Single‐Site Forecasts for 130 Photovoltaic Systems at Distribution System Operator Level, Using a Hybrid‐Physical Approach, to Improve Grid‐Integration and Enable Future Smart‐Grid Operation

Abstract

With rising shares of photovoltaic (PV) generated electricity in the power grids, PV power forecasting is rapidly gaining importance for energy suppliers and grid operators. But specifically, the distribution system operators (DSOs) require detailed forecasts at higher spatial resolutions, to manage grid congestions and avoid voltage band violations, which are local phenomena. In view of future smart grid operation, the authors propose a forecasting scheme using numerical weather prediction (NWP) data to feed a PV performance model and compute single‐site power forecasts. The intraday outcomes of this rather conventional forecasting scheme are combined with a supervised‐learning model, based on recurrent neural networks (type: long short‐term memory) which is fed by the 15 min measurement data of the two previous days for each PV system. This approach accounts for different local effects, not covered by the NWP‐based physical model, such as partial shading or snow cover. Furthermore, it mitigates the impacts of “imperfect” data on side of the DSOs, used to parametrize the PV performance model. This hybrid‐physical approach has been applied to about 130 PV systems, validated over a 2 year period, and yielded a normalized root‐mean‐square error of 8.3% and 8.7% in 2020 and 2021.

Hybrid-Physical Probabilistic Forecasting for a Set of Photovoltaic Systems using Recurrent Neural Networks

“…In Figure 7, we consider samples from system 44. This system is the second best of our ranking, and has been identified as problematic for the PV performance model because of a double-pitched roof, not reflected by the system description features [KFBB22]. On both sides of Figure 7, the systematic shift of the PV performance model is clearly visible.…”

“…Results are given in Table 1. In [KFBB22], skill scores are computed using a 24h persistence model, adjusted according to the clear sky PV power for the day under consideration. This is a common baseline in the solar energy domain [PC12].…”

“…The underlying hypothesis is that solar irradiance is fairly smooth over limited regional areas, and the production curve specific to a PV system will be mainly influenced by how it converts this solar energy to PV power according to its specifications. In [KFBB22], authors of the present paper introduced a model which performs intra-day probabilistic forecasts of PV power production. It combines the PV performance model above with a model based on Long-Short-Term Memory (LSTM) cells [HS97].…”

Intelligent Modeling and Optimization of Solar Plant Production Integration in the Smart Grid Using Machine Learning Models

“…Another forecasting approach is proposed, integrating numerical weather prediction data with a recurrent neural network (RNN) model fed by 15 min measurements from the preceding 2 days, to enhance photovoltaic power forecasts for distribution system operators in smart grids. [ 7 ] According to the Electric Power Research Institute, solar power plants are at the forefront of the renewable energy sector because of their ability to remotely monitor and safeguard their linked components and to automatically improve their operations. The study suggests three solutions.…”

A novel GBDT-BiLSTM hybrid model on improving day-ahead photovoltaic prediction

“…To improve the forecasting accuracy and remove the reliance on a single weather factor, Ref. 6 considers the weather features comprehensively, including wind force, wind direction, temperature, etc. Furthermore, Ref.…”