Electrical Energy Consumption Forecast Using Support Vector Machines

Energy consumption is a demand for the amount of energy that must supply the building at any given time. Building energy consumption has continued increased over the last few decades all over the world, and Heating, Ventilating, and Air-Conditioning (HVAC), which has a catalytic role in regulating the temperature in the room, mostly accounted for of building energy use. Models created for in this study support vector machine and support vector machine-based models of genetic algorithm to obtain the value of accuracy or error rate or the smallest error value Root Mean Square Error (RMSE) in predicting energy consumption in buildings is more accurate. After testing the two models of support vector machines and support vector machines based on the genetic algorithm is the testing results obtained by using support vector machines where RMSE value obtained was 2,613. Next was the application of genetic algorithms to the optimization parameters C and γ values obtained RMSE error of 1.825 and a genetic algorithm for feature selection error RMSE values obtained for 1,767 of the 7 predictor variables and the selection attribute or feature resulting in the election of three attributes used. After that is done the optimization parameters and the importance of the value of feature selection mistake or error of the smallest RMSE of 1.537. Thus the support vector machine algorithm based on genetic algorithm can give a solution to the problems in the prediction of energy consumption rated the smallest mistake or error.

Download Full-text

A Hybrid Method Based on SVM Integrated Improved PSO Algorithm for Electrical Energy Consumption Forecasting of Crude Oil Pipeline

Volume 3: Operations, Monitoring, and Maintenance; Materials and Joining ◽

10.1115/ipc2020-9434 ◽

2020 ◽

Author(s):

Lei Xu ◽

Lei Hou ◽

Zhenyu Zhu ◽

Yu Li ◽

Ting Lei

Keyword(s):

Support Vector Machine ◽

Particle Swarm Optimization ◽

Energy Consumption ◽

Crude Oil ◽

Electrical Energy ◽

Support Vector ◽

Electrical Energy Consumption ◽

Swarm Optimization ◽

Improved Particle Swarm Optimization ◽

Oil Pipelines

Abstract Electrical energy consumption forecasting of crude oil pipelines plays a critical role in energy consumption target setting, batch scheduling, and unit commitment. For actual crude oil pipelines, because of its uncertainty, nonlinearity, intermittency, fluctuations and complexity, it is challenging to establish the electrical energy consumption forecasting model. And it is difficult to describe the non-linear characteristics of electrical energy consumption forecasting by traditional methods. Therefore, a novel hybrid electrical energy consumption forecasting system based on the combination of support vector machine (SVM) and improved particle swarm optimization (IPSO) is proposed, which includes four parts: data pre-processing part, optimization part, forecasting part, and evaluation part. In the pre-processing stage, in order to avoid large deviation caused by sampling stochasticity of small samples, the training set and the test set are divided by stratified sampling method. During the modeling process, the non-linear relationship in electrical energy consumption forecasting is efficiently represented by support vector machine, and the parameters of support vector machine regression are optimized by the improved particle swarm optimization algorithm. According to the established IPSO-SVM model, evaluation part is conducted to make a comprehensive evaluation for this framework. By comparing the evaluation indicators of IPSO-SVM with that of eight state-of-the-art forecasting methods, the effectiveness of IPSO-SVM method is evaluated. Based on the operation data of four crude oil pipelines in China, the results show that the proposed IPSO-SVM hybrid model has the best forecasting performance than other benchmark models, and its forecasting results are the closest to the actual data. It is concluded that the proposed approach can be an efficient technique for electrical energy consumption forecasting of crude oil pipelines.

Download Full-text