Approches pour la réduction de l'impact de défaut dans le transport d'énergie du parc éolien offshore via VSC-HVDC

Abstract

In this work a power generation system of an offshore wind farm, a control system and a transmission system using VSC-HVDC stations connected to the onshore AC main grid is presented. Three configurations were studied, modeled and validated by simulation. The contribution of the research work for improving the technical and economic side are described as follows: A new MPPT (Maximum Power Point Tracking) algorithm is used for extraction of the maximum power available from the wind turbine in offshore wind farms. This MPPT technique improves the conversion efficiency from wind turbine in small and large scale (offshore wind farm). It facilitates manufacturers in developing simple MPPT devices, which are cheaper, robust, reliable and able to achieve maximum energy efficiency. Another main contributions of this work deals with the reduction in size, cost and the faults impact in AC and DC system built to transmit the power of an offshore wind farm (OWF) to the onshore main AC grid via two stations 3L-NPC VSC-HVDC. The developed solution uses the nonlinear observers based on the extended Kalman filter (EKF). This filter estimâtes the speed and the rotor position of each generator of the offshore wind farm, the DC bus voltage of the offshore DC-AC inverter and at the both stations-3L-NPC VSC-HVDC (offshore and onshore). Moreover, development of the Extended Kalman filter reduces the impact of AC and DC faults. Two controls techniques approaches were used, the indirect control associating EKF algorithm for the control of the offshore DC-AC converter and the other dq control associating EKF algorithm for the control of both converters AC-DC and DC-AC at offshore and onshore stations. The integration of nonlinear observers (EKF) in the control algorithm solves the problem of measurement uncertainties, uncertainties in modeling, malfunctioning, error in measurement due to sensor failure and the problem associated with the impact of faults (AC and DC) on the quality of energy in transmission systems. These estimations also contribute to make the overall cost of the system cheaper and less cumbersome system size and also to reduce the impact of faults (AC and DC) on the system. The third developed contribution of this thesis concerns with the use of nonlinear observers based on EKF algorithm. This filter permits to estimate the speed and the rotor position of each generators of the OWF and the DC bus voltage of the offshore DC-AC inverter. The contribution focuses on the development of two controls algorithms for both stations. The first, nonlinear modified control is applied to the first converter of the VSC-HVDC offshore station to ensure the power transfer generated by the wind farm to the onshore VSC-HVDC station. The second nonlinear modified control ensure the regulation of DC bus voltage and uses an adaptive reference model control (MRAC) to compensate the overcurrent and the overvoltage during the AC and DC faults. The developed control scheme allows reducing the impact of AC fault at PCC (point of common coupling) of onshore of the AC main grid side. The impact depth of AC fault on the magnitude of the onshore AC main grid currents have been claimed to reduce up to 60% in comparison to the reported research by (Erlich, Feltes et Shewarega, 2014) compared to the proposed control based on MRAC which reduces the impact depth at 35%. During the conditions of the AC and DC faults, the reduction in the amplitude of the AC grid currents and the improvement in response time were observed and the stability of the system has also been enhanced by the use of MRAC. The fourth developed contribution of this work presents a new command based on the sliding mode (SM) applied for the VSC-HVDC stations that connect the offshore wind farm to the onshore AC main grid. The wind farm consists of ten turbines coupled with permanent magnet synchronous generators (VSWT / PMSGs) and connected in parallel and each is controlled by its own DC-DC converter to extract the MPPT. A comparison between the performance of the SM control and the non-linear control using PI controllers for both conditions (with and without DC fault) was analyzed and the superiority of the SMC has been established through various case studies. For experimental validation, a small-scale prototype of the system is built and tested in GREPCI laboratory using dSPACE-DS1104 Controller Board. The analysis and the simulation of the studied systems are developed under the Matlab/Simulink/Simpowersystem. Simulation and experimentation results of the developed configurations are very found satisfactory in terms of steady-state and dynamic, response, system stability and power quality.

Date	12 Oct 2016
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Ambrish Chandra (Supervisor)