 ### Conductance incremental method

Figure 1 shows the output characteristic curve of photovoltaic panels at a specific temperature and specific light intensity (ie, light intensity). Calculate dP/dU on the power/voltage curve, and use the relationship between dP/dU and 0 to determine the working state of the photovoltaic panel. When the photovoltaic panel is working in state B in Figure 1, the output power of the photovoltaic panel reaches the maximum. At this time:

From formula (1.1), we can get:

Based on the above principles, the flow chart of the actual execution algorithm is shown in Figure 2. The digital implementation of the conductance increment method requires division operations, such as dividing current by voltage. Because division consumes a lot of resources in the digital processor, the flow of Figure 4.4 will bring lower computational efficiency. In order to avoid the division operation of equation (1.2), when judging whether equation (1.1) is true, as long as dU≠0, dP×dU=0 can be the same as the judgment result of dP×dU=0. For the case of dU=0, it is considered that when dU=0, the judgment of equation (1.1) is also established, that is, dPxdU=0 is established. In this way, a division judgment problem can be transformed into a multiplication, thereby avoiding the problem of dividing by 0, which is easy to simulate and implement in practice. Figure 3 is a flowchart based on this idea, where ΔU represents the amount of change to the photovoltaic input voltage reference each time.

### Perturbation Observation

The perturbation observation method is also a commonly used MPPT implementation algorithm. Its principle is: first use a digital controller to generate a voltage reference, and control the output voltage of the photovoltaic panel to follow the voltage reference, where the voltage reference changes at all times; then observe the change of the output power of the photovoltaic panel after the disturbance, determine the direction of the power change, and then determine the next reference voltage setting, until the photovoltaic panel finally works at the maximum power point.

As mentioned above, the P-U characteristic curve of a photovoltaic cell is a function of a single peak value, and there is a single maximum power point. Therefore, in the implementation of the disturbance observation method, the program has been constantly changing the reference, and then measuring the magnitude and direction of the output change caused by the change of the reference input signal. For example, when the solar photovoltaic panel is working on the left side of the maximum power point of the photovoltaic cell characteristic curve, the MPPT algorithm will increase the benchmark; when the operating point is located on the right side of the maximum power point, the algorithm will move the operating characteristic curve toward the direction of the decrease of the reference voltage, and finally the photovoltaic panel will swing near the maximum power point. Obviously, when it is stable, the working state of the photovoltaic cell is constantly disturbing, which will cause a certain power loss, because there is no disturbance, it is impossible to track the maximum power point.

In order to discuss the disturbance observation method more intuitively, Figure 4 shows a schematic diagram of the working process of the disturbance observation method.