Search

The accuracy of machine learning-based power outage prediction models (OPMs) is sensitive to how well event severity is represented in their training datasets. Unbalanced or overly dispersed event severity can result in random errors in outage predictions and underestimation in severe events or overestimation in weak ones. To improve accuracy in the prediction of storm-caused power outages, we introduce a novel method called “Conditioned OPM” that divides an OPM training dataset into subsets of events representative of the predicted event's severity by calculating the quantile weight distance (QWD) between severe weather-related events in the dataset and the predicted event. Based on 102 storm events (including two hurricanes, Irene and Sandy), that have occurred since 2005 over Eversource Energy's Connecticut service territory, we quantified the weather differences among predicted events, which we classified into three groups of severity: low, moderate, and high. The Conditioned OPM creates a subset of the historical events based on their classified severity group and uses that subset as the training dataset to predict the power outages. The study shows that the accuracy of event severity classification was 0.76, and the mean absolute percentage error (MAPE) decreased by about 30%; this method was also tested on forecast events and exhibited a low (20%) MAPE.

In this manuscript, we discuss the importance of using accurate weather forecasts based on remote sensing observations for improving the accuracy of electric distribution grid power outage prediction. Weather variable accuracy is critical to limit the significant power outage modeling errors arising from the strong non-linearities in the relationships between hydrometeorological variables and power outages. We consolidate recent findings related to the most important variables for outage prediction, and to the propagation of their uncertainties in the University of Connecticut Outage Prediction Model (UConn-OPM). UConn-OPM is an operational machine learning-based framework predicting power outage in electric distribution networks. Combining remote sensing datasets with weather analysis used for OPM training, and post-storm verification, would directly contribute to improving the OPM prediction accuracy.

In this paper, a novel complex chaotic system is constructed. The characteristics of new complex chaotic system are analyzed by symmetric, dissipative and stability, then dynamical performances studied using bifurcation diagram, Lyapunov exponent spectrum and complexity. On the basis of this, an image diffusion algorithm is proposed based on the model of law of gravity. Security performances of the proposed algorithm are researched through the key space, statistics, information entropy, noise attack. The experimental results illustrates that the novel complex chaotic system has abundant dynamical behaviors and large parameters range, the security analysis shows that the proposed image encryption algorithm possess higher security features. Therefore, the research will provide theoretical guidance and experimental basis for chaotic secure communication and information security.

The performance of axial-flow compressors is seriously affected by inlet distortion due to a decrease of stability margin. To increase the operational range of compressors, the inlet flow distortion has attracted increasing attention. This paper describes the stall detection based on Deterministic Learning (DL) theory to predict the onset of flow instability for an axial compressor operating in the inlet flow distortion. The inlet flow distortion with plugboard was simulated. Experiments are conducted on a low-speed axial flow compressor test rig of Beihang University at different rotating speeds. Firstly, by installing high response dynamic pressure transducers arranged circumferentially around the casing of the axial compressor, the dynamic data for inlet flow distortion are collected. Secondly, based on deterministic learning theory, the system dynamics underlying stall inception patterns are identified. Finally, based on modeling results, rapid detection of small oscillation faults is used to perform the detection of stall precursors. Results show that this approach successfully detects inception signal of rotating stall under inlet distortion in the compressor test rig.

In this paper, we investigate the problem of human gait recognition based on temporal data sequences by utilizing deterministic learning theory. We employ joint-angle trajectories of lower limbs for gait recognition. The joint angle data generated from human gait locomotion is chosen as the temporal data since it represents the dynamic characteristics of human gait and contains lots of valuable information for gait recognition. What's more, the joint angle trajectories are periodic or periodic-like (recurrent) which makes the radial basis function (RBF) network easily satisfy the partial persistence of excitation (PE) condition. Firstly, discrete-time joint angle data obtained by motion-capture equipment or image-processing algorithms forms the temporal data sequences generated from human gait locomotion, locally-accurate approximation of the underlying gait system dynamics is achieved by using RBF networks. We then prove the convergence of the approximation error and related parameters. Consequently, the joint angle data sequences can effectively represent human gait locomotion by using the knowledge of approximated gait dynamics which is kept in constant RBF networks. Finally, similarity definition for temporal gait data sequences generated from different persons or from different status of one person is given, and a method for recognition of gait temporal data sequences is proposed. We use less complicated simulation examples of compass-like biped robots gait recognition to demonstrate the effectiveness of our schemes.

Rotating stall is one of aerodynamic instabilities in which reduced flow rate gives rise to flow separation and the formation of stall cells, causing enormous vibratory stresses in the blades to limit the performance of compressors. When stall is imminent, an easily recognizable flow breakdown process, known as spike-type stall, is observed in most modern compressors. It is more meaningful for active control to capture the emergence of spike-type initial perturbation during transients into rotating stall or surge. Based on the deterministic learning theory, this paper conducts the approximately accurate modeling and rapid detection of spike-type stall precursors for the Mansoux-C3 model. Firstly, based on a high-dimensional ODE model (the Mansoux model), it gives the mathematical expression of the Mansoux-C3 model. According to the analysis of the stall inception process of the Mansoux-C3, the spike-type stall inception is considered to occur. For the spike-type stall, we perform the locally-accurate identification of the system dynamics via deterministic learning theory. Secondly, based on the approximately accurate modeling of the dominant system dynamics, rapid detection for spike-type stall precursors is achieved by rapid detection of small oscillation faults. Simulation studies show that the detection for rotating stall via deterministic learning is an effective method for the spike-type stall.

Blood pressure (BP), the pressure exerted by circulating blood upon the walls of blood vessels, is an important physiological parameter and can provide some information for personal healthcare. There are two ways to measure blood pressure of human being, invasive and noninvasive methods. Invasive method can measure BP continuously and accurately, however, it is inconvenient to operate and may cause infection. Although traditional noninvasive methods such as Korotkoff and Oscillometric are easy to obtain BP value, they all need a cuff and not very convenient. Therefore, in this paper, we implement a cuff-less continuous blood pressure monitoring device, which is consisted of two acquisition nodes and an Android smart-phone. The data can be transmitted by a Bluetooth module in the system. The pulse transit time (PTT) can be obtained by using the ECG and pulse wave signals, which were synchronously acquired by our designed system, for estimating blood pressure. Our proposed method can be used for non-constrained and continuous blood pressure monitoring.

ECG and pulse signal are important human physiological signals, where the information of human cardiovascular system can be deduced. The analysis of ECG and pulse signal is great significant for prevention and detection of cardiovascular diseases. Traditional pulse signal acquisition systems only can acquire the unilateral pulse signal. However, the pulse signals in two arms could be different due to the different geometries of two arms' blood vessels, the difference of vascular elasticity and the characteristics of surrounding tissues, and so on. In this paper, multi-channel signals acquisition system based on C8051f410 microcontroller is proposed. The single channel ECG signal and dual channel pulse signals can be synchronously acquired, displayed and analyzed. This system can be used for future study on the diseases related to the different structures and functions of bilateral vascular systems.

Semiconductor manufacturers provide low-cost, low-power and short-range solutions for the application of Wireless Sensor Network (WSN), such as ZigBee, Bluetooth, Wi-Fi etc. The quality of radio frequency signal, which carries information in wireless communication channel, is the key to WSN. But it is easily influenced by the external environment. Received Signal Strength Indicator (RSSI) is a measurement for the quality of radio frequency signal which is often used in field of space localization. This paper comprehensively analyses the characteristics of RSSI possibly influenced by the factors such as temperature, the height of sensor node's position, the type of antenna and the electromagnetic interference of human body. We found that the attenuation of Received Signal Strength (RSS) varies about 5.0 dBm with the change of temperature in every 10 centigrade. The ground can reduce the radiation area of wireless signal, so that RSS is dramatically influenced by the height of its position. Patch antenna causes the attenuation of RSS about 11.2 dBm larger than helix antenna in the distance of 25 m. If the sensor node is wore in human body, the posterior position leads to the attenuation of RSS about 13.0∼16.2 dBm larger than anterior position in a specified distance range due to the electromagnetic interference of human body. Finally, some improvement devices were raised here for the WSN in the application of home healthcare according to the qualified effects of various environmental factors.

An improved 2D weighted centroid localization algorithm for Wireless Sensor Networks (WSNs) is proposed. Compared with most previous approaches, the improved algorithm uses a optimized threshold to restrict the number of anchors involved. Simulation results show that our improved weighted centroid localization algorithm increased the positioning accuracy by at least 15% than traditional Monte Carlo Localization (MCL) algorithm. Furthermore, the algorithm is extended from 2D to 3D and comparison with MCL method both in accuracy and in efficiency are carried out as well, showing that 3D-WCL algorithm has a high accuracy about 13% localization average error significantly with little hardware cost, which is suitable for practical application, such as medical surveillance in hospital etc.

In this paper, a QoS-guaranteed cross-layer resource allocation algorithm for multiclass services in downlink LTE system is proposed, which takes EXP rule, channel quality variance, real-time services and non-real-time services and minimum transmission rate into account. The key features of the proposed algorithm are that all the users and resource blocks will be allocated step by step respectively. Numerical results demonstrate that the proposed algorithm effectively guarantees the user QoS requirement for multiclass services. At the same time, it mainly maintains the throughput and fairness performances in a high level.