Pred-SF: A Precipitation Prediction Model Based on Deep Neural Networks.

How to predict precipitation accurately and efficiently is the key and difficult problem in the field of weather forecasting. At present, we can obtain accurate meteorological data through many high-precision weather sensors and use them to forecast precipitation. However, the common numerical weather forecasting methods and radar echo extrapolation methods have insurmountable defects. Based on some common characteristics of meteorological data, this paper proposes a Pred-SF model for precipitation prediction in target areas. The model focuses on the combination of multiple meteorological modal data to carry out self-cyclic prediction and a step-by-step prediction structure. The model divides the precipitation prediction into two steps. In the first step, the spatial encoding structure and PredRNN-V2 network are used to construct the autoregressive spatio-temporal prediction network for the multi-modal data, and the preliminary predicted value of the multi-modal data is generated frame by frame. In the second step, the spatial information fusion network is used to further extract and fuse the spatial characteristics of the preliminary predicted value and, finally, output the predicted precipitation value of the target region. In this paper, ERA5 multi-meteorological mode data and GPM precipitation measurement data are used for testing to predict the continuous precipitation of a specific area for 4 h. The experimental results show that Pred-SF has strong precipitation prediction ability. Some comparative experiments were also set up for comparison to demonstrate the advantages of the combined prediction method of multi-modal data and the stepwise prediction method of Pred-SF.

Numerical Forecast Correction of Temperature and Wind Using a Single-Station Single-Time Spatial LightGBM Method.

Achieving high-performance numerical weather prediction (NWP) is important for people's livelihoods and for socioeconomic development. However, NWP is obtained by solving differential equations with globally observed data without capturing enough local and spatial information at the observed station. To improve the forecasting performance, we propose a novel spatial lightGBM (Light Gradient Boosting Machine) model to correct the numerical forecast results at each observation station. By capturing the local spatial information of stations and using a single-station single-time strategy, the proposed method can incorporate the observed data and model data to achieve high-performance correction of medium-range predictions. Experimental results for temperature and wind prediction in Hainan Province show that the proposed correction method performs well compared with the ECWMF model and outperforms other competing methods.

A fast relay feedback auto-tuning tilt-integral-derivative (TID) controller method with the fractional-order Ziegler-Nichols approach.

The relay feedback auto-tuning method, which was an early commercialized approach, has maintained its popularity due to its simplicity and robustness. However, the classical proportional integral derivative (PID) controller auto-tuning method often results in unacceptable overshoot, especially for integrating and higher-order processes. By integrating the TID controller with the relay feedback technique, we significantly enhance dynamic control performance without relying on prior knowledge of the system. However, the direct application of the classical auto-tuning method to the TID controller encounters challenges due to the additional fractional-order transfer function s-1n. Therefore, we have developed a fractional-order Ziegler-Nichols (FOZ-N) approach, specifically designed to adjust the parameters of the TID controller. The proposed FOZ-N method is fast, simple, and capable of achieving the desired performance. In contrast to the previous Ziegler-Nichols tuning method, the TID controller parameters Kt and Ki are determined to shift the critical point (-1/Ku,j0) to the FOZ-N point (-0.5,-j0.7) on the Nyquist curve, ensuring system robustness and dynamic performance. The impact of the fractional order parameter s-1n and ratio r is explored through time-domain analysis, where these parameters are determined to ensure optimal dynamic performance. Additionally, we provide a detailed tuning procedure along with a helpful example. To demonstrate the advantages of the proposed auto-tuning TID controller over the Ziegler-Nichols PID controller, Optimal PID controller, simple internal model control (SIMC) PID controller, and Ziegler-Nichols FOPID controller, we present a simulation illustration involving multiple different systems. To validate the practical outcomes of this paper, we present experimental results on the temperature control of a Peltier cell.

Handling the Challenges of Small-Scale Labeled Data and Class Imbalances in Classifying the N and K Statuses of Rubber Leaves Using Hyperspectroscopy Techniques.

The nutritional status of rubber trees (Hevea brasiliensis) is inseparable from the production of natural rubber. Nitrogen (N) and potassium (K) levels in rubber leaves are 2 crucial criteria that reflect the nutritional status of the rubber tree. Advanced hyperspectral technology can evaluate N and K statuses in leaves rapidly. However, high bias and uncertain results will be generated when using a small size and imbalance dataset to train a spectral estimaion model. A typical solution of laborious long-term nutrient stress and high-intensive data collection deviates from rapid and flexible advantages of hyperspectral tech. Therefore, a less intensive and streamlined method, remining information from hyperspectral image data, was assessed. From this new perspective, a semisupervised learning (SSL) method and resampling techniques were employed for generating pseudo-labeling data and class rebalancing. Subsequently, a 5-classification spectral model of the N and K statuses of rubber leaves was established. The SSL model based on random forest classifiers and mean sampling techniques yielded optimal classification results both on imbalance/balance dataset (weighted average precision 67.8/78.6%, macro averaged precision 61.2/74.4%, and weighted recall 65.7/78.5% for the N status). All data and code could be viewed on the:Github https://github.com/WeehowTang/SSL-rebalancingtest. Ultimately, we proposed an efficient way to rapidly and accurately monitor the N and K levels in rubber leaves, especially in the scenario of small annotation and imbalance categories ratios.

Underwater power compensated white light source based on synthetic white laser.

The semiconductor white laser light source is used as a light source for underwater illumination. The required standard color temperature of white light is obtained at the underwater target surface. We studied the power compensation of a synthetic white laser source and its application to underwater illumination. First, the power ratios of the red (638 nm), green (520 nm), and blue (450 nm) lasers at a color temperature of 6500 K were obtained by using chromaticity theory. Next, the three-color and synthetic white laser parameters were obtained with transmission distance, according to the exponential attenuation characteristics of different light in clear water and seawater medium. The three-color laser power at the output was compensated, and the underwater target illumination surface reached the standard 6500 K color temperature of the white laser, improving the illumination. Finally, an experimental system for underwater white laser illumination based on power compensation was established. The errors between experimental and theoretical results of color temperature and illuminance are no more than 0.43% and 22.15%. This power-compensated synthetic white laser light source has both the advantages of long-range underwater detection and the spectral advantages of LED white light sources. The white laser light source meets specific requirements by compensating for power and optimizing white light characteristics for underwater lighting applications.

Enhancement of lower limb motor imagery ability via dual-level multimodal stimulation and sparse spatial pattern decoding method.

Recently, motor imagery brain-computer interfaces (MI-BCIs) with stimulation systems have been developed in the field of motor function assistance and rehabilitation engineering. An efficient stimulation paradigm and Electroencephalogram (EEG) decoding method have been designed to enhance the performance of MI-BCI systems. Therefore, in this study, a multimodal dual-level stimulation paradigm is designed for lower-limb rehabilitation training, whereby visual and auditory stimulations act on the sensory organ while proprioceptive and functional electrical stimulations are provided to the lower limb. In addition, upper triangle filter bank sparse spatial pattern (UTFB-SSP) is proposed to automatically select the optimal frequency sub-bands related to desynchronization rhythm during enhanced imaginary movement to improve the decoding performance. The effectiveness of the proposed MI-BCI system is demonstrated on an the in-house experimental dataset and the BCI competition IV IIa dataset. The experimental results show that the proposed system can effectively enhance the MI performance by inducing the α, ß and γ rhythms in lower-limb movement imagery tasks.

Electrokinetic Translocation of a Deformable Nanoparticle through a Nanopore.

The nanopore-based biosensing technology is built up on the fluctuation of the ionic current induced by the electrokinetic translation of a particle penetrating the nanopore. It is expected that the current change of a deformable bioparticle is dissimilar from that of a rigid one. This study theoretically investigated the transient translocation process of a deformable particle through a nanopore for the first time. The mathematical model considers the Poisson equation for the electric potential, the Nernst-Planck equations for the ionic transport, the Navier-Stokes equations for the flow field, and the stress-strain equation for the dynamics of the deformable bioparticle. The arbitrary Lagrangian-Eulerian method is used for the fully coupled particle-fluid dynamic interaction. Results show that the deformation degree of the particle, the velocity deviation, and the current is different from the rigid particle. The deformation degree of the particle will reach the maximum when the particle passes a nanopore. Because of the deformation of particles, the total force applied on deformable particles is larger than that of rigid particles, resulting in larger velocity deviation and current deviation. The influences of the ratio of the nanoparticle radius to the Debye length and surface charge density of the nanopore are also studied. The research results illustrate the translocation mechanism of a deformable nanoparticle in the nanopore, which can provide theoretical guidance for the biosensing technology based on the nanopore.

Gradual increase of perturbation load induces a longer retention of locomotor adaptation in children with cerebral palsy.

The goal of this study is to determine whether the size and the variability of error have an impact on the retention of locomotor adaptation in children with cerebral palsy (CP). Eleven children with CP, aged 7-16 years old, were recruited to participate in this study. Three types of force perturbations (i.e., abrupt, gradual and noisy loads) were applied to the right leg above the ankle starting from late stance to mid-swing in three test sessions while the subject walked on a treadmill. Spatial-temporal gait parameters were recorded using a custom designed 3D position sensor during treadmill walking. We observed that children with CP adapted to the resistance force perturbation and showed an aftereffect consisting of increased step length after load release. Further, we observed a longer retention of the aftereffect for the condition with a gradual load than that with an abrupt load. Results from this study suggested that the size of error might have an impact on the retention of motor adaptation in children with CP with a longer retention of motor adaptation for the condition with a small size of error than that with a large error. In addition, enhanced variability of error seems facilitate motor learning during treadmill training. Results from this study may be used for the development of force perturbation based training paradigms for improving walking function in children with CP.

sEMG Based Gait Phase Recognition for Children with Spastic Cerebral Palsy.

The goal of this study was to examine the optimal strategies for the recognition of gait phase based on surface electromyogram (sEMG) of leg muscles while children with cerebral palsy (CP) walked on a treadmill. Ten children with CP were recruited to participate in this study. sEMG from eight leg muscles and leg position signals were recorded while subjects walked on a treadmill. The position signals of left and right legs were used to develop a five gait sub-phases classifier, i.e., mid stance, terminal stance, pre-swing, mid swing, and terminal swing. Seven feature sets of sEMG signals were tested in recognizing the five gait sub-phases of children with CP. Results from this study indicated that the recognition performance of mean absolute value and zero crossing was better than that with other feature sets when using support vector machine (average classification accuracy was 89.40%). Further, we found that the performance of gait phase recognition is relatively better in pre-swing than other sub-phases, and the performance of gait phase recognition is relatively poorer in mid-swing than other sub-phases. Results from this study may be used to develop an intention-driven robotic gait training system/paradigm for assisting walking in children with CP through robotic training.

Detection of Nitrogen Content in Rubber Leaves Using Near-Infrared (NIR) Spectroscopy with Correlation-Based Successive Projections Algorithm (SPA).

Near-infrared spectroscopy is an efficient, low-cost technology that has potential as an accurate method in detecting the nitrogen content of natural rubber leaves. Successive projections algorithm (SPA) is a widely used variable selection method for multivariate calibration, which uses projection operations to select a variable subset with minimum multi-collinearity. However, due to the fluctuation of correlation between variables, high collinearity may still exist in non-adjacent variables of subset obtained by basic SPA. Based on analysis to the correlation matrix of the spectra data, this paper proposed a correlation-based SPA (CB-SPA) to apply the successive projections algorithm in regions with consistent correlation. The result shows that CB-SPA can select variable subsets with more valuable variables and less multi-collinearity. Meanwhile, models established by the CB-SPA subset outperform basic SPA subsets in predicting nitrogen content in terms of both cross-validation and external prediction. Moreover, CB-SPA is assured to be more efficient, for the time cost in its selection procedure is one-twelfth that of the basic SPA.

Fractional order PIλ controller synthesis for steam turbine speed governing systems.

The current state of the art of fractional order stability theory is hardly to build connection between the time domain analysis and frequency domain synthesis. The existing tuning methodologies for fractional order PIλDµ are not always satisfy the given gain crossover frequency and phase margin simultaneously. To overcome the drawbacks in the existing synthesis of fractional order controller, the synthesis of optimal fractional order PIλ controller for higher-order process is proposed. According to the specified phase margin, the corresponding upper boundary of gain crossover frequency and stability surface in parameter space are obtained. Sweeping the order parameter over λ∈(0,2), the complete set of stabilizing controller which guarantees both pre-specifying phase frequency characteristic can be collected. Whereafter, the optimal fractional order PIλ controller is applied to the speed governing systems of steam turbine generation units. The numerical simulation and hardware-in-the-loop simulation demonstrate the effectiveness and satisfactory closed-loop performance of obtained fractional order PIλ controller.

Applying a pelvic corrective force induces forced use of the paretic leg and improves paretic leg EMG activities of individuals post-stroke during treadmill walking.

OBJECTIVE: To determine whether applying a mediolateral corrective force to the pelvis during treadmill walking would enhance muscle activity of the paretic leg and improve gait symmetry in individuals with post-stroke hemiparesis. METHODS: Fifteen subjects with post-stroke hemiparesis participated in this study. A customized cable-driven robotic system based over a treadmill generated a mediolateral corrective force to the pelvis toward the paretic side during early stance phase. Three different amounts of corrective force were applied. Electromyographic (EMG) activity of the paretic leg, spatiotemporal gait parameters and pelvis lateral displacement were collected. RESULTS: Significant increases in integrated EMG of hip abductor, medial hamstrings, soleus, rectus femoris, vastus medialis and tibialis anterior were observed when pelvic corrective force was applied, with pelvic corrective force at 9% of body weight inducing greater muscle activity than 3% or 6% of body weight. Pelvis lateral displacement was more symmetric with pelvic corrective force at 9% of body weight. CONCLUSIONS: Applying a mediolateral pelvic corrective force toward the paretic side may enhance muscle activity of the paretic leg and improve pelvis displacement symmetry in individuals post-stroke. SIGNIFICANCE: Forceful weight shift to the paretic side could potentially force additional use of the paretic leg and improve the walking pattern.