With an exploratory study of science education for middle or junior high schools, this paper explains why a physics teacher endeavours to develop an updated and adapted curriculum and pedagogy to support and empower adolescents by initiating their learning in physics and moving on to quantum mechanics at their early ages (Grades 8–9) in a science program during our complex contemporary times. Furthermore, the study discusses how the teacher develops such a curriculum and utilizes technological pedagogical tools to engage and encourage young students, particularly girls, to learn mathematical sciences connected to their everyday lives and lived experiences, not as a school subject but as required scientific knowledge and skills for 21^st-century achievements. Such initiation can create opportunities to learn physics over a longer period of time, broaden girls’ participation, increase their self-confidence, and practise essential STEM skills such as critical thinking, problem-solving, collaboration, innovation, and so forth.

Teaching and learning in physics have long been challenging because the subject involves numerous abstract concepts that require multidimensional higher-order thinking skills. Take free-body diagrams (FBDs) as an example: in drawing them, students need to construct an accurate FBD (with the correct magnitude, direction, and location of force vectors) to represent their conceptual understanding of vectors and motion to calculate the unknown variables. Studies show that many students struggle with vector concepts and misconceptions when drawing FBDs even when such e-learning tools as simulations are available, and the steps for drawing FBDs are suggested. This paper describes a tangible three-dimensional-printed (3D-printed) FBD model that can help students visualize the relationship among force vectors. The proposed 3D-printed FBD model was piloted in a Grade 10 class with 16 students in a Hong Kong secondary school. Analysis of the students’ FBD drawing performance before and after the model’s introduction revealed observable improvements.

Stimulated by market and technologies, the audio industry is booming. Many earphones are no longer satisfactory with simply providing excellent products but tend to provide customers with more opportunities to adjust the tuning, one of which is to add the earphones with different acoustic silencer foam. This study begins by introducing different types of earphones and the history. After that, the specific research is raised by questioning the prevailing perception of acoustic silencer foam in earphones. By comparing and analyzing the differences among the frequency response curves of headphones with different types of acoustic silencer foam, it is concluded that the smaller the cells of acoustic silencer foam and the presence of membrane will increase its ability to reduce the loudness at high frequency. This study reveals two significant characteristics which affect the performance of acoustic silencer foam and provides a reliable theoretical basis for the tuning technology of foam.

Faraday's law of electromagnetic induction is a difficult point in middle school physics teaching. Because multiple variables are involved in students' exploration experiments, students are often confused, especially because traditional experiments cannot effectively obtain mathematical relationships between variables. This study aims at the application of knowledge transfer in deep learning theory, and guides students to modify the experimental device for exploring acceleration in kinematics, adding coils, magnets and other devices. So as to realize the control of variables such as magnetic flux change (ΔΦ), coil turns (n), time change (Δt). This experiment allows students to break the stereotyped influence of existing boundaries in the field of knowledge, and promotes students' ability to transfer knowledge and innovate. The experimental instrument has strong expansibility and provides at least three or more exploration programs, which provides a guarantee for students' diverse exploration programs.

Ultra-short-period (USP) planets are rare Earth-sized exoplanets with the shortest possible orbital periods of all known planets. To date, only slightly over 100 USPs have been detected, as traditional methods used in detecting USPs are often biased and time-consuming. We introduce a novel GPU fast phase folding technique coupled with a Deep Convolutional Neural Network (DCNN) specifically developed for searching for short period planets. The DCNN was trained on a set of 2,000,000 synthetic USP samples and performs exceedingly well in identifying both true and false positive transit signals, with a 99.5% validation accuracy over the training set and a 100% recovery rate of all existing USPs. Compared to the traditional Box Least Squares method, our method is shown to be ~1000 times faster in searching for transit signals in photometric light curves. This neutral network is being used in searching for additional USPs in Kepler data.