Research On Fault Diagnostic Method For Micro-grid Based On Analytic Model

Micro-grid fault diagnosis system for the fault components’rapid detection and location in Micro-grid. Its ultimate aim is to ensure a stable and safe operation of the Micro-grid, making the power supply reliability. Meanwhile the fault diagnosis system provides a convenient for quick fix after fault accident happened. Currently, the research of fault diagnosis is use of "Expert System", "Artificial Neural Networks", "Bayesian Networks", "Petri Net" and "Multi-agent Technology", which does not apply to diverse topologies, flexible operation mode of Micro-grid. Based on analytical model for Micro-grid fault diagnosis for the changing topology of the Micro-grid has broad prospects.This paper based on the analysis the characteristics of the Micro-grid, composed of the elements. Introduced the common protection methods of Micro-grid, analysis the characteristics of these protection methods. For these characteristics established a class of analytic models and applied in Micro-grid fault diagnosis, and to study the solution of the analytic model has built. The analytic model is essentially constituted by the logic equations, the variables included reflects the rules of fault diagnosis. Each solution of equations corresponding to a possible fault modes, which launched the diagnosis results and conclusions. The outcome of this research is reflected in the following three aspects:1. According to the characteristics of the Micro-grid protection proposed analytical model of fault diagnosis. The analytical model is essentially reflects the analytical expression of fault diagnosis rules, the advantage is retain the logic completely between suspicious components and protection movement in the fault zone. Makes the fault diagnosis system has a strong fault tolerance.2. Research on the simplification of the analytical model and it’s solve method. To transform the established analytical model, decoupling, deduced simplified model expression. Research on the analytical method of the alarm signal, derived the complete solution set of the model.3. Bring the idea of risk decision into the analytical model, and solved. There are many uncertain factors in the Micro-grid fault diagnosis process, analysis the rules of these uncertain factors and constructed the evaluation of fault diagnosis decisions. The Micro-grid fault diagnosis problem is then transformed into an integer programming problem of risk-based decision making. Use the improved particle swarm algorithm proposed in this paper to solve the model, obtain the optimal solution. Finally launch of the results and conclusions of fault diagnosis.