Fixed-gain Filter Research Articles

Design of a Fuzzy-Based Adaptive Gain Filter for PMSM Servo Systems With Maneuverability

In order to overcome the disadvantages of complexity and non-adaptability of the conventional Kalman filter (KF) in servo systems with maneuverability, a novel fuzzy-based adaptive gain filter (FAGF) is proposed in this paper. Firstly, based on the principle of KF, a third-order fixed gain filter (FGF) model is established to realize the state observation of the PMSM servo system. Then, the influence of filter gain on the observation performance of a time-varying system is analyzed by deriving the covariance matrices of the observed states. On this basis, an adjustable gain is designed by fuzzy logic to replace the fixed gain of the conventional observer. The algorithm can adaptively balance the fast and smooth performance of the observer according to the maneuvering conditions of the system. Finally, the superiority of the proposed FAGF is verified by both simulations and experiments.

Design of Full-Order State Observer for Two-Mass Joint Servo System Based on the Fixed Gain Filter

This article mainly designs a novel full-order state observer for the two-mass elastic servo system. In order to observe all the states of the two-mass system without the aid of any additional sensors, the model of the observer is constructed that includes two relative parts. Based on the theory of steady-state Kalman filter (KF), the fixed gain filter (FGF) scheme is studied. To overcome the heavy computational processing burden from the KF, the mechanism of the only one adjustable parameter is studied for the FGF-based observer. The adjustable parameter can be further refined by stability analysis and control characteristics within a small range, and the exact value can be calculated according to the desired observation performance, by compromising both the noise sensitivity and tracking capability. The proposed full-order observer method is experimentally verified in a two-mass system to provide easy parameter tuning, decreased computational complexity, and flexible observation performance in all cases.

Design of Full-Order State Observer for Motor Drive Systems Based on the Fixed Gain Filter

This paper mainly discusses about the full-order state observation technology for the permanent magnet synchronous motor (PMSM) drive system. First, the concept of the Kalman filter (KF) is synthesized to design a novel fixed gain filter (FGF) with an optimal fixed feedback gain matrix, which ensures a fast and accurate observation for position, speed, acceleration, and load torque. To alleviate the heavy computational load of the KF, second, the feedback gain is configured as a single tunable parameter only, which can be calculated offline. The regulation parameter is refined within a certain small range by fulfilling the system stability criterion. Furthermore, the system dynamic performance indexes, such as noise smoothing capability and bandwidth, are both analyzed to deduce the expected exact value of the regulation parameter. The proposed FGF can provide an accurate observation equivalent to the conventional KF, while the proposed FGF provides a faster tracking performance, an easier parameter tuning mechanism, and a lower program complexity. The performance of the FGF is verified by simulations and experiments in all cases.

A Fast and Precise Grid Synchronization Method Based on Fixed-Gain Filter

As the demand for electricity is growing and the penetration of renewable energy resources is increasing, grid synchronization with high accuracy is essential for utility networks. This paper presents a novel fixed-gain filter (FGF) scheme with an optimal fixed feedback gain matrix to estimate the (angle) position and frequency of the grid, which ensures a fast and accurate synchronization under varying grid conditions. Instead of using the conventional phase-locked loop (PLL) scheme that suffers from the difficulty of parameter selection, the concept of the Kalman filter (KF) is synthesized to design the second-order FGF and third-order FGF for synchronization. To overcome the heavy computational processing for the KF, the feedback gain is subject to a single tunable parameter only, which can be calculated offline and refined within a certain small range fulfilling the system stability criterion. Compared with the PLL approach, the new FGF provides a faster and more precise tracking performance, an easier parameter tuning mechanism, a lower program complexity, and a better grid stability, which is verified by simulations and experiments in all cases.

Angle Estimation of Extended Targets in Main-Lobe Interference With Polarization Filtering

In the presence of the interference from a main-lobe direction, the monopulse measurements always get erroneous angle estimates. By polarization filtering, the interference can be suppressed. However, an error from the cross-polarization of the target and the unpolarized component of the interference is introduced in the monopulse processing. To effectively extract the angle of the extended target accurately after polarization filtering, duplex polarization filtering and polarization filtering extractor based on a covariance matrix are developed. In addition, it can be found that the variance of the angle estimates varies when the polarizations of the target and the interference fluctuate, which degrades the performance of the tracking filter. For this problem, an adaptive fixed-gain filter for stabilizing the variance of the target angle estimates is proposed.

A General Solution to Optimal Fixed-Gain (<formula formulatype="inline"><tex Notation="TeX">$\alpha$</tex></formula>-<formula formulatype="inline"><tex Notation="TeX">$\beta$</tex></formula>-<formula formulatype="inline"><tex Notation="TeX">$\gamma$</tex></formula> etc.) Filters

A considerable number of papers in the literature provide solutions for fixed-gain filters, such as the $\alpha$ - $\beta$ filter, that minimize the mean squared error under various dynamic models. This note demonstrates that those specialized models are unnecessary as the tools to explicitly find optimal fixed gains for arbitrary observable systems are already in the literature. The results produced by this method agree with published results for specialized algorithms. The solution of this letter also obviates the need for special “aeolotropic” filter design. Code implementing the algorithm and simulations is available in Matlab on IEEEXplore.

Switching-Type $H_{\infty }$ Filter Design for T–S Fuzzy Systems With Unknown or Partially Unknown Membership Functions

This paper is concerned with the H∞ filter design problem for Takagi-Sugeno (T-S) fuzzy systems with unknown or partially unknown membership functions. If the membership functions are allowed to be unknown or partially unknown, then a fuzzy system may describe a wide class of nonlinear systems. However, in this case, the filter design of fuzzy systems, which is based on parallel distributed compensator strategy, is infeasible. To tackle this difficulty, a switching mechanism, which depends on the lower and upper bounds of the unknown membership functions, is introduced to construct the H∞ filter with varying gains. Some examples given in the simulation section show that the proposed method can achieve a better disturbance attenuation performance than a simple fixed-gain filter design approach.

A Practical Two-Input Single-Output Fuzzy Logic Controller with an Alpha-Beta Prefilter

This paper deals with the design of a practical two-input single-output (TISO) fuzzy logic controller for a nonlinear system. We firstly define two generalized errors for a nonlinear system and develop a stable feedback control scheme. Then a novel fuzzy logic controller is designed and incorporated into this scheme to achieve a better tracking response. The key feature of the proposed control scheme is that a discrete prefilter, called alpha-beta filter is placed in front of the fuzzy logic controller. The significance of using the Kalman-like fixed-gain filter is its capability of suppressing noise and feeding the fuzzy logic controller with smooth signals. From an implementation point of view, the discrete TISO fuzzy logic controller with the alpha-beta prefilter is believed to be quite simple, general and useful for practical applications. The validity of the proposed control scheme is verified through practical testing on the experimental setup called magnetic levitation system. The test results strongly suggest that the newly proposed control scheme is simple yet effective for use in a variety of feedback control systems.

Radar Beacon Tracking with Downlinked Heading and Airspeed

Position, ground velocity, and the wind are estimated using flight data acquired by the Discrete Address Beacon System (DABS). The measurements consist of radar range and bearing plus telemetered airspeed and heading sampled every 4 s. The data are used to determine statistics for the random errors in range and bearing, the random compass and airspeed errors, and the flight technical error in holding heading and airspeed. The random errors were found to be small in comparison to airspeed and heading-dependent compass biases. After correction for biases a fixed-gain filter provided wind estimates with an uncertainty of 2 knots (1 a) over a bandwidth of 0.05 Hz. The air-derived data were effective in improving tracking during vehicle accelerations or when a component of the DABS data was missing or inaccurate. Deterioration of the quality of measured data was detectable by monitoring the measurement residuals. The results have implications for systems which attempt to do large-area wind mapping with radar and telemetered air data.

Theory and applications of simultaneous realization of transfer and sensitivity functions

The theory developed on the problem of simultaneous realization of a transfer function and a prescribed sensitivity has been used to design variable parameter filters. In particular, variable centerfrequency fixed bandwidth and variable-bandwidth fixed-gain filters are considered in detail.