A fault prediction method that uses improved case-based reasoning to continuously predict the status of a shaft furnace

Yan, Aijun; Wang, Weixian; Zhang, Chunxiao; Zhao, Hui

doi:10.1016/j.ins.2013.04.025

Cited by 36 publications

(16 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Set the minimum support degree threshold value minsup=25%, the result of event SPM is shown as Table 3. 4 <CR> 58.6% ESL 5 <MR, BP> 27.4% ESL 6 <MR, TO> 55.3% ESL 7 <MR, CR> 42.4% ESL 8 <BP, TO> 27.4% ESL 9 <BP, CR> 27.4% ESL 10 <TO, CR> 27.4% ESL 11 <MR, (BP, TO)> 27.4% ESL 12 <MR, BP, CR> 27.4% ESL 13 <MR, TO, CR> 27.4% ESL 14 <(BP, TO), CR> 27.4% ESL 15 <MR, (BP, TO), CR> 27.4% Table 4 is the sequence-pattern support matrix SPS of cases and the result of similarity measure.…”

Section: Calculating Instancementioning

confidence: 99%

“…The idea of CBR is analogical reasoning, which comes from the cognitive mode when human solving problems: the experiences of solving problems in the past can be used as references to deal with new similar problem [1]. On account of good explainability and operability of CBR, so far it has been successfully applied to many fields, such as emergency preparedness [2], bankruptcy prediction [3], fault diagnosis [4], software design [5], image retrieval [6], stock selection [7] etc.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Two- Step CBR Method Based on Sequential Data

Xiao¹

2016

IJDTA

View full text Add to dashboard Cite

show abstract

Section: Calculating Instancementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Two- Step CBR Method Based on Sequential Data

Xiao¹

2016

IJDTA

View full text Add to dashboard Cite

show abstract

“…A single performance indicator from the AUC can summarise the ROC curve as the higher the AUC value, the better the performance of the method. The credibility in the region of security surveillance means low false alarm rate (FAR) and low missed alarm rate (MAR) of the MAS [17]. The false alarm (FA) means that the MAS is triggered when it should not be triggered, while the missed alarm (MA) means that the MAS is not triggered when it should be triggered [18].…”

Section: Introductionmentioning

confidence: 99%

A Method for Selecting Optimal Number of Sensors to Improve the Credibility

Chen¹,

Yu²,

et al. 2016

Journal of Sensors

View full text Add to dashboard Cite

With the development of sensors, it is possible to embed many sensors within a certain space, which makes the monitor and alarm system with multisensor possible. There are two important parameters in a monitor and alarm system, namely, the false alarm rate and the missed alarm rate. In this work, a method for selecting optimal number of sensors in the sensor array is presented to improve the credibility. The influence factors of the weights and the false alarm rate and the missed alarm rate of one sensor and total number of sensors are discussed. An experimental setup was developed. The monitoring methods of common strategies and the proposed optimal number of sensors strategy are compared graphically by the receiver operating characteristic curves and the area under receiver operating characteristic curve values. The receiver operating characteristic curves graphically prove that the optimal number of sensors’ method presents the best performance, and it is shown that the optimal number of sensors’ method has the highest area under receiver operating characteristic value (0.9631). This method may aid future users of the monitor and alarm system by providing an optimal number of sensors strategy for high credibility.

show abstract

“…The common method is to calculate weighted average of history solutions according to similarity. It is also a good way to extract rules for case revise by data mining technique [15], and the rules are added to each case as a special attribute, but this will increase the burden on database storage. Besides that, the algorithm like expert system and neural networks are introduced to establish revise models based on training data, but it needs enough history data with high similarity [16].…”

Section: Introductionmentioning

confidence: 99%