Hybrid deep learning and optimized variational mode decomposition for point-interval runoff prediction

Accurate wind power prediction can effectively alleviate the pressure of the power system peak frequency regulation, and is more conducive to the economic dispatch of the power system. To enhance wind power forecasting accuracy, a hybrid approach for wind power interval prediction is proposes in this study. Firstly, an Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) is applied to decompose the initial wind power sequence into multiple modes, and Variational Mode Decomposition is used to further decompose the high-frequency non-stationary components. Next, Fuzzy Entropy (FE) is utilized to assess the complexity of the post-decomposed Intrinsic Mode Functions (IMFs), and different forecasting methods are employed accordingly, the point predictions were obtained by linearly summing the component predictions.Additionally, an improved sparrow search algorithm (ISSA) is used to seek the optimal hyperparameters of the prediction algorithm. Finally, the prediction intervals are constructed using the point prediction results based on kernel density estimation (KDE). The root mean square errors (RMSE) of deterministic predictions are 2.8458 MW and 1.8605 MW, with uncertainty coverage rates of 95.83% and 97.92% at a 95% confidence level.

This paper proposes a wind power probabilistic model (WPPM) using the reflection method and multi-kernel function kernel density estimation (KDE). With the increasing penetration of renewable energy sources (RESs) into power systems, several probabilistic approaches have been introduced to assess the impact of RESs on the power system. A probabilistic approach requires a wind power scenario (WPS), and the WPS is generated from the WPPM. Previously, WPPM was generated using a parametric density estimation, and it had limitations in reflecting the characteristics of wind power data (WPD) due to a boundary bias problem. The paper proposes a WPPM generated using the KDE, which is a non-parametric method. Additionally, the paper proposes a reflection method correcting for the boundary bias problem caused by the double-bounded characteristic of the WPD and the multi-kernel function KDE minimizing the effect of tied values. Six bandwidth selectors are used to calculate the bandwidth for the KDE, and one is selected by analyzing the correlation between the normalized WPD and the calculated bandwidth. The results were validated by generating WPPMs with WPDs in six regions of the Republic of Korea, and it was confirmed that the accuracy and goodness-of-fit are improved when the proposed method is used.

High-precision monthly runoff prediction results are of great significance to regional water resource management. However, with the changes in human activity, climate, and underlying surface conditions, the runoff sequence presents highly nonlinear and random characteristics. In order to improve the accuracy of runoff prediction, this study proposed a runoff prediction model based on fuzzy information granulation (FIG) and back propagation neural network (BPNN) improved with genetic algorithm (FIG-GA-BP). First, FIG was used to process the original runoff data to generate three sequences of minimum, average, and maximum that can reflect the rule of runoff changes. Then, genetic algorithms (GA) were used to obtain the optimal initial weights and thresholds of the BPNN through selection, crossover, and mutation. Finally, BPNN was used to predict the generated three sequences separately to obtain the prediction interval. The proposed model was applied to the monthly runoff interval prediction of Linjiacun and Weijiabu hydrological stations in the main stream of the Wei River and Zhangjiashan hydrological station on Jing River, a tributary of the Wei River. Compared with the interval prediction model FIG-BP, FIG-WNN, and traditional BP model. The results show that the FIG-GA-BP interval prediction model had a good prediction effect, with higher prediction accuracy and a narrower range of prediction intervals. Therefore, this model has superiority and practicability in monthly runoff interval prediction.

In this paper, we present a Markov chain Monte Carlo (MCMC) simulation algorithm for estimating parameters in the kernel density estimation of bivariate insurance claim data via transformations. Our data set consists of two types of auto insurance claim costs and exhibits a high-level of skewness in the marginal empirical distributions. Therefore, the kernel density estimator based on original data does not perform well. However, the density of the original data can be estimated through estimating the density of the transformed data using kernels. It is well known that the performance of a kernel density estimator is mainly determined by the bandwidth, and only in a minor way by the kernel. In the current literature, there have been some developments in the area of estimating densities based on transformed data, where bandwidth selection usually depends on pre-determined transformation parameters. Moreover, in the bivariate situation, the transformation parameters were estimated for each dimension individually. We use a Bayesian sampling algorithm and present a Metropolis-Hastings sampling procedure to sample the bandwidth and transformation parameters from their posterior density. Our contribution is to estimate the bandwidths and transformation parameters simultaneously within a Metropolis-Hastings sampling procedure. Moreover, we demonstrate that the correlation between the two dimensions is better captured through the bivariate density estimator based on transformed data.

In many data stream mining applications, traditional density estimation methods such as kernel density estimation, reduced set density estimation can not be applied to the density estimation of data streams because of their high computational burden, processing time and intensive memory allocation requirement. In order to reduce the time and space complexity, a novel density estimation method Dm-KDE over data streams based on the proposed algorithm m-KDE which can be used to design a KDE estimator with the fixed number of kernel components for a dataset is proposed. In this method, Dm-KDE sequence entries are created by algorithm m-KDE instead of all kernels obtained from other density estimation methods. In order to further reduce the storage space, Dm-KDE sequence entries can be merged by calculating their KL divergences. Finally, the probability density functions over arbitrary time or entire time can be estimated through the obtained estimation model. In contrast to the state-of-the-art algorithm SOMKE, the distinctive advantage of the proposed algorithm Dm-KDE exists in that it can achieve the same accuracy with much less fixed number of kernel components such that it is suitable for the scenarios where higher on-line computation about the kernel density estimation over data streams is required.We compare Dm-KDE with SOMKE and M-kernel in terms of density estimation accuracy and running time for various stationary datasets. We also apply Dm-KDE to evolving data streams. Experimental results illustrate the effectiveness of the proposed method.

With the increasing energy demand, energy development has been widely concerned. As a kind of clean energy, wind power is the focus of energy development. But in the use of wind power, randomness is a major hidden danger in the safe operation of the system. So we can find that the research topic is critical to make wind speed prediction more efficient and accurate. In this study, the original sequence is divided by the variational mode decomposition (VMD) method, and the neural network usually act as a prediction model on the wind speed, and the training error is obtained. In interval prediction, an interval prediction method combining empirical error with quantile statistical Markov Chain Monte Carlo (MCMC) is creatively proposed. According to the randomness of MCMC, a large number of independent error prediction cases can be generated. According to the quantiles of the results of multiple random simulations, the quantiles are estimated to form a series of interval predictions with a given confidence. Finally, the point prediction value is added to the error prediction interval to accomplish the wind speed prediction interval completely. The prediction eventually show that: (1) the model processed by VMD is more accurate and stable than the single neural network prediction method; (2) in the interval prediction, the combination of empirical error and MCMC method shows better prediction effect; (3) the interval prediction with different confidence represents the consistency of statistical significance. In this paper, the proposal of a neural network method of wind speed interval prediction based on empirical error and MCMC is helpful for the relevant departments to adjust the dispatching plan in time, formulate reasonable control strategy of the wind farm, and improve stability and reliability of the system operation.

Refined offshore wind speed prediction: Leveraging a two-layer decomposition technique, gated recurrent unit, and kernel density estimation for precise point and interval forecasts

Kernel density estimation is an important data smoothing technique. It has been applied most successfully for univariate data whilst for multivariate data its development and implementation have been relatively limited. The performance of kernel density estimators depends crucially on the bandwidth selection. Bandwidth selection in the univariate case involves selecting a scalar parameter which controls the amount of smoothing. In the multivariate case, the bandwidth matrix controls both the degree and direction of smoothing so its selection is more difficult. So far most of the research effort has been expended on automatic, data-driven selectors for univariate data. There is, on the other hand, a relative paucity of multivariate counterparts. Most of these multivariate bandwidth selectors are focused on the restricted case of diagonal matrices. In this thesis practical algorithms are constructed, with supporting theoretical justifications, for unconstrained bandwidth matrices. The two main classes of univariate bandwidth selectors are plug-in and cross validation. These unidimensional selectors are generalised to the multidimensional case. The univariate framework for theoretically analysing kernel density estimators is extended to a general multivariate version. This framework has at its core the quantification of the relative rates of convergence which provide a guide to the asymptotic behaviour of bandwidth selectors. Simulation studies and real data analysis are employed to illustrate their finite sample behaviour. It is found that unconstrained selectors possess good asymptotic and finite sample properties in a wide range of situations. Buoyed by this success, two extensions are embarked upon. The first is variable bandwidth selection, generalising the above case where the bandwidth is fixed throughout the sample space. The variation of the bandwidths is controlled by the local properties of the data. The novel contribution is to use non-parametric clustering to summarise these local properties, along with unconstrained bandwidth matrices. The second is in kernel discriminant analysis where unconstrained bandwidth matrices are shown to produce more accurate discrimination.

Problems with censored data arise frequently in survival analyses and reliability applications. The estimation of the density function of the failure times is often of interest. Two inherent problems in density estimation for lifetime data are the spill-over at the origin and the smoothing parameter selection. To address these issues, we propose the use of asymmetric kernels (like inverse Gaussian) with bandwidths selected by a Bayes criterion. We show strong pointwise consistency of the density estimator, and for suitable choices of the prior, we show that one can obtain meaningful bandwidths with the same rates of convergence as for the classical asymptotically optimal bandwidths.

Kernel density estimation is a widely used statistical tool and bandwidth selection is critically important. The Sheather and Jones’ (SJ) selector [A reliable data-based bandwidth selection method for kernel density estimation, J. R. Stat. Soc. Ser. B 53 (1991), pp. 683–690] remains the best available data-driven bandwidth selector. It can, however, perform poorly if the true density deviates too much in shape from normal. This paper first develops an alternative selector by following ideas in Park and Marron [On the use of pilot estimators in bandwidth selection, Nonparametr. Stat. 1 (1992), pp. 231–240] to reduce the impact of the normal reference density. The selector can bring drastic improvement to less smooth densities that the SJ selector has difficulty with, but may be slightly worse off otherwise. We then propose to combine the alternative selector and SJ selector by using the smaller of the two bandwidths, which has the effect of automatically picking the better one for a particular density. In our extensive simulation, study using the 15 benchmark densities in Marron and Wand [Exact mean integrated squared error, Ann. Statist. 20 (1992), pp. 712–736], the combined selector significantly improves the SJ selector for 5 difficult densities and retains the superior performance of the SJ selector for the other 10. A ready-to-use R function is provided.

Short-term wind speed forecasting is essential for enhancing the efficiency and dependability of wind renewable energy installations. Although often used, conventional point predictions generated by machine learning techniques frequently fail to accurately capture the natural uncertainty associated with wind speed variation. Modeling this type of uncertainty is crucial for providing credible information as the level of uncertainty increases. Prediction intervals (PIs) offer a probabilistic framework for quantifying forecast uncertainty. This paper presents a hybrid forecasting methodology that combines support vector regression (SVR) with adaptive kernel density estimation (AKDE) to estimate wind speed prediction intervals over various short-term horizons (10, 30, 60, and 120 minutes). In contrast to standard kernel density estimation (KDE), which employs a uniform bandwidth and may overlook local data attributes, the adaptive KDE approach adjusts the bandwidth in accordance with the local distribution of forecast errors, thereby facilitating more precise and locally tuned uncertainty quantification. The efficacy of the proposed SVR-AKDE model is evaluated against conventional KDE-based interval estimation. Outcomes are assessed by recognized PI quality indicators, including prediction interval coverage probability (PICP), prediction interval normalized average width (PINAW), and coverage width-based criterion (CWC). Simulation findings confirm the efficacy of our approach and demonstrate that the SVR-AKDE-based PI forecasting consistently provides enhanced coverage and narrower widths compared to traditional KDE. This approach provides a comprehensive solution for short-term wind speed forecasting with quantifiable uncertainty, therefore enhancing its application in operational wind energy control.

Recently Hazelton and Turlach (2009) proposed a weighted kernel density estimatorfor the deconvolution problem. In the case of Gaussian kernels and measurement er-ror, they argued that the weighted kernel density estimator is a competitive estimatorover the classical deconvolution kernel estimator. In this paper we consider weightedkernel density estimators when sample observations are contaminated by double expo-nentially distributed errors. The performance of the weighted kernel density estimatorsis compared over the classical deconvolution kernel estimator and the kernel density es-timator based on the support vector regression method by means of a simulation study.The weighted density estimator with the Gaussian kernel shows numerical instabilityin practical implementation of optimization function. However the weighted densityestimates with the double exponential kernel has very similar patterns to the classicalkernel density estimates in the simulations, but the shape is less satisfactory than theclassical kernel density estimator with the Gaussian kernel.Keywords: Deconvolution, kernel density estimator, support vector regression, weightedkernel density estimator.

Estimating the kernel density function of a random vector taking values on Riemannian manifolds are considered. More Precisely, we consider the problem of the estimation of the probability density of n i.i.d. random objects on the product of two compact Riemannian manifolds without boundary. The proposed methodology adapts the Pelletier’s approach which is the kernel density estimation on non Euclidean setting. Under sufficient regularity assumptions on the underlying density, L2 convergence rates are obtained. Riemannian products are the most natural and fruitful generalization of Cartesian products. These manifold models are of interest to many fields of applied mathematics, natural sciences and technology. Let us recall here as examples, the use of these products in the domain of computer vision and in biomedical sciences. In this note, we construct an estimator of the density on a product manifold from one of the estimators existing in the literature. Among the estimation methods, we mainly cite a method based on Fourier developments proposed in (Hendriks, 1990). Another method based on geodesic distance is proposed later by B. Pelletier in (Pelletier, 2005). This method used in the present work, is a generalization of the kernel method in the affine case. Another method recently developed by Kim et al. in (Kim, 2013) use the concept of exponential map to define a kernel estimator of the density on a Riemannian manifold. The aim in the present work, is to show that Pelletier's method is the best suited to the case of the Riemannian product.

Kernel density estimation (KDE) is a widely used method in geography to study concentration of point pattern data. Geographical networks are 1.5 dimensional spaces with specific characteristics, analyzing events occurring on networks (accidents on roads, leakages of pipes, species along rivers, etc.). In the last decade, they required the extension of spatial KDE. Several versions of Network KDE (NKDE) have been proposed, each with their particular advantages and disadvantages, and are now used on a regular basis. However, scant attention has been given to the temporal extension of NKDE (TNKDE). In practice, when the studied events happen at specific time points and are constrained on a network, the methodologies used by geographers tend to overlook either the network or the temporal dimension. Here we propose a TNKDE based on the recent development of NKDE and the product of kernels. We also adapt classical methods of KDE (Diggle's correction, Abramson's adaptive bandwidth and bandwidth selection by leave‐one‐out maximum likelihood). We also illustrate the method with Montreal road crashes involving a pedestrian between 2016 and 2019.

Numerous facets of scientific research implicitly or explicitly call for the estimation of probability densities. Histograms and kernel density estimates (KDEs) are two commonly used techniques for estimating such information, with the KDE generally providing a higher fidelity representation of the probability density function (PDF). Both methods require specification of either a bin width or a kernel bandwidth. While techniques exist for choosing the kernel bandwidth optimally and objectively, they are computationally intensive, since they require repeated calculation of the KDE. A solution for objectively and optimally choosing both the kernel shape and width has recently been developed by Bernacchia and Pigolotti (2011). While this solution theoretically applies to multidimensional KDEs, it has not been clear how to practically do so.A method for practically extending the Bernacchia–Pigolotti KDE to multidimensions is introduced. This multidimensional extension is combined with a recently-developed computational improvement to their method that makes it computationally efficient: a 2D KDE on 105 samples only takes 1 s on a modern workstation. This fast and objective KDE method, called the fastKDE method, retains the excellent statistical convergence properties that have been demonstrated for univariate samples. The fastKDE method exhibits statistical accuracy that is comparable to state-of-the-science KDE methods publicly available in R, and it produces kernel density estimates several orders of magnitude faster. The fastKDE method does an excellent job of encoding covariance information for bivariate samples. This property allows for direct calculation of conditional PDFs with fastKDE. It is demonstrated how this capability might be leveraged for detecting non-trivial relationships between quantities in physical systems, such as transitional behavior.