In this paper, we briefly consider a queueing model with energy harvesting and multi-threshold control by the service modes. The available service modes distinguish by the service rate, the number of required for a customer service energy units and the probability of error occurrence during service. The account of the possibility of error occurrence is very important in modelling wireless sensor networks. The use of a larger number of energy units corresponds in real-life wireless communication systems to the ability to send a stronger signal what implies the faster service (transmission) rate and a lower probability of incorrect transmission (due to the noise in the wireless transmission channel). Incorrect transmission implies the necessity of repeated transmission and additional consumption of energy. Under the fixed thresholds defining the control strategy, behavior of the system is described by multi-dimensional Markov chain what allows to formulate and solve optimization problems.