Objectives of Programme's Learning Outcomes

• Imagine, design, build and operate machines, equipment and processes to provide a solution to a complex problem of energy production, conversion and transmission by integrating the needs, constraints, context and technical, economic, societal, ethical and environmental issues.
• Implement a chosen solution in the form of a drawing, schematic, diagram or plan that conforms to standards, a model, a prototype, software and/or a digital model.
• Integrate rational energy management.
• Evaluate the approach and results in order to adapt or optimize the proposed solution.
• Mobilize a structured set of scientific knowledge and skills and specialized techniques to meet, with expertise and adaptability, the missions of the civil engineer in energy engineering.
• Identify and discuss potential applications of new and emerging technologies in the energy field.
• Assess the validity of models and results given the state of the science and the characteristics of the problem.
• Plan, manage and carry out projects according to their objectives, resources and constraints, ensuring the quality of activities and deliverables.
• Define and frame the project in terms of its objectives, resources and constraints.
• Work effectively in a team, develop leadership, make decisions in multidisciplinary, multicultural and international contexts.
• Interact effectively with other actors to carry out joint projects in various contexts (multidisciplinary, multicultural and international).
• Communicate and exchange information in a structured manner - orally, graphically and in writing, in French and in one or more other languages - at the scientific, cultural, technical and interpersonal levels, adapting to the goal pursued and the audience concerned.
• Argue and convince, both orally and in writing, in front of a client, a colleague, teachers and juries.
• Act as a responsible, open-minded, and critical professional in an autonomous professional development process.
• Analyze your personal functioning and adapt your professional attitudes.
• Demonstrate openness and critical thinking by comparing the technical and non-technical aspects of the problems analyzed and the solutions proposed.
• Make critical use of the various means available for independent research and training.
• Contribute through research to the innovative solution of a problem in engineering sciences.
• Communicate, in writing and orally, on the process and its results by highlighting the scientific quality criteria of the research carried out, as well as the potential for theoretical or technical innovation and the possible non-technical issues.

Learning Outcomes of UE

Understand and identify the needs in terms of data analytics in modern energy systems.
Master data science technologies for data filtering, analysis and correlation within energy data systems
Master fundamentals of Machine Learning (supervised and unsupervised learning).
Implement and apply these techniques in Python to selected use-cases from the energy systems community.
Process energy data represented with temporal series by machine learning models.

Use the presented approaches like guides and implement them to find a solution to a given problem (case study) in data analysis for energy systems.

UE Content: description and pedagogical relevance

AA ‘Machine Learning and Data Analysis for Energy Systems':
Needs in terms of data analysis in modern energy systems
Data Analysis and filtering
Supervised learning: polynomial regression, classification (logistic regression), model selection and diagnosis (bias-variance trade-off, cross-validation, regularization), introduction to advanced models (neural networks: the MultiLayer Perceptron, etc.),
Unsupervised learning: clustering (K-means, evaluation of partitions), Principal Component Analysis
Application of the concepts on energy use-cases, using the Python programming language.

Theoretical lectures are punctuated by examples that present concrete cases of application and/or are the object of reflection allowing critical analysis of the methods presented through question and answer exchanges with the students.

The practical sessions are designed to develop critical analysis skills. The protocols present the studies to be carried out and the tools for doing so, in the form of a prepared computer code structure, to be modified according to the analysis paths to be chosen proactively, justified by appropriate questioning and then validated by the results obtained. This requires students to be up to date with the material covered in the course.

AA ‘Deep Learning for Energy Systems and Time Series':
Introduction and presentation of deep neural networks ;
Basics and types of recurrent neural networks (RNN, LSTM, GRU, etc.) ;
Optimization and regularization techniques of deep neural networks and RNNs ;
Evaluation metrics of recurrent neural networks.

Prior Experience

Fundamentals of Statistics