Deep learning has made remarkable achievements in many fields. However, though fuzzy neural networks with natural interpretability are widely used in prediction and control scenarios, there are very few studies on the deepening of fuzzy systems. By integrating rough set theory, fuzzy rough neural network has unique advantages benefiting from the complementarity of the fuzzy set theory and the rough set theory as well as the powerful learning ability of neural networks. Similarly, research on deep fuzzy rough neural networks is even rarer. In this study, in order to improve the performance of the fuzzy rough neural network and expand its application range, the deep fuzzy rough neural network model is constructed and optimized by stacking blocks of fuzzy rough neural network to imitate the deepening deep neural network based on multiobjective evolution. Each fuzzy rough neural network block is interpretable, and its stacked architecture also has high interpretability. To automatically generate deep fuzzy rough neural network models with high efficiency, a distributed parallel multiobjective neuroevolution framework is developed, thus blocks can be stacked flexibly and deep architecture can be optimized considering multiple optimization objectives of accuracy, interpretability and generalization simultaneously. In addition, multiobjective evolution is combined with Wang-Mendel method, pseudoinverse and BP to effectively learn specific parameters. Finally, based on the time series prediction problems, the superiority of the multiobjective deep fuzzy rough neural network evolutionary framework is verified.

This paper investigates the development of low-carbon buildings and carbon emission control in Jiangsu Province, China, utilizing a mathematical model. Through correlation analysis and principal component analysis, the carbon emissions of the entire life cycle of residential buildings are evaluated, and a Grey Prediction Model is established. The study shows that the annual carbon emission from air conditioners is 370.92 kg, given an annual electricity consumption of 1324.71 kW and a carbon emission of 0.28 kg per kWh. It identifies the key carbon emission indicators, including precipitation, temperature, energy consumption, building area, construction materials, water, natural gas, and waste. Principal component analysis ranks building area as the most significant factor. Using the GM (1,1) model, the carbon emissions of Jiangsu Province in 2024 were predicted to be 1.5576 million tons by historical data. Emission reduction suggestions are proposed, such as constructing thicker walls, increasing green spaces, reducing construction waste, and promoting balanced economic development. Moreover, prioritizing insulation materials in building design can reduce winter energy consumption since energy consumption is higher in winter than in summer. This research supports China’s goals of achieving a carbon peak by 2030 and carbon neutrality by 2060 while encouraging low-carbon technological innovation and improving people’s living standards. This study also emphasizes the importance of locally tailored strategies for effective emissions reduction.

With the proliferation of shared personal data online, users encounter difficulties in revoking data access permissions and requesting data deletions, thus increasing the risk of privacy breaches. Machine unlearning offers a solution and is useful for sensing-computing integrated chips and systems, but the state-of-the-art shard-ed, isolated, sliced, and aggregated (SISA) method produces models with high complexity and limited generalization ability. This study proposes a novel framework, named multiobjective evolutionary algorithm-SISA (MOEA-SISA), which integrates feature decomposition-based differential grouping (FDbDG) to improve optimization efficiency through dynamic grouping of decision variables. Three optimization objectives can hence be proposed: model accuracy, model complexity, and generalization ability. By optimizing these three objectives, the trained model becomes closer to the retrained results while preventing excessive forgetting. Experimental models including ViT, VGG-16, and ResNet-50 are used for sub-class forgetting tasks. Compared with various state-of-the-art methods, MOEA-SISA offers advantages across these models, especially in terms of model complexity and generalization ability. Tests against membership inference attacks demonstrate that MOEA-SISA effectively retains accuracy and enhances the generalization ability in sub-class forgetting scenarios. Additionally, the proposed approach offers significant advantages for improving the efficiency and performance of sensing-computing integrated chips and systems.

Wireless short-range transmission is essential for precise wellbore trajectory control and real-time formation evaluation. Its signal propagation characteristics are influenced by multiple factors, including antenna type, drill collar, mud, and formation resistivity. Most prior studies are based on Magnetic-field Antennas (MFA) and primarily focus on the effects of formation resistivity variations, whereas the investigations on the influence of drill collars and mud resistivity are limited. In this study, a three-dimensional finite-element electromagnetic model of the “antenna–drill collar–mud–formation” system was developed to investigate wireless short-range transmission. The model was used to characterize and compare the electromagnetic field distributions of MFA and Electric-field Antennas (EFA) under in situ conditions. On this basis, a set of parametric sensitivity analyses on transmission performance was performed to quantify the effects of key factors, including drill-collar conductivity and mud resistivity. The results reveal fundamentally different electromagnetic field distributions for the two antenna types: (1) MFA is dominated by localized circumferential magnetic flux loops, whereas EFA transmits signals through axially extended eddy-current channels. (2) The drill collar exerts opposite effects on the two antennas, suppressing signal levels for MFA while significantly enhancing transmission for EFA, resulting in signal amplitudes that are ${10}^3-{10}^5$ times higher. (3) In addition, mud resistivity has little influence on MFA, whereas increasing mud resistivity leads to the pronounced attenuation of EFA signals. These findings provide a quantitative basis for antenna selection and performance optimization in wireless short-range transmission systems under different Logging-While-Drilling (LWD) conditions.