Self-Learning Data-Based Models as Basis of a Universally Applicable Energy Management System

Abstract

In the transfer from fossil fuels to renewable energies, grid operators, companies and farms develop an increasing interest in smart energy management systems which can reduce their energy expenses. This requires sufficiently detailed models of the underlying components and forecasts of generation and consumption over future time horizons. In this work, it is investigated via a real-world case study how data-based methods based on regression and clustering can be applied to this task, such that potentially extensive effort for physical modeling can be decreased. Models and automated update mechanisms are derived from measurement data for a photovoltaic plant, a heat pump, a battery storage, and a washing machine. A smart energy system is realized in a real household to exploit the resulting models for minimizing energy expenses via optimization of self-consumption. Experimental data are presented that illustrate the models' performance in the real-world system. The study concludes that it is possible to build a smart adaptive forecast-based energy management system without expert knowledge of detailed physics of system components, but special care must be taken in several aspects of system design to avoid undesired effects which decrease the overall system performance.Energies 2020, 13, 2084 2 of 42 to exploit these installations by increasing self-consumption, i.e., maximizing the amount of locally produced energy that is directly used.However, to perform such a task in an automated way, smart energy systems are needed which are able to determine and execute optimal operation strategies by rescheduling controllable loads and control of storage devices in accordance with the expected energy production profile. In simple system setups such as integrated PV-storage systems, relay-based self-consumption maximization strategies might suffice and are commonly found in practice [6]. For more complex system setups with several controllable devices, in many cases, more advanced optimization strategies are proposed, which require sufficiently detailed models of system components [7]. In the case of a large office building, a farm, or an industrial premise, loads and devices exhibiting very specific or complex consumption behaviors might prohibit the usage of advanced optimization methods due to the effort of deriving and implementing the necessary white or gray box models. This situation was reported by Žáčeková et al. [8] and Sturzenegger et al. [9] for the usage of model predictive control (MPC) in the minimization of the energy demand of office buildings. Furthermore, the application of forecast methods for the expected power generation is investigated in several studies, indicating financial benefits for a smart energy system or its integration into the overall power system (e.g., [6,10,11]). If these benefits are to be exploited in a smart energy system, forecast models for volatile generation and consumption must also be developed, increasing the initial modeling effort even further.In the literature, there are basi...

Towards modelling of energy storages for use in an intelligent energy management system

“…Solving the parameter identification problem of training the model yields optimal parameter values of p * 0 = 5.42 kWh and p * 1 = 0.93 % min . Due to the physical interpretation of these parameters, p 0 can be validated by comparing it to the nominal energy capacity of the battery, published by its manufacturer as 5.5 kWh [3]. For the example dataset, the forecast matches the measured data well (Figure 1) and achieves a MAPE of 10.5 % on the test data, see Table 1 for a detailed overview of the errors.…”

“…The model function with only two parameters used here is numerically equivalent to a first order polynomial approach used by [3]. With such a simple physics-based model, the only difference is that the physical interpretation of the parameters allows…”

“…In the same context, [4] have shown the use of data-based models for battery and milk cooling as energy storage systems, and [5] presented for the battery system an extension into a probabilistic model. While data-based models of a battery were already used in the EMS presented by [3], the milk cooling system is harder to predict based on the available data. Using the same two example systems, this work will cover the approach, results and challenges of instead using physicsbased parameter identification models to forecast energy storages' SoC values.…”

“…This is the setting of the project SmartFarm2 and its predecessor SmartFarm, whose approach is to focus on controlling the existing electricity consuming devices automatically with an intelligent energy management system (EMS). Such an intelligent EMS employs numerical algorithms to compute and execute optimal control strategies for all connected devices in a forecastbased way [3]. This approach requires accurate forecasts of available generated power and power demand for all connected devices as well as the amount of energy stored in energy storages for the respective forecast horizon.…”

Recent Trends and Issues of Energy Management Systems Using Machine Learning

“…The complexity in decision-making for energy production and distribution is attributed to the challenges posed by uncertainties in energy load forecasting. To address these challenges, self-learning data-based forecasting models employing regression and clustering methods have been proposed for cost-efficient operation [70]. Moreover, load forecasting methods based on LSTM and gradient tree boosting algorithms have been developed to design autonomous EMS with integrated uncertainty estimation methods [71].…”