This paper introduces a new variational Gaussian filtering approach for estimating the state of a nonlinear dynamic system. We first assume that the predictive distribution of the state is Gaussian and derive an itera...
详细信息
Virtual concerts have become an emerging form of music performance, and it casts a perspective on the development and exploration of real-time interactive virtual spaces as well as the creation of digital twins. This ...
详细信息
The Mossformer model excels in speech separation but has not been effectively applied to music source separation. Music sources have complex characteristics and higher sampling rates, making separation tasks more chal...
详细信息
ISBN:
(数字)9798350380347
ISBN:
(纸本)9798350380354
The Mossformer model excels in speech separation but has not been effectively applied to music source separation. Music sources have complex characteristics and higher sampling rates, making separation tasks more challenging. We addressed a rarely explored task of separating piano concerto recordings into individual piano and orchestral tracks. This process involves intricate coordination between the piano and orchestra, creating highly complex audio signals in both time and frequency domains. Our main contributions include: (1) adapting the speech separation model for the novel task of piano concerto source separation, constructing and processing a specialized dataset.(2) introducing channel attention in the separation module to dynamically adjust feature focus based on instrument characteristics, enhancing key features. Experiments on the Piano Concerto Dataset (PCD) showed improved separation performance, with a 0.22dB average Signal-to-Distortion Ratio (SDR) increase over the baseline model.
As a highly developed region, Guangdong province has substantial industrial emissions. Its subtropical monsoon climate, characterized by abundant hydrothermal conditions, contributes to a substantial biomass potential...
详细信息
Modular batteries can be aggregated to deliver frequency regulation services for power grids. Although utilizing the idle capacity of battery modules is financially attractive, it remains challenging to consider the h...
详细信息
The quotient of two multivariate Gaussian densities can be written as an unnormalized Gaussian density, which has been applied in some recently developed multiple-model fixed-interval smoothing algorithms. However, th...
详细信息
The extensive integration of distributed resources such as distributed photovoltaic system and electric vehicles enhances the uncertainty of the load side greatly. Corresponding demand response strategies can be devel...
The extensive integration of distributed resources such as distributed photovoltaic system and electric vehicles enhances the uncertainty of the load side greatly. Corresponding demand response strategies can be developed for different distributed resource types and characteristics of households’ load, which is of great significance for new energy consumption and stable operation of the power system. However, only the net load power of households can be obtained through smart meters instead of the distributed resource types of households. Based on this, a distributed resource classification and identification model based on feature extraction is proposed in this paper, and effectively solves the problems of low accuracy and strong data dependence of existing recognition methods. Firstly, a generalized weather class generation algorithm based on clustering and voting is established, which is used to determine the weather class of each day to provide a basis for the following feature extraction. Then, based on the typical net load profiles under different generalized weather classes, a two-stage feature extraction and a classification identification model based on integrated learning algorithms are established to identify whether a user contains DPVS and EV, respectively. Finally, based on the category labels of each user for each stage, the final classification of the category to which the user belongs is carried out. Simulation experiments show that the recognition model built using the typical features extracted in the paper has good recognition accuracy.
Accurate prediction of distributed photovoltaic power can help the arrange the scheduling plan reasonably to Maintain the safe and stable operation of the distribution network. However, distributed photovoltaic power ...
Accurate prediction of distributed photovoltaic power can help the arrange the scheduling plan reasonably to Maintain the safe and stable operation of the distribution network. However, distributed photovoltaic power generation exists lack of meteorological data and real-time monitoring data, the distribution range is large and dispersed, the installed capacity is relatively small and so on, the centralized photovoltaic power prediction method is not applicable to distributed photovoltaic power prediction, so it is necessary to carry out a regional power prediction study for distributed photovoltaic. Distributed PV is located in close proximity, in a similar external environment, and the power output between stations has strong spatial and temporal correlation, which can be captured to improve the accuracy of power forecasting. To this end, a regional distributed PV power prediction model based on static-dynamic spatiotemporal correlation modeling is proposed. The k-means is firstly used to cluster the regional distributed photovoltaic sites into several sub-regions, according to the geographical location and output characteristics that reflect the long-term static correlation between power stations. Then, graph convolutional neural network are applied to capture the dynamic correlation between subregions according to the power. In this way, the static and dynamic correlation are effectively considered in power prediction modeling, and the prediction accuracy has been improved. The superiority of the proposed method is verified by a real world dataset in China.
Atmospheric oxidizing capacity(AOC)is an essential driving force of troposphere chemistry and self-cleaning,but the definition of AOC and its quantitative representation remain *** by national demand for air pollution...
详细信息
Atmospheric oxidizing capacity(AOC)is an essential driving force of troposphere chemistry and self-cleaning,but the definition of AOC and its quantitative representation remain *** by national demand for air pollution control in recent years,Chinese scholars have carried out studies on theories of atmospheric chemistry and have made considerable progress in AOC *** paper will give a brief review of these ***,AOC indexes were established that represent apparent atmospheric oxidizing ability(AOIe)and potential atmospheric oxidizing ability(AOIp)based on aspects of macrothermodynamics and microdynamics,respectively.A closed study refined the quantitative contributions of heterogeneous chemistry to AOC in Beijing,and these AOC methods were further applied in Beijing-Tianjin-Hebei and key areas across the *** addition,the detection of ground or vertical profiles for atmospheric OH·,HO_(2)·,NO_(3)·radicals and reservoir molecules can now be obtained with domestic instruments in diverse ***,laboratory smoke chamber simulations revealed heterogeneous processes involving reactions of O_(3)and NO_(2),which are typical oxidants in the surface/interface atmosphere,and the evolutionary and budgetary implications of atmospheric oxidants reacting under multispecies,multiphase and multi-interface conditions were ***,based on the GRAPES-CUACE adjoint model improved by Chinese scholars,simulations of key substances affecting atmospheric oxidation and secondary organic and inorganic aerosol formation have been *** numerical simulations of AOIe and AOIp were performed,and regional coordination of AOC was *** optimized plan for controlling O_(3)and PM2.5was analyzed by scenario simulation.
Modular batteries can be aggregated to deliver frequency regulation services for power grids. Although utilizing the idle capacity of battery modules is financially attractive, it remains challenging to consider the h...
详细信息
暂无评论