the MSC connected to the DC side, so its average value is not convenient to measure. This study presents a DC-link current estimation algorithm for the current feed-forward control in the LSC of the stand-alone BDFG. With the help of the equivalent circuit of the BDFG and the power balance, the rapid calculation of the average DC-link current can be estimated as well. With this estimation of the current instead of the actual DC-link current, the DC-link voltage and AC side current of the LSC can achieve stability quickly. The effectiveness and correction is validated by the simulation and experiments.

Study on transient beam loading compensation for China ADS proton linac injector II**Supported by National Natural Science Foundation of China (91426303, 11525523)

Significant transient beam loading effects were observed during beam commissioning tests of prototype II of the injector for the accelerator driven sub-critical (ADS) system, which took place at the Institute of Modern Physics, Chinese Academy of Sciences, between October and December 2014. During these tests experiments were performed with continuous wave (CW) operation of the cavities with pulsed beam current, and the system was configured to make use of a prototype digital low level radio frequency (LLRF) controller. The system was originally operated in pulsed mode with a simple proportional plus integral and deviation (PID) feedback control algorithm, which was not able to maintain the desired gradient regulation during pulsed 10 mA beam operations. A unique simple transient beam loading compensation method which made use of a combination of proportional and integral (PI) feedback and feedforward control algorithm was implemented in order to significantly reduce the beam induced transient effect in the cavity gradients. The superconducting cavity field variation was reduced to less than 1.7% after turning on this control algorithm. The design and experimental results of this system are presented in this paper.

A Study on the Feedforward Control Algorithm for Dynamic Positioning System Using Ship Motion Prediction

In the present study we verified performance of feed-forward control algorithm using short term prediction of ship motion information by taking advantage of developed numerical simulation model of FPSO motion. Up until now, various studies have been conducted about thrust control and allocation for dynamic positioning systems maintaining positions of ships or marine structures in diverse sea environmental conditions. In the existing studies, however, the dynamic positioning systems consist of only feedback control gains using a motion of vessel derived from environmental loads such as current, wind and wave. This study addresses dynamic positioning systems which have feedforward control gain derived from forecasted value of a motion of vessel occurred by current, wind and wave force. In this study, the future motion of vessel is forecasted via Brown`s Exponential Smoothing after calculating the vessel motion via a selected mathematical model, and the control force for maintaining the position and heading angle of a vessel is decided by the feedback controller and the feedforward controller using PID theory and forecasted vessel motion respectively. For the allocation of thrusts, the Lagrange Multiplier Method is exploited. By constructing a simulation code for a dynamic positioning system of FPSO, the performance of feedforward control system which has feedback controller and feedforward controller was assessed. According to the result of this study, in case of using feedforward control system, it shows smaller maximum thrust power than using conventional feedback control system.

Fast estimation of combustion metrics in DI diesel engines for control-oriented applications

The present work has been focused on the development of low-throughput semi-empirical models to predict the peak firing pressure, indicated mean effective pressure and brake mean effective pressure in direct injection diesel engines. The models have been calibrated and assessed on a 1.6L GM Euro 6 diesel engine, on the basis of the results of experimental tests conducted at a dynamic test bench at GMPT-E (General Motors Powertrain-Europe). Model validation was carried out over “New European Driving Cycle” and “Worldwide Harmonized Light vehicle Test Procedure” missions. The performance of the semi-empirical models has been compared, in terms of prediction accuracy, robustness, inversion efficiency and computational expense, with that of a previously developed real-time physical combustion model. The thus developed semi-empirical models are characterized by a very low computational effort, and are therefore suitable for the development of innovative feed-forward control algorithms in the Engine Control Unit.

Combination of map-based and adaptive feedforward control algorithms for active engine mounts

Active engine mounts significantly contribute to ensure the comfort in vehicles with emission-reducing engine technologies, e.g., cylinder-on-demand (COD), downsizing or turbochargers. To control active engine mounts, either adaptive or non-adaptive feedforward control is commonly employed. Since both approaches have previously been treated separately, this study proposes methods to connect them in terms of multiple-input-multiple-output Newton/FxLMS adaptive filters with self-trained, grid-based look-up tables. The look-up tables are incorporated as parameter-maps or parallel-maps, respectively. By combining the two feedforward control strategies, their inherent advantages, i.e., the adaptivity of adaptive filtering and the direct impact as well as the tracking behavior of map-based feedforward control, are utilized. The proposed control structures are illustrated by simulation and experimentally demonstrated in a vehicle with a V8-COD engine. While both methods significantly reduce the convergence time of the adaptive filter, the parallel implementation additionally improves the tracking behavior during fast engine run-ups.

Coupling Model and Controller Design for Four-layer Register System

A nonlinear and coupling model is established according to the multi-layer register system working principle in gravure printed electronic equipments, and then, the linear model of the register system is constructed based on one order Taylor formula. In order to improve the speed and accuracy of register control, a feedforward PID control algorithm is presented according to the linear model in multi-layer register system. This algorithm is unique in that it uses the PID to adjust the inputs of the register system, and feedforward compensation for the known disturbance, making the register control accuracy has been greatly improved. The simulation shows that the proposed register controller is able to realize a high precision control for the register system and endowed with better control performance than PID controller in register control.

Hysteresis Modeling of Piezoelectric Actuators and Feed-Forward Compensation Algorithm Research

Piezoelectric ceramic actuators have wide application prospects in precision positioning and measurement technology, but its hysteresis characteristics cause a lot of restrictions to their applications. To solve this problem, this paper summary the hysteresis characteristic of piezoelectric ceramics, and collect the data through the experiment, analyze and research the piezoelectric ceramic hysteresis curves at different excitation signal effect, and the fitting curve is got based on the experimental data in order to establish hysteresis model of piezoelectric ceramic actuators by using of the curve fitting toolbox of MATLAB. Then the inverse model is obtained and the inverse compensation feed-forward control algorithm is proposed. The experiment results show that the piezoelectric ceramic actuators meet the basic linear relationship between the input voltage signal and the output displacement, and eliminate hysteresis nonlinear of the piezoelectric actuators.

An Extended Static and Dynamic Feedback–Feedforward Control Algorithm for Insulin Delivery in the Control of Blood Glucose Level

The potential for successful automatic control of blood glucose concentration (BGC) has entered a new era because of recent technological advancements in insulin pumps and blood glucose sensors. However, a critical advancement necessary for full automation and long-term use is a control algorithm that can effectively maintain tight control of BGC under extreme variation of important disturbances such as activity, stress, and food consumption. Because feedforward control (FFC) models disturbances directly, it has the potential to eliminate the effects of disturbances completely. A Wiener-type feedforward control law is limited to the inclusion of only input (i.e., modeled disturbances and the manipulated variable) dynamics. Using a semicoupled modeling network that includes pseudo-blood insulin concentration, this work presents a more phenomenological FFC law that includes input dynamics, blood insulin and blood glucose dynamics, and blood glucose levels. Modeling results on 15 adults with type 1 diabetes ...

Model-based feedforward control of the VGT in a diesel engine based on empirical models of compressor and turbine efficiencies

Current diesel engines commonly employ a variable geometry turbocharger (VGT) to improve drivability and fuel efficiency, and this paper addresses model-based feedforward control of the VGT based on empirical models of compressor and turbine efficiencies to improve transient response. Though compressor and turbine efficiencies affect compressor air flow and turbine power, respectively, it is challenging to understand exactly the compressor and turbine efficiencies in real engines. In order to cope with this problem, we propose empirical models of compressor and turbine efficiencies. The input states are proposed based on maps of the compressor and turbine efficiencies, which are provided by turbocharger manufacturers. Parameters of the efficiency models are identified with 225 data of steady-state engine experiments to reflect engine operating conditions. The proposed compressor and turbine efficiency models were applied to the model-based VGT feedforward control algorithm to verify the effectiveness of the efficiency models. The proposed modelbased VGT feedforward control algorithm based on the compressor and turbine efficiency models was experimentally verified with 2.2 L common-rail-direct-injection diesel engines.

Experimental and analytical study of secondary path variations in active engine mounts

Active engine mounts (AEMs) provide an effective solution to further improve the acoustic and vibrational comfort of passenger cars. Typically, adaptive feedforward control algorithms, e.g., the filtered-x-least-mean-squares (FxLMS) algorithm, are applied to cancel disturbing engine vibrations. These algorithms require an accurate estimate of the AEM active dynamic characteristics, also known as the secondary path, in order to guarantee control performance and stability. This paper focuses on the experimental and theoretical study of secondary path variations in AEMs. The impact of three major influences, namely nonlinearity, change of preload and component temperature, on the AEM active dynamic characteristics is experimentally analyzed. The obtained test results are theoretically investigated with a linear AEM model which incorporates an appropriate description for elastomeric components. A special experimental set-up extends the model validation of the active dynamic characteristics to higher frequencies up to 400Hz. The theoretical and experimental results show that significant secondary path variations are merely observed in the frequency range of the AEM actuator׳s resonance frequency. These variations mainly result from the change of the component temperature. As the stability of the algorithm is primarily affected by the actuator׳s resonance frequency, the findings of this paper facilitate the design of AEMs with simpler adaptive feedforward algorithms. From a practical point of view it may further be concluded that algorithmic countermeasures against instability are only necessary in the frequency range of the AEM actuator׳s resonance frequency.