Unveiling overlooked aspects of model predictive control for building air conditioning systems
Not Available
Engineering topics
Publications and source records attributed to Bhattacharya, Arnab.
Not Available
Critical infrastructure networks, such as power and transportation networks, can be modelled as cyber-physical systems. As the complexity of such systems grow, there is need for developing metrics and design tools that will co-optimize the physical components of the system and the control policies to guarantee resilience against cyber and natural threats. To that end, we develop a simulation-based control co-design computational framework that will concurrently determine the system and control parameters of a cyber-physical system to meet pre-specified resilience, operational and economic objectives. Here, the capabilities of the developed co-design engine is demonstrated by designing the physical components and control parameters of a microgrid system that will meet its resiliency objectives when subjected to various cyber and physical threats.
Security of cyber-physical systems (CPS) continues to pose new challenges due to the tight integration and operational complexity of the cyber and physical components. To address these challenges, this article presents a domain-aware, optimization-based approach to determine an effective defense strategy for CPS in an automated fashion—by emulating a strategic adversary in the loop that exploits system vulnerabilities, interconnection of the CPS, and the dynamics of the physical components. Our approach builds on an adversarial decision-making model based on a Markov Decision Process (MDP) that determines the optimal cyber (discrete) and physical (continuous) attack actions over a CPS attack graph. The defense planning problem is modeled as a non-zero-sum game between the adversary and defender. We use a model-free reinforcement learning method to solve the adversary’s problem as a function of the defense strategy. We then employ Bayesian optimization (BO) to find an approximate best-response for the defender to harden the network against the resulting adversary policy. This process is iterated multiple times to improve the strategy for both players. We demonstrate the effectiveness of our approach on a ransomware-inspired graph with a smart building system as the physical process. Numerical studies show that our method converges to a Nash equilibrium for various defender-specific costs of network hardening.
Collaboration and Negotiation is a critical high-level function of an Autonomous Intelligent Cyber-Defense Agent (AICA) that enables communication among agents, central cyber C2, and human operators. Maintaining the Confidentiality, Integrity, and Availability (CIA) triad while achieving mission goals requires stealthy AICA agents to exercise: 1) minimal communication as needed for avoiding detection, 2) verification of information received with possibly limited resources, and 3) active learning during operations to address dynamic conditions. Moreover, negotiations to jointly identify and execute a Course of Action (COA) solution will require building consensus under distributed and/or decentralized multiagent settings with information uncertainties. This chapter presents algorithmic approaches for enabling the collaboration and negotiation function. Strengths and limitations of potential techniques are identified, and a representative example is illustrated. Recommendations for future development are also discussed.
Cyber-Physical Systems (CPSs) like the power grid are critically important but also increasingly vulnerable; ensuring reliable system operation in the face of disruptions is becoming more and more challenging. Multi-Level Optimization (MLO) is a powerful way to model adversarial interactions, which naturally makes it applicable to studying CPS security. However, MLO typically does not address underlying system dynamics, and incorporating nonlinear dynamics is generally infeasible. In this paper, we show how to combine MLO with the Koopman Operator (KO) to remedy this. The KO maps nonlinear dynamics to a lifted space in which those dynamics are linear, thus making it ideal for use with MLO. Moreover, the structure of the KO also provides convenient ways to incorporate domain knowledge into the data-driven process of learning the KO representation of a given system. Here we then demonstrate the use of MLO-KO on a small example problem taken from the power grid domain, discuss the scalability and computational cost of MLO-KO, and identify future research directions for this work.
Critical infrastructures such as cyber-physical energy systems (CPS-E) integrate information flow and physical operations that are vulnerable to natural and targeted failures. Safe, secure, and reliable operation and control of CPS-E is critical to ensure societal well-being and economic prosperity. Automated control is key for real-time operations and may be mathematically cast as a sequential decision-making problem under uncertainty. Emergence of data-driven techniques for decision making under uncertainty, such as reinforcement learning (RL), have led to promising advances for addressing sequential decision-making problems for risk-based robust CPS-E control. However, existing research challenges include understanding the applicability of RL methods across diverse CPS-E applications, addressing the effect of risk preferences across multiple RL methods, and development of open-source domain-aware simulation environments for RL experimentation within a CPS-E context. This article systematically analyzes the applicability of four types of RL methods (model-free, model-based, hybrid model-free and model-based, and hierarchical) for risk-based robust CPS-E control. Problem features and solution stability for the RL methods are also discussed. We demonstrate and compare the performance of multiple RL methods under different risk specifications (risk-averse, risk-neutral, and risk-seeking) through the development and application of an open-source simulation environment. Motivating numerical simulation examples include representative single-zone and multizone building control use cases. Finally, six key insights for future research and broader adoption of RL methods are identified, with specific emphasis on problem features, algorithmic explainability, and solution stability.
Cyber-Physical Systems (CPSs) have a large input space, with discrete and continuous elements across multiple layers. Hybrid Attack Graph (HAG) provide a flexible and efficient approach to generate attack sequences for a CPS. Analysis and testing of large-scale HAGs are prohibitively costly. We propose a dimension reduction via property-preserving multi-layer graph sampling algorithms. Existing property-preserving graph sampling approaches generate a representative subgraph of an original large-sized graph while preserving the key properties, such as node and edge distribution, clustering coefficients, and betweenness. On the other hand, we propose impact-driven sampling strategies to transform the input data to a lower-dimensional representation while retaining key properties of the data.
The dynamics-aware economic dispatch (DED) problem embeds low-level generator dynamics and operational constraints to enable near real-time scheduling of generation units in a power network. DED produces a more dynamic supervisory control policy than traditional economic dispatch (T-ED) that reduces overall generation costs. However, the incorporation of differential equations that govern the system dynamics makes DED an optimization problem that is computationally prohibitive to solve. In this work, we present a new data-driven approach based on differentiable programming to efficiently obtain offline parametric solutions to the underlying DED problem. In particular, we employ the recently proposed differentiable predictive control (DPC) for offline learning of explicit neural control policies based on identified Koopman operator (KO) model of the system dynamics. We demonstrate the high solution quality and five orders of magnitude computational-time savings of the DPC method over the original optimization-based DED approach on a 9-bus test power grid network.
Security assessment of cyber-physical energy systems (CPESs) such as the electric power grid is a critical operation to maintain availability, reliability, and quality of service in the presence of persistent threats from malicious cyber actors. Existing security assessment approaches such as penetration testing and red teaming rely on subject matter expert experience and forensic cyber analysis of historical events to perform realistic, threat-informed assessments of CPES defense. CPESs have a large attack surface because of the heterogeneity and complexity of underlying topology, devices, measurements, and vulnerabilities. The aforementioned approaches lead to partial coverage of the attack surface with a large set of unknown but possible exploits. There is a need to automate the CPES attack surface discovery and contextualize it for relevant, highly probable, real-world attack scenarios. We propose a methodology and framework to facilitate the discovery of the CPES attack surface. We present a multilayer attack graph with ranked attack sequences to describe CPES failure scenarios. We present a work-in-progress framework that lists key components to automate the attack modeling and sequence generation. We demonstrate the published National Electric Sector Cybersecurity Organization Resource CPES failure scenario to highlight the trustworthiness of generated attack sequences.
Generalized Disjunctive Programming (GDP) provides an alternative framework to model optimization problems with both discrete and continuous variables. The key idea behind GDP involves the use of logical disjunctions to represent discrete decisions in the continuous space, and logical propositions to denote algebraic constraints in the discrete space. Compared to traditional mixed-integer programming (MIP), the inherent logic structure in GDP yields tighter relaxations that are exploited by global branch and bound algorithms to improve solution quality. In this paper, we present a general GDP model for optimal control of hybrid systems that exhibit both discrete and continuous dynamics. Specifically, we use GDP to formulate a model predictive control (MPC) model for piecewise-affine systems with implicit switching logic. As an example, the GDP-based MPC approach is used as a supervisory control to improve energy efficiency in residential buildings with binary on/off, relay-based thermostats. A simulation study is used to demonstrate the validity of the proposed approach, and the improved solution quality compared to existing MIPbased control approaches.
Not Available
The dynamics-aware economic dispatch (DED) problem embeds low-level generator dynamics and operational constraints to enable near real-time scheduling of generation units in a power network. DED produces a more dynamic supervisory control policy than traditional economic dispatch (T-ED) that reduces overall generation costs. However, in contrast to T-ED, DED is a nonlinear, non-convex optimization problem that is computationally prohibitive to solve. We introduce a machine learning-based operator-theoretic approach for solving the DED problem efficiently. Specifically, we develop a novel discrete-time Koopman Operator (KO) formulation that embeds domain information into the structure of the KO to learn high-fidelity approximations of the generator dynamics. Using the KO approximation, the DED problem can be reformulated as a computationally tractable linear program (abbreviated DED-KO). We demonstrate the high solution quality and computational-time savings of the DED-KO model over the original DED formulation on a 9-bus test system.
Cyber-physical systems (CPSs) consist of computing and communication devices integrated with physical components such as sensors and actuators. Increasing connectivity to the Internet for remote monitoring and control has made CPSs more vulnerable to deliberate attacks, which are distinctly different from random perturbations in the system. This provides a way for purely cyber attacks to have physical consequences. Stuxnet is a prominent example of such an attack, one in which the malware acted over an extended period of time while deliberately remaining undetected. Such attacks can be described as Advanced Persistent Threats (APTs) -- long-term, stealthy attacks. Here, we extend our previous work on hypergames to develop defender strategies that are robust to deception and do not rely on attack detection. We prove that the defender can bound the attacker payoff with these strategies even when the attacker can choose between different attack modes, and we numerically demonstrate our approach on a realistic building control system. Finally, we discuss next steps in extending this approach towards an operational capability.
For effective integration of building operations into the evolving demand response programs of the power grid, real-time decisions concerning the use of building appliances for grid services must excel on multiple criteria, ranging from the added value to occupants' comfort to the quality of the grid services. In this paper, we present a data-driven stochastic decision-support framework to dynamically rank load control alternatives in a commercial building, addressing the needs of multiple decision criteria (e.g. occupant comfort, grid service quality) under uncertainties in occupancy patterns. We adopt a stochastic multi-criteria decision algorithm recently applied to prioritize residential on/off loads, and extend it to i) consider complex load control decisions (e.g. dimming of lights, changing zone temperature set-points) in a commercial building; and ii) systematically integrate zonal occupancy patterns to better identify short-term (and time-varying) opportunities for grid service participation. We evaluate the performance of the proposed framework for curtailment of air-conditioning, lighting, and plug-loads in a multi-zone commercial office building for a range of design choices. With the help of a prototype system that integrates an interactive \textit{Data Analytics and Visualization} frontend we demonstrate a way for the building operators to monitor and change in real-time the available flexibility in energy consumption and to develop trust in the decision recommendations by interpreting the rationale behind the ranking.
In this paper discusses the use and integration of disparate but complementary tools for dealing with solvability, operability, and security challenges in Cyber-Physical Systems (CPSs): the Koopman Operator (KO) for solvability, Disjunctive Programming (DP) for operability, and Multi-Level Optimisation (MLO) for security. These methods can obviate the need for some of the traditional assumptions used in modelling CPSs. This paper demonstrates the methods' capabilities and considers ways to advance each method individually. It concludes by discussing how to integrate the different methods and identify useful synergies generated by doing so.
Increasing deployment of advanced sensing, controls, and communication infrastructure enables buildings to provide services to the power grid, leading to the concept of grid-interactive efficient buildings. Since occupant activities and preferences primarily drive the availability and operational flexibility of building devices, there is a critical need to develop occupant-centric approaches that prioritize devices for providing grid services, while maintaining the desired end-use quality of service. In this paper, we present a decision-making framework that facilitates a building owner/operator to effectively prioritize loads for curtailment service under uncertainties, while minimizing any adverse impact on the occupants. The proposed framework uses a stochastic (Markov) model to represent the probabilistic behavior of device usage from power consumption data, and a load prioritization algorithm that dynamically ranks building loads using a stochastic multi-criteria decision-making algorithm. The proposed load prioritization framework is illustrated via numerical simulations in a residential building use-case, including plug-loads, air-conditioners, and plug-in electric vehicle chargers, in the context of load curtailment as a grid service. Suitable metrics are proposed to evaluate the closed-loop performance of the proposed prioritization algorithm under various scenarios and design choices. Scalability of the proposed algorithm is established via computational analysis, while time-series plots are used for intuitive explanation of the ranking choices.
Adversary emulation is an offensive exercise that provides a comprehensive assessment of a system’s resilience against cyber attacks. However, adversary emulation is typically a manual process, making it costly and hard to deploy in cyber-physical systems (CPS) with complex dynamics, vulnerabilities, and operational uncertainties. In this paper, we develop an automated, domain-aware approach to adversary emulation for CPS. We formulate a Markov Decision Process (MDP) model to determine an optimal attack sequence over a hybrid attack graph with cyber (discrete) and physical (continuous) components and related physical dynamics. We apply model-based and model-free reinforcement learning (RL) methods to solve the discrete-continuous MDP in a tractable fashion. As a baseline, we also develop a greedy attack algorithm and compare it with the RL procedures. We summarize our findings through a numerical study on sensor deception attacks in buildings to compare the performance and solution quality of the proposed algorithms.
In this paper, we provide an introduction to the Koopman Operator (KO) designed to be accessible to those not already familiar with the field. Our aim is to expose domain experts and controls practitioners to the concept and the capabilities the KO provides in the interest of promoting wider KO use; as such, we deliberately focus more on the uses and computational aspects of the KO than on the wealth of analytical results in the literature. Those researchers may then be able to pose new applications-based research questions for the KO community. We begin by defining the KO and describing its key components. Secondly, we discuss computational methods used to calculate the KO in practice. Thirdly, we provide an overview of different uses for the KO and its associated capabilities. Finally, we discuss some open areas of research for future KO work.