In this paper, we propose a novel person specific fall detection system based on a monocular camera, which can be applied for assisting the independent living of an older adult living alone at home. A single camera co...
详细信息
In this paper, we propose a novel person specific fall detection system based on a monocular camera, which can be applied for assisting the independent living of an older adult living alone at home. A single camera covering the living area is used for video recordings of an elderly person's normal daily activities. From the recorded video data, the human silhouette regions in every frame are then extracted based on the codebook background subtraction technique. Low-dimensionality representative features of extracted silhouetted are then extracted by convolutionalneuralnetwork-based autoencoder (CNN-AE). Features obtained from the CNN-AE are applied to construct an one class support vector machine (OCSVM) model, which is a data driven model based on the video recordings and can be applied for fall detection. From the comprehensive experimental evaluations on different people in a real home environment, it is shown that the proposed fall detection system can successfully detect different types of falls (falls towards different orientations at different positions in a real home environment) with small false alarms.
ABSTRACTExisting researches predominantly focus on optimizing time-of-use tariffs for single-type consumers, overlooking the fact that an actual distribution system comprises multi-type consumers with distinct power c...
详细信息
ABSTRACTExisting researches predominantly focus on optimizing time-of-use tariffs for single-type consumers, overlooking the fact that an actual distribution system comprises multi-type consumers with distinct power consumption behaviors and preferences. To better cater to the diverse needs of different consumer types, this study proposes an optimization method for multi-type consumers of time-of-use tariffs using price-based demand response and a consumer classification model. Firstly, a joint classification model for multi-type consumers is constructed based on convolutional neural network autoencoder and hierarchical clustering algorithm. Then, by applying the constructed model to historical power consumption data, consumers are classified into several groups. Secondly, a fuzzy heuristic algorithm is introduced to optimize time-of-use tariffs for these consumer groups. Through fuzzy reasoning, the optimal tariffs during different time periods are determined for each group using the fuzzy heuristic algorithm. The effectiveness and adaptability of the proposed method is verified through an application example involving time-of-use tariff optimization for an actual distribution network. Simulation results demonstrated that the maximum power demands for consumers powered by the power company reduced by 6%~9%. The load rate for the studied power company increased from 0.77 to 0.81 ~ 0.84. In addition, compared with two latest methods, when the proposed method was used for ToU tariff optimization, reduction rate of the maximum power demand was improved by 0.75%~2.05% and 0.98%~2.52%, respectively. Load rate for the power company was improved by 0.01 ~ 0.03 and 0.02 ~ 0.03.
暂无评论