The accurate prediction of photovoltaic (PV) power output is crucial for optimal energy management. However, PV generation systems are influenced by various meteorological factors, resulting in the fluctuation and intermittency issues their power. To enhance accuracy power, this study proposes a novel hybrid model, Convolutional Neural Network-Long Short-Term Memory-Kolmogorov-Arnold networks (CLKAN), integrated with Improved Sparrow Search Algorithm (ISSA), Variational Mode Decomposition...

Urban mobility is undergoing significant technological transformations, with the application of sharing and autonomous driving technologies. It will reshape people's travel behavior patterns. However, direction change heavily influenced by urban spatial features, including density population, distance between residents job, public transportation infrastructure, diversity local place, as well form. In response to this evolving landscape, study integrates macro-level predictions from IAM...

Lots of similarity-based algorithms have been designed to deal with the problem link prediction in past decade. In order improve accuracy, a novel cosine similarity index CD based on distance between nodes and value vectors is proposed this paper. Firstly, node coordinate matrix can be obtained by distances which are different from row regarded as coordinates nodes. Then, used their index. A local community density LD also proposed. series CD-based indices include CD-LD-k, CD*LD-k, CD-k CDI...

A cyber-physical system (CPS) represents the coupling network between a communication and power grid. Information flow (IF) as medium of networks has an important influence on performance assessment this type network. However, traditional IF models cannot be applied to CPS directly because they do not take distinctive effect associated operational mechanism into account. In paper, we propose information model (IFM) based graph theory traffic dynamics considering functional architecture...

Diagnosis of power transformer abnormality is very important for system reliability. This paper presents a novel approach fault diagnosis based on probabilistic neural network and dissolved gas-in-oil analysis (DGA) technique. A new hybrid evolutionary algorithm combining particle swarm optimization (PSO) back- propagation (BP)algorithm, referred to as HPSO-BP algorithm, proposed select optimal value PNN parameter. The developed in such way that PSO used do global search give good direction...

Link prediction is an important task in the field of network analysis and modeling, predicts missing links current networks new future networks. In order to improve performance link prediction, we integrate global, local, quasi-local topological information Here, a novel stacking ensemble framework proposed for this paper. Our approach employs random forest-based recursive feature elimination select relevant structural features associated with constructs two-level model involving various...

In the pilot provinces of China’s electricity spot market, power generation companies usually adopt separate bidding mode, which leads to a low willingness demand-side response and poor flexibility in interaction mechanism between supply demand. Based on analysis demand day-ahead market with participation sales companies, this paper abstracted game process “power grid-sales company-users” tripartite competition environment into two-layer (purchase layer/sales layer) model proposed...

Feng and Huang in 2016 defined a new notion called weighted topological entropy (pressure) obtained the corresponding variational principle for compact dynamical systems. In this paper, it was our hope to carry out further study from following three aspects:(1) Inspired well-known classical theory, we define various (measure-theoretic) entropies investigate their relationships.(2) The formula of subsets transformations by factor maps is generalized version.(3) A which comes Brin-Katok...

The complex properties of an urban public transport network rise from its intrinsic evolutional mechanism. Three models are established which are, respectively, a stop network, transfer and line network. An analysis the basic characteristics these - including degree distribution, average path length clustering coefficient is completed. Moreover, this paper puts forward method for searching Hub node. This not only combines global efficiency local efficiency, but also provides quantitative...

In this article, researcher collaboration patterns and research topics on Intelligence Security Informatics (ISI) are investigated using social network analysis approaches. The networks exhibit scale-free property small-world effect. From these networks, the authors obtain key researchers, institutions, three important topics.

A method of identification a vulnerable branch considering the propagation characteristics cascading failures is proposed. To determine components that trigger or expand failures, fault chain model power system established by selecting with maximum outage probability among all branches as at next step in failure process. The process set can be compressed into directed and weighted graph merging repeated different processes accumulating corresponding weight values between components:...

At present, digital substation automation system has been widely used in China's power system, and IEC 61850 standard also preliminarily applied the construction of substation. However, there are a series problems engineering practice at present stage China, especially configuration intelligent equipment protection devices, which restrict further promotion system. Starting from several key this paper studies overall method testing debugging equipment, field equipment.

In this article, a unified method for fault locating and length measurement of the branch cable is proposed. First, typical two-port equivalent networks are extracted by means topology analysis cable, their scattering parameters (S-parameters) adopted to achieve proposed method, due excellent performance on characterizing wave propagation. Then, taking advantage inherent relation elements in S-parameters matrix, analytical formulae evaluating identifying location derived. This simple but...

A two-dimensional numerical gas-solid flow models without chemical reactions is developed to simulate in the fluidized bed boiler through MFIX software. Platen heating surfaces considered upper of furnace. Considering that velocity and concentration are most factor effecting on erosion, there three main area suffered serious erosion: (1) There a maximum at position ( X = 57 cm, Y= 60 cm) right side wall, which will produce impact erosion. (2) In vicinity (X 140 suffer tear. (3) As platen...