State Space Models

We develop and apply several novel methods quantifying dynamic multi-agent team interactions. These interactions are detected information-theoretically and captured in two ways: via (i) directed networks (interaction diagrams) representing significant coupled dynamics between pairs of agents, and (ii) state-space plots (coherence diagrams) showing coherent structures in Shannon information dynamics. This model-free analysis relates, on the one hand, the information transfer to responsiveness of the agents and the team, and, on the other hand, the information storage within the team to the teamʼs rigidity and lack of tactical flexibility. The resultant interaction and coherence diagrams reveal implicit interactions, across teams, that may be spatially long-range. The analysis was verified with a statistically significant number of experiments (using simulated football games, produced during RoboCup 2D Simulation League matches), identifying the zones of the most intense competition, the extent and types of interactions, and the correlation between the strength of specific interactions and the results of the matches.

This paper addresses fault diagnosis in dynamic systems represented by discrete state-space models. The main idea of the paper is to propose a systematic way to implement a Petri net-based fault diagnosis system. This procedure consists of three main steps. In the first step, a fault diagnosis system is built based on the Luenberger observer. In the second step, the obtained fault diagnoser equations are transformed to a suitable format. Finally, in the third step, the obtained equations are implemented by a Petri net called continuous-time delay Petri net (CTDPN) that can realize difference equations. Based on this method, a systematic approach is proposed for realizing a classical fault diagnoser by CTDPN. By integrating the concept of state-space observers and PNs in this paper, new and effective methods are developed for the analysis and fault diagnosis of systemsknown as hybrid systems-that have both continuous and discrete variables. The performance of the proposed method is thoroughly investigated, and the obtained results show that the proposed CTDPN can precisely detect the occurred faults, their types and their occurrence time instances.

Fırçasız DC Motorlar yüksek verimlilik, yüksek güç faktörü,
yüksek tork, temel kontrol ve düşük bakım maliyetleri
nedeniyle askeri uygulamalarda tercih sebebi olmaktadır. Bu
makalede, fırçasız bir DC Motor pozisyonunun, optimal
olarak (LQR) kontrol edilmesi çalışılmıştır. Bunlara ek
olarak, tasarlanan pozisyon kontrolör yapısı, hazırlanan test
düzeneğinde, xpc target ile gerçek zamanlı çalıştırılmıştır. Bu
çalışmada, tasarlanan optimal denetleyicinin gerçek sistem ve
şartlar altında performansı incelenmektedir. Geliştirilmiş olan
bu kontrolörün bozucu etkiler ve değişken pozisyon girdileri
için çeşitli deney sonuçları sunulmuş ve irdelenmiştir. Bu
sonuçlardan denetleyici performansının tatmin edici olduğu
anlaşılmaktadır.

Scientists, engineers and the like are a strange lot. Unperturbed by societal norms, they direct their energies to finding better alternatives to existing theories and concocting solutions to unsolved problems. Driven by an insatiable curiosity, they record their observations and crunch the numbers. This tome is about the science of crunching. It’s about digging out something of value from the detritus that others tend to leave behind. The described approaches involve constructing models to process the available data. Smoothing entails revisiting historical records in an endeavour to understand something of the past. Filtering refers to estimating what is happening currently, whereas prediction is concerned with hazarding a guess about what might happen next.
This book describes the classical smoothing, filtering and prediction techniques together with some more recently developed embellishments for improving performance within applications. It aims to present the subject in an accessible way, so that it can serve as a practical guide for undergraduates and newcomers to the field. The material is organised as an eleven-lecture course. The foundations are laid in Chapters 1 and 2, which explain minimum-mean-square-error solution construction and asymptotic behaviour. Chapters 3 and 4 introduce continuous-time and discrete-time minimum-variance filtering. Generalisations for missing data, deterministic inputs, correlated noises, direct feedthrough terms, output estimation and equalisation are described. Chapter 5 simplifies the minimum-variance filtering results for steady-state problems. Observability, Riccati equation solution convergence, asymptotic stability and Wiener filter equivalence are discussed. Chapters 6 and 7 cover the subject of continuous-time and discrete-time smoothing. The main fixed-lag, fixed-point and fixed-interval smoother results are derived. It is shown that the minimum-variance fixed-interval smoother attains the best performance. Chapter 8 attends to parameter estimation. As the above-mentioned approaches all rely on knowledge of the underlying model parameters, maximum-likelihood techniques within expectation-maximisation algorithms for joint state and parameter estimation are described. Chapter 9 is concerned with robust techniques that accommodate uncertainties within problem specifications. An extra term within Riccati equations enables designers to trade-off average error and peak error performance. Chapter 10 applies the afore-mentioned linear techniques to nonlinear estimation problems. It is demonstrated that step-wise linearisations can be used within predictors, filters and smoothers, albeit by forsaking optimal performance guarantees. Chapter 11 rounds off the course by exploiting knowledge about transition probabilities. HMM and minimum-variance-HMM filters and smoothers are derived. The improved performance offered by these techniques needs to be reconciled against the significantly higher calculation overheads.

PARA TODOS LOS SISTEMAS QUE A CONTINUACION SE MUESTRAN: ´
1. Realizar la simulaci´on del sistema en Simscape, u es una entrada escal´on. Con condiciones iniciales cero.
Graficar las variables de estado y la salida.
2. Obtener el modelo en el espacio del estado, de la forma: ˙x = Ax + Bu, y = Cx
3. Simular el modelo del punto 2, u es una entrada escal´on. Con condiciones iniciales cero. Comparar las variables
de estado, y la salida con la simulaci´on del punto 1.
4. Calcular el vector de estado x(t) = L
−1{(sI − A)
−1}x(0) + {(sI − A)
−1Bu(s)} y la respuesta del sistema
y(t) = Cx(t). Con condiciones iniciales cero.
5. Simular la soluci´on del punto 4, u es una entrada escal´on. Comparar las variables de estado, y la salida con la
simulaci´on del punto 1.
6. Determinar si el sistema es controlable.
7. Determinar si el sistema es observable

: sistema mec´anico traslacional, entrada u(t) = f(t),
Sistema mec´anico rotacional, la salida es y = θ1, la entrada u = T(t).

Bu çalışmada durum - uzay fonksiyonu verilen bir sistemin transfer fonksiyonunun çıkarılması, bu transfer fonksiyonuna bağlı sistemin kararlılık analizi ve denetleyici tasarımı Matlab programı yardımıyla ve Laplace dönüşümleri gerçekleştirilerek yapılmış, ilgili grafikler Matlab programı ile elde edilmiştir.

EM algorithm is popular in maximum likelihood estimation of parameters for state-space models. However, extant approaches for the realization of EM algorithm are still not able to fulfill the task of identification systems, which have external inputs and constrained parameters. In this paper, we propose new approaches for both initial guessing and MLE of the parameters of a constrained state-space model with an external input. Using weighted least square for the initial guess and the partial differentiation of the joint log-likelihood function for the EM algorithm, we estimate the parameters and compare the estimated values with the " actual " values, which are set to generate simulation data. Moreover, asymptotic variances of the estimated parameters are calculated when the sample size is large, while statistics of the estimated parameters are obtained through bootstrapping when the sample size issmall. The results demonstrate that the estimated values are close to the " actual " values.Consequently, our approaches are promising and can applied in future research.

The problem of vibrations cause the collapse of the systems and cause significant economic losses if not avoided. For this reason, the researchers have dealt with this issue in all years, but the problem is not over. In this paper will highlight the problem of controlling the vibrations resulting from fluid flow inside the pipes, in addition to the study of reducing the vibrations of these pipes. The ideal control is designed based on, the state space theory and according to the linear behavior of time, because the stability of the pipes are found from linear change. Different factors were studied on the controller's performance and then the optimal factors were obtained to obtain the best performance. The research included the study of the response and the natural frequency and the study of the dynamic behavior of different types of stabilization of the pipes in the presence with no hydraulic damping (active control) and monitoring the response and stability of each case of stabilization. Where, the control investigation include using the analytical solution of general equation of motion for pipe by using state space technique. There, the parameters studied for active control pipe vibration included the effect of hydraulic damper position, base width of hydraulic damper, damping, and the flow Pressure, on the pipe vibration response, with various pipe boundary conditions supported.

If computer programs were smarter, they would, like people, recognize sequences of events, form models of their environment, and formulate rules based on experience.  This paper describes the development of a program designed to address the difficult computational problems involved in integrating the inductive and deductive reasoning necessary to perform such tasks.  “Theme One” is a prototype program composed of “Index”, a learning algorithm for sequential data, and “Study”, an algorithm for building logical models.  The project goal is an interactive research tool that assists students and investigators in the exploration of qualitative data using artificial intelligence.

Modeling is a process of simplifying an event. One of the purposes of modeling is forecasting. A good model is capable of explaining data entirely in a simple way. A good forecasting is born from a precise modeling. This article will discuss the evaluation of time series data modeling without considering location factor, i.e. univariate time series data modeling in each location using Autoregressive (AR), multivariate time series data modeling in all locations using Vector Autoregressive (VAR), in addition to modeling which takes into account location factor, i.e. space-time modeling using Space Time Autoregressive (STAR) and Generalized Space Time Autoregressive (GSTAR) models. The scope of problem in this article is only determined autoregressive time series pattern in each location. In addition, the study only one variable and used Queen Contiguity spatial weight matrix. Data used was derived from simulation by generating first order stationary autoregressive pattern data, AR(1). Generated locations were 3, each of which with 100 time series, 10 percent data was used as evaluation tool between methods, i.e. by determining Root Mean Square Error (RMSE) for forecasting. The method was repeated 100 times to determine the consistency of RMSE. The results show that STAR method is the best among AR, VAR, and GSTAR methods for cases of same time order, indicated by the smallest RMSE frequency of STAR had the most, i.e. 37% in location 1, 32% in locations 2 and 3.

With the advent of Big Data, inference in very large datasets is becoming increasingly common. Furthermore, business models and operational strategies demand ever more often an online, user real-time feedback solution in contrast with standard batch methods. This involves performing inference on streaming data even before it is persisted. A flexible and elegant way of performing this task is by modelling data as a Dynamic Generalised Linear Model (DGLM), fundamentally a State-Space Model (SSM) which combines a linear transition state model with an observational model following a distribution from the exponential family. Here we will explain the basics of a fully Bayesian online and sequential state estimation using sequential Monte Carlo (SMC) methods and how this class of algorithms can be implemented to perform state and parameter estimation simultaneously. The merits and shortcomings of these methods will also be discussed.

One of the bench mark problem used by many researchers in the area of control engineering is the Ball and Beam system (BBS) which possess severe nonlinearity and instability characteristics. The BBS connected with servo motor results in an open loop unstable system due to the presence of multiple poles at the origin. This process has the difficulty in controller design because of assumed nonlinear relation between beam angle and ball displacement. This paper deals with design methodology of full state feedback (FSFB) controller with pre compensator and the performance of which is compared with linear quadratic controller (LQC). The state feedback controller with pre compensator yields better performance in terms of transient response specifications compared with linear quadratic controller. A simulation is carried out using MATLAB to evaluate the proposed control algorithm on the modelled Ball and Beam system. Keyword-BBS, State-Space model, full state feedback controller, linear quadratic controller __________________________________________________________________________________________________

An approach to obtaining a parsimonious polynomial model from time series is proposed. An optimal minimal nonuniform time series embedding schema is used to obtain a time delay kernel. This scheme recursively optimizes an objective functional that eliminates maximum number of false nearest neighbors between successive state space reconstruction cycles. A polynomial basis is then constructed from this time delay kernel. A sparse model from this polynomial basis is obtained by solving a regularized least squares problem. The constraint satisfaction problem is made computationally tractable by keeping the ratio between number of constraints to number of variable small by using fewer samples spanning all regions of the reconstructed state space. This helps the structure selection process from an exponentially large combinatorial search space. A forward stagewise algorithm is then used for fast discovery of the optimization path. Results are presented for Mackey-Glass system.

Studying the time-related course of psychological processes is a challenging endeavor, particularly over long developmental periods. Accelerated longitudinal designs (ALD) allow capturing such periods with a limited number of assessments in a much shorter time framework. In ALDs, participants from different cohorts are measured repeatedly but the measures provided by each participant cover only a fraction of the time range of the study. It is then assumed that the common trajectory can be studied by aggregating the information provided by the different converging cohorts. We conducted a Monte Carlo study to evaluate the practical relevance of using discrete- and continuous-time latent change score models for recovering the trajectories of a developmental process from ALD data under different sampling conditions. We focused on exponential trajectories typically found in the development of cognitive abilities from childhood to early adulthood. The results support the appropriateness of ALD designs to study such processes under various conditions of sampling. When all cohorts are drawn from the same population, both discrete- and continuous-time models are able to recover the parameters defining the underlying developmental process. However, discrete-time models yield biased estimates when time lags between observations are not constant. When cohorts are not from the same population and, thus, lack convergence, both types of models show bias in various parameters. We discuss the findings in the context of developmental methodology, encourage researchers to adopt continuous time models to analyze data from ALDs, and provide recommendations about how to implement such research designs.

Summary
1. Species’ abundances vary in space and time. Describing these patterns is a cornerstone of macroecology.
Moreover, trends in population size are an important criterion for the assessment of a species’ conservation status.
Because abundance trends are not homogeneous in space, we need to quantify variation in abundance trends
across the geographical range of a species. A basic difficulty exists in that data sets that cover large geographic
areas rarely include population abundance data at high temporal resolution. Whilst both broad-scale geographic
distribution data and site-specific population trend data are becoming more widely available, approaches are
required which integrate these different types of data.
2. We present a hierarchical model that integrates observations from multiple sources to estimate spatio-temporal
abundance trends. The model links annual population densities on a spatial grid to both long-term count data
and to opportunistic occurrence records from a citizen science programme. Specific observation models for both
data types explicitly account for differences in data structure and quality.
3. We test this novel method in a virtual study with simulated data and apply it to the estimation of abundance
dynamics across the range of a butterfly species (Pyronia tithonus) in Great Britain between 1985 and 2004. The
application to simulated and real data demonstrates how the hierarchical model structure accommodates various
sources of uncertainty which occur at different stages of the link between observational data and the modelled
abundance, thereby it accounts for these uncertainties in the inference of abundance variations.
4. We show that by using hierarchical observation models that integrate different types of commonly available
data sources, we can improve the estimates of variation in species abundances across space and time. This will
improve our ability to detect regional trends and can also enhance the empirical basis for understanding range
dynamics.

Transmission of wattage through High Voltage electricity (HVDC) has attracted the eye of several researchers in recent years. In this paper, to investigate the step response of rectifier and inverter controllers in a twelve-pulse current source converter-based HVDC transmission system, a well-developed linear continuous model is presented. In which, the converter AC system is represented by damped LLR equivalents at the first harmonic and at the third harmonic. Also, they're equipped with a double-tuned harmonic filter and second-order high pass filter to suppress the AC harmonics and a capacitor for reactive power compensation. The DC system is secured with rectifier current control, inverter current control, inverter voltage control, and inverter angle of extinction control. The HVDC transmission system model is implemented in the MATLAB/Simulink environment and the step response of the rectifier and inverter controllers has been investigated by observing the rectifier and inverter side DC quantities.

Information on stock status is available only for a few of the species forming the catch assemblage of rapido fishery of the North-central Adriatic Sea (Mediterranean Sea). Species that are caught almost exclusively by this gear, either as target (such as Pectinidae) or accessory catches (such as flatfishes apart from the common sole), remain unassessed mainly due to the lack of data and biological information. Based on cluster analysis, the catch assemblage of this fishery was identified and assessed using CMSY model. The results of this data-poor methodology showed that, among the species analyzed, no one is sustainably exploited. The single-species CMSY results were used as input to an extension of the same model, to test the effect of four different harvest control rule (HCR) scenarios on the entire catch assemblage, through 15-years forecasts. The analysis showed that the percentage of the stocks that will reach B msy at the end of the projections will depend on the HCR applied. Forecasts showed that a reduction of 20% of fishing effort may permit to most of the target and accessory species of the rapido trawl fishery in the Adriatic Sea to recover to B msy levels within 15 years, also providing a slight increase in the expected catches.

The aim of this paper is to present a State Estimator for Plunger Lift wells based on the Extended Kalman Filter (EKF) algorithm. The state estimator is a joint operation of the Plunger Lift (PL) dynamic model in State Space approach and EKF algorithm. The model is constituted by a set of discrete differential algebraic equations (DAEs) discretized and modeled in the form of equations in state space taking into account the measurement signals in the presence of noise. The state estimator is able to process the measurement signal thus providing estimates of the state variables that in this problem are slug velocity and casinghead pressure. The results of the computational simulation performed with data from a real well are predictions of important well performance items such as production rate, average piston upward velocity, cycle time, and pressures with values very close to those found in the well. The State Estimator for Plunger Lift is therefore a robust prediction system for the variables that influence the performance of a PL well since it considers the noises present in the process and in the measurement. The quality of the prediction is influenced by the correct adjustment of the parameters of the EKF algorithm. 1. Introduction The natural flowing wells are characterized by a high reservoir pressure, in which the fluids contained in their interior reach the surface freely. That is, the flow of fluids-oil, water and gas-from the reservoir to the facilities is due only to the reservoir energy, which usually occurs at the beginning of the well's productive life (Beggs, 2003). After that stage, artificial lift techniques (Brown, 1988) have to be applied to provide the additional energy required to raise reservoir fluids to the surface by using specific equipment to reduce the bottomhole flowing pressure, and consequently, increasing the inflow. Plunger Lift (PL) is one of that artificial lifting methods for oil and gas wells in which the plunger is a piston that moves freely up and down through the tubing and depends on the wellbore pressure to make oil or gas rise. In cyclic operation of the PL the well alternates between two conditions: flow, wherein the flow control valve is open, and closed, which is the period in which the flow control valve remains closed. During shut-in, the gas flow through the line is interrupted allowing the piston to fall to the bottom. After a predetermined time, the valve opens and the pressure on the surface falls below the wellhead pressure. The differential force through the piston, due to the reduction of pressure in the tubing above the slug and the high pressure of the well below the piston, is responsible for raising the liquid slug and the piston to the surface. Studies about models of PL system and intelligent methods for PL are discussed in this section. These theoretical elements compose the basis of the State Estimator for Plunger Lift (SEPL). Many researchers works on the development of different models for PL system. Some of their studies were chosen for the development of the state estimator for PL such as (Rosina, 1983) and (Baruzzi and Alhanati, 1995) (Rosina, 1983) presented a dynamic model for conventional PL only for the period from the opening of the production line until the moment when the piston reaches the surface with an experimental program used for the validation of the model. The model of (Rosina, 1983) supports (Baruzzi and Alhanati, 1995) work that described a method for predicting where it is possible to have liquid accumulation only in the tubing during buildup. Based on the premise that gas can accumulate only in the tubing-casing annular space, they have shown that there is a minimal Gas Liquid Ratio (GLR) to achieve that necessary condition for the model to work. In addition, the researchers developed a dynamic model for PL including an afterflow stage. They performed an afterflow time sensitivity analysis providing some recommendations for optimization of the PL system. About the application of intelligent techniques for artificial lift, the work (Salahshoor et al., 2013) presented a solution for casingheading instability problem for a gas-lift production well, based on integration of (O.A. Dourado), adriao@dca.ufrn.br (A.D. D oria Neto), pablo@dca.ufrn.br (P.J. Alsina).

This paper evaluates five models of Nowcasting that forecast Mexico's quarterly GDP: a Dynamic Factor Model (MFD), two Bridge Equation Models (BE) and two Principal Components Models (PCA). The results indicate that the BE forecasts average is statistically better than the rest of the models considered, according to the accuracy test of Diebold-Mariano (1995). In addition, using real-time information, it is found that the BE average is more accurate than the median of the forecasts provided by the analysts surveyed by Bloomberg and the median of the experts who answer Banco de México’s Survey of Professional Forecasters.

Glaucoma is a major cause of blindness and its mechanisms are not fully understood. The progression of the disease can be slowed by early diagnosis, but this is a difficult task because available data is typically noisy and has high variability. Several artificial intelligence approaches have been used in this context, although they generally don’t exploit the temporal nature of the data. Here we propose a new algorithm for modeling and clustering time series using the Non-stationary Dynamic Bayesian Network framework. Our method is tested both on a simulated and a glaucoma dataset, obtaining promising performances and enabling meaningful insight in the field.

Poverty is a fundamental problem that existed and still occurs in Indonesia. This analysis was conducted to describe the factors that influence the welfare low status in sub-districts of North Sulawesi Province. Global regression has obtained influence variables to poverty status are woman leader in household (X 1), Child not in formal education (X 2), physical defect (X3), chronic desease victim (X5), self-owned house (X6), household using electricity (X8) and household using its own toilet (X10). Geographical weighted regression analysis (GWR) is a locally model, in order to estimates of parameters at each location. The result of GWR has different parameter estimation at each the location. Mix Geographical weighted regression model (MGWR) has combined local and global parameters. As well as the GWR, MGWR model results has give different parameter estimation at each location. GWR Model is the best model to explain the influence of the low welfare status in sub-districts of North Su...

Information on stock status is available only for a few of the species forming the catch assemblage of rapido fishery of the North-central Adriatic Sea (Mediterranean Sea). Species that are caught almost exclusively by this gear, either as target (such as Pectinidae) or accessory catches (such as flatfishes apart from the common sole), remain unassessed mainly due to the lack of data and biological information. Based on cluster analysis, the catch assemblage of this fishery was identified and assessed using CMSY model. The results of this data-poor methodology showed that, among the species analyzed, no one is sustainably exploited. The single-species CMSY results were used as input to an extension of the same model, to test the effect of four different harvest control rule (HCR) scenarios on the entire catch assemblage, through 15-years forecasts. The analysis showed that the percentage of the stocks that will reach Bmsy at the end of the projections will depend on the HCR applied....

The VASIMR propulsion system is an ion propulsion system for spacecraft that uses magnetic fields to accelerate plasma to produce thrust. Undesired heat produced in the helicon section of VASIMR must be monitored and removed safely to avoid damage to system components, especially when higher power operating regimes are explored. This article demonstrates a strategy for distributed temperature estimation, based on OES measurement, and a model where the states represent the distributed temperature profile. OES provides a noninvasive measurement technique, which can be used as an output "correction" term for a state-estimation scheme. In this application, it is shown that the 2048 OES channels recorded can be accurately represented by only three principal components for temperature estimation. Use of the principal components as corrector terms in the state-space model dramatically improve model accuracy and the capability of the model to recover from unknown initial conditions and multiple system input changes.

Inferring music time structures has a broad range of applications in music production, processing and analysis. Scholars have proposed various methods to
analyze different aspects of time structures, such as beat, downbeat, tempo and meter. Many state-of-the-art (SOFA) methods, however, are computationally expensive. This makes them inapplicable in real-world industrial settings where the scale of the music collections can be millions. This paper proposes a new state space and a semi-Markov model for music time structure analysis. The proposed approach turns the commonly used 2D state spaces into a 1D model through a jump-back reward strategy. It reduces the state spaces size drastically. We then utilize the proposed method for causal, joint beat, downbeat, tempo, and meter tracking, and compare it against several previous methods. The proposed method delivers similar performance with the SOFA joint causal models with a much smaller state space and a more than 30 times speedup.

In comparison to classical cascade control architecture of DC motors, the state feedback control offers advantages in terms of design complexity, hardware realization and adaptivity. This paper presents a methodic approach to state space control of a DC motor. The state space model identified from experimental data provides the basis for a linear quadratic regulator (LQR) design. The state feedback linear control is augmented with a feedforward control for compensation of Coulomb friction. The controller is successfully applied and the closed loop behavior is evaluated on the experimental testbed under various reference signals.