An enhanced fault detection method for centrifugal chillers using kernel density estimation based kernel entropy component analysis

Xia, Yudong; Ding, Qiang; Jing, Nijie; Tang, Yijia; Jiang, Aipeng; Jiangzhou, Shu

doi:10.1016/j.ijrefrig.2021.04.019

Cited by 16 publications

(4 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kernel density analysis: This method was used to study the spatial agglomeration characteristics of tourist flows in tourist attractions in the Yangtze River Delta and calculate their heat (visit frequency) [46,47]. The formula is as follows:…”

Section: Gis Spatial Analysis Methodsmentioning

confidence: 99%

Spatial Structure Characteristics of Tourist Attraction Cooperation Networks in the Yangtze River Delta Based on Tourism Flow

Wang

Hang

2021

Sustainability

View full text Add to dashboard Cite

This study aimed to examine the spatial structure of the tourist attraction cooperation network in the Yangtze River Delta, from the perspective of tourist flow. This study conducted spatial and social network analyses of 470 popular tourist attractions in the Yangtze River Delta region of China, accounting for the occurrence and co-occurrence of tourist attraction information in tourist travel notes. The analyzed tourist attractions show an obvious spatial agglomeration effect, including four high-density agglomeration areas and two medium-density agglomeration areas. Degree centrality, closeness centrality, and betweenness centrality were used to examine the tourism function, distribution function, and connection function of nodes in the network; nodes were divided into various types of roles according to their function. There are eight condensed subgroups, but their scales are unbalanced. In these condensed subgroups, several tourist attractions with an intermediate function can be selected as transit and stopover points on tourist routes. This study can contribute to the understanding of tourists’ spatial behavior, clarify the role and status of nodes in the cooperation network of tourist attractions based on tourism flow, and help them to formulate measures for the joint marketing of tourist attractions, and promote the development of tourism in the Yangtze River Delta region.

show abstract

Section: Gis Spatial Analysis Methodsmentioning

confidence: 99%

Spatial Structure Characteristics of Tourist Attraction Cooperation Networks in the Yangtze River Delta Based on Tourism Flow

Wang

Hang

2021

Sustainability

View full text Add to dashboard Cite

show abstract

“…It also has to learn the mapping relationship between the system input and the output in a large amount of historical data in order to construct the model that can perform fault prediction in the system. In [21], a fault prediction model based on dynamic kernel principal component analysis and wavelet packet decomposition was designed to accurately detect the incipient fault in the bearing. In literature [22], fault prediction is investigated on the basis of probability density estimation, which can quantitatively detect the state of the system to detect the incipient fault in time.…”

Section: Introductionmentioning

confidence: 99%

“…Lemma 1. The approximation error of OLAD and the basis function comprise 𝜛 k in Equation (21). The upper bound of the smooth nonlinear function of 𝜛 k that is assumed to be the error of state estimation and parameter estimation, which can be expressed as…”

mentioning

confidence: 99%

Fault detection and prediction scheme for nonlinear stochastic distribution systems

Zhang

Yao

Sun

et al. 2023

Asian Journal of Control

View full text Add to dashboard Cite

Model‐based fault detection technique has a broad range of applications because of the small change to the system when the system state is known to be available and the low cost. For nonlinear stochastic distribution systems containing uncertain disturbance term, a model‐based fault detection and failure time prediction scheme is proposed in this paper, and observers are designed to detect whether the incipient fault has occurred in the system. The residual is obtained by comparing the output of the actual system with the output of the observer. When the residual exceeds the threshold value obtained by derivation, it is determined that the fault has occurred in the system. The fault size can then be estimated in real time and used to determine the time to failure (TTF) or the remaining useful life of the system. The TTF of the system is obtained by comparing the magnitude of the current system fault with the fault threshold. Finally, the feasibility of the presented fault detection scheme is proved by the Lyapunov stability theory and the validity of the scheme is proved by computer simulation.

show abstract

“…A lower fault level will give a smaller impact on the system operation, making it harder to identify the incipient defects. Over the years, while a number of data-driven approaches have been successfully applied for fault diagnosis in chillers (Han et al, 2011;Yan et al, 2014;Huang et al, 2018;Wang et al, 2018;Xia et al, 2021a), timely identifying the faults at their incipient stages is still a significant challenge. For example, the diagnosis accuracies for the incipient faults of refrigerant over charging and lubricant over charging were only 48 and 54.3%, respectively, in a recently reported study (Huang et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Incipient Chiller Fault Diagnosis Using an Optimized Least Squares Support Vector Machine With Gravitational Search Algorithm

et al. 2021

Self Cite

View full text Add to dashboard Cite

Operational faults in centrifugal chillers will lead to high energy consumption, poor indoor thermal comfort, and low operational safety, and thus it is of significance to detect and diagnose the anomalies timely and effectively, especially for those at their incipient stages. The least squares support vector machine (LSSVM) has been regarded as an effective algorithm for multiclass classification. One of the most difficult issues in LSSVM is parameter tuning. Therefore, this paper reports a development of a gravitational search algorithm (GSA) optimized LSSVM method for incipient fault diagnosis in centrifugal chillers. Considering the inadequacies of conventional principle component analysis (PCA) algorithm for nonlinear data transformation, kernel principle component analysis (KPCA) was firstly employed to reduce the dimensionality of the original input data. Secondly, an optimized “one against one” multi-class LSSVM classifier was developed and its penalty constant and kernel bandwidth were tuned by GSA. Based on the fault samples of seven typical faults at their incipient stages in chillers from ASHRAE RP 1043, the proposed GSA optimized LSSVM fault diagnostic model was trained and validated. For the purpose of demonstrating the priority of the proposed fault diagnosis method, the obtained results were compared to that of using the LSSVM classifier optimized by another two algorithms, namely, the conventional cross-validation method and particle swarm optimizer. Results showed that the best fault diagnosis performance could be achieved using the proposed GSA-LSSVM classifier. The overall average fault diagnosis accuracy for the least severity faults was reported over 95%.

show abstract

An enhanced fault detection method for centrifugal chillers using kernel density estimation based kernel entropy component analysis

Cited by 16 publications

References 38 publications

Spatial Structure Characteristics of Tourist Attraction Cooperation Networks in the Yangtze River Delta Based on Tourism Flow

Spatial Structure Characteristics of Tourist Attraction Cooperation Networks in the Yangtze River Delta Based on Tourism Flow

Fault detection and prediction scheme for nonlinear stochastic distribution systems

Incipient Chiller Fault Diagnosis Using an Optimized Least Squares Support Vector Machine With Gravitational Search Algorithm

Contact Info

Product

Resources

About