Font Size: a A A

Load Forecasting In The Substation And Reactive Power Control

Posted on:2006-10-11Degree:MasterType:Thesis
Country:ChinaCandidate:Y C LiuFull Text:PDF
GTID:2192360155966899Subject:Power system and its automation
Abstract/Summary:PDF Full Text Request
Voltage is a significant index for power quality. Power substation is the nexus of power net and user, which act as a key part to stabilize voltage and reduce power loss. Now most substations are equipped with OLTC (on-load tap-changer) transformers and power capacitors .The OLTC transformers and power capacitors are the main equipments to adjust voltage and reactive power. In most substations, nine-section algorithm is used as control theory. The relationship of voltage and reactive power is ignored in nine-section algorithm. The OLTC and breaker will switch much more when using nine-section algorithm.Accurate electric load forecasting is the foundation to switch compensating capacitors properly. On account of the fact that power system is an intense non-linear, time-varying system which consists of a great many of dynamic unmodeled characteristics and tends to become more complicated, this paper adopts ANN (Artificial Neural Network) to complete the forecasting task of short term and ultra short term load. The fundamental principles of ANN are introduced and the application of Kohonen Network and BP (Back Propagation) Network in load forecasting is analyzed. Some examples are also included.To avoid possible power oscillation when switching compensating capacitors, a method based on load forecasting and reactive power compensation is proposed. The basic idea is to establish a closed-loop reactive power compensation system, which treats the load forecasting result as one of input parameters. The actual load is compared to the forecasted value and is judged whether it is a fluctuant load. Take the over-loading withstand ability of transformers into consideration as well, a decision whether the compensating capacitors should be switched on or off and if so, when to switch, is made. As the decision is made in this way, bus voltages and load power factors can be kept as well as the switch operations can be reduced as much as possible, which will prolong devices' life-span and promote system security consequently.In addition, the program realizing real-time switching control of compensating capacitors based on load forecasting is presented. The schematic diagram as well as hardware design is given too.
Keywords/Search Tags:reactive power compensation, load forecasting, ANN (Artificial Neural Network), compensating capacitors
PDF Full Text Request
Related items