Model Predictive Control Approach Research Articles

Performance Improvement of Predictive Voltage Control for Interlinking Converters of Integrated Microgrid

Voltage quality in Renewable Energy System may be harmed by varying power demand and fluctuating energy source outputs. The model predictive voltage (MPV) control approach is suggested in this study. The proposed control strategy enables appropriate power interactions between the AC and DC subgrids while sharing power fluctuations in a coordinated way. Both the AC and DC subgrids support each other in accordance with their normalized relative changes in AC frequency and DC voltage, respectively. A typical photovoltaic (PV) wind-battery-based hybrid AC/DC microgrid is modelled and investigated, and the usefulness of the proposed scheme is validated under the renewable energy variations and load perturbations. MATLAB/Simulink has validated the efficacy of the suggested strategy. A set of comparative tests was conducted between conventional Proportional-Integral (PI) control, Sliding Mode Control (SMC) strategies, and a novel Model Predictive Voltage Control (MPV) approach under diverse conditions. The aim was to comprehensively assess the advantages of the proposed MPV method. Results demonstrate that, in contrast to traditional PI and SMC control strategies, the MPV technique exhibits distinct benefits in terms of superior control performance and achieves exceptionally low Total Harmonic Distortion (THD) values. Ultimately, the proposed solution not only raises the performance standard but also ensures the stable operation of the Renewable Energy System (RES).

Bidirectional DC-DC Converter and Improved Electrical Vehicle Dynamic Response Control

Background: In automotive applications where bidirectional power flow is necessary to lighten the power system, dual active bridge (DAB) converters are frequently employed. Variations in the required output voltage, erratic input voltage, and shifting loads all have an impact on this converter. As a result, converter performance has to be improved. To increase efficiency, the current stress of the DC-DC converter must be optimised. This paper proposes a control scheme for the coupled inductor bidirectional DC-DC (CIB DC-DC) converter utilising both model predictive control (MPC) and a proportional-integral (PI) controller. The integration of these control techniques aims to enhance the performance and efficiency of the converter. Methods: The MPC algorithm is employed to predict the converter's future behaviour based on a dynamic model, taking into account system constraints and performance criteria. By optimising the control action over a finite time horizon, the MPC algorithm ensures an optimal response, considering the current state and anticipated changes. Additionally, a PI controller is incorporated to augment the control strategy. The proportional component of the PI controller enables a fast initial response to the error between the desired and actual converter outputs. The integral component eliminates steady-state errors and provides robustness against disturbances, resulting in improved overall system performance. Results: The proposed control scheme is implemented and evaluated through simulations and experimental tests on a prototype converter. The results demonstrate the effectiveness of the combined MPC and PI controller approach. Conclusion: The coupled inductor bidirectional DC-DC (CIB DC-DC) converter using MPC can provide precise control of power flow between two voltage domains, enabling efficient bidirectional power transfer. The predictive capabilities of MPC allow it to adapt to varying load conditions and respond quickly to changes, ensuring stable operation and accurate regulation of voltage and current. Overall, the coupled inductor bidirectional DC-DC converter controlled using MPC over PI offers improved performance, efficiency, and flexibility compared to traditional control methods. MPC can handle the complex dynamics response and non-linear characteristics of the converter, making it suitable for bi-directional vehicle charging applications, where precise control and high efficiency can be achieved.

An Iterative Adaptive Dynamic Programming Approach for Macroscopic Fundamental Diagram-Based Perimeter Control and Route Guidance

Macroscopic fundamental diagrams (MFDs) have been widely adopted to model the traffic flow of large-scale urban networks. Coupling perimeter control and regional route guidance (PCRG) is a promising strategy to decrease congestion heterogeneity and reduce delays in large-scale MFD-based urban networks. For MFD-based PCRG, one needs to distinguish between the dynamics of (a) the plant that represents reality and is used as the simulation tool and (b) the model that contains easier-to-measure states than the plant and is used for devising controllers, that is, the model-plant mismatch should be considered. Traditional model-based methods (e.g., model predictive control (MPC)) require an accurate representation of the plant dynamics as the prediction model. However, because of the inherent network uncertainties, such as uncertain dynamics of heterogeneity and demand disturbance, MFD parameters could be time-varying and uncertain. Conversely, existing data-driven methods (e.g., reinforcement learning) do not consider the model-plant mismatch and the limited access to plant-generated data, for example, subregional OD-specific accumulations. Therefore, we develop an iterative adaptive dynamic programming (IADP)-based method to address the limited data source induced by the model-plant mismatch. An actor-critic neural network structure is developed to circumvent the requirement of complete information on plant dynamics. Performance comparisons with other PCRG schemes under various scenarios are carried out. The numerical results indicate that the IADP controller trained with a limited data source can achieve comparable performance with the “benchmark” MPC approach using perfect measurements from the plant. The results also validate the IADP’s robustness against various uncertainties (e.g., demand noise, MFD error, and trip distance heterogeneity) when minimizing the total time spent in the urban network. These results demonstrate the great potential of the proposed scheme in improving the efficiency of multiregion MFD systems. Funding: The work was jointly funded by the National Natural Science Foundation of China under [Grant 72071214] (R. Zhong), the Dit4Tram project from the European Union’s Horizon 2020 Research and Innovation Programme under [Grant 953783] (N. Geroliminis), and the Research Student Attachment Programme of The Hong Kong Polytechnic University (C. Chen). Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2023.0091 .

Influence of usage and model inaccuracies on the performance of smart hot water heaters: lessons learned from a demand response field test

Domestic hot water heaters are considered to be easily integrated as flexible loads for demand response. While literature grows on reproducible simulation and lab tests, real-world implementation in field tests considering state estimation and demand prediction-based model predictive control approaches is rare. This work reports the findings of a field test with 16 autonomous smart domestic hot water heaters. The heaters were equipped with a retrofittable sensor/actuator setup and a real-time price-driven model predictive control unit, which covers state estimation, demand prediction, and optimization of switching times. With the introduction of generic performance indicators (specific costs and thermal efficiency), the results achieved in the field are compared by simulations to standard control modes (instantaneous heating, hysteresis, night-only switching). To evaluate how model predictive control performance depends on the user demand prediction and state estimation accuracy, simulations assuming perfect predictions and state estimations are conducted based on the data measured in the field. Results prove the feasible benefit of RTP-based model predictive control in the field compared to a hysteresis-based standard control regarding cost reduction and efficiency increase but show a strong dependency on the degree of utilization.

Model predictive control (MPC) applied to a simplified model, magnetic nanoparticle hyperthermia (MNPH) treatment process

Magnetic nanoparticle hyperthermia (MNPH) has emerged as a promising cancer treatment that complements conventional ionizing radiation and chemotherapy. MNPH involves injecting iron-oxide nanoparticles into the tumor and exposing it to an alternating magnetic field (AMF). Iron oxide nanoparticles produce heat when exposed to radiofrequency AMF due to hysteresis loss. Minimizing the non-specific heating in human tissues caused by exposure to AMF is crucial. A pulse-width-modulated AMF has been shown to minimize eddy-current heating in superficial tissues. This project developed a control strategy based on a simplified mathematical model in MATLAB SIMULINK® to minimize eddy current heating while maintaining a therapeutic temperature in the tumor. A minimum tumor temperature of 43 [°C] is required for at least 30 [min] for effective hyperthermia, while maintaining the surrounding healthy tissues below 39 [°C]. A model predictive control (MPC) algorithm was used to reach the target temperature within approximately 100 [s]. As a constrained MPC approach, a maximum AMF amplitude of 36 [kA/m] and increment of 5 [kA/m/s] were applied. MPC utilized the AMF amplitude as an input and incorporated the open-loop response of the eddy current heating in its dynamic matrix. A conventional proportional integral (PI) controller was implemented and compared with the MPC performance. The results showed that MPC had a faster response (30 [s]) with minimal overshoot (1.4 [%]) than PI controller (115 [s] and 5.7 [%]) response. In addition, the MPC method performed better than the structured PI controller in its ability to handle constraints and changes in process parameters.

Stability analysis for an ad-hoc model predictive control in DC/DC converters with a constant power load

This paper presents a stability analysis for a horizon-one continuous control set model predictive control (MPC) for DC/DC converters connected to a constant power loads (CPLs). Different MPC approaches have been previously presented, each tailored to a specific type of converter but without a formal stability proof, failing to ensure robust performance across various operating conditions. Unlike previous works in the scientific literature, our approach is general and applicable to any second-order DC/DC converter, including the Buck, Boost, Buck-Boost, and non-inverting Buck-Boost converters. A formal stability analysis was conducted using the proposed approach, initially for the open-loop system and subsequently for the closed-loop system. The trace-determinant or triangle criterion was employed for discrete-time systems in R2. Three analyses were conducted to evaluate the performance of the proposed MPC, showing the stability issues of DC/DC converters with CPLs in open loops. This work proposes solutions using the suggested technique. Numerical simulations utilizing the Boost and Buck-Boost converters demonstrate that, in open-loop operation, even minor perturbations in the CPL value can destabilize the behavior of electrical variables. In addition, simulations show that, for the Boost converter, the settling time was about 1.5ms. Meanwhile, in the Buck-Boost case, it was about 5ms. Standardized control indicators, such as the integral absolute error criterion (IAE), the integral of the time-multiplied absolute error criterion (ITAE), and the integral of the time-multiplied square error criterion (ITSE), confirm the effective performance of the proposed one-step MPC. The values range between 1.521×10−2 and 4.591×10−6 for the Boost converter, and between 6.607×10−3 and 3.366×10−6 for the Buck-Boost topology.

Cooperative Decision-Making Approach for Multiobjective Finite Control Set Model Predictive Control Without Weighting Parameters

Finite control set model predictive control (FCS-MPC) has gained increasing popularity as an emerging control strategy for electrical drive systems. However, it is still a challenging task to optimize weighting parameters, as multiple objectives are involved in the customized cost function. A cooperative decision-making approach for FCS-MPC is proposed in this article, to solve the optimization problems with manifold control objectives. By splitting the cost function, the optimization problem underlying multi-objective FCS-MPC is separated into a series of decomposed optimization problems. By doing so, the dimension of the decomposed problem is reduced to one. To collect the information for decision-making, an efficient sorting algorithm is applied for each control objective. The theory behind the cooperative decision-making approach is comprehensively analyzed, to validate both the effectiveness and efficiency of the proposed scheme. More specifically, the highlight is the adaptive mechanism on the number of desired candidates, to obtain a decent performance for torque and flux. The candidate which minimizes the switching frequency is selected as the optimal. The proposed scheme is experimentally verified and compared with the existing FCS-MPC without weighting parameters.

Simultaneous robust model predictive control and state estimation for nonlinear systems with state‐ and input‐dependent uncertainties

AbstractThe convergence and stability of uncertain nonlinear systems is a challenging problem in the nonlinear control area. Besides, in many practical cases, all states are not measurable and are affected by measurement noise. Based on this motivation, the first objective of this paper is to design a novel output feedback robust model predictive control approach for nonlinear systems with state‐ and input‐dependent uncertainties and measurement noise. This approach combines state estimation and robust model predictive control (MPC) into one min–max optimization and by solving the optimization, these two tasks are performed simultaneously. The studied nonlinear system comprises a linear part, a nonlinear part, and a function that denotes the state‐ and input‐dependent uncertainties. Therefore, the other objective is to reduce the computational complexity; thus, the system's nonlinear term and the aforementioned uncertainties are converted into additional disturbances. Subsequently, the optimization problem becomes a quadratic form, which leads to global convergence with the appropriate selection of objective function weights. Besides, this paper explores the convergence of the closed‐loop system states and the sufficient synthesis conditions to guarantee input‐to‐state stability. The implementation on a numerical example and a CSTR process demonstrate the applicability and reliability of the proposed approach.

Enhanced gas-lift system operation using LPV nonlinear model predictive control

In this work, we propose a novel Linear Parameter Varying (LPV) Model Predictive Control (MPC) scheme for oil production maximisation in wells operated with gas-lift systems. The control of a gas-lift system poses several engineering challenges, including the nonlinear nature of its behaviour and different kinds of process constraints. Recent studies indicate that MPC strategies stand out as promising solutions to address many of these issues – with results already published for gas-lift applications. However, when nonlinear models are used, the corresponding nonlinear MPC (NMPC) algorithms entail substantial computational costs, potentially complicating the feasibility of (real-time) embedded implementation. Furthermore, certain complexities in standard gas-lift system representations hinder the use of typical NMPC solvers (e.g. CasADi), thus making approximation-based schemes unavoidable. Accordingly, we propose an LPV formulation to describe the gas-lift system that ensures closed-loop stability and recursive feasibility when implemented as a model for MPC. In this regard, we synthesise an LPV MPC scheme and compare it to an NMPC through CasADi. Several realistic nonlinear numerical simulations are presented and our results indicate that the proposed LPV MPC scheme, in addition to being three times faster (in average), achieved an increase in the total amount of produced oil when compared to an NMPC via CasADi. Then, aiming to maximise oil production rate, we introduce a modified version of the LPV MPC approach, resulting in an increase in the total amount of produced oil of approximately 3% in the studied scenario, when compared to the unmodified LPV MPC scheme.

Minimizing total annualized cost per tonne of feed processed of a semicontinuous distillation process utilizing data-driven model predictive control

Semicontinuous distillation is a separation technique used to purify multicomponent mixtures with low to medium throughput. This research addresses the problem of designing a Data-driven Model Predictive Control (MPC) approach that enables minimizing the Total Annualized Cost (TAC) of the semicontinuous process per tonne of feed processed while maintaining the required product purity. In lieu of typically unavailable first principles models, the manuscript demonstrates the implementation of data-driven technique using data collected from an Aspen Plus Dynamics simulation as a test bed. A subspace model identification technique is adapted to develop a multi-model framework to capture the dynamic behavior of the process and then utilized within a Shrinking Horizon MPC (SHMPC) scheme, to achieve the required objective. The simulation results demonstrate a lowering of the TAC/tonne of feed by 11.4% compared to the traditional PI setup used in the previous studies.

A Gaussian-Process-Based Model Predictive Control Approach for Trajectory Tracking and Obstacle Avoidance in Autonomous Underwater Vehicles

To achieve the efficient and precise control of autonomous underwater vehicles (AUVs) in dynamic ocean environments, this paper proposes an innovative Gaussian-Process-based Model Predictive Control (GP-MPC) method. This method combines the advantages of Gaussian process regression in modeling uncertainties in nonlinear systems, and MPC’s constraint optimization and real-time control abilities. To validate the effectiveness of the proposed GP-MPC method, its performance is first evaluated for trajectory tracking control tasks through numerical simulations based on a 6-degrees-of-freedom, fully actuated, AUV dynamics model. Subsequently, for 3D scenarios involving static and dynamic obstacles, an AUV horizontal plane decoupled motion model is constructed to verify the method’s obstacle avoidance capability. Extensive simulation studies demonstrate that the proposed GP-MPC method can effectively manage the nonlinear motion constraints faced by AUVs, significantly enhancing their intelligent obstacle avoidance performance in complex dynamic environments. By effectively handling model uncertainties and satisfying motion constraints, the GP-MPC method provides an innovative and efficient solution for the design of AUV control systems, substantially improving the control performance of AUVs.

Eco-Driving on Hilly Roads in a Mixed Traffic Environment: A Model Predictive Control Approach

Human driving behavior significantly affects vehicle fuel economy and emissions on hilly roads. This paper presents an ecological (eco) driving scheme (EDS) on hilly roads using nonlinear model predictive control (NMPC) in a mixed traffic environment. A nonlinear optimization problem with a relevant prediction horizon and a cost function is formulated using variables impacting the fuel economy of vehicles. The EDS minimizes vehicle fuel usage and emissions by generating the optimum velocity trajectory considering the longitudinal motion dynamics, the preceding vehicle’s state, and slope information from the digital road map. Furthermore, the immediate vehicle velocity and angle of the road slope are used to tune the cost function’s weight utilizing fuzzy inference methods for smooth maneuvering on slopes. Microscopic traffic simulations are used to show the effectiveness of the proposed EDS for different penetration rates on a real hilly road in Fukuoka City, Japan, in a mixed traffic environment with the conventional (human-based) driving scheme (CDS). The results show that the fuel consumption and emissions of vehicles are significantly reduced by the proposed NMPC-based EDS compared to the CDS for varying penetration rates. Additionally, the proposed EDS significantly increases the average speed of vehicles on the hilly road. The proposed scheme can be deployed as an advanced driver assistance system (ADAS).

Integrating Model Predictive Control With Stormwater System Design: A Cost‐Effective Method of Urban Flood Risk Mitigation During Heavy Rainfall

AbstractThe integration of green‐gray infrastructures with advanced control approaches is revolutionizing the stormwater system retrofitting, emerging as an innovative strategy to mitigate urban flood risks. However, a major challenge lies in balancing the substantial investments of these infrastructure projects with their environmental benefits, such as reduced flooding volume and lower peak flow. Model predictive control (MPC), a dynamic and intelligent control approach, optimizes these environmental benefits but is underutilized in the system design phase for cost‐effectiveness analysis. This study introduces a multi‐scenario model framework that incorporates MPC and other control approaches into stormwater system designs, including the implementation of controlled storage tanks and green infrastructures. This framework provides comprehensive modeling tools for practitioners to evaluate the flood control benefits and costs across various infrastructure designs and control scenarios, ultimately identifying solutions that are both environmentally and economically viable. A case study conducted in a small urban catchment area in Shenzhen City, China, demonstrates the effectiveness of this framework. The results indicate that MPC outperforms other control scenarios, particularly under heavy or extreme rainfall conditions. Notably, MPC not only provides superior environmental benefits but also yields considerable cost savings, ranging from 1,787 to 9,371 USD per hectare compared to static control, equating to a 5% reduction in cost relative to rule‐based control. Such findings suggest that integrating MPC is a cost‐effective alternative to extensive infrastructure expansion for flood management, which significantly enhances the benefit contribution of controlled infrastructures without substantial additional expenses.

EV Charging Scheduling Under Demand Charge: A Block Model Predictive Control Approach

This paper studies the online scheduling of electric vehicle charging by a service provider subject to a demand charge in a distribution system. Demand charge imposes a penalty on the peak power consumption over each billing period, representing a substantial cost for the service provider with a large number of clients. Because the demand charge is calculated at the end of the billing period, it poses challenges in real-time scheduling when energy demand forecasts are inaccurate, resulting in either overly conservative power consumption or substantial demand charge. We propose a block model predictive control approach that decomposes the demand charge into a sequence of stage costs. Optimality conditions on demand patterns are also presented and analyzed. Numerical simulations demonstrate the efficacy of the proposed approach. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This paper addresses a significant practical problem of minimizing the demand charge on the real-time scheduling of deferrable demands. In particular, we consider a setting where a commercial electric vehicle (EV) charging service provider has to manage the online scheduling of a large number of arriving EVs at a charging facility subject to a maximum charging power constraint and a tariff with the demand charge. A major practical challenge is to balance the tradeoff between maximizing profit in scheduling as much EV charging as possible and the need to minimize penalty on the peak charging power. We propose a model predictive control strategy that decomposes the overall demand charge into a sequence of terminal costs. Also addressed is the practical constraint arising from the mismatched EV charging decision period and the power measurement period used to compute the demand charge. Using real data collected at the Adaptive Charging Network (ACN) testbed in simulations, the proposed approach yields 8-12% improvement in operational profit over existing benchmarks, while it has yet been tested in actual charging systems. In the future research, we will address the charging scheduling under demand charge over multiple charging stations.

Distributed multi-horizon model predictive control for network of energy hubs

The increasing penetration of renewable energy resources has transformed the energy system from a traditional hierarchical energy delivery paradigm to a distributed structure. Local energy hubs activates synergies among energy carriers rendering flexibility to the system and gives rise to possible energy trading among networked local energy hubs. Joint operation of such hubs and peer-to-peer trading between them can improve energy efficiency and support the integration of renewable energy resources. However, for such complex systems involving multiple stakeholders, both computational tractability and privacy concerns need to be accounted for. In this work, a novel multi-horizon distributed MPC framework is introduced for the control of energy hub networks. The multi-horizon approach increases the prediction horizon for MPC without compromising the time discretization or making the problem computationally intractable. The distributed scheme is based on a consensus alternating direction method of multipliers algorithm. It combines the superior performance of the centralized approach with the privacy preservation of the decentralized approach. A benchmark three-hub network is used to investigate the performance of the proposed method in simulation and compare it to the decentralized, centralized and standard model predictive control (MPC) approaches. The results show superior performance of the distributed multi-horizon MPC in terms of total cost, computational time, and robustness to demand and prices variations. Finally, the performance was also experimentally validated in real time by implementing the controller on a real energy hub system.

A Tube Linear Model Predictive Control Approach for Autonomous Vehicles Subjected to Disturbances

The path tracking performance of autonomous vehicles is degraded by common disturbances, especially those that affect the safety of autonomous vehicles (AVs) in obstacle avoidance conditions. To improve autonomous vehicle tracking performances and their computational efficiency when subjected to common disturbances, this paper proposes a tube linear model predictive controller (MPC) framework for autonomous vehicles. A bicycle vehicle dynamics model is developed and employed in the tube MPC control design in the proposed framework. A robust invariant set is calculated with an efficient linear programming (LP) method, and it is used to guarantee that the constraints are satisfied under common disturbance conditions. The results show that the computational cost of robust positively invariant sets that are constructed by the LP method is much less than that obtained by the traditional method. In addition, all the trajectories of the tube linear MPC successfully avoided obstacles when under disturbance conditions, but only about 80% of the trajectories obtained with the traditional MPC successfully avoided obstacles under disturbance conditions. The proposed framework is effective.

An improved ANFIS model predictive current control approach for minimizing torque and current ripples in PMSM-driven electric vehicle

An improved ANFIS model predictive current control approach for minimizing torque and current ripples in PMSM-driven electric vehicle

Using Control Barrier Functions to Incorporate Observability: Application to Range-Based Target Tracking

Abstract In many nonlinear systems, the observability of the system is dependent on its state and control input. Thus, incorporating observability into a control scheme can enhance an observer's ability to recover accurate estimates of unmeasured states, minimize estimation error, and ultimately, allow the original control objective to be achieved. The accommodation of observability, however, may conflict with the original control goal at times. In this paper, we propose the use of control barrier functions (CBFs) to enforce observability and thereby facilitate the convergence of the state estimate to the true state while accommodating the original control objectives. Motivated by practical applications for autonomous robots operating in global positioning system-denied environments, we focus on the problem of target tracking for a unicycle model when only the distance to the target is measured. The proposed approach is compared in simulation with a model predictive control (MPC) approach that treats an observability-related metric as part of the cost function, where several different options for the observability metric are explored. It is found that the CBF-based approach achieves control and estimation performance that is comparable to that of the MPC approach, but with significantly less computational complexity. These findings are further experimentally verified in range-based target tracking with a swimming robotic fish.

Dynamic modelling and a dual vector modulated improved model predictive control with auto tuning feature of active front‐end converters for distributed energy resources

AbstractThe operational performance of grid‐connected active front‐end converters (AFEs) faces challenges arising from the intricate interplay among phase‐locked loop (PLL) non‐linearities, grid impedance, and conventional control strategies, resulting in compromised stability. This study introduces a refined approach to dynamic model predictive control (MPC) by integrating recursive least squares (RLS) for the precise estimation of physical model parameters, thereby addressing stability concerns. Unlike conventional methodologies, the proposed enhanced RLS‐based MPC approach, equipped with an auto‐tuning feature, allows for the design of controllers without a prerequisite understanding of exact external dynamics. Notably, this technique exhibits exceptional disturbance rejection capabilities. The evaluation of the cost function at each sampling interval facilitates the determination of optimal switching states based on predicted variables. Gate pulses for the switches of the AFEs are generated accordingly. Employing a simulation platform, the proposed control structure's performance across varied conditions is comprehensively assessed, encompassing alterations in grid impedance and system non‐linearities. The method adeptly integrates inherent non‐linearities within the system, showcasing exceptional robustness in diverse dynamic scenarios. To further substantiate the efficacy of the proposed control system over conventional approaches, simulation results are validated using a laboratory hardware platform equipped with Typhoon HIL and dSPACE real‐time emulators, providing tangible evidence of the proposed control system's effectiveness in real‐world hardware setups. The multifaceted approach, encompassing precise parameter estimation, predictive control, auto‐tuning, disturbance rejection, robust design, and real‐time evaluation, collectively establishes a resilient foundation for enhancing and maintaining the overall stability of the system across diverse operating scenarios.

Automation of superconducting cavity cooldown process using two-layer surrogate model and model predictive control method

Superconducting cavity is the key equipment of the superconducting accelerator, which provides higher acceleration voltage and higher frequency power per unit length, and saves equipment space. Superconducting cavities need to be gradually cooled from ambient temperature (300 K) to the superconducting temperature (4.2 K or below) during the test and operation. The temperature difference on the cavity must be strictly limited during the cooldown process to prevent excessive thermal stress on the surface of the superconducting cavity. Since this cooldown process for the superconducting cavity is a typical large hysteresis, non-linear process that is difficult to control automatically using decoupled proportion integral derivative (PID) methods directly, a less efficient manual control scheme is normally adopted. In this paper, 3D numerical simulation, 1D pipe and 0D tank model with artificial neural network (ANN) were combined to generate a two-layer surrogate model that can balance computational accuracy and speed, to improve the automation and cooling efficiency of the superconducting cavity cooldown process. In order to achieve automatic control of the cooling procedure for the superconducting cavity, a model predictive control (MPC) approach was also built on the basis of this two-layer surrogate model. According to the results of the experiment test, the improved method could realize a quick and smooth cooldown process of the superconducting cavity, during which the temperature difference on the cavity could satisfy the requirements. Additionally, the improved automatic cooldown method was more adaptable and saved 29 % more time than the original manual control method. The foundation for a more intelligent automated control of future large cryogenic systems or other system with the large hysteresis, non-linear properties, was laid.