Abstract:One key task for the planning and organizing of Inspection and Maintenance (I&M) works in an integrated energy system (IES) is to handle its weakest vulnerabilities, that is, when and where in this system the I&M actions or countermeasures ought to be performed. This requires a pinpoint spatiotemporal distribution prediction of these vulnerabilities, then more preparation times can be brought about for utilities to appropriately allocate the limited I&M manpower and equipment. With such a motivation, in light of both the internal and surrounding conditions, this paper proposes a prognosis framework, namely the Importance-Fuzzy High-utility Pattern Identification with lifetime-dependent factors (IFHPIlf). The IFHPIlf aims for the component-vulnerability patterns with higher profits rather than simply higher frequencies. In that case, when applied in imbalanced distributed databases, this framework can incorporate the High-Impact-Low-Probability (HILP) components as well, without the extra steps of separating and assessing rare and common components. For the purpose of assessing the utilities of each component and transaction, a parallel learning architecture is formulated to evaluate the corresponding two attributes, unit price and quantity, respectively. To estimate the unit price, the risk level of each component will be quantitatively rated via the built BP Time-dependent-lifetime Importance (BPTdlI) model, where the direct impacts of each component’s lifetime distribution on the reliability distribution of the whole system, as well as the entire underlying system failure-related hazard cut sets, are integrated. Ergo, the data dependence in temporal scales can be taken into account; To qualitatively differentiate the perilous components, the quantity will be calculated in the established Importance-Fuzzy High-utility Pattern Identification (IFHPI) model, wherein the time-dependent-lifetime important FIS are conjugated with the Fuzzy Interest-Utility Measures (FIUM). Hence, both the discrete and continuous features will be handled in the same entity, and the determinations can be straightforwardly based on their influences. At last, the flexibility and feasibility of this framework during applications are demonstrated in terms of an empirical case study.

Data-driven Approach for Spatiotemporal Distribution Prediction of Fault Events in Power Transmission Systems

Early Warning System for Spatiotemporal Prediction of Fault Events in a Power Transmission System

A Framework for Dynamic Prediction of Reliability Weaknesses in Power Transmission Systems Based on Imbalanced Data

Fault Prediction for Distribution Network Based on CNN and LightGBM Algorithm

A Dynamic Risk-Early-warning Methodology of Distribution System Faults Incorporating Spatiotemporal Imbalanced Data Distributions

Temporal Difference Learning Based Critical Component Identifying Method with Cascading Failure Data in Power Systems

Research on Association Matching Analysis Technology Between Power Grid Faults and Meteorological Disaster Information

An Ensemble System to Predict the Spatiotemporal Distribution of Energy Security Weaknesses in Transmission Networks

Data-Based Line Trip Fault Prediction in Power Systems Using LSTM Networks and SVM.

Cause-Effect Modeling and Spatial-Temporal Simulation of Power Distribution Fault Events

Data-driven Solutions for Power System Fault Analysis and Novelty Detection.

Fault Prediction for Electromechanical Equipment Based on Spatial-Temporal Graph Information

Fault Location and Classification for Distribution Systems Based on Deep Graph Learning Methods

A Physics-Informed Data-Driven Fault Location Method for Transmission Lines Using Single-Ended Measurements with Field Data Validation

Data Mining And Analysis Of Tree-Caused Faults In Power Distribution Systems

Physics-Informed Data-Driven Transmission Line Fault Location Based on Dynamic State Estimation

Exploratory Analysis of Massive Data for Distribution Fault Diagnosis in Smart Grids

A vulnerability spatiotemporal distribution prognosis framework for integrated energy systems within intricate data scenes according to importance-fuzzy high-utility pattern identification

A Transformer Heavy Overload Spatiotemporal Distribution Prediction Ensemble under Imbalanced and Nonlinear Data Scenarios

An Integrated Risk Assessment Model for the Multi-Perspective Vulnerability of Distribution Networks under Multi-Source Heterogeneous Data Distributions

A Risk Warning Method for Urban Distribution Network Based on Associated Recognition of Typical Fault and Environment Scenario