A non-intrusive load monitoring algorithm based on real-time feature extraction and deep learning model

Overview of non-intrusive load monitoring and identification techniques

In this paper, a new method of non intrusive load monitoring (NILM) feature extraction is proposed.The method is based on the analysis and identification of the load transient state and steady state characteristics of the load in the power supply and power outages using the back propagation neural network (BP).The current signal through current sensor is detected in the main switchboard NILM, compared with detecting active power, reactive power and harmonic components of power signals and load the instantaneous change of the state traditional NILM, this new method is more convenient and reduces the amount of computation.The new NILM method integrates artificial intelligence identification technology and load current acquisition technolog.

Knowing the power consumption of individual household appliances is useful for end-user as well as utilities. There are two ways for appliance load monitoring (ALM), namely intrusive load monitoring (ILM) and non-intrusive load monitoring (NILM). This paper focuses on the NILM approach, and discusses a simple yet effective method to improve its accuracy by constructing a better knowledge-base. The proposed methodology is initially verified with the simulation using the Reference Energy Disaggregation Data (REDD) dataset, and later tested on a lab-scale hardware setup as well. Test results reveal that careful construction of knowledge-base can increase the performance of NILM algorithms. MATLAB is used as the programming platform.

What we measure, we can improve. In accordance to this approach, Indian Institute of Science (IISc), Bangalore has developed a Micro-grid Monitoring System in the campus through the installation of Smart Meters, covering almost 250 nodes including substations, centers, departments, administration, hostels and other utilities. This will help the institute in various ways such as capacity planning, substation loading, phase imbalance correction, over-voltage monitoring, billing and so on. Smart Meters measure the power consumption at a single point in the building giving a picture of the energy consumption of the building as a whole. It is necessary to also understand the scenario of the constituent loads at the point where the smart meter is installed so that ways could be found to reduce consumption. Personalised, concise and reliable feedback providing appliance level breakdown of energy consumption in the premises is the key in implementing energy efficiency programs. Taking this into consideration the area of Non Intrusive Load Monitoring (NILM) was explored. In NILM the aggregate smart meter data is separated into constituent loads by machine learning techniques. The NILM system is trained through previous data sets and then the algorithm will disaggregate the total power into individual appliances based on its experience. A benchmark NILM algorithm called Factorial Hidden Markov Model was used for proper load disaggregation. Finally an attempt was made to develop a Smartphone app to visualize results and bring the data to the people.

Although a diversity of approaches towards Non-Intrusive Load Monitoring (NILM) are now available, complementary studies must be carried out in order to improve load modeling and consequently increase load recognition accuracy. This paper presents a study on active power appliances' modeling when only low-frequency granularities are considered. The approach analyzes the distribution of power changes in finite-state loads using an effective feature transform to stabilize power variance. Subtractive clustering method is utilized for detection of power transition recurrences. It allows identifying most likely power changes presented in individual load profiles using a prior knowledge of power uncertainties. The principal aim of this paper is to create a probabilistic device-specific model of household resistive loads, which can be used in event-based load recognition. The results using real data prove the efficacy of the proposed method.

The fog has different features and effects within every single environment. Detection whether there is fog in the image is considered a challenge and giving the type of fog has a substantial enlightening effect on image defogging. Foggy scenes have different types such as scenes based on fog density level and scenes based on fog type. Machine learning techniques have a significant contribution to the detection of foggy scenes. However, most of the existing detection models are based on traditional machine learning models, and only a few studies have adopted deep learning models. Furthermore, most of the existing machines learning detection models are based on fog density-level scenes. However, to the best of our knowledge, there is no such detection model based on multi-fog type scenes have presented yet. Therefore, the main goal of our study is to propose an adaptive deep learning model for the detection of multi-fog types of images. Moreover, due to the lack of a publicly available dataset for inhomogeneous, homogenous, dark, and sky foggy scenes, a dataset for multi-fog scenes is presented in this study (https://github.com/Karrar-H-Abdulkareem/Multi-Fog-Dataset). Experiments were conducted in three stages. First, the data collection phase is based on eight resources to obtain the multi-fog scene dataset. Second, a classification experiment is conducted based on the ResNet-50 deep learning model to obtain detection results. Third, evaluation phase where the performance of the ResNet-50 detection model has been compared against three different models. Experimental results show that the proposed model has presented a stable classification performance for different foggy images with a 96% score for each of Classification Accuracy Rate (CAR), Recall, Precision, F1-Score which has specific theoretical and practical significance. Our proposed model is suitable as a pre-processing step and might be considered in different real-time applications.

Non-intrusive load monitoring (NILM) is to estimate individual appliance’s power consumption from aggregated smart meter data, which is useful for optimized energy management and provisioning of customized services. While deep learning (DL) has achieved state-of-the-art NILM performance, it is still constrained by the dependency on large amounts of data and intensive computations on training. In this paper, we propose a pre-training approach to address the generalization of DL models for NILM. We develop a meta-learning based approach and an ensemble learning based approach, which pre-train a base model and then fine-tune it with few-short learning when applied to an unknown dataset. The models are validated with two real-world datasets and shown to achieve a superior transferability performance compared with traditional DL and transfer learning methods.

Fault diagnosis of power distribution networks is a significant task for power system operation and maintenance, which can support rapid and accurate identification, localisation and repair of various faults that may occur in power distribution networks. To improve the accuracy and efficiency of fault diagnosis, this paper proposes a deep adaptive learning model for dealing with the data-driven identification problems in power systems. In the deep adaptive learning model, an adaptive shortcut learning scheme is designed to adaptively adjust the aggregation between the convolutional layers and the attention modules. In this manner, the hidden features can be effectively captured by adaptive learning with the proposed scheme. Therefore, the proposed model can effectively make online decisions on the fault types in high-voltage AC and DC test transmission. The experimental case study and comparison study in this paper demonstrate the reliable performance of the proposed model for online fault diagnosis of power distribution networks, which also show its capability and feasibility for online implementation.

To better respond to people’s demands for multimedia learning, appropriate learn-ing paths should be offered based on their actual learning demands and different knowledge levels. Adaptive online learning model integrates and improves exist-ing learning frameworks to offer a set of knowledge paths that can cater to dif-fer7ent preferences, tastes, and knowledge levels of learners, no need for them to be aware of this. Based on the improved ant colony algorithm, an adaptive learn-ing system model that can satisfy learners’ demands is built herein with reference to the foraging approach of ants to traverse the paths, thereby to find the best learning path, while the classification method for some learning objects can de-termine the search parameters. This innovative approach proposed hereof can help improve learners' academic performance and learning efficiency.

Introduction: Rhegmatogenous retinal detachment (RRD) is one of the most common fundus diseases. Many rural areas of China have few ophthalmologists, and ophthalmologic ultrasound examination is of great significance for remote diagnosis of RRD. Therefore, this study aimed to develop and evaluate a deep learning (DL) model, to be used for automated RRD diagnosis based on ophthalmologic ultrasound images, in order to support timely diagnosis of RRD in rural and remote areas. Methods: A total of 6,000 ophthalmologic ultrasound images from 1,645 participants were used to train and verify the DL model. A total of 5,000 images were used for training and validating DL models, and an independent testing set of 1,000 images was used to test the performance of eight DL models trained using four different DL model architectures (fully connected neural network, LeNet5, AlexNet, and VGG16) and two preprocessing techniques (original, original image augmented). Receiver operating characteristic (ROC) curves were used to analyze their performance. Heatmaps were generated to visualize the process of the best DL model in the identification of RRD. Finally, five ophthalmologists were invited to diagnose RRD independently on the same test set of 1,000 images for performance comparison with the best DL model. Results: The best DL model for identifying RRD achieved an area under the ROC curve (AUC) of 0.998 with a sensitivity and specificity of 99.2% and 99.8%, respectively. The best preprocessing method in each model architecture was the application of original image augmentation (average AUC = 0.982). The best model architecture in each preprocessing method was VGG16 (average AUC = 0.998). Conclusion: The best DL model determined in this study has higher accuracy, sensitivity, and specificity than the ophthalmologists’ diagnosis in identifying RRD based on ophthalmologic ultrasound images. This model may provide support for timely diagnosis in locations without access to ophthalmologic care.

A feature-free method for testing adaptive network learning models is presented. The test is based on a property called the mean matching law, a. property shared by many adaptive network models of learning. As an application of this method, we prove that cue competition effects obtained with statistically independent cues cannot be explained by many previous adaptive network learning models, including those based on the delta learning rule. These results point to the need to incorporate competitive learning properties into adaptive network learning models.

A requirement of Smart Grids is the ability to predict the energy consumption patterns of their users. In the residential domain, this is usually not feasible due to the inability of the grid to dialog with (legacy) domestic appliances. To overcome this issue Non Intrusive Load Monitoring (NILM) was introduced, a task in which a predictor is used to disaggregate household power consumption. Many of the newer approaches make use of Neural Networks to accomplish this task, due to their superior ability to detect patterns in temporal (thus sequential) data. These models unfortunately require a huge amount of data to achieve good performance, and have the tendency to overfit the training data, making them difficult to predict future consumptions. For these reasons, adapting them to optimally predict a (future) house’s consumption requires expensive and often prohibitive data collection phases. We propose a solution in the form of a neuro-symbolic framework that refines neural network predictions via a constrained optimization problem modelling the characteristics of the appliances of a house. This combined approach achieves superior performance with respect to the neural network alone over two out of five appliances and comparable results for the remaining ones, without requiring further training data.

Due to the increase in electricity demand and the rapid depletion of fossil fuels, energy management has become a critical issue over the last two decades. Thus, researchers, utility suppliers, governments, and policymakers are working in tandem to develop novel solutions. In recent years, solutions based on Intrusive Load Monitoring (ILM) and Non-intrusive Load Monitoring (NILM) have garnered the interest of many researchers. However, NILM systems are less difficult to implement and more cost-effective than ILM systems. Even though available NILM-based solutions can identify single-state devices with acceptable accuracy, identifying the various operating states of multi-state devices remains a problem. This research work proposes a novel supervised learning algorithm to correctly identify the operating states of multi-state residential devices. Results obtained through extensive simulations indicate that the proposed algorithm can achieve device and state identification accuracy of 93 percent and 91 percent, respectively.

International audience

The performance of non-intrusive load monitoring (NILM) system is closely related to the uniqueness and comprehensiveness of the load features extracted from the equipment. However, existing feature extraction methods for NILM are difficult to extract interpretable intuitive features representing the uniqueness of load, and lack of comprehensive attention to the features on different time scales. Therefore, this paper proposes a multi-time-scale shapelet based feature extraction framework. First, shapelet, the most unique segment of a sequence, which can distinguish it from other sequences is introduced as a feature and multi-time-scale current shapelets are proposed to extract comprehensive characteristics of electric appliances and offer intuitive interpretability. Meanwhile, a shapelet transform is constructed, so that the essence of shapelet based NILM is transformed from sequence matching to a general classification problem. Next, an ensemble bagging classifier is proposed on multi-time scales to effectively balance the classification errors on each time scale and enhance the performance of NILM system. In addition, this framework can support small sample training to facilitate rapid deployment of NILM applications at the initial practical stage. The results show that this framework can effectively focus on the uniqueness of load on different time scales, provide interpretability for NILM, and achieve good classification effects.