Alexander Stotsky

Robustness and performance of an indirect adaptive control scheme in presence of bounded disturbances

Robustness and performances in the presence of disturbances of an indirect adaptive control scheme driven by the prediction error which is designed by organizing a sliding mode are examined in this paper. A new idea for performance improvement in the presence of disturbances is presented. The idea is applied to plants with unknown relative degree. It is proved that the proposed scheme can provide arbitrarily improved transient performances. The results are confirmed by simulation.

Adaptive Estimation of the Engine Friction Torque

Errors in an estimate of friction torque in modern spark ignition automotive engines have a direct impact on driveability performance of a vehicle and necessitate a development of real-time algorithms for adaptation of the friction torque. The friction torque in the engine control unit is presented as a look-up table with two input variables (engine speed and indicated engine torque). Algorithms proposed in this paper estimate the friction torque during engine start and idle. Newton’s second law for rotational dynamics is used as a reference model during engine start. The friction torque is estimated via a deviation from the reference model. The values of the friction torque at the nodes of the look-up table are updated, if new measured data of the friction torque is available. New recursive and computationally efficient algorithms are developed for adaptation of the nodes of the look-up tables. The algorithms are tested on a Volvo vehicle equipped with a six cylinder prototype engine.

Recursive trigonometric interpolation algorithms

Abstract Two new recursive algorithms for the calculation of the coefficients of the trigonometric polynomial that is fitted to measured data in the least-squares sense in a window of a sufficiently large size that is moving in time are proposed. Both algorithms use the strictly diagonally dominant property of the information matrix for a sufficiently large window size. The first one is based on a recursive inversion of the information matrix and the second one uses a priori estimate of the coefficients of the polynomial obtained from the Kaczmarz projection method. The algorithms are computationally suitable for selection of the order of a trigonometric polynomial in each step of a moving window in real-time applications. The results are verified by simulations.

Statistical Engine Misfire Detection

AbstractNew recursive filtering algorithms for misfire detection based on the trigonometric interpolation method are proposed for spark ignition automotive engines. The technique improves the performance of the filtering algorithms, allowing a flexible choice of the size of the moving window. Correction algorithms are introduced for the recursive trigonometric interpolation method that ensure robustness with respect to round-off errors which are present in the finite precision implementation environment. New real-time statistical algorithms based on a hypothesis testing for a misfire detection are proposed. The statistical decision-making mechanism makes it possible to achieve misfire detection at a certain significance level with an automatically selected sample size depending on the signal quality, which in turn improves the robustness of the misfire detection algorithm. This work was done within the Volvo Six Sigma program.

Proactive control of wind turbine with blade load constraints

This paper describes an easy-to-implement, proactive control strategy for wind turbines, incorporating constraints on blade loads. The control strategy is aimed for rejecting wind gusts and is based on upwind speed measurements and a new statistical wind gust detection mechanism. The control action comprises simultaneous driveline and collective pitch control with constraints on flapwise bending moment. The controller is evaluated by simulation on a transient between two steady-state operational modes of the wind turbine. A new driveline backstepping-based controller with integral action for compensation of steady-state errors is also proposed and verified by simulations.

Model Based Control of Wind Turbines: Look-Ahead Approach

A new composite turbine control architecture that consists of feedforward and feedback parts based on the upwind speed measurements and wind speed measurements at the turbine site, respectively, is described. The algorithm starts with preprocessing of a low-rate sampled upwind speed via the spline interpolation method. A run-ahead model driven by the signals from a preprocessing block models the turbine response and produces the feedforward part of turbine controller. The turbine control system is driven by both the feedforward part that comes from the run-ahead model and feedback part based on the wind speed measured at the turbine site. It is proved that the controller is stable despite the difference between the time-shifted preview measurements (expected wind speed) and the actual wind speed measured at the turbine site. Existing industrial proportional–integral–derivative turbine controllers can easily be upgraded with the preview part of the control architecture described in this article. Improved blade load regulation via the blade pitch angle control guarantees a hard upper bound on the flapwise bending moment. The results are confirmed by simulation with a wind speed record from the Hönö turbine outside Gothenburg, Sweden.

Variable structure control of engine idle speed with estimation of unmeasurable disturbances

A new controller for throttle and spark advance to control the engine speed at idle under unknown time varying disturbances is proposed in this paper. By using measurements of the engine speed the disturbance estimator is designed to reconstruct a disturbance torque. The controller is formulated so that the throttle is used as much as possible as a main tool to produce a torque and spark advance is used to compensate intake to torque production delay. The stability of the system is proved via the Lyapunov method.

A new frequency domain system identification method

A new frequency domain system identification method based on a multi-frequency input signal is proposed. Frequency contents of the oscillating signal are estimated using a modified Kaczmarz algorithm proposed in this paper. Lyapunov stability analysis is performed for this new Kaczmarz algorithm and transient bounds for estimation error are established. Moreover, a new method for estimation of the variance of the measurement noise in Kaczmarz algorithms is also described. A comparison of a transient performance of modified Kaczmarz algorithm and a recursive least-squares algorithm is presented. The results are applied to a frequency domain identification of a DC motor.

Frequency Determination in Control Applications: Excitation Based Approach

New algorithms for estimation of the frequencies of oscillating waveform signals are described. Model of the signals is presented in the form of linear difference equation with unknown coefficients, which define the frequencies and amplitudes. Coefficients are estimated utilizing the property of the persistence of excitation of oscillating signals. Exponentially damped and oscillating signals are described in a unified framework. A property of excitation is proved for exponentially damped signal that contains a single frequency via diagonal dominance of an information matrix. Two applications of this frequency estimation technique are considered. The first one is filtering of the wind speed signal in wind turbine control applications, and the second one is the frequency estimation of exponentially damped signal motivated by the engine knock detection applications.

Computationally efficient filtering algorithms for engine torque estimation

A new computationally efficient filtering algorithm for reconstruction of the first harmonic of a periodic signal is presented. The algorithm allows to recover the combustion quality information from the engine speed measurements which are noise contaminated. The algorithm is applied to the torque estimation problem for a V8 spark ignition engine.