A Fusion Network with Stacked Denoise Autoencoder and Meta Learning for Lateral Walking Gait Phase Recognition and Multi-Step-Ahead Prediction.

Lateral walking gait phase recognition and prediction are the premise of hip exoskeleton application in lateral resistance walk exercise. In this work, we presented a fusion network with stacked denoise autoencoder and meta learning (SDA-NN-ML) to recognize gait phase and predict gait percentage from IMU signals. Experiments were conducted to detect the four lateral walking gait phases and predict their percentage in 10 healthy subjects across different speeds. The performance of SDA-NN-ML and other widely used algorithms including Support Vector Machine (SVM), Adaptive Boosting (AdaBoost) and Long Short Term Memory (LSTM) were evaluated. The cross-subject recognition accuracy of SDA-NN-ML (89.94%) decreased by 4.62% compared to the training accuracy, which outperformed SVM (8.60%), AdaBoost (5.61%), and LSTM (7.12%). For real-time and cross-subject prediction of gait phase percentage, the RMSE of SDA-NN-ML (0.2043) outperformed that of a single regression network (0.2426). With a signal noise ratio of 100:30, the cross-subject recognition accuracy decreased by a mere 5.70%, while the prediction result (RMSE) of SDA-NN-ML increased by 0.0167 when compared to the noise-free results. SDA-NN-ML demonstrates a stable multi-step-ahead prediction ability with an accuracy higher than 82.50% and an RMSE of less than 0.23 when the ahead time is less than 200 ms. The results demonstrated that the proposed method has high accuracy and robust performance in lateral walking gait recognition and prediction.

Cellulose ultrafine fibers embedded with titania particles as a high performance and eco-friendly separator for lithium-ion batteries.

Mixtures of cellulose acetate (M.W. ∼3 × 104 g/mol) dissolved in 75% v/v acetic acid in water (17% w/w) and ground anatase titania particles with diameters of 197 ±â€¯75 nm (0%, 5% and 10% w/w) were electrospun at 17 kV with a fiber collection distance and a feed rate of 10 cm and 0.6 mL/h. Then, the fiber was treated with 0.5 M potassium hydroxide in ethanol. Rough regenerated cellulose (RC)-titania separators with diameters of ∼310 nm and uniformly dispersed titania particles showed ∼78% porosities. They decomposed at 300 °C, higher than the decomposition temperature of polyethylene separators (220 °C). Added titania particles increased the electrolyte wettability and lithium transference number (from 0.22 to 0.62). RC - 10% titania separator retained the capacity with 79 mA h/g after 30 cycles and had excellent discharge capacity. These fascinating properties make RC-titania separator promising for lithium ion battery.

Three-dimensional reconstruction of a solid-oxide fuel-cell anode.

The drive towards increased energy efficiency and reduced air pollution has led to accelerated worldwide development of fuel cells. As the performance and cost of fuel cells have improved, the materials comprising them have become increasingly sophisticated, both in composition and microstructure. In particular, state-of-the-art fuel-cell electrodes typically have a complex micro/nano-structure involving interconnected electronically and ionically conducting phases, gas-phase porosity, and catalytically active surfaces. Determining this microstructure is a critical, yet usually missing, link between materials properties/processing and electrode performance. Current methods of microstructural analysis, such as scanning electron microscopy, only provide two-dimensional anecdotes of the microstructure, and thus limited information about how regions are interconnected in three-dimensional space. Here we demonstrate the use of dual-beam focused ion beam-scanning electron microscopy to make a complete three-dimensional reconstruction of a solid-oxide fuel-cell electrode. We use this data to calculate critical microstructural features such as volume fractions and surface areas of specific phases, three-phase boundary length, and the connectivity and tortuosity of specific subphases.