Study Of Load Shedding Strategy Considering Risk Assessment

The power shortage of the system will be generated after a catastrophicfault or a certain power loss, which leads the frequency of the system drops.When the spinning reserve is insufficient to maintain the frequency stability, acertain amount of load always need to be cut to compensate for the powershortage in order to maintain the stable operation of the system. So it isnecessary to choose economical and reliable load shedding strategy.The conventional under frequency load shedding (UFLS) measure is usedin the actual power systems, but it often miss the best time to shed load as acorrective action, and the amount of load shedding can not completely matchthe actual power shortage. It is difficult to satisfy the modern power system’srequirements of power quality. For the above shortcomings, an event-drivenoptimal load shedding strategy is proposed based on consideration of risk, theapproach of feed forward control greatly enhance the result and efficiency ofload shedding.In order to ensure that the optimal load shedding strategy can minimize thecosts under various fault conditions, the paper study the optimal power flowmodel with its algorithms deeply, establish the model of optimal power flow considered cost, using primal-dual interior point method to solve the model,write MATLAB program to achieve its function, finally get the optimal amount ofload shedding of each load node after any fault. The example shows that theload shedding model can make sure the stable operation of the system withminimum cost of the amount of load shedding.In order to determine the optimal load shedding strategy which can be usedin any case of fault, the quick sort algorithm is used according to the model ofelement outage rate, to determine the expected fault state set of the system.Then the optimal load shedding model is used to get the amount of loadshedding of each load node in all fault conditions. The cluster algorithm is usedto reduce the number of load shedding devices, it cluster the faults by powershortage. Consideration of the load shedding risk indicators of cost andfrequency deviation, a0-1programming model is established to make theamount of load shedding discretized and optimized for all the faults in each faultset, and a set of feeder removal program is achieved according to thecorresponding fault set. The comparison of the ultimate recovery effect with thetraditional approach indicates that the load shedding strategy mentioned hereinmay improve system stability and economy, reduce the risk of system operation.In order to verify the affect of the frequency dynamic process using thestrategy mentioned herein, this paper analyze the power system frequencycharacteristics and dynamic component model, build the example model usingthe simulation software-DIgSILENT. The comparison of frequency dynamic process and indicators between the traditional UFLS program and the optimalload shedding strategy under a variety of faults confirm that the optimal loadshedding strategy mentioned herein can reduce the frequency fluctuation andaccelerate the recovery rate of the frequency.