Conformal prediction (CP) is a learning framework controlling prediction coverage of prediction sets, which can be built on any learning algorithm for point prediction. This work proposes a learning framework named conformal loss-controlling prediction, which extends CP to the situation where the value of a loss function needs to be controlled. Different from existing works about risk-controlling prediction sets and conformal risk control with the purpose of controlling the expected values of loss functions, the proposed approach in this article focuses on the loss for any test object, which is an extension of CP from miscoverage loss to some general loss. The controlling guarantee is proved under the assumption of exchangeability of data in finite-sample cases and the framework is tested empirically for classification with a class-varying loss and statistical postprocessing of numerical weather forecasting applications, which are introduced as point-wise classification and point-wise regression problems. All theoretical analysis and experimental results confirm the effectiveness of our loss-controlling approach.

As the cornerstone of adjusting microgrid output, the electrical load interface unit is difficult to continuously output desired dynamics under various uncertainties. To handle the problem, a variable form linear active disturbance rejection control (VFLADRC), including extended state differential compensation, is proposed for load interface unit according to the separating response. In this technology, the extended state differential is embedded in the linear extended state observer (LESO) to construct a modified LESO, alongside the control law is generated via a feedforward way. By doing this, the attenuation capability of generalized disturbance response in the system output is improved, thereby optimizing output dynamics. In addition, the stability and output characteristics of the studied scheme are theoretically investigated. It is revealed that the electrical load interface unit with VFLADRC exhibits a higher level of robustness. Finally, comparative experiments are conducted on a 40-kW test platform, which validates the correctness of the analysis and the superiority of the proposed approach.

Global navigation satellite system reflectometry (GNSS-R) Delay-Doppler map measures the sea surface roughness, which has recently been applied to retrieve sea surface wind speed. However, current studies on GNSS-R wind speed retrieval only use the spatial domain of the delay-Doppler map without considering the variations patterns in the map, which is regarded as frequency domain information of the map. In this study, we propose a joint frequency-spatial-domain wind speed retrieval network (FSNet) based on reflectivity data provided by the Cyclone Global Navigation Satellite System (CyGNSS) mission. We construct a matchup dataset between the CyGNSS satellite data and ECMWF model data from January 1, 2018, to December 31, 2019. The wind speed range is 0–25 m/s. Using the proposed FSNet, frequency and spatial features are simultaneously extracted. The frequency domain feature supplements the spatial-domain information of the mid and high-level features in the neural network. Rather than directly concatenating the frequency-domain features with the spatial-domain features, we designed a feature fusion module to fuse frequency and spatial features for wind speed retrieval adaptively. Experiments show that our FSNet wind speed retrieval has a root mean square error (RMSE) of 1.63 m/s for a wind range of 0-25 m/s. This accuracy is 25.4% better than the operational algorithm provided by the CyGNSS Level 2 wind speed product. For a higher wind range of 16-25m/s, FSNet performed even better, improving the RMSE by 31%.

Quantitative precipitation estimation (QPE) is an essential task in meteorology and hydrology and is of great significance for disaster prevention and control. The starting point of QPE is to establish a point-by-point mapping relationship between atmospheric observations and rain gauges. Traditional methods called Z-R relationships fit the parameters in a given paradigm to perform QPE under meteorology prior guidance. Methods based on machine learning (ML) construct the QPE models from statistical views, which could benefit from large historical data. However, in operational applications, these methods are challenging to estimate severe precipitation accurately. The reason is that severe precipitation is usually caused by convective systems. The point-by-point QPEs only focus on fixed isolated points and are difficult to characterize convective systems that cause precipitation effectively. In this letter, a spatiotemporal attention model is proposed for one-hour QPE. For each pixel, the spatiotemporal attention guides the model to find and focus on the most worthy attention region at each moment instead of a definite isolated point, making the model view more flexible and insightful. In experiments, the radar data from 2015 to 2016 in North China are used to train and evaluate the model. Compared with other methods, the results show that the spatiotemporal attention model could effectively improve the accuracy of QPE, especially for intense precipitation. The case study also shows that the operation of our model is more consistent with meteorological perspectives.

Remote sensing image captioning (RSIC) has great significance in image understanding, which describes the image content in natural language. Existing methods are mainly based on deep learning and rely on the encoder–decoder model to generate sentences. In the decoding process, recurrent neural network (RNN) and long short-term memory (LSTM) are normally applied to sequentially generate image captions. In this letter, the transformer encoder–decoder is combined with grid features to improve the RSIC performance. First, the pretrained convolutional neural network (CNN) is used to extract grid-based visual features, which are encoded as vectorial representations. Then, the transformer outputs semantic descriptions to bridge visual features and natural language. Besides, the self-critical sequence training (SCST) strategy is applied to further optimize the image captioning model and improve the quality of generated sentences. Extensive experiments are organized on three public datasets of RSCID, UCM-Captions, and Sydney-Captions. Experimental results demonstrate the effectiveness of SCST strategy and the proposed method achieves superior performance compared with the state-of-the-art image captioning approaches on the RSCID dataset.

Tropical cyclone (TC) intensity estimation is critical for disaster forecasting and severe weather warning. In recent years, the performance of various TC intensity estimation models has been gradually enhanced, but the accuracy still needs to be improved. In this article, 71 intensity-related features are extracted from satellite infrared images of TCs. These features are grouped by eye features, circle features, texture features, and time-series features. Using the random forest model as the underlying algorithm of conformal prediction (CP), an intensity applicable CP framework is proposed. On the one hand, the proposed network can achieve point estimation of the TC intensity. On the other hand, it is also possible to realize the intensity interval estimation that satisfies a given significance level. In the experiments, the root mean square error of the point estimation algorithm is 7.86 kt (1 kt ≈ 0.51 m/s), and the performance of the proposed algorithm is better than the comparison algorithms. In addition, interval estimation enriches decision-making information. The experimental results show that the proposed model is a competitive and promising method for estimating the TC intensity.

DC microgrid can efficiently integrate the renewable energy sources, such as fuel cells and photovoltaic (PV) panels. However, due to the voltage mismatch between the low output voltage of renewable energy source and the high voltage of DC bus, a high voltage gain step-up converter interface between them is required. In this paper, a novel nonisolated DC/DC converter with high voltage gain is proposed. In addition to the high gain characteristic, the proposed converter provides ripple-free input current, which can prolong the service life of fuel cell and improve the generation efficiency of PV panels. Besides, lower voltage stresses of power switches and didoes contribute to enhancing converter efficiency. The operation principle, characteristics analysis and comparative study are also demonstrated in this paper. Finally, a 400W 40V/400V scaled-down experimental prototype is built to verify the converter performance and the theoretical analysis.