Electrical load forecasting in power systems based on quantum computing using time series-based quantum artificial...

Area of Science:

• Artificial Intelligence
• Quantum Computing
• Power Systems Engineering

Background:

• Reliable power system operation necessitates precise energy management, challenged by unpredictable consumer behavior and load uncertainties.
• Accurate load forecasting is crucial for efficient energy management, reducing complexity and improving system reliability.
• Existing forecasting methods often struggle with inherent uncertainties in power system data.

Purpose of the Study:

• To implement a quantum computing-based artificial neural network for accurate short-term load forecasting.
• To evaluate a hybrid quantum/classical approach for predicting future load values.
• To demonstrate the potential of quantum artificial intelligence in addressing forecasting challenges within smart grids.

Main Methods:

• A hybrid quantum/classical artificial neural network was developed for load forecasting.
• A time series-based technique was employed, utilizing only historical load data.
• The model was tested on two distinct load types from an experimental laboratory setting.

Main Results:

• The quantum computing-based strategy successfully predicted future load values.
• The hybrid model demonstrated effectiveness in short-term load forecasting.
• Experimental results validated the accuracy of the quantum-enhanced approach.

Conclusions:

• Quantum computing-based artificial intelligence shows significant potential for forecasting applications in smart grids.
• The hybrid quantum/classical neural network offers a promising solution for managing uncertainties in power system load forecasting.
• This approach enhances energy management by providing reliable predictions based solely on historical load data.