Unsupervised/supervised learning concept for 24-hour load forecasting

Djukanović, M.; Babic, Blagoje; Šobajić, D.J.; Pao, Yoh‐Han

doi:10.1049/ip-c.1993.0046

Cited by 54 publications

(18 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The output of the forecasting model is the national power consumption. The input delay vector is In_Del = [1,2,3,4,6], and the output one is Out_Del = [0, 1, 2]. The length of training set involves 8600 data lines, which means a bit more than a year.…”

Section: The Forecasting Resultsmentioning

confidence: 99%

An advanced model for electric load forecasting

Nastac

Ulmeanu

2013

2013 IEEE Integration of Stochastic Energy in Power Systems Workshop (ISEPS)

View full text Add to dashboard Cite

The paper describes an extension of the adaptive retraining model applied for electric load forecasting. Previously, the retraining procedure was successfully used for different kinds of applications that includes industrial, biomedical of financial data. This time the data comes from the most important power suppliers at the national level. Several cases are analyzed in order to find suitable parameters for a useful implementation.

show abstract

Section: The Forecasting Resultsmentioning

confidence: 99%

An advanced model for electric load forecasting

Nastac

Ulmeanu

2013

2013 IEEE Integration of Stochastic Energy in Power Systems Workshop (ISEPS)

View full text Add to dashboard Cite

show abstract

“…Several authors have already illustrated how a preventive classi cation of historical load data can enhance forecasting accuracy of A NN models [12,13]. T his can be seen very clearly when addressing holidays and unscheduled power demand.…”

Section: Neural-fuzzy C Lustering Learning M Ethodologymentioning

confidence: 97%

“…T his task has required a week-type classi cation aimed to identify similarities among patterns of power pro les. Recent works have illustrated the eOE ectiveness of an unsupervised learning approach for this task [12].…”

Section: A Pproaches For C Lassi Cationsmentioning

confidence: 99%

“…, x 7 ] represents a week power demand (x : 24 hours load demand), then clustering operation in the unlabeled power demand data X 2 f p is the assignment of labels to the vectors in X . C -partitions of X are a set of [12] clustering algorithm is employed in power demand classi cation, where it then solves an optimization problem formulation given by…”

Section: Power Demand C Lusteringmentioning

confidence: 99%

See 1 more Smart Citation

Electric Load Forecasting Model for the State of Bahrain Network

Qamber¹

2001

Electric Power Components and Systems

View full text Add to dashboard Cite

T he prese nt paper pre sents a c lass o f week se rie s fo recasting manner. T his c lass, emplo yed in the e xtent o f lo ng-te rm e lec tric load fo recasting, is based o n Fuzzy Infere nce C lustering (FIC ) method alo ng with A NN. T he FIC is fo llo wed to arrange the o pe rating wee k co rrec tly. In the pro po sed algo rithm, the FIC was used to e valuate the c lasse s. T he po we r o f the propo sed architecture is sho wn by one wee k in advance o f Bahrain po we r syste m grid. T he unfamiliar use o f informatio n obtaining fro m the arrangem ent stage pe rmits the procedure used in nding a re le vant e nhancement o f the fo recast correc tness fo r irregular load situatio ns. T he e mployed fuzzy c luste ring algorithm is use ful also for linguistic mode ling of an e lec tric power demand used for highe r hierarc hy dec ision syste m. T he algo rithm is e xplained in detail and veri ed through a numerical e xample o f state o f Bahrain netwo rk.

show abstract

“…While it is common practice to develop a single learned model (e.g., a classification or regression model, or a single ensemble of multiple base-learners) to characterize a given dataset, for many problems it is practically advantageous to partition the population into multiple, relatively homogeneous segments and then develop separate models for each segment [2,7,20,18]. For example, an e-tailor may attract different types of browsers, from casual shoppers to bulk purchasers, and one may want to model their purchasing inclinations separately.…”

Section: Introductionmentioning

confidence: 99%

A framework for simultaneous co-clustering and learning from complex data

Deodhar

Ghosh

2007

Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

For difficult classification or regression problems, practitioners often segment the data into relatively homogenous groups and then build a model for each group. This two-step procedure usually results in simpler, more interpretable and actionable models without any loss in accuracy. We consider problems such as predicting customer behavior across products, where the independent variables can be naturally partitioned into two groups. A pivoting operation can now result in the dependent variable showing up as entries in a "customer by product" data matrix. We present a modelbased co-clustering (meta)-algorithm that interleaves clustering and construction of prediction models to iteratively improve both cluster assignment and fit of the models. This algorithm provably converges to a local minimum of a suitable cost function. The framework not only generalizes co-clustering and collaborative filtering to model-based coclustering, but can also be viewed as simultaneous co-segmentation and classification or regression, which is better than independently clustering the data first and then building models. Moreover, it applies to a wide range of bi-modal or multimodal data, and can be easily specialized to address classification and regression problems. We demonstrate the effectiveness of our approach on both these problems through experimentation on real and synthetic data.

show abstract

Unsupervised/supervised learning concept for 24-hour load forecasting

Cited by 54 publications

References 8 publications

An advanced model for electric load forecasting

An advanced model for electric load forecasting

Electric Load Forecasting Model for the State of Bahrain Network

A framework for simultaneous co-clustering and learning from complex data

Contact Info

Product

Resources

About