Adequacy Assessment of Power System and Capacity Credit Estimation with Renewable Source Integration

Abstract

Ever increasing global population creates the serious issue of cumulative increase in demand for electrical power. To cope with the situation, many methods have been suggested in the past. These methods include expansion plan as well renewable integration in islanding and interconnection. In the era of green energy, instead of expansion planning of conventional generation, it is advisable to choose renewable integration. Among all the available renewable sources wind and solar are more popular.
Reliability estimation is a statistical calculation. This demands unavailability rates of generating units taking part in the evaluation process. The index for PV is determined using the plant configuration. Mostly three types of configuration are accessed and failure rate is determined. Wind energy conversion system converts wind energy into electrical energy. To anticipate the variability of wind speed in the adequacy assessment, wind speed is predicted using some of the popular methods; e.g. autoregressive moving average (ARMA) method, Weibull distribution and adaptive neuro-fuzzy inference system (ANFIS). In case of ARMA and Weibull distribution, mathematical models are developed and collected wind speed is fitted to them. The best fitting is determined by finding F-value in case of ARMA. Statistical error in case of Weibull distribution determines the fitting. The approach in case of ANFIS is little different. ANFIS divides the total data into two groups; training and checking data. ANFIS trains the input with the training data and checks the accuracy using the checking data. The comparative plots between the actual and predicted wind speed decides the suitable prediction technique among all the considered techniques. Power from wind turbine generator (WTG) is segregated into different categories to form multi-states. Multi output states are reduced to 2 states using Apportioning method to determine the unavailability rate. With the determination of failure rates, adequacy is estimated by finding the expected loss of load index.
Capacity value of a newly added generating unit is generally determined using reliability based methods. These methods are estimating capacity value considering additional load taken by the system, integrating a theoretical capacity or adding a practical capacity by maintaining the annual system risk. Unlike conventional, it is not that easy to include random renewable sources in the adequacy assessment as they rely on chaotic resources. Thus, to include renewable sources in the process of adequacy assessment, hourly generated power are considered as negative load and subtracted from the system load. x
The reliability based method of capacity credit estimation demands enormous data collection as well as recursive probabilistic calculation. Again repetitive calculation of capacity value of identical renewable sources may occur due to dependency of renewable generation on the random resource. Thus, a database is constructed with the reliability indices after adding sequential value of renewable sources to the system along with all the capacities values corresponding to each reliability based method. To calculate the capacity value, expected loss of loads (LOLEs) after adding the equivalent capacities are chosen as the target. To find the best match, the database is represented as a multidimensional tree and nearest neighbor search algorithm is used to search the query in the tree.