Academia.eduAcademia.edu

Figure 2 – uploaded by FXintegrity Publishing

See full PDFdownloadDownload figure
n [The HTM model leverages a sparse distributed representation by utilizing the HTM neural network which is based on the CLA. This network employs SDRs to encode input data leading to enhanced noise robustness and flexibility in handling symbolic, numeric, and string data without the requirement of distinct training and testing phases. The CLA enables stable predictions by imitating the brain's regenerative learning processes and generates internal activations that mimic the brain's predictive capabilities. These characteristics make the HTM model relevant for fault prediction purposes. It is illustrated in Fig.3 and depicted in Algorithm 2. 5.Model of the Proposed HTM System

Figure 3 n [The HTM model leverages a sparse distributed representation by utilizing the HTM neural network which is based on the CLA. This network employs SDRs to encode input data leading to enhanced noise robustness and flexibility in handling symbolic, numeric, and string data without the requirement of distinct training and testing phases. The CLA enables stable predictions by imitating the brain's regenerative learning processes and generates internal activations that mimic the brain's predictive capabilities. These characteristics make the HTM model relevant for fault prediction purposes. It is illustrated in Fig.3 and depicted in Algorithm 2. 5.Model of the Proposed HTM System

Related Figures (15)

1. Introduction Keywords: Hierarchical Temporal Memory;Partial Discharge;Fault Detection Methods;Transmission Line; Simulation and Descriptive Method.
1. Introduction Keywords: Hierarchical Temporal Memory;Partial Discharge;Fault Detection Methods;Transmission Line; Simulation and Descriptive Method.
2.1 Review Table
2.1 Review Table
2.2 Identified Research Gap
2.2 Identified Research Gap
"ig.1. Visual representation of the flow in the HTM detection proce:
"ig.1. Visual representation of the flow in the HTM detection proce:
(Source:https://www.kaggle.com/c/vsb-power-line-faultdetection/data)
(Source:https://www.kaggle.com/c/vsb-power-line-faultdetection/data)
Table 2: Onitsha-Alaoji Key TL/ Transformer Parameters
Table 2: Onitsha-Alaoji Key TL/ Transformer Parameters
Fig.4. Cortical Learning User-interface Architecture for Fault Detection Analysis in HTM
Fig.4. Cortical Learning User-interface Architecture for Fault Detection Analysis in HTM
Fig.5. Sequence diagram showing Entity Relations of the HTM system
Fig.5. Sequence diagram showing Entity Relations of the HTM system
Table 3:Fault dataset sample
Table 3:Fault dataset sample
Table 4: Pattern Formation based on temporally aggregated data samples The data in the Tables revealed that PDis the fault type that is being detected in this study. It involves faults arising from insulator damage, wire melting, and vegetation growing too close to cable lines which may cause faults. The PD is commonly found as a fault leading to transmission line errors. Kaggle Dataset established the PD fault diagnosis primarily through target and phase variables, where a target variable value of 0 indicates no fault present and 1 implies a potential fault. The phase variable represents the three-phase power line condition
Table 4: Pattern Formation based on temporally aggregated data samples The data in the Tables revealed that PDis the fault type that is being detected in this study. It involves faults arising from insulator damage, wire melting, and vegetation growing too close to cable lines which may cause faults. The PD is commonly found as a fault leading to transmission line errors. Kaggle Dataset established the PD fault diagnosis primarily through target and phase variables, where a target variable value of 0 indicates no fault present and 1 implies a potential fault. The phase variable represents the three-phase power line condition
Fig.7. PSD plots of fault condition at Onitsha-Alaoji TL (fault applied at 40km of total line length) rig.6. PSD plots of pre-fault condition at Onitsha-Alaoji TL
Fig.7. PSD plots of fault condition at Onitsha-Alaoji TL (fault applied at 40km of total line length) rig.6. PSD plots of pre-fault condition at Onitsha-Alaoji TL
Table 5: Results of the proposed HTM sequence learning model
Table 5: Results of the proposed HTM sequence learning model
Table 6: Results of the ELM sequence learning model
Table 6: Results of the ELM sequence learning model
Table 7: Results of the OS-ELM sequence learning model
Table 7: Results of the OS-ELM sequence learning model
Related topics:
Hierarchical Temporal Memory
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved