TW202103435A - Global maximum power point tracking method of solar cell characterized by using a section sampling to track to the vicinity of the maximum power point and then using an alpha-factor perturbation and observation method to improve the tracking accuracy and control the operation point at the maximum power point - Google Patents

Abstract

Disclosed is a global maximum power point tracking method of solar cell, which is realized by a control circuit. The maximum power point tracking method includes the following steps: setting N shading pattern sampling sections, wherein a sampling current value and a sampling voltage value of a sampling point are measured at each one of the N shading pattern sampling sections, and the N is a positive integer; based on a reference voltage value, calculating the voltage per-unit of each sampling voltage value and calculate the product of the sampling current value of each shading pattern sampling section and the voltage per-unit to obtain N products, and selecting the two sampling points corresponding to the first largest product and the second largest product among the N products to be two initial operation points; and based on the two initial operation points, respectively performing a calculation of the <alpha>-factor perturbation and observation method so as to obtain two candidate maximum power points, wherein the one with the larger power value in the two candidate maximum power points is used to be the maximum power point for determining a voltage command.

Description

A method for tracking the global maximum power of solar cells

The present invention relates to a solar cell maximum power tracking method, in particular to a solar cell global maximum power tracking method suitable for partial shading and uniform illumination.

Energy is an indispensable foundation for national development. Most industries rely on energy as a source of production power. With the advancement of technology, the demand for energy is also increasing year by year. The global energy consumption distribution ratio in 2017 shows that at present, Energy consumption is dominated by non-renewable energy, with a proportion of 73.5%, followed by renewable energy at 26.5%. In recent years, environmental changes and global warming issues have made environmental awareness highly valued. Therefore, renewable energy has become the focus of attention. Renewable energy includes water power, wind energy, solar energy, biomass energy, geothermal energy, and tides. According to the development trend of global renewable energy in recent years, the power capacity generated by solar energy is second only to hydropower, and the power capacity generated by renewable energy in 2017 reached 2,195 billion watts. Because of its abundant sources, no fuel cost, and low environmental impact, solar energy is currently one of the most eye-catching renewable energy sources. From 2007-2017 global photovoltaic units and annual new installations, solar power has become an energy development. The main trend.

Among the renewable energy sources, solar energy has become one of the main energy sources in the future due to its clean, low environmental damage and abundant resources. However, most of the current commercially available solar cells have a photoelectric conversion efficiency of less than 20%, and their electrical characteristics are non-linear. The curve also has a Maximum Power Point (MPP), and this curve will be affected by illuminance and temperature. Therefore, how to extract the maximum power from the solar cell to improve the overall efficiency is an important issue. Solar power generation systems generally need to operate solar cells at the maximum power point. This technology is called Maximum Power Point Tracking (MPPT) technology.

In practical applications, the solar power system is not composed of a single solar cell, but is composed of multiple solar cells in series or parallel to meet the voltage or current requirements. In addition, the solar power system is usually installed outdoors and is susceptible to dust, shade and surroundings. The impact of buildings, etc., makes the solar panels receive uneven illumination, causing the power-voltage characteristic curve of the solar cell to produce multiple peaks and maximum power. The position of the rate point is not fixed. This situation is called Partial Shading Conditions (PSCs), and the maximum power tracking technology suitable for partial shading is called Global Maximum Power Point Tracking (GMPPT) technology.

The related research of solar maximum power tracking algorithm can be divided into two main points:

When the solar power system tracks to the maximum power point, transient tracking loss will occur. In order to reduce the transient tracking loss and respond to changes in the external environment, how to have an ideal transient response is an important issue.

When the solar power system is in a stable state for a long time, in order to avoid the steady-state tracking loss caused by the oscillation of the operating point near the maximum power point, how to make the solar power system have good tracking accuracy in the steady state is also in the solar power system An indispensable link.

Conventional maximum power tracking methods include Open Circuit Voltage (OCV), Short Circuit Current (SCI), Perturb and Observe (P&O), and Incremental Conductance ,INC), Fuzzy Logic Control (FLC) method, Artificial Neural Network (ANN) method and Ripple Correlated Control (RCC) method, the above methods work under a single uniform illuminance and temperature The maximum power point can be tracked effectively, and the hardware complexity can be reduced, and the maximum power tracking effect can be improved. However, when there is partial shading in the operating environment, the performance of these maximum power tracking methods will be reduced, or even the maximum power point cannot be tracked. The documents proposed to solve the problem of partial shading are mainly divided into hardware and software. Types of:

1. Hardware type maximum power point tracking method:

Change the centralized power circuit architecture to a distributed power circuit architecture, or solve the problem of partial shading through a reconfigurable circuit architecture. Including: arranging the solar cell module array structure type immediately according to the partial shading situation, the multi-stage power converter independently converts the solar cell module electric energy type, the module integration type, the self-adaptive balanced electric energy type, etc. Since most of this method is a decentralized architecture, each decentralized individual can be independently controlled, so it can ensure that the maximum power point of the whole area can be tracked, but it has a higher cost and complexity.

2. Software type maximum power point tracking method:

Generally used in centralized solar power generation system architecture, only a set of power converters are needed, and multiple peak points of the P-V characteristic curve under partial shading can be located through software to track the global maximum power point. Some literature uses search methods based on numerical theory to gradually narrow the search range to find the maximum power point in the entire domain. Some literature uses Fibonacci Search Algorithm (FSA) as the basis to narrow the search range. Window Search Algorithm (WSA)-based search method, which uses the triangular range formed by the open circuit voltage and short circuit current on the power-voltage characteristic curve and operates according to the current corresponding to the operating point. By gradually narrowing the operating range to track the global maximum power point, the above method can improve the tracking accuracy and hit rate, and shorten the tracking time, but the algorithm is more complex and requires a microprocessor with powerful computing capabilities. The firmware integration of the old solar power generation system is difficult.

An excellent maximum power tracking method must have the characteristics of fast tracking speed, low tracking loss, low implementation cost (soft and hardware complexity), good system compatibility, and easy expansion, among which low implementation cost includes simple algorithm It can be implemented by a low-cost microcontroller, without additional sensing devices and circuits (such as illuminance meters, thermometers, and conversion circuits). For commercial solar power generation systems, based on cost and volume considerations, the utilization and conversion efficiency of solar cells have become extremely important, but the output power of solar cells will vary according to the amount of sunlight and temperature at the time, so , It is necessary to develop a maximum power tracking law, which can still maintain the maximum power output of the solar array under different operating environmental conditions, and has a fast and accurate tracking response. The conventional technology can exert high performance under stable weather conditions, but under partial shade conditions, because the power-voltage characteristic curve becomes more complicated, it exhibits multiple peaks and generates multiple local maximum powers. Because the conventional maximum power tracking method stops searching when it reaches the peak value, it will encounter difficulties in searching the global maximum power point, which will cause the tracking efficiency of the solar power generation system to decrease. Since partial shading is quite common for large-scale solar power generation systems, a novel global maximum power tracking algorithm is urgently needed in this field.

One purpose of this case is to disclose a method for tracking the global maximum power of solar cells, which uses section sampling to track to the vicinity of the maximum power point, and then improves the tracking accuracy with the α- factor perturbation observation method to achieve stable control of the operating point The purpose of the maximum power point.

Another purpose of this case is to disclose a solar cell global maximum power tracking method, which can be implemented by a low-cost microcontroller, without additional sensing devices and circuits, and thus can achieve low software and hardware complexity and system Good compatibility and easy expansion.

Another purpose of this case is to expose a method for tracking the global maximum power of solar cells. Compared with the deterministic cuckoo bird search method of the prior art, the rise time is improved by 34.67%, the stable time is improved by 25.18%, and the energy is tracked on average. The loss is reduced by 35.68%, and the steady-state tracking accuracy can reach 99.99%.

Another purpose of this case is to disclose a method for tracking the global maximum power of solar cells, which can track the global maximum power point under different shading patterns, and its steady-state tracking accuracy under the five shading patterns tested Both are higher than 99.00%.

To achieve the foregoing objective, a global maximum power tracking method for solar cells is proposed, which is implemented by a control circuit. The maximum power tracking method includes the following steps: setting N shading pattern sampling sections, The shading pattern sampling section measures a sampling current value of a sampling point and a sampling voltage value. N is a positive integer; calculate the voltage per unit value of each sampling voltage value according to a reference voltage value and calculate each said The product of the sampling current value and the voltage per unit value of the shading pattern sampling section is used to obtain N products, and the two sampling points corresponding to the largest first two of the N products are selected as Are two starting operating points; and performing an α- factor perturbation observation method according to the two starting operating points to obtain two candidate maximum power points, whichever has the larger power value As the maximum power point to determine a voltage command, the α- factor disturbance observation method calculation includes: △ V _pv ( n ) = α × △ V _pv ( n -1) , P(n-2) < P(n-1) ) And P(n) < P(n-1), where △V _pv (n) is the current voltage disturbance, △V _pV (n-1) is the previous voltage disturbance, and α is less than 1 The constant is used to reduce steady-state oscillation. P(n) is the current power value, P(n-1) is the power value of the previous sample, and P(n-2) is the power value of the previous two samples.

In one embodiment, the reference voltage value is equal to an open circuit voltage value of a solar cell.

In an embodiment, the minimum amplitude of the illumination change of the N shading pattern sampling sections is 100 W/m ² .

In one embodiment, the control circuit includes: a boost converter having an input terminal, a control terminal and an output terminal, the input terminal is used for coupling with a solar cell system, and the control terminal is used for A pulse width modulation signal is received, and the output terminal is used for coupling with a load; and a microcontroller is used for generating the voltage command and providing the pulse width modulation signal according to the voltage command.

In one embodiment, the microcontroller has a digital signal processor for performing an analog-to-digital conversion operation and a digital filtering operation on the current voltage and the current current respectively, and executes a proportional-integral based on the voltage command -Differential control operation and a pulse width modulation operation to output the pulse width modulation signal.

In order to enable your reviewer to further understand the structure, features and purpose of this case, a detailed description of the preferred specific embodiments and diagrams are attached as follows.

100‧‧‧Solar cell system

200‧‧‧Boost converter

300‧‧‧Microcontroller

400‧‧‧Load

Step a‧‧‧Set N shading pattern sampling sections, measure a sampling current value and a sampling voltage value of a sampling point in each of the N shading pattern sampling sections, N is a positive integer

Step b‧‧‧Calculate the voltage per unit value of each of the sampled voltage values according to a reference voltage value and calculate the product of the sample current value and the voltage per unit value of each of the shading pattern sampling sections to obtain N products, two of the sampling points corresponding to the largest first two of the N products are selected as two starting operation points

Step c‧‧‧According to the two initial operating points, perform an α- factor perturbation observation method to obtain two candidate maximum power points, and use the one with the larger power value among the two candidate maximum power points as the maximum power Point to determine a voltage command

FIG. 1 shows a flowchart of an embodiment of the solar cell global maximum power tracking method in this case.

Fig. 2 shows the equivalent circuit diagram of a single diode of a solar cell.

Figure 3a shows the output current-voltage curve of the solar cell under different illuminance values.

Figure 3b shows the output power-voltage curve of the solar cell under different illuminance values.

Fig. 4a shows the current-voltage curve of the ^{solar cell under different temperature values of 1000W/m 2 of full illumination.}

Figure 4b shows the power-voltage curve of the solar cell at a full illuminance of 1000 W/m ^{2 at different temperature values.}

Figure 5a shows the output current-voltage curve of the solar cell under shade.

Figure 5b shows a graph of the output power vs. voltage of the solar cell under shade.

Figure 6 shows a schematic diagram of the hardware architecture of the global maximum power tracking system adopted in this case.

FIG. 7 is a schematic diagram of the structure of the solar power generation system 7 in series and in parallel.

Figure 8a shows a schematic diagram of 120 shading patterns.

Fig. 8b shows a schematic diagram of output current-voltage curves under 120 shading patterns.

Fig. 8c shows a schematic diagram of output power-voltage curves under 120 shading patterns.

Fig. 8d shows the distribution of the maximum power points in the whole area of 120 illuminance patterns.

Fig. 9 is a schematic diagram showing the sampling of the output current-voltage curve of the solar cell in this case.

Figure 10a shows a schematic diagram of the power-voltage curve from the left half plane crossing the maximum power point and then tracking to the right half plane.

Figure 10b shows a schematic diagram of the power-voltage curve from the right half plane crossing the maximum power point and then tracking to the left half plane.

Figure 11 shows the flow chart of the α-factor disturbance observation method.

Figure 12 shows the main firmware flow chart of the global maximum power tracking method in this case.

Figure 13 shows the experimental system setup diagram of the global maximum power tracking method in this case.

Figure 14 shows a schematic diagram of the definition of the performance measurement standard of the global maximum power tracking method in this case.

FIG. 15a shows the power-voltage curve of shading pattern 1. FIG.

Figure 15b shows the simulated power and voltage tracking waveforms of shading pattern 1.

FIG. 15c shows the power-voltage curve of shading pattern 2. FIG.

Figure 15d shows the simulated power and voltage tracking waveforms of shading pattern 2.

Figure 16a shows the success of the conventional cuckoo bird search method with different variable factors β to catch up to the maximum power hit rate.

FIG. 16b shows the power-voltage curve and voltage tracking waveform diagram of the shading pattern 1 simulated by the deterministic cuckoo bird search method of the prior art.

Fig. 16c shows the power-voltage curve and voltage tracking waveform diagram of the shading pattern 2 simulated by the deterministic cuckoo bird search method of the prior art.

Figure 17a shows the power-voltage curve of the 7slp system under uniform illumination (1000W/m ^{2 ).}

Figure 17b shows a simulation diagram of the power and voltage tracking waveforms of this case under standard test conditions.

Fig. 17c shows a simulation diagram of the power and voltage tracking waveforms of the conventional technology's variable step disturbance observation method under standard test conditions.

Figure 17d shows a simulation diagram of the power and voltage tracking waveforms of the deterministic cuckoo bird search method of the conventional technology under standard test conditions.

Figure 18a shows the actual measurement diagram of the simulator in shading pattern 1 of this case.

Figure 18b shows the measured tracking waveform of the shading pattern 1 in this case.

Figure 18c shows the actual measurement diagram of the simulator in shading pattern 2 of this case.

Figure 18d shows the measured tracking waveforms of the shading pattern 2 in this case.

Please refer to FIG. 1, which shows a flowchart of an embodiment of the solar cell global maximum power tracking method in this case.

As shown in the figure, the global maximum power tracking method of solar cells in this case is realized by a control circuit. The maximum power tracking method includes the following steps: setting N shading pattern sampling sections, The pattern sampling section measures a sampling current value and a sampling voltage value at a sampling point, and N is a positive integer; (step a); calculate the voltage per unit value of each sampling voltage value according to a reference voltage value and calculation The product of the sampling current value and the voltage per unit value of each of the shading pattern sampling sections is used to obtain N products, and the two products corresponding to the largest first two of the N products are selected. The sampling points are used as two initial operating points; (step b); and an α- factor perturbation observation method is performed according to the two initial operating points to obtain two candidate maximum power points, and the two candidate maximum power points The point with the larger power value is used as the maximum power point to determine a voltage command. The α- factor disturbance observation method includes: △ V _pv ( n ) = α × △ V _pv ( n -1) , P(n- 2) < P(n-1) and P(n) < P(n-1), where △V _pv (n) is the current voltage disturbance and △V _pv (n-1) is the previous voltage The amount of disturbance, α is a constant less than 1 to reduce steady-state oscillation, P(n) is the current power value, P(n-1) is the power value of the previous sample, P(n-2) is the previous two samples Power value; (Step c).

Wherein, the reference voltage value is, for example, but not limited to, equal to a solar cell open circuit voltage value; the minimum amplitude of the illumination change of the N shading pattern sampling sections is, for example, but not limited to, 100W/m ² .

The control circuit includes, for example, but not limited to: a boost converter having an input terminal, a control terminal and an output terminal. The input terminal is used for coupling with a solar cell system, and the control terminal is used for receiving a Pulse width modulation signal, and the output terminal is used for coupling with a load; and a microcontroller is used for generating the voltage command and providing the pulse width modulation signal according to the voltage command.

Wherein, the microcontroller has a digital signal processor for performing an analog-to-digital conversion operation and a digital filtering operation on the current voltage and the current current respectively, and executes a proportional-integral control operation and a digital filtering operation according to the voltage command. The pulse width modulation operation is performed to output the pulse width modulation signal.

The following will explain the principle of this case:

Electrical characteristics of solar cells:

Please refer to FIG. 2, which shows the equivalent circuit diagram of a single diode of a solar cell.

As shown in the figure, the electrical characteristic of the solar cell is a non-linear power source, and reverse current is not allowed. Its voltage and current show an exponential curve relationship. Therefore, when the output voltage of the solar cell changes, its output current will also change. According to the equivalent circuit, the relationship between the output voltage and current of the solar cell is shown in equation (1).

Among them, I _pv is the output current of the solar cell, I _ph is the photoelectric conversion current, I _S is the reverse saturation current of the diode, q is the carrier charge (1.60×10 ^-19 C), and V _pv is the output of the solar cell Voltage, R _S is the equivalent series resistance, k is the Boltzmann constant (1.38065×10 ^-23 J/°K), A is the ideality factor (between 1 and 2), T is the absolute temperature value, and N is the solar cell The number of series, R _P is the equivalent parallel resistance.

The relationship between the photoelectric conversion current I _ph and the illuminance value is shown in equation (2).

Among them, S is the illuminance value in W/m ² , and I _SC is the short-circuit current of the solar cell.

It can be seen from equation (2) that when the illuminance value S increases, the output power of the semiconductor increases due to the increase of the irradiated light energy, and the photoelectric conversion current I _{ph of the} solar cell also increases.

Generally speaking, since the parallel resistance value of solar cells is much larger than the series resistance value, equation (1) and equation (2) can be used to sort out equation (3).

In order to observe the influence of the change of the illuminance value and the ambient temperature value on the output characteristic curve of the solar cell, the equation (3) can be rewritten into the equation (4).

It is known from equation (4) that due to the natural logarithmic relationship, the solar cell output voltage V _pv only slightly changes when the illuminance rises. When the illuminance value S rises, the output electric energy of the semiconductor increases due to the increase of the irradiated light energy, and the photoelectric conversion current I _{ph of the} solar cell also increases. And because the output short-circuit current I _SC is almost proportional to the illuminance value S , the short-circuit current of the solar cell increases significantly, and the output current I _{pv at the} operating point also increases.

Please refer to FIGS. 3a and 3b together. FIG. 3a shows the output current-voltage curve diagram of the solar cell under different illuminance values; FIG. 3b shows the output power-voltage curve diagram of the solar cell under different illuminance values.

Among them, the ambient temperature is fixed at 25°C, and the different illuminance values are 200W/m ² , 400W/m ² , 600W/m ² , 800W/m ² and 1000W/m ² , as shown in the figure, these five illuminances The value is based on the five solar cell output curves drawn after the calculation of equation (3), and the characteristic curve will change with the change of the illuminance value.

In order to facilitate the observation of the influence on the output characteristic curve of the solar cell when the illuminance or ambient temperature changes, equation (4) can be rearranged into equation (5).

Please refer to Figures 4a and 4b together, where Figure 4a shows the current-voltage curve of the solar cell at different temperature values ^{at a full illumination of 1000W/m 2} ; Figure 4b shows the solar cell at different temperatures ^{at a full illuminance of 1000W/m 2} Power-voltage curve at the lower value.

According to the Standard Test Condition (STC) of the solar cell module, set the temperature and illuminance to 25° C and full illuminance (1000W/m ² ) respectively, and use MATLAB mathematical software and formula (3) to draw the standard test The output current-voltage and power-voltage characteristic curves of solar cell modules under conditions. As shown in the figure, the output characteristic curve of the solar cell will also be affected by the ambient temperature value. From equation (5), when the ambient temperature value rises, the conversion current will decrease due to the characteristics of the equivalent diode and the output of the solar cell will be reduced. The current I _pv rises slightly; and from equation (3), it is known that the ambient temperature value is proportional to the solar cell output voltage V _pv , but the solar cell output current I _pv will also increase as the temperature value rises, and it is affected It is much larger than the output voltage V _pv , so the ambient temperature value has little effect on the output voltage V _pv of the solar cell. On the contrary, the equivalent series resistance cross-voltage attenuation increases due to the slight increase in the output current I _pv , so that the output voltage V of the solar cell _PV causes a significant drop, so when the temperature of the solar cell rises, its output power also drops.

The aforementioned characteristic curves are all introduced for a single solar cell. However, in practical applications, the output power that a single solar cell can provide is limited, so multiple solar cells are connected in series or parallel to form a solar cell module To meet the required specifications. Since it is no longer powered by a single solar cell, when a shaded situation occurs, the shaded solar cell not only fails to provide energy normally, but also consumes energy as a load in the form of heat, which is called a hot spot. The solar cell will be short-circuited by the bypass diode of the solar cell, so the solar characteristic curve will show a multi-peak situation.

Please refer to Figures 5a and 5b. Figure 5a shows the output current-voltage curve of the solar cell under shade; Figure 5b shows the output power-voltage curve of the solar cell under shade. Figure.

Take 7 series and 1 parallel solar cell modules as an example, when shading occurs, and the illuminance of the solar cells are 100W/m ² , 200W/m ² , 300W/m ² , 400W/m ² , 500W/m ² , 600W/m ² and 700W/m ² , when the temperature is 25℃, as shown in Figure 5a, because it is a system of 7 series and 1 parallel, the curve will show 7 peaks, which means that the system of several series is at most There will be several peaks. As shown in Figure 5b, the position of the maximum power point appears at the fourth peak of the curve, but the position of the maximum power point will change with different illuminances. This shows that the conventional maximum power tracking method is not suitable for shading situations. the reason.

The hardware architecture of the global maximum power tracking system adopted by the solar cell in this case:

Please refer to FIG. 6, which shows a schematic diagram of the hardware architecture of the global maximum power tracking system adopted in this case.

The output characteristic curve of the solar cell is a nonlinear exponential function curve, and there is a maximum power point, and the solar cell will produce a different characteristic curve according to the ambient temperature and sunlight intensity at the time, so it can output the operating point of the maximum power It is also different. In order to capture the maximum power output at any time, the operating point must be operated at the maximum power point on the solar cell characteristic curve to maximize its energy. This control method is called maximum power tracking technology.

As shown in the figure, the hardware architecture of the solar maximum power tracking system adopted in this case includes a solar cell system 100, a boost converter 200, and a microcontroller 300.

The boost converter 200 has an input terminal, a control terminal and an output terminal. The input terminal is used for coupling with the solar cell system 100. The control terminal is used for receiving a pulse width modulation signal, and the The output terminal is used for coupling with a load 400.

The microcontroller 300 has a digital signal processor for sampling, an analog-to-digital conversion operation, and a digital filtering operation on the voltage and current output by the solar cell system 100, and then performing the operation of the maximum power tracking method to generate a The voltage command generates a duty cycle through a proportional-integral-derivative control operation and a pulse width modulation operation to control the boost converter 200 to achieve the purpose of maximum power tracking.

Wherein, since the output voltage of the solar cell system 100 of the prior art is generally too low, the boost converter 200 is used to increase the output voltage of the solar cell system 100. The boost converter 200 is, for example, but not limited to A step-up DC-DC converter; the microcontroller 300 is implemented by, for example, but not limited to, a low-cost digital signal processor. In this case, the low-cost dsPIC33FJ16GS502 digital signal processor launched by the American company Microchip is used as the digital control core for maximum power tracking.

Please refer to FIG. 7, which shows a schematic diagram of the solar power generation system 7 in series and 1 in parallel.

In this case, 7 solar cell modules of LDK-85 manufactured by LDK Solar are connected in series as the input source of the system, that is, the structure of 7 series and 1 parallel, which is represented by 7slp. As shown in Figure 7a, the electrical of solar cell modules The specifications are shown in Table 1.

As mentioned above for the electrical characteristics of solar cells, the output power-voltage curve of solar cell modules will be affected by the illuminance and ambient temperature, but the influence of slight illuminance changes on the power-voltage curve is not obvious, and different illuminances affect the power-voltage The influence of the curve is much greater than the influence of the ambient temperature. Therefore, this case assumes that all solar cell modules are operated at an average ambient temperature of 25°C, and the standard test condition (STC) illuminance 1000W/m ^{2 is} regarded as full illuminance. And the illuminance is changed from 0W/m ² to 1000W/m ² in a fixed range.

If the amplitude of the illuminance change is set too large, the simulated data will be incomplete; on the contrary, if the amplitude of the illuminance change is set too small, the simulated data will be too large. In order to balance the data integrity and the amount of data, and to improve the identification of different illuminances In this case, 100W/m ² is selected as the minimum illuminance change range. Within the test illuminance change range, 100W/m ^{2 is} used as the minimum illuminance during the test, and 1000W/m ^{2 is} used as the maximum illuminance during the test, that is, there is a total of 10 kinds of illuminance.

The 7slp solar power system is a series structure, and the same illuminance arranged in different ways will have the same shading effect, so the calculation of the possible shading style only considers the number of combinations instead of the number of arrangements, and does not repeat the shading of illuminance. The shade pattern calculation is represented by equation (6).

Among them, n represents the total number of possible illuminances, and m represents the total number of illuminances appearing in the solar power generation system.

Please also refer to Figures 8a~8d, where Figure 8a shows a schematic diagram of 120 shading patterns, Figure 8b shows a schematic diagram of the output current-voltage curve under 120 shading patterns, and Figure 8c shows a schematic diagram of 120 shading patterns. A schematic diagram of the output power vs. voltage curve under one shading pattern. Fig. 8d shows the distribution of the maximum power point interval for the 120 illuminance patterns.

If the illuminance of each solar cell module is not repeated, for the 7slp architecture used in this case, C(10,7)=120 different shading patterns will be produced. Because this case is simulated based on the shading pattern of the 7slp solar power system under non-repetitive illuminance, as shown in Figure 8a. The e illuminance combination under shading pattern is (400W/m ² , 500W/m ² , 600W/m ² , 700W/m ² , 800W/m ² , 900W/m ² , 1000W/m ² ), as shown in Figure 8b, Figure 8c shows that there are multiple regional maximum power points and a global maximum power point on the output characteristic curve of the solar cell, and the characteristic curves formed by different shading styles will also be different.

In this case, MATLAB software is used to simulate the distribution of the global maximum power point of various shading styles. At the same time, the maximum power point voltage value under each shading style is also recorded. The statistics of the results are shown in Figure 8d. It is possible to know the number of times the global maximum power point (GMPP) appears in each interval. Under the 7slp architecture, GMPP has the highest probability of appearing in the fifth interval, reaching 47.5%, followed by the fourth interval, reaching 35%, and the third interval The probability of the first and seventh intervals is 13.33% and 3.33%, respectively, the probability of the second interval is 0.833%, and the probability of the first interval and the seventh interval is 0.

Please refer to FIG. 9, which shows a sampling diagram of the output current-voltage curve of the solar cell in this case.

As shown in the figure, the 7 marked dots are the locations of the 7 sampling points, and the asterisk is the location of the maximum power point of the whole area under a shading pattern, and the illuminance combination of the shading pattern is also (400W/m ² , 500W/m ² , 600W/m ² , 700W/m ² , 800W/m ² , 900W/m ² , 1000W/m ² ). There are 7 smooth sections in the figure, where samples are taken in each of the 7 smooth sections, and then the voltage and current values of the 7 sampling points are substituted into equation (7) to derive 7 illuminances under the shading pattern.

Since the parallel resistance value of solar cells is much larger than the series resistance value, equation (5) can be simplified to equation (8).

In addition, the temperature change during sampling in the smooth section has little effect on the estimated illuminance, so equation (8) can be simplified into equation (9), and the calculation equation of illuminance can be derived as described in equation (7).

Regarding how to select sampling points in the 7 smooth sections, this case uses MATLAB to simulate the selectable range among the 7 sections under various shading patterns and take the intersection of the results. The starting point is the beginning of the smooth section of each section. When the absolute value of current change to voltage change is less than the set value, it means that the current change is too large and the sampling point is no longer smooth. In the section, the selectable range of 7 sections is shown in Table 2.

Among them, the value in brackets in Table 2 is calculated from the open circuit voltage value of LDK-85 solar cell 21.6V.

With the distribution of the 120 shading patterns in each section obtained by the previous simulation, record the maximum power point voltage of the different shading patterns in each section, and then calculate the maximum power point voltage of the different shading patterns in each section The values are averaged, as shown in Table 3.

Among them, the Per-unit value in Table 3 is calculated based on the LDK-85 solar cell open circuit voltage value of 21.6V.

It can be seen from Fig. 8d that the maximum power point without shading pattern falls in the first interval and the seventh interval, but for the convenience of subsequent calculations, the data in the first interval and the seventh interval in Table 3 are the best and the worst shading. The average value of the maximum power point voltage of the two sections in the shade pattern is taken as the average value of the maximum power point voltage of the two sections, and the best combination of illuminance is 400W/m ² , 500W/m ² , 600W/m ² , 700W/m ² , 800W/m ² , 900W/m ² and 1000W/m ² , the worst combination of illumination is 100W/m ² , 200W/m ² , 300W/m ² , 400W/m ² , 500W/m ² , 600W/m ² And 700W/m ² , and the data in other intervals can be obtained by taking the average of the global maximum power point voltage corresponding to the number of times in each section in Fig. 8d.

By sampling the 7 sets of voltage and current values, the illuminance combination under a certain shading pattern can be derived, and the 7 sets of current values are sorted from large to small, and then the maximum power point voltage values in Table 4 are marked with each value. The product ( P = I × V ), that is, the largest current value is multiplied by 0.80, and the second largest current value is multiplied by 1.62, so 7 sets of product results will be obtained, and the largest two sets are selected, and then The product of the corresponding voltage per unit value and the open circuit voltage value is used as the two sets of reference voltage values for the second-stage α- factor disturbance observation method. The operating points are controlled at the maximum power point through the α- factor disturbance observation method, and finally power tracking is selected The larger value is the maximum power point, and the purpose of choosing the two groups with the largest product result here is to avoid the situation where the product result is too close and misjudges the maximum power point.

Alpha factor disturbance observation method:

The working principle of the α- factor perturbation observation method is similar to that of the perturbation observation method. In order to improve the problem of the fixed-step perturbation observation method oscillating near the maximum power point, many variable-step perturbation observation methods have been proposed in the literature, but they involve many parameters. And the algorithm is more complicated, so this case is based on the spirit of the variable step perturbation observation method, using the α- factor perturbation observation method, as shown in equation (10).

△ D ( n ) = α ×△ D ( n -1) (10)

Among them, △ D ( n ) is the current duty cycle disturbance, △ D ( n-1 ) is the previous duty cycle disturbance, and α is the set α factor.

It can be seen that the current duty cycle disturbance quantity △ D ( n ) is only related to the previous duty cycle disturbance quantity △ D ( n-1 ) and α factor, and its relational expression is simple, and the α factor in the formula must be less than 1 Reduce the steady-state oscillation problem by reducing the disturbance of the duty cycle. In order to ensure that the system responds quickly, this method needs to meet the following two conditions to execute equation (10).

P ( n -1)> P ( n ) (11)

P ( n -1)> P ( n -2) (12)

Among them, P(n) is the current power value, P(n-1) is the power sampling value of the previous time, and P(n-2) is the power sampling value of the previous two times.

Please refer to Figures 10a~10b. Figure 10a shows a schematic diagram of the power-voltage curve crossing the maximum power point from the left half plane and then tracking to the right half plane. Figure 10b shows the power-voltage curve crossing the right half plane. Schematic diagram of tracking to the left half plane after the maximum power point.

As shown in the figure, the common point of the two is crossing the maximum power point, and crossing the maximum power point, that is, the operating point is located near the maximum power point. In order to eliminate the oscillation problem of the fixed-step disturbance observation method near the maximum power point, this method Multiply the duty cycle disturbance amount by the α factor to reduce the disturbance degree, thereby achieving the effect of improving the accuracy of steady-state tracking. This conditional design can take into account whether it is started from the open-circuit voltage terminal or the short-circuit current terminal, and when the conditions are not met , Which means that there is still a certain distance between the current operating point and the maximum power point. At this time, a larger duty cycle disturbance can be used to shorten the transient tracking time.

Please refer to FIG. 11, which shows a flow chart of the α- factor disturbance observation method.

As shown in the figure, first sample the voltage and current value of the solar cell, and calculate the power value, then determine whether the conditional formula (11) and the conditional formula (12) are established, and determine the magnitude of the disturbance of the duty cycle through the result, and then use the disturbance Observe the principle of the observation method and calculate the duty cycle disturbance required by the current operating point position, and then update the duty cycle value. Repeat the above process to achieve the purpose of maximum power tracking.

Please refer to FIG. 12, which shows a flowchart of the firmware main program of the global maximum power tracking method in this case.

As shown in the figure, the firmware main program structure mainly includes the function setting of the peripheral pins of the digital signal processor, internal working environment setting, register setting, interrupt setting, timer setting, analog/digital converter and pulse width adjustment. Change and other peripheral settings. First provide power to the digital signal processor, then initialize the variables used, set the internal oscillator, set the timer, select the input and output ports, set and enable the analog/digital converter and pulse width modulation module And interrupt vector setting, then the main program enters an endless loop and waits for an interrupt to occur. The interrupt part only uses ADC interrupt in this case, which can be divided into three parts: sampling, filtering and maximum power tracking algorithm. The sampling part sends the signal to the digital signal processor through the analog/digital converter in order to prevent the influence of high-frequency noise. In this case, a finite impulse response filter (Finite Impulse Response Filter, FIR Filter) was used to eliminate noise. As the actual sunshine changes more Slow, so the voltage command update time does not need to be too fast. Here, 0.2 seconds is designed as the voltage command change time, because the ADC module enters an interrupt every 1ms. When the count is less than 0.2 seconds, only the solar cell module input voltage Perform FIR filtering with the current and use PID calculation to follow the voltage command provided by the last maximum power tracking algorithm. When the count reaches the condition, it enters the maximum power tracking subroutine and repeats the above actions to complete the maximum power tracking.

Comparison and analysis of this case and conventional technology:

Then use the simulation software MATLAB to realize the global maximum power tracking method in this case, and simulate and record the tracking performance of the various illuminance patterns generated by the 7 series and 1 parallel solar cell modules, and the simulation results are actually used for the maximum power tracking experiment. Finally, the experimental results are analyzed and compared.

Please refer to Figure 13, which shows the experimental system setup diagram of the global maximum power tracking method in this case.

As shown in the figure, during the experiment in this case, the input source uses the TerraSAS ETS 600X8 D-PVE solar cell simulator introduced by AMETEK to simulate the solar cell model LDK-85 introduced by LDK Solar, and the power level is boosted. For the converter, the load end uses the 63108A electronic load introduced by Chroma and operates in constant voltage mode and stabilized to 200V for experiments. The control stage uses the digital signal processor dsPIC33FJ16GS502 introduced by Microchip to cooperate with the MPPT tracking written by Microchip. The algorithm performs global maximum power tracking, and the waveform measurement part uses the DPO4054 oscilloscope introduced by Tektronix.

When TerraSAS ETS 600X8 D-PVE simulates the characteristic curve of the solar cell, the internal program uses 1024 voltage and current points and uses interpolation to draw the current-voltage characteristic curve of the solar cell. The user can also directly provide 1024 points. Input the open circuit voltage V _oc , short circuit current I _sc , maximum power point voltage V _mpp , maximum power point current I _mpp , and fill factor (Fill Factor, FF) and temperature coefficient and other relevant information, the corresponding solar cell current-voltage characteristic curve can be established. In this case, 7 series LDK Solar solar cell modules, model LDK-85, manufactured by LDK Solar are used as the system input source. , And the electrical specifications of the monolithic solar cell modules are shown in Table 1, and the corresponding specifications are shown in Table 4 after 7 strings and 1 parallel.

Definition of performance evaluation parameters:

Please refer to Figure 14 and Table 5 together, which show a schematic diagram of the definition of the performance measurement standard of the global maximum power tracking method in this case.

Before the simulation, the performance evaluation standard must be defined to achieve the fairness and accuracy of the comparison. The gray area shown in Figure 14 is to track the power loss, and the setting of 10 seconds is mainly based on the simulation results and the stability time of each method The maximum value is about 5 seconds. Multiply it by 2 to take into account both the transient response and steady-state response tracking power loss.

Simulation and comparison of shade situations:

It can be seen from Figure 8d that the global maximum power point of the 7slp solar power system in this case does not appear in the first peak and the seventh peak. In order to verify that the maximum power tracking technology proposed in this case can track under different shading patterns To the maximum power point of the whole region, select 5 shade styles with the maximum power point of the whole region distributed in different intervals from 120 shading styles for simulation. The illuminance combinations of the 5 shading styles are shown in Table 6.

When performing simulations of different shading styles and comparison methods, the maximum power tracking command change time is represented by steps, and a step is defined as 0.2 seconds, and the maximum simulation time is set to 10 seconds, that is The maximum number of steps is 50 steps. Carrying out the proposed method in the case of analog shade, the factor α observation of the disturbance factor α is set to 0.5.

Please also refer to Figures 15a~15d, where Figure 15a shows the power-voltage curve of shading pattern 1, Figure 15b shows the simulated power and voltage tracking waveforms of shading pattern 1, and Figure 15c shows it The power-voltage curve of shading pattern 2. FIG. 15d shows the simulated power and voltage tracking waveforms of shading pattern 2.

As shown in Figures 15a and 15c, the figure shows the position of the global maximum power point and its value. As shown in Figures 15b and 15d, the unit of the abscissa is the number of steps. The figure shows the α used in this case. The factor disturbance observation method effectively changes the voltage step size, so the required rise time and stabilization time are shorter, and it can also greatly reduce the tracking power loss. The simulation results of each measurement standard for the 5 shading patterns are summarized in Table 7 Shown.

Please also refer to Figures 16a~16c, where Figure 16a shows the success of the conventional technology's deterministic cuckoo bird search method with different variable factors β to the maximum power hit rate, and Figure 16b shows the conventional technology's deterministic cuckoo The power-voltage curve and voltage tracking waveform of shading pattern 1 simulated by the bird search method. Figure 16c shows the power-voltage curve and voltage of the shading pattern 2 simulated by the decisive cuckoo bird search method of the prior art. Trace the waveform graph.

Then compare the deterministic cuckoo bird search method of the conventional technology with the method proposed in this case. The deterministic cuckoo bird search method of the conventional technology mentions that the selection of the variable factor β will affect the successful hit rate in the shaded situation. Therefore, this case also divides the variable factor β into 16 equal parts, and simulates 120 different shade situations, and then selects the higher hit rate to compare with the method proposed in this case. As shown in Figure 16a, when the variation factor β is 1/16 to 3/16, the hit rate is as high as 119 times or more, but the stability time and tracking power loss under these three variation factors are higher, so the hit is weighed For reasons such as power loss and tracking power loss, this case selects the conventional technology deterministic cuckoo bird search method with a variation factor β of 6/16 to simulate and compare with this case, and the illuminance combinations of the 5 shading patterns are also the same as in Table 6. Table 8 shows the simulation results of the measurement standards of the 5 shade patterns of the deterministic cuckoo bird search method of the conventional technology.

Finally, the results of this case and the deterministic cuckoo-bird search method of the conventional technology under 120 shade patterns are summarized as shown in Table 9.

It can be seen from the above table that this case has improved 34.67% and 25.18% in terms of rise time and settling time respectively, and reduced 35.68% in tracking power loss, and the steady-state tracking accuracy can reach 99.99%.

Please refer to Figures 17a~17d. Figure 17a shows ^{the power-voltage curve of the 7slp system under uniform illumination (1000W/m 2} ), and Figure 17b shows the power and voltage tracking of this case under standard test conditions. Waveform simulation diagram. Figure 17c shows the power and voltage tracking waveform simulation diagram of the conventional technique's variable step perturbation observation method under standard test conditions. Figure 17d shows the conventional technique's deterministic cuckoo bird search method in the standard. Power and voltage tracking waveform simulation diagram under test conditions.

As shown in Figure 17a, the figure has marked the location of the maximum power point and its maximum power is 597.558W, as shown in Figures 17b~17d, the standard test condition is that the ambient temperature is 25℃, and the illuminance is 1000W/m ² at full illumination.

The evaluation values of the 10 kinds of uniform illuminance in this case are shown in Table 10.1 and Table 10.2.

The variable step disturbance observation method of the conventional technology (M value is 1) for each evaluation value of 10 kinds of uniform illuminance is shown in Table 11.1 and Table 11.2.

The evaluation values of the deterministic cuckoo bird search method (variation factor β = 6/16) of the conventional technology in 10 kinds of uniform illuminance are shown in Table 12.1 and Table 12.2.

It can be seen from Table 11.1 and Table 11.2 that when the illuminance is below 400W/m ² , the rise time and settling time of the conventional technology's variable step perturbation observation method are both the set simulation time maximum value of 10s, that is, when the illuminance is below 400W/m 2 In these four uniform illuminance situations, the method has not completed the maximum power tracking within the set time, and it is known from Table 10.1 and Table 10.2 that this problem will not occur in the measurement standards of this case. Furthermore, by observing the steady-state average power of Tables 10 to 12 under 1000W/m ² illuminance, it can be verified that all three methods can successfully track to the maximum power value in Fig. 17a.

The measured environmental conditions are the same as the simulation experiment, and the program also updates the maximum power tracking command every 0.2 seconds. Similarly, in order to verify that this case can track the global maximum power point under different shading styles, five shade styles with global maximum power points distributed in different intervals from 120 shade styles were selected for actual measurement, and these 5 types The illuminance combinations of shading styles are shown in Table 6.

Please refer to Figures 18a~18d together, where Figure 18a shows the actual measurement of the simulation machine in the shading pattern 1, Figure 18b shows the waveform of the actual measurement tracking in the shading pattern 1, and Figure 18c shows the actual situation. Figure 18d shows the actual measurement tracking waveform of the shading pattern 2 in this case.

Among them, in the actual measurement tracking waveform part, this case uses a digital storage oscilloscope to record the output voltage, current and power waveform data points of the solar cell, and then uses MATLAB software to draw it to facilitate the calculation of various evaluation values. There are 5 shading patterns The evaluation results of each actual measurement are shown in Table 13.

With the design disclosed above, this case has the following advantages:

1. The global maximum power tracking method for solar cells disclosed in this case uses section sampling and tracking to the vicinity of the maximum power point, and then improves the tracking accuracy with the α- factor perturbation observation method to achieve stable control of the operating point at the maximum power Point of purpose.

2. The global maximum power tracking method for solar cells disclosed in this case can be implemented with a low-cost microcontroller, without additional sensing devices and circuits, thereby achieving low software and hardware complexity and system compatibility Good, and easy to expand and other features.

3. The global maximum power tracking method of solar cells disclosed in this case has improved rise time by 34.67%, improved stability time by 25.18%, and reduced average tracking power loss by 35.68 compared to the deterministic cuckoo bird search method of the prior art. %, and the steady-state tracking accuracy can reach 99.99%.

4. The global maximum power tracking method for solar cells disclosed in this case can track the global maximum power point under different shading patterns, and its steady-state tracking accuracy is higher than the five shading patterns tested. 99.00%.

The disclosure in this case is a preferred embodiment, and any partial changes or modifications that are derived from the technical ideas of the case and can be easily inferred by those who are familiar with the art will not deviate from the scope of the patent right of the case.

In summary, regardless of the purpose, means and effects of this case, it is showing its technical characteristics that are very different from conventional knowledge, and its first invention is suitable for practical use, and it is also in compliance with the patent requirements of the invention. I implore the examiner to observe carefully and pray. Granting patents as soon as possible will benefit the society and feel the virtues.

Step a‧‧‧Set N shading pattern sampling sections, and measure a sampling current value and a sampling voltage value of a sampling point in each of the N shading pattern sampling sections. N is a positive integer

Step b‧‧‧Calculate the voltage per unit value of each of the sampled voltage values according to a reference voltage value and calculate the product of the sample current value and the voltage per unit value of each of the shading pattern sampling sections to obtain N products, two of the sampling points corresponding to the largest first two of the N products are selected as two starting operation points

Step c‧‧‧According to the two initial operating points, perform an α- factor perturbation observation method to obtain two candidate maximum power points, and use the one with the larger power value among the two candidate maximum power points as the maximum power Point to determine a voltage command

Claims (5)

A global maximum power tracking method for solar cells is realized by a control circuit. The maximum power tracking method includes the following steps: setting N shading pattern sampling sections, and each amount in the N shading pattern sampling sections Measure a sampled current value and a sampled voltage value of a sampling point, N is a positive integer; calculate the voltage per unit value of each sampled voltage value according to a reference voltage value and calculate the total value of each sampled section of the shading pattern The product of the sampled current value and the voltage per unit value is used to obtain N products, and the two sampling points corresponding to the largest first two of the N products are selected as two initial operating points; And according to the two initial operating points, an α- factor perturbation observation method is performed to obtain two candidate maximum power points, and the one with the larger power value among the two candidate maximum power points is used as the maximum power point to determine one Voltage command, the α- factor disturbance observation method calculation includes: △ V _pv ( n ) = α × △ V _pv ( n- 1) , P(n-2)<P(n-1) and P(n)<P (n-1), where △V _pv (n) is the current voltage disturbance, △V _pv (n-1) is the previous voltage disturbance, and α is a constant less than 1 to reduce steady-state oscillation, P (n) is the current power value, P(n-1) is the power value of the previous sample, P(n-2) is the power value of the previous two samples. For example, the solar cell global maximum power tracking method described in the scope of the patent application, wherein the reference voltage value is equal to an open circuit voltage value of the solar cell. For example, the solar cell global maximum power tracking method described in the scope of the patent application, wherein the minimum amplitude of the illumination change of the N shading pattern sampling sections is 100W/m ² . For example, the solar cell global maximum power tracking method described in the scope of patent application, wherein the control circuit includes: a boost converter with an input terminal, a control terminal and an output terminal, and the input terminal is used to interact with A solar cell system is coupled, the control terminal is used to receive a pulse width modulation signal, and the output terminal is used to couple to a load; and a microcontroller is used to generate the voltage command and respond to the voltage Command to provide the pulse width modulation signal. For example, the solar cell global maximum power tracking method described in the scope of patent application, wherein the microcontroller has a digital signal processor for performing an analog-to-digital conversion operation and a digital signal for the current voltage and the current current, respectively Filtering operation, and executing a proportional-integral-derivative control operation and a pulse width modulation operation according to the voltage command to output the pulse width modulation signal.

Images