Inverse Transfer Function Research Articles

Tip-sample distance control using photothermal actuation of a small cantilever for high-speed atomic force microscopy

We have applied photothermal bending of a cantilever induced by an intensity-modulated infrared laser to control the tip-surface distance in atomic force microscopy. The slow response of the photothermal expansion effect is eliminated by inverse transfer function compensation. By regulating the laser power and regulating the cantilever deflection, the tip-sample distance is controlled; this enables much faster imaging than that in the conventional piezoactuator-based z scanners because of the considerably higher resonant frequency of small cantilevers. Using this control together with other devices optimized for high-speed scanning, video-rate imaging of protein molecules in liquids is achieved.

A study on an estimation system of inverse transfer function using adaptive filter estimating minimum‐phase and allpass transfer function

AbstractA structure is proposed for an adaptive system with an adaptive filter located before the unknown system (pre‐inverse adaptive system) for estimation of the inverse of the transfer function (inverse transfer function) of the unknown system. In general, when an adaptive transversal filter is used as an adaptive filter, the delay signal of the output of the unknown system is needed in the adaptive algorithm for the weights. Since an adaptive filter is inserted in the front stage, this signal cannot be observed, so that a replica of the unknown system is needed. In this paper, an adaptive system that does not require this replica is discussed. Estimation of the inverse transfer function of the minimum phase of the unknown system is performed by an adaptive exponential filter and an inverse copy of the weights of the exponential filter placed in front of the unknown system. The signal within the adaptive algorithm consists of the observable input signal to the adaptive exponential filter and the estimation error. Estimation of the inverse transfer function for the allpass transfer function of the unknown system is performed by the adaptive transversal filter and the reversing copy of the weight to the transversal filter located before the unknown system. The signal in the adaptive system consists of the observable input signal to the exponential filter and the estimation error. Convergence of the weight is studied from the point of view of monotonic increase of the gradient. The unique feature of the approach is that the algorithm of the two adaptive filters consists of a gradient algorithm with guaranteed convergence for the weights and of copies of the weights after updating. Finally, a performance evaluation of the adaptive system and a comparison with conventional systems are performed by numerical simulation. © 2007 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 90(9): 10– 17, 2007; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjc.20305

A compensation scheme for CVT transient effects using artificial neural network

Capacitive voltage transformers (CVTs) provide instrument-level voltage signals to meters and protective relays in high voltage (HV) systems. The transients in CVTs could lead to protective relay mal-operation. This paper proposes the use of an artificial neural network (ANN) scheme to correct CVT secondary waveform distortions due to CVT transient. This is accomplished by using samples of voltage signals to achieve a good approximation of the inverse transfer function of CVTs, thus an accurate estimation of the primary voltages. Simulations are performed and the impacts of different parameters are studied. Performance results show that the proposed scheme is accurate and reliable. The proposed scheme can also be implemented on a digital signal processor board for real-time application.

Application of the ultrasonic characterization methods for highly attenuating plastic materials

In this paper, prediction of the back surface reflection, which is based on attenuation and phase velocity dispersion estimation method, in highly attenuating plastic (polyvinylidene fluoride—PVDF) has been described. Estimation of the attenuation law is based on the inverse transfer function of the object approximation in the frequency domain. The oscillating character of the inverse transfer function of the highly attenuating plastic material gives rise to an ill-posed problem for approximation. It has been solved in two ways: application of the Tikhonov regularization process and iterative adjustment of the approximation parameters. The estimated attenuation coefficient α 0 and power n have been used for calculation of phase velocity dispersion using the Kramers–Kronig relations and having attenuation coefficient measured at single-frequency value. The reconstructed waveforms of the ultrasonic back surface reflections using the estimated attenuation and phase velocity dispersion curves have been presented and compared with the experimental one.

Compensation of Machined Profile at Quadrant Motion by Real-Time Corrected Inverse Transfer Function

Compensation of Machined Profile at Quadrant Motion by Real-Time Corrected Inverse Transfer Function

Real-time substructuring technique for the shaking table test of upper substructures

In this study, a real-time substructuring technique for the shaking table test is proposed. The proposed technique adopts the upper part of the object structure as the experimental substructure and the lower part as the numerical substructure. The force exerted on the interface of the two separated substructures is calculated on-line using the feed-back from an absolute floor acceleration measurement. The acceleration at the interface of the two substructures is calculated through an on-line time history analysis of the numerical substructure and is used as a reference signal in the shaking table controller. The inverse transfer function of the shaking table is employed to minimize the errors between the reference signal and generated ground motion for the experimental substructure. Through a shaking table test, the validity and accuracy of the proposed technique is demonstrated, and a good agreement between experimental and numerical results is obtained.

A Study on Pre-Inverse Adaptive System Based on Estimation of Minimum-Phase and Allpass Transfer Function

In this paper, we discuss a pre-inverse adaptive system in which the adaptive filter is located before the unknown system. We then propose an adaptive system that does not require a replica of the unknown system, as in the case of the filtered-x adaptive system. With the proposed method, we can therefore avoid the instability that arises from error in the replica of the unknown system. In the proposed method, the inverse transfer function of the minimum-phase part of the unknown system is estimated by an adaptive exponential filter. The inverse transfer function of the allpass part of the unknown system is estimated by an adaptive transversal filter, whose inverse can be easily realized by reversing the order of the weights. Finally, a numerical verification of the analytical results, the convergence performance of the adaptive systems, and a performance comparison with the filtered-x adaptive system are obtained by computer simulation

Improvement of “SAFIA” source separation method under reverberant conditions

A sound source separation method with robustness in real environments is needed to achieve clear speech communications and improve noise immunity in a noisy environment. One of the problems in source separation using multiple microphones is to improve performance under reverberant conditions. This paper applies two improvements to SAFIA, a conventional source separation method, in order to improve its performance under reverberant conditions. One of the improvements is smoothing of the interchannel level difference in the time direction in order to reduce SAFIA's decision error. In this way the noise reduction rate (NRR) can be improved by 7.0 to 8.0 dB while the distortion of the target speech is reduced. The other is the addition of a mechanism that separates signals overlapping on the same frequency to the first stage of SAFIA. The mechanism calculates the ratio of the transfer functions of the source and the microphones from the ratio of the received signals of the channels, so that the inverse transfer function between the sources and the microphones can be estimated. These techniques improved the NRR of the proposed method (enhanced SAFIA) by 7.3 dB and its distortion by 12.6 dB on average compared with the conventional SAFIA, which is a great improvement in performance. © 2005 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 89(3): 22–37, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ ecjc.20160

Correction of Saturated Current Transformers Secondary Current Using ANNs

Current transformers (CTs) provide instrument-level current signals to meters and protective relays. Protective relays' accuracy and performance are directly related to steady-state and transient performance of CTs. CT saturation could lead to protective relay maloperation or even prevent tripping. This paper proposes the use of an artificial neural networks scheme to correct CT secondary waveform distortions. The proposed module uses samples of current signals to achieve the inverse transfer function of CT. Simulation studies are preformed and the influence of changing different parameters is studied. Performance studies results show that the proposed algorithm is accurate and reliable. The proposed algorithm has also been implemented and tested on a digital signal processor board. Details of the implementation and experimental studies are provided in this paper.

Real-Time Synthesis and Control by Corrected Inverse Transfer Function of an NC Table(Precision positioning and control technology)

Real-Time Synthesis and Control by Corrected Inverse Transfer Function of an NC Table(Precision positioning and control technology)

Repetitive PD control strategy with inverse transfer function compensation for CVCF inverter

A novel repetitive control strategy for the output waveform of single-phase CVCF inverters is presented. In this scheme, the inverse transfer function of inverter is used as a compensator to obtain stable and satisfy harmonic rejection. Besides, PD controller is adopted to improve transient performance. Simulation and experimental results, which are gotten from a DSP-based 400Hz, 5.5KW inverter, indicate that the proposed control scheme can achieve not only low THD during steady-state operation but also fast transient response during load step change.

Direct repetitive control of SPWM inverter for UPS purpose

A novel repetitive controller directly combined with an open loop SPWM inverter is presented in this paper. To cope with the high-resonant peak of the open loop inverter that may cause instability, a zero-phase-shift notch filter other than the inverse transfer function of the inverter or a conventional second-order filter is incorporated in the controller. The proposed method has good harmonic rejection and large tolerance to parameter variations. To further reduce the steady-state error, a low-pass-filter Q(z) algorithm is applied. The DC bias problem is also taken into consideration and solved with the repetitive controller itself. The method is implemented with a digital signal processor and achieves low THD% (1.4%-1.7%) with nonlinear loads and fast error convergence (3-5 fundamental periods). It proves to be a cost-effective solution for common UPS products where high-quality output voltage is more stressed than fast dynamic response.

Site Response at Treasure and Yerba Buena Islands, California

A variety of methods are utilized to reinvestigate the physical relationship between the seismic response of Treasure Island (TI) and Yerba Buena Island (YBI) in California. These islands are a soil (TI) and rock (YBI) site pair separated by 2 km. The site pair has been used previously by researchers to identify soil response to earthquake shaking. Linear regime ground motions (MW4.0-MW4.6 and PGA: 0.014–0.017 g) recorded in the TI vertical array indicate a coherent wavefield in the sediments and an incoherence between the rock and sediments. Our analyses show that the greatest change in the wavefield occurred between the rock and soil layers, corresponding to a significant impedance contrast. The waveforms change very little as they propagate through the sediments, indicating that the site response is a cumulative effect of the entire soil structure and not a result of wave propagation within individual soil layers. In order to highlight the complexity of the site response, correlation analysis was used to demonstrate that the rock and soil ground motions were not highly coherent between the two sites. YBI was, therefore, shown to be an inappropriate reference site for TI. One-dimensional (1D) vertical wave propagation and inverse techniques were used to differentiate between 1D site response and more complex site behavior. Both 1D methods (vertical wave propagation and inverse transfer functions) proved incapable of capturing the site response at TI beyond the initial four seconds of motion. Finite difference waveform modeling, based on a two-dimensional velocity structure of the northern San Francisco Bay was needed to explain the linear site response at TI as horizontally propagating surface waves trapped in the bay sediments. A simplified velocity structure for the San Francisco Bay including a single 100 m basin layer (constant shear-wave velocity of 400 m/s) over a 1.5 km/s layer of Franciscan bedrock was able to trap energy in the basin and produce surface waveform ringing similar to that observed in the TI data. Due to surface waves propagating in the San Francisco Bay sediments, any 1D model will not fully characterize site response at TI. All 1D models will fail to produce the late arriving energy observed in the ground motions.

How to handle colored observation noise in large least-squares problems

An approach to handling colored observation noise in large least-squares (LS) problems is presented. The handling of colored noise is reduced to the problem of solving a Toeplitz system of linear equations. The colored noise is represented as an auto regressive moving-average (ARMA) process. Stability and invertability of the ARMA model allows the solution of the Toeplitz system to be reduced to two successive filtering operations using the inverse transfer function of the ARMA model. The numerical complexity of the algorithm scales proportionally to the order of the ARMA model times the number of observations. This makes the algorithm particularly suited for LS problems with millions of observations. It can be used in combination with direct and iterative algorithms for the solution of the normal equations. The performance of the algorithm is demonstrated for the computation of a model of the Earth’s gravity field from simulated satellite-derived gravity gradients up to spherical harmonic degree 300.

Site Response at Treasure and Yerba Buena Islands, California

A variety of methods are utilized to reinvestigate the physical relationship between the seismic response of Treasure Island (TI) and Yerba Buena Island (YBI) in California. These islands are a soil (TI) and rock (YBI) site pair separated by 2 km. The site pair has been used previously by researchers to identify soil response to earthquake shaking. Linear regime ground motions (MW4.0-MW4.6 and PGA: 0.014–0.017 g) recorded in the TI vertical array indicate a coherent wavefield in the sediments and an incoherence between the rock and sediments. Our analyses show that the greatest change in the wavefield occurred between the rock and soil layers, corresponding to a significant impedance contrast. The waveforms change very little as they propagate through the sediments, indicating that the site response is a cumulative effect of the entire soil structure and not a result of wave propagation within individual soil layers. In order to highlight the complexity of the site response, correlation analysis was used to demonstrate that the rock and soil ground motions were not highly coherent between the two sites. YBI was, therefore, shown to be an inappropriate reference site for TI. One-dimensional (1D) vertical wave propagation and inverse techniques were used to differentiate between 1D site response and more complex site behavior. Both 1D methods (vertical wave propagation and inverse transfer functions) proved incapable of capturing the site response at TI beyond the initial four seconds of motion. Finite difference waveform modeling, based on a two-dimensional velocity structure of the northern San Francisco Bay was needed to explain the linear site response at TI as horizontally propagating surface waves trapped in the bay sediments. A simplified velocity structure for the San Francisco Bay including a single 100 m basin layer (constant shear-wave velocity of 400 m/s) over a 1.5 km/s layer of Franciscan bedrock was able to trap energy in the basin and produce surface waveform ringing similar to that observed in the TI data. Due to surface waves propagating in the San Francisco Bay sediments, any 1D model will not fully characterize site response at TI. All 1D models will fail to produce the late arriving energy observed in the ground motions.

ELECTROMAGNETIC HOLOGRAPHY ON CYLINDRICAL SURFACES USING K-SPACE TRANSFORMATIONS - Abstract

Spectral decomposition in 2-D kz-m space is used to develop transfer functions that relate modal electromagnetic fields on concentric cylindrical surfaces. It is shown that all time-average radiated power is generated by superluminal modes (phase velocity vz > c) which are confined to the baseband |kz| < k0. Subluminal modes, with kz outside of the radiated band, are radially evanescent but permit recovery of imaging resolution that exceeds the usual diffraction limit provided by the radiated fields. Outward translation between cylinder surfaces is found to have a stable low-pass 2-D transfer characteristic in kz-m space, where spatial resolution decreases with increased radius. The inverse transfer functions for inward translation of field components (termed backpropagation) employ a high-pass process that amplifies subluminal evanescent modes, thus potentially enhancing resolution while also amplifying measurement noise. A 2-D filter with flat elliptical passband and Gaussian roll-off is used to mitigate noise amplification with backpropagation. Outward translation and backpropagation are tested using sampled data on finite-length cylinders for various noise levels.

A study on an estimation of inverse transfer function using an adaptive exponential filter

AbstractConsidering an adaptive system estimating the inverse of the transfer function (inverse transfer function) of an unknown system, a configuration is proposed in which the adaptive system is placed in front of the unknown system (pre‐inverse modeling). In general, when a transversal adaptive filter is used, the output delay signal of the unknown system is needed in the adaptive algorithm of the tap coefficient. This signal cannot be observed because an adaptive filter is inserted in the prior stage, so that a replica of the unknown system is needed. In this paper, an adaptive system that does not require this replica is considered. The inverse transfer function of the minimum‐phase section of the unknown system is estimated with an adaptive exponential filter by means of a gradient algorithm. Since the signal in the tap algorithm is the delayed output signal of the observable adaptive system, no replica is required. Convergence of the tap coefficients is rigorously studied theoretically from the point of view of monotonic increase of the gradient. Also, the minimum‐phase section of the unknown system is estimated with a transversal adaptive filter by using a gradient algorithm. A method of calculation of the inverse transfer function with this estimated value is presented. The algorithm of the two adaptive filters is a gradient algorithm that is easily realized with guaranteed convergence. Finally, numerical verification of the study results, performance evaluation of the adaptive system, and comparison with the conventional system are carried out by computer simulation. © 2002 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 85(12): 47–55, 2002; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjc.1127

Improvement of Shaking-Table Control by Real-Time Compensation of Reaction Force Caused by a Specimen. 1st Report. Scheme for Controller Design and Verification by Simulation.

An improved shaking-table control method that can compensate the reaction force caused by a non-linear specimen in real-time has been developed. This method measures the reaction force and generates a drive force to cancel it. The compensation signal to generate the drive force is calculated by a reaction-force compensator, which consists of the inverse transfer function of the shaking-table actuating system, a filter to reduce the noise in the measured reaction-force signal, and a proportional element to stabilize the shaking-table characteristics. The motion of a shaking table loading a non-linear specimen was simulated. This simulation shows that the compensator effectively cancels the reaction force and that increasing the SN ratio of the measured reaction force increases this effect. Moreover, two means to measure the reaction force are presented. The first means are the loadcells between the table and the specimen. The second means are the loadcells between pistons and the table.

PA—Precision Agriculture: An Analytical Grain Flow Model for a Combine Harvester, Part II: Analysis and Application of the Model

A dynamic grain flow model describes how the grain flow at the end of the threshing process reacts to feedrate variations during harvest. These input flow variations have different origins as there are variations in travel speed, cutting width and locally variable grain yield. In this study, an analysis of the model is performed and a specific application in the domain of precision farming is illustrated.A closer look at the construction of the analytical grain flow model reveals some complicated dynamics in the system and an internal return loop with a significant time delay. Especially, the latter property makes it difficult to simplify the model into a linear transfer function, an interesting form for further applications. Due to a specific property of grain spreading in the return flow, it is possible to compensate for the variations in the return flow fraction. The grain smoothing effect of the combine harvester for 0% return flow was approximated by a fourth-order linear transfer function.Some producers of combine harvesters already possess a commercial product for online grain yield measurement. These sensors are typically mounted at the end of the threshing process since this is the only place where a flow of clean grain is present. Nevertheless, this position has also a drawback. Grain flow signals, measured at the end of the threshing process contain unconditionally the influence of disturbing machine dynamics. In practice, this means that feedrate variations and subsequently estimated yield variations, are smoothed reducing the economic interest of the resultant yield maps. The inverse simplified linear transfer function makes it possible to reduce these latter errors and to improve immediately the results of further site-specific grain yield analysis.

Correction of current transformer distorted secondary currents due to saturation using artificial neural networks

Current transformer saturation can cause protective relay misoperation or even prevent tripping. This paper presents the use of artificial neural networks (ANN) to correct current transformer (CT) secondary waveform distortions. The ANN is trained to achieve the inverse transfer function of iron-core toroidal CTs which are widely used in protective systems. The ANN provides a good estimate of the true (primary) current of a saturated transformer. The neural network is developed using MATLAB and trained using data from EMTP simulations and data generated from actual CTs. In order to handle large dynamic ranges of fault currents, a technique of employing two sets of network coefficients is used. Different sets of network coefficients deal with different fault current ranges. The algorithm for running the network was implemented on an Analog Devices ADSP3101 digital signal processor. The calculating speed and accuracy proved to be satisfactory in real-time application.