Dynamic Load Modeling Based on Extreme Learning Machine

2012 ◽  
Vol 195-196 ◽  
pp. 1043-1048
Author(s):  
Zhong Hui Liu ◽  
Zhen Shu Wang ◽  
Mei Hua Su
Keyword(s):  
Power System ◽  
Extreme Learning Machine ◽  
Dynamic Simulation ◽  
Dynamic Load ◽  
Reactive Power ◽  
Power Model ◽  
Load Modeling ◽  
Load Model ◽  
Load Characteristics ◽  
Learning Machine

The dynamic load characteristics have significant impact on the power flow, transient stability computation, voltage stability calculation of the power system, and so on. Noticing that traditional mechanism loads model has difficulty in precisely describing the dynamic characteristics of synthetic load, this paper presents a non-mechanism dynamic load model based on Extreme Learning Machine (ELM). The Power Fault Recorder and Measurement System (PFRMS) is used to obtain data for load modeling. Take voltage and real/reactive power with different time delay as inputs, and take real/reactive power as output, train the ELM using the samples formed by fault data, the real power model and reactive power model are established respectively. The number of hidden layer nodes which has impact on the ELM model is also discussed. Dynamic simulation experiment is conducted at power system dynamic simulation laboratory. The simulation result shows that the ELM load model is simple and flexible, its parameters are easy to be identified. The ELM load model can describe the dynamic load characteristics accurately.

An Investigative Study on Impact of Frequency Dynamics in Load Modeling

2020 ◽  
Vol 01 (01) ◽  
Author(s):  
Musa Mohammed ◽  
◽  
Abubakar Abdulkarim ◽  
Adamu Sa’du Abubakar ◽  
Abdullahi Bala Kunya ◽  
...  
Keyword(s):  
Power System ◽  
Reactive Power ◽  
Low Voltage ◽  
Model Performance ◽  
System Stability ◽  
Normal System ◽  
Area Network ◽  
Load Modeling ◽  
Feed Forward Neural Network ◽  
Load Model

Load modeling plays a significant impact in assessing power system stability margin, control, and protection. Frequency in the power system is desired to be kept constant, but in a real sense, it is not constant as loads continually change with time. In much literature, frequency dynamics are ignored in the formulation of load models for the basic assumption that it does not affect the models. In this paper, the composite load model was formulated with Voltage-Frequency Dependency (V-FD) on real and reactive powers and applied to estimate the load model. 2- Area network 4- machines Kundur test network was used for testing the developed model. The model was trained with measurements from a low voltage distribution network supplying the Electrical Engineering department at Ahmadu Bello University, Zaria. Both training and testing data were captured under normal system operation (dynamics). To evaluate the V-FD model performance, Voltage-Dependent (VD) model was examined on the same measured data. The work makes use of the Feed Forward Neural Network (FFNN) as a nonlinear estimator. Results obtained indicate that including frequency dynamics in modeling active power reduces the accuracy of the model. While in modeling reactive power the model performance improves. Hence, it can be said that including frequency dynamics in load modeling depends on the intended application of the model.

scholarly journals Study of power flow in electricity system using extreme learning machine

2021 ◽  
Vol 926 (1) ◽  
pp. 012028
Author(s):  
M Darwis ◽  
I C Gunadin ◽  
S M Said
Keyword(s):  
Power System ◽  
Extreme Learning Machine ◽  
Power Flow ◽  
Reactive Power ◽  
Flow Analysis ◽  
Load Flow ◽  
Machine Method ◽  
Fast Processing ◽  
Learning Machine ◽  
Hidden Layer

Abstract Load Flow or Power Flow Analysis in the power system in used to determine the power system parameters such as voltage, current, active power, and reactive power contained in the power grid. The method that has long been used in the calculation of load flow or power flow is the Newton-Raphson iteration method. As for its development, to complete the power flow study, it is carried out by implementing the Artificial Intelligence method, one of which is the Extreme Learning Machine method. This method is used in the simulation of the simple 39 Bus system calculation from IEEE. In this Extreme Learning Machine, the testing analysis is carried out with 2 inputs, 1 hidden layer, 5 neurons, and 2 outputs and the number of datasets is 39 to produce MAE and MAPE respectively 2.02 and 0.76% and with a very fast processing time of 0.010s

scholarly journals Dynamic load modeling of some low voltage devices

2009 ◽  
Vol 22 (1) ◽  
pp. 61-70 ◽  
Author(s):  
Lidija Korunovic ◽  
Dobrivoje Stojanovic
Keyword(s):  
Induction Motor ◽  
Dynamic Load ◽  
Low Voltage ◽  
Supply Voltage ◽  
Incandescent Lamp ◽  
Mercury Lamp ◽  
Load Modeling ◽  
Load Model ◽  
Fluorescent Lamps

This paper presents the results of dynamic load modeling for some frequently used low voltage devices. The modeling of long-term dynamics is performed on the basis of step changes of supply voltage of the heater, incandescent lamp, mercury lamp, fluorescent lamps, refrigerator, TV set and induction motor. Parameters of dynamic exponential load model of these load devices are identified, analyzed and mutually compared.

Solving realistic reactive power market clearing problem of wind-thermal power system with system security

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Sravanthi Pagidipala ◽  
Sandeep Vuddanti
Keyword(s):  
Power System ◽  
Reactive Power ◽  
Power Market ◽  
Operating Conditions ◽  
Objective Functions ◽  
Market Clearing ◽  
Load Model ◽  
Multi Objective ◽  
Reactive Power Market ◽  
Single Objective

Abstract This paper proposes a security-constrained single and multi-objective optimization (MOO) based realistic security constrained-reactive power market clearing (SC-RPMC) mechanism in a hybrid power system by integrating the wind energy generators (WEGs) along with traditional thermal generating stations. Pre-contingency and post-contingency reactive power price clearing plans are developed. Different objective functions considered are the reactive power cost (RPC) minimization, voltage stability enhancement index (VSEI) minimization, system loss minimization (SLM), and the amount of load served maximization (LSM). These objectives of the SC-RPMC problem are solved in a single objective as well as multi-objective manner. The choice of objective functions for the MOO model depends on the load model and the operating condition of the system. For example, the SLM is an important objective function for the constant power load model, whereas the LSM is for the voltage-dependent/variable load model. The VSEI objective should be used only in near-critical loading conditions. The SLM/LSM objective is for all other operating conditions. The reason for using multiple objectives instead of a single objective and the rationale for the choice of the appropriate objectives for a given situation is explained. In this work, the teaching learning-based optimization (TLBO) algorithm is used for solving the proposed single objective-based SC-RPMC problem, and a non-dominated sorting-based TLBO technique is used for solving the multi-objective-based SC-RPMC problem. The fuzzy decision-making approach is applied for extracting the best-compromised solution. The validity and efficiency of the proposed market-clearing approach have been tested on IEEE 30 bus network.

scholarly journals Accurate Load Modeling Based on Analytic Hierarchy Process

2016 ◽  
Vol 2016 ◽  
pp. 1-11 ◽  
Author(s):  
Zhenshu Wang ◽  
Xiaohui Jiang ◽  
Shaorun Bian ◽  
Yangyang Ma ◽  
Bowen Fan
Keyword(s):  
Power System ◽  
Analytic Hierarchy Process ◽  
Load Modeling ◽  
Analytic Hierarchy ◽  
Load Model ◽  
Load Models ◽  
Power Load ◽  
Hierarchy Process ◽  
Weight Coefficients ◽  
Traction Power

Establishing an accurate load model is a critical problem in power system modeling. That has significant meaning in power system digital simulation and dynamic security analysis. The synthesis load model (SLM) considers the impact of power distribution network and compensation capacitor, while randomness of power load is more precisely described by traction power system load model (TPSLM). On the basis of these two load models, a load modeling method that combines synthesis load with traction power load is proposed in this paper. This method uses analytic hierarchy process (AHP) to interact with two load models. Weight coefficients of two models can be calculated after formulating criteria and judgment matrixes and then establishing a synthesis model by weight coefficients. The effectiveness of the proposed method was examined through simulation. The results show that accurate load modeling based on AHP can effectively improve the accuracy of load model and prove the validity of this method.

scholarly journals A data-driven approach for online aggregated load modeling through intelligent terminals

2019 ◽  
Vol 15 (1) ◽  
pp. 155014771982599 ◽  
Author(s):  
Yi Tang ◽  
Liangliang Zhu ◽  
Jia Ning ◽  
Qi Wang
Keyword(s):  
Power System ◽  
Induction Motor ◽  
Power Distribution ◽  
Distribution Networks ◽  
Data Driven ◽  
Load Modeling ◽  
Load Model ◽  
Power Models ◽  
Data Driven Approach ◽  
Aggregated Load

Load model has significant impact on power system simulation. Current load modeling approaches are inadequate on revealing the accuracy and time-variation of load compositions. The application of wireless sensors dispersed in power distribution networks provides further opportunities for load modeling. In this article, a data-driven online aggregated load modeling approach is proposed systematically. First, all the electricity consumers are clustered according to big data of power consumption behaviors. In each cluster, typical users are designated to stand for the characteristics of the cluster, and intrusive measurement is adapted to capture these typical users’ time-varying information by employing wireless intelligent terminals, which can identify the composition of static load and induction motor load online. Second, the load models of other users in each cluster are assumed identical to typical users, including static impedance–current–power models and induction motor models. Finally, the composite load model is achieved by hierarchical aggregation and bottom-to-up stepwise equivalence. Simulations demonstrate that the load model built by proposed approach reflects higher accuracy and adaptability in power system.

Sign in / Sign up

Export Citation Format

Share Document