Input-output Behavior Research Articles

Quantifying the Autonomy of Structurally Diverse Automata: A Comparison of Candidate Measures.

Should the internal structure of a system matter when it comes to autonomy? While there is still no consensus on a rigorous, quantifiable definition of autonomy, multiple candidate measures and related quantities have been proposed across various disciplines, including graph-theory, information-theory, and complex system science. Here, I review and compare a range of measures related to autonomy and intelligent behavior. To that end, I analyzed the structural, information-theoretical, causal, and dynamical properties of simple artificial agents evolved to solve a spatial navigation task, with or without a need for associative memory. By contrast to standard artificial neural networks with fixed architectures and node functions, here, independent evolution simulations produced successful agents with diverse neural architectures and functions. This makes it possible to distinguish quantities that characterize task demands and input-output behavior, from those that capture intrinsic differences between substrates, which may help to determine more stringent requisites for autonomous behavior and the means to measure it.

Neural network surgery: Combining training with topology optimization

With ever increasing computational capacities, neural networks become more and more proficient at solving complex tasks. However, picking a sufficiently good network topology usually relies on expert human knowledge. Neural architecture search aims to reduce the extent of expertise that is needed. Modern architecture search techniques often rely on immense computational power, or apply trained meta-controllers for decision making. We develop a framework for a genetic algorithm that is both computationally cheap and makes decisions based on mathematical criteria rather than trained parameters. It is a hybrid approach that fuses training and topology optimization together into one process. Structural modifications that are performed include adding or removing layers of neurons, with some re-training applied to make up for any incurred change in input–output behaviour. Our ansatz is tested on several benchmark datasets with limited computational overhead compared to training only the baseline. This algorithm can achieve a significant increase in accuracy (as compared to a fully trained baseline), rescue insufficient topologies that in their current state are only able to learn to a limited extent, and dynamically reduce network size without loss in achieved accuracy. On standard ML datasets, accuracy improvements compared to baseline performance can range from 20% for well performing starting topologies to more than 40% in case of insufficient baselines, or reduce network size by almost 15%.

Physics-Aware Deep-Learning-Based Proxy Reservoir Simulation Model Equipped With State and Well Output Prediction

Data-driven methods have been revolutionizing the way physicists and engineers handle complex and challenging problems even when the physics is not fully understood. However, these models very often lack interpretability. Physics-aware machine learning (ML) techniques have been used to endow proxy models with features closely related to the ones encountered in nature; examples span from material balance to conservation laws. In this study, we proposed a hybrid-based approach that incorporates physical constraints (physics-based) and yet is driven by input/output data (data-driven), leading to fast, reliable, and interpretable reservoir simulation models. To this end, we built on a recently developed deep learning–based reduced-order modeling framework by adding a new step related to information on the input–output behavior (e.g., well rates) of the reservoir and not only the states (e.g., pressure and saturation) matching. A deep-neural network (DNN) architecture is used to predict the state variables evolution after training an autoencoder coupled with a control system approach (Embed to Control—E2C) along with the addition of some physical components (loss functions) to the neural network training procedure. Here, we extend this idea by adding the simulation model output, for example, well bottom-hole pressure and well flow rates, as data to be used in the training procedure. Additionally, we introduce a new architecture to the E2C transition model by adding a new neural network component to handle the connections between state variables and model outputs. By doing this, it is possible to estimate the evolution in time of both the state and output variables simultaneously. Such a non-intrusive data-driven method does not need to have access to the reservoir simulation internal structure, so it can be easily applied to commercial reservoir simulators. The proposed method is applied to an oil–water model with heterogeneous permeability, including four injectors and five producer wells. We used 300 sampled well control sets to train the autoencoder and another set to validate the obtained autoencoder parameters. We show our proxy’s accuracy and robustness by running two different neural network architectures (propositions 2 and 3), and we compare our results with the original E2C framework developed for reservoir simulation.

Analysis of smart contracts balances

We define a technique for analyzing updates of smart contracts balances due to transfers of digital assets. The analysis addresses a lightweight smart contract language and consists of a two-step translation. First, we define the input-output behaviors of smart contract functions by means of a simple functional language with static dispatch. Then we associate the terms of this intermediate language with cost equations that compute the loss or gain of digital assets. The resulting equations can be fed to an off-the-shelf cost analyzer to provide upper bounds to the loss or gain. Our analysis has been prototyped and we report its assessments and discuss extensions with additional features.

Novel Control-Oriented Models for Liquid Transport in Falling Film Evaporator Tubes

The paper presents novel control-oriented transport models for evaporating liquid films in the tubes of a falling film evaporator. In this context, our goal consists in qualitatively mapping the experimentally observed input-output behavior. Two transport models are proposed, where the difference between them is that one allows overtaking of liquid particles and the other does not. The transport models are equipped with two new models of evaporation which are different from the commonly assumed uniform evaporation. The models are initially developed from the conservation laws in the form of partial differential equations. Using the method of characteristics we then obtain the input-output relations for the proposed models in the form of time-delay equations. The time-delay representation is advantageous for simulation and for the future control design. In a simulation study, we observe the principal properties of the models and find that they correspond well with the experimentally observed input-output behavior.

Retroactivity induced operating regime transition in an enzymatic futile cycle.

Activated phosphorylation-dephosphorylation biochemical reaction cycles are a class of enzymatic futile cycles. A futile cycle such as a single MAPK cascade governed by two underlying enzymatic reactions permits Hyperbolic (H), Signal transducing (ST), Threshold-hyperbolic (TH) and Ultrasensitive (U) operating regimes that characterize input-output behaviour. Retroactive signalling caused by load due to sequestration of phosphorylated or unphosphorylated form of the substrate in a single enzymatic cascade without explicit feedback can introduce two-way communication, a feature not possible otherwise. We systematically characterize the operating regimes of a futile cycle subject to retroactivity in either of the substrate forms. We demonstrate that increasing retroactivity strength, which quantifies the downstream load, can trigger five possible regime transitions. Retroactivity strength is a reflection of the fraction of the substrate sequestered by its downstream target. Remarkably, the minimum required retroactivity strength to evidence any sequestration triggered regime transition demands 23% of the substrate bound to its downstream target. This minimum retroactivity strength corresponds to the transition of the dose-response curve from ST to H regime. We show that modulation of the saturation and unsaturation levels of the enzymatic reactions by retroactivity is the fundamental mechanism governing operating regime transition.

Capacitance pressure sensor with S-type electrode for improved sensitivity

This paper aims at designing a differential pressure sensor. The objective of the work is to design and fabricate the electrodes of a capacitive pressure sensor, so as to measure absolute and differential pressure accurately with improved sensitivity. In place of conventional parallel plate diaphragm, S-type electrodes are proposed in the present work. The work comprises of study of the proposed design in terms of a mathematical model, input-output behavior along with detailed analysis of pressure distribution pattern. Output capacitance obtained for changes in pressure is converted to voltage with the suitable signal conditioning circuit and data acquisition system to acquire the signal on to a PC. A neural network model is designed to compensate the nonlinearities present in the sensor output. Input-output characteristics of the designed sensor shows an improved response as compared with existing pressure sensors.

Linear‐like properties arise naturally in the adaptive control setting

SummaryClassical discrete‐time adaptive controllers typically provide asymptotic stabilization and tracking; usually the affect of the noise is at best bounded‐input bounded‐output. Recently we have shown that if you design a discrete‐time adaptive controller in just the right way, then in a variety of situations you not only obtain exponential stability, but also a bounded gain on the noise in every p−norm, as well as a never‐before‐seen linear‐like convolution bound on the input–output behavior. Quite surprisingly, the approach is very natural, and relies on the use of the unmodified, original projection algorithm to carry out parameter estimation; if the set of plant uncertainty is not convex, then a multi‐estimator and switching are used. The goal of this paper is to provide an overview of the approach, discuss the results‐to‐date, and list some of the open problems.

Modeling cell-specific dynamics and regulation of the common gamma chain cytokines.

SUMMARYThe γ-chain receptor dimerizes with complexes of the cytokines interleukin-2 (IL-2), IL-4, IL-7, IL-9, IL-15, and IL-21 and their corresponding “private” receptors. These cytokines have existing uses and future potential as immune therapies because of their ability to regulate the abundance and function of specific immune cell populations. Here, we build a binding reaction model for the ligand-receptor interactions of common γ-chain cytokines, which includes receptor trafficking dynamics, enabling quantitative predictions of cell-type-specific response to natural and engineered cytokines. We then show that tensor factorization is a powerful tool to visualize changes in the input-output behavior of the family across time, cell types, ligands, and concentrations. These results present a more accurate model of ligand response validated across a panel of immune cell types as well as a general approach for generating interpretable guidelines for manipulation of cell-type-specific targeting of engineered ligands.

Neural Identification for Control

We present a new method for learning control law that stabilizes an unknown nonlinear dynamical system at an equilibrium point. We formulate a system identification task in a self-supervised learning setting that jointly learns a controller and corresponding stable closed-loop dynamics hypothesis. The input-output behavior of the unknown dynamical system under random control inputs is used as the supervising signal to train the neural network-based system model and the controller. The proposed method relies on the Lyapunov stability theory to generate a stable closed-loop dynamics hypothesis and corresponding control law. We demonstrate our method on various nonlinear control problems such as n-link pendulum balancing and trajectory tracking, pendulum on cart balancing, and wheeled vehicle path following.

Finite horizon robust synthesis using integral quadratic constraints

AbstractWe present a robust synthesis algorithm for uncertain linear time‐varying (LTV) systems on finite horizons. The uncertain system is described as an interconnection of a known LTV system and a perturbation. The input–output behavior of the perturbation is specified by time‐domain Integral Quadratic Constraints (IQCs). The objective is to synthesize a controller to minimize the worst‐case performance. This leads to a nonconvex optimization. The proposed approach alternates between an LTV synthesis step and an IQC analysis step. Both inducedand terminal Euclidean norm penalties on output are considered for finite horizon performance. The proposed algorithm ensures that the robust performance is nonincreasing at each iteration step. The effectiveness of this method is demonstrated using numerical examples.

Variadic equational matching in associative and commutative theories

In this paper we study matching in equational theories that specify counterparts of associativity and commutativity for variadic function symbols. We design a procedure to solve a system of matching equations and prove its termination, soundness, completeness, and minimality. The minimal complete set of matchers for such a system can be infinite, but our algorithm computes its finite representation in the form of solved set. From the practical side, we identify two finitary cases and impose restrictions on the procedure to get an incomplete algorithm, which, based on our experiments, describes the input-output behavior and properties of Mathematica's flat and orderless pattern matching.

State and parameter estimation for model-based retinal laser treatment

Abstract We present an approach for state and parameter estimation in retinal laser treatment by a novel setup where both measurement and heating is performed by a single laser. In this medical application, the temperature that is induced by the laser in the patient’s eye is critical for a successful and safe treatment. To this end, we pursue a model-based approach using a model given by a heat diffusion equation on a cylindrical domain, where the source term is given by the absorbed laser power. The model is parametric in the sense that it involves an absorption coefficient, which depends on the treatment spot and plays a central role in the input-output behavior of the system. After discretization, we apply a particularly suited parametric model order reduction to ensure real-time tractability while retaining parameter dependence. We augment known state estimation techniques, i.e., extended Kalman filtering and moving horizon estimation, with parameter estimation to estimate the absorption coefficient and the current state of the system. Eventually, we show first results for simulated and experimental data from porcine eyes. We find that, regarding convergence speed, the moving horizon estimation slightly outperforms the extended Kalman filter on measurement data in terms of parameter and state estimation, however, on simulated data the results are very similar.

Surface-Based Path Following Control: Application of Curved Tapes on 3-D Objects

In this article, a novel approach for the versatile wrinkle-free application of (curved) precut adhesive tapes on freeform 3-D surfaces is presented. Straight and curved tape application paths are mapped onto the 3-D object as geodesics and as lines with imposed geodesic curvature, respectively. The proposed surface-based path following control concept extends the classical path following control by a novel parallel contact frame and a parallel projection operator. Using a static state feedback, the robotic system is transformed into a system with linear input-output behavior in the path coordinates. This allows to traverse a path on a 3-D object with a draping roll without turning around the surface normal vector. The latter prevents distortions and wrinkles of the applied tape. Experimental results with a Kuka LBR iiwa 14 R820 demonstrate the feasibility of the proposed approach.

Static and Dynamic Modeling of the Electromagnets of the Maglev Vehicle Transrapid

This work provides a new approach for modeling the electromagnets of the magnetic levitation (Maglev) vehicle Transrapid intending to effectively and efficiently represent the statics and the dynamics in a frequency range relevant for Maglev control design and for the use in mechatronic simulation models. It includes the effects of magnetic reluctances, fringing and leakage flux, magnetic saturation, and eddy currents. The modeling follows a systematic approach by setting up the equivalent magnetic and electric circuits, and coupling the equations using Ampère's circuital law and Faraday's law of induction resulting in a system of differential algebraic equations. It allows to study effects occurring in the magnet design process or different operating points, resulting, e.g., from aerodynamic uplift, for which an extended validity range is needed. Since the computational effort of such models is significant, additionally, a numerical procedure for model reduction is presented yielding a simplified version, which possesses a nearly identical input-output behavior and is usable for control design or in large vehicle models. The proposed approach can be applied to any levitation or guidance magnet of the vehicle and is shown exemplarily for a Transrapid's levitation magnet. The model is validated for the parameters of the latest Transrapid vehicle, called TR09, showing good correspondence with the measured static force-current-gap characteristics and the inductance.

Variable Cut-Off Frequency Observer-Based Positioning for Ball-Beam Systems Without Velocity and Current Feedback Considering Actuator Dynamics

This paper develops an observer-based positioning scheme for ball-beam systems considering actuator dynamics. The practical constraints are handled systematically, including the mechanical dynamical nonlinearities, mismatched load disturbances, and parameter uncertainties. This result provides contributions as follows. First, parameter-independent observers exponentially estimate the ball velocity, motor speed, and its acceleration to remove the velocity, motor speed, and current feedback. Second, the auto-tuner automatically adjusts the desired closed-loop input-output behaviors to update its cut-off frequency in the transient operations. Third, observer-based active damping injection reduces the closed-loop ball position and actuator speed dynamics to 1 by pole-zero cancellation. Finally, disturbance observers act as a dynamic compensator by estimating the disturbances from model-plant mismatches such as dynamic nonlinearities, mismatched load disturbances, and parameter variations. The experimental study verifies the applicability of the proposed technique using the Quanser Ball-Beam hardware driven by an SRV02 servomotor.

An Efficient Deep Learning Scheme To Predict the Electronic Structure of Materials and Molecules: The Example of Graphene-Derived Allotropes.

Computations based on density functional theory (DFT) are transforming various aspects of materials research and discovery. However, the effort required to solve the central equation of DFT, namely the Kohn-Sham equation, which remains a major obstacle for studying large systems with hundreds of atoms in a practical amount of time with routine computational resources. Here, we propose a deep learning architecture that systematically learns the input-output behavior of the Kohn-Sham equation and predicts the electronic density of states, a primary output of DFT calculations, with unprecedented speed and chemical accuracy. The algorithm also adapts and progressively improves in predictive power and versatility as it is exposed to new diverse atomic configurations. We demonstrate this capability for a diverse set of carbon allotropes spanning a large configurational and phase space. The electronic density of states, along with the electronic charge density, may be used downstream to predict a variety of materials properties, bypassing the Kohn-Sham equation, leading to an ultrafast and high-fidelity DFT emulator.

Parallel Gradient-Based EM Optimization for Microwave Components Using Adjoint- Sensitivity-Based Neuro-Transfer Function Surrogate

This article proposes a novel parallel gradient-based electromagnetic (EM) optimization approach to microwave components using adjoint-sensitivity-based neuro-transfer function (neuro-TF) surrogate. In the proposed technique, the surrogate model is trained using not only the input-output behavior but also the adjoint sensitivity information generated from the EM simulation simultaneously. By exploiting adjoint EM sensitivity for surrogate modeling, the proposed technique can obtain accurate surrogate models with larger valid range using the same amount of fine model evaluations compared with the existing gradient-based surrogate optimization without adjoint sensitivity. Furthermore, because the surrogate model is developed using adjoint EM sensitivity, the gradients calculated using the developed surrogate model in the proposed technique are much more accurate. The accurate gradients lead to further speedup of the surrogate optimization and improved quality of surrogate optimal solution in each surrogate optimization iteration. Since the surrogate model is valid in a large neighborhood and the gradients are sufficiently accurate, the proposed technique can achieve the optimal EM solution faster than the existing gradient-based surrogate optimization without adjoint sensitivity. Three examples of EM optimizations of microwave components are used to demonstrate the proposed technique.

Feedback control of boundary layer Tollmien-Schlichting waves using a simple model-based controller

The attenuation of spatially evolving instability Tollmien-Schlichting (T-S) waves in the boundary layer of a flat plate with zero pressure gradients using an active feedback control scheme is theoretically and numerically investigated. The boundary layer is excited artificially by various perturbations to create a three-dimensional field of instability waves. Arrays of actuators and sensors are distributed locally at the wall surface and connected together via a feedback controller. The key elements of this feedback control are the determination of the dynamic model of the flat plate boundary layer between the actuators and the sensors, and the design of the model-based feedback controller. The dynamic model is established based on the linear stability calculation which simulates the three-dimensional input-output behaviour of the boundary layer. To simplify the control problem, an uncoupled control mode of the dynamic model is made to capture only those dynamics that have greatest influences on the input-output behaviour. A Proportional-Integral-Derivative (PID) controller, i.e. a lead-lag compensator, combining with a standard Smith predictor is designed based on the system stability criterion and the specifications using frequency-response methods. Good performance of the feedback control with the uncoupled control mode is demonstrated by the large reduction of the three-dimensional disturbances in the boundary layer. This simple feedback control is realistic and competitive in a practical implementation of T-S wave cancellation using a limited number of localised sensors and actuators.

Data-Driven Modeling of Wireless Power Transfer Systems With Multiple Transmitters

This article develops a new method of data-driven modeling for a class of multiple-transmitter single-receiver wireless power transfer (WPT) systems. A continuous-time multiple-input single-output (MISO) model with pure time delays is used to characterize the input–output behavior of the system, where the transfer functions associated with each input channel are not constrained to have the same denominator. Moreover, the time delays are allowed to be a fraction of the sampling interval in order to account for the delay effects that stem from circuit components and wireless communication, which are, by nature, often a fraction of the sampling interval. An optimal refined instrumental variable method is proposed to estimate the parameters and time delays of the MISO model based on sampled input–output data. In contrast to the conventional circuit-theory-based modeling methods that rely on circuit parameters and result in models which are often complex, the proposed data-based method yields parsimonious models, whose parameters are directly estimated from input–output data. Due to the easy availability of sampled data in control engineering applications, the proposed method is clearly more user-friendly, having a broad prospect for efficient operation of WPT systems, such as prediction, optimization, and control. Numerical and experimental results are presented to validate the effectiveness and merit of the proposed method.