Equation Based New Methods for Residential Load Forecasting

Abstract

This work proposes two non-linear and one linear equation-based system for residential load forecasting considering heating degree days, cooling degree days, occupancy, and day type, which are applicable to any residential building with small sets of smart meter data. The coefficients of the proposed nonlinear and linear equations are tuned by particle swarm optimization (PSO) and the multiple linear regression method, respectively. For the purpose of comparison, a subtractive clustering based adaptive neuro fuzzy inference system (ANFIS), random forests, gradient boosting trees, and long-term short memory neural network, conventional and modified support vector regression methods were considered. Simulations have been performed in MATLAB environment, and all the methods were tested with randomly chosen 30 days data of a residential building in Memphis City for energy consumption prediction. The absolute average error, root mean square error, and mean average percentage errors are tabulated and considered as performance indices. The efficacy of the proposed systems for residential load forecasting over the other systems have been validated by both simulation results and performance indices, which indicate that the proposed equation-based systems have the lowest absolute average errors, root mean square errors, and mean average percentage errors compared to the other methods. In addition, the proposed systems can be easily practically implemented.

Efficient Energy Management Based on Convolutional Long Short-Term Memory Network for Smart Power Distribution System

“…In [26], the fuzzy logic controller method is proposed for residential load forecasting using two inputs to draw the uncertain relation of the input and output. Furthermore, standard machine learning algorithms have been introduced to deal with the energy load forecasting problem such as random forest with decision trees, linear regression and gradient-boosted trees, which were evaluated on Spanish electricity load data with a tenminute frequency [27].…”

Electricity demand and price forecasting model for sustainable smart grid using comprehensive long short term memory

“…In (Khodaei et al 2018) proposed a multi-objective model for cost emission operation of an industrial consumer with help of fuzzy-based heat and power hub models. The fuzzy logic is a non-linear system that uses IF-THEN logic to work and its' ability to give answers among "true" and "false" is more commonly used in temperature and machine control (Alam and Ali 2020). Therefore, computers cannot explicitly process and analyze the fuzzy system and usually requires the use of specific rules for simulation (Suganthi et al 2015).…”

“…Therefore, computers cannot explicitly process and analyze the fuzzy system and usually requires the use of specific rules for simulation (Suganthi et al 2015). Moreover, it also becomes a slow system since it works by the fuzzy system which relies for each input on the numbers of inputs and the membership function (Alam and Ali 2020).…”

Residential load forecasting by a PSO-tuned ANFIS2 method considering the COVID-19 influence

“…Moreover, the performance of the ANN method highly depends on the conditions such as the availability of a huge amount of previous data or historical data, ensuring a good relationship between the inputs and output and appropriate tuning of the hidden and output layers. However, the adaptive neuro-fuzzy inference system (ANFIS) is reported to show better performance than the ANN system as it combines the beneficial features of both ANN and the fuzzy system (Alam and Ali, 2020b). However, like the fuzzy system, if the number of the input of ANFIS becomes more than three, a sluggish response due to high computational burden makes the system practically nonviable.…”