The growing electricity demand, the high penetration of renewables, and the deregulation of electricity markets have caused the electricity grids to be operated with high capacity and close to their stability limits. Power companies face increasing security threats in the operation of power networks. State estimation is an important tool to determine the system states. State estimation is an integral part of energy management systems for security analysis and other power system applications. A prerequisite for state estimation is that the system must be fully observable by available measurements. Before the introduction of synchronized measurement technology, state estimation was done based on supervisory control systems and data acquisition. The advantage of micro-phasor measurement units over these systems is that they measure the phase angle directly and all measurements are coordinated. The micro-phasor measurement unit measures the voltage phasor of the bus in which it is installed and the current phasors of all the branches that are connected to that bus. Direct measurement of all system states is possible by placing phasor measurement units in all buses of a network without running any state estimator. The micro-phasor measurement unit and the telecommunication system related to it are very expensive, so the voltage phasor of the buses adjacent to the buses with micro-phasor measurement unit can be obtained with the help of the branch current phasor. The main goal of optimal micro-phasor measurement unit placement is to determine the minimum number of units to be installed so that the entire power system is fully observable for state estimation. In the first stage, by using the optimal micro-phasor measurement units placement method in the presence of zero injection nodes and various contingencies, we witnessed the improvement of the visibility of the distribution system. Also, the presentation of the proposed model in the presence of different configurations of the distribution network resulted in obtaining the optimal configuration of the minimum number of micro-phasor measurement units, optical fiber cables, and network reliability index. Also, the proposed method improves some technical characteristics of the distribution network, such as the voltage profile and reducing the system power loss. In the second stage, by using the two-stage probabilistic method, in the first step, we saw a decrease in the number of micro-phasor measurement units, and in the second step, we saw an improvement in the accuracy of the distribution system state estimation in the presence of various network constraints.