Gridded Fields Research Articles

Real-time Distributed Computing Model of Low-Voltage Flow Data in Digital Power Grid under Autonomous and Controllable Environments

Motivated by the progress in artificial intelligence and edge computing, this paper proposes a real-time distributed computing model for low-voltage flow data in digital power grids under autonomous and controllable environments. The model utilizes edge computing through wireless offloading to efficiently process and analyze data generated by low-voltage devices in the power grid. Firstly, we evaluate the performance of the system under consideration by measuring its outage probability, utilizing both the received signal-to-noise ratio (SNR) and communication and computing latency. Subsequently, we analyze the system’s outage probability by deriving an analytical expression. To this end, we utilize the Gauss-Chebyshev approximation to provide an approximate closed-form expression. The results of our experimental evaluation demonstrate the effectiveness of the proposed model in achieving real-time processing of low-voltage flow data in digital power grids. Our model provides an efficient and practical solution for the processing of low-voltage flow data, making it a valuable contribution to the field of digital power grids.

Preparation of layered Ni-rich LiNi0.9Co0.05Mn0.05O2 cathode materials with excellent electrochemical properties by controllable lithium supply and sintering

Layered Ni-rich cathode materials (LiNixCoyMn1-x-yO2) are promising cathode materials for next-generation lithium-ion batteries due to their high reversible capacity and energy density. However, it is still facing the challenges in unstable structure and weak electrochemical properties, inhibiting their large-scale application. Herein, Ni-rich LiNi0.9Co0.05Mn0.05O2 cathode materials are synthesized using Ni0.9Co0.05Mn0.05(OH)2 precursors and Li source by two-steps heating treatment processes. Their structure and properties are regulated by controlling Li supply amounts and sintering temperature to further expand the applications of Ni-rich cathode materials. Experimental results suggest that the preheating at 500 °C effectively establishes a structural foundation (R-3m) for the following sintering of LiNi0.9Co0.05Mn0.05O2 materials. Furthermore, it is found that Li supply amount and sintering temperature have a certain correlation with the structural stability and electrochemical properties of materials. At Li supply amount 1.04 (Li/transition metals atom ratio) and sintering temperature 700 °C, the prepared LiNi0.9Co0.05Mn0.05O2 materials exhibit optimal electrochemical properties (77.0 % capacity retention after 400 cycles at 1C, and about 128.0 mAh g‐1 discharge capacity at 10C). This study provides a rational regulating strategy for synthesizing Ni-rich LiNi0.9Co0.05Mn0.05O2 cathode materials with controllable structural and electrochemical properties, further expanding their applications in the fields of smart grids and electric vehicles.

Grid data asset relationship and intelligent classification integrating knowledge graph and Internet of Things

Abstract With the development of smart grids, power grids have accumulated massive amounts of data in various links such as power generation, transmission, substation, distribution, power consumption, and dispatch. More and more big data applications are beginning to be applied in various professional fields of the power grid. Promote the application and value discovery of smart grid big data through data fusion inside and outside the grid. Grid data has become an important asset for enterprise development, but power grid enterprises lack effective technical means to solve the whole life cycle monitoring and relationship of power grid data assets. Aiming at the relationship between power grid data assets, this paper proposes a set of grid data asset relationship and intelligent classification framework that integrates knowledge graph and Internet of Things. First, the grid knowledge graph extraction relationship is carried out by ProjE algorithm. Then, the relationship between power grid data assets and intelligent classification framework that integrates knowledge graph and Internet is proposed. Finally, the corresponding classification application is proposed by using intelligent classification algorithm. Experimental results show that the intelligent classification accuracy rate can reach 93.12% under the relationship between the knowledge graph and the Internet data assets, which has a new idea for the future development of the relationship between power grid data assets.

Application of Neuro-Fuzzy Techniques for Energy Scheduling in Smart Grids Integrating Photovoltaic Panels

In recent years, most of the research in the field of smart grids integrating renewable energy sources assumed energy efficiency as a scheduling objective. However, the aspects of energy consumption or energy demand have not been described clearly, even though they have been proven to be an effective way of reducing energy consumption. In this context, this study aimed to cover a key research challenge in the field, such as the development of an intelligent strategy for solving energy consumption scheduling problems. The added value of our proposal consists of classifying individual consumption profiles assigned to each operation cycle phase, instead of considering an average of non-varying consumption of household appliances. Within this hybrid approach, the proposed explainable system, based on self-organizing maps of neural networks, fuzzy clustering algorithm, and scheduling technics, correlates the complex interrelation between power generated from renewable energy sources in a smart grid, prosumers’ load behaviors, and the consumption profile of controllable or uncontrollable appliances. The tests were made using green energy consumption and production from real monitored data sets. The load-shifting algorithm that was used to reduce energy consumption from the national energy grid proved its effectiveness. In fact, consumers paid 25% less for the energy they used from the national energy grid during the times when the amount of electricity produced from renewable sources was reduced as a result of weather conditions.

Research on Intelligent Organization and Application of Multi-source Heterogeneous Knowledge Resources for Energy Internet

ABSTRACTTo improve the informationization and intelligence of the energy Internet industry and enhance the capability of knowledge services, it is necessary to organize the energy Internet body of knowledge from existing knowledge resources of the State Grid, which have the characteristics of large scale, multiple sources, and heterogeneity. At the same time, the business fields of State Grid cover a wide range. There are many sub-fields under each business field, and the relationship between fields is diverse and complex. The key to establishing the energy Internet body of knowledge is how to fuse the heterogeneous knowledge resources from multiple sources, extract the knowledge contents from them, and organize the different relationships. This paper considers transforming the original knowledge resources of State Grid into a unified and well-organized knowledge system described in OWL language to meet the requirements of heterogeneous resource integration, multi-source resource organization, and knowledge service provision. For the State Grid knowledge resources mainly in XML format, this paper proposes a Knowledge Automatic Fusion and Organization idea and method based on XSD Directed Graph. According to the method, the XML corresponding XSD documents are transformed into a directed graph in the first stage during which the graph neural network detects hidden knowledge inside the structure to add semantic information to the graph.In the second stage, for other structured knowledge resources (e.g., databases, spreadsheets), the knowledge contents and the relationships are analyzed manually to establish the mappings from structured resources to graph structures, using which the original knowledge resources are transformed into graph structures, and merged with the directed graphs obtained in the first stage to achieve the fusion of heterogeneous knowledge resources. And expert knowledge is introduced for heterogeneous knowledge fusion to further extend the directed graph. And in the third stage, the expanded directed graph is converted to the body of knowledge in the form of OWL. This paper takes the knowledge resources in the field of human resources of the State Grid as an example, to establish the ontology of the human resources training field in a unified manner, initially demonstrating the effectiveness of the proposed method.

SGAM-Based Analysis for the Capacity Optimization of Smart Grids Utilizing e-Mobility: The Use Case of Booking a Charge Session

The description of the functionality of a smart grid’s architectural concept, analyzing different Smart Grid (SG) scenarios without disrupting the smooth operation of the individual processes, is a major challenge. The field of smart energy grids has been increasing in complexity since there are many stakeholder entities with diverse roles. Electric Vehicles (EVs) can transform the stress on the energy grid into an opportunity to act as a flexible asset. Smart charging through an external control system can have benefits for the energy sector, both in grid management and environmental terms. A suitable model for analyzing and visualizing smart grid use cases in a technology-neutral manner is required. This paper presents a flexible architecture for the potential implementation of electromobility as a distributed storage asset for the grid’s capacity optimization by applying the Use Case and Smart Grid Architecture Model (SGAM) methodologies. The use case scenario of booking a charge session through a mobile application, as part of the TwinERGY Horizon 2020 project, is deployed to structure the SGAM framework layers and investigate the applicability of the SGAM framework in the integration of electromobility as a distributed storage asset into electricity grids with the objective of enhanced flexibility and decarbonization.

Generation of Non-Linear Technique Based 6 Hourly Wind Reanalysis Products Using SCATSAT-1 and Numerical Weather Prediction Model Outputs

We combined observations of ocean surface winds from Indian SCATterometer SATellite-1 (SCATSAT-1) with a background wind field from a numerical weather prediction (NWP) model available at National Centre for Medium-Range Weather Forecast (NCMRWF) to generate a 6-hourly gridded hybrid wind product. A distinctive feature of the study is to produce a global gridded wind field from SCATSAT-1 scatterometer passes with spatio-temporal data gaps at regular synoptic hours relevant for forcing models and other NWP studies. We are following the concept from the modern particle filter technique, which does not represent the model probability density function (PDF) as Gaussian. We generated the 6-hourly hybrid winds for 2018 and validated them using the wind speed from daily gridded level-4 SCATSAT-1 winds (L4AW), Cross Calibrated Multi-Platform (CCMP) dataset and global buoy data from National Data Buoy Centre (NDBC). The results suggest the potential of the technique to produce scatterometer winds at the desired temporal frequency with significantly less noise and bias along the swath. The study shows that the generated hybrid winds are of prime quality compared with the already existing daily products available from Indian Space Research Organization (ISRO).

Impact of upstream variability on the Loop Current dynamics in numerical simulations of the Gulf of Mexico

The Loop Current (LC), which is the main mesoscale dynamic feature of the Gulf of Mexico (GoM), has a major impact on the circulation and its variability in the interior Gulf. The LC is a highly variable and dynamic feature. It changes shape from a short jet connecting the two openings of the GoM in an almost straight line ("retracted phase") to a long loop invading most of the eastern part of the GoM ("extended phase"). When it is in the extended phase, it can shed large anticyclonic eddies, called Loop Current Eddies, which then migrate to the western GoM. In this study, the processes controlling the LC dynamics are investigated using two multi-decadal simulations of the Gulf of Mexico HYbrid Coordinate Ocean Model differing in their open boundary conditions (BCs) and altimetry-derived gridded fields. The LC in the simulation with BCs derived from monthly climatology state variables frequently remains in its retracted phase significantly longer than observed. In contrast, the duration of the retracted phase is notably shorter in the simulation in which the BCs have realistic daily variability. By examining the flow properties through the Yucatan Channel from which the LC originates, we find that increased intensity of this current and a westward shift of the mean core is associated with the LC transitions from the retracted to the extended phase. This transition is accompanied by an increase of both cyclonicity of the flow in the west and anticyclonicity in the east of the core of this jet. Moreover, the number of anticyclonic eddies entering in the GoM through the Yucatan Channel is significantly higher when the LC extends in the GoM. Consequently, this study demonstrates the importance of realistic flow variability at the lateral boundaries for accurate simulation of the LC system in a model, and reveals characteristics of the upstream flow associated with different LC behavior that can potentially aid in forecasting the LC system.

Exploration and Application of Self‐Healing Strategies in Lithium Batteries

AbstractLithium batteries (LBs) are developed tremendously owing to their excellent energy density as well as cyclic persistence, exhibiting promising applications from portable devices to e‐transportation and grid fields. However, with the ever‐increasing demand for intelligent wearable electronics, more requests are focused on high safety, good durability, and satisfied reliability of LBs. The self‐healing route, which can simulate the ability of organic organisms to repair damage and recover initial function through its intrinsic vitality, is believed to be an efficient strategy to alleviate the unavoidable physical or chemical fatigue and damage issues of LBs, beneficial for the realization of the above mentioned high requests. In this review, the applicability and development of self‐healing materials are summarized in electrodes, electrolytes, and interfacial layers in recent years, focusing on exploring the feasibility of different self‐healing strategies in LBs, discussing the advantages and disadvantages of existing strategies in different parts of batteries, and indicating the possible research directions for beginners who are interested in this field. Finally, the critical challenges and the future research directions as well as opportunities are prospected.

SEMDA: Secure and Efficient Multidimensional Data Aggregation in Smart Grid without a Trusted Third Party

Smart grids are a combination of traditional power system engineering as well as information and communications technology. Smart grid terminals provide convenient services to users by aggregating their data in real time. However, terminals can derive user privacy information from real-time data on smart devices. Therefore, security data aggregation has been widely studied in the field of smart grid. Most existing schemes are one-dimensional data aggregation or rely on a trusted third party. In reality, multidimensional data (such as a user’s electricity consumption or user’s main usage time, etc.) makes sense for terminals to flexibly adjust supply and demand strategies. In this paper, we propose an efficient and secure multidimensional data aggregation scheme that supports batch validation without a trusted third party. Firstly, we apply the Chinese remainder theorem to encode the user’s multidimensional data and realize the independence of each dimension in terminal decryption. Secondly, we adopt a secure key negotiation protocol that does not require a trusted third party. Finally, based on paillier homomorphic encryption and bilinear pairing, we construct an encryption scheme that can reuse the key and blind factor and support batch verification. The analysis results show that our scheme is secure for users’ privacy protection. Experimental results show that, compared with existing 1 dimensional aggregation schemes, our scheme has almost no growth in computational overhead for terminal decryption.

Preface

The 2nd International Conference on Smart Grid and Energy Internet (SGEI 2022) was successfully held on 21st-23rd October, 2022 in Sanya, China (virtual conference). SGEI 2022 promoted innovative technologies and practical applications in the grid and energy fields, aiming to fully realize the interconnection, openness, sharing and synergy of various energy resources, and to shape a green, low-carbon, efficient and low-cost smart energy system.We had the honor of having invited Prof. Jizhong Zhu, President of IEEE PES SBLC (China), to serve as our Conference General Chair. The conference was composed of keynote speeches, oral presentations, and online Q&A discussion, attracting 100 individuals. Firstly, keynote speakers were each allocated 30-45 minutes to hold their speeches. Then in the next part, oral presentations, the excellent papers selected were presented by their authors in sequence.During the conference, three distinguished professors were invited to address their keynote speeches. Among them, Prof. Behnam Mohammadi-Ivatloo from University of Tabriz, Iran addressed a thought-provoking speech on Flexibility Resources for Integrating more Renewable Energy Resources in Power Systems. In the speech, different challenges of integrating renewable energy resources were discussed and different flexibility options were presented for hosting a higher share of renewable power resources in the power systems. And then, Prof. Amjad Anvari-Moghaddam from Aalborg University, Denmark made a speech on Operation and Control of Renewable Energy Systems. To accommodate large numbers of renewable resources, energy distribution and transmission networks need to be adapted and expanded to avoid network congestion and failures. This report covered the aforementioned promising areas in green energy transition and discussed the current and future opportunities and challenges exist in this context. All the sophisticated keynote speakers combined their own interest fields with the conference theme to make dramatic speeches, triggering heated discussion in the conference. And every participant praised this conference for disseminating useful and insightful knowledge.List of Committee member is available in this pdf.

Kernel Density Derivative Estimation of Euler Solutions

Conventional Euler deconvolution is widely used for interpreting profile, grid, and ungridded potential field data. The Tensor Euler deconvolution applies additional constraints to the Euler solution using all gravity vectors and the full gravity gradient tensor. These algorithms use a series of different-sized moving windows to yield many solutions that can be employed to estimate the source location from the entire survey area. However, traditional discrimination techniques ignore the interrelation among the Euler solutions, so they cannot be employed to separate adjacent targets. To overcome this difficulty, we introduced multivariate Kernel Density Derivative Estimation (KDDE) as an extension of Kernel Density Estimation, which is a mathematical process to estimate the probability density function of a random variable. The multivariate KDDE was tested on a single cube model, a single cylinder model, and three composite models consisting of two cubes with various separations using gridded data. The probability value calculated by the multivariate KDDE was used to discriminate spurious solutions from the Euler solution dataset and isolate adjacent geological sources. The method was then applied to airborne gravity data from British Columbia, Canada. Then, the results of synthetic models and field data show that the proposed method can successfully locate meaningful geological targets.

Spatio‐temporal changes in the mean and extreme temperature indices for Serbia

AbstractThis study provides a comprehensive analysis of changes in mean and extreme temperature indices for Serbia using a high‐resolution daily gridded temperature data set during the period of 1951–2020. The annual and seasonal trends of the indices in the gridded fields were computed and tested for statistical significance using the nonparametric Mann–Kendall test and the Sen's slope estimator. The t‐test was applied to check the stability of means of temperature indices between two standard periods (1961–1990 and 1991–2020). A correlation analysis was used in order to investigate the connections of the large‐scale circulation patterns and the mean and extreme temperature indices. Results showed that a warming trend dominated for the last 70 years, with slightly more intense changes for the mean maximum temperatures (TXM) than for the mean minimum temperatures (TNM). An increase was recorded for indices based on the maximum temperature (TXM, hottest days, warm days, summer days) and minimum temperature (TNM, coldest nights, warm nights, tropical nights). The highest increase was observed for warm nights (7 days⋅decade−1) during the summer season. The negative change in cool days (−1.5 days⋅decade−1) and cool nights (−2 days⋅decade−1) was recorded in almost all seasons, being significant for the both annual and summer values. The correlation between the temperature indices and large‐scale circulation patterns revealed that the East Atlantic (EA) pattern had a highly positive correlation with warm temperature indices (hottest days, warm days, warm nights, summer days, and tropical nights) and a negative correlation with cold indices (coldest nights, cool days, cool nights, frost days, and ice days). The North Atlantic Oscillation, the most important teleconnection pattern during the winter months in Europe, had weaker influence on temperature indices in Serbia than the EA pattern, but stronger than the EA/Western Russia pattern.

Anode corrosion in aqueous Zn metal batteries

Given their low cost and intrinsic safety, aqueous Zn metal batteries (AZMBs) are drawing increasing attention in the field of smart grids and large-scale energy storage. However, the Zn metal anode in aqueous electrolyte suffers from a critical issue, corrosion, which must be fully addressed before the practical implementation of AZMBs. In this perspective, the mechanisms of aqueous Zn metal anode corrosion in both alkaline and neutral electrolytes are compared and discussed. The methods for studying the corrosion processes and the strategies for Zn corrosion protection in AZMBs are also summarized. Finally, some expectations about potential research directions for making corrosion-resistant AZMBs a commercial reality are provided.

Observation based climatology Martian atmospheric waves perturbation Datasets

The Martian atmospheric waves perturbation Datasets (MAWPD) version 2.0 is the first observation-based climatology dataset of Martian atmospheric waves. It contains climatology-gridded temperature, gravity waves, and tides spanning the whole Martian year. MAWPD uses the Data INterpolating Empirical Orthogonal Functions method (DINEOF) reconstruction method for data assimilation with the observational data from the Mars Global Surveyor (MGS), Mars Reconnaissance Orbiter (MRO), Mars Atmosphere and Volatile EvolutioN (MAVEN), Mars Pathfinder (MP), Mars Phoenix Lander (MPL), Mars Exploration Rover (MER) and Mars Express (MEX) temperature retrievals. The dataset includes gridded fields of temperature (Level 1 data) as well as the physical quantities of GWs (Level 2 data, amplitude, and potential energies), SPWs and tides (Level 2 data, amplitude, and phase). The MAWPD, based entirely on multiple reliable observations, provides climatological background atmospheric information of temperature and wave disturbances on Mars. The dataset is not only useful for observation-based scientific studies concerning Martian atmospheric waves, e.g., circulation, dust storms, and wave excitation mechanism, but also for cross-validating with model-based datasets or model results.

Research Progress on the Application of Nanomaterials in Lithium-Ion Battery Cathode Materials

Energy scarcity and environmental pollution are getting worse as science and technology advance, which has a significant influence on people's living conditions. Therefore, in light of the present sustainable development plan, the discovery and implementation of new energy sources are crucial for sustainable development. Lithium-ion batteries have been playing an increasingly significant role in the field of smart grids, electric vehicles, and personal electronic devices due to the fact that solar, wind, hydropower, and nuclear energy are new renewable energy sources in this context, but these energy sources are seasonal, regional, and discontinuous. There are currently more demanding specifications for lithium-ion batteries, including high energy density and high power density, due to the growth of electric vehicles and other energy storage technologies. In order to enhance the electrochemical performance of lithium-ion batteries, this thesis focuses on the incorporation of nanomaterials into the cathode materials. It is discovered that there is still much opportunity for nanomaterial development in the cathode materials of batteries.

Statistical Downscaling to Improve the Subseasonal Predictions of Energy-Relevant Surface Variables

Abstract Owing to the increasing share of variable renewable energies in the electricity mix, the European energy sector is becoming more weather sensitive. In this regard, skillful subseasonal predictions of essential climate variables can provide considerable socioeconomic benefits to the energy sector. The aim of this study is therefore to improve the European subseasonal predictions of 100-m wind speed and 2-m temperature, which we achieve through statistical downscaling. We employ redundancy analysis (RDA) to estimate spatial patterns of variability from large-scale fields that allow for the best prediction of surface fields. We compare explanatory powers between the patterns obtained using RDA against those derived using principal component analysis (PCA), when used as predictors in multilinear regression models to predict surface fields, and show that the explanatory power of the former is superior to that of the latter. Subsequently, we employ the estimated relationship between RDA patterns and surface fields to produce statistical probabilistic predictions of gridded surface fields using dynamical ensemble predictions of RDA patterns. We finally demonstrate how a simple combination of dynamical and statistical predictions of surface fields significantly improves the accuracy of subseasonal predictions of both variables over a large part of Europe. We attribute the improved accuracy of these combined predictions to improvements in reliability and resolution.

A Federated Named Entity Recognition Model with Explicit Relation for Power Grid

The power grid operation process is complex, and many operation process data involve national security, business secrets, and user privacy. Meanwhile, labeled datasets may exist in many different operation platforms, but they cannot be directly shared since power grid data is highly privacy-sensitive. How to use these multi-source heterogeneous data as much as possible to build a power grid knowledge map under the premise of protecting privacy security has become an urgent problem in developing smart grid. Therefore, this paper proposes federated learning named entity recognition method for the power grid field, aiming to solve the problem of building a named entity recognition model covering the entire power grid process training by data with different security requirements. We decompose the named entity recognition (NER) model FLAT (Chinese NER Using Flat-Lattice Transformer) in each platform into a global part and a local part. The local part is used to capture the characteristics of the local data in each platform and is updated using locally labeled data. The global part is learned across different operation platforms to capture the shared NER knowledge. Its local gradients from different platforms are aggregated to update the global model, which is further delivered to each platform to update their global part. Experiments on two publicly available Chinese datasets and one power grid dataset validate the effectiveness of our method.

Daily High-Resolution Temperature and Precipitation Fields for the Contiguous United States from 1951 to Present

Abstract In this paper, a new set of daily gridded fields and area averages of temperature and precipitation is introduced that covers the contiguous United States (CONUS) from 1951 to present. With daily updates and a grid resolution of approximately 0.0417° (nominally 5 km), the product, named nClimGrid-Daily, is designed to be used particularly in climate monitoring and other applications that rely on placing event-specific meteorological patterns into a long-term historical context. The gridded fields were generated by interpolating morning and midnight observations from the Global Historical Climatology Network–Daily dataset using thin-plate smoothing splines. Additional processing steps limit the adverse effects of spatial and temporal variations in station density, observation time, and other factors on the quality and homogeneity of the fields. The resulting gridded data provide smoothed representations of the point observations, although the accuracy of estimates for individual grid points and days can be sensitive to local spatial variability and the ability of the available observations and interpolation technique to capture that variability. The nClimGrid-Daily dataset is therefore recommended for applications that require the aggregation of estimates in space and/or time, such as climate monitoring analyses at regional to national scales. Significance Statement Many applications that use historical weather observations require data on a high-resolution grid that are updated daily. Here, a new dataset of daily temperature and precipitation for 1951–present is introduced that was created by interpolating irregularly spaced observations to a regular grid with a spacing of 0.0417° across the contiguous United States. Compared to other such datasets, this product is particularly suitable for monitoring climate and drought on a daily basis because it was processed so as to limit artificial variations in space and time that may result from changes in the types and distribution of observations used.

Reconstructing ocean subsurface salinity at high resolution using a machine learning approach

Abstract. A gridded ocean subsurface salinity dataset with global coverage is useful for research on climate change and its variability. Here, we explore the feed-forward neural network (FFNN) approach to reconstruct a high-resolution (0.25∘ × 0.25∘) ocean subsurface (1–2000 m) salinity dataset for the period 1993–2018 by merging in situ salinity profile observations with high-resolution (0.25∘ × 0.25∘) satellite remote-sensing altimetry absolute dynamic topography (ADT), sea surface temperature (SST), sea surface wind (SSW) field data, and a coarse-resolution (1∘ × 1∘) gridded salinity product. We show that the FFNN can effectively transfer small-scale spatial variations in ADT, SST, and SSW fields into the 0.25∘ × 0.25∘ salinity field. The root-mean-square error (RMSE) can be reduced by ∼11 % on a global-average basis compared with the 1∘ × 1∘ salinity gridded field. The reduction in RMSE is much larger in the upper ocean than the deep ocean because of stronger mesoscale variations in the upper layers. In addition, the new 0.25∘ × 0.25∘ reconstruction shows more realistic spatial signals in the regions with strong mesoscale variations, e.g., the Gulf Stream, Kuroshio, and Antarctic Circumpolar Current regions, than the 1∘ × 1∘ resolution product, indicating the efficiency of the machine learning approach in bringing satellite observations together with in situ observations. The large-scale salinity patterns from 0.25∘ × 0.25∘ data are consistent with the 1∘ × 1∘ gridded salinity field, suggesting the persistence of the large-scale signals in the high-resolution reconstruction. The successful application of machine learning in this study provides an alternative approach for ocean and climate data reconstruction that can complement the existing data assimilation and objective analysis methods. The reconstructed IAP0.25∘ dataset is freely available at https://doi.org/10.57760/sciencedb.o00122.00001 (Tian et al., 2022).