Savvas Panagi

Dynamic Flexibility Requests in Local Flexibility Markets: Quantifying the DSO Willingness to Pay

Tue, 04 Aug 2026 00:00:00 +0000

A Benchmarking Case Study for Local Flexibility Markets: Network, Scenarios, and Open Inputs

Mon, 03 Aug 2026 00:00:00 +0000

Towards an open-source real-time operator-training platform: Analysis of computational efficiency

Wed, 01 Jul 2026 00:00:00 +0000

Tuning performance of grey-box models for thermal building applications

Fri, 01 May 2026 00:00:00 +0000

cresROADS 2026

Tue, 14 Apr 2026 00:00:00 +0000

Brought together researchers, TSOs, DSOs, industry partners, and developers working on the future of power system simulation, operation, and digitalization.

Many thanks to CRESYM, TwinEU, and all collaborators involved in these activities.

Enhanced Optimal Power Flow Using a Trained Neural Network Surrogate for Distribution Grid Constraints

Tue, 14 Apr 2026 00:00:00 +0000

2nd DENSE Summer School

Mon, 30 Mar 2026 00:00:00 +0000

Hosted at the École Centrale de Nantes. The program included lab tours and collaborative working package meetings, fostering knowledge exchange and networking among researchers and industry partners.

Coordinated curtailment of uncontrollable distributed energy resources in isolated power systems

Sun, 01 Mar 2026 00:00:00 +0000

A thermal-electrical co-optimization framework for active distribution grids with electric vehicles and heat pumps

Thu, 29 Jan 2026 00:00:00 +0000

PSCC 2026

Thu, 08 Jan 2026 00:00:00 +0000

As power systems become increasingly complex, operator training requires simulation environments that can accurately reproduce system dynamics while meeting strict real-time constraints.

In this work, we evaluate the real-time performance of the open-source dynamic simulator Dynawo as the core engine of a modern operator-training platform. Using a large-scale model of the French transmission system and a range of realistic contingency scenarios, we investigate how network-model fidelity, topology representation, and solver settings affect computational performance.

Increasing the RES Hosting Capacity of the Cyprus Distribution System Focusing on Export Limitation Schemes

Wed, 17 Dec 2025 00:00:00 +0000

Secondment at The University of Manchester

Thu, 20 Nov 2025 00:00:00 +0000

During the secondment, stochastic scheduling and operational techniques were investigated and implemented for the optimal operation of electric vehicles and heat pumps.

Impact of Uncontrolled Electric Vehicle Charging on Unbalanced Suburban Low-Voltage Networks

Mon, 20 Oct 2025 00:00:00 +0000

ISGT Europe 2025

Mon, 20 Oct 2025 00:00:00 +0000

Participated in IEEE ISGT Europe 2025 in Malta and presented a Monte Carlo-based modeling and simulation framework designed to evaluate the impact of uncontrolled electric-vehicle charging on three-phase suburban low-voltage networks, incorporating realistic data from surveys, smart meters, and rooftop PV installations.

Smart Energy Systems Conference 2025

Mon, 22 Sep 2025 00:00:00 +0000

Valuable insights into heating and cooling system modeling and control. Glad to share the experience with DENSE doctoral candidates Hasibuzzaman Mahmud and Milad Mohammadyari.

1st DENSE Summer School

Wed, 10 Sep 2025 00:00:00 +0000

Explored advanced topics in sector-coupled energy system optimization, multi-energy system modeling, and digitalization in sustainable energy systems, including lab tours, collaborative WP meetings, and a final board session.

IEA-DHC Symposium 2025

Mon, 08 Sep 2025 00:00:00 +0000

Participated in the 2025 District Heating and Cooling Symposium and presented research aiming to balance physical interpretability with data-driven adaptability in thermal building models.

Optimal Operation of Electric Vehicles to Enhance the Flexibility Provision in Active Distribution Grids

Sun, 29 Jun 2025 00:00:00 +0000

PowerTech 2025

Sun, 29 Jun 2025 00:00:00 +0000

Introduced a day-ahead optimization framework that leverages EV scheduling to support distribution system operation—offering upward and downward flexibility while respecting both network constraints and user preferences.

Sizing of Fast Frequency Response Reserves for improving frequency security in low-inertia power systems

Sun, 01 Jun 2025 00:00:00 +0000

DTU PES Summer School 2025

Sun, 18 May 2025 00:00:00 +0000

Intensive training on electricity market design, conic relaxations, system reliability with deep learning, hybrid markets for renewable integration, and optimization under uncertainty. Presented Fast Frequency Response for Low Inertia Systems in the poster session.

4th Doctoral Colloquium

Sat, 05 Apr 2025 00:00:00 +0000

Shared progress on flexibility modeling and optimization within the DENSE MSCA project.

1st DENSE Thematic Workshop

Tue, 01 Apr 2025 00:00:00 +0000

Discussed how EVs can support grid operation through intelligent control and optimization. Research collaboration between SPS Lab (CUT) and Electricity Authority of Cyprus.

CRESYM Power System Dynamics Workshop

Mon, 20 Jan 2025 00:00:00 +0000

Covered angle, frequency, and voltage dynamics with case studies and real-time interaction.

Adaptive Reduced System Modeling for Real-time Dynamic Simulations

Wed, 01 Jan 2025 00:00:00 +0000

DENSE

Wed, 01 Jan 2025 00:00:00 +0000

Dependable engineering of Smart Energy Systems (DENSE)

Funding Agency: Horizon Europe MSCA (Marie Skłodowska-Curie Actions Doctoral Network)
Start Date: January 2024
Duration: 4 years
Total budget: €2.6m
Website:
SPS Lab:

What is DENSE?

DENSE (Dependable Smart Energy Systems) is a 4-year Marie Skłodowska-Curie Actions Doctoral Network (HORIZON-TMA-MSCA-DN) funded under Horizon Europe. It brings together 10 partners from applied research, academia, and industry across 6 countries, with the aim of implementing 12 doctoral research projects.

The network trains early-stage researchers in the innovative dependable engineering of Smart Energy Systems (SESs), with a main focus on robustness as well as preventive and corrective actions under uncertainty.

Why it matters

The European Green Deal aims to transform the EU into a climate-neutral, fair, and prosperous society by 2050. DENSE supports this ambition by training and connecting the next generation of researchers and innovators needed to progress toward dependable Smart Energy Systems.

Dependability of complex networks such as SESs characterizes their ability to deliver service that can justifiably be trusted. It comprises attributes such as availability, reliability, safety, integrity, and maintainability. DENSE tackles this challenge with a focus on operational robustness and preventive/corrective actions in SESs.

Research objectives

Methods for designing scalable, hierarchical, and predictive coordination and operation algorithms for SESs at transnational and communal level
Tools for robust data-driven planning, control, and adaptation, including open-source solutions for joint planning and optimization across heat and power
Methods for robust estimation, fault detection, and preventive security management in SESs under uncertainties and disturbances with different time-scales
Software and hardware demonstrators of systems and components with control, management, and analysis tools from individual doctoral projects
Advanced validation methods for performance and functionality assessment of SESs, including heat pumps, green mobility, and cross-sectoral applications in smart communities and buildings

My role

As a Doctoral Candidate in DENSE, my research focuses on the optimal operation of electric vehicles and heat pumps within active distribution networks, in collaboration between the Cyprus University of Technology (SPS Lab) and the Electricity Authority of Cyprus (EAC).

Source

Project page adapted from the description.

Impact of fast frequency response on renewable energy source curtailment and load shedding

Wed, 01 Jan 2025 00:00:00 +0000

TRAISIM

Wed, 01 Jan 2025 00:00:00 +0000

Training Simulator for Power System Operators (TRAISIM)

Funding Agency: CRESYM
Start Date: January 2025
Partners: Cyprus University of Technology (CUT), Réseau de Transport d’Électricité (RTE), Collaborative Research for Energy System Modelling (CRESYM)
Website:
Code:

Overview

TRAISIM is developing an open-source real-time training simulator that helps power system operators prepare for a grid that is becoming more dynamic, more digital, and more demanding to operate.

The project brings together high-fidelity power-system simulation, operator-focused interfaces, and modern digital-twin thinking in a single platform. Its purpose is to make advanced operator training more accessible, transparent, and adaptable for transmission system operators, researchers, and educators.

Built on top of Dynawo, TRAISIM offers an open alternative to closed and vendor-locked training environments. This creates space for collaboration, easier experimentation, and a clearer pathway from research to real operational use.

Why It Matters

Control rooms are evolving quickly. Operators must respond to renewable integration, changing network conditions, automation, and growing system complexity, all while maintaining security and reliability.

TRAISIM is designed to support that transition by offering a realistic environment where operators can train on credible scenarios, test responses to disturbances, and build confidence with tools that reflect the needs of future grids.

The project is also a strong demonstration that open-source technology can support large-scale, real-time operator training without sacrificing ambition or practical relevance.

What TRAISIM Delivers

TRAISIM combines several key elements into one training environment:

a realistic physical simulator for large transmission networks
a control-room style human-machine interface
orchestration tools that align simulation time and operator interaction
trainer and scenario-management functions for guided exercises
protection and automation behaviour for more credible event progression
AI-enabled model adaptation to improve responsiveness where it matters most

Together, these capabilities help create a platform that is realistic enough for serious training, flexible enough for research, and open enough to grow with new operational needs.

Progress So Far

In its first phase, TRAISIM has shown that open-source real-time training for large transmission systems is a realistic and promising direction. The project has already established a strong benchmarking foundation, identified the main performance bottlenecks, and defined a focused roadmap for the next stage of development.

The platform is being demonstrated on a very large transmission network model inspired by the French grid, highlighting TRAISIM’s relevance for realistic TSO-scale applications.

Current work is focused on:

improving simulation speed and robustness
using adaptive model selection to reduce computational burden
strengthening coordination between the simulator and the training environment
preparing reusable methods, software improvements, and research outputs for the wider community

Early results from the adaptive model selection work are especially encouraging, showing major reductions in model size while maintaining very high predictive accuracy.

Publications

TRAISIM’s first public outputs are already available and showcase both the project vision and its early technical progress:

, accepted at PSCC 2026
TRAISIM: Training Simulator for Power System Operators, poster presented at CRESROADS 2026

These publications present the first benchmark results, explain the platform concept, and outline the next steps in solver optimisation, AI-enabled model adaptation, and open-source dissemination.

Synergy with SPS-Lab Research

TRAISIM is closely connected to other SPS-Lab research directions:

IBM (Interpolation-Based Method): improved handling of controller-induced discontinuities in hybrid simulations
Modeling & Simulation: scalable numerical methods for large and complex power-system models
TwinEU (Pilot 8, Task 5.5): digital twin concepts and infrastructure that support the broader training-platform vision

Contact:

Source

Project page adapted from the description.

DENSE Welcome Meeting

Tue, 05 Nov 2024 00:00:00 +0000

Met doctoral candidates and supervisors from the DENSE program.

Comparison of Fast Frequency Response Methods in the Low-Inertia Grid of Cyprus

Mon, 21 Oct 2024 00:00:00 +0000

SyNERGY MED 2024

Mon, 21 Oct 2024 00:00:00 +0000

Influential platform for innovation and collaboration in Mediterranean energy and digital transformation.

Pizza and Presenting your Research

Thu, 17 Oct 2024 00:00:00 +0000

Event organized by the Learning Development Network at CUT.

1st SUNRISE Summer School

Mon, 09 Sep 2024 00:00:00 +0000

Hands-on sessions and lectures supported by Typhoon HIL.

Grid-Forming Control Workshop (L2EP)

Mon, 08 Jul 2024 00:00:00 +0000

Organized by L2EP, Laboratoire d’Électrotechnique et d’Électronique de Puissance de Lille.

Graduation

Wed, 26 Jun 2024 00:00:00 +0000

Graduation day at the National Technical University of Athens — a milestone marking the completion of my studies in Electrical and Computer Engineering.

Grid-Forming Inverters Workshop (AIT)

Thu, 20 Jun 2024 00:00:00 +0000

I participated in the Technical Workshop on Qualification and Validation of Grid Services from Grid-Forming Inverters of Hybrid and Renewable Energy Plants, held on June 20, 2024, in Vienna, Austria. Hosted at the Austrian Institute of Technology (AIT), this workshop offered a profound exploration into the latest advancements and challenges in grid-forming inverter technology.

CIRED 2024 Vienna Workshop

Wed, 19 Jun 2024 00:00:00 +0000

I actively participated and presented a poster showcasing my research on the impact of Fast Frequency Response (FFR) on Renewable Energy Source curtailment and load shedding. The CIRED 2024 workshop in Vienna focuses on increasing distribution network hosting capacity.

Kickoff of MSCA PhD (DENSE)

Sat, 01 Jun 2024 00:00:00 +0000

DENSE is a 4-year Marie Skłodowska-Curie Actions Doctoral Network under Horizon Europe.

DSO-TSO-ESS

Wed, 01 May 2024 00:00:00 +0000

Overview

Project 1: In collaboration with the TSO and DSO of Cyprus, developed technical requirements for integrating Energy Storage Systems (ESS) into the power grid and technical standards for grid-forming inverters to support voltage and frequency regulation.

Focus areas

ESS grid integration: technical requirements for connecting Energy Storage Systems to the Cyprus power system
Grid-forming inverters: technical standards to support voltage and frequency regulation
TSO–DSO collaboration: alignment of requirements across transmission and distribution stakeholders

Context

This work was carried out at the Sustainable Power Systems Laboratory (Cyprus University of Technology), in collaboration with the Transmission System Operator and Distribution System Operator of Cyprus.

ESS and Grid Forming technical requirements

Wed, 01 May 2024 00:00:00 +0000

Started work related to energy storage systems and grid-forming technical requirements, in collaboration with system stakeholders.

My First KNX Home Automation Project: A Step-by-Step Guide

Tue, 06 Feb 2024 00:00:00 +0000

Introduction

This is my first time diving into home automation in a bottom-up building, and I’m starting with something basic, lighting control. The client wanted a smarter and more energy-efficient home, so I decided to use KNX technology to automate the lights. It’s an exciting journey, and I’m eager to share my experience in the world of KNX automation for the first time. This article will provide an in-depth look at the planning, implementation, and configuration stages of the lighting KNX automation project.

Overview of the Project

The project involved installing KNX-compatible sensors and actuators to control the lighting in different rooms. Lighting automation is a fundamental aspect of home automation and provides an excellent introduction to KNX technology. By automating the lighting, I hoped to achieve greater convenience, energy efficiency, and customization options for the client’s home environment. Throughout the project, I focused on ensuring seamless integration between the KNX devices and maintaining flexibility for future expansion into other automation functionalities.

Devices used in the project:

1 x Surge Arrestor
1 x MDT power suply 640mA (STV-064002)
1 x MDT switch actuator 16 fold 16A (AKK-161603)
3 x MDT push button lite 55 4FOLD (BE-TAL550401)
2 x MDT smart push button 6way with screen (BE-TAS8601)
1 x RGB LED controller (AKD-0324V01)

In the KNX installation, a Type 3 surge arrestor plays a crucial role, shielding against low-energy voltage spikes and surges. However, at the main distribution board, it’s imperative to install a Type 1 surge arrestor for comprehensive protection. This robust device not only guards against direct lightning strikes but also adeptly handles high-energy surges, ensuring the resilience of the system. The power supply chosen for the KNX setup has a nominal current of 640 mA, strategically accommodating future expansions with ease. To facilitate seamless scene control, we’ve deployed three push buttons/sensors: two discreetly positioned within the bedroom (one at the entrance of the bedroom and another next to the bed), and a third positioned outside the bathroom. For enhanced functionality, we’ve integrated two advanced sensors/buttons featuring screens—one strategically placed in the corridor and another at the house entrance—enabling intuitive control and monitoring. Complementing these controls, a 16-fold actuator for various loads, ensuring smooth and efficient functionality throughout. Lastly, to elevate the ambiance of the living space, we’ve installed an RGB controller in the living room. This dynamic addition allows us to craft captivating lighting moods, enhancing the overall atmosphere of the house.

Deploying Diverse Regression Machine Learning Algorithms: An Implementation Overview

Sun, 19 Nov 2023 00:00:00 +0000

In this current post, we go into the exploration and analysis of various regression machine learning methods. Our objective is to utilize a range of machine learning techniques to forecast the load demand of the Cyprus power system.

In the initial phase, the ‘data’ dataframe include all essential features and measurements spanning a 16-month period. Throughout the analysis, our focus is directed towards two important features: temperature and hour. While incorporating additional features could potentially enhance result accuracy, for code explanation and methodology simplification, we have opted to retain only these two key features.

import matplotlib.pyplot as plt
import numpy as np

# Create a scatter plot with demand as color
plt.scatter(data['hour'], data['T2M'], c=data['Demand'], cmap='plasma', s=50)

# Add colorbar
cbar = plt.colorbar()
cbar.set_label('Demand (MW)')

# Set labels for axes
plt.xlabel('Hour of the Day')
plt.ylabel('Temperature (°C)')

plt.title('Temperature vs Hour (Color-coded by Demand)')

plt.show()

The first analysis illustrates the fluctuation of demand in megawatts (MW) in response to variations in temperature and hour. In instances of low demand, the figure showcases a darker color, transitioning to brighter hues as demand increases.

Data’s preparation

Prior to normalizing the data, it’s crucial to recognize that the ‘hour’ feature is cyclical. A detailed analysis on handling cyclical features has been discussed in a previous post, accessible . Furthermore, the guidance on scaling data developed in other post , Normalisation and Standardisation method implemented. Moreover, the data’s was divided into test and training data.

from sklearn.preprocessing import StandardScaler
sc=StandardScaler()
data.loc[:,['hour_sin','hour_cos','T2M_norm']]=sc.fit_transform(data.loc[:,['hour_sin','hour_cos','T2M']])

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)

Assessing Machine Learning Regression Models: RMSE and R-squared Evaluation

When it comes to checking how well a machine learning regression model is doing, we use two important measures: RMSE and R-squared. RMSE looks at the average difference between the predicted values and the actual values. If the RMSE is low, it means the model is doing a good job. R-squared, on the other hand, tells us how much of the variability in the data our model can explain. A higher R-squared is better, showing that the model fits the data well. So, by looking at both RMSE and R-squared, we can figure out if our regression model is accurate and how well it explains the data.

from sklearn.metrics import r2_score
from sklearn.metrics import mean_squared_error
# Calculate RMSE
rmse = np.sqrt(mean_squared_error(y_test, X_test[:,5]))

# Calculate R-squared
r_squared = r2_score(y_test, X_test[:,5])

print(f'RMSE: {rmse}')
print("R-squared:", r_squared)

Current TSO’s demand forecast accuracy: RMSE= 30.194821557544707 , R-squared= 0.9700502334373836

1. Multiple Linear Regression

Multiple Linear Regression is a statistical technique that helps us make predictions by analyzing the relationships between multiple variables. Imagine you have a dataset with one outcome you want to predict (let’s call it ‘Y’) and several factors that might influence it (let’s call them ‘X1,’ ‘X2,’ etc.).

Multiple Linear Regression:

$$ Y = b_0 + b_1X_1 + b_2X_2 + \ldots + b_kX_k $$

The goal is to find the best values for b0,b1,…,bk that minimize the difference between our predicted Y and the actual Y in our dataset. Once we find these values, we can use them to make predictions. For example, if we want to predict someone’s weight (Y), we’d use their height (X1), age (X2), and other relevant factors. The coefficients (b0,b1,…,bk) tell us how much Y is expected to change for a one-unit change in each X variable, holding other variables constant. A positive coefficient means an increase in Y with an increase in the corresponding X, and a negative coefficient means the opposite.

Implementation of Multiple Linear Regression

# Training the Simple Linear Regression model on the Training set
from sklearn.linear_model import LinearRegression
MLR = LinearRegression()
MLR.fit(X_train[:,:3], y_train)

# Predicting the Test set results
y_predMLR = MLR.predict(X_test[:,:3])

from sklearn.metrics import r2_score
from sklearn.metrics import mean_squared_error
# Calculate RMSE
rmse = np.sqrt(mean_squared_error(y_test, y_predMLR))

# Calculate R-squared
r_squared = r2_score(y_test, y_predMLR)

print(f'RMSE: {rmse}')
print("R-squared:", r_squared)

2. Polyonimial Regression

Polynomial regression focuses on capturing nonlinear patterns by introducing higher-degree polynomial terms of a single independent variable. Polynomial regression is particularly useful when the relationship between the variables is nonlinear, providing a more accurate fit to the data compared to the linear assumptions of multiple linear regression.

Polynomial Regression:

$$ Y = b_0 + b_1X + b_2X^2 + \ldots + b_nX^n $$

Implementation of Polyonimial Regression

from sklearn.preprocessing import PolynomialFeatures
from sklearn.linear_model import LinearRegression
from sklearn.pipeline import make_pipeline

# Assuming X_train is your feature matrix with 3 columns, and y_train is your target variable

# Create a polynomial regression model with degree 2 (you can change the degree as needed)
degree = 2
PR = make_pipeline(PolynomialFeatures(degree), LinearRegression())

# Train the model
PR.fit(X_train[:,:3], y_train)

# Make predictions
y_predPR = PR.predict(X_test[:,:3])

from sklearn.metrics import mean_squared_error
# Calculate RMSE
rmse = np.sqrt(mean_squared_error(y_test, y_predPR))

# Calculate R-squared
r_squared = r2_score(y_test, y_predPR)

print(f'RMSE: {rmse}')
print("R-squared:", r_squared)

Implementation Support Vector Regression

# Training the SVR model on the whole dataset
from sklearn.svm import SVR
SVR = SVR(kernel = 'rbf')
SVR.fit(X_train[:,:3], y_train)

# Predicting a new result
y_predSVR=SVR.predict(X_test[:,:3])

print(np.concatenate((y_predSVR.reshape(len(y_predSVR),1), y_test.reshape(len(y_test),1)),1))
from sklearn.metrics import mean_squared_error
# Calculate RMSE
rmse = np.sqrt(mean_squared_error(y_test, y_predSVR))
# Calculate R-squared
r_squared = r2_score(y_test, y_predSVR)

print(f'RMSE: {rmse}')
print("R-squared:", r_squared)

Implementation Disicion Tree Regression

# Training the Decision Tree Regression model on the whole dataset
from sklearn.tree import DecisionTreeRegressor
DTR = DecisionTreeRegressor(random_state = 0)
DTR.fit(X_train[:,:3], y_train)
# Predicting a new result
y_predDTR=DTR.predict(X_test[:,:3])
from sklearn.metrics import mean_squared_error
# Calculate RMSE
rmse = np.sqrt(mean_squared_error(y_test, y_predDTR))
# Calculate R-squared
r_squared = r2_score(y_test, y_predDTR)
print(f'RMSE: {rmse}')
print("R-squared:", r_squared)

Implementation Random Forest Regression

# Training the Random Forest Regression model on the whole dataset
from sklearn.ensemble import RandomForestRegressor
RFR = RandomForestRegressor(n_estimators = 10, random_state = 0)
RFR.fit(X_train[:,:3], y_train)
# Predicting a new result
y_predRFR=RFR.predict(X_test[:,:3])
from sklearn.metrics import mean_squared_error
# Calculate RMSE
rmse = np.sqrt(mean_squared_error(y_test, y_predRFR))
# Calculate R-squared
r_squared = r2_score(y_test, y_predRFR)
print(f'RMSE: {rmse}')
print("R-squared:", r_squared)

Results

	TSO’s Forecast	Multiple Linear Regression	Polyonimial Regression	Support Vector Regression	Disicion Tree Regression	Random Forest Regression
RMSE	30.2	109.9	63.1	59.7	27.3	29.4
R-Squared	0.970	0.603	0.869	0.883	0.976	0.972

Your thoughts and questions are important for me. Feel free to share your insights or inquire about anything in the comments section below. Let’s keep the conversation going!

CIGRE Cyprus National Conference 2023

Tue, 14 Nov 2023 00:00:00 +0000

Excited to make my debut at CIGRE Cyprus National Conference 2023! Work focuses on integrating Battery Energy Storage System (BESS) with Fast Frequency Response (FFR) and Frequency Control Reserve (FCR) in the Cyprus power system to address challenges like under-frequency load shedding.

Experience

Tue, 24 Oct 2023 00:00:00 +0000

Basic definition of KNX automations

Sat, 02 Sep 2023 00:00:00 +0000

The KNX Association was founded in 1990, its headquarters are located in Brussels (Belgium) and its original name was “EIB Association”. In general, the aim of the Association was to promote smart homes, smart buildings and in particular the EIB (European Installation Bus) system, which jointly developed by some very well-known manufacturers.

KNX is an alternative method for electrical installation and is widely used in a new buildings. KNX installations are designed to provide centralized control, monitoring, and automation of various devices and systems within a building.The software that is used to programming KNX devices is called ETS.

Difference between KNX and conventional installation

In the case of the use of the most widespread means of transmission of KNX technology, “Twisted Pair”, is installed parallel to the 230V line and a control cable (transmission medium CNS data). This fact, if compared to the conventional technology of electrical installations has the following effects:

When the KNX devices are installed decentralized, the cost of wiring vs of conventional technical facilities is drastically reduced.
The number of system functions increases significantly.
The transparency of the facility increases.

Feature Scaling (Normalisation and Standardisation) / Python implementation

Tue, 29 Aug 2023 00:00:00 +0000

following the previous article about , feature scaling is also important when we are going to use a machine learning algorithms. Is a method to scale the features in a comparative form. But, how can we scaling it ? what is the difference between normalization and standardization ? and when we can use it ? All this questions is gone solve at the current article/blog post.

Equations

Normalization (Min-Max Transform):

$$ X' = \frac{X - X_{\text{min}}}{X_{\text{max}} - X_{\text{min}}} $$

Standardization (Z-Score Transform):

$$ X' = \frac{X - \mu}{\sigma} $$

Base on the very helpful Udemy courses that i have take (Machine learning A-Z) i will try to provide what’s the problem if we are not scaling the features and how scaling can help be giving a simple example. Let’s imagine that we have 2 columns of income and age, also we have 3 different persons as the following table:

Person	Income	Age (Years)
1	70.000€	45
2	60.000€	44
3	52.000€	40

So which 2 of this 3 persons is more similar together? We are clearly sure that the person 1-3 is the worst similar couple that we can take so ignore this pattern. If we try to calculate the difference between the remain combinations, person 1 from 2 has salary difference of 10.000€ and age different of 1 year, but person 2 from 3 has salary different of 8.000€ and age different of 4 years. So numerically the best pair is the person 2-3 which has less totally “distance” (8004<10001). Is this true? No, because is not acceptable to compare income with age. If we want to compare different units of measurement like income with age we have to change the units to a common comparable form.

What is the different of each transform (Normalisation and Standardisation) ?

Normalization, transform the data’s is a scale of [0-1], in contrast the standardisation is not predetermined scale from before and as larger is the standard deviation (σ) as closest to the zero was the standardisation data’s.

When we use each transform ?

Normalisation is recommended when the features follows normal distribution, in contrast the standardisation can be used in the most of situations without any problem. But this universal usage of standardisation does not mean that is always the best choice to scaling the features.

Going back to the problem we have described and try to normalisation the values ,we will have the following table:

Person	Income	Age
1	1	1
2	0.444	0.8
3	0	0

What about the most common pair now?

Person [1-2] = (1-0.444)+(1-0.8) = 0.756

Person [2-3] = (0.444-0)+(0.8-0) = 1.244

So is obvious that the most similar couple is the person 1 with person 2 and not the above results he had calculated before normalization.

To implement this 2 methodologies with python, i will propose to use sklearn.preprocessing library.

Standardisation:

import numpy as np
from sklearn.preprocessing import StandardScaler

# Create a sample dataset with two features (columns)
data = np.array([[10, 2],
 [20, 4],
 [30, 6],
 [40, 8]])

# Create a StandardScaler object
scaler = StandardScaler()

# Fit the scaler to the data and transform the data
normalized_data = scaler.fit_transform(data)

print("Original Data:\n", data)
print("\nNormalized Data:\n", normalized_data)

Normalisation:

import numpy as np
from sklearn.preprocessing import MinMaxScaler

# Create a sample dataset with two features (columns)
data = np.array([[10, 2],
 [20, 4],
 [30, 6],
 [40, 8]])

# Create a MinMaxScaler object
scaler = MinMaxScaler()

# Fit the scaler to the data and transform the data
normalized_data = scaler.fit_transform(data)

print("Original Data:\n", data)
print("\nNormalized Data:\n", normalized_data)

Your thoughts and questions are important for me. Feel free to share your insights or inquire about anything in the comments section below. Let’s keep the conversation going!

How to handle cyclical features, for distance methodologies like k-means machine learning algorithms

Sat, 26 Aug 2023 00:00:00 +0000

Often in situations of use machine learnings methods, we have to consider how to handle the cyclic features. For example in K-Means algorithm it use Euclidean distance in order to sort the available data’s in clusters. In this situations the distance between the hour 0 (00:00) from 23 (23:00) is bigger than what really is. Base on the literature in order to overcome this problem they are use the sin and cosine method to represent each hour in a different cyclic form. With the help of and his corresponding article [1] about i will try to analytical explain the method and with a simple example i will represent how this help in distance measures.

Equations

$$ \begin{align*} x &= \sin\left(\frac{a \cdot 2\pi}{\max(a) + 1}\right) \\ y &= \cos\left(\frac{a \cdot 2\pi}{\max(a) + 1}\right) \end{align*} $$

Base on the above equations we transform real values to a new cyclical form. So, lets think a simple cyclical problem of seasons values and try to solve it for the proof of operation. If we declare the seasons Winter, Spring , Summer and Autumn to a number value it will be 0,1,2 and 3 corresponding. As we know after Autumn the next season is the winter but a distance of 3 (3-0) is the maximum that can occur in our problem if we use the declared number values. After the implementation of transform we have the following form of datas:

Now let’s try to represent the above data’s in x and y axes of scatter plot (x_axes=Sin , y_axes=Cos) :

Finally, how does this help in K-Means algorithm and other distance methods that i will use ? Euclidian distance calculated based on the below formulas.

In our simple problem there are 2D points and after applying the formulas the distance is the following:

So is this Euclidean distance true? if we think the reality this represent exactly how far is each season for each other.

Your thoughts and questions are important for me. Feel free to share your insights or inquire about anything in the comments section below. Let’s keep the conversation going!

References:

[1] :

How to use 0-5v plc to control the speed of high current dc motor

Tue, 22 Aug 2023 00:00:00 +0000

A lot of DIY projects need a programmable microcontroller to control the operation of it. In the current article we will discuss a suitable industrial module to control a dc motor speed. The following article is my inspiration after i am trying to building a small project that need to connect Arduino with motor controller to control the speed.

Base on my personal experience I have bought the following controller:

If the above link is not available any more you can find similar controller by keyword CCMHCN

With this type of controller, you can use PWM Arduino output or some other microcontroller output and control the yellow and red signal from the above image. In the other side of controller you have to connect the motor and the power supply. If we assume that the power supply nominal voltage was 12V, this mean that the 5v input signal gives 12v output signal and 2.5v input signal gives 6v output signal and so on.

In the following code an example of use Arduino to test the operation of the controller is represent. Using serial communication you can give a range of values between 0-10 and it will transform from 0 to 255 PWM output which is 0-5v.

#include "MapFloat.h" //Library Source: https://github.com/radishlogic/MapFloat
float PWMVal = 0;
const int pinOut = 3; // PWM Pin of Arduino Nano

void setup() {
 Serial.begin(9600);
}

void loop() {
 while (Serial.available()>0){
 String myString = Serial.readString(); // Read as String
 float myFloat = myString.toFloat(); // Convert it to float
 float PWMVal = mapFloat (myFloat, 0,10.0,0,255); // 0->0 , 10->255
 analogWrite(pinOut, PWMVal); // Output
 Serial.print("LEVEL = ");
 Serial.println(myFloat);
 Serial.print ("PMW VALUE = ");
 Serial.println (PWMVal);
 Serial.println("----------------------------");
 }
}

Malta Trip

Sat, 19 Aug 2023 00:00:00 +0000

Setting off on a new adventure, my girlfriend and I eagerly ventured to the enchanting island of Malta. Roaming through its picturesque streets, we uncovered a captivating fusion of ancient allure and contemporary energy that the island exuded. From exploring historic sites to savoring the flavors of local delicacies, our journey immersed us in the rich cultural mosaic of Malta. Malta’s unique charm and vibrant spirit made this escapade an unforgettable chapter in our travel story.

Joined SPS Lab as Research Associate

Mon, 15 May 2023 00:00:00 +0000

Joined the Sustainable Power Systems Lab of Cyprus University of Technology and started working on operating marginal policy for frequency reserves in the power system of Cyprus. Main responsibilities included dimensioning the necessary inertia, FFR, FCR, FRR, and RR reserves.

Grigoris Farakos Award

Fri, 24 Mar 2023 00:00:00 +0000

Commendation in honor of Grigoris Farakos, for my distinction in the three highest averages during the 4th year of studies in the direction of energy.

Romania Trip

Sun, 19 Mar 2023 00:00:00 +0000

Embarking on another trip with my girlfriend, this time to Bucharest, Romania, was a truly enriching experience. We wandered through the bustling streets, discovering the perfect blend of old-world charm and modern vibrancy that the city had to offer. Exploring historic landmarks, indulging in local cuisine, and immersing ourselves in the cultural tapestry of Bucharest were highlights of our adventure. The memories we created together in the heart of Romania’s capital will forever hold a special place in our hearts.

Diploma Thesis Presentation

Thu, 09 Mar 2023 00:00:00 +0000

On March 9, 2023, I proudly presented my thesis on the optimal operation of smart grids and energy communities, marking the culmination of my academic journey towards obtaining a Bachelor’s degree. This moment signifies the completion of my academic responsibilities and represents a significant milestone in my pursuit of knowledge and expertise in the field of electrical and computer engineering.

TSOC-OMP

Sun, 29 Jan 2023 00:00:00 +0000

Summary

The growing integration of renewable energy sources is compelling power system operators to decommission conventional power plants, resulting in a loss of inertia and frequency control capabilities. This shift has led the system to operate at marginal or even lower limits of frequency reserves, diminishing overall reliability. Consequently, there is a critical need to resize reserves using innovative techniques and explore new frequency reserve products to mitigate these challenges. This project focuses on the dimensional aspects of Inertia, FFR (Fast Frequency Response), FCR (Frequency Containment Reserve), FRR (Frequency Restoration Reserve), and RR (Restoration Reserve) to address and optimize the system’s response.

Project Objectives

The TSOC-OMP (TSOC Operating Margin Policy) project was initiated to address the critical challenges facing the Cyprus power system due to increasing renewable energy penetration. The main objectives include:

Operating Margin Assessment: Comprehensive review and analysis of the current operating margin policy for the Cyprus Transmission System Operator (TSOC)
Frequency Security Enhancement: Development of strategies to maintain frequency security in low-inertia conditions
Reserve Sizing Optimization: Optimization of various frequency reserve types (FFR, FCR, FRR, RR) for the Cyprus power system
Policy Recommendations: Formulation of new operating margin policies that accommodate the changing generation mix

Methodology

The methodologies used in this project focus on:

Theoretical Analysis

System Limits Calculation: Mathematical modeling of frequency response characteristics under low-inertia conditions
Reserve Requirements Assessment: Analytical determination of minimum reserve requirements for different contingency scenarios

Dynamic Simulations

Time-Domain Analysis: Detailed simulation of frequency dynamics following large disturbances
Scenario Analysis: Evaluation across multiple operating conditions and generation mixes
Performance Metrics: Assessment of frequency nadir, rate of change of frequency (RoCoF), and settling time

Probabilistic Techniques

Risk Assessment: Probabilistic evaluation of frequency security under uncertainty
Monte Carlo Simulations: Statistical analysis of system performance across various scenarios
Reliability Indices: Calculation of frequency-related reliability metrics

Machine Learning Approaches

Data-Driven Optimization: Gradient-descent-based methods for reserve sizing
Dynamic Security Assessment: Integration of DSA with linear predictions for optimal FFR allocation
Pattern Recognition: Identification of critical operating conditions requiring enhanced reserves

Key Findings

The project has yielded several important insights for the Cyprus power system:

Fast Frequency Response (FFR) Optimization: Development of novel data-driven methods for optimal FFR sizing in low-inertia grids
Operating Margin Redefinition: Identification of new margin requirements that account for reduced system inertia
Reserve Coordination: Strategies for effective coordination between different frequency reserve types
System Resilience: Enhanced understanding of system behavior under extreme conditions

Impact and Outcomes

The TSOC-OMP project has contributed to:

Policy Development: Informed the development of new operating margin policies for TSOC
Operational Guidelines: Establishment of practical guidelines for frequency reserve management
Research Advancement: Publication of findings in leading power systems journals
Industry Collaboration: Strengthened collaboration between academia and the Cyprus TSO

The project findings have been published in peer-reviewed journals, including research on , demonstrating the practical application of the developed methodologies in real-world power system operation.

Industrial contacts

Vrahimis Koutsoloukas (TSOC)
Antonis Lazari (TSOC)

Source

Project page adapted from the description.

Internship at Electricity Authority of Cyprus

Sat, 01 Oct 2022 00:00:00 +0000

During my internship at the Electricity Authority of Cyprus (EAC), I became familiar with the Cyprus distributor power system and related operational topics.

Internship at LDK Consultants

Mon, 01 Aug 2022 00:00:00 +0000

At LDK Consultants, I worked as a junior engineer under an internship program offered by my university. My responsibilities were in the field of energy audits and renewable energy projects. Photo captured during an energy audit.

Hungary Trip

Tue, 17 May 2022 00:00:00 +0000

A trip to Hungary with my girlfriend was amazing. We explored Budapest, enjoyed the local food, and took in the sights. It was a special time together that we’ll always remember fondly. This trip may have been the best I’ve ever had.

Netherlands Trip

Sat, 13 Nov 2021 00:00:00 +0000

During my study years, I had the wonderful opportunity to embark on a memorable trip to the Netherlands with my friends. Exploring the charming streets of Amsterdam, cycling along picturesque canals, and immersing ourselves in the rich cultural heritage were highlights of our adventure. We visited world-renowned museums, sampled delicious Dutch cuisine, and soaked in the vibrant atmosphere of this beautiful country.

Favourite tennis court

Fri, 22 Oct 2021 00:00:00 +0000

One of my favourite tennis courts in Athens. Located at the National Technical University of Athens, I used to play more than two days a week while studying.

First attempt at opening a business

Mon, 29 Oct 2018 00:00:00 +0000

In 2018, after a lot of research, I started my first online printing business (Art Your Own). More specifically, I bought the first equipment and started a company printing customized gifts like mugs, T-shirts, keyrings, and many other items. Until today I maintain my involvement with it mostly for entertainment purposes.

Started studies at NTUA

Tue, 09 Oct 2018 00:00:00 +0000

In 2018, I embarked on my academic journey at the National Technical University of Athens (NTUA), where I enrolled in the Department of Electrical and Computer Engineering. This marked the beginning of an exciting chapter in my education, as I delved into the world of electrical engineering and embraced the challenges and opportunities that lay ahead. Photo captured from the building of the Electrical and Computer Engineering department.

Cyprus National Guard

Sat, 09 Sep 2017 00:00:00 +0000

The Cypriot army, known as the Cyprus National Guard, plays a crucial role in safeguarding the sovereignty and security of the island. Conscription is mandatory for male citizens, and service typically lasts for 12 months. I served in the artillery department with the specialty of a communicator.

Lyceum Graduation Day

Fri, 23 Jun 2017 00:00:00 +0000

On my Lyceum graduation day, emotions ran high. Surrounded by classmates and loved ones, receiving that diploma was a moment of immense pride and accomplishment. It marked the end of one journey and the beginning of another, filled with anticipation and excitement for what lay ahead.

When Our Paths Crossed

Fri, 04 Nov 2016 00:00:00 +0000

In 2016, I met my partner. It was a significant moment that brought a positive change to my life. Since then, we’ve built a strong connection and shared many memorable experiences.

Growing Up

Wed, 19 Feb 2003 00:00:00 +0000

While I was celebrating my birthdays at kindergarten!

Childhood Memories

Wed, 19 Jan 2000 00:00:00 +0000

Photo taken from the living room of my parent’s house. The chair is still alive with different color!

Born

Fri, 19 Feb 1999 00:00:00 +0000

I was born in Paralimni, Cyprus. Paralimni is a town located on the eastern coast of the island of Cyprus, near the popular tourist destination of Ayia Napa. In this picture you can see my godfather, my godmother and around me are my sister, my cousins and my aunt.

Savvas Panagi

Dynamic Flexibility Requests in Local Flexibility Markets: Quantifying the DSO Willingness to Pay

A Benchmarking Case Study for Local Flexibility Markets: Network, Scenarios, and Open Inputs

Towards an open-source real-time operator-training platform: Analysis of computational efficiency

Tuning performance of grey-box models for thermal building applications

cresROADS 2026

Enhanced Optimal Power Flow Using a Trained Neural Network Surrogate for Distribution Grid Constraints

2nd DENSE Summer School

Coordinated curtailment of uncontrollable distributed energy resources in isolated power systems

A thermal-electrical co-optimization framework for active distribution grids with electric vehicles and heat pumps

PSCC 2026

Increasing the RES Hosting Capacity of the Cyprus Distribution System Focusing on Export Limitation Schemes

Secondment at The University of Manchester

Impact of Uncontrolled Electric Vehicle Charging on Unbalanced Suburban Low-Voltage Networks

ISGT Europe 2025

Smart Energy Systems Conference 2025

1st DENSE Summer School

IEA-DHC Symposium 2025

Optimal Operation of Electric Vehicles to Enhance the Flexibility Provision in Active Distribution Grids

PowerTech 2025

Sizing of Fast Frequency Response Reserves for improving frequency security in low-inertia power systems

DTU PES Summer School 2025

4th Doctoral Colloquium

1st DENSE Thematic Workshop

CRESYM Power System Dynamics Workshop

Adaptive Reduced System Modeling for Real-time Dynamic Simulations

DENSE

Dependable engineering of Smart Energy Systems (DENSE)

What is DENSE?

Why it matters

Research objectives

My role

Related publications

Source

Impact of fast frequency response on renewable energy source curtailment and load shedding

TRAISIM

Training Simulator for Power System Operators (TRAISIM)

Overview

Why It Matters

What TRAISIM Delivers

Progress So Far

Publications

Synergy with SPS-Lab Research

Source

DENSE Welcome Meeting

Comparison of Fast Frequency Response Methods in the Low-Inertia Grid of Cyprus

SyNERGY MED 2024

Pizza and Presenting your Research

1st SUNRISE Summer School

Grid-Forming Control Workshop (L2EP)

Graduation

Grid-Forming Inverters Workshop (AIT)

CIRED 2024 Vienna Workshop

Kickoff of MSCA PhD (DENSE)

DSO-TSO-ESS

Overview

Focus areas

Context

ESS and Grid Forming technical requirements

My First KNX Home Automation Project: A Step-by-Step Guide

Introduction

Overview of the Project

Devices used in the project:

Deploying Diverse Regression Machine Learning Algorithms: An Implementation Overview

CIGRE Cyprus National Conference 2023

Experience

Basic definition of KNX automations

Difference between KNX and conventional installation

Feature Scaling (Normalisation and Standardisation) / Python implementation

How to handle cyclical features, for distance methodologies like k-means machine learning algorithms

How to use 0-5v plc to control the speed of high current dc motor

Malta Trip

Joined SPS Lab as Research Associate

Grigoris Farakos Award

Romania Trip

Diploma Thesis Presentation

TSOC-OMP

Summary

Project Objectives

Methodology