Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Probabilistic estimation of hosting capacity and operating reserve including optimisation of PV installation capacity

Hobbs, Dale (2019) Probabilistic estimation of hosting capacity and operating reserve including optimisation of PV installation capacity. Honours thesis, Murdoch University.

PDF - Whole Thesis
Download (2MB) | Preview


In recent times the need to deploy additional sustainable generation means has become more apparent due to the ever-changing landscape of the global energy generation sector. Australia’s changing consumer needs means new technologies like renewable generation sources such as solar PV systems have increased in popularity over time, though their full capability has not yet been met. Though their intermittent generation is cause for concern in maintaining a stable and quality power supply. This thesis aims to address the issues by developing a probabilistic methodology for the day ahead estimate of the maximum hosting limits capacity and minimum operating reserve requirements of a microgrid containing high levels of PV penetration.

Before the commencement and development of the project, a wide range of methods from literature were analysed regarding microgrids and their use in this project. The comprehensive range of concepts of microgrids and their distributed generation were divulged and incorporated into the project methodology. To understand how to provide the probabilistic estimate of the maximum hosting capacity, three previously methods in literature were analysed, each providing more technically advanced approaches than the last. The same research approach was used to understand the methodology of developing a probabilistic estimate of the operating reserve. These methods range in methodology, from the Monte Carlo simulation method to advanced artificial neural networks.

To provide the day ahead estimates, an artificial neural network is developed to generate the network parameter forecasts required, providing with it, a probabilistic range of input to a network model. The maximum hosting capacity limit will ensure the amount of renewable generation expected will not exceed the performance indexes required for, voltage level, line loading limits and generator reverse power flow. The minimum operating reserve will provide an estimate of the reserve generation required if there were to be a sudden drop in the renewable energy supply. The estimates are created by modelling the IEEE 13-bus network in PowerFactory containing high levels of renewable generation. The programming functionality in this package has been utilised to automate the immense simulation, calculation and data collection processes required on a case by case basis. Statistical analysis is performed to define the probability of these estates occurring. The probabilities of these estimates will help network operators in making decisions for the control of the microgrid. Adding to these estimates were the PV generation capacity optimisations to increase the maximum hosting capacity limit.

Several test cases were created to analyse the performance of the modelling automation developed. Each of these cases created a different insight into the estimation and optimisation cases and their interaction with the performance indexes. The probabilistic estimations derived produced a normal distribution of values for each of the cases tested. Probability statistics are applied to provide the probability of such estimates occurring for the next day's operation. The optimisation successfully provided the maximised PV generation, setting a maximum hosting capacity within the performance index limits.

The methodology developed was successful in providing the probabilistic estimation required and optimising the PV installed capacity. This method offered the use of advanced technology, such as artificial neural networks, to provide more reliable predictions into the network operation.

Item Type: Thesis (Honours)
Murdoch Affiliation: Engineering and Energy
Supervisor(s): Arefi, Ali
Item Control Page Item Control Page


Downloads per month over past year