markdown____

Conditional density estimation is an important problem in a variety of areas such as system identification, machine learning, artificial intelligence, empirical economics, macroeconomic analysis, quantitative finance and risk management.

This work considers the general problem of conditional density estimation, i.e., estimating and predicting the density of a response variable as a function of covariates. The semi-parametric models proposed and developed in this work combine fuzzy and probabilistic representations of uncertainty, while making very few assumptions regarding the functional form of the response variable's density or changes of the functional form across the space of covariates. These models possess sufficient generalization power to approximate a non-standard density and the ability to describe the underlying process using simple linguistic descriptors despite the complexity and possible non-linearity of this process.

These novel models are applied to real world quantitative finance and risk management problems by analyzing financial time-series data containing non-trivial statistical properties, such as fat tails, asymmetric distributions and changing variation over time.

/pdf/conditional-density-models-integrating-fuzzy-and-2wibhpsaju.pdf

Conditional Density Models Integrating Fuzzy and Probabilistic Representations of Uncertainty

System identification is fundamental in many recent state-of-the-art developments in power electronic such as modeling, parameter tracking, estimation, self-tuning and adaptive control, health monitoring, and fault detection. Therefore, this paper presents a comprehensive review of parametric, non-parametric, and dual hybrid system identification for dc–dc switch mode power converter (SMPC) applications. The paper outlines the key challenges inherent with system identification for power electronic applications; speed of estimation, computational complexity, estimation accuracy, tracking capability, and robustness to disturbances and time-varying systems. Based on the literature in the field, modern solutions to these challenges are discussed in detail. Furthermore, this paper reviews and discusses the various applications of system identification for SMPCs; including health monitoring and fault detection.

/pdf/advances-on-system-identification-techniques-for-dc-dc-1sio5kvh0d.pdf

Advances on System Identification Techniques for DC–DC Switch Mode Power Converter Applications

DC distribution systems typically consist of several feedback-controlled switched-mode converters forming a complex power distribution system. Consequently, a number of issues related to stability arise due to interactions among multiple converter subsystems. Recent studies have presented methods such as passivity-based stability criterion, where the stability and other dynamic characteristics of an interconnected system can be effectively analyzed using bus-impedance measurement. Studies have presented online measurement techniques, where the bus impedance is obtained by combining together the measurements of input and output impedances of single converters in the system. Since the converters are coupled, the presented measurement techniques require several measurement cycles in order to sequentially measure the individual impedances to be combined to obtain the overall bus impedance. This paper presents a measurement technique based on injection of orthogonal binary sequences. Applying this method, all the impedances in the system can be simultaneously measured during one measurement cycle. Therefore, the overall measurement time of the bus impedance is reduced compared to conventional measurement techniques. Furthermore, the method guarantees that the system dynamics do not change between measurements, and therefore, the computed bus impedance is not distorted. Experimental results are presented and used to demonstrate the effectiveness of the proposed method in the design of a stabilizing controller for a notional dc system using positive feed-forward control.

/pdf/real-time-stability-analysis-and-control-of-multiconverter-2pnw2m07lu.pdf

Real-Time Stability Analysis and Control of Multiconverter Systems by Using MIMO-Identification Techniques

Evolving participatory learning (ePL) modeling joins the concepts of participatory learning and evolving fuzzy systems. It uses data streams to continuously adapt the structure and functionality of fuzzy models. This paper suggests an enhanced version of the ePL approach, called ePL+, which includes both an utility measure to shrink rule bases, and a variable cluster radius mechanism to improve the cluster structure. These features are useful in adaptive fuzzy rule-based modeling to recursively construct local fuzzy models with variable zone of influence. Moreover, ePL+ extends ePL to multi-input, multi-output fuzzy system modeling. Computational experiments considering financial returns volatility modeling and forecasting are conducted to compare the performance of the ePL+ approach with state of the art fuzzy modeling methods and with GARCH modeling. The experiments use actual data of S&P 500 and Ibovespa stock market indexes. The results suggest that the ePL+ approach is highly capable to model volatility dynamics, in a robust, flexible, parcimonious, and autonomous way.

Enhanced evolving participatory learning fuzzy modeling: an application for asset returns volatility forecasting

Switched-mode converters have become extensively used in powering different consumer products. The increased mass production using low-cost components with high parameter variation has set new challenges to designers. Recent studies have shown that frequency-domain characterization yields most useful information about the dynamics and quality of a converter. This has been widely recognized and fast correlation-based techniques have been developed enabling the use of frequency-response measurements both in design and production phases. These techniques are based on a broad-band excitation signal that is injected into a converter and its response is analyzed. Most of the previous papers have used the pseudorandom binary sequence (PRBS) as the excitation. The conventional binary sequence do not, however, suit well for the processes suffering from strong nonlinear distortions. This paper proposes an excitation signal that can be used similarly to the conventional PRBS but provides much more accurate frequency-response estimates. Inverse-repeat binary sequence (IRS) is applied to cancel the effect of even-order nonlinearities and the frequency response is measured by means of spectrum method. The proposed technique is verified with practical measurements.

Frequency-Response Measurement of Switched-Mode Power Supplies in the Presence of Nonlinear Distortions

Circular correlation based identification of switching power converter with uncertainty analysis using fuzzy density approach

This paper introduces a new class of GARCH method, termed the GARCH-Fuzzy-Density method for density forecasting the realizations of a process in which the higher-order moments may be time-varying. The method is based on a probabilistic Takagi-Sugeno fuzzy system and GARCH model. Traditional GARCH models usually assume that the shape of the conditional distribution of a process is conditional only on the first two moments. However, it is well documented that in empirical applications the conditional distribution beyond the first two moments may be conditional on higher-order moments such as skewness. Therefore, traditional GARCH models are insufficient for capturing all aspects of such processes. To resolve the problem mentioned above, the GARCH-FuzzyDensity model was developed. The method is capable of modeling conditional distributions in which the higher-order moments may be time-varying. Therefore, the GARCH-FuzzyDensity model provides more accurate density forecasts for the higher-order moment varying processes than the traditional GARCH models.

The GARCH-FuzzyDensity method for density forecasting

Frequency-domain characterization of switched-mode power supplies have become increasingly popular during the last years instead of time-domain analysis. The characterizing frequency responses of power supplies have been shown to be highly efficient tools in various design and validation procedures. This trend has yielded a number of applications for obtaining the frequency responses fast and accurately. One of the most popular methods has been the use of a broadband excitation, such as pseudo-random binary sequence (PRBS) and correlation-based techniques which have several advantages over the conventional sine-sweep-based methods. This paper applies the Dirac-ideal maximum-length binary sequence (D-MLBS) to measure the frequency responses and introduces technique to evaluate the measured responses using distributional-based models thus providing a tool for various statistical analyses. The proposed methods are verified by experimental data from a high-frequency switched-mode converter.

Evaluating the confidence of frequency response function of a switched-mode converter using distributional models

http://www.nt.ntnu.no/users/skoge/prost/proceedings/ifac11-proceedings/data/html/papers/1060.pdf

Tomi Helin

Papers

Circular correlation based identification of switching power converter with uncertainty analysis using fuzzy density approach

The GARCH-FuzzyDensity method for density forecasting

Evaluating the confidence of frequency response function of a switched-mode converter using distributional models

Continuous Monitoring of Industrial Processes through Cross-Correlation Techniques