Photovoltaic power generation MPPT control technology based on fuzzy strategy

With the development of the global economy, energy issues are becoming increasingly acute, and more and more countries are beginning to pay attention to the issues of energy utilization and conversion efficiency. Photovoltaic power generation has the advantages of being pollution-free, noise-free, inexhaustible, and so on, and therefore has attracted more and more attention. However, due to the nonlinear characteristics of the photovoltaic system itself and the complex manufacturing process of photovoltaic cells, its conversion efficiency is generally 14% to 15%. In order to allow the solar cell array to output more electrical energy under the same sunshine and temperature conditions, the maximum power point tracking (MPPT) problem is proposed.

MPPT is essentially an optimization process. By measuring voltage, current and power and comparing the changing relationship between them, the position relationship between the current working point and the peak point is determined, and then the current (or voltage) is controlled to move to the current working point and the peak power point, and finally the current (or voltage) is controlled to swing back and forth within a certain range near the peak power point. Fuzzy control has strong adaptability and good robustness. As a new control concept, it is very suitable for solar photovoltaic power generation, which contains many uncertainties and is difficult to describe with an accurate mathematical model.

1 Photovoltaic characteristics

A photovoltaic cell is equivalent to a large-area equivalent diode with an extremely thin PN cross section parallel to the light-receiving surface. Its equivalent circuit is shown in Figure 1.

In Figure 1, I is the output current of the solar cell; Id is the working current of the diode; Irsh is the leakage current; ILG is the photovoltaic cell current source; Rsh is the parallel equivalent resistance of the photovoltaic cell; Rs is the series equivalent resistance of the photovoltaic cell. The output characteristic equation of the photovoltaic cell obtained from Figure 1 is:

Where:

The previous equation shows that the larger the parallel resistance Rsh is, the less it will affect the value of the short-circuit current. Therefore, Rsh can be ignored in the design, and the simplified photovoltaic cell output characteristic equation is obtained:

In formulas (1) to (4), I is the output current of the photovoltaic cell; V is the output voltage of the photovoltaic cell; Ios is the dark saturation current of the photovoltaic cell; T is the surface temperature of the photovoltaic cell; K is the Boltzmann constant (1.38×10-23J/K); λ is the sunlight intensity; q is the unit charge (1.6×10-19C); k1 is the temperature coefficient of the short-circuit current; ISCR is the short-circuit current of the photovoltaic cell under standard test conditions (photovoltaic cell temperature 25℃, sunlight intensity 1000 W/m2); ILG is the photocurrent; EGO is the band gap width of the semiconductor material; Tr is the reference temperature (301.18 K); Ior is the dark saturation current under Tr; A and B are ideal factors, generally between 1 and 2.

With the development of the global economy, energy issues are becoming increasingly acute, and more and more countries are beginning to pay attention to the issues of energy utilization and conversion efficiency. Photovoltaic power generation has the advantages of being pollution-free, noise-free, inexhaustible, and so on, and therefore has attracted more and more attention. However, due to the nonlinear characteristics of the photovoltaic system itself and the complex manufacturing process of photovoltaic cells, its conversion efficiency is generally 14% to 15%. In order to allow the solar cell array to output more electrical energy under the same sunshine and temperature conditions, the maximum power point tracking (MPPT) problem is proposed.

MPPT is essentially an optimization process. By measuring voltage, current and power and comparing the changing relationship between them, the position relationship between the current working point and the peak point is determined, and then the current (or voltage) is controlled to move to the current working point and the peak power point, and finally the current (or voltage) is controlled to swing back and forth within a certain range near the peak power point. Fuzzy control has strong adaptability and good robustness. As a new control concept, it is very suitable for solar photovoltaic power generation, which contains many uncertainties and is difficult to describe with an accurate mathematical model.

1 Photovoltaic characteristics

A photovoltaic cell is equivalent to a large-area equivalent diode with an extremely thin PN cross section parallel to the light-receiving surface. Its equivalent circuit is shown in Figure 1.

In Figure 1, I is the output current of the solar cell; Id is the working current of the diode; Irsh is the leakage current; ILG is the photovoltaic cell current source; Rsh is the parallel equivalent resistance of the photovoltaic cell; Rs is the series equivalent resistance of the photovoltaic cell. The output characteristic equation of the photovoltaic cell obtained from Figure 1 is:

Where:

The previous equation shows that the larger the parallel resistance Rsh is, the less it will affect the value of the short-circuit current. Therefore, Rsh can be ignored in the design, and the simplified photovoltaic cell output characteristic equation is obtained:

In formulas (1) to (4), I is the output current of the photovoltaic cell; V is the output voltage of the photovoltaic cell; Ios is the dark saturation current of the photovoltaic cell; T is the surface temperature of the photovoltaic cell; K is the Boltzmann constant (1.38×10-23J/K); λ is the sunlight intensity; q is the unit charge (1.6×10-19C); k1 is the temperature coefficient of the short-circuit current; ISCR is the short-circuit current of the photovoltaic cell under standard test conditions (photovoltaic cell temperature 25℃, sunlight intensity 1000 W/m2); ILG is the photocurrent; EGO is the band gap width of the semiconductor material; Tr is the reference temperature (301.18 K); Ior is the dark saturation current under Tr; A and B are ideal factors, generally between 1 and 2.

When the load RL changes from 0 to infinity, the output characteristic curve of the solar cell can be obtained as shown in Figure 2. When the load resistance RL is adjusted to a certain value Rm, a point M is obtained on the curve, and the product of the corresponding working voltage and working current is the largest, that is, Pm=ImVm. This point M is now defined as the maximum power output point (MPP).

2 Maximum Power Point Tracking of Photovoltaic Systems

In photovoltaic systems, it is usually required that the output power of photovoltaic cells be maintained at the maximum, that is, to make photovoltaic cells work at the maximum power point, so as to improve the conversion efficiency of photovoltaic cells. MPPT is a process of continuous measurement and adjustment to achieve the best result. It does not need to know the precise mathematical model of the photovoltaic array, but constantly changes the setting value of the controllable parameters during operation, so that the current working point gradually approaches the peak power point, so that the photovoltaic system operates near the peak power point.

For resistive loads, the intersection of its load line and the IV curve determines the operating point of the photovoltaic cell. Different loads RL determine different operating points. Therefore, under different temperature and sunshine intensity conditions, when the maximum power point drifts, the photovoltaic cell can be adjusted to work at the maximum power point again. Regarding the maximum power point tracking algorithm of photovoltaic cells, many previous literatures have proposed a variety of methods, such as voltage feedback method, perturbation observation method, power feedback method, straight line approximation method, actual measurement method and incremental conductance method.

However, these methods cannot track and respond quickly when the environment of photovoltaic modules changes complexly. Conventional methods can only converge to the local highest operating point, but not the real highest point of the PV curve. Therefore, the duty cycle perturbation method is proposed. Figure 3 shows the structure of a general photovoltaic power generation system. The MPPT controller adjusts the input/output relationship by adjusting the duty cycle D of the PWM signal, thereby achieving the impedance matching function.

3 MPPT Implementation Based on Fuzzy Control

3.1 Basic principles of fuzzy control

The foundation of fuzzy control is fuzzy logic, which is closer to human thinking and language expression than traditional logic systems. In some complex systems, especially when there are qualitative inaccuracies and uncertain information in the system, the effect of fuzzy control is often better than conventional control. The basic structure of the fuzzy control system is shown in Figure 4.

The fuzzy control system generally controls the process according to the output error and the change of the error. First, the actual measured precise error e and the error change Δe are transformed into fuzzy quantities through fuzzy processing. At the sampling time k, the error and error change are defined as:

Where yr and yk represent the set value and the process output at time k respectively; ek is the output error at time k. These quantities are used to calculate the fuzzy control rules, and then transformed into precise quantities to control the process.

3.2 Design of fuzzy controller

The design of fuzzy logic controller mainly includes the following contents:

(1) Determine the input variables and output variables of the fuzzy controller;

(2) Summarize and summarize the control rules of fuzzy controllers;

(3) Determine the methods of fuzzification and defuzzification;

(4) Select the domain and determine the relevant parameters.

The solution of fuzzy design is often not unique, and heuristic trial methods must be used to a large extent to obtain the best choice. For the initial design, simulation can be performed first. If the control performance does not meet the requirements, it is necessary to redefine the membership function and sometimes even the input/output.

3.2.1 Fuzzy subsets of input/output quantities and domains

The input and output variables of the fuzzy system are the input power change E; the input last step length A(n-1); and the output step length A(n). The linguistic variables E and A are defined as 8 and 6 fuzzy subsets respectively, namely:

E={NB, NM, NS, NO, PO, PS, PM, PB)

A={NB, NM, NS, PS, PM, PB}

Where: NB, NM, Ns, NO, PO, PS, PM, PB represent fuzzy concepts such as negative large, negative medium, negative zero, positive zero, positive small, positive medium, positive large, etc., and their domains are defined as 14 and 12 levels, namely:

E={-6,-5,-4,-3,-2,-1,-0,+0,+1,+2,+3,+4,+5,+6)

A={-6,-5,-4,-3,-2,-1,+1,+2,+3,+4,+5,+6}

3.2.2 MPPT fuzzy control algorithm

In Figure 5, e(n) represents the actual value of the difference between the output power at the nth moment and the n-1th moment; E(n) represents the value of this difference in the fuzzy set domain; a(n) represents the actual value of the step length at the nth moment; A(n) represents the value of this step length in the fuzzy set domain; Ke and Ka are quantization factors respectively.

By analyzing the characteristic curve between the photovoltaic cell output P and the duty cycle D, and considering the influence of external environmental factors on the photovoltaic cell output power, the final control rules obtained by adjusting the actual simulation results are shown in Table 1.

4 System Modeling and Simulation

Matlab's fuzzy logic toolbox expands Matlab's ability to design fuzzy logic systems and has become an important tool for solving engineering problems using fuzzy methods. Here, the fuzzy logic toolbox in Matlab7.1 is combined for auxiliary design. The fuzzy logic toolbox gives a mamdani type controller by default, and selects the "intersection" method as min; the "union" method as max; the reasoning method as min; the clustering method as max; and the defuzzification method as the centroid method. Figure 6 shows the fuzzy logic toolbox interface.

After the fuzzy controller is designed, the photovoltaic cell model is built using Simulink, as shown in Figure 7.

Secondly, the MPPT fuzzy control system is built as shown in Figure 8.

In the figure, subsystem is the photovoltaic cell model; the S function only realizes the function of D(n)=D(n-1)+a(n). After repeated experiments, the quantization factor Ka is 0.01; Ke is 10. The intensity of the simulated external factors suddenly increases from 600 W/m2 to 900 W/m2, the surface temperature T=25℃, and the maximum simulation step time is set to 0.025 s, and the running time is 10 s. The output power waveform is shown in Figure 9.

Figure 10 is the tracking waveform of the output power of the perturbation observation method. By comparison, it can be found that the use of fuzzy logic control to track the maximum power point of photovoltaic cells is not only fast, but also has basically no fluctuation after reaching the maximum power point, that is, it has good dynamic and steady-state performance.

5 Conclusion

When performing maximum power point tracking in a solar power generation system, fuzzy control is used to intelligently adjust the step size according to the tracking situation and changes in external factors such as battery surface temperature and sunlight intensity.

Simulink is used to build a model and perform simulations. The results show that it is feasible to apply fuzzy control to maximum power point tracking and that it exhibits good control performance.