Flood control and management is a fundamental issue for hydrology researchers and managers. Regarding the design and construction of different hydraulic structures such as reservoirs and dams, as effective techniques for flood control, accurate estimation of the magnitude and return periods of flood is required for appropriate estimation of the dimension and resilience of structures. Design flood estimation is done through frequency analysis with the key stationary assumption. Nowadays, factors such as land use change, inappropriate management and climate change has influenced stationary conditions of flood peaks. Therefore, in the presence of NONSTATIONARY conditions, the estimation based on stationary assumption is not confident and may lead to large errors. In this study for non-stationary flood frequency analysis, the GAMLESS model for location, scale and shape parameter estimation are introduced while visual inspection of NONSTATIONARY are as well presented and developed for quantile estimation. Six hydrometery stations in different provinces in the north of Iran were selected. Frequency analysis in stationary and non-stationary conditions was performed for each station. Results indicated that location and scale parameters have linear and quadratic trend. In addition, in Nodekhormalo station the design flood estimated by NONSTATIONARY assumption was around 3 times higher than of that obtained for the stationary conditions. Results also demonstrated that in stations with increasing non-stationary trend, return period of large floods was decreasing and for the same return periods, flood quantiles has increased.

In this article, a just-in-time adaptive distance metric learning for application in NONSTATIONARY environments is addressed. This generative model enables adaptive similarity-based classifiers to classify the time-labeled inquiry patterns which are coming from a nonparametric stochastic process with superior performance. Here, the performance points to accurately classification in low-dimensional feature space with minimum model adaptation during the system life time. While there are adaptive forms of feature extraction methods, which transform training patterns to a low-dimensional space and/or improve classifier accuracy, they are vulnerable to nonparametric changes in data and must continuously update their parameters. In the proposed method, an optimal transformation matrix transforms time-labeled instances from the original space to a new feature space to maximize the probability of selecting the correct class label for incoming instances using similarity-based classifiers. To this end, for a given time-labeled instance, nonparametric intra-class and extra-class distributions are proposed. The proposed method is also furnished to a temporal detector to provide the most convenient time for the adaptation phase. Experimental results on real and synthesized datasets that include real and artificial changes demonstrate the performance of the proposed method in terms of accuracy and dimension reduction in dynamic environments.

Aims: Over the past twenty years, Iran has experienced a rise in extreme temperatures, particularly in hot events like extreme temperatures, as indicated by recent studies. This research seeks to analyze the annual maximum temperatures (AMT) in the dry Province of Kerman, Iran, focusing on both stationary (S) and NONSTATIONARY (NS) behavior. Materials & Methods: Trend, homogeneity, and stationarity tests were utilized to identify the critical characteristics of the AMTs from 1979 to 2019. Frequency analysis of the AMTs was conducted using both stationary Generalized Extreme Value (S-GEV) and NONSTATIONARY GEV (NS-GEV) models, estimating distribution parameters through a maximum likelihood estimator(MLE). In addition to the time-varying NS-GEV (TNS-GEV) investigations, soil moisture (SM) was incorporated as a covariate.  Findings: Results demonstrate that, compared to the S-GEV case, the NS-GEV frequency analyses significantly impact the return values of the AMTs, leading to an increase. The NS-GEV estimations for 50-year return levels were significantly higher than those in the S-GEV. The study’s findings revealed that the average Akaike Information Criterion (AIC) for both the S-GEV and TNS-GEV estimations decreased from 110 to 71 across all 12 selected stations in Kerman Province. The AIC value for the NS-GEV with the soil moisture (SM) covariate was approximately 94. Thus, the TNS-GEV frequency analysis of AMTs resulted in improved AIC values compared to the NS-GEV with soil moisture as the covariate. Conclusion: Given the NONSTATIONARY (NS) conditions caused by natural and/or human activities, it is recommended to utilize NS frequency analysis for estimating hydrologic variables across different design periods. It has been noted that NS-GEV frequency analyses lead to higher return levels of AMTs than S-GEV analyses.

In this study, we propose a novel method for computing both primal and dual filters for NONSTATIONARY biorthogonal wavelets, offering an advanced approach to wavelet filter design. The key challenge in image compres-sion that this study addresses is the inefficiency of conventional station-ary wavelets, which rely on fixed filter banks that do not adapt to local variations in an image. This limitation results in suboptimal compression performance, particularly for images with varying statistical properties and localized features. To address this, we use a NONSTATIONARY biorthogonal fil-ter banks, which modify basis functions at different scaling levels, leading to improved frequency resolution, signal representation, and compression efficiency. Our technique employs cardinal Chebyshev B-splines to derive explicit formulas for the primal filters, enabling precise calculation of filter coeffi-cients essential for wavelet transforms. Additionally, we enforce normality and biorthogonality conditions within NONSTATIONARY multiresolution anal-ysis to maintain the relationship between primal and dual wavelet filters at each scaling level. This structured approach allows for explicit formulation of the dual filters while ensuring accurate decomposition and reconstruc-tion. Experimental results confirm that the proposed method improves compression efficiency over conventional Daubechies biorthogonal filters, increasing the number of zero coefficients in compressed images. This leads to better visual quality and reduced storage requirements while maintaining computational efficiency. Such improvements are particularly beneficial in applications requiring high-fidelity image reconstruction, such as medical imaging, satellite data processing, and video compression. MATLAB sim-ulations validate the effectiveness of the approach, making it a promising alternative for image processing and data compression applications.

Different approaches are presented for systems identification in the literature. Benchmark problems for the system identification and damage detection of civil engineering structures are established, and different methods are illustrated by international participants. In this paper, the dynamic characteristics of a three story shear frame, subjected to NONSTATIONARY white noise excitation are identified by the use of Natural Excitation Technique (NExT), Wavelet and Hilbert transforms. Because the ambient vibration imposed on the system is NONSTATIONARY, the response acceleration of the system is also NONSTATIONARY. Therefore, a method is used to turn NONSTATIONARY signals into stationary ones. Natural Excitation Technique is applied to extract free vibration responses of the system from the available stationary signals. Continuous Wavelet Transform (CWT) of free vibration decay decomposes the signals to a set of sub-signals corresponding to natural vibration modes. The mother wavelet used is modified complex morlet wavelet. Analytical complex signals are extracted from the mentioned sub-signals using Hilbert Transform. The Hilbert transform is applied to each modal response to obtain the instantaneous phase angle and amplitude as functions of time t. Then, a linear least-square fit algorithm is used to fit the instantaneous phase angle and the log of instantaneous amplitude. From the slopes of these linear least-square lines, the natural frequency and damping ratio of each mode can be identified. Based on a single measurement of the free vibration time history at a proper location of the MDOF linear system, all natural frequencies and damping ratios can be identified. When the responses at all degrees of freedom are measured, the complete system dynamic characteristics can all be identified, including the mode shapes, damping and stiffness matrices. The applications of the proposed method are illustrated in detail using a linear three degrees of freedom shear frame. Simulation results show that the accuracy of the method in identifying the system characteristics is remarkable.