Distributed Optimal Control of Cyber-Physical Systems: Controller Synthesis, Architecture Design and System Identification

The centralized paradigm of a single controller and a single plant upon which modern control theory is built is no longer applicable to modern cyber-physical systems of interest, such as the power-grid, software defined networks or automated highways systems, as these are all large-scale and spatially distributed. Both the scale and the distributed nature of these systems has motivated the decentralization of control schemes into local sub-controllers that measure, exchange and act on locally available subsets of the globally available system information. This decentralization of control logic leads to different decision makers acting on asymmetric information sets, introduces the need for coordination between them, and perhaps not surprisingly makes the resulting optimal control problem much harder to solve. In fact, shortly after such questions were posed, it was realized that seemingly simple decentralized optimal control problems are computationally intractable to solve, with the Wistenhausen counterexample being a famous instance of this phenomenon. Spurred on by this perhaps discouraging result, a concerted 40 year effort to identify tractable classes of distributed optimal control problems culminated in the notion of quadratic invariance, which loosely states that if sub-controllers can exchange information with each other at least as quickly as the effect of their control actions propagates through the plant, then the resulting distributed optimal control problem admits a convex formulation.

The identification of quadratic invariance as an appropriate means of "convexifying" distributed optimal control problems led to a renewed enthusiasm in the controller synthesis community, resulting in a rich set of results over the past decade. The contributions of this thesis can be seen as being a part of this broader family of results, with a particular focus on closing the gap between theory and practice by relaxing or removing assumptions made in the traditional distributed optimal control framework. Our contributions are to the foundational theory of distributed optimal control, and fall under three broad categories, namely controller synthesis, architecture design and system identification.

We begin by providing two novel controller synthesis algorithms. The first is a solution to the distributed H-infinity optimal control problem subject to delay constraints, and provides the only known exact characterization of delay-constrained distributed controllers satisfying an H-infinity norm bound. The second is an explicit dynamic programming solution to a two player LQR state-feedback problem with varying delays. Accommodating varying delays represents an important first step in combining distributed optimal control theory with the area of Networked Control Systems that considers lossy channels in the feedback loop. Our next set of results are concerned with controller architecture design. When designing controllers for large-scale systems, the architectural aspects of the controller such as the placement of actuators, sensors, and the communication links between them can no longer be taken as given -- indeed the task of designing this architecture is now as important as the design of the control laws themselves. To address this task, we formulate the Regularization for Design (RFD) framework, which is a unifying computationally tractable approach, based on the model matching framework and atomic norm regularization, for the simultaneous co-design of a structured optimal controller and the architecture needed to implement it. Our final result is a contribution to distributed system identification. Traditional system identification techniques such as subspace identification are not computationally scalable, and destroy rather than leverage any a priori information about the system's interconnection structure. We argue that in the context of system identification, an essential building block of any scalable algorithm is the ability to estimate local dynamics within a large interconnected system. To that end we propose a promising heuristic for identifying the dynamics of a subsystem that is still connected to a large system. We exploit the fact that the transfer function of the local dynamics is low-order, but full-rank, while the transfer function of the global dynamics is high-order, but low-rank, to formulate this separation task as a nuclear norm minimization problem. Finally, we conclude with a brief discussion of future research directions, with a particular emphasis on how to incorporate the results of this thesis, and those of optimal control theory in general, into a broader theory of dynamics, control and optimization in layered architectures.

On the Role of Delays in Biological Systems : Analysis and Design

This work quantifies the nature of delays in genetic regulatory networks and their effect on system dynamics. It is known that a time lag can emerge from a sequence of biochemical reactions. Applying this modeling framework to the protein production processes, delay distributions are derived in a stochastic (probability density function) and deterministic setting (impulse function), whilst being shown to be equivalent under different assumptions. The dependence of the distribution properties on rate constants, gene length, and time-varying temperatures is investigated. Overall, the distribution of the delay in the context of protein production processes is shown to be highly dependent on the size of the genes and mRNA strands as well as the reaction rates. Results suggest longer genes have delay distributions with a smaller relative variance, and hence, less uncertainty in the completion times, however, they lead to larger delays. On the other hand large uncertainties may actually play a positive role, as broader distributions can lead to larger stability regions when this formalization of the protein production delays is incorporated into a feedback system.

Furthermore, evidence suggests that delays may play a role as an explicit design into existing controlling mechanisms. Accordingly, the reccurring dual-feedback motif is also investigated with delays incorporated into the feedback channels. The dual-delayed feedback is shown to have stabilizing effects through a control theoretic approach. Lastly, a distributed delay based controller design method is proposed as a potential design tool. In a preliminary study, the dual-delayed feedback system re-emerges as an effective controller design.

Photonic and Device Design Principles for Ultrahigh-Efficiency (>50%), Spectrum-Splitting Photovoltaics

The sun has the potential to power the Earth's total energy needs, but electricity from solar power still constitutes an extremely small fraction of our power generation because of its high cost relative to traditional energy sources. Therefore, the cost of solar must be reduced to realize a more sustainable future. This can be achieved by significantly increasing the efficiency of modules that convert solar radiation to electricity. In this thesis, we consider several strategies to improve the device and photonic design of solar modules to achieve record, ultrahigh (> 50%) solar module efficiencies. First, we investigate the potential of a new passivation treatment, trioctylphosphine sulfide, to increase the performance of small GaAs solar cells for cheaper and more durable modules. We show that small cells (mm2), which currently have a significant efficiency decrease (~ 5%) compared to larger cells (cm2) because small cells have a higher fraction of recombination-active surface from the sidewalls, can achieve significantly higher efficiencies with effective passivation of the sidewalls. We experimentally validate the passivation qualities of treatment by trioctylphosphine sulfide (TOP:S) through four independent studies and show that this facile treatment can enable efficient small devices. Then, we discuss our efforts toward the design and prototyping of a spectrum-splitting module that employs optical elements to divide the incident spectrum into different color bands, which allows for higher efficiencies than traditional methods. We present a design, the polyhedral specular reflector, that has the potential for > 50% module efficiencies even with realistic losses from combined optics, cell, and electrical models. Prototyping efforts of one of these designs using glass concentrators yields an optical module whose combined spectrum-splitting and concentration should correspond to a record module efficiency of 42%. Finally, we consider how the manipulation of radiatively emitted photons from subcells in multijunction architectures can be used to achieve even higher efficiencies than previously thought, inspiring both optimization of incident and radiatively emitted photons for future high efficiency designs. In this thesis work, we explore novel device and photonic designs that represent a significant departure from current solar cell manufacturing techniques and ultimately show the potential for much higher solar cell efficiencies.

Mesoscale Optoelectronic Design of Wire-Based Photovoltaic and Photoelectrochemical Devices

The overarching theme of this thesis is mesoscale optical and optoelectronic design of photovoltaic and photoelectrochemical devices. In a photovoltaic device, light absorption and charge carrier transport are coupled together on the mesoscale, and in a photoelectrochemical device, light absorption, charge carrier transport, catalysis, and solution species transport are all coupled together on the mesoscale. The work discussed herein demonstrates that simulation-based mesoscale optical and optoelectronic modeling can lead to detailed understanding of the operation and performance of these complex mesostructured devices, serve as a powerful tool for device optimization, and efficiently guide device design and experimental fabrication efforts. In-depth studies of two mesoscale wire-based device designs illustrate these principles—(i) an optoelectronic study of a tandem Si|WO3 microwire photoelectrochemical device, and (ii) an optical study of III-V nanowire arrays.

The study of the monolithic, tandem, Si|WO3 microwire photoelectrochemical device begins with development and validation of an optoelectronic model with experiment. This study capitalizes on synergy between experiment and simulation to demonstrate the model’s predictive power for extractable device voltage and light-limited current density. The developed model is then used to understand the limiting factors of the device and optimize its optoelectronic performance. The results of this work reveal that high fidelity modeling can facilitate unequivocal identification of limiting phenomena, such as parasitic absorption via excitation of a surface plasmon-polariton mode, and quick design optimization, achieving over a 300% enhancement in optoelectronic performance over a nominal design for this device architecture, which would be time-consuming and challenging to do via experiment.

The work on III-V nanowire arrays also starts as a collaboration of experiment and simulation aimed at gaining understanding of unprecedented, experimentally observed absorption enhancements in sparse arrays of vertically-oriented GaAs nanowires. To explain this resonant absorption in periodic arrays of high index semiconductor nanowires, a unified framework that combines a leaky waveguide theory perspective and that of photonic crystals supporting Bloch modes is developed in the context of silicon, using both analytic theory and electromagnetic simulations. This detailed theoretical understanding is then applied to a simulation-based optimization of light absorption in sparse arrays of GaAs nanowires. Near-unity absorption in sparse, 5% fill fraction arrays is demonstrated via tapering of nanowires and multiple wire radii in a single array. Finally, experimental efforts are presented towards fabrication of the optimized array geometries. A hybrid self-catalyzed and selective area MOCVD growth method is used to establish morphology control of GaP nanowire arrays. Similarly, morphology and pattern control of nanowires is demonstrated with ICP-RIE of InP. Optical characterization of the InP nanowire arrays gives proof of principle that tapering and multiple wire radii can lead to near-unity absorption in sparse arrays of InP nanowires.

Synthetic, kinetic and mechanistic studies in early transition metal systems. I. α- and β-hydrogen elimination reactions in derivatives of permethyltitanocene, -zirconocene, and -hafnocene. II. The reactions of aluminum alkyls with derivatives of permethylniobocene

The thermal decomposition of Cp*Ti(CH_3)_2 (Cp*≡ ƞ^5-C_5Me_5) toluene solution follows cleanly first-order kinetics and produces a single titanium product Cp*(C_5Me_4CH_2)Ti(CH_3) concurrent with the evolution of one equivalent of methane. Labeling studies using Cp*_2Ti- (CD_3)_2 and (Cp*-d_(15))_2Ti(CH_3)_2 show the decomposition to be intramolecular and the methane to be produced by the coupling of a methyl group with a hydrogen from the other TiCH_3 group. Activation parameters, ΔH^‡ and ΔS^‡, and kinetic deuterium isotope effects have been measured. The alternative decomposition pathways of α-hydrogen abstraction and a-hydrogen elimination, both leading to a titanium-methylidene intermediate, are discussed.

The insertion of unactivated acetylenes into the metal-hydride bonds of Cp*_2MH_2 (M = Zr, Hf) proceeds rapidly at low temperature to form monoand/ or bisinsertion products, dependent upon the steric bulk of the acetylene substituents. Cp*_2M(H)(C(Me)=CHMe), Cp*_2M(H)(CH=CHCMe_3), Cp*_2M(H)-(CH=CHPh), Cp*_2M(CH=CHPh)_2, Cp*_2M(CH=CHCH_3)_2 and Cp*_2Zr- (CH=CHCH_2CH_3)_2 have been isolated and characterized. To extend the study of unsaturated-carbon ligands, Cp*_2M(C≡CCH_3)_2 have been prepared by treating Cp*_2MCl_2 with LiC≡CCH_3. The reactivity of many of these complexes with carbon monoxide and dihydrogen is surveyed. The mono(2- butenyl) complexes Cp*_2M(H)(C(Me)=CHMe) rearrange at room temperature, forming the crotyl-hydride species Cp*_2M(H)(ƞ^3-C_4H_7). The bis(propenyl) and bis(l-butenyl) zirconium complexes Cp*_2Zr(CH=CHR)_2 (R = CH_3, CH_2CH_3) also rearrange, forming zirconacyclopentenes. Labeling studies, reaction chemistry, and kinetic measurements, including deuterium isotope effects, demonstrate that the unusual 6-hydrogen elimination from an sp^2-hybridized carbon is the first step in these latter rearrangements but is not observed in the former. Details of these mechanisms and the differences in reactivity of the zirconium and hafnium complexes are discussed.

The reactions of hydride- and alkyl-carbonyl derivatives of permethylniobocene with equimolar amounts of trialkylaluminum reagents occur rapidly producing the carbonyl adducts Cp*_2Nb(R)(COAlR'_3) (R = H, CH_3, CH_2CH_3, CH_2CH_2Ph, C(Me)=CHMe; R' = Me, Et). The hydride adduct Cp*_2NbH_3•AlEt_3 has also been formed. In solution, each of these compounds exists in equilibrium with the uncomplexed species. The formation constants for Cp*_2Nb(H)(COA1R'_R) have been measured. They indicate the steric bulk of the Cp* ligands plays a deciding factor in the isolation of the first example of an aluminum Lewis acid bound to a carbonyl-oxygen in preference to a metalhydride. Reactions of Cp*_2Nb(H)CO with other Lewis acids and of the one:one adducts with H_2, CO and C_2H_4 are also discussed.

Cp*_2Nb(H)(C_2H_4) also reacts with equimolar amounts of trialkylaluminum reagents, forming a one:one complex that ^1H NMR spectroscopy indicates contains a Nb-CH_2CH_2-Al bridge. This adduct also exists in equilibrium with the uncomplexed species in solution. The formation constant for Cp*_2N+/b(H)(CH_2CH_2ĀlEt_3) has been measured. Reactions of Cp*_2Nb(H)(C_2H_4) with other Lewis acids and the reactions of Cp*_2N+b(H)- (CH_2CH_2ĀlEt_3) with CO and C_2H_4 are described, as are the reactions of Cp_*2Nb(H)(CH_2=CHR) (R = Me, Ph), Cp*_2Nb(H)(CH_3C≡CCH_3) and Cp*_2Ti-(C_2H_4) with AlEt_3.

Engineering and Characterization of Cytochrome P450 Enzymes for Nitrogen-Atom Transfer Reactions

The creation of novel enzyme activity is a great challenge to protein engineers, but nature has done so repeatedly throughout the process of natural selection. I begin by outlining the multitude of distinct reactions catalyzed by a single enzyme class, cytochrome P450 monooxygenases. I discuss the ability of cytochrome P450 to generate reactive intermediates capable of diverse reactivity, suggesting this enzyme can also be used to generate novel reactive intermediates in the form of metal-carbenoid and nitrenoid species. I then show that cytochrome P450 from Bacillus megaterium (P450_BM3) and its isolated cofactor can catalyze metal-nitrenoid transfer in the form of intramolecular C–H bond amination. Mutations to the protein sequence can enhance the reactivity and selectivity of this transformation significantly beyond that of the free cofactor. Next, I demonstrate an intermolecular nitrene transfer reaction catalyzed by P450_BM3 in the form of sulfide imidation. Understanding that sulfur heteroatoms are strong nucleophiles, I show that increasing the sulfide nucleophilicity through substituents on the aryl sulfide ring can dramatically increase reaction productivity. To explore engineering nitrenoid transfer in P450BM3, active site mutagenesis is employed to tune the regioselectivity intramolecular C–H amination catalysts. The solution of the crystal structure of a highly selective variant demonstrates that hydrophobic residues in the active site strongly modulate reactivity and regioselectivity. Finally, I use a similar strategy to develop P450-based catalysts for intermolecular olefin aziridination, demonstrating that active site mutagenesis can greatly enhance this nitrene transfer reaction. The resulting variant can catalyze intermolecular aziridination with more than 1000 total turnovers and enantioselectivity of up to 99% ee.

Biological Activity of Rhodium Metalloinsertors and the Design of Bifunctional Conjugates

The Barton laboratory has established that octahedral rhodium complexes bearing the sterically expansive 5,6-chrysene diimine ligand can target thermodynamically destabilized sites, such as base pair mismatches, in DNA with high affinity and selectivity. These complexes approach DNA from the minor groove, ejecting the mismatched base pairs from the duplex in a binding mode termed metalloinsertion. In recent years, we have shown that these metalloinsertor complexes also exhibit cytotoxicity preferentially in cancer cells that are deficient in the mismatch repair (MMR) machinery.

Here, we establish that a sensitive structure-activity relationship exists for rhodium metalloinsertors. We studied the relationship between the chemical structures of metalloinsertors and their effect on biological activity for ten complexes with similar DNA binding affinities, but wide variation in their lipophilicity. Drastic differences were observed in the selectivities of the complexes for MMR-deficient cells. Compounds with hydrophilic ligands were highly selective, exhibiting preferential cytotoxicity in MMR-deficient cells at low concentrations and short incubation periods, whereas complexes with lipophilic ligands displayed poor cell-selectivity. It was discovered that all of the complexes localized to the nucleus in concentrations sufficient for mismatch binding; however, highly lipophilic complexes also exhibited high mitochondrial uptake. Significantly, these results support the notion that mitochondrial DNA is not the desired target for our metalloinsertor complexes; instead, selectivity stems from targeting mismatches in genomic DNA.

We have also explored the potential for metalloinsertors to be developed into more complex structures with multiple functionalities that could either enhance their overall potency or impart mismatch selectivity onto other therapeutic cargo. We have constructed a family of bifunctional metalloinsertor conjugates incorporating cis-platinum, each unique in its chemical structure, DNA binding interactions, and biological activity. The study of these complexes in MMR-deficient cells has established that the cell-selective biological activity of rhodium metalloinsertors proceeds through a critical cellular pathway leading to necrosis.

We further explored the underlying mechanisms surrounding the biological response to mismatch recognition by metalloinsertors in the genome. Immunofluorescence assays of MMR-deficient and MMR-proficient cells revealed that a critical biomarker for DNA damage, phosphorylation of histone H2AX (γH2AX) rapidly accumulates in response to metalloinsertor treatment, signifying the induction of double strand breaks in the genome. Significantly, we have discovered that our metalloinsertor complexes selectively inhibit transcription in MMR-deficient cells, which may be a crucial checkpoint in the eventual breakdown of the cell via necrosis. Additionally, preliminary in vivo studies have revealed the capability of these compounds to traverse the complex environments of multicellular organisms and accumulate in MMR-deficient tumors. Our ever-increasing understanding of metalloinsertors, as well as the development of new generations of complexes both monofunctional and bifunctional, enables their continued progress into the clinic as promising new chemotherapeutic agents.

Distributed optimization in power networks and general multi-agent systems

The dissertation studies the general area of complex networked systems that consist of interconnected and active heterogeneous components and usually operate in uncertain environments and with incomplete information. Problems associated with those systems are typically large-scale and computationally intractable, yet they are also very well-structured and have features that can be exploited by appropriate modeling and computational methods. The goal of this thesis is to develop foundational theories and tools to exploit those structures that can lead to computationally-efficient and distributed solutions, and apply them to improve systems operations and architecture.

Specifically, the thesis focuses on two concrete areas. The first one is to design distributed rules to manage distributed energy resources in the power network. The power network is undergoing a fundamental transformation. The future smart grid, especially on the distribution system, will be a large-scale network of distributed energy resources (DERs), each introducing random and rapid fluctuations in power supply, demand, voltage and frequency. These DERs provide a tremendous opportunity for sustainability, efficiency, and power reliability. However, there are daunting technical challenges in managing these DERs and optimizing their operation. The focus of this dissertation is to develop scalable, distributed, and real-time control and optimization to achieve system-wide efficiency, reliability, and robustness for the future power grid. In particular, we will present how to explore the power network structure to design efficient and distributed market and algorithms for the energy management. We will also show how to connect the algorithms with physical dynamics and existing control mechanisms for real-time control in power networks.

The second focus is to develop distributed optimization rules for general multi-agent engineering systems. A central goal in multiagent systems is to design local control laws for the individual agents to ensure that the emergent global behavior is desirable with respect to the given system level objective. Ideally, a system designer seeks to satisfy this goal while conditioning each agent’s control on the least amount of information possible. Our work focused on achieving this goal using the framework of game theory. In particular, we derived a systematic methodology for designing local agent objective functions that guarantees (i) an equivalence between the resulting game-theoretic equilibria and the system level design objective and (ii) that the resulting game possesses an inherent structure that can be exploited for distributed learning, e.g., potential games. The control design can then be completed by applying any distributed learning algorithm that guarantees convergence to the game-theoretic equilibrium. One main advantage of this game theoretic approach is that it provides a hierarchical decomposition between the decomposition of the systemic objective (game design) and the specific local decision rules (distributed learning algorithms). This decomposition provides the system designer with tremendous flexibility to meet the design objectives and constraints inherent in a broad class of multiagent systems. Furthermore, in many settings the resulting controllers will be inherently robust to a host of uncertainties including asynchronous clock rates, delays in information, and component failures.

An integrated design approach to power systems : from power flows to electricity markets

Power system is at the brink of change. Engineering needs, economic forces and environmental factors are the main drivers of this change. The vision is to build a smart electrical grid and a smarter market mechanism around it to fulfill mandates on clean energy. Looking at engineering and economic issues in isolation is no longer an option today; it needs an integrated design approach. In this thesis, I shall revisit some of the classical questions on the engineering operation of power systems that deals with the nonconvexity of power flow equations. Then I shall explore some issues of the interaction of these power flow equations on the electricity markets to address the fundamental issue of market power in a deregulated market environment. Finally, motivated by the emergence of new storage technologies, I present an interesting result on the investment decision problem of placing storage over a power network. The goal of this study is to demonstrate that modern optimization and game theory can provide unique insights into this complex system. Some of the ideas carry over to applications beyond power systems.

Distributed Control and Optimization for Communication and Power Systems

We are at the cusp of a historic transformation of both communication system and electricity system. This creates challenges as well as opportunities for the study of networked systems. Problems of these systems typically involve a huge number of end points that require intelligent coordination in a distributed manner. In this thesis, we develop models, theories, and scalable distributed optimization and control algorithms to overcome these challenges.

This thesis focuses on two specific areas: multi-path TCP (Transmission Control Protocol) and electricity distribution system operation and control. Multi-path TCP (MP-TCP) is a TCP extension that allows a single data stream to be split across multiple paths. MP-TCP has the potential to greatly improve reliability as well as efficiency of communication devices. We propose a fluid model for a large class of MP-TCP algorithms and identify design criteria that guarantee the existence, uniqueness, and stability of system equilibrium. We clarify how algorithm parameters impact TCP-friendliness, responsiveness, and window oscillation and demonstrate an inevitable tradeoff among these properties. We discuss the implications of these properties on the behavior of existing algorithms and motivate a new algorithm Balia (balanced linked adaptation) which generalizes existing algorithms and strikes a good balance among TCP-friendliness, responsiveness, and window oscillation. We have implemented Balia in the Linux kernel. We use our prototype to compare the new proposed algorithm Balia with existing MP-TCP algorithms.

Our second focus is on designing computationally efficient algorithms for electricity distribution system operation and control. First, we develop efficient algorithms for feeder reconfiguration in distribution networks. The feeder reconfiguration problem chooses the on/off status of the switches in a distribution network in order to minimize a certain cost such as power loss. It is a mixed integer nonlinear program and hence hard to solve. We propose a heuristic algorithm that is based on the recently developed convex relaxation of the optimal power flow problem. The algorithm is efficient and can successfully computes an optimal configuration on all networks that we have tested. Moreover we prove that the algorithm solves the feeder reconfiguration problem optimally under certain conditions. We also propose a more efficient algorithm and it incurs a loss in optimality of less than 3% on the test networks.

Second, we develop efficient distributed algorithms that solve the optimal power flow (OPF) problem on distribution networks. The OPF problem determines a network operating point that minimizes a certain objective such as generation cost or power loss. Traditionally OPF is solved in a centralized manner. With increasing penetration of volatile renewable energy resources in distribution systems, we need faster and distributed solutions for real-time feedback control. This is difficult because power flow equations are nonlinear and kirchhoff's law is global. We propose solutions for both balanced and unbalanced radial distribution networks. They exploit recent results that suggest solving for a globally optimal solution of OPF over a radial network through a second-order cone program (SOCP) or semi-definite program (SDP) relaxation. Our distributed algorithms are based on the alternating direction method of multiplier (ADMM), but unlike standard ADMM-based distributed OPF algorithms that require solving optimization subproblems using iterative methods, the proposed solutions exploit the problem structure that greatly reduce the computation time. Specifically, for balanced networks, our decomposition allows us to derive closed form solutions for these subproblems and it speeds up the convergence by 1000x times in simulations. For unbalanced networks, the subproblems reduce to either closed form solutions or eigenvalue problems whose size remains constant as the network scales up and computation time is reduced by 100x compared with iterative methods.