Imitation and Transfer Q-learning-Based Parameter Identification for Composite Load Modeling

2020 ◽  
pp. 1-1
Author(s):  
Jian Xie ◽  
Zixiao Ma ◽  
Kaveh Dehghanpour ◽  
Zhaoyu Wang ◽  
Yishen Wang ◽  
...  
Keyword(s):  
Parameter Identification ◽  
Load Modeling ◽  
Q Learning

Development of the Power Load Modeling System with Denoising and Parameter Identification

2013 ◽  
Vol 805-806 ◽  
pp. 712-715
Author(s):  
Li Di Wang ◽  
Qing Ying Ge ◽  
Zhe Li ◽  
Tai Gang Nian
Keyword(s):  
Parameter Identification ◽  
Dynamic Model ◽  
Variable Load ◽  
Signal Acquisition ◽  
Load Modeling ◽  
Load Model ◽  
Process Measurement ◽  
Power Load ◽  
Measurement Signal ◽  
Zip Model

The power load modeling system is designed with denoising and parameter identification. This system consists of signal acquisition, signal preprocessing, parameter identification, different load modeling methods such as ZIP model and Dynamic modeling. Original signal can be read from Excel file, which is the simulated signal or measurement signal. Then some kinds of denoising methods can be selected, which are mean filtering, medial filtering and wavelet denoising. After being denoised, the load signal is suitable for the parameter identification process. ZIP model is used to simulate the static load model, and the dynamic model is used to simulate the dynamic load model which is changeable during different periods. With the parameter identification and simulation process, measurement power load signal is used in the experiment, the dynamic model is more suitable for the variable load voltage features description.

Parameter Identification for the Dynamic Load Modeling Based on Denoising Method of the Measurement Data

2011 ◽  
Vol 217-218 ◽  
pp. 907-910
Author(s):  
Li Di Wang ◽  
Jiang Feng Tang ◽  
Jun Sheng Shi
Keyword(s):  
Parameter Identification ◽  
Dynamic Load ◽  
Measurement Data ◽  
Wavelet Denoising ◽  
Load Modeling ◽  
Denoising Method ◽  
Specific Approach ◽  
Identification Process ◽  
Filtering Method ◽  
Mean Filtering

Different denoising methods are used in parameter identification for the dynamic load modeling and the specific approach is proposed. The effects of different denoising methods including mean filtering, medial filtering and wavelet denoising are discussed. Mean filtering method is not helpful to contain the step changes of the measurement voltage, thus is unsuitable for the parameter identification process. Medial filtering method and wavelet denoising methods are suitable for the parameter identification in dynamic load modeling. Furthermore, experiment results based on the measurement data show that the wavelet denoising method is more efficient in some aspects such as the accuracy of identification and SSE.

scholarly journals Two-Stage WECC Composite Load Modeling: A Double Deep Q-Learning Networks Approach

2020 ◽  
Vol 11 (5) ◽  
pp. 4331-4344 ◽  
Author(s):  
Xinan Wang ◽  
Yishen Wang ◽  
Di Shi ◽  
Jianhui Wang ◽  
Zhiwei Wang
Keyword(s):  
Learning Networks ◽  
Load Modeling ◽  
Two Stage ◽  
Q Learning

scholarly journals Robust Time-Varying Parameter Identification for Composite Load Modeling

2019 ◽  
Vol 10 (1) ◽  
pp. 967-979 ◽  
Author(s):  
Chong Wang ◽  
Zhaoyu Wang ◽  
Jianhui Wang ◽  
Dongbo Zhao
Keyword(s):  
Parameter Identification ◽  
Time Varying ◽  
Load Modeling ◽  
Time Varying Parameter

Comparsion of Two Kinds of Optimization Algorithm for Load Model Parameter Identification

2013 ◽  
Vol 380-384 ◽  
pp. 1521-1524
Author(s):  
Li Jie Ding ◽  
Zhou Jing Zhang ◽  
Ying Liu ◽  
Qi Huang ◽  
Jun Wang
Keyword(s):  
Power System ◽  
Parameter Identification ◽  
Optimization Algorithm ◽  
Differential Method ◽  
Numerical Results ◽  
Load Modeling ◽  
Key Technology ◽  
Model Parameter ◽  
Load Model ◽  
Parameter Identifications

Parameter Identification is the key technology in measurement-based load modeling. In order to identify parameters in power system ,the differential method which is based on the multiple curves fitting and interpolated method are compared in the paper. Numerical results illustrate that the differential method can improve the accuracy for load modeling parameter identifications.

Sign in / Sign up

Export Citation Format

Share Document