Machine Learning Course for Loss Detection
Online 300 h course in Machine Learning applied to loss detection. 6 modules covering Regulation (EU) 2016/679 (GDPR), Regulation (EU) 2024/1689 on artificial intelligence and Python (scikit-learn, LightGBM). Assigned tutor and open enrolment.
Take the controls.
Run a real engineering trade-off from this field. Move the sliders, watch the model react in real time and beat the challenge.
What you will be able to do
Distinguish and quantify technical and non-technical losses in distribution networks.
Prepare smart meter data for detection models.
Train supervised models to identify fraud.
Apply unsupervised anomaly detection techniques.
Evaluate models using metrics suited to imbalanced classes.
Deploy models in inspection campaigns with ethical and legal criteria.
Develop a complete loss detection project using machine learning.
Who it's for
- Data scientist in utilities looking to specialise in Machine Learning applied to loss detection.
- Energy loss analyst looking to specialise in Machine Learning applied to loss detection.
- Machine learning engineer looking to specialise in Machine Learning applied to loss detection.
- Fraud control manager looking to specialise in Machine Learning applied to loss detection.
- Metering engineer looking to specialise in Machine Learning applied to loss detection.
- Advanced analytics consultant looking to specialise in Machine Learning applied to loss detection.
- Engineering or science graduates seeking to enter electricity distribution.
What you'll work with
- Regulation (EU) 2016/679 (GDPR)
- Regulation (EU) 2024/1689 on artificial intelligence
- IEC 62056 (DLMS/COSEM)
- Royal Decree 1110/2007 on metering points
- ISO/IEC 23894
- ISO/IEC 27001
- IEC 61968-9
6 modules · 300 h
1.1Technical losses in lines and transformers1.2Non-technical losses: fraud and metering errors1.3Energy balance per transformer substation1.4Economic and regulatory impact of losses1.5Types of electricity theft1.6Billing and reading errors1.7Smart meter and concentrator data1.8Loss indicators by zone1.9Field inspection procedures1.10Workshop: energy balance of a transformer substation
2.1Integrating consumption, customer and network data2.2Feature engineering for load curves2.3Meter event variables2.4Handling incomplete series2.5Extreme class imbalance2.6Labelling with inspection results2.7Geographic and socio-economic data2.8Privacy and anonymisation2.9Time-based partitions for validation2.10Laboratory: feature engineering with hourly curves
3.1Logistic regression and decision trees3.2Random forest and gradient boosting3.3XGBoost and LightGBM for classification3.4Neural networks on time series3.5Convolutional networks with load curves3.6Oversampling and weighting techniques3.7Metrics: precision, recall and AUC3.8Gain curves and prioritisation of inspections3.9Interpretability with SHAP3.10Practice: fraud classifier with LightGBM
4.1Anomaly detection with isolation forest4.2Autoencoders for anomalous patterns4.3Customer clustering and atypical profiles4.4Comparison with similar customers4.5Graph models of the low-voltage network4.6Estimation of technical losses by load flow4.7Detection of tampering through meter events4.8Combining expert rules and models4.9Active learning with inspections4.10Case study: anomaly detection without labels
5.1Integration with inspection campaigns5.2Model lifecycle and MLOps5.3Monitoring data drift5.4Measuring field hit rate5.5Economic return on detection5.6Legal and consumer protection aspects5.7Bias and fairness in inspection selection5.8Experiences of distributors in Europe and Latin America5.9Communication with field teams5.10Case study: campaign plan based on the model
6.1Problem and scope definition6.2Data acquisition and preparation6.3Feature engineering6.4Training supervised models6.5Exploration of unsupervised models6.6Evaluation and prioritisation of inspections6.7Estimation of economic benefit6.8Deployment plan6.9Report and presentation6.10Final review with the tutor
How you'll learn
100 % online training on the SEIUM campus, with downloadable material, recorded classes and an assigned technical tutor. There are no fixed timetables: students progress at their own pace within the enrolment period, which makes it possible to combine the course with work.
Each module combines theory, guided reading of the original standard and applied exercises. The working software is Python (scikit-learn, LightGBM), PyTorch, MLflow, Apache Spark, SHAP, in free, academic or evaluation versions. Tests are studied through real training procedures for a fraud classifier using meter data, anomaly detection in load curves without labels, and prioritisation of inspection campaigns. Assessment is continuous by module, plus the final integrative project.
Build it end to end
Your final project follows the real engineering workflow, step by step, reviewed by your tutor.
Distinguish and quantify technical and non-technical losses in distribution networks.
Prepare smart meter data for detection models.
Train supervised models to identify fraud.
Apply unsupervised anomaly detection techniques.
Evaluate models using metrics suited to imbalanced classes.
Where this takes you
Requirements
- No prior degree is required.
- Internet connection and a computer capable of running the programme's software.
- Sufficient Spanish to follow technical regulatory texts.
Ready to start?
Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.
Frequently asked questions
The programme comprises 300 hours in total, spread over 3 to 5 months. That is between 10 and 15 hours a week on average, compatible with full-time work.
We work with Python (scikit-learn, LightGBM), PyTorch, MLflow, Apache Spark, SHAP. Free, academic or evaluation versions are used, and the tutor indicates the open-source alternative where one exists.
An SEIUM own diploma on passing continuous assessment and the integrative project. It is in-house training, not an official qualification from the Ministry.
Yes. Modules passed in a course or postgraduate diploma in the same area are recognised towards the advanced programme. Check the recognition policy when you enrol.
Related programmes
Course in PV and Storage Integration in Buildings
Online 300-hour course in PV and storage integration in buildings. 6 Modules Covering Royal Decree 244/2019 on Self-Consumption, Low Voltage Electrotechnical Regulation (ITC-BT-40) and PVsyst. Assigned…
Course in Reliability and Predictive Maintenance in Wind Energy
Online 300 h course in reliability and predictive maintenance in wind energy. 6 modules covering IEC 61400-25 (communications for monitoring and control), IEC 61400-26-1 (wind turbine…
Course in Power Flow Control with FACTS
Online 300 h course in power flow control with FACTS. 6 modules with Regulation (EU) 2015/1222 (CACM), Regulation (EU) 2017/1485 (SO GL) and MATPOWER. Assigned tutor…
More in Energy and Networks
All Energy and Networks programmes → · How to apply · Fees & financing · Scholarships