Electrical demand aggregation effects on the performance of deep learning-based short-term load forecasting of a residential building

Shaqour, Ayas; Ono, Tetsushi; Hagishima, Aya; Farzaneh, Hooman

doi:10.1016/j.egyai.2022.100141

Cited by 32 publications

(15 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…e core of the employment recommendation module is based on GRU aggregation [36] of employment preference and ability requirement features; i.e., the aggregation weights are computed adaptively and dynamically using neural networks, which are calculated as follows:…”

Section: Establishment Of Recommendation Systemmentioning

confidence: 99%

The Establishment of College Student Employment Guidance System Integrating Artificial Intelligence and Civic Education

Huang

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

A logical and appropriate job advising system for college students is the only way to address the current dire work scenario for recent graduates in a society that is becoming more and more competitive. The process of building new career advice systems for college students is still fraught with many major issues. Existing problems are manifested in the lack of a perfect employment guidance system for college students and a lack of effective ideological and political education, which leads to the inability of college students to establish a correct employment concept. At the same time, the recommendation method used in the current college student employment guidance system is not reasonable enough and only considers students’ one-way preference for employment units. Then, the history of one user’s employment is likely to lead to unreliable negative samples, which affects the recommendation performance. In addition, the current employment recommendation system cannot fully integrate the service function, management function, education function, and research function of the employment guidance system. In this paper, we address the aforementioned issues and propose an unique career counseling system for college students by combining algorithmic recommendation with ideological and political education. The system adaptively aggregates the preferences and needs of students and employers and fully integrates several functions of the student career competence and career guidance system. The experimental results show that the establishment of a college students’ employment guidance system integrating artificial intelligence (AI) and civic education not only has better interpretability compared with the traditional employment guidance system but also has a higher area under curve (AUC). Furthermore, we also discuss the importance of ideological and political education and referral systems in the work of the career guidance system through ablation experiments. The results show that the AUC values of integrating ideological and political education and recommender systems were higher than those of integrating only ideological and political education or recommender systems. Therefore, establishing an employment guidance system for college students that integrates AI and civic education can effectively guide students’ employment. At the same time, the college student employment guidance system proposed in this paper can offer employment guidance courses in colleges and universities, help graduates to solve their doubts, and make graduates adjust their psychological state in the process of job hunting.

show abstract

Section: Establishment Of Recommendation Systemmentioning

confidence: 99%

The Establishment of College Student Employment Guidance System Integrating Artificial Intelligence and Civic Education

Huang

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…In this part of the study, we collected oil viscosity data from the experimental results from our presented study and previous work, and a robust correlation was developed based on the obtained data to predict oil viscosity for investigated reservoirs oil. Different types of machine learning algorithms, such as black box and white box models, have recently been applied in industry. , However, genetic programming algorithms were applied in this study to build a novel robust correlation for the oil in the examined resource. We divided the data set into 80% training and 20% testing sets.…”

Section: Resultsmentioning

confidence: 99%

“…Different types of machine learning algorithms, such as black box and white box models, have recently been applied in industry. 72,73 However, genetic programming algorithms were applied in this study to build a novel robust correlation for the oil in the examined resource. We divided the data set into 80% training and 20% testing sets.…”

Section: Methodsmentioning

confidence: 99%

Effect of Different Water Content and Catalyst on the Performance of Heavy Oil Oxidation in Porous Media for In Situ Upgrading

Mehrabi-Kalajahi

Hadavimoghaddam

Varfolomeev

et al. 2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Enhanced heavy oil recovery by in situ combustion (ISC) is still restricted as an attractive thermal method because of difficulties with ignition, inefficient combustion, and unstable combustion fronts. This study illustrates how different water contents and CoFe 2 O 4 nanoparticles coated with oleic acid (OA), individually and synergistically, can be used in heavy oil oxidation for enhanced oil recovery through ISC. By means of X-ray diffraction, field-emission scanning electron microscopy, transmission electron microscopy, and dynamic light scattering, we characterized the synthesized bimetallic oxide catalysts with and without oleic acid as a capping agent. The size of the CoFe 2 O 4 and CoFe 2 O 4 @OA nanoparticles was evaluated by the ImageJ method in order to estimate the size distribution in different environments. The size−frequency analysis showed that the dimension of CoFe 2 O 4 nanoparticles in the presence of oleic acid as a capping agent due to uniformity and well distribution decreased to an average size of ∼20 nm. The oxidation of the heavy oil was studied by a self-designed thermo-effect cell containing a porous medium (PMTEC) and a laboratory-scale combustion tube. According to the results, the addition of 20% water content compared to 10% and 30% water content, and oil-dispersed CoFe 2 O 4 @OA compared to CoFe 2 O 4 nanoparticles led to a decrease in both low-temperature oxidation and high-temperature oxidation into lower temperatures through the reduction of energy barriers and the promotion of coke formation. The viscosity of recovered oil decreased from 2071 to 246 mPa•s after oxidation in porous media with a 20% water content and a CoFe 2 O 4 @OA catalyst. In addition, the SARA analysis showed that the oxidation of heavy oil in porous media with water content and catalyst reduced the heavy fractions such as resin and asphaltene contents from 20.9 and 5.9% to 9.7 and 2.5%, respectively, and increased the light fractions including saturate and aromatic contents from 28.7 and 44.3% to 41.3 and 46.5%, respectively. In addition, based on experimental data, a robust correlation was developed to predict oil viscosity using a genetic programming algorithm. The results showed that the developed method could predict oil viscosity with enough accuracy. On the basis of the viscosity reduction data, the combustion efficiency of heavy oil is in turn: heavy oil + CoFe 2 O 4 @OA > heavy oil + continued...

show abstract

“…The ML approach can be classified into three main categories: The first is supervised learning, which maps the relationship between dependent variables and the target variable. It can be used for numerous building energy applications, such as predicting human behavior, risks, energy demand, and renewable energy generation, as well as identifying equipment types and activities [10][11][12] [13]. The second is unsupervised learning, which is primarily used for autonomous pattern recognition in buildings and can be useful for identifying energy lifestyle patterns, consumer types, and anomaly detection [14]…”

Section: Proceedings Of the 8 Th International Exchange And Innovatio...mentioning

confidence: 99%

Recent Advances in Reinforcement Learning Applications for Building Energy Management: A Mini Review.

Shaqour¹,

Hagishima²

2022

Proceedings of International Exchange and Innovation Conference on Engineering &Amp; Sciences (IEICES)

View full text Add to dashboard Cite

In 2019, buildings accounted for 55% of the global electricity demand, making them a key contributor to global emissions and a core target for energy efficiency, energy reduction, and policies and measures promoting renewable energy usage. Reinforcement learning (RL) is an agent-based modelling technique that has proven successful in many applications, particularly in artificial intelligence. RL has attracted research attention owing to its utilization in building energy management (BEM) applications. In this work, the latest research advances that utilize this method are investigated and discussed, primarily its usage in modelling complex building energy problems, building energy consumption control, optimization for comfort and cost savings, and the enhancement of demand forecasting algorithms. Furthermore, the combination of RL with other deep learning methods is discussed. As a state-of-the-art technology in smart grid building applications, RL is applied for control purposes and forecasting enhancement.

show abstract

Electrical demand aggregation effects on the performance of deep learning-based short-term load forecasting of a residential building

Cited by 32 publications

References 44 publications

The Establishment of College Student Employment Guidance System Integrating Artificial Intelligence and Civic Education

The Establishment of College Student Employment Guidance System Integrating Artificial Intelligence and Civic Education

Effect of Different Water Content and Catalyst on the Performance of Heavy Oil Oxidation in Porous Media for In Situ Upgrading

Recent Advances in Reinforcement Learning Applications for Building Energy Management: A Mini Review.

Contact Info

Product

Resources

About