ĐẠI HỌC QUỐC GIA TP.HCM CỘNG HÒA XÃ HỘI CHỦ NGHĨA VIỆT NAM ---------- Độc lập - Tự do - Hạnh phúc TRƯỜNG ĐẠI HỌC BÁCH KHOA KHOA:KH & KT Máy tính ____ NHIỆM VỤ ĐỒ ÁN TỐT NGHIỆP BỘ MÔN: KT MÁY TÍNH ______ Chú ý: Sinh viên phải dán tờ này vào trang nhất của bản thuyết trình HỌ VÀ TÊN: VŨ HOÀNG HẢI _____________________ MSSV: 1952669 ______ NGÀNH: COMPUTER ENGINEERING _______ LỚP: CC19KTM1 ___________ 1. Đầu đề luận án: A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II. Nhiệm vụ (yêu cầu về nội dung và số liệu ban đầu): PHASE 1 • Investigate the problem of scheduling and multi-task multi-team scheduling • Research theory of NSGA-II • Implement a solution for multi-task multi-team scheduling using PMSBX-NSGA- II with with real data from Bien Dong POC • Enhance the algorithm for generalized case • Evaluate the algorithm for a simulated case of multi-battery multitask scheduling PHASE 3 • Deploy the algorithm • Analyze performance parameters Testing on devices • Further algorithm optimization • Investigate simulative models and summarize strategies. PHASE 4 • Convexing and tuning parameters • Verifying effectiveness on devices • Performance evaluation • Even further optimization.
Ngày giao nhiệm vụ luận án: 10/01/2022 4. Ngày hoàn thành nhiệm vụ: 03/06/2022 5. Họ tên giảng viên hướng dẫn: Phần hướng dẫn: 1) PGS. Quản Thành Thơ 100 % 2) ThS.
Mai Đức Trung 100 % Nội dung và yêu cầu LVTN đã được thông qua Bộ môn. CHỦ NHIỆM BỘ MÔN GIẢNG VIÊN HƯỚNG DẪN CHÍNH (Ký và ghi rõ họ tên) (Ký và ghi rõ họ tên) PGS. Phạm Quốc Cường PGS. Quản Thành Thơ TRƯỜNG ĐẠI HỌC BÁCH KHOA CỘNG HÒA XÃ HỘI CHỦ NGHĨA VIỆT NAM KHOA KH & KT MÁY TÍNH Độc lập - Tự do - Hạnh phúc ---------------------------- Ngày tháng năm PHIẾU ĐÁNH GIÁ LUẬN VĂN/ ĐỒ ÁN TỐT NGHIỆP (Dành cho người hướng dẫn/phản biện) 1.
Họ và tên SV: Vũ Hoàng Hải MSSV: 1952669 Ngành (chuyên ngành): Kỹ thuật máy tính 2. Đề tài: A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II. Họ tên người hướng dẫn/phản biện: PGS. Quản Thành Thơ 4.
Tổng quát về bản thuyết minh: Số trang: 117 Số chương: 6 Số bảng số liệu: 14 Số hình vẽ: 48 Số tài liệu tham khảo: 24 Phần mềm tính toán: Không có Hiện vật (sản phẩm): Không có 5. Những ưu điểm chính của LV/ ĐATN: - A brand new method in the field of scheduling and power delivery, that is non-existent in current literature, capable of adaptive to tailored datasets. - PMSBX-NSGA-II is a breakthrough extension of NSGA-II by altering the crossover and mutation genetic operators to a polynomial style expression, perfectly suited with real value dataset context. - The method is applied and tested on comprehensive data gathering methodologies and is able to naturally simulate the complex usage and distribution of LiB to devices.
- The foundation is extended and enhanced from existing published paper in similar scheduling context. Những thiếu sót chính của LV/ĐATN: 7. Đề nghị: Được bảo vệ Bổ sung thêm để bảo vệ Không được bảo vệ 8. Các câu hỏi SV phải trả lời trước Hội đồng: a.
Đánh giá chung (bằng chữ: Xuất sắc, Giỏi, Khá, TB): Điểm : 10 /10 Ký tên (ghi rõ họ tên) TRƯỜNG ĐẠI HỌC BÁCH KHOA CỘNG HÒA XÃ HỘI CHỦ NGHĨA VIỆT NAM KHOA KH & KT MÁY TÍNH Độc lập - Tự do - Hạnh phúc ---------------------------- Ngày tháng năm PHIẾU ĐÁNH GIÁ LUẬN VĂN/ ĐỒ ÁN TỐT NGHIỆP (Dành cho người hướng dẫn/phản biện) 1. Họ và tên SV: Vũ Hoàng Hải MSSV: 1952669 Ngành (chuyên ngành): Kỹ thuật máy tính 2. Đề tài: A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II. Họ tên người hướng dẫn/phản biện: PGS.
Phạm Quốc Cường 4. Tổng quát về bản thuyết minh: Số trang: 117 Số chương: 6 Số bảng số liệu: 14 Số hình vẽ: 48 Số tài liệu tham khảo: 24 Phần mềm tính toán: Không có Hiện vật (sản phẩm): Không có 5. Những ưu điểm chính của LV/ ĐATN: - Self-build datasets used to train the model, which is not only meaningful for the research process of the thesis but also for future related studies. - Proposing the PMSBX-NSGA-II model to help improve local learning more, to solve the problem that the messages of the same class have different content, while the contents of different classes may have similar content that the classification models text type independent of messages encountered.
- Performing experiments to evaluate the PMSBX-NSGA-II model, the results show that the model is higher than the baseline NSGA-II model. Những thiếu sót chính của LV/ĐATN: - The PMSBX-NSGA-II model still has limitations such as the fact that it is not an end-to-end model because the model has to go through a K-SOM training step before it reaches the graph neural network. back-propagation to learn the weights of the text-semantic feature extraction model suitable for the dataset, so that when the dataset is larger, learning the weights to extract the appropriate feature for the dataset whether that is important. Đề nghị: Được bảo vệ Bổ sung thêm để bảo vệ Không được bảo vệ 8.
Các câu hỏi SV phải trả lời trước Hội đồng: a. Đánh giá chung (bằng chữ: Xuất sắc, Giỏi, Khá, TB): Điểm : 9 /10 Ký tên (ghi rõ họ tên) A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II. Vu Hoang Hai Undergraduate Faculty of Computer Science and Engineering project author email hai.vn Quan Thanh Tho, PhD. Associate Professor Faculty of Computer Science and Engineering project instructor email qttho@hcmut.vn Mai Duc Trung Associate Master Faculty of Computer Science and Engineering project instructor email mdtrung@hcmut.vn Pham Cong Thien PhD.
Student Faculty of Computer Science and Engineering project instructor email pcthien.vn Computer Engineering Project (CO4347) A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II. STATEMENT OF ORIGINALITY I hereby declare the following thesis proposal, including all conducted researches, methods of analysis and performance evaluation results, is authentic, legitimate and of personal work, con- ducted under the supervision and guidance of Associate Professor Dr. Quan Thanh Tho, Mr. Mai Duc Trung, and in affiliation with Mr.
Pham Cong Thien. All resources used have been stated in accordance with the rules and regulations applied at Ho Chi Minh University - Department of Computer Science and Engineering, Vietnam National University Ho Chi Minh City, as well as complying with international standards of the Institute of Electrical and Electronics Engineers (IEEE). All quotes, citations and references that were extracted - in full or parts, from various pub- lications, regardless of types and locations are duly indicated in the bibliography section. The intellectual content, graphic and document design, and descriptive expressions used in this these is fully self-designed and belongs to the just author, albeit assistance from external resources.
I fully uphold to the laws and regulations stated in this declaration and will be responsible for any allegation of misconduct, plagiarism or deliberate negligence of the thesis’ authenticity. All third-parties, including the instructors and representatives partaking in this thesis, will not face judgment on behalf. Ho Chi Minh University of Technology, June 2023, Thesis author Vu Hoang Hai Computer Engineering Project (CO4347) A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II. ACKNOWLEDGMENTS Words cannot express my deepest gratitude to the instructors, Dr.
Quan Thanh Tho for his openness, his wealth of knowledge and substance of a genius, continu- ously motivating and supporting the work I have done, pushing boundaries never possible in the realm of my abilities; Mr. Mai Duc Trung for his priceless assistance in revising and evaluating proposed metrics, results, and report advisory; Mr. Pham Cong Thien, PhD Student, for his coaching role in providing research methodologies, breakthrough tools and practical datasets from the headquarters ongoing master projects. I have benefited greatly from them and I am extremely grateful for being the central task force and a reliable, faithful student/partner.
In addition, I would like to give my sincerest thank to the URA Taskforce members their relentless teamwork and partnership spirits that pushes forward the project’s objectives with breakthrough ideas. This endeavor would not have been possible without their support. My parents, family and friends are the corporal driving force for my spirits, motivation, and confidence. They deserve endless love and blessings for their continual support, encouragement, understanding and belief in my accomplishments.
Thank you all for the being the core of my life. Computer Engineering Project (CO4347) A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II. ABSTRACT Forecasting as an art and science is an indispensable tool, used for a wide variety of purposes, both critical and non-critical assessments. One major field in the predictive augmentation is the application of scheduling, and more specifically, the optimization realm.
These types of scheduling problems require accurate detailing of current features, and provide precise information of sched- uled tasks in a future time range, based on multiple constraints and circumstances. These forecasts are applied for auditing, inspecting and administrating maintenance of devices to safeguard them from unforeseeable events. Outlining instructions for scheduled job beforehand provide a sleuth of benefits in terms of efficiency. They allow devices to be flexible on initiating requests, see clear identifiers of bound failures and save on costs, replacement parts, and other relevant expenses.
In an evermore con- nected world, it is essential to have a readily available forecasting system to eliminate challenges that old-fashioned document-style counterparts can incur. In this computer engineering project, the author wishes to implement an advanced algorithm for the problem of optimizing battery energy supply to critical devices at oceanic oil rigs, using PMSBX-NSGA-II. This proposed method is an extension of NSGA-II, enabling deep compatibility with our dataset involving "supply orders". The enhancement of the PMSBX-NSGA-II greatly re- lies on customized genetic operators: polynomial mutation and simulated binary crossover, which are tailored to the switchable multi-source supply apparatus.
The experiment shows significant capabilities of PMSBX-NSGA-II in both ensuring supply demands are met and resources are in- telligently utilized, all at the same time keeping the algorithm fast and stable. Computer Engineering Project (CO4347) A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II. Contents List of Figures iv List of Tables vii A Project introduction 1 1 Problem statement 1 2 Power supply and the importance of adequate regulation 2 3 Project objective, scope and approach 4 3.1 Definitive approach to scheduling. 5 4 Research gap and contributions 6 5 Contents of the project 7 B Preliminaries 8 1 Formal definition 8 1.1 The power supply apparatus .2 Inputs and outputs of the scheduling model .1 Introduction to genetic algorithm .1 Components of a genetic algorithm .2 Sequence of operation of genetic algorithm .3 Operators of genetic algorithm .2 Non-dominated Sorting .1 Multi-objective optimization problem.
20 Computer Engineering Project (CO4347) i A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II.2 Dominance of solutions .3 The Pareto-optimal set .3 Non-dominated sorting algorithm, version II .1 Determining the frontiers .2 Determining crowding distance .3 Summary of steps of NSGA-II. 27 C Related research 28 1 The taxonomy of supply orders scheduling 28 2 Common approaches to scheduling energy supply orders 29 2.1 Brute force algorithm .2 Divide and conquer algorithm. 34 D Scheduling using PMSBX-NSGA-II 38 1 Introduction to PMSBX-NSGA-II 38 2 Techniques of PMSBX-NSGA-II 39 2.1 Traditional bit-string encoding scheme .2 PMSBX-NSGA-II using real parameters .2 The advanced approach: Simulated Binary Crossover .1 Old-school bit-flipping .2 Dealing with polynomials: PMSBX-NSGA-II mutation. 50 3 PMSBX-NSGA-II Framework 51 E Experiments and results 53 Computer Engineering Project (CO4347) ii A multi-objective scheduling approach to power devices from Lithium-ion batteries using PMSBX-NSGA-II.
1 Dataset and the supply resources 53 1.