Machines are an integral part of the actual workshop production system. If machinery operates for prolonged periods under high energy consumption, it will accelerate equipment aging and increase the likelihood of malfunctions, thereby disrupting production cycles. Conversely, sustained idle states or low energy utilization can lead to significant resource waste. Therefore, achieving energy consumption equilibrium is a critical and highly valuable research area that merits systematic exploration. It addresses the green flexible job shop scheduling problem with balanced machine energy consumption (GFJSP-BMEC) and proposes an optimization model aimed at minimizing two objectives: makespan and the weighted sum of inter-machine energy consumption differences and total energy consumption. To this end, an improved non-dominated sorting genetic algorithm Ⅱ (INSGA-Ⅱ) is developed to optimize these objectives simultaneously. Extensive numerical experiments are conducted to verify the impact of the machine energy consumption balancing strategy on makespan and total energy consumption. Furthermore, INSGA-Ⅱ is compared with non-dominated sorting genetic algorithm Ⅱ and multi-objective particle swarm optimization to demonstrate its effectiveness and superiority in solving GFJSP-BMEC.