Short-term forecasting of individual household electricity loads with investigating impact of data resolution and forecast horizon

Yildiz, Baran; Bilbao, José Ignacio; Dore, Jonathon; Sproul, Alistair B.

doi:10.1051/rees/2018003

Cited by 18 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Pemilihan metode statistik seperti regresi linier, regresi berganda, moving average, autoregressive moving average, dan lainnya dipilih berdasarkan pada jenis data dan akurasi peramalan. Akurasi peramalan bermanfaat untuk menekan biaya operasi [11] [12] dalam memproduksi daya listrik, pengaturan sistem distribusi transmisi atau perencanaan manajemen energi [3]. Peramalan beban yang berlebihan mengakibatkan kelebihan supply dan sebaliknya kekurangan supply berimbas pada kontinuitas pelayanan [13].…”

Section: Pendahuluanunclassified

Peningkatan Akurasi pada Prediksi Beban Listrik Menggunakan Metode Moving Average

Yuniarti¹,

Wardiman²,

Wirangga³

et al. 2021

JSE

View full text Add to dashboard Cite

This paper discusses improving the accuracy of electrical load forecasting by imputation on empty load data. It is important to estimate the demand for electricity loads for the power plant operating system, fuel supply and maintenance of the power system. The forecast of the electrical load is carried out on the basis of the historical data of electrical load which is generally represented in the load curve. The load curves in research at the Singkarak substation Borang show that there are several load patterns, some missing data and data that is suddenly increasing. The percentage of blank data in 2015 was 1.8379%, while the highest in September at was 0.5137% or 45 hours. To fill in the missing data, three imputation techniques were used, i.e., filling in the data from the previous day's data at the same time ; perform regression analysis on the month the data was missing; and using the mean values from monthly data. The results of forecast using the moving average method provide a forecast of the electrical load on January 1, 2016 wll be 138 kW. The Mean Absolute Error (MAE) for the best load forecast is 9.59, using a data set equipped with the imputation of the mean.

show abstract

Section: Pendahuluanunclassified

Peningkatan Akurasi pada Prediksi Beban Listrik Menggunakan Metode Moving Average

Yuniarti¹,

Wardiman²,

Wirangga³

et al. 2021

JSE

View full text Add to dashboard Cite

show abstract

“…Note that, in (11), the operation diag(·) constructs a diagonal matrix from a vector, and the value of * has no impact on the log-determinant of this Jacobian. In order to get the series of transformations in (6), multiple coupling layers like (9) and (10) are combined in an alternating way to construct a normalizing flow [31]. As a result, the log-determinant of the Jacobian matrix ∂f θ (x) ∂x T in (4) is just a sum of lower-triangular matrices' log-determinants, which makes the efficient computation of the training objective in (5) possible.…”

Section: Flow-based Generative Modelsmentioning

confidence: 99%

“…Despite these advances, residential load forecasting, especially for a single or a small number of households, remains a challenging problem for several reasons. Firstly, individual load naturally exhibits higher volatility L. Zhang compared to a larger aggregation of loads because of the randomness of human behaviors and smaller base loads [7]- [9]. This makes achieving very accurate point forecasts fundamentally difficult and the standard metric of the distance between forecasted and realized values becomes less useful as a figure of merit [10].…”

Section: Introductionmentioning

confidence: 99%

Scenario Forecasting of Residential Load Profiles

Zhang

2020

IEEE J. Select. Areas Commun.

View full text Add to dashboard Cite

Load forecasting is an integral part of power system operations and planning. Due to the increasing penetration of rooftop PV, electric vehicles and demand response applications, forecasting the load of individual and a small group of households is becoming increasingly important. Forecasting the load accurately, however, is considerable more difficult when only a few households are included. A way to mitigate this challenge is to provide a set of scenarios instead of one point forecast, so an operator or utility can consider a range of behaviors. This paper proposes a novel scenario forecasting approach for residential load using flow-based conditional generative models. Compared to existing scenario forecasting methods, our approach can generate scenarios that are not only able to infer possible future realizations of residential load from the observed historical data but also realistic enough to cover a wide range of behaviors. Particularly, the flow-based models utilize a flow of reversible transformations to maximize the value of conditional density function of future load given the past observations. In order to better capture the complex temporal dependency of the forecasted future load on the condition, we extend the structure design for the reversible transformations in flow-based conditional generative models by strengthening the coupling between the output and the conditional input in the transformations. The simulation results show the flow-based designs outperform existing methods in scenario forecasting for residential load by both providing more accurate and more diverse scenarios. arXiv:1906.07373v1 [eess.SY]

show abstract

“…Among those, short-term load forecast is more reliable and efficient. STLF improves the efficiency and reliability of smart grid including home energy management, demand response implementation, electricity price market design [18]- [20], [23]. Two techniques are commonly used for STLF: statistical techniques such as the linear regression model, exponential model etc.…”

Section: Introductionmentioning

confidence: 99%

Short-term Power Load Forecast of an Electrically Heated House in St. John’s, Newfoundland, Canada

Marma

Iqbal

Seary

2020

EJECE

View full text Add to dashboard Cite

A highly efficient deep learning method for short-term power load forecasting has been developed recently. It is a challenge to improve forecasting accuracy, as power consumption data at the individual household level is erratic for variable weather conditions and random human behaviour. In this paper, a robust short-term power load forecasting method is developed based on a Bidirectional long short-term memory (Bi-LSTM) and long short-term memory (LSTM) neural network with stationary wavelet transform (SWT). The actual power load data is classified according to seasonal power usage behaviour. For each load classification, short-term power load forecasting is performed using the developed method. A set of lagged power load data vectors is generated from the historical power load data, and SWT decomposes the vectors into sub-components. A Bi-LSTM neural network layer extracts features from the sub-components, and an LSTM layer is used to forecast the power load from each extracted feature. A dropout layer with fixed probability is added after the Bi-LSTM and LSTM layers to bolster the forecasting accuracy. In order to evaluate the accuracy of the proposed model, it is compared against other developed short-term load forecasting models which are subjected to two seasonal load classifications.

show abstract

Short-term forecasting of individual household electricity loads with investigating impact of data resolution and forecast horizon

Cited by 18 publications

References 16 publications

Peningkatan Akurasi pada Prediksi Beban Listrik Menggunakan Metode Moving Average

Peningkatan Akurasi pada Prediksi Beban Listrik Menggunakan Metode Moving Average

Scenario Forecasting of Residential Load Profiles

Short-term Power Load Forecast of an Electrically Heated House in St. John’s, Newfoundland, Canada

Contact Info

Product

Resources

About