Method for improving expansibility of deep neural network power flow model

Abstract

The invention discloses a method for improving the expansibility of a deep neural network power flow model, which mainly comprises the following steps: (1) acquiring basic data of a power system; (2)determining a feature vector; (3) establishing an original DNN power flow model; (4) training the original DNN power flow model to obtain a trained original DNN power flow model; (5) expanding the original DNN power flow model to obtain an expanded DNN power flow model; and (6) calculating the probabilistic power flow of the expanded system to obtain a probabilistic power flow result. The method can be widely applied to the calculation of the probabilistic power flow of the power system, and is particularly suitable for the situation in which the DNN of the original system is not suitable forbeing used due to the expansion of the system.

Description

technical field

[0001] The invention relates to the field of electric power system and automation thereof, in particular to a method for improving the expansibility of a deep neural network power flow model. Background technique

[0002] In recent years, with the vigorous development of renewable energy, the power system is facing more and more uncertainties. These uncertainties have significant impacts on power system operation, planning and control. Probabilistic power flow can fully consider the impact of various uncertainties on system power flow, thus providing valuable information for power system analysis.

[0003] However, the relationship between various uncertain factors and corresponding power flow results is very complicated, which makes it very difficult to solve the probability power flow. Traditional probabilistic power flow calculation methods can be divided into the following two types: analytical method and simulation method. The solution ideas of the ab...

Examples

Embodiment 1

[0103] see Figure 1 to Figure 5 , a method for improving the scalability of a deep neural network power flow model, which mainly includes the following steps:

[0104] 1) Obtain the basic data of the power system.

[0105] 2) Determine the eigenvectors. A DNN is a model that extracts features between sample inputs and outputs. In order for DNN to effectively extract the features of the probabilistic power flow, the input feature vector should contain features related to renewable energy and loads. The output eigenvector should contain the power flow calculation results of interest.

[0106] Further, the main steps to determine the eigenvectors are as follows:

[0107] 2.1) Set uncertain factors, including new energy uncertainty and load uncertainty. Set the power flow results, including the voltage amplitude and phase angle of each node, the active power and reactive power of each branch.

[0108] 2.2) Calculate the active injection power P of node i inj,i , reactive p...

Embodiment 2

[0199] An experiment to verify the method of improving the scalability of the deep neural network power flow model mainly includes the following steps:

[0200] 1) Acquisition of trend samples

[0201] In this embodiment, the IEEE39 node system and the IEEE118 node system are used for simulation. Original system: In the IEEE39 node system, the present invention introduces a wind farm on the busbars 23, 24, and 25. The maximum output of the wind farm is 260MW, and introduces a photovoltaic power station on the busbars 17, 18, and 19. The maximum output of the photovoltaic power station is 200MW. In the IEEE118 node system, the present invention introduces wind farms on bus bars 59, 80 and 90 with a maximum output of 260MW, and introduces photovoltaic power stations on bus bars 13, 14, 16 and 23 with a maximum output of 200MW. Expanded system: In the IEEE39 node system, a new bus 40 and branches 26-40 are added on the basis of the original system. In the IEEE118 node system, a...

technical field

[0001] The invention relates to the field of electric power system and automation thereof, in particular to a method for improving the expansibility of a deep neural network power flow model. Background technique

[0002] In recent years, with the vigorous development of renewable energy, the power system is facing more and more uncertainties. These uncertainties have significant impacts on power system operation, planning and control. Probabilistic power flow can fully consider the impact of various uncertainties on system power flow, thus providing valuable information for power system analysis.

[0003] However, the relationship between various uncertain factors and corresponding power flow results is very complicated, which makes it very difficult to solve the probability power flow. Traditional probabilistic power flow calculation methods can be divided into the following two types: analytical method and simulation method. The solution ideas of the ab...

Examples

Embodiment 1

[0103] see Figure 1 to Figure 5 , a method for improving the scalability of a deep neural network power flow model, which mainly includes the following steps:

[0104] 1) Obtain the basic data of the power system.

[0105] 2) Determine the eigenvectors. A DNN is a model that extracts features between sample inputs and outputs. In order for DNN to effectively extract the features of the probabilistic power flow, the input feature vector should contain features related to renewable energy and loads. The output eigenvector should contain the power flow calculation results of interest.

[0106] Further, the main steps to determine the eigenvectors are as follows:

[0107] 2.1) Set uncertain factors, including new energy uncertainty and load uncertainty. Set the power flow results, including the voltage amplitude and phase angle of each node, the active power and reactive power of each branch.

[0108] 2.2) Calculate the active injection power P of node i inj,i , reactive p...

Embodiment 2

[0199] An experiment to verify the method of improving the scalability of the deep neural network power flow model mainly includes the following steps:

[0200] 1) Acquisition of trend samples

[0201] In this embodiment, the IEEE39 node system and the IEEE118 node system are used for simulation. Original system: In the IEEE39 node system, the present invention introduces a wind farm on the busbars 23, 24, and 25. The maximum output of the wind farm is 260MW, and introduces a photovoltaic power station on the busbars 17, 18, and 19. The maximum output of the photovoltaic power station is 200MW. In the IEEE118 node system, the present invention introduces wind farms on bus bars 59, 80 and 90 with a maximum output of 260MW, and introduces photovoltaic power stations on bus bars 13, 14, 16 and 23 with a maximum output of 200MW. Expanded system: In the IEEE39 node system, a new bus 40 and branches 26-40 are added on the basis of the original system. In the IEEE118 node system, a...

Images