Data-Driven Domestic Flexible Demand: Observations from experiments in cold climate

TL;DR

Data-driven domestic flexible demand for heating in cold climates is challenged by highly stochastic thermal responses and limited direct HP power control. The study compares Subspace Identification (SI) with one-step and multi-step predictors, the Fundamental Lemma (FL), and Quantile Regression (QR) in an experimental Norwegian home equipped with four heat pumps and IoT sensors, under Nord Pool spot prices. Findings show SI-based methods offer better prediction accuracy and computational efficiency than FL, with MS-SI enabling probabilistic insights via QR to manage peak penalties. The results support scalable, convex-regression–based predictive control (EMPC) for large-scale domestic flexible demand, leveraging thermal inertia and affordable IoT to balance economics and comfort in multi-energy systems.

Abstract

In this chapter, we report on our experience with domestic flexible electric energy demand based on a regular commercial (HVAC)-based heating system in a house. Our focus is on investigating the predictability of the energy demand of the heating system and of the thermal response when varying the heating system settings. Being able to form such predictions is crucial for most flexible demand algorithms. We will compare several methods for predicting the thermal and energy response, which either gave good results or which are currently promoted in the literature for controlling buildings. We will report that the stochasticity of a house response is -- in our experience -- the main difficulty in providing domestic flexible demand from heating. The experiments were carried out on a regular house in Norway, equipped with four air-to-air Mitsubishi heat pumps and a high-efficiency balanced ventilation system. The house was equipped with multiple IoT-based climate sensors, real-time power measurement, and the possibility to drive the HVAC system via the IoT. The house is operating on the spot market (Nord Pool NO3) and is exposed to a peak energy demand penalty. Over a period of three years, we have collected data on the house (temperatures, humidity, air quality), real-time power and hourly energy consumption, while applying various flexible demand algorithms responding to the local energy costs. This has produced large variations in the settings of the heating system and energy demand, resulting in rich data for investigating the house response. This chapter aims at providing important insights on providing flexible demand from houses in cold climates.

Figures (13)

• Figure 1: Experimental house, Jan. 2022, at latitude 63° 25' 50" N.
• Figure 2: Spot market in NO1 (black) and NO3 (red), both histograms (upper graph) and time series (lower graph). The experimental house is situated in NO3.
• Figure 3: Monthly hourly peak energy penalty policy. The red stairs in the upper graph show the different thresholds of monthly hourly peak energy (in kWh) and the corresponding penalty (in NOK). The blue histogram shows the hourly total energy consumption in the experimental house. The lower graph shows the hourly total energy consumption (in red) and the monthly peak hourly energy (in black). The blue levels show the thresholds, numbered by their respective costs (in NOK). The current penalty policy is based on averaging the three peak energy demand in each month.
• Figure 4: Map of the day-ahead hourly spot prices (here for the $10^\mathrm{th}$ of Oct. 2023, 11-12am) in the Nord Pool system. Nord Pool defines the energy and power markets for a large part of Europe (highlighted in green on the map).
• Figure 5: Distribution of the outdoor temperature in the area of the experimental house. Some level of heating is used on a large part of the year.