An integrated PMU architecture for power system applications

Time synchronized phasors obtained using Phasor Measurement Units (PMU) spread across wide areas have revolutionized power system monitoring and control. These synchronized measurements must be accurate and fast in order to comply with the latest IEEE standards for synchrophasor measurements. The speed at which a PMU provides an output depends on the group delay associated with that PMU and the permissible group delay in-turn decides the utility of a PMU for either control or measurement application. Based on the group delay compensation techniques, in the literature, two individual types of PMUs, such as causal and non-causal PMUs have been introduced. This paper presents an approach where both causal and non-causal PMUs are combined in an integrated PMU architecture. This method not only illustrates the group delay performance of two PMUs in a single module, but also can be used for multiple functions. In this environment several PMU algorithms have been compared with respect to their group delays and their effect on the response time. Application of the integrated PMU architecture to a four-machine 10-bus power system has been demonstrated using a six-input PMU with three-phase voltage and current signals as inputs. Different causal compensation schemes are introduced due to the availability of voltage and current-based frequency and ROCOF signals. Impact of these compensation schemes on PMU accuracy is evaluated through the Total Vector Error (TVE) index. The influence of these compensation schemes on measurements like power and impedance is also investigated. Finally, outputs from the integrated PMU architecture are fed into a Power System Stabilizer (PSS) to control the small-signal stability performance of a power system during dynamic conditions.

Non-Intrusive Load Monitoring Based on Deep Pairwise-Supervised Hashing to Detect Unidentified Appliances

Non-intrusive load monitoring (NILM) is a fast developing technique for appliances operation recognition in power system monitoring. At present, most NILM algorithms rely on the assumption that all fluctuations in the data stream are triggered by identified appliances. Therefore, NILM of identifying unidentified appliances is still an open challenge. To pursue a scalable solution to energy monitoring for contemporary unidentified appliances, we propose a voltage-current (V-I) trajectory enabled deep pairwise-supervised hashing (DPSH) method for NILM. DPSH performs simultaneous feature learning and hash-code learning with deep neural networks, which shows higher identification accuracy than a benchmark method. DPSH can generate different hash codes to distinguish identified appliances. For unidentified appliances, it generates completely new codes that are different from codes of multiple identified appliances to distinguish them. Experiments on public datasets show that our method can get better F1-score than the benchmark method to achieve state-of-the-art performance in the identification of unidentified appliances, and this method maintains high sustainability to identify other unidentified appliances through retraining. DPSH can be resilient against appliance changes in the house.