Electrical load forecasting: A deep learning approach based on K-nearest neighbors

The paper presents an analysis and classification method to evaluate the working condition of angle grinders by means of infrared (IR) thermography and IR image processing. An innovative method called BCAoMID-F (Binarized Common Areas of Maximum Image Differences—Fusion) is proposed in this paper. This method is used to extract features of thermal images of three angle grinders. The computed features are 1-element or 256-element vectors. Feature vectors are the sum of pixels of matrix V or PCA of matrix V or histogram of matrix V. Three different cases of thermal images were considered: healthy angle grinder, angle grinder with 1 blocked air inlet, angle grinder with 2 blocked air inlets. The classification of feature vectors was carried out using two classifiers: Support Vector Machine and Nearest Neighbor. Total recognition efficiency for 3 classes (TRAG) was in the range of 98.5–100%. The presented technique is efficient for fault diagnosis of electrical devices and electric power tools.

show abstract

“…Sometimes outliers should be removed. The description of the NN classifier can be found in papers [ 33 , 34 , 35 , 36 ].…”

Section: Analyzed States Of the Angle Grindermentioning

confidence: 99%

Ventilation Diagnosis of Angle Grinder Using Thermal Imaging

Głowacz

2021

Sensors

145

View full text Add to dashboard Cite

show abstract

“…A combination of three different techniques, extreme gradient boosting (XGB), light gradient boosting (LEGBM), and multilayer perception (MLP), are selected, the performance superiority is approved, including the training time [ 23 ]. A proposed model competes with the traditional ones based on modifying non-parameter kernel-density estimation to validate the prediction intervals and NSGA-II is adopted for multi-objective optimization [ 24 ]. A new STLF proposed technique based on time series and nonlinear relationship of load data hybrid with the multi-temporal spatial scale method [ 25 ].…”

Section: Related Workmentioning

confidence: 99%

A Real-Time Electrical Load Forecasting in Jordan Using an Enhanced Evolutionary Feedforward Neural Network

Alhmoud

Khurma

Al-Zoubi

et al. 2021

Sensors

View full text Add to dashboard Cite

Power system planning and expansion start with forecasting the anticipated future load requirement. Load forecasting is essential for the engineering perspective and a financial perspective. It effectively plays a vital role in the conventional monopolistic operation and electrical utility planning to enhance power system operation, security, stability, minimization of operation cost, and zero emissions. Two Well-developed cases are discussed here to quantify the benefits of additional models, observation, resolution, data type, and how data are necessary for the perception and evolution of the electrical load forecasting in Jordan. Actual load data for more than a year is obtained from the leading electricity company in Jordan. These cases are based on total daily demand and hourly daily demand. This work’s main aim is for easy and accurate computation of week ahead electrical system load forecasting based on Jordan’s current load measurements. The uncertainties in forecasting have the potential to waste money and resources. This research proposes an optimized multi-layered feed-forward neural network using the recent Grey Wolf Optimizer (GWO). The problem of power forecasting is formulated as a minimization problem. The experimental results are compared with popular optimization methods and show that the proposed method provides very competitive forecasting results.

show abstract

“…One method that has currently stood out for classification and regression is the knearest neighbors (k-NN) [34]. Despite being a method that has been used for several years, many variations of this algorithm are currently being evaluated to improve its capacity.…”

Section: Introductionmentioning

confidence: 99%

Classification of Contaminated Insulators Using k-Nearest Neighbors Based on Computer Vision

et al. 2021

View full text Add to dashboard Cite

Contamination on insulators may increase the surface conductivity of the insulator, and as a consequence, electrical discharges occur more frequently, which can lead to interruptions in a power supply. To maintain reliability in an electrical distribution power system, components that have lost their insulating properties must be replaced. Identifying the components that need maintenance is a difficult task as there are several levels of contamination that are hard to notice during inspections. To improve the quality of inspections, this paper proposes using k-nearest neighbors (k-NN) to classify the levels of insulator contamination based on images of insulators at various levels of contamination simulated in the laboratory. Computer vision features such as mean, variance, asymmetry, kurtosis, energy, and entropy are used for training the k-NN. To assess the robustness of the proposed approach, a statistical analysis and a comparative assessment with well-consolidated algorithms such as decision tree, ensemble subspace, and support vector machine models are presented. The k-NN showed up to 85.17% accuracy using the k-fold cross-validation method, with an average accuracy higher than 82% for the multi-classification of contamination of insulators, being superior to the compared models.

show abstract

Electrical load forecasting: A deep learning approach based on K-nearest neighbors

Cited by 88 publications

References 68 publications

Ventilation Diagnosis of Angle Grinder Using Thermal Imaging

Ventilation Diagnosis of Angle Grinder Using Thermal Imaging

A Real-Time Electrical Load Forecasting in Jordan Using an Enhanced Evolutionary Feedforward Neural Network

Classification of Contaminated Insulators Using k-Nearest Neighbors Based on Computer Vision

Contact Info

Product

Resources

About