This Course Learning Outcome consists of generated ideas for the capstone project using preliminary and conceptual designs, relevant codes and engineering tools that the student has learned from previous course subjects and its application on the current capstone design project.
As we know, proper structuring of course scheme in curriculum contributes to better learning. Thus, the courses that mapped well with design concepts, codes, and engineering tools are of vital importance to the student as the attainment of Course Learning Outcomes is concerned.
For example, as the students have learned and gained sufficient background knowledge and application in the prerequisite subjects, it is ensured that he/she won't be experiencing too much difficulty in the higher subject courses. Similarly, this subject course requires sufficient background of the concepts and tools that the student has grasped from previous years, depending on the research category and priority area that he/she has chosen.
My capstone thesis was titled "Site Specific Wind Mapping using Deep Learning Methods in Extracting Building Characteristics in One City in the Philippines".
The study aimed at quantifying the resistance capacities of building components using drone photogrammetry, a surveying technique using UAV, and machine learning methods, a subpart of AI in extracting building properties.
The capstone project introduces a methodology for site specific wind risk mapping using machine learning and wind simulations for generation of wind fragility curves.
This thesis was divided into two phase: (1) is the use of machine learning, a part of Artificial Intelligence, wherein Electronics Engineering students will make use of computer algorithms to automatically recognize and extract building characteristics such as roof slope, roof type, structural material and building dimensions as seen from a single image, and (2) the use of drone photogrammetry, which involves taking measurement using unmanned aerial vehicles such as drones, which is typically used in surveying and mapping applications and the generation of fragility curves, a probabilistic tool that is used to determine the failure probability of a structure, element or component given a 3-sec gust wind speed from a previous typhoon which hit the selected study area.
These two phases are of vital importance in both Electronics Engineering and Civil Engineering courses for it not only requires the usage of engineering tools such as drones for surveying, programming languages such as python for training the machine learning algorithm, and engineering softwares such as ANSYS for wind analysis of structures and QGIS for analysis of geospatial data of study area, but it also require the expertise and appropriate skills, design concepts that needs to be refreshed by the researchers from previous subjects, and referral codes.
For example, the researchers made use of NSCP provisions in the thesis in calculating gust effect factors, velocity, and wind pressures.
Also, we adapted the HAZUS-Multi Hazard hurricane model, a standardized methodology developed by Federal Emergency Management Agency (FEMA) in defining the damage state of a structural elements that will be subjected to extreme wind speeds using repetitive simulations and structural analysis.
Last September 28, 2022, I had attended a webinar on drone photogrammetry, spearheaded by Engineers Network Technical Training Institute (ENTTI). The webinar tackled the importance of drone photogrammetry to modern land surveying, and the integration of drone photogrammetry in Architecture, Engineering, and Construction Industry. Attending this 3-hour credited webinar with other engineering professionals was such a huge help for me to know and be familiarized with the basic concepts of drone photogrammetry, photogrammetry processing, and its significant benefits to modern land development projects. This will be very useful on our capstone project since we will be using drones to collect thousands of residential building images that will be used for training the machine learning algorithm.
As I crafted the manuscript (Chap I-III), I have realized that I was able to apply the concepts that was taught from previous courses subconsciously. Mastering the gained knowledge and skills from previous courses and applying it to solve world problems is way too different from studying within the premises of a classroom. With that, I am very thankful to my instructors who always make ways to deliver the necessary topics per subject course even though there were class interruptions and challenges faced due to pandemic. I was able to understand how to use engineering softwares, both for surveying and programming use since they were discussed way back in my first and second years in college. Also, I was able to apply the NSCP provisions without spending too much time understanding on how to use and apply them in my current project proposal. And I am hopeful that our capstone project for the next semester will be succesful for this will solve community issue concerning disaster risk management.