In ultra dense WiFi networks, access points are deployed with high density in order to support Internet access for large amount of mobile users at the same time. Since the mobile users may not be uniformly distributed in the network, the heavy loaded access points and the interference among different access points would cause the network throughput degradation. Further, although increasing transmission power levels of access points could enhance network throughput, the energy efficiency would be low. Energy consumption is an increasingly important issue today so that the network performance needs to be improved in an energy efficient manner. Under this condition, we propose a distributed power adjustment scheme to intelligently maximize the throughput of the entire network without increasing transmission power in the presence of dynamic distributed mobile users. To validate the performance of our proposed scheme, we simulate various network conditions and the results show that our proposed scheme can sustain throughput with less energy consumption with less than 200 iterations.