An approach to the problem of designing a stable corrective digital filter is presented. Approach based on an assumption of a stabilisation level and minimisation of an approximation or an assumption of the approximation level and minimisation a stabilisation level of an ideal corrective filter. The correction is performed by a cascade connection of both systems. The obtained solution is an anticipative system.
Calibrating signals for measuring systems intended for measurement of dynamic signals are generally limited to standard signals. The unit step input in time domain and sinusoidal signal in frequency domain are the most often applied here. However dynamic signals have in the practice quite arbitrary shapes which significantly differ from the standard signals. In consequence, the results of a calibration process realised by means of standard input signals are in practice useless, except of some very particular cases. On the other hand the process of calibration can not be realised by means of...
The proposed method of creating of batch procedures for determining of signal derivatives bases on idea of so-called „ averaged differentiation”. It is shown, that application of moments of weight function allows to design the correction procedures , which can be used for recovery of instantaneous values of derivatives on the base of „averaged” on certain time interval derivatives of noisy signal x ( t ). The problem of accuracy of proposed correction procedures, also in case of presence of disturbances influencing the processed signal x ( t ), is deeply discussed. The obtained formulae can...
A method for correcting dispersive distortion inside a pressure bar, on the basis of experimentally determining the bar's dispersive characteristic is presented in the paper. For this purpose the continuous wavelet transform together with the Gabor analyzing function was applied. In this method, a one-point measurement of the strain inside the bar was used. The results of correction of impact force waveforms, generated by longitudinal impacts of steel spheres into a steel bar, are presented.
Some new methods of fault diagnosis of electronic linear multiports based on bilinear transformation in multidimensional spaces are presented. They are extensions of the input-output methods 3D, 4D and 6D previously elaborated by authors for diagnosis of twoport electronic circuits. The methods can be applied in a diagnosis of linear electronic circuits with the aid of different technologies: conventional measurement systems, testing buses and neural networks. They can be utilised in practice for localisation and identification of single, double and triple faults. Examples illustrating the...
Digital filters are commonly used in engineering. In many different technical problems it is necessary to apply digital filtration with required accuracy of signal processing in defined frequency domain with possible great attenuation of disturbances. This paper concerns correctors of known measuring windows like Hamming and Barthann according to requirements mentioned above.
Measurement, in the wide sense, is defined as a process of empirical, objective assignment of symbols to attributes of objects and events of the real world, in such a way as to describe them. Strictly defined measurement is defined as measurement that conforms to the paradigm of the physical sciences. Weakly defined measurement is measurement in the wide sense, but which is not strongly defined. Strongly and weakly defined measurements are analysed and compared. They are compared with other forms of symbolic representation.
The Guide to the expression of uncertainty in measurement (GUM) (ISO,1995), contains internationally agreed recommendations for the evaluation and expression of uncertainties in a measurement process, providing valuable advise for constituting the model that relates input to output quantities. If a probability density function is associated with each of input quantities, a probability “region” for the outputs quantities (the measurands) can be derived and its level of confidence may be evaluated by analytical or numerical methods. This paper analyzes the uncertainty evaluation of a process...
In this paper we propose the weak t-norm (wt-norm) based arithmetics to describe the propagation of systematic as well as statistical inexactness. We use the general term “inexactness” to express all possible approaches of inaccuracy evaluations. In the presented theory the measurement results are fuzzy intervals. The arithmetic operators are obtained by applying the generalisation of Zadeh’s extension principle based on the wt-norm. The averaging operator defined in the framework of wt-norm arithmetic decreases the statistical component, and does not change the systematic component of...