Multivariate time series (MTS) forecasting is critical in various real-world applications. Recent studies leverage large language models (LLMs) for MTS forecasting, achieving impressive improvements in prediction accuracy and generalization over deep learning-based models. However, these methods often overlook the inherent nonstationarity and domain-specific nature of MTS, as well as the MTS-text modality representation gap, thus limiting the full potential of LLMs in MTS forecasting. In this article, we present A$^{2}$RA-NSMTSllm, a novel multiscale model that adversarially aligns retrieval-augmented LLMs for nonstationary MTS forecasting (A$^{2}$RA-NSMTSllm). Specifically, A$^{2}$RA-NSMTSllm employs a frequency-guided multiscale decomposition normalization–denormalization Framework, which adaptively removes and recovers the nonstationarity based on multiscale distribution dynamics modeling. To bridge the gap between MTS and LLM semantic space, it develops a time series–text alignment reconstruction-enhanced generative adversarial network, where the generator effectively learns to produce LLM-friendly TS embeddings without losing key temporal features. Furthermore, A$^{2}$RA-NSMTSllm introduces a TS knowledge base retrieval-augmented mechanism to enhance domain-specific forecasting, dynamically retrieving and integrating relevant knowledge to guide LLM temporal reasoning and forecasting. Extensive experiments show that A$^{2}$RA-NSMTSllm achieves superior forecasting performance and generalization, outperforming the latest CALF by 8.19%/5.27%, 11.01%/8.66%, and 20.94%/10.84% in MSE/MAE across full-shot, few-shot, and zero-shot scenarios.

The Moon’s distinctive spatial weathering environment and the internal dynamics of its evolutionary process have produced intricate lunar characteristics that make remote sensing data interpretation challenging. The variety of features and the absence of observational data have resulted in the issue of ill-posed inversion in lunar remote sensing. The Hapke radiative transfer model employed in the current inversion is often oversimplified by idealized assumptions, including applying a single-parameter Gaussian distribution to represent macroscopic roughness, significantly diverging from the actual lunar surface characteristics. In addition, the macroscopic roughness parameter of the Hapke model has some coupling with the single-scattering albedo, making inversion difficult. This study proposes an enhanced Hapke model defined as RaySHapke, to accurately represent the actual lunar surface parameters and simulate shading, shadowing, and multiple scattering effects by raytracing. This approach presents the macroroughness shadowing function and proposes an approximate Bayesian inversion framework to enhance the stability and efficiency of the inversion process. The results indicate that raytracing can proficiently replicate the terrain effect. Comparison of the original Hapke model with the RaySHapke model suggests that the raytracing shade function may enhance the accuracy and stability of the inversion of single-scattering albedo and phase function parameters. This study provides a solid foundation for interpreting the physical properties of the lunar surface, as well as lunar science and engineering analysis.

Given the crucial role of the gas utilization ratio (GUR) in reflecting blast furnace operation and energy consumption, accurately predicting its development trend holds significant value for blast furnace operators. However, in the harsh ironmaking environment, GUR-affecting variables are prone to significant nonstationary noise. Moreover, these variables are coupled and correlated, meaning that improper regulation of one variable can destabilize the furnace and lead to substantial GUR fluctuations. This poses a major challenge for achieving accurate GUR prediction. To tackle this issue, this article proposes a denoising multiscale spectral graph wavelet neural network (DMSGWNN) for online dynamic forecasting of the GUR, which is an end-to-end learning method that removes variable noise and captures complex variable correlations simultaneously. First, a regularized self-representation (RSR) model is constructed to eliminate nonstationary noise in blast furnace process variables. Then, a novel multiscale spectral graph wavelet neural network (MSGWNN) is proposed to capture the complex correlations among input variables and extract their multiscale representations through spectral graph wavelet (SGW) transform with the heat kernel scaling function and Gaussian kernel wavelet functions. Finally, the effectiveness of the proposed DMSGWNN method is verified using actual blast furnace ironmaking process data from a blast furnace in China, achieving an average predictive hit rate (HR) as high as 98.06% for GUR prediction.

Time series forecasting plays an important role in numerous real-world domains. Considerable studies have been devoted to prediction by learning temporal features, utilizing improved variants of deep neural networks. However, the variable temporal patterns inherent in complex time series prohibit deep models from discovering reliable dependencies, impairing prediction accuracy. Going beyond previous models, we propose CorrDCN, a decomposed convolutional network with the capability of seasonal autocorrelation 2D-variation modeling. We design the Frequency Guided Decomposition block adaptively configured based on the input series. This facilitates precise series decomposition while allowing CorrDCN to personalize modeling for the decomposed components. Further, we utilize a concise cascaded MLP structure to progressively learn trend variations and integrate local features with global correlations to adequately model seasonal variations. In particular, to tackle the limitations of the 1D structure in simultaneous modeling, we represent seasonal variations in 2D space by reshaping a seasonal 2D tensor based on autocorrelation. This reshaping operation embeds the local features and global correlations of the seasonal series into the rows and columns of the 2D tensor, and thus such seasonal 2D variations can be easily captured by the Multi-scale Inception layers. CorrDCN shows competitive performance on six benchmark datasets. Compared to the mainstream prediction models TimesNet, Non-stationary Transformer and FEDformer, CorrDCN achieves averaged MSE reductions of 4.9%, 17.1% and 19.4%, respectively.

Timely and accurate traffic prediction is crucial for public safety and rational allocation of resources such as roads. However, it still remains an open challenge for timely accurate traffic forecasting, due to the highly nonlinear temporal correlation and dynamical spatial dependence of traffic data. In order to fully capture the temporal and spatial dependences, we propose a dual-channel spatio-temporal wavelet transform graph neural network (DSTwave) for traffic forecasting. Specifically, the wavelet transform neural network is used to obtain the low- and high-frequency parts from the original traffic sequence signals, and in order to accurately capture the spatio-temporal dependence of the low- and high-frequency components in the long - and short-term patterns, the dual-channel ST-GCN with trend-seasonal feature decomposition is carefully designed. In addition, Dynamic-adaptive adjacency matrix is introduced, which can flexibly adapt to changing data. A large number of experiments on two real datasets show that the proposed model has high prediction accuracy.

Domain generalization (DG) is a challenging task that aims to train a robust model with only labeled source data and can generalize well on unseen target data. The domain gap between the source and target data may degrade the performance. A plethora of methods resort to obtaining domain-invariant features to overcome the difficulties. However, these methods require sophisticated network designs or training strategies, causing inefficiency and complexity. In this paper, we first analyze and reclassify the features into two categories, i.e., implicitly disentangled ones and explicitly disentangled counterparts. Since we aim to design a generic algorithm for DG to alleviate the problems mentioned above, we focus more on the explicitly disentangled features due to their simplicity and interpretability. We find out that the shape features of images are simple and elegant choices based on our analysis. We extract the shape features from two aspects. In the aspect of networks, we propose Multi-Scale Amplitude Mixing (MSAM) to strengthen shape features at different layers of the network by Fourier transform. In the aspect of inputs, we propose a new data augmentation method called Random Shape Warping (RSW) to facilitate the model to concentrate more on the global structures of the objects. RSW randomly distorts the local parts of the images and keeps the global structures unchanged, which can further improve the robustness of the model. Our methods are simple yet efficient and can be conveniently used as plug-and-play modules. They can outperform state-of-the-art (SOTA) methods without bells and whistles.

The accurate radiometric calibration is very important for remote sensing optical sensors. Numerous astronomical measurements and model analyses have shown that the Moon is extremely stable in reflectance properties and can serve as a great absolute reference source. Jilin-1 satellite has made long-period observations of the Moon over a wide range of phase angles. However, the spectra displayed fluctuations after radiometric calibration with the provided coefficients. This is inconsistent with the lunar reflectance spectra, which are smooth and generally no sharp absorption. In this study, we conducted radiometric calibration for PMS2 onboard Jilin-1 Guangpu02 (GP02) satellite using the Moon as the calibration source. The calibration results were validated based on different lunar geological terrain, i.e., maria and highlands. The lunar phase curve obtained through the application of the lunar calibration coefficients derived in this study show good agreement with the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) and PLEIADES data. The results indicate that: 1) the spectra become much smoother after lunar calibration, consistent with the characteristics of lunar spectra; 2) the lunar calibration results demonstrate exceeding linearity, with average linear fit uncertainty of 0.69% and goodness-of-fit$({R{^{2}}})$of 0.9997 for the calibration coefficients; and 3) the average relative difference between the phase curves of PMS2 and SeaWiFS is 3.6% and 4.66% with PLEIADES. The obtained results confirm the effectiveness of our lunar calibration method. Although this method was specifically applied to the PMS2, the lunar calibration method is expected to be applicable to other Earth-orbiting satellites.

Time series forecasting is an important problem involving many fields, including the prediction of extreme weather early warning, electricity consumption planning, and long-term traffic congestion. Compared with one-step-ahead prediction, multi-horizon forecasting demands high prediction capacity of the model. Recent studies have shown the great potential of Transformer to improve the prediction accuracy. However, there are three problems with Transformer that restrict its performance, i.e. error accumulation, short-term and long-term dependencies. First, due to the teacher forcing strategy, the ground truth of target values are given during training and replaced by previous step output during testing. This difference between training and testing can lead to error accumulation. Second, time series data have a strong dependence on their local time information. But in classical Transformer architecture, the dot-product self-attention is computed by point-wise values, which are insensitive to local context. Thus, they may fail to distinguish between a turning point, an outlier and the part of patterns. Third, most methods optimize only one objective function and don't model the distributions of data, which is difficult to capture the long-term intricate patterns of time series. To solve these issues, we propose a Transformer-based time series forecasting model in this paper, named Adversarial Convolutional Transformer(ACT). First, we change the decoding mode from step-by-step way to one-step way, which can predict the entire sequence at one forward step to relieve the error accumulation issue. Next, we propose the convolutional attention block, which incorporates local context into the self-attention mechanism and captures the short-term dependencies of data. Then, we introduce adversarial training to the model to capture the long-term repeating patterns. Experiments on five challenging datasets demonstrate that ACT can bring solid improvements in accuracy.

TanSat was successfully launched on December 22, 2016, and has been acquiring global measurements of CO2and O2spectral bands in reflected sunlight since early February 2017. The atmospheric carbon dioxide grating spectrometer (ACGS) is a spaceborne three-band grating hyperspectral spectrometer suite onboard TanSat. The ACGS is designed to measure high-spectral-resolution, coboresighted spectra of reflected sunlight within the molecular oxygen (O2) A-band range from 0.758 to 0.778$\mu \text{m}$and the weak and strong absorption bands of carbon dioxide (WCO2and SCO2) ranging from 1.594 to 1.624$\mu \text{m}$and from 2.042 to 2.082$\mu \text{m}$, respectively. The spectral resolving power ($\lambda /\Delta \lambda $) of the ACGS is ~19 000, ~12 800 and ~12 250 in the O2A-band, WCO2band and SCO2band, respectively. The inflight radiometric calibration accuracy is better than 5%, which satisfies the required specification. The wavelength calibration accuracy of the O2A-band is ~0.19 pm, that of the WCO2band is ~0.27 pm, and that of the SCO2band is ~4.75 pm, all of which meet the 0.05 full-width at half-maximum (FWHM) requirement. The spectroscopic performance of the ACGS exceeds the mission requirements by a margin. The ACGS has noise levels that are comparable to or smaller than those observed during prelaunch testing, and the noise has remained stable in the three bands during inflight operations. The signal-to-noise ratio (SNR) levels of the three bands meet the specified requirements. As expected, the ACGS radiometric performance in the O2A, WCO2, and SCO2bands was fairly good during its first 17 months inflight.

The microwave radiation imager (MWRI) located onboard the FengYun-3C (FY-3C) satellite provides a considerable amount of critical information for numerical weather predictions. Obtaining accurate geolocation results from the FY-3C MWRI data is of great importance. In this letter, we improve the traditional coastline inflection method (CIM) and propose an$\ell _{0}$sparse approximation model for geolocation error estimation and correction. Specifically, we propose using the jump point of the step function to estimate the true coastline point. This approach can characterize the geolocation errors more accurately than the CIM, which further improves the geolocation accuracy. In the theoretical part, we provide a complete solution to obtain the step function through an iterative blind deconvolution. For a practical use, we demonstrate the effectiveness of the proposed method for geolocation error estimation through quantitative results obtained on the FY-3C MWRI data. The experimental results show that the proposed method can achieve an improvement of up to 33.33% in the standard deviation of geolocation errors (approximately 0.00030) compared to the traditional CIM (approximately 0.00045). Furthermore, we also apply the proposed method to the FY-3C satellite and improve the geolocation accuracy of the MWRI data through geolocation error correction.

Nominal grid data of geostationary remote sensing satellites are fundamental for generating the subsequent products. It can be obtained by normalized projection models, mainly based on the imaging mode. However, there are only definitions and primary descriptive equations for the normalized geostationary projection (NGP) model in the existing literature, while the corresponding imaging mode and the physical interpretation are missing, thus hindering the understanding of the produced nominal grid dataset as well as the subsequent products based on the grid. This paper first derived the imaging mode for NGP based on the limited literature. In addition, another new imaging mode was introduced and analyzed based on NGP. The corresponding projection model [nonstandard normalized geostationary projection (NNGP)] was proposed, which is entirely consistent with the situation of America's Geostationary Operational Environmental Satellite-R Series (GOES-R) and Chinese Fengyun-4A (FY-4A). Furthermore, this paper proposed a novel nominal projection model for frame imaging, which is consistent with the imaging mode of China's Gaofen-4. Finally, extensive experiments were designed to comparatively analyze the three nominal grids and demonstrate a detailed difference. By providing a theoretical basis for nominal grid selection, this research is highly significant for the efficient near-real-time production and further applications of geostationary images, as well as the conversion between different datasets resulting from different nominal grid data. In addition, our models are sufficiently tested during the on-orbit running of FY-4A, the first satellite of China's second-generation three-axis stabilized geostationary meteorological satellite series. The algorithms provide the technical support for the high-precision image navigation and registration and play a significant role in robustly producing the meteorological data with similar quality to those from GOES-R.

Focus on the first China domestic coking flue gas desulfurization and denitriation integrated device, in order to solve the problem that the entrance parameters fluctuate and a detection lag exists due to the upstream coking workshop, which is extremely unfavorable to the optimal control of desulfurization and denitriation process. An intelligent integrated prediction model of flue gas SO2 concentration, O2 content and NOx concentration was proposed: the mechanism models of SO2, NOx concentration and O2 content were established according to the principle of material balance and reaction kinetics, respectively. For the prediction error, raw data was pretreated and the auxiliary variables were determined by principal component analysis, in order to improve the training speed and generalization ability of neural network, an improved RBFNN combining optimal stopping principle and dual momentum adaptive learning rate was proposed and used to compensate the error. Based on the practical data of two 55-hole and 6-meter top charging coke ovens in the coking group, the effectiveness and superiority of proposed model and method were verified by simulation via comparison of various models.