Residential appliance monitoring based on low frequency smart meter measurements

Abstract

A Non-Intrusive Load Monitoring (NILM) method for residential appliances based on uncorrelated spectral components of an active power signal is presented. This method utilizes the Karhunen Loéve (KL) expansion to breakdown the active power signal into subspace components so as to construct a unique information rich appliance signature. Unlike existing NILM techniques that rely on multiple measurements at high sampling rates, this method works effectively with a single active power measurement taken at a low sampling rate. After constructing the signature data base, subspace component level power conditions were introduced to reduce the number of possible appliance combinations. Then, an algorithm was presented to identify the turned on appliance combination in a given time window. After identifying the turned on appliance combination, another algorithm was introduced to disaggregate the energy contribution of each individual appliance. The case study conducted using tracebase public data set demonstrate the ability of the proposed method to accurately identify and disaggregate individual energy contributions of turned on appliance combinations that contain single state, multi state and continuous varying appliances. Finally, the proposed method was modified to accommodate usage behavior patterns of each residence adaptively. The modification was validated using six US households in REDD public database. This significantly improves the convergence speed of the turned on appliances identification process.

Smart Grids Data Analysis: A Systematic Mapping Study

“…C4. Load segregation Non-intrusive appliance load monitoring [35], [66], [92], [130], [147], [207], [219], disaggregate smart home sensor data [135], [144], [260], [267] C5. Power loads / consumption analysis consumption clustering ( [109], [136], [137], [182], [190], [198], [199], [206], [210], [221], [259], [263]), consumption prediction ( [?…”

Non-intrusive load monitoring under residential solar power influx

“…Here, the KLE based spectral feature extraction method was used as unique signature information of active power signals might not be apparent in the time domain profiles due to their low sampling rate. KLE allows the extraction of hidden features of time domain active power signals by utilizing their uncorrelated spectral components [18,23,38]. This process is shown in Figure3. A given SC is then further decomposed to generate more SCs to improve the resolution.…”

Non-Intrusive Load Identification Method Based on Improved KM Algorithm

“…This particular type of load may not have the fixed signatures, thus making some of NILM methods fail to identify these loads. Therefore, several methods are proposed to deal with the modern appliances, such as deep learning-based technique [19], particle filtering method [20] and KL-based expansion [21]. These methods in general can also cope with the following appliances [22], [23]: single state, multi-state, and continuously varying appliances [24], among which multi-state appliances are more common and more complicated to be identified.…”