When the battery or supercapacitor is connected to the output of a PV system, the conventional voltage equation expressing its mathematical model usually must be replaced by the current relationship to study the maximum power point tracking (MPPT) control theory. However, hitherto, there is a lack of an attempt to disclose the current constraint relationships at the maximum power point (MPP), which leads to the potential risk of MPPT failure. To solve this problem, in this paper, the MPPT constraint conditions on the basis of currents are built and then a new MPPT control strategy is proposed. In this strategy, a linearized model parameter of a PV cell is used as the bridge to find the current relationships. On the basis of them, some expressions involving the duty cycle are built to directly calculate the control signal of the MPPT controller. Meanwhile, an implementation method is designed to match this proposed MPPT strategy. Finally, some simulation experiments are conducted. The simulation results verify that the proposed MPPT constraint expressions are accurate and workable and that the proposed MPPT strategy and its implementation process are feasible and available. In addition, the simulation results also show that the proposed MPPT strategy has a better MPPT speed and the same MPPT accuracy when the P&O method and fuzzy algorithm are compared. By this work, the MPPT constraint conditions based on current relationships are first found, representing a breakthrough in disclosing the inherent relationships between different currents when the PV system is operating around the MPP.