Giovanni Mercurio Casolino

Load areas in distribution systems

The updated and new monitoring and control functionalities for distribution systems make available a huge amount of information. Indeed, not all the available information is needed to be considered; a significant simplification could be achieved by selecting and treating only the relevant information. With this aim, the concept of Load Area (LA) was recently introduced: a LA is a group of prosumers whose power injection has a similar impact on the distribution grid operating conditions. In the paper, the concept of LA is adopted to select the relevant data needed for the monitoring and control of distribution systems; a possible compact equivalent representation of the relationships between those data is proposed. The results from two test cases confirm the viability of the proposed modeling and highlight a relevant reduction in the amount of information used to describe the network.

Load Area application to radial distribution systems

The evolution of the distribution systems involves updated and new control functionalities; along with them, a huge amount of information is made available and has to be handled. Actually, only part of this information is necessary; a remarkable simplification can be obtained by considering only the relevant information. For this scope, a method was recently proposed based on the concept of Load Area, a group of prosumers whose power injection has a similar impact on the distribution grid operating conditions. In this paper, the method previously developed for a generic distribution system is applied to a radial distribution system to obtain its equivalent radial representation; for it, the computational times for the power-flow problem are shown. The results confirm the viability of the method and the reduction of computational time for the ubiquitous power-flow problem.

MODELING VOLTAGE REGULATION IN OBJECT-ORIENTED ANALYSIS OF DISTRIBUTION SYSTEMS INCLUDING DISPERSED GENERATION

Abstract Dispersed Generation (DG) affects the operation of distribution systems; the related modeling and analyzing techniques are affected too. A novel approach to the distribution load flow problem has been recently proposed, based on the Object Oriented (OO) paradigm and including DG. In this paper, the OO modeling and algorithm are extended to account for the inclusion of ULTC transformers voltage regulation. To validate the proposed steady-state model and to verify its assumption of stability, a detailed dynamic model is developed in Matlab/Simulink environment for a simple case study. The numerical results give evidence of the effectiveness of the proposed steady-state model which accounts for the interaction between DG and ULTC voltage regulation.

A Software Environment for Developing Object-Oriented Applications in Distribution Management Systems

Thanks to the increasing capability of computing systems, the development of distribution management systems for electrical distribution systems can be approached from the perspective of flexibility, expansibility and maintainability. For this purpose, new methods have been recently proposed, in which the modeling of the distribution systems as well as algorithms and computational methods have been developed with an object-oriented (OO) approach. In this paper, an extensible software environment for the development and the implementation of OO methods, aimed at distribution network applications, is presented. It is an open platform, based on timed Petri nets and on a discrete event scheduler, useful to simulate custom applications with specifically developed libraries and code. A significant example on distribution system state estimation is presented, which highlights the features of the environment

Specialized methods for the implementation of Load Areas in radial distribution networks

The change towards future distribution networks requires new and updated monitoring and control functions, which involve a large amount of information. The treatment of a significant part of this information can be simplified by introducing the concept of Load Area, a group of prosumers whose power injection has a similar impact on the distribution grid operating conditions. Thanks to the Load Area concept, a compact yet accurate representation of distribution networks can be obtained. The Load Areas allow to select the relevant data for both monitoring and control functionalities of distribution systems, as well as to represent the relationships between such data in a compact representation. In this paper, the Load Area concept is recalled, and specialized methods are applied to get a compact representation that exploit the radiality. The results from two test cases confirm the viability of the proposed methods and highlight the computational gains deriving from their adoption.

Software-in-the-loop simulation of a test system for automotive electric drives

The Automotive market is rapidly changing, as the increasing attention to the engine emissions pushes car makers to use vehicles less dependent on gasoline and diesel fuel, resulting in lower operating costs and emissions. Today, electric vehicles seem to be the best answer to the market demand. In this scenario, electric drives for traction play a key role, thus the necessity to develop effective and flexible test systems, able to ensure performance, quality and safety. In this paper a new automotive electric drives test system is proposed. The test system for electric drives is based on an inverter controlled through a real-time platform in which the Electric Vehicle model is implemented. The inverter emulates the behaviour of the motor and load at the load drive terminals. This paper describes the architecture of the system and shows the real-time simulation results. Three control methods are used for the load drive and compared in the paper.

Computational aspects of object-oriented distribution state estimation

Object-oriented (OO) state estimation modeling has been recently proposed, for both radial and weakly meshed distribution systems. The OO algorithm is based on the Newtons iterative method, widely adopted to find a solution to the state estimation problem. For radial topologies, the OO algorithm derives from the Newtons method. Weakly meshed topologies are treated with radial equivalents; the OO approach introduces some approximations. The convergence performance of the OO algorithm is investigated on a 69-branch system

Load areas in reconfiguration of distribution systems

With the moving of distribution systems towards a renewed architecture, new monitoring and control functionalities have emerged; with them a large amount of information need to be treated. Not all this information is strictly necessary; a relevant part of it can be simplified by the concept of Load Area: a group of prosumers whose power injection has a similar impact on the distribution grid operating conditions. The Load Areas allow to select the relevant data for both monitoring and control functionalities of distribution systems, and to describe the relationships between such data in a compact representation. In this paper, the concept of LA is examined when multiple configurations of the network have to be considered, so as to examine the changing of LAs when the grid is reconfigured. Two test cases are evaluated; the results show that even in case of reconfiguration of the network, the number of LAs to be considered remains limited, allowing a relevant reduction in the information required to represent the network.

Computationally-efficient simulation of electric industrial plants

The validation of a power control system in industrial plants requires a modelling for simulation of the electric power system that is fairly accurate and computationally-efficient. The second aspect is particularly relevant if a real-time simulation is required. In this paper the classical models of the network and the induction motors are revised so as to allow an adequate trade-off between computational-efficiency and accuracy of the results. In particular, the model for the induction motors are improved to achieve an adequate accuracy in all the operating range; the network solution is improved to increase computational efficiency. Numerical simulations are presented to show the limited computational burden in simulating a large industrial plant, as well as the good accuracy of the results in comparison with the characteristics of actual machines.

Test facility for control system validation in electric industrial plants

The management of an electric industrial plant requires a control system able to guarantee a safe operation in all situations that may arise, limiting possible outage conditions. A key issue is the validation of the control structure which requires a real-time simulation of the electric power system that is fairly accurate and computationally-efficient, together with tools for data exchange that allow a reliable communication with all the control devices. In this paper, the components and the architecture of a laboratory test facility for the validation of control systems is presented, which is based on a real time digital simulator of the electric industrial plants. Specific details are given concerning the communication structure needed for data exchange between the real-time simulator and the control system. The results of validation tests are presented to show the behavior of the industrial plant in different operating conditions and to prove the validation of the control structure by means of the proposed test facility.