Conventional Feedback Control Research Articles

An artificial neural network based approach to air supply control in large indoor spaces considering occupancy dynamics

Occupancy dynamics can significantly influence indoor thermal environments, especially in large indoor spaces. It is difficult for conventional feedback control systems to respond promptly to occupancy dynamics because of the substantial thermal inertia of large spaces, which leads to unfavorable thermal conditions in environments regulated by such systems. To address this challenge, this study proposes an air supply control approach based on artificial neural networks (ANNs). In the proposed approach, a large space is divided into multiple zones and an ANN model is used to characterize the relationship between occupancy dynamics and the supply air flow rates of each zone, thereby expediting the response of the air-conditioning system to occupancy dynamics. First, a multi-zone thermal environment model was developed to accurately emulate the thermal behavior of each zone. Next, employing the developed model of the environment, the optimal air flow rates required for each zone to maintain the desired thermal environment were estimated for various boundary conditions, which were used as pretraining data for four candidate ANNs. Finally, the best-performing ANN candidate, Long Short-Term Memory (LSTM), was adopted in a case study building via a comparison against several conventional air supply control methods. The results from the case studies demonstrate that the proposed approach can effectively expedite the system response to occupancy dynamics, thereby minimizing the occurrence of overcooling and overheating, and lowering the occupancy-weighted thermal discomfort level by 73.1 %. The proposed approach holds promise for real-time applications based on digital twin architecture.

Deep reinforcement learning for PID parameter tuning in greenhouse HVAC system energy Optimization: A TRNSYS-Python cosimulation approach

The control of indoor temperature in greenhouses is crucial as it directly impacts the crop's thermal comfort and the performance of heating, ventilation, and air-conditioning (HVAC) systems. Conventional feedback controllers, like on/off, can sometimes make HVAC system work at full capacity when only half that capacity is needed. In contrast, the proportional-integral-derivative (PID) controller, provides precise control based on its P, I, and D parameters. However, it lacks a formal design procedure for optimizing a specified objective function. Previous studies have utilized conventional PID tuning approaches to track room setpoint temperature for residential buildings, data centers, and office buildings, with limited research in greenhouse applications. To address this gap, this study proposes a flexible PID controller that employs a deep reinforcement learning (DRL) algorithm to optimize its parameters, by tracking the setpoints and energy consumption of a greenhouse planted with tomatoes. This approach is different from the typical method of using the trained RL agent directly in HVAC controls. Through a self-made TRNSYS-Python cosimulation framework, the DRL agent interacts directly and in real time with the greenhouse and its plants. Consequently, optimized PID parameters were established and tested in the simulated environment. The resulting performance, in terms of both energy consumption and its ability to maintain the crop’s comfort temperature, was compared with the simulated on/off and manually tuned PID controllers. Compared to the on/off baseline control, the proposed PID optimized parameters reduce energy use by 8.81% to 12.99% and the manually tuned PID parameters with the Ziegler-Nichols tuning method reduce energy use by 7.17 %. Additionally, the proposed method had a deviation of 2.07% to 3.13%, while the manually tuned PID controller and the on/off controller had deviations of 7.27% and 3.27%, respectively, from the minimum comfortable temperature. This study serves as a framework for improving the energy efficiency of greenhouse HVAC system operations.

Model predictive control for thermal comfort and energy optimization of an air handling unit system in airport terminals using occupant feedback

The dynamic fluctuations in occupant flow within airport terminals contribute to delays in the system's response under traditional feedback control. This imbalance between supply and demand can result in discomfort and unnecessary energy consumption. While Occupant-Based Model Predictive Control (OBMPC) has gained research interest as it leverages occupant forecasts to enhance control performance, there is limited knowledge regarding its application in the air-conditioning systems of airport terminals, mainly due to the intricate dynamics of occupant flow in terminal buildings. In this study, we proposed a model predictive control strategy based on dynamic occupant flow with the goal of fast responding to occupant flow and reducing energy consumption. We integrated mathematical formulas and the Anylogic simulation software to predict occupant variations in the baggage claim hall. The model predictive control strategy of the air conditioning system was proposed with the predicted occupant flow. The proposed strategy was evaluated throughout the whole cooling season in the validated simulation model of the air handling unit system of the baggage claim hall at an airport terminal. The results demonstrated that the proposed strategy could effectively respond to variations in occupancy and achieve 10% energy savings throughout the entire cooling season, compared to conventional feedback control. Our work presents a viable solution to address system regulation lag and reduce energy consumption without jeopardizing thermal comfort.

Novel virtual sensors development based on machine learning combined with convolutional neural-network image processing-translation for feedback control systems of internal combustion engines

Physical sensors are commonly used to record performance data of internal combustion engines (ICEs) for online feedback control and calibration, but they are prone to diagnostic and increased development costs. Lookup tables are commonly used in conventional calibration and feedback control; however, the table parameters increase with the advancement of ICE technologies under transient operations. Consequently, the calibration and control systems are time-consuming. This work proposes novel virtual sensors to address these issues by predicting the combustion, performance, and emission of ICEs using neural networks and image processing/translation. The novel sensors are targeted for onboard feedback control systems under transient driving. Firstly, a virtual diesel engine (VDE) was developed and calibrated against experimental data taken from a production 2.2 L turbocharged diesel engine. The VDE was calibrated under WLTC, JC08, and NEDC transient operations and was used to generate teaching data. Next, the virtual sensors are developed using five machine learning (ML) regressors. The result shows that the coefficient of determination R2 from all ML regressors exceeded 0.94, and the XG-Boost outperforms other ML techniques with R2 > 0.977. XG-Boost parameter estimations were 8 times faster than that on a desktop simulation. Then, an image classification model using a deep convolutional neural network (D-CNN) is constructed, and the dependency of performance parameters and exhaust emissions with the rate of heat release (R.H.R) and in-cylinder pressure profile is confirmed. The performance parameters and emissions dependency was compared individually with R.H.R. and the in-cylinder pressure profile. As a result, a strong correlation between the performance and R.H.R. was observed. Finally, a generative adversarial network (GAN) model was constructed to translate the in-cylinder pressure profile to R.H.R. profile. A novel method to develop virtual sensors for advanced feedback control of any type of ICEs is proposed for the first time.

Reconfigurable supply-based feedback control for enhanced energy flexibility of air-conditioning systems facilitating grid-interactive buildings

Air-conditioning systems have great potential to provide energy flexibility services to the power grids of high-renewable penetration, due to their high power consumption and inherent energy flexibilities. Direct load control by switching off some operating chillers is the simplest and effective means for air-conditioning systems in buildings to respond to urgent power reduction requests of power grids. However, the implementation of this approach in today's buildings, which widely adopt demand-based feedback controls, would result in serious problems including disordered cooling distribution and likely extra energy consumption. This study, therefore, proposes a reconfigurable control strategy to address these problems. This strategy consists of supply-based feedback control, incorporated with the conventional demand-based feedback control, a control loop reconfiguration scheme and a setpoint reset scheme, facilitating effective control under limited cooling supply and smooth transition between supply-based and demand-based feedback control modes. The proposed control strategy is deployed in a commonly-used digital controller to conduct hardware-in-the-loop control tests on an air-conditioning system involving six AHUs. Test results show that the reconfigurable control achieves commendable control performance. Proper chilled water distribution enables even thermal comfort control among the building zones during demand response and rebound periods. Temperature deviation of the building zones is controlled below 0.2 K most of the time. 11.6 % and 27 % of power demand reductions are achieved during demand response and rebound periods respectively, using the proposed reconfigurable control compared with that using conventional controls.

Multi-agent based distributed cooperative control of air-conditioning systems for building fast demand response

The utilization of air-conditioning systems for fast demand response is particularly valuable in addressing urgent power balance requirements, owning to their large electricity consumption and inherent energy flexibilities. Limiting the number of operating chillers has proven to be a cost-effective means for direct load control in many studies. However, the conventional feedback control fails to distribute the limited cooling supply evenly, resulting in inherent uneven temperature rises among individual air-conditioned spaces when using this direct load control. To tackle this issue, a distributed cooperative control approach is put forward in this research. The proposed control is achieved based on a multi-agent system (i.e., valve agents and damper agents) in a self-organizing way. Each agent performs on-site control using the collected information of its own and its neighbors in a distributed architecture. The validation tests demonstrate that the proposed control approach is capable of efficiently managing the uneven temperature increases in different zones/spaces of the building. During the demand response event, a significant reduction of 2,562 kWh electricity is achieved, accounting for 19.7% of the electricity consumption using the conventional control.

Adaptive recurrent neural network for uncertainties estimation in feedback control system

In this paper, a recurrent neural network (RNN) is used to estimate uncertainties and implement feedback control for nonlinear dynamic systems. The neural network approximates the uncertainties related to unmodeled dynamics, parametric variations, and external disturbances. The RNN has a single hidden layer and uses the tracking error and the output as feedback to estimate the disturbance. The RNN weights are online adapted, and the adaptation laws are developed from the stability analysis of the controlled system with the RNN estimation. The used activation function, at the hidden layer, has an expression that simplifies the adaptation laws from the stability analysis. It is found that the adaptive RNN enhances the tracking performance of the feedback controller at the transient and steady state responses. The proposed RNN based feedback control is applied to a DC–DC converter for current regulation. Simulation and experimental results are provided to show its effectiveness. Compared to the feedforward neural network and the conventional feedback control, the RNN based feedback control provides good tracking performance.

Decentralized proportional-integral controller based on dynamic decoupling technique using Beckhoff TwinCAT-3.1

An improved technique for the design of decentralized dynamic decoupled proportional-integral (PI) controllers to control many variables of column flotation was developed and implemented in this paper. This work was motivated by challenges when working with multiple inputs and multiple outputs (MIMO) systems that are not controllable by conventional linear feedback controllers. Conventional feedback control design consists of various drawbacks when it comes to complex industrial processes. The introduction of decentralization, decoupling, and many advanced controls design methods overcomes these drawbacks. Hence, the design and implementation of control systems that mitigate stability for MIMO systems are important. The developed closed-loop model of the flotation process is implemented in a real-time platform using TwinCAT 3.1 automation software and CX5020 Beckhoff programmable logic controllers (PLC) through the model transformation technique. The reasons for using the CX5020 as an implementation environment were motivated by the reliability, and is built according to new industry standards, allowing transformation, which makes it more advantageous to be used more than any other PLCs. This is done to validate the effectiveness of the recommended technique and prove its usability for any multivariable system. Comparable numerical results are presented, and they imply that industrial usage of this method is highly recommended.

Enhancing Dynamic Stability Performance of Hybrid AC/DC Power Grids Using of a New Robust Controller

This research investigates the dynamic performance of a multi-terminal DC (MTDC) network linked to a wind farm side converter. The suggested approach is based on loop shaping, mutual quality decomposition, and an effective H controller architecture (MQDM). The transfer function was built as the wind power plant shifted from the Master-Slave control approach to the conventional feedback control method. The singular value curves of the original open-loop system and the suggested solution are compared in order to derive the weighting functions of the robust H controller. The control gain is then calculated by utilizing the MQDM approach to get the system's transfer function from the state space. By using the single value order reduction approach and frequency response analysis to generate a robust controller's gain, the suggested methodology was put into practice. The suggested method for controlling the DC grid voltage on the MTDC converter on the wind farm side has now been implemented in the MATLAB environment. Applying various uncertainties to the DC grid, wind turbines, and AC grid using simulation data demonstrates how reliable the suggested approach is.

Implementasi Sistem Kendali Keseimbangan Statis Pada Robot Quadruped Menggunakan Reinforcement Learning

The basic thing to consider when building a quadruped robot is the issue of balance. These factors greatly determine the success of the quadruped robot in carrying out movements such as stabilizing the body on an inclined plane, walking movements and others. Conventional feedback control methods by performing mathematical modeling can be used to balance the robot. However, this method still has weaknesses. The application of conventional feedback control methods often results in an inaccurate controller, so it must be manually tuned for its application. In this study, reinforcement learning methods were used using Q-Learning algorithms. The use of reinforcement learning methods was chosen because no mathematical calculations are needed to control the balance of quadruped robots. The process of learning the system to train the agent's abilities is carried out using a Gazebo simulator. The learning results show that the system could run well as evidenced by the higher value of sum rewards per episode.

Output-based event-triggered control for discrete-time systems with three types of performance analysis

<abstract><p>This paper considers an output-based event-triggered control approach for discrete-time systems and proposes three new types of performance measures under unknown disturbances. These measures are motivated by the fact that signals in practical systems are often associated with bounded energy or bounded magnitude, and they should be described in the $ \ell_{2} $ and $ \ell_\infty $ spaces, respectively. More precisely, three performance measures from $ \ell_{q} $ to $ \ell_{p} $, denoted by the $ \ell_{p/q} $ performances with $ (p, q) = (2, 2), \ (\infty, 2) $ and $ (\infty, \infty) $, are considered for event-triggered systems (ETSs) in which the corresponding event-trigger mechanism is defined as a function from the measured output of the plant to the input of the dynamic output-feedback controller with the triggering parameter $ \sigma (>0) $. Such a selection of the pair $ (p, q) $ represents the $ \ell_{p/q} $ performances to be bounded and well-defined, and the three measures are natural extensions of those in the conventional feedback control, such as the $ H_\infty $, generalized $ H_2 $ and $ \ell_1 $ norms. We first derive the corresponding closed-form representation with respect to the relevant ETSs in terms of a piecewise linear difference equation. The asymptotic stability condition for the ETSs is then derived through the linear matrix inequality approach by developing an adequate piecewise quadratic Lyapunov function. This stability criterion is further extended to compute the $ \ell_{p/q} $ performances. Finally, a numerical example is given to verify the effectiveness of the overall arguments in both the theoretical and practical aspects, especially for the trade-off relation between the communication costs and $ \ell_{p/q} $ performances.</p></abstract>

Simulation-guided feedforward-feedback control of melt pool temperature in directed energy deposition

High-performance closed-loop control systems are crucial to process stability and part quality in additive manufacturing processes. In this work, a simulation-guided feedforward-feedback control framework is developed for the directed energy deposition process to effectively control the melt pool temperature. The developed framework consists of two stages: offline iterative finite element simulation for feedforward control and online photodiode-based temperature monitoring for feedback control. The experimental results show that the feedforward-feedback control method can significantly improve the control performance when compared to the conventional feedback control method.

Fractional order adaptive MRAC controller design for high-accuracy position control of an industrial robot arm

Most of conventional feedback controllers become inefficient in hard industrial environments like in steel industry, because of uncertainties in the plant model, or process dynamics variation due to nonlinear actuators, and changes in the character of the disturbances. This paper proposes an adaptive control design based on fractional order model reference adaptive control (FOMRAC) strategy in order to deal with an uncertain horizontal positioning control of an unloading machine in a rotary hearth furnace for hot rolling operation. The proposed FOMRAC scheme uses the MIT rule as an adaptive mechanism with two main modifications comparatively to the conventional MRAC: the reference model is an adequate fractional order system and the parameter adjustment rule contains a fractional order integrator. Stability analysis of the proposed control scheme is performed using the Lyapunov stability theorem. Numerical simulations are presented to show the effectiveness of the proposed fractional adaptive schemes applied to an industrial robot arm loading round steel blocks from inside a rotary hearth furnace. After comparison with the conventional MRAC, it is shown that the performances of FOMRAC are superior to classical control schemes.

Federated Learning on the Road Autonomous Controller Design for Connected and Autonomous Vehicles

The deployment of future intelligent transportation systems is contingent upon seamless and reliable operation of connected and autonomous vehicles (CAVs). One key challenge in developing CAVs is the design of an autonomous controller that can accurately execute near real-time control decisions, such as a quick acceleration when merging to a highway and frequent speed changes in a stop-and-go traffic. However, the use of conventional feedback controllers or traditional learning-based controllers, solely trained by each CAV’s local data, cannot guarantee a robust controller performance over a wide range of road conditions and traffic dynamics. In this paper, a new federated learning (FL) framework enabled by large-scale wireless connectivity is proposed for designing the autonomous controller of CAVs. In this framework, the learning models used by the controllers are collaboratively trained among a group of CAVs. To capture the varying CAV participation in the FL training process and the diverse local data quality among CAVs, a novel dynamic federated proximal (DFP) algorithm is proposed that accounts for the mobility of CAVs, the wireless fading channels, as well as the unbalanced and non-independent and identically distributed data across CAVs. A rigorous convergence analysis is performed for the proposed algorithm to identify how fast the CAVs converge to using the optimal autonomous controller. In particular, the impacts of varying CAV participation in the FL process and diverse CAV data quality on the convergence of the proposed DFP algorithm are explicitly analyzed. Leveraging this analysis, an incentive mechanism based on contract theory is designed to improve the FL convergence speed. Simulation results using real vehicular data traces show that the proposed DFP-based controller can accurately track the target CAV speed over time and under different traffic scenarios. Moreover, the results show that the proposed DFP algorithm has a much faster convergence compared to popular FL algorithms such as federated averaging (FedAvg) and federated proximal (FedProx). The results also validate the feasibility of the contract-theoretic incentive mechanism and show that the proposed mechanism can improve the convergence speed of the DFP algorithm by 40% compared to the baselines.

Adaptive Internal Model Control Based on Parameter Adaptation

In actual industrial control, many objects have the characteristics of large lag and time-varying parameters, which makes it difficult for traditional PID control to obtain satisfactory control effects. Therefore, engineers need a control algorithm with a better control effect and a simple structure, and thus the internal model control is introduced into industrial process control. Internal model control has received a lot of attention in the control field because of its excellent control effect, and in ideal conditions, it can perfectly suppress external disturbances, and the output strictly tracks the input. On the basis of the internal model control, a two-degrees-of-freedom internal model control was developed. Compared with the conventional feedback control, the internal model control structure mainly embeds an internal model consistent with the control object in the control object, so the deviation between the internal model and the control object determines the quality of the control effect. In this paper, the internal model is changed to an adjustable parameter model, and the model parameters are adjusted in real time using the parameter adaptive algorithm, so that the model output error is as small as possible, or even zero. In order to solve the influence of interference on the system, compensation based on MRAC theory is used. In order to verify the feasibility of the algorithm, it was applied to the landing process control of fixed-wing Unmanned aerial vehicle) UAV and achieved satisfactory results.

A Control Strategy of LCL-Type Grid-Connected Inverters for Improving the Stability and Harmonic Suppression Capability

The conventional inverter-side current single-loop feedback control scheme is weak in suppressing the grid-side current harmonics, posing a challenge for an inverter to inject high-quality current under distorted grid voltage. With capacitor current compensation added, the control scheme achieves controllability of the grid-side current harmonics so that it can effectively suppress some specific harmonic components. However, due to the stability requirements, only a few low-order harmonic resonance controllers can be applied, which limits the mitigation of high-order harmonics. To tackle this problem, the grid-side current feedback control with inductor–capacitor–inductor (LCL) resonance damping is proposed in this paper. In this case, a higher LCL resonance frequency can be set compared to the inverter-side current single-loop feedback control scheme. Thereby, more resonance controllers can be applied to suppress high-order grid-side current harmonics. The active damping method of capacitor current proportional feedback plus capacitor voltage proportional feedback is adopted because of its high robustness to grid impedance variations. Furthermore, this paper reveals that the applied active damping method has a limitation in that it only considers a single inverter under inductive grid impedance, which cannot eliminate the risk of resonance caused by the interaction of multiple inverters and the grid. To address this issue, a phase lead compensator (PLC) is proposed, eliminating the resonance risk by removing the non-passive region of the inverter output admittance. To retain the advantage of the inverter-side current single-loop feedback control scheme, i.e., only a few measuring devices are required, a digital differentiator is used to calculate the capacitor current from the capacitor voltage. The difference between the measured inverter-side current and the calculated capacitor current is taken to approximate the grid-side current for the feedback control. The control performance is comparable to using the grid-side current for feedback. Simulation and experimental results demonstrate that the proposed control scheme endows the inverter with good stability and current quality without extra measurement devices.

Model-Based Closed-Loop Control for High-Power Current Source Rectifiers Under Selective Harmonic Elimination/Compensation PWM With Fast Dynamics

Selective harmonic elimination/compensation PWM (SHE/SHC-PWM) is widely used in current source rectifiers (CSRs) under normal/distorted grid conditions due to its best line current harmonic performance with reduced switching losses. However, due to the off-line nature of SHE/SHC-PWM that operates over fundamental frequency, its dynamic performance under closed-loop control is usually not satisfactory compared with the space-vector modulation or carrier-based PWM (CB-PWM) for CSR. To improve the closed-loop dynamics of CSR under SHE/SHC-PWM, a model-based closed-loop control scheme is proposed in this article. In case of transient when the dc current reference of CSR changes, the new input references of the SHE/SHC-PWM module, i.e., the modulation index and delay angle, are directly calculated and updated in real-time based on the mathematical model of CSR derived in this article. As a result, the dynamic performance for dc current tracking under closed-loop SHE/SHC-PWM operations can be improved greatly due to this added model-based feedforward path, while the conventional feedback controller is still adopted but only for steady-state error corrections. The application of the proposed method on back-to-back current source converter (CSC) motor drives is also introduced. Simulation and experimental results validate this proposed control for CSR with fast dynamics.

Event-triggered stabilisation of sampled-data singular systems: a hybrid control approach

This paper provides a design of event-triggered feedback stabilisation scheme for a class of sampled-data singular systems, in which the input matrix corresponding to the algebraic equation is not full column rank. First, unlike the conventional piecewise-constant sampled-data feedback controller, a novel hybrid impulsive and sampled-data state feedback controller is designed, which can effectively match with the algebraic equation at sampling instants. An improved discrete event-triggering mechanism with checking period is proposed to achieve exponential stabilisation. Based on these, by applying the input delay approach and defining a new input-delay dependent Lyapunov functional, a sufficient stability condition is presented which ensures that the solution of the event-triggered control system exists and is unique, and the system is exponentially stable with definite decay rate. The estimate of decay coefficient is also given explicitly. Then, an optimisation-based method is developed to jointly design the hybrid controller gains and the event parameters. Finally, two simulation examples including a linearised 2-generator and 6-bus power network system are given to illustrate the validity of the new approach.

Carrier envelope phase stabilization with a true common-path collinear ƒ–2ƒ interferometer

A variety of technologies have been so far demonstrated for the carrier-envelope phase (CEP) control of mode-locked oscillators. An instrumental part of these technologies is the configuration of interferometers and the application of feedback control. Of the devices, the collinear ƒ–2ƒ interferometer developed in this work is compact and robust against disturbances because the optical paths of the ƒ and 2ƒ components are collinear within the interferometer. To compensate for the delay time between the ƒ and 2ƒ components, a birefringent time plate made of α-BBO is installed in the interferometer. We achieved an in-loop CEP stability of 27 mrad (rms) using conventional feedback control with an acousto-optic modulator to control the pump power. We believe that the collinear ƒ–2ƒ interferometer offers the best choice as an interferometer for CEP stabilization of the front-end oscillator in the chirped-pulse amplification system.

Critical Assessment of Control Strategies for Industrial Systems with Input–Output Constraints

Safety and economic considerations pose constraints on inputs and outputs of industrial processes. A conventional feedback control system is not equipped to handle these constraints, and hence, industrial practitioners resort to the addition of ad hoc elements such as selector blocks, in conjunction with more sophisticated control strategies such as split range control. However, this approach lacks an organized and prudent methodology and is usually carried out with the aid of experiential knowhow of plant operators and control engineers. Some recent developments in the literature address this issue, however simultaneously alluding to model predictive control (MPC) as a more natural approach for such systems. In this work, we evaluate the efficacy of the aforementioned approach of simple ad hoc additions to accommodate constraints in a process. To this end, with the aid of a continuously stirred tank reactor model, we illustrate the ease of deployment of selector blocks for handling constraints in simple systems and assess their limitations as the system becomes more complex. Next, we consider a heat exchanger employed in a South-East Asian processing facility where the performance of the currently used selector-based architecture is deemed unsatisfactory by the operators. The current control structure was evaluated for conformity with the approaches suggested in the recent literature and was found to be sound; hence, the unsatisfactory performance stems from the inherent limitations of the proportional integral control structure and/or improper tuning. The implementation of MPC on the exchanger system leads to performance amelioration and enhanced robustness.