Hybrid Predictive Control Research Articles

Networked explicit hybrid predictive control of an industrial benchmark process

Hybrid systems have acquired deep attention in research as many systems in industries and laboratories manifest hybrid dynamic behaviors. Optimal control design for hybrid systems exhibits challenges for controlling the interacting continuous and discrete dynamics. A double-tank hybrid system (DTHS) is proposed in this paper with a mixed-integer actuation mechanism. The proposed DTHS aims to control the liquid level of the two tanks simultaneously by a networked hybrid predictive controller. Hybrid frameworks are provided to investigate the system through reachability analysis. An explicit hybrid model predictive control (EHMPC) has been developed to control the levels of the DTHS. Hardware-In-the-Loop (HIL) simulation is performed for assessing the performance of the EHMPC with a hardware real-time networked controller. The networked real-time controller is constructed and implemented using the Node.js frameworks. The simulation and realtime results are presented and discussed to evaluate the hybrid controller performance.

Hybrid predictive based control of precipitation in a water-scarce region: A focus on the application of intelligent learning for green irrigation in agriculture sector

A growing need for irrigation in agriculture results from recent climatic parameter uncertainties brought on by climate change, global warming, and other factors. The present-day tumultuous, unpredictable, ever-changing, and ambiguous nature of the onset, cessation, and duration of adverse weather conditions poses a formidable obstacle for farmers in formulating informed judgments pertaining to agricultural practices. In this study, the metrological simulation was carried out based on different input variables, including wind speed, wind direction, relative humidity, and minimum and maximum temperature, to predict the rainfall in the arid agricultural area of Kano, Nigeria. For this purpose, an adaptive neuro-fuzzy inference system (ANFIS), feed-forward neural network (FFNN), and multi-linear regression (MLR) were utilized. Five evaluation criteria for predictive control, including determination coefficient (R2), Nash–Sutcliffe efficiency (NSE), mean square error (MSE), mean absolute error (MAE), and root mean square error (RMSE), were used to figure out how accurate the models were based on how the features were chosen. The output proved the reliable accuracy of intelligent regression learning. The results depicted that MLR-M1 with R2 = 0.9989, NSE = 0.9872, and RMSE = 0.0016 performs the best at predicting rainfall, even though all three computational models (ANFIS, FFNN, and MLR) produced good results. The predictive models justified reliable tools for the management of water resources, especially in the agricultural sector.

Finite-time boundedness of networked control systems via hybrid predictive control based on cloud storage method

In this paper, we investigate the problem of predictive control for a class of networked control systems (NCSs) based on the cloud storage method (CSM). To improve the system’s transient performance and actively compensate for the time-varying delays (TVDs) and data packet dropouts (DPDs), a hybrid networked predictive control (NPC) method is proposed. The switched control system is established based on a class of switching signals varying with the TVDs and DPDs. Novel finite-time hybrid stabilization controllers are proposed by using NPC theory. Then, sufficient conditions are developed to ensure the finite-time boundedness (FTB) of the augmented switched systems. Two experiments illustrate the performance to verify the effectiveness of the proposed results.

Hybrid Data-Model Predictive Control for Enabling Participation of Renewables in Regulating Reserves Service

A long-horizon hybrid data-model predictive direct power control (HD-MPDPC) scheme is proposed. The hybrid control scheme is for grid-side power electronic converters (PEC) in grid-connected renewable energy sources (RES). Computational burden of conventional MPDPC is alleviated by reducing the number of candidate voltage vectors to be examined in the cost function. This is achieved by employing data-driven forecast of RES output power to identify and eliminate the voltage vectors that do not counteract active and reactive power errors appropriately. RES power is forecasted using recurrent neural networks. The method for eliminating voltage vector candidates is inspired from direct power control, where output of active and reactive power hysteresis controllers along with the sector in which grid voltage vector lies are used to determine the switching states of the grid-side PEC. Thanks to reduced computational complexity, the hybrid predictive control scheme dispatches RES power more reliably over long horizons. This in turn enables RES as a regulating reserves service provider in power systems. Hardware-in-the-loop studies of a grid-connected wave energy conversion system verify reduced computational complexity of HD-MPDPC and its effectiveness in controlling RES output power over long horizons.

Robust Predictive Control of Grid-Tied Modular Multilevel Converters for HVDC Systems With Virtual-Flux Based Online Inductance Estimation

Direct model predictive control (DMPC) has become an effective control alternative for the grid-tied modular multilevel converter. However, classical DMPC of the grid-tied MMC relies on accurate system model. System parameter variations will deteriorate the control performances of power, grid current, and circulating current. Besides, the inherent fewer switching actions of DMPC will also lead to poor circulating current performance. To solve these issues, this work proposes a robust DMPC solution for the grid-tied MMC. Novelties lie in two aspects. First, a virtual-flux (VF) based online inductance estimator is proposed, which utilizes the VF estimation error caused by inductance deviation. It achieves fast and accurate estimation of AC equivalent inductance, and reduces the dependency on system parameters. Thereafter, we propose a hybrid predictive control frame with a simple proportional-resonant (PR) controller, which considerably suppresses the circulating current. Finally, experimental results verify the effectiveness of the proposed solution.

Development and performance evaluation of a chiller plant predictive operational control strategy by artificial intelligence

Traditionally, chiller plants are controlled and monitored by a predetermined control strategy to ensure appropriate operation based on the designed system configuration. With the use of new technology of variable speed drive (VSD) for compressors, smart control strategies could be leveraged to enhance the system efficiency in lieu of traditional control strategies. For example, using orderly and straightforward switching procedures without considering various factors in switching the units, including the high-efficiency partial load range benefitted from the VSD, the actual performance of the units as a whole and the variable chilled water flow rate, result in the chiller plant not operating at maximum performance and efficiency. To address these issues, a hybrid predictive operational chiller plant control strategy is proposed to optimize the performance of the chiller plant. Artificial intelligence is employed as the data mining algorithm, with big data analysis based on the actual acquired voluminous operation data by fully considering the characteristics of chiller plants without additional installation of large-sized and high-priced equipment. Artificial neural network (ANN) was employed in the control strategy to predict the future outdoor temperature, building cooling load demand and the corresponding power consumption of the chiller plants. At the same time, particle swarm optimization (PSO) was applied to search for the optimized setpoints, e.g., chilled water supply temperature, operating sequence, chilled water flow rate, for the chiller plants. The developed control strategy has been launched in a chiller plant with a cooling capacity of 7,700 kW installed in a hospital in Hong Kong. The system coefficient of performance (COP) and overall energy consumption of the chiller plants were enhanced by about 8.6% and reduced by about 7.9%, respectively, compared with the traditional control strategy. This real-time, continuous, automatic optimization control strategy can determine the most efficient combination of operating parameters of a chiller plant with different control settings. This ensures that the chiller plant operates in its most efficient mode year-round under various operational conditions.

System Identification and Controller Design of a Novel Autonomous Underwater Vehicle

Autonomous underwater vehicle is an effective tool for humans to explore the ocean. It can be used for the monitoring of underwater structures and facilities, which puts forward more accurate and stable requirements for the system operation of the autonomous underwater vehicle. This paper studies the system and structural design, including the parameter identification design and control system design, of a novel autonomous underwater vehicle called “Arctic AUV”. The dynamic mathematical model of the “Arctic AUV” was established, and the system parameter identification method based on the multi-sensor least squares centralized fusion algorithm was proposed. The simplification of the mathematical model of the robot was theoretically derived, and the online parameter identification and motion control were combined, so that the robot could cope with the influence of the arctic water velocity and external turbulence. Based on the hybrid control scheme of adaptive PID and predictive control, the accurate motion control of the “Arctic AUV” was realized. A prototype of “Arctic AUV” was developed, and system parameter identification experiments were carried out in indoor pool water. Hybrid adaptive and predictive control experiments were also carried out. The validity of the parametric design method in this paper was verified, and by comparative experiment, the effect of the control method proposed in this paper was better than the traditional method.

Coordination of low-power nonlinear multi-agent systems using cloud computing and a data-driven hybrid predictive control method

This paper studies the coordinated control issue for low-power nonlinear multi-agent systems (LNMASs). To improve the calculation ability of the LNMAS for massive real-time data and alleviate the communication burden of individual agents, cloud computing systems are employed to establish the multi-agent system instead of low-power microprocessors. To reduce the adverse effects caused by the network delays and data loss in all communication channels and coordinate the outputs of all agents, a novel data-driven networked hybrid predictive control (DDNHPC) method using only input and output information of the system is proposed. In the DDNHPC method, the nonlinear agents are described as dynamic data models for controller design. Based on the dynamic data models, the hybrid predictive method including direct output predictions and further switching output predictions is designed to establish the active outputs coordination relationship among all agents. Simulation and cloud-based experimental examples are performed to illustrate the effectiveness and applicability of the proposed method. The achievement also provides a solution to implement the coordination of large-scale intelligent manufacturing.

MPC for heating systems with minimum up-and down-time requirements

Abstract We address the problem of operating up- and down-time restricted components with activation thresholds using linear predictive control instead of hybrid predictive control. Accounting for operational requirements of this kind usually requires optimal control problems with discrete variables, which are computationally demanding. As a workaround, run-time requirements are often ignored in the control design and enforced independently from the optimal input value. We present conditions and an algorithm that ensures minimum run-time requirements in linear MPC. The control performance for a heating application is compared to linear MPC with rule-based post-processing and hybrid MPC. The results show that our approach ensures the up-and down-time requirements at a high control performance and is significantly faster than hybrid MPC.

Finite Control Set MPC with Fixed Switching Frequency Applied to a Grid Connected Single-Phase Cascade H-Bridge Inverter

Finite control set model predictive control (FCS-MPC) has been widely investigated in recent years due to its ability to handle optimization problems with multiple control objectives in a diverse variety of systems. Moreover, its direct implementation in digital-based systems has made it an attractive strategy in static power converter applications. However, its characteristics such as variable switching frequency and spread harmonic spectrum limit the use of standard MPC due to power losses, audible noise, steady-state performance, and resonances. To mitigate these problems and extend the FCS-MPC applications to new areas, this paper proposes a new hybrid predictive control scheme, capable of achieving a harmonic spectrum distribution similar to that obtained with a pulse-width modulation scheme. The proposed strategy is based on a system model to generate an optimization, and, at the same time, an input restriction in the cost function of the standard FCS-MPC. This new approach is validated through experimental tests carried out in a grid-connected Cascaded H-bridge inverter.

A general closed-loop power-decoupling control for reduced-switch converter-fed IM drives

Twice ripple power exists in single-phase to three-phase reduced-switch induction motor drives, which degrades the performance of the drive system, such as larger total harmonic distortion of the grid current, the larger ripple voltage in the dc link and the larger torque ripple. Most existing power decoupling schemes are open-loop decoupling methods. However, the open-loop decoupling method has the disadvantages of heavy computation and strong dependency of parameters. This paper analyzes all steady state solutions of the decoupling capacitor voltage from the point of view of solving differential equations and presents a unified multi-proportional-resonant closed-loop decoupling method to suppress the ripple voltage on the dc bus. The whole drive system adopts a hybrid predictive control strategy with dual loops to coordinate the operation of each part. Comprehensive experimental results are presented to verify the effectiveness of the presented method.

Hybrid Ventilation System and Soft-Sensors for Maintaining Indoor Air Quality and Thermal Comfort in Buildings

Maintaining both indoor air quality (IAQ) and thermal comfort in buildings along with optimized energy consumption is a challenging problem. This investigation presents a novel design for hybrid ventilation system enabled by predictive control and soft-sensors to achieve both IAQ and thermal comfort by combining predictive control with demand controlled ventilation (DCV). First, we show that the problem of maintaining IAQ, thermal comfort and optimal energy is a multi-objective optimization problem with competing objectives, and a predictive control approach is required to smartly control the system. This leads to many implementation challenges which are addressed by designing a hybrid ventilation scheme supported by predictive control and soft-sensors. The main idea of the hybrid ventilation system is to achieve thermal comfort by varying the ON/OFF times of the air conditioners to maintain the temperature within user-defined bands using a predictive control and IAQ is maintained using Healthbox 3.0, a DCV device. Furthermore, this study also designs soft-sensors by combining the Internet of Things (IoT)-based sensors with deep-learning tools. The hardware realization of the control and IoT prototype is also discussed. The proposed novel hybrid ventilation system and the soft-sensors are demonstrated in a real research laboratory, i.e., Center for Research in Automatic Control Engineering (C-RACE) located at Kalasalingam University, India. Our results show the perceived benefits of hybrid ventilation, predictive control, and soft-sensors.

Hybrid modelling and predictive control of utility-scale variable-speed variable-pitch wind turbines

Wind energy has proven to be the highest reliable source of renewables due to the maturity of the technology. Wind turbine (WT) systems are complex systems with undergoing development process that requires innovative methods of design and control. In this paper, WTs are studied through hybrid systems framework. Hybrid models of WT are extracted with representable dynamics from nonlinear complex design code. The WT model is formulated into a mixed-logical dynamical model (MLD) and a piecewise affine (PWA) model. Then, a receding horizon control strategy is applied to WT hybrid model resulting in a hybrid model predictive control (HMPC) with mixed-integer programming (MIP) problem. The performance of the proposed controller is compared against the baseline controller within a simulation environment for the National Renewable Energy Laboratory (NREL) 5MW benchmark WT as a case study. The analysis and investigation of HMPC highlight its capability as a potential tool for exploiting the control objectives of WT systems.

Hybrid Model Predictive Control of a Variable Displacement Engine Mode Management

The use of Hybrid Predictive Control for model-based propulsion control applications is considered. A Variable Displacement Engine (VDE) control problem is considered, involving both continuous-time dynamics and discrete control actions in the form of activating/deactivating the engine cylinders. Hybrid Model Predictive Control is one of the most successful hybrid control schemes and builds upon predictive control methods developed for engine torque management. The ways in which preview information can be used to improve controller performance are considered, as well as simplifications to the hybrid control algorithms to reduce the computational burden. Several hybrid control design approaches are compared using a simulation of a VDE engine. The aim is to optimize the total system behaviour to provide good torque tracking, reduced fuel consumption and smooth cylinder switching. The main contribution is the demonstration that hybrid predictive control can provide a practical solution to an engine control application with the potential to enhance performance and with options to reduce complexity.

Hybrid predictive control with simple linear control based circulating current suppression for modular multilevel converters

Hybrid predictive control with simple linear control based circulating current suppression for modular multilevel converters

A Hybrid Predictive Control for a Current Source Converter in an Aircraft DC Microgrid

A current source converter (CSC) is a promising topology for interfacing aircraft generators with the onboard DC microgrid. In this paper, a hybrid predictive control is proposed for the CSC with an output LC filter in such application. Deadbeat predictive control with larger sampling time is applied to the output circuit, generating reference source currents. Finite-set model predictive control with smaller sampling time is applied to the input circuit to achieve sinusoidal source currents, which is simplified by saving the source current predictions. The proposed scheme eliminates both the proportional-integral controller and the weighting factor, which are required in the existing studies. Besides, it has lower control complexity. A SiC-MOSFET-based prototype is used to verify the validity of the proposed scheme. Experimental results under 150 V/350–800 Hz AC input and 270 V DC output demonstrate the superior control performance.

Current control designed with model based predictive control for six-phase motor drives

Finite control set model-based predictive control techniques are distinguished by being an interesting alternative to traditional field oriented control techniques for multiphase drives due to their fast dynamic response and flexibility in the introduction of constraints. However, those predictive control techniques have some drawbacks regulating the (x−y) current components which can cause machine losses as well as a high computational burden. This paper presents a comparative study of an enhanced predictive current control technique with a conventional predictive control technique and two hybrid predictive control techniques applied to an asymmetrical six-phase induction motor drive in terms of current tracking, total harmonic distortion of stator currents and computational burden. Experimental results are reported to demonstrate the benefits of the different current control techniques by using the mean squared error and total harmonic distortion of stator currents as quality figures of merit and the number of floating point operations to measure the computational burden of each predictive control, thus concluding the advantages and limitations of each technique at transient and steady regimes.

A robust chiller sequencing control method for enhancing cooling supply reliability and energy efficiency

Control of heating, ventilation and air-conditioning systems (HVAC) has become an active area of research due to the significant energy expenditures and maintenance cost. Among the various controls of HVAC systems, chiller sequencing control is of great importance for multi-chiller plant to keep cooling reliability, system stability and energy efficiency. However, conventional chiller sequencing control usually lacks consideration of system stability and the accessional cost caused by the sequencing actions. Especially, those short-time or unnecessary sequencing actions become the main reason for operators to shut down automatic control. To enhance the reliability of chiller sequencing control, a robust hybrid predictive control is developed. This control is developed based on the typical total cooling load-based control and use an ARX prediction model to double check necessity of the sequencing actions. Case studies were carried out on a TRNSYS dynamic simulation with the configuration data from a real secondary school. Results showed that the switch number can be reduced more than 20% and operation cost can be reduced up to 4.25% without scarified in thermal comfort and energy efficiency.

Hybrid modeling and predictive control of intelligent vehicle longitudinal velocity considering nonlinear tire dynamics

A hybrid model predictive control (HMPC) strategy is proposed in this paper to autonomously regulate intelligent vehicle longitudinal velocity considering nonlinear tire dynamics. Since the tire longitudinal dynamics, which has significant influence on vehicle longitudinal velocity control performance, exhibits highly nonlinear dynamical behaviors, the piecewise affine (PWA) identification is conducted firstly based on experimental data to accurately model the tire longitudinal dynamics. On this basis, due to that the intelligent vehicle needs to be operated in two distinct modes (drive and brake) for autonomous velocity regulation and because of the affine submodel switching behaviors of the PWA-identified tire model, the intelligent vehicle longitudinal dynamics control process considered in this work can be regarded as a hybrid system with both continuous variables and discrete operating modes. Thus, the mixed logical dynamical framework is further used to model the intelligent vehicle longitudinal dynamics, and a HMPC controller, which allows us to optimize the switching sequences of the operation modes (binary control inputs) and the torques acted on the wheels (continuous control inputs), is tuned based on online mixed-integer quadratic programming. Simulation results finally demonstrate the effectiveness of the proposed HMPC controller for intelligent vehicle longitudinal velocity regulation under typical driving conditions.

A hybrid predictive sequencing control for multi-chiller plant with considerations of indoor environment control, energy conservation and economical operation cost

Reliable automatic control of heating, ventilating and air-conditioning (HVAC) systems is essential to address buildings energy efficiency and response to the worldwide challenges of human health, energy conversation and carbon emissions. Among the various controls of HVAC systems, chiller sequencing control is of great importance for multi-chiller plant to keep cooling reliability, system stability and energy efficiency. However, conventional chiller sequencing control usually lacks consideration of system stability and the accessional cost caused by the sequencing actions. Especially, those short-time or unnecessary sequencing actions become the main reason for operators to shut down automatic control. To enhance the reliability of chiller sequencing control, a hybrid predictive control is developed. This control is developed based on a typical sequencing control strategy and use an autoregressive with exogenous (ARX) prediction model to double check the necessity of sequencing commands so as to reduce the unexpected sequencing actions. Case studies were carried out on a secondary school in subtropical city. Results showed that the switch number and economical operation cost can be reduced more than 20% and 6.72%, respectively. Additionally, energy consumption was also reduced while keeping acceptable indoor environment, which is of great importance for carbon emissions’ reduction and sustainable city development.