Lithium-ion batteries, which are vital for powering mobile devices, experience performance degradation over time due to capacity fading and other aging phenomena, thereby presenting safety risks. This work introduces the DeTransformer-Physics model, a predictive model for battery capacity. This model synergistically integrates physical knowledge and a de-noising autoencoder to enhance predictive accuracy. Firstly, it employs a denoising autoencoder as a preprocessing step to reconstruct denoised input data, thereby facilitating the extraction of more effective input data. Subsequently, a model describing the procedure for lithium-ion battery capacity degradation is derived and incorporated into a neural network, with the model training constrained by physical principles. Finally, using the lithium-ion battery degradation dataset from NASA, we demonstrate that incorporating the denoising autoencoder and embedding physical knowledge substantially improves the predictive accuracy of the model.

The continuous increase in carbon dioxide (CO2) concentrations has become a primary driver of global warming. Tracking CO2levels in high-emitting enterprises is crucial for achieving carbon neutrality. Although high-precision ground-based sensors provide real-time monitoring, their widespread deployment is limited by cost, leading to data gaps in certain areas. To tackle this issue, we design an aerial-ground cooperative sensing system that enables three-dimensional CO2observation of high-emitting enterprises. The system comprises a long-term monitoring module employing ground-based sensors and a periodic monitoring module using Unmanned Aerial Vehicles (UAVs) with portable sensors. Based on the collected data, we develop a Kriging model to estimate the CO2concentrations at unobserved locations and construct a fine-grained three-dimensional CO2distribution map. The experimental results demonstrate that our proposed scheme can achieve large-scale CO2assessment using limited sensing resources, thereby serving as a valuable reference for effectively regulating corporate CO2emissions. Our implementation has been deployed around a high-emitting enterprise in Nanjing since February 2024.

The electrosurgical generator (ESG) is a medical power supply device that utilizes the thermal effect of high-frequency current flowing through the body tissues. It needs to maintain constant output power when the tissue load varies over a wide range during treatment, which means that wide-output high-frequency voltage should be provided by ESGs. Moreover, the frequency of the output voltage had better be stable to ensure a good therapeutic effect. However, most ESGs control the power by adjusting the frequency or the shape of the output voltage. This paper proposes a two-stage structure with a resonant DC/DC converter as the first stage and a class DE resonant inverter as the second stage. Due to the resonant structure, not only do all the semiconductors achieve ZVS operation, but also the converter is suitable for wide-range voltage regulation. The simulation results and the experimental waveforms verify the theoretical analysis and effectiveness.