The randomness and volatility of wind power greatly affect the safety and economy of the power systems, and the wake effect of the wind farm aggravates the wind energy loss and the wind power fluctuation. Taking into consideration the wake effect of the wind farm, a new coordinated wind power smoothing control strategy for multi-wind turbines (M-WT) and energy storage systems (ESS) is proposed. The proposed method is based on a multi-agent deep reinforcement learning (MADRL), in which the relationship between output power and wake effect is firstly analyzed, and a power smoothing control model of the M-WT and ESS is established. MADRL is then introduced to optimize the power control of M-WT and ESS. In order to further increase the learning and training efficiency, an improved MADRL algorithm based on the partitioned experience buffer and prioritized experience replay is proposed, where the experience buffer is divided into positive, negative, and neutral experiences, and the experiences are sampled according to experience priority. The effectiveness of the proposed strategy is verified on the SimWindFarm platform. The results show that the proposed control strategy can maximize the economic benefits while further smoothing wind power fluctuations and increasing power generation.
Measurement of the ground reaction forces (GRF) during walking is typically limited to laboratory settings, and only short observations using wearable pressure insoles have been reported so far. In this study, a new proxy measurement method is proposed to estimate the vertical component of the GRF (vGRF) from wearable accelerometer signals. The accelerations are used as the proxy variable. An orthogonal forward regression algorithm (OFR) is employed to identify the dynamic relationships between the proxy variables and the measured vGRF using pressure-sensing insoles. The obtained model, which represents the connection between the proxy variable and the vGRF, is then used to predict the latter. The results have been validated using pressure insoles data collected from nine healthy individuals under two outdoor walking tasks in non-laboratory settings. The results show that the vGRFs can be reconstructed with high accuracy (with an average prediction error of less than 5.0%) using only one wearable sensor mounted at the waist (L5, fifth lumbar vertebra). Proxy measures with different sensor positions are also discussed. Results show that the waist acceleration-based proxy measurement is more stable with less inter-task and inter-subject variability than the proxy measures based on forehead level accelerations. The proposed proxy measure provides a promising low-cost method for monitoring ground reaction forces in real-life settings and introduces a novel generic approach for replacing the direct determination of difficult to measure variables in many applications.
9 December 2022 in IEEE Transactions on Transportation Electrification
A hybrid energy storage system (HESS) is adopted to tackle the traction network voltage fluctuation problem caused by high power and large energy demand during the starting and braking of urban rail trains. The system is composed of on-board ultracapacitors and ground lithium batteries, aiming to smooth out the power fluctuation to realize “peak-shaving and valley-filling”. Based on deep reinforcement learning (DRL) online sequence decision, a dynamic power allocation strategy is proposed to improve the energy-saving and voltage stabilization of DC traction networks as well as HESS life protection. Furthermore, to enhance the DRL’s efficiency under time-varying operating conditions, an annealing bias - priority experience replay twin delayed deep deterministic policy gradient algorithm (A-TD3) is proposed to train the replay buffer in DRL. The online learning and optimization strategy is implemented via the mechanism of “trial and error” and “feedback” of the agent. RT-LAB semi-physical real-time simulation systems are adopted to verify the effectiveness of the proposed strategy. Compared with the traditional filtering algorithms and DRL algorithms, the results show that the proposed method converges faster and is more energy saving and stable while effectively protecting the energy storage components.
Rapid and accurate muscle segmentation is essential for the diagnosis and monitoring of many musculoskeletal diseases. As gold standard, manual annotation suffers from intensive labor and high inter-operator reproducibility errors. In this study, deep learning (DL) based automatic muscle segmentation from MR scans is investigated for post-menopausal women, who normally experience a decline in muscle volume. The performance of four Deep Learning (DL) models was evaluated: U-Net and UNet++ and two modified U-Net networks, which combined feature fusion and attention mechanisms (Feature-Fusion-UNet, FFU, and Attention-Feature-Fusion-UNet, AFFU). The models were tested for automatic segmentation of 16-lower limb muscles from MRI scans of two cohorts of post-menopausal women (11 subjects in PMW-1, 8 subjects in PMW-2; from two different studies so considered independent datasets) and 10 obese post-menopausal women (PMW-OB). Furthermore, a novel data augmentation approach is proposed to enlarge the training dataset. The results were assessed and compared by using the Dice similarity coefficient (DSC), relative volume error (RVE), and Hausdorff distance (HD). The best performance among all four DL models was achieved by AFFU (PMW-1: DSC 0.828 ± 0.079, 1-RVE 0.859 ± 0.122, HD 29.9 mm ± 26.5 mm; PMW-2: DSC 0.833 ± 0.065, 1-RVE 0.873 ± 0.105, HD 25.9 mm ± 27.9 mm; PMW-OB: DSC 0.862 ± 0.048, 1-RVE 0.919 ± 0.076, HD 34.8 mm ± 46.8 mm). Furthermore, the augmentation of data significantly improved the DSC scores of U-Net and AFFU for all 16 tested muscles (between 0.23% and 2.17% (DSC), 1.6%–1.93% (1-RVE), and 9.6%–19.8% (HD) improvement). These findings highlight the feasibility of utilizing DL models for automatic segmentation of muscles in post-menopausal women and indicate that the proposed augmentation method can enhance the performance of models trained on small datasets.
29 August 2019 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Freezing of gait (FOG) is an episodic gait disturbance affecting locomotion in Parkinson’s disease. As a biomarker to detect FOG, the Freeze index (FI), which is defined as the ratio of the areas under power spectra in ‘freeze’ band and in ‘locomotion’ band, can negatively be affected by poor time and frequency resolution of time-frequency spectrum estimate when short-time Fourier transform (STFT) or Wavelet transform (WT) is used. In this study, a novel high-resolution parametric time-frequency spectral estimation method is proposed to improve the accuracy of FI. A time-varying autoregressive moving average model (TV-ARMA) is first identified where the time-varying parameters are estimated using an asymmetric basis function expansion method. The TV-ARMA model is then transformed into frequency domain to estimate the time-frequency spectrum and calculate the FI. Results evaluated on the Daphnet Freezing of Gait Dataset show that the new method improves the time and frequency resolutions of the time-frequency spectrum and the associate FI has better performance in the detection of FOG than its counterparts based on STFT and WT methods do. Moreover, FOGs can be predicted in advance of its occurrence in most cases using the new method.