TL;DR
This paper introduces a novel fault-tolerant method for pinpointing the source of transient voltage disturbances (TVDS) in smart grids, even with the complexities of distributed generation (DG). By fusing information from multiple power quality monitors (PQMs) and accounting for their individual reliability, the approach significantly improves location accuracy and robustness compared to existing methods.
What Problem Does This Paper Address?
Smart grids are increasingly complex, with a growing number of distributed generators (DGs) and a vast network of power quality monitors (PQMs). Transient voltage disturbances (like sags and swells) are common and can disrupt sensitive loads. Accurately locating the source of these disturbances is crucial for identifying causes, assigning responsibility, and implementing effective mitigation strategies. Traditional methods, often relying on simple direction judgments from PQMs, struggle with the unreliability of individual monitors, data errors, and the significant impact of DGs on power flow, leading to incorrect locations. This paper aims to address this by developing a more robust and fault-tolerant TVDS localization method.
Methods
The core of the proposed approach is Information Fusion (IF), a technique that combines data from multiple sources to make more reliable decisions. The authors introduce several key innovations:
- Credibility Factors: They define three factors to quantify the reliability of direction judgments from each PQM. These include:
- Disturbance Intensity: How strong the disturbance signal is at the PQM.
- Fluctuation Characteristics: How the disturbance signal changes over time, affecting its stability.
- State Estimation: The accuracy of data from "virtual" PQMs (derived from estimation when actual monitors are missing).
- DG Influence Rules: Two specific rules are proposed to analyze how DG integration affects direction judgments. These rules help identify when a DG might cause a misjudgment and how to correct for it.
- Fusion Credibility: Using Dempster-Shafer evidence theory, the method fuses the credibilities from different PQMs and different criteria (like disturbance power and disturbance energy) to arrive at a more robust overall credibility score.
- Conflict Weakening: A strategy is implemented to reduce the influence of conflicting direction judgments from different PQMs, preventing them from unduly affecting the final location.
- Verification and Correction: A specific process is designed to verify and correct the initial location result, especially when the TVDS is located on a line connected to the main grid and a DG.
Datasets
The proposed approach was validated using a simulation model of the IEEE 13-node test feeder. This is a standard benchmark system in power systems research, modified to include a distributed generator (DG) and simulated transient voltage disturbance events. The paper does not introduce a new dataset but uses a well-established testbed to demonstrate its method's effectiveness.
Key Insights and Ideas
- Individual PQM Reliability Matters: The accuracy of TVDS localization is not just about having many PQMs, but about understanding the reliability of each one. Factors like signal strength and signal stability significantly impact how trustworthy a PQM's judgment is.
- DGs are Game-Changers: Distributed generators fundamentally alter power flow dynamics, making traditional direction-finding algorithms prone to errors. Explicitly accounting for DG influence is essential for accurate localization.
- Fusion is Key to Robustness: Combining information from multiple, diverse sources (PQMs with varying reliability, different disturbance analysis methods) through information fusion dramatically improves fault tolerance.
- Conflict Isn't Always Bad, But Needs Management: Conflicting information from PQMs can arise. The "conflict weakening" strategy helps manage these conflicts, ensuring they don't disproportionately skew the final result.
- Virtual PQMs Can Help, But With Caution: When actual PQMs are unavailable, state estimation can provide data, but the inherent estimation errors must be factored into the credibility assessment.
Key Results
The proposed IF-based approach demonstrated superior performance compared to traditional matrix-based (TM) and disturbance measuring (DM) methods.
- High Accuracy: In simulations involving direction misjudgments, the proposed method achieved a 93% accuracy rate, significantly outperforming the TM method (20%) and the DM algorithm (75%).
- Fault Tolerance: The method successfully located TVDS even when multiple direction misjudgments occurred, a scenario where traditional methods failed.
- DG Integration Handling: The approach effectively handled scenarios with DG integration, correctly identifying and adjusting for misjudgments caused by these sources.
- Reliability Evaluation: The paper also introduced a reliability coefficient (H) to quantify the confidence in the location result, providing an additional layer of insight.
Limitations and Future Work
- Topology Limitations: While the proposed method is robust for tree-like distribution grids, further investigation is needed for more complex topologies (e.g., circular grids) and multiple DG configurations.
- Parameter Tuning: The parameters used in the conflict weakening strategy (30% and 20%) were determined empirically and may require adjustment for different practical scenarios.
- Future Work: The authors suggest extending the IF-based location strategy to arbitrary topologies and multiple DG integrations. This will involve developing more sophisticated influence rules and correction strategies for these complex scenarios, potentially through further theoretical analysis, simulation experiments, and field tests.
Summary updated