Power equipment condition monitoring is an important practice of utility companies. It helps to prevent unexpected outages of power systems and reduces maintenance costs. There are two general approaches to condition monitoring:​ passive and active. Passive condition monitoring is conducted by detecting abnormal signals generated by the equipment of interest. This approach is reactive since it waits for abnormal equipment conditions to occur. Active condition monitoring is conducted by injecting a test signal and analyzing the equipment response. It is a proactive approach. However, due to its intrusive nature, active condition monitoring can only be performed occasionally in an offline mode, and it involves high costs.

The proposed Ph.D. research aims to transform active condition monitoring from offline to online. This is achieved by utilizing natural power disturbances as the test signals. The proposed approach has the potential to enable frequent and automatic testing of equipment once the monitoring system is set up. In addition, the statistical trends of equipment responses under repetitive tests will provide a much more accurate picture of equipment conditions. In this research, the proposed concept is applied to solving two challenging problems:​ substation grounding grid condition monitoring and open conductor detection.

This thesis presents two online methods for monitoring the substation grounding grid impedance and thus revealing the conditions of the substation grounding grid. Natural disturbances in the form of a ground-fault current and a 3rd harmonic current are used as test signals, respectively. Both methods’ performance and effectiveness are demonstrated through analytical studies and extensive verification, sensitivity, and application studies. This thesis further presents two online schemes to detect the open conductor condition in power plants. Natural voltage sag disturbance and 3rd harmonic power are used as test signals, respectively. The performance and practicability of both methods have been demonstrated using simulation and experimental studies. These two methods can work individually or in a parallel way to increase the effectiveness and reliability of the open conductor detection.

The power-disturbance-based condition monitoring concept as presented in this thesis has the potential to transform many offline, intrusive condition monitoring methods into online automatic condition monitoring methods, thus creating a new opportunity to improve the “smartness” of power equipment and power systems.