scholarly journals Clustering-based forecasting method for individual consumers electricity load using time series representations

2018 ◽  
Vol 8 (1) ◽  
pp. 38-50 ◽  
Author(s):  
Peter Laurinec ◽  
Mária Lucká
Keyword(s):  
Time Series ◽  
Smart Grid ◽  
Series Representation ◽  
Smart Meter ◽  
Grid Data ◽  
Forecasting Accuracy ◽  
Forecasting Methods ◽  
Forecasting Method ◽  
Series Representations ◽  
Electricity Load

Abstract This paper presents a new method for forecasting a load of individual electricity consumers using smart grid data and clustering. The data from all consumers are used for clustering to create more suitable training sets to forecasting methods. Before clustering, time series are efficiently preprocessed by normalisation and the computation of various model-based time series representation methods. Final centroid-based forecasts are scaled by saved normalisation parameters to create the forecast for every consumer. Our method is compared with the approach that creates forecasts for every consumer separately. Evaluation and experiments were conducted on three smart meter datasets from residences of Ireland and Australia, and factories of Slovakia. The achieved results proved that our clustering-based method improves forecasting accuracy mainly for residential consumers.We can also proclaim that it can be found such time series representation and clustering setting that will our forecasting method perform more accurately than fully disaggregated approach. Our method is also more scalable since it is necessary to train the model only on clusters and not for every consumer separately

scholarly journals Load forecasting of refrigerated display cabinet based on CEEMD–IPSO–LSTM combined model

Open Physics ◽  
2021 ◽  
Vol 19 (1) ◽  
pp. 360-374
Author(s):  
Yuan Pei ◽  
Lei Zhenglin ◽  
Zeng Qinghui ◽  
Wu Yixiao ◽  
Lu Yanli ◽  
...  
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Time Series Data ◽  
Load Forecasting ◽  
Series Data ◽  
Forecasting Model ◽  
Combined Model ◽  
Forecasting Accuracy ◽  
Forecasting Method ◽  
Consumption Reduction

Abstract The load of the showcase is a nonlinear and unstable time series data, and the traditional forecasting method is not applicable. Deep learning algorithms are introduced to predict the load of the showcase. Based on the CEEMD–IPSO–LSTM combination algorithm, this paper builds a refrigerated display cabinet load forecasting model. Compared with the forecast results of other models, it finally proves that the CEEMD–IPSO–LSTM model has the highest load forecasting accuracy, and the model’s determination coefficient is 0.9105, which is obviously excellent. Compared with other models, the model constructed in this paper can predict the load of showcases, which can provide a reference for energy saving and consumption reduction of display cabinet.

APPLING GREY FORECASTING METHOD TO FORECAST THE PORTFOLIO’S RATE OF RETURN IN STOCK MARKET OF IRAN

2012 ◽  
Vol 01 (07) ◽  
pp. 01-16
Author(s):  
Ali Mohammadi ◽  
Sara Zeinodin Zade
Keyword(s):  
Time Series ◽  
Stock Market ◽  
Predictive Ability ◽  
Grey Model ◽  
Rate Of Return ◽  
Forecasting Accuracy ◽  
Grey Forecasting ◽  
Best Value ◽  
Forecasting Method ◽  
Time Series Method

Stock market is one of the most important investment market, which influenced by many factors, therefore it needs a robust and accurate forecasting. In this study ,grey model used as a forecasting method and examined if it is the most reliable forecasting method in comparison of time series method. The information of portfolio’s rate of return is gathered from 50 accepted companies in Tehran stock market, which were announced as the best companies last year. Mean Square of the errors (MSE) is computed by different value of α in grey model which could be varied between .1 to .9 ,to examined if α=.5 is the best value that our model could take .Then the predictive ability of the model is compared with different type of time series based forecasting methods Experimental results confirm forecasting accuracy of grey model. Tracking signal is computed for grey model to see whether grey model forecasting is in control or not. At the last portfolio’s rate of return is forecasted for next periods.

Comparison Analysis of Aircraft Aluminum Alloy Corrosion Damage Forecasting Method

2012 ◽  
Vol 268-270 ◽  
pp. 348-351
Author(s):  
Zhi Guo Liu ◽  
Zhi Tao Mu ◽  
Zeng Jie Cai
Keyword(s):  
Time Series ◽  
Aluminum Alloy ◽  
Corrosion Damage ◽  
Neural Net ◽  
Series Method ◽  
Comparison Analysis ◽  
Forecasting Accuracy ◽  
Fitting Method ◽  
Forecasting Method ◽  
Time Series Method

Three different analysis methods was put forward to carried out aircraft aluminum alloy structure corrosion damage forecasting,and comparison analysis of different method which included basic forecasting caculation principle and forecasting accuracy and forecasting extensionality also was discussed.The forecasting calculation result shows that the prediction accuracy of neural net and time series method is higher than the data fitting method,and the prediction extensionality of time series method is the best among the three method which discussed.

The use of the Hurst exponent to investigate the quality of forecasting methods of ultra-high-frequency data of exchange rates

2019 ◽  
Vol 65 (2) ◽  
pp. 200-223
Author(s):  
Robert Szóstakowski
Keyword(s):  
Time Series ◽  
Hurst Exponent ◽  
Foreign Exchange Market ◽  
Fractal Theory ◽  
Data Series ◽  
Time Data ◽  
Forecasting Accuracy ◽  
Forecasting Error ◽  
Forecasting Methods ◽  
Tick Data

Over the last century a variety of methods have been used for forecasting financial time data series with different results. This article explains why most of them failed to provide reasonable results based on fractal theory using one day tick data series from the foreign exchange market. Forecasting AMAPE errors and forecasting accuracy ratios were calculated for statistical and machine learning methods for currency time series which were divided into sub-segments according to Hurst ratio. This research proves that the forecasting error decreases and the forecasting accuracy increases for all of the forecasting methods when the Hurt ratio increases. The approach which was used in the article can be successfully applied to time series forecasting by indicating periods with the optimal values of the Hurst exponent.

scholarly journals Advising an Appropriate Forecasting Method for a Snacks Item (Biscuit) Manufacture Company in Bangladesh

2018 ◽  
Vol 66 (1) ◽  
pp. 55-58
Author(s):  
Nandita Barman ◽  
M Babul Hasan ◽  
Md Nayan Dhali
Keyword(s):  
Time Series ◽  
Inventory Management ◽  
Linear Trend ◽  
Exponential Smoothing ◽  
Demand And Supply ◽  
Forecasting Methods ◽  
Forecasting Method ◽  
Short Term Forecasting ◽  
Exponential Smoothing Method ◽  
Manufacture Company

In this paper, we study the most appropriate short-term forecasting methods for the newly launched biscuit factory produces different types of biscuits. One of them is nut-orange twisted biscuits. As it is a newly launched biscuit factory, it does not use any scientific method to find future demand of their products to produce for the purpose of sales. Having an error free production as well as a good inventory management we try to find an appropriate forecasting method for the sets of data we analyzed for that specific production. Several forecasting methods of time series forecasting such as the Moving Averages, Linear Regression with Time, Exponential Smoothing, Holt‘s Method, Holt-Winter‘s Method etc. can be applied to estimate the demand and supply for these companies. This paper focuses on selecting an appropriate forecasting technique for the newly launched biscuit company. For this, we analyze Exponential Smoothing method as used to time series. We observe from the empirical results of the analysis that if the data has no trend as well as seasonality, Exponential Smoothing Forecasting Method processes as the most appropriate forecasting method for the factory. If the data experiences linear trend in it then Holt’s Forecasting Method processes as the most appropriate forecasting method for the sets of data we analyzed. Dhaka Univ. J. Sci. 66(1): 55-58, 2018 (January)

scholarly journals Aplikasi Metode Grey Forecasting Pada Peramalan Kebutuhan Bahan Bakar Alternatif Ramah Lingkungan di PT. Indocement Tunggal Prakarsa Tbk

2015 ◽  
Vol 14 (2) ◽  
Author(s):  
Nanda Lokita Nariswari ◽  
Cucuk Nur Rosyidi
Keyword(s):  
Time Series ◽  
Alternative Energy ◽  
Historical Data ◽  
Raw Materials ◽  
Time Series Forecasting ◽  
Short Term ◽  
Grey Forecasting ◽  
Forecasting Methods ◽  
Forecasting Method ◽  
A Company

<span><em>Forecasting is one of the methods required by a company to plan the demand of raw materials in the </em><span><em>future, in order to avoid the emergence of various problems such as stock out. However, not all </em><span><em>forecasting methods can be used to forecast demand in the short term a specially a condition where the </em><span><em>company only has a few historical data. Grey method is a forecasting method which can be used to </em><span><em>predict the short-term demand. The purpose of this study is to determine how well the Grey method used </em><span><em>to predict the demand of alternative energy and compared with other forecasting methods. Mean Squared </em><span><em>Error (MSE) is used as a measure of the goodness of the method. The result of the study indicates that the </em><span><em>Grey Forecasting Methods MSE value that is smaller than other time series forecasting methods.</em></span></span></span></span></span></span></span><br /></span>

scholarly journals APLIKASI METODE ARIMA BOX-JENKINS UNTUK MERAMALKAN KASUS DBD DI PROVINSI JAWA TIMUR

2019 ◽  
Vol 13 (2) ◽  
pp. 183
Author(s):  
Muhammad Bintang Pamungkas
Keyword(s):  
Time Series ◽  
Significance Test ◽  
Time Series Forecasting ◽  
Forecasting Model ◽  
Upward Trend ◽  
Independent Variables ◽  
Forecasting Methods ◽  
Forecasting Method ◽  
Dependent Variables ◽  
Seasonal Data

The Box-Jenkins forecasting method is one of the time series forecasting methods. This method uses past values as dependent variables and independent variables are ignored. Box-Jenkins (ARIMA) method has advantages that can be used on non-stationary data, can be used on all data patterns including seasonal data patterns so this method can be used to predict cases of DHF in East Java Province. This research was conducted to determine the best model with seasonal ARIMA forecasting model and also to analyze the result of DHF case forecasting in East Java Province. The analysis result shows that the best model for DHF case in East Java Province is ARIMA (1,1,2)(2,1,1)12. The best model has fulfilled the test requirement that is parameter significance test and diagnostics check. Forecasting results show the number of DHF cases in 2017-2018 will experience an upward trend. The total number of DHF cases in 2017 was 14,277 cases and increased to 22,284.54 DHF cases in 2018. The forecasting results showed that the highest peak of DHF cases occurred in January 2017 with 1,914.22 cases and then decrease in the next month until the lowest case occurred in October with 768.46. The forecast for 2018 also shows that the highest DHF cases occurred in January with 3455.55 and declined to the lowest in October with 1126.49 cases. MAPE value in the forecast is 43.51%. The MAPE value indicates that the forecasting is good enough, adequate and feasible to use.

scholarly journals Proposed Improvement of Forecasting Using Time Series Forecasting of Fast Moving Consumer Goods

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
Tasya Regina ◽  
Panca Jodiawan
Keyword(s):  
Time Series ◽  
Historical Data ◽  
Moving Average ◽  
Time Series Forecasting ◽  
Class A ◽  
Forecasting Methods ◽  
Forecasting Method ◽  
Double Exponential ◽  
Data Points ◽  
Double Exponential Smoothing

<p>The company discussed in this paper is a national distributor firm that distributes FMCG products. The PPIC division in the company is responsible for forecasting the demand using the combination of the moving average method and intuition according to the interest of the company. However, the PPIC staff never measures the accuracy of their forecasting method. This research paper aims to evaluate the forecasting methods used to predict the demands of 12 classes of A SKU. Four-time series forecasting methods are particularly implemented, i.e., ARIMA, moving average (MA), double exponential smoothing (DES), and linear regression (RL). Forecasting using the ARIMA method is carried out by considering the stationarity of the average and variance of the historical data points. Forecasting using DES is carried out by using the optimal alpha and gamma values of the ARIMA method. The results show that the performance of each forecasting method varies, depending on which demands of class A SKU are predicted. Based on these results, the current forecasting method utilized by the company should be improved using the time series forecasting methods leading to the smallest error values for each class of A SKU.</p>

Sign in / Sign up

Export Citation Format

Share Document