Research On Power System Black-Start Restoration And Its Decision Support Technique

As modern society is dependent on the power supply excessively, the damage due to widespread blackout is more and more severe. As one of self-rescue measures, the research on power system black-start restoration and its decision support technique is of great importance to optimizing control routines, relieving restoration burdens, reducing restoration duration and loss resulting from faults. After fruits achieved and issues existed about black start have been investigated thoroughly, planning of black-start scheme is designated to be the kernel of this thesis. According to the classification of black-start stages, technical problems and decision methods prior to the establishment of main network are studied by means of centralized optimization. Decision support technique is introduced to construct the decision support frame assisting dispatchers with making and choosing schemes. The prototype, the application of decision support frame to Hebei's southern power network, shows that the approach proposed in this thesis could provide dispatchers a new access to constituting practical schemes quickly and effectively.Main contributions of this thesis are described as following:1. Simulation programmes integrated on a uniform data platform are developed so that technical verifications such as generator's self-excitation, overvoltage when energizing transmission lines, etc. are simplified. A smooth way has been paved for developing decision support system of black start and schemes'automatic test and assessment.2. A skeleton-network reconfiguration strategy based on assessing node importance and an optimization method of reconfiguration control aiming at minimizing reconfiguration cost are proposed in order to decrease reconfiguration's load and complexity considerably. First, the importance of nodes is scaled through investigating networks'topological characteristic and objective function representing network structure and restoration constrains is constructed. Discrete particle swarm optimization is employed in identifying skeleton networks as restoration target. Furthermore, minimization of reconfiguration cost considering restoration capacity and distance is regarded as optimization objective and Prim algorithm is adopted to form detailed restoration sequences based on certain skeleton network. Besides, values of control variables are confirmed by nonlinear optimization.3. Hierarchical data envelopment analysis is proposed to evaluate black-start schemes planed for initial stages of restoration. The comparison of preferable schemes'reconfiguration effects could provide more comprehensive references for dispatchers making decision. Combining Analytic Hierarchy Process with Data Envelopment Analysis reasonably, hierarchical data envelopment analysis takes account of technical problems and experiential factors synthetically. Further comparison of reconfiguration effects could detect important schemes'feasibility with respect to subsequent restoration so that schemes are investigated more thoroughly.4. A universal decision support frame for power system black start is proposed and a prototype of black-start decision support system is developed. With regard to planning of black-start scheme, the frame integrates numerical calculations of technical test and data processing of decision analysis. The prototype, the frame's application to Hebei's southern power network, proves that the method could make black-start schemes'investigation and reasoning more convenient and improve dispatchers'ability of dealing with large-scale blackout.