Data-driven Control Research Articles

Suppression of quasiperiodic thermoacoustic oscillations via genetic programming

We use genetic programming (GP) to discover data-driven control laws for the suppression of quasiperiodic oscillations in a prototypical thermoacoustic system. We rank the control laws based on a predefined cost function that accounts for the pressure amplitude and the actuation effort. We then breed subsequent generations of control laws via a tournament process. We find that GP closed-loop control is more effective than GP open-loop control and conventional periodic forcing, producing a similarly high degree of amplitude suppression but with the lowest actuation effort. We also find that GP closed-loop control can identify unforeseen actuation mechanisms, providing new insight into the physical coupling between the heat release rate and pressure fields.

Data-Driven Control of Distributed Event-Triggered Network Systems

The present paper deals with data-driven event-triggered control of a class of unknown discrete-time interconnected systems (a.k.a. network systems). To this end, we start by putting forth a novel distributed event-triggering transmission strategy based on periodic sampling, under which a model-based stability criterion for the closed-loop network system is derived, by leveraging a discrete-time looped-functional approach. Marrying the model-based criterion with a data-driven system representation recently developed in the literature, a purely data-driven stability criterion expressed in the form of linear matrix inequalities (LMIs) is established. Meanwhile, the data-driven stability criterion suggests a means for co-designing the event-triggering coefficient matrix and the feedback control gain matrix using only some offline collected state-input data. Finally, numerical results corroborate the efficacy of the proposed distributed data-driven ETS in cutting off data transmissions and the co-design procedure.

Data-Driven Control of Disturbance Compensator/PI Controller for Yaw-rate Control of Autonomous Driving

Data-Driven Control of Disturbance Compensator/PI Controller for Yaw-rate Control of Autonomous Driving

Koopman-Operator-Based Robust Data-Driven Control for Wheeled Mobile Robots

The Koopman operator theory offers a way to construct explicit control-oriented high-dimensional linear dynamical models for the original nonlinear systems, solely using the input–output data of the dynamical system. The modeling accuracy of the Koopman model largely depends on the basis functions (lifting functions), dimensionality, and data quality. However, there has not been a systematic way to solve the problems mentioned above. In this article, a Koopman-operator-based robust data-driven control framework is proposed for wheeled mobile robots, via incorporating tools from control theory, to solve the problem of modeling errors of the Koopman model. By employing an extended state observer, the modeling errors of the Koopman model, including unknown external disturbances, are online estimated and compensated in the control signal in real time. Then, sliding-mode control is used to synthesize the controller. Importantly, the method of virtual control input is proposed, to cope with the model errors arising from the rotational motion of all the mobile robots. Besides, stability analysis is conducted, and the optimal dimensionality of the Koopman model is experimentally selected. Finally, experimental tests on an omnidirectional mobile robot are carried out to verify the effectiveness of the proposed control scheme, in terms of tracking performance and robustness.

Real-Time Feasibility of Data-Driven Predictive Control for Synchronous Motor Drives

The data-driven control paradigm allows overcoming conventional troubles in the controller design related to model identifications procedures. Raw data are directly exploited in the control input selection by forcing the future plant dynamics to be coherent with previously collected samples. This article focuses, in particular, on the data-enabled predictive control algorithm. A relevant disadvantage of this algorithm is the fact that the complexity of the online control program grows with the dimension of the dataset. This issue becomes particularly relevant when considering embedded applications, such as the control of synchronous motor drives, characterized by challenging real-time constraints. This work proposes a systematic approach for dramatically reducing the complexity of such algorithms. Such methodology enables real-time feasibility of the constrained version of this control structure, which was previously precluded. Simulations and experimental results are provided to validate the method, considering the current control of an interior permanent magnet motor as test-case.

Data-Driven Steering of Concentric Tube Robots in Unknown Environments via Dynamic Mode Decomposition

Concentric Tube Robots (CTRs) are a type of continuum robot capable of manipulating objects in restricted spaces and following smooth trajectories. CTRs are ideal instruments for minimally invasive surgeries. Accurate control of CTR's motion in presence of contact with tissue and external forces will allow safe deployment of the robot in a variety of minimally invasive surgeries. Here, we propose a data-driven controller that can repeatedly and precisely direct the robot along predetermined deployment trajectories. The proposed controller doesn't rely on a mathematical model of the robot and employs Extended Dynamic Mode Decomposition (EDMD) to learn the nonlinear dynamics of the robot and the interaction forces on the fly. This enables the robot to follow desired trajectories in the presence of unknown perturbations, such as external forces. Experiments are carried out to evaluate the accuracy of the controller in steering the robot on arbitrary trajectories. Results demonstrate that the robot can track trajectories with a mean accuracy of 2.4 mm in repeated trials. Furthermore, we simulate scenarios where the robot is in contact with a rigid obstacle and is cutting through phantom tissue. Results show the robot can reach various static targets with a minimum accuracy of 2 mm.

Comparison of data-driven prediction methods for comprehensive coke ratio of blast furnace

Abstract The emission of blast furnace (BF) exhaust gas has been criticized by society. It is momentous to quickly predict the comprehensive coke ratio (CCR) of BF, because CCR is one of the important indicators for evaluating gas emissions, energy consumption, and production stability, and also affects composite economic benefits. In this article, 13 data-driven prediction techniques, including six conventional and seven ensemble methods, are applied to predict CCR. The result of ten-fold cross-validation indicates that multiple linear regression (MLR) and support vector regression (SVR) based on radial basis function are superior to the other methods. The mean absolute error, the root mean square error, and the coefficient of determination (R 2) of the MLR model are 1.079 kg·t−1, 1.668, and 0.973, respectively. The three indicators of the SVR model are 1.158 kg·t−1, 1.878, and 0.975, respectively. Furthermore, AdaBoost based on linear regression has also strong prediction ability and generalization performance. The three methods have important significances both in theory and in practice for predicting CCR. Moreover, the models constructed here can provide valuable hints into realizing data-driven control of the BF process.

Data-driven control of agent-based models: An Equation/Variable-free machine learning approach

We present an Equation/Variable free machine learning (EVFML) framework for the control of the collective dynamics of complex/multiscale systems modeled via microscopic/agent-based simulators. The approach obviates the need for construction of surrogate, reduced-order models. The proposed implementation consists of three steps: (A) from high-dimensional agent-based simulations, machine learning, in particular, non-linear manifold learning (Diffusion Maps (DMs)) identifies a set of coarse-grained variables that parametrize the low-dimensional manifold on which the emergent/collective dynamics evolve. The out-of-sample extension and pre-image problems, i.e. the construction of non-linear mappings from the high-dimensional input space to the low-dimensional manifold and back, are solved by coupling DMs with the Nyström extension and Geometric Harmonics, respectively; (B) having identified the manifold and its coordinates, we exploit the Equation-free approach to perform numerical bifurcation analysis of the emergent dynamics; then (C) based on the previous steps, we design data-driven embedded wash-out controllers that drive the agent-based simulators to their intrinsic, imprecisely known, emergent open-loop unstable steady-states, thus demonstrating that the scheme is robust against numerical approximation errors and modeling uncertainty. The efficiency of the framework is illustrated by controlling emergent unstable (i) traveling waves of a deterministic agent-based model of traffic dynamics, and (ii) equilibria of a stochastic financial market model with mimesis.

A general infrastructure for data-driven control design and implementation in tokamaks

A general infrastructure for tokamak controllers based on data-driven neural net models is presented. The paradigm allows for more flexible choices of both the underlying model and the desired controlled variables and targets. The system is implemented and tested on the DIII-D tokamak, enacting simultaneous pressure and temperature control via a finite-set model-predictive controller. Traditional control methods such as proportional–integral–derivative (PID) have proven effective for decoupled control tasks, but scale poorly when trying to achieve more complicated goals such as full state control. This is exactly where model-based controllers succeed.

The valve opening control of the motor-operated valve using virtual internal model tuning and vata-driven prediction

This paper deals with applications of data-driven control, which is a controller tuning method that directly utilizes the data, for mechanical systems. Particularly, we focus on the valve opening control of the motor-operated valve as one of the representative mechanical control units which are used in much industrial equipment and are needed to be precisely controlled. If we have a mathematical model of such a system, a model-based controller design is the most rational approach. Otherwise, data-driven controller tuning is also expected to be a rational approach. However, there exists a nonlinear characteristic, such as the limit of a speed implemented in the nonlinear compensator. Since conventional data-driven controller design can not be used because data-driven control is generally derived for only linear time-invariant systems, we have to modify the method of driven controller tuning method so as to be applicable for such systems. This paper proposes a controller tuning method for the valve-opening control system implemented as a cascade control system with a limit of the speed. Here, we apply Virtual Internal Model Tuning, which is abbreviated as VIMT, to tune the inner linear controller. Then, we also apply data-driven prediction to compute the outer loop controller including the limit of the speed. Finally, we verified the usefulness of the proposed method using experimental verification.

Guarantees for data-driven control of nonlinear systems using semidefinite programming: A survey

This survey presents recent research on determining control-theoretic properties and designing controllers with rigorous guarantees using semidefinite programming and for nonlinear systems for which no mathematical models but measured trajectories are available. Data-driven control techniques have been developed to circumvent a time-consuming modelling by first principles and because of the increasing availability of data. Recently, this research field has gained increased attention by the application of Willems’ fundamental lemma, which provides a fertile ground for the development of data-driven control schemes with guarantees for linear time-invariant systems. While the fundamental lemma can be generalized to further system classes, there does not exist a comparable data-based system representation for nonlinear systems. At the same time, nonlinear systems constitute the majority of practical systems. Moreover, they include additional challenges such as data-based surrogate models that prevent system analysis and controller design by convex optimization. Therefore, a variety of data-driven control approaches has been developed with different required prior insights into the system to ensure a guaranteed inference. In this survey, we will discuss developments in the context of data-driven control for nonlinear systems. In particular, we will focus on methods based on system representations providing guarantees from finite data, while the analysis and the controller design boil down to convex optimization problems given as semidefinite programming. Thus, these approaches achieve reasonable advances compared to the state-of-the-art system analysis and controller design by models from system identification. Specifically, the paper covers system representations based on extensions of Willems’ fundamental lemma, set membership, kernel techniques, the Koopman operator, and feedback linearization.

Learning controllers for nonlinear systems from data

This article provides an overview of a new approach to designing controllers for nonlinear systems using data-driven control. Data-driven control is an important area of research in control theory, and this novel method offers several benefits. It can recreate from a data-centred perspective many of the results available in the model-based case, including local stabilization based on Taylor or polynomial expansion, absolute stabilization, as well as approximate and exact feedback linearization. Moreover, the method is analytically and computationally simple, and permits to infer regions of attraction and invariant sets, also when the data are corrupted by noise.

Sistem Arsitektur Manajemen Bangunan untuk Memaksimalkan Swakonsumsi pada Bangunan Universitas: Studi Kasus

Due to its intermittent nature, significant adoption of solar PV into the grid can decrease grid reliability. One solution to increase it is to increase PV self-consumption with two methods: adding Energy Storage System (ESS) and conducting Demand Side Management (DSM). University building has a distinct characteristic in its complex dynamics. Therefore, there is a lack of research to control both methods of increasing self-consumption. This paper aimed to do an integrated literature review on increasing self-consumption and then propose a system architecture recommendation for university building management based on the review. The Smart Grid Architectural Model (SGAM) evaluated the case study object. The result showed that a data-driven controller has been chosen as the most suitable controller for the university building management system. The data needed to build a data-driven controller could be obtained through readily available sensors in the case study object, making it feasible for implementation.

Distributed Data-driven Iterative Learning Control for Consensus Tracking

High performance consensus tracking of networked dynamical systems working repetitively has found applications in a range of areas. Existing iterative learning control (ILC) designs for this problem either require a system model that can be difficult or expensive to obtain in practice, and/or cannot guarantee the monotonic convergence of the tracking error norm. They often have difficulties handling varying networks too. This paper proposes a data-driven norm optimal ILC (DD-NOILC) framework to address these limitations using the recent development in data-driven control, in particular, the so called Willems’ fundamental lemma. The novel design guarantees that even without using any model information, the proposed DD-NOILC framework can achieve the same convergence performance as the model-based NOILC framework, i.e., monotonic convergence of the tracking error norm to zero. Furthermore, using the alternating direction method of multipliers (ADMM), a distributed implementation of the framework is developed such that each subsystem's input can be updated locally, making the proposed distributed DD-NOILC algorithm suitable for large-scale and varying networks. Convergence properties of the proposed algorithms are analysed rigorously, and numerical examples are provided to verify the effectiveness of the distributed DD-NOILC algorithm.

Sistem Arsitektur Manajemen Bangunan untuk Memaksimalkan Swakonsumsi pada Bangunan Universitas: Studi Kasus

Due to its intermittent nature, significant adoption of solar PV into the grid can decrease grid reliability. One solution to increase it is to increase PV self-consumption with two methods: adding Energy Storage System (ESS) and conducting Demand Side Management (DSM). University building has a distinct characteristic in its complex dynamics. Therefore, there is a lack of research to control both methods of increasing self-consumption. This paper aimed to do an integrated literature review on increasing self-consumption and then propose a system architecture recommendation for university building management based on the review. The Smart Grid Architectural Model (SGAM) evaluated the case study object. The result showed that a data-driven controller has been chosen as the most suitable controller for the university building management system. The data needed to build a data-driven controller could be obtained through readily available sensors in the case study object, making it feasible for implementation.

Closed-loop Aspects of Data-Enabled Predictive Control

In recent years, the amount of data available from systems has drastically increased, motivating the use of direct data-driven control techniques that avoid the need of parametric modeling. The aim of this paper is to analyze closed-loop aspects of these approaches in the presence of noise. To analyze this, a unified formulation of several approaches, including Data-enabled Predictive Control (DeePC) and Subspace Predictive Control (SPC) is obtained and the influence of noise on closed-loop predictors is analyzed. The analysis reveals potential closed-loop correlation problems, which are closely related to well-known results in closed-loop system identification, and consequent control issues. A case study reveals the hazards of noise in data-driven control.

Physically Consistent Multiple-Step Data-Driven Predictions Using Physics-Based Filters

Data-driven control can facilitate the rapid development of controllers, offering an alternative to conventional approaches. In order to maintain consistency between any known underlying physical laws and a data-driven decision-making process, preprocessing of raw data is necessary to account for measurement noise and any inconsistencies it may introduce. In this paper, we present a physics-based filter to achieve this and demonstrate its effectiveness through practical applications, using real-world datasets collected in a building on the École Polytechnique Fédérale de Lausanne (EPFL) campus. Two distinct use cases are explored: indoor temperature control and demand response bidding.

A Generative Verification Framework on Statistical Stability for Data-Driven Controllers

This study proposes a novel framework for evaluating the stability of data-driven controllers and the concept of statistical stability. The proposed framework can be used when it is challenging to show stability through conventional control theory. The novelty of this paper lies in that it provides a method for scientifically analyzing the stability of data-driven controllers, thereby improving the quality of data-driven controllers. The proposed framework consists of three parts: the generative model, controller optimizer, and verification model. A variational autoencoder is used to classify and randomly generate data, and the generated data are used to train the controller. A support vector machine is used to classify areas where the controller is statistically stable. The statistical stability of an optimal controller designed using a deep neural network structure is analyzed using the proposed framework.

Data-Driven Output-Feedback Control for Unknown Switched Linear Systems

This paper proposes a data-driven control method to stabilize unknown switched linear systems under arbitrary switching. We consider the case where the system state is not measurable and design an output feedback controller only using measured input-output data. First, the system with multiple outputs is transformed into a single-output system with observability preserved. Then, a data-based state-space representation that has the same input-output relationship as the original system is constructed using the input-output data of the single-output system, based on which the data-driven controller is designed. Sufficient conditions for asymptotic stability of the closed-loop system under arbitrary switching are established in terms of linear matrix inequalities (LMIs). Compared with the existing method, the proposed method decreases the dimension of the constructed data-based state-space representation, which may reduce the computational burden of the controller design. The effectiveness of the proposed controller is illustrated by an example.

Data-driven Norm Optimal Iterative Learning Control for Point-to-Point Tasks

Iterative learning control (ILC) is suitable for high-performance repetitive tasks since it learns from past trials to improve the tracking performance. Existing ILC designs often require a model, which can be difficult or expensive to obtain in practice. To address this problem, we recently developed a data-driven norm optimal ILC using the latest developments from data-driven control, namely, the Willems’ fundamental lemma. In this paper, we show that the idea can also be extended to point-to-point ILC tasks that focus on tracking some intermediate points of the whole trial. We propose a novel data-driven point-to-point norm optimal ILC algorithm that can achieve the same performance as the model-based algorithm but without using an analytical model. The design requires the available data to be persistently exciting of a sufficiently high order. To relax this requirement, a receding horizon based algorithm and a trial partition based algorithm are further developed with well-defined, but different convergence properties. Numerical examples are given to illustrate the proposed algorithms’ effectiveness.