XGBoost-Based Day-Ahead Load Forecasting Algorithm Considering Behind-the-Meter Solar PV Generation

With the rapid expansion of renewable energy, the penetration rate of behind-the-meter (BTM) solar photovoltaic (PV) generators is increasing in South Korea. The BTM solar PV generation is not metered in real-time, distorts the electric load and increases the errors of load forecasting. In order to overcome the problems caused by the impact of BTM solar PV generation, an extreme gradient boosting (XGBoost) load forecasting algorithm is proposed. The capacity of the BTM solar PV generators is estimated based on an investigation of the deviation of load using a grid search. The influence of external factors was considered by using the fluctuation of the load used by lighting appliances and data filtering based on base temperature, as a result, the capacity of the BTM solar PV generators is accurately estimated. The distortion of electric load is eliminated by the reconstituted load method that adds the estimated BTM solar PV generation to the electric load, and the load forecasting is conducted using the XGBoost model. Case studies are performed to demonstrate the accuracy of prediction for the proposed method. The accuracy of the proposed algorithm was improved by 21% and 29% in 2019 and 2020, respectively, compared with the MAPE of the LSTM model that does not reflect the impact of BTM solar PV.

Clifford Fuzzy Support Vector Machine for Regression and Its Application in Electric Load Forecasting of Energy System

Electric load forecasting is a prominent topic in energy research. Support vector regression (SVR) has extensively and successfully achieved good performance in electric load forecasting. Clifford support vector regression (CSVR) realizes multiple outputs by the Clifford geometric algebra which can be used in multistep forecasting of electric load. However, the effect of input is different from the forecasting value. Since the load forecasting value affects the energy reserve and distribution in the energy system, the accuracy is important in electric load forecasting. In this study, a fuzzy support vector machine is proposed based on geometric algebra named Clifford fuzzy support vector machine for regression (CFSVR). Through fuzzy membership, different input points have different contributions to deciding the optimal regression hyperplane. We evaluate the performance of the proposed CFSVR in fitting tasks on numerical simulation, UCI data set and signal data set, and forecasting tasks on electric load data set and NN3 data set. The result of the experiment indicates that Clifford fuzzy support vector machine for regression has better performance than CSVR and SVR of other algorithms which can improve the accuracy of electric load forecasting and achieve multistep forecasting.

Correlation between Weather Variables and Electricity Demand

In the industrial era 4.0 as it is today, along with the increasing need for electrical energy, energy efficiency is an essential factor in achieving energy production cost efficiency. Efficiency will be achieved if electricity production can be adjusted to the customer’s electrical load. However, adjusting the electricity production poses a challenge, namely the difficulty of predicting the daily electricity load of customers. There are many factors that affect the electric load. One of the main factors is the weather. Therefore, this study focuses on the correlation of weather parameters on load demand. The weather parameters consist of temperature, rainfall, solar radiation, and wind speed. Case studies were conducted in Bali Island and Central Java Province. This paper looked for correlations between weather parameters with the load demand, especially in the island of Bali and Central Java Province.

A Novel Short-Term Residential Electric Load Forecasting Method Based on Adaptive Load Aggregation and Deep Learning Algorithms

Short-term residential load forecasting is the precondition of the day-ahead and intra-day scheduling strategy of the household microgrid. Existing short-term electric load forecasting methods are mainly used to obtain regional power load for system-level power dispatch. Due to the high volatility, strong randomness, and weak regularity of the residential load of a single household, the mean absolute percentage error (MAPE) of the traditional methods forecasting results would be too big to be used for home energy management. With the increase in the total number of households, the aggregated load becomes more and more stable, and the cyclical pattern of the aggregated load becomes more and more distinct. In the meantime, the maximum daily load does not increase linearly with the increase in households in a small area. Therefore, in our proposed short-term residential load forecasting method, an optimal number of households would be selected adaptively, and the total aggregated residential load of the selected households is used for load prediction. In addition, ordering points to identify the clustering structure (OPTICS) algorithm are also selected to cluster households with similar power consumption patterns adaptively. It can be used to enhance the periodic regularity of the aggregated load in alternative. The aggregated residential load and encoded external factors are then used to predict the load in the next half an hour. The long short-term memory (LSTM) deep learning algorithm is used in the prediction because of its inherited ability to maintain historical data regularity in the forecasting process. The experimental data have verified the effectiveness and accuracy of our proposed method.