Image-based mains signal disaggregation and load recognition

Abstract

The mains signal is a complex fusion of various electrical equipment load signals in a building. In the non-intrusive load monitoring recognition, our main aim is to be able to extract as much load features as possible from the complex aggregate mains signal in a simpler way through a computer vision-based approach as opposed to the powers series signal approach. Power series methods, which are one dimensional in nature, suffer from poor aggregate and load signal feature localization necessitating a larger training dataset spanning very long time periods and normally require signal formatting and pre-processing. We use Gramian angular summation fields to transform the power series into a reduced image dataset that contains a rich set of localized signal features. A computer vision approach allows us to capture as much information as possible, and then propose an image-based mains load recognition system with high performance. In this paper for the entire recognition system, we use convolutional neural networks that very well adapted to vision recognition. The load signal image disaggregation is achieved through the powerful stacked denoising autoencoder noise extraction network. To test the proposed system, some simulations and comparisons are carried out and the results show that our easier to handle method can achieve acceptable performance.

Scattering Transform for Classification in Non-Intrusive Load Monitoring

“…However, the method in [27] needed a multifeature method to transform the 1D load data into 2D matrix data. Authors in [28] proposed a 2D CNN structure that recognizes the load status. The proposed method used Gramian angular fields (GAFs) for encoding appliance low-frequency power series (1D) to an image (2D).…”

Differential evolution algorithm with population knowledge fusion strategy for image registration

“…Image registration is a complex task in image processing, which refers to match different images of the same scene taken at multiple times, in multiple viewpoints or with multiple sensors [21,22]. Remote sensing image registration methods proposed in literature consist of two categories: feature-based registration methods and intensity-based registration methods.…”

Time-Lapse Image Method for Classifying Appliances in Nonintrusive Load Monitoring

“…Additionally, the results of the GAF with the denoising autoencoder (DAE) have shown better classification accuracy than the factorial hidden Markov model (FHMM) [18]. Recently, the GAF method has demonstrated good performance in combination with the stacked denoising autoencoder (sDAE) in noisy environments using real datasets [19], and it has also been utilized for short-term load forecasting [20]. However, studies using the GAF [17][18][19][20] have been simulated with only one type of GAF: the GADF or GASF transformation.…”

“…Recently, the GAF method has demonstrated good performance in combination with the stacked denoising autoencoder (sDAE) in noisy environments using real datasets [19], and it has also been utilized for short-term load forecasting [20]. However, studies using the GAF [17][18][19][20] have been simulated with only one type of GAF: the GADF or GASF transformation. The GAF technique may not typically use regression information of time-series data, which leads to the difficulty of derivations for operation patterns of appliances.…”

“…Previous works have been primarily focused on transforming the transition points of energy consumption and classifying the appliances with features of images [17][18][19][20][21]. However, these methods cannot represent all changes in the appliance.…”