RESEARCH ON ELECTRICITY CONSUMPTION PREDICTION BASED ON DEEP INTEGRATION MODELS
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Harbin Institute of Technology
Abstract
With the development of power system and the improvement of intelligence
level, power consumption forecast plays a vital role in power system operation
management, energy dispatching and market trading. Effective power demand
forecasting is the key to power system planning and operation, and is of great
significance to achieve safe, efficient and sustainable energy supply. However, the
traditional deep learning model has the problem of falling into local optimal in the
optimization process, which leads to the unstable performance of the model. To
overcome this problem, the particle swarm optimization (PSO) algorithm is improved
in this study to improve the performance of deep learning models in power
consumption prediction tasks.
This study collates and summarizes the challenges encountered when dealing
with nonlinear, non-stationary, and high-dimensional data. To overcome these
challenges, an improved particle swarm optimization (PSO) algorithm is proposed to
optimize the parameters of deep learning models, thereby enhancing the model's
fitting ability and generalization performance. The improved PSO algorithm in this
study adopts dynamic weight adjustment and multi-stage optimization strategy, which
effectively realizes the balance between global search and local search, and greatly
improves the performance of the model in complex power systems. In the process of
model construction, the stack ensemble learning method is adopted, and five machine
learning methods including long short-term memory network (LSTM) are used to
build a deeply integrated power consumption model prediction model.
To verify the validity and applicability of the model, extensive experimental tests
are carried out on real world power system data sets. The experimental results show
that the PSO-Stacking model in this study has a root-mean-square error (RMSE) of
0.095, a mean absolute error (MAE) of 0.074, and an R square (R²) of 0.862, which are
robust performance indicators. These results demonstrate the effectiveness of
improved particle swarm optimization algorithm and stacked ensemble learning
model in power consumption prediction tasks. Compared with the traditional deep
learning model, the optimized deep learning model using the improved PSO algorithm
shows considerable improvement in accuracy, stability and response speed.
Description
This paper mainly discusses the related technologies of power consumption
prediction, proposes an adaptive ensemble learning classification algorithm based on
improved particle swarm optimization (PSO) algorithm, and constructs a deeply
integrated power consumption prediction model through stacked ensemble learning
method. Although LSTM performs well in processing sequence data, its performance
is still limited in the face of complex and heterogeneous data sets. Therefore, in order
to further improve the performance of the model, this paper introduces an improved
version of PSO algorithm to optimize the hyperparameters in the ensemble learning
process, especially adding dynamic weighting factors and compression factors. In this
paper, a variety of machine learning methods, including LSTM networks, are
combined to determine the best combination of different base learners through
adaptive ensemble learning strategies. With this approach, multiple learners are able
to complement each other's strengths and weaknesses, resulting in a powerful
integrated model that effectively improves the accuracy of power consumption
predictions. The experimental results show that compared with traditional methods,
the improved model exhibits higher prediction accuracy and stronger robustness,
especially when dealing with complex and large-scale data sets.
Future research will continue to explore how more advanced machine learning
and deep learning techniques, such as Markov models, can be used to further improve
the accuracy of power consumption predictions. In addition, with the rapid
development of the energy Internet, power consumption forecasting will need to
integrate more data sources, such as grid data, meteorological data, market data, etc.,
to support the efficient operation of the energy Internet. This puts higher demands on
data processing and analysis capabilities, so how to develop more powerful data
processing tools, especially when dealing with large, heterogeneous data sets, remains
a key issue.
Keywords
Deep learning, electricity consumption forecasting, Particle Swarm Optimization algorithm, Optimization of parameter searching for Stacking
Citation
Alkhattabi, Moayad. (2024). Research on Electricity Consumption Prediction Based on Deep Integration Models (Master’s thesis, Harbin Institute of Technology).