FIELD OF INVENTION

Photovoltaic systems with improved conversion efficiency of solar energy.

OBJECT OF INVENTION

The first object is for extracting maximum power from PV arrays using a fixed configuration under different shading conditions.

A major challenge in PV systems is to make it energy efficient. One of the major factors which contribute to the reduction of PV power is the partial shading. The reduction in power depends on module interconnection scheme and shading pattern. Different interconnection schemes are used to reduce the losses due to partial shading. This paper presents a fixed interconnection scheme for the PV arrays which enhances the PV power under different shading conditions. The proposed scheme facilitates to distribute the effect of shading over the array thereby reducing the losses due to partial shading. A comparison is made between Electrical Array Reconfiguration (EAR) scheme and proposed scheme for a 5X5 PV array. The performance of the system is investigated for different shading conditions and the MATLAB/SIMULINK results are presented to show that power extracted from the PV array under partial shading conditions is improved when compared to other interconnection schemes.

DESCRIPTION OF INVENTION
In recent years, the utilization of renewable energy has become an attractive alternative to fossil fuels due to the growing concern on the environmental issues [1]-[2]. Photovoltaic is emerging as a promising solution due to the improvement in the semiconductor technology. Since Photovoltaic systems are relatively expensive, researchers in the field of PV are concentrating more on improving the conversion efficiency of solar energy [3]-[4]. There are various factors that contribute to the reduction of output power from the PV arrays. One of the major factors is the partial shading [5]. Partial shading is caused due to the passage of clouds, buildings, towers and trees [6]-[8]. The losses due to partial shading are mainly due to the electrical configuration of the modules in the array because a shaded module in series with other un-shaded modules limits the string current thereby reducing the maximum power generated by the array [9]-[10].

The reduction in power depends on module interconnection scheme and shading pattern. Different interconnection schemes such as Series-Parallel (SP), Total Cross Tied (TCT) and Bridge Linked (BL) have been proposed in literature [11]-[12] to interconnect the modules in the array. Reduction of mismatch losses in the array by changing the interconnection scheme of the modules in PV arrays have been addressed [13]-[15]. Investigation on fault tolerance [16] in different interconnection schemes reveals that TCT or BL are comparatively less susceptible to electrical mismatches. The study on the reliability which is an important parameter to assess the operational life time of PV array indicates that life of an array is almost doubled when crossties are introduced in the array [17]. Employing modularized network based on crossties increases the operating life of PV arrays by 30% [17]. The electrical performance of a PV module with SP, TCT, BL, SS (Simple series) and HC (honey comb) interconnection configuration is analyzed [18]. The five interconnection schemes are compared for their maximum power and fill factor and it is found that the TCT configuration shows a superior performance over the other four configurations. An electrical reconfiguration scheme [19] is proposed in which the reconfiguration strategy is based on a controllable matrix of switches configuring the PV modules in a single string of parallel connected rows connected to the load. Under partial shading conditions, the matrix control algorithm reconfigures the PV modules to optimize the current of the single string, i.e., the output power of the system. However, this method involves greater complexity and cost, since it requires a large number of switches and sensors for implementation. The effect of change in interconnections among the modules within a shaded PV array on its MPP and a clear relationship between the inter connections of the PV modules and their power output was proposed [20].

This paper presents a fixed configuration method to arrange the modules so as to enhance the PV power generation under partial shaded conditions. This structure reduces the occurrence of shading in the modules of same row thereby maximizing the output power from the array. A comparison is made between proposed scheme and different configurations in addition to EAR configuration using a 5X5 PV array. The system performance is investigated for different shading conditions on a 5x5 PV array with conventional interconnection schemes and the results reveal that the proposed structure exhibits superior performance under partially shaded conditions.

A. Conventional interconnection schemes
In a PV array, the interconnection scheme refers to the manner in which modules are interconnected in the array. The three widely used interconnection schemes reported in literature [21] are the Series-Parallel (SP), Total Cross Tied (TCT) and Bridge Linked (BL) configuration. The SP scheme consists of a series of panels in a string and a number of such strings are connected in parallel. The TCT configuration is obtained by connecting ties across each junction. The BL interconnection scheme is derived from the connections in a bridge rectifier. Fig.1 shows the three interconnection schemes in a 3x3 PV array.
A study on the available module interconnections and their impact on power production are reported in many literatures [19]. Comparison of different inter connections schemes based on shading conditions and shading level reveals the influence of the interconnection schemes on generated PV power. The TCT scheme is considered to be the better scheme to reduce the losses under most of the shading conditions, although none of the scheme is found to be effective under wide shading conditions. In this study, a fixed interconnection scheme is proposed to reduce the mismatch losses due to partial shading.

B. Methodology used in the proposed invention:
The physical locations of modules within the column in the array are fixed during installation according to the proposed algorithm. Consider an m x n PV array having n modules connected in parallel and such m parallel groups are connected in series, forming a matrix referred as initial configuration, Amn. Each panel is represented as Ay where T is row index and 'j' is the column index to which the panel belongs. In this case, i =1, 2....m and j = 1, 2....n. The array Bmn is formed, by relocating the modules of array A, as follows. Consider the square root of 'm' truncated to an integer as 'f. Let 'p' be the 'm mod f which decides the method of reconfiguration. In case the value of 'p' is not equal to zero, step I is carried out otherwise step II is to be performed. Step-I:

The modules in each column are shifted down by (j-1)f steps. That is Ay is relocated as Bkj in the new arrangement where k = (i+ (j-1)f) and the value of k is normalized by subtracting with m, if the value of k is more than m. Step-I I

First the columns are divided into different groups with each group having 'm/f columns. The modules in each column of first group are shifted down by (j-1)f steps. Similarly, the modules in the next group of columns are shifted one step more than previous group of columns and so on. That is the module Ay is relocated as Bkj where k = i+(j-1)f for the modules in columns of first group and k = i+(j-1)f+1 for the modules in the columns of second group and so on. The arrangement of the modules using the proposed scheme is explained for 3X3, 5X5 and 6X6 PV array configurations. 3X3 PV array:

Consider a 3 x 3 PV array as shown in fig.2 (a). Here m = n = 3 and square root of 'm' truncated to an integer (f) is equal to 1. The value of 'p' (m mod f) ) is equal to zero. Therefore, the modules in the first three columns (m/f = 3) are shifted uniformly. The modules in the first column remain same as k is equal to i. The modules in second column (Ai2) are relocated as Bk2 where k = i+1. Similarly, the modules in the third column (Ai3) are relocated as Bk3 where k=i+2 as shown in fig.2 (b).

5X5 PV array:
Consider a 5X5 array as shown in fig.3 (a). Here m = n = 5 and square root of m truncated to an integer (f) is 2. The value of p is equal to 1. Therefore step-l is performed. Any module Ay is relocated as Bkj where k = i+ (j-1)f. That is the position of modules in first column of B is same as A. The modules in the second column (Ai2) are relocated as B^ where k = i+2. Similarly for third, fourth and fifth columns the value of the k is i+4, i+6, i+8 respectively as shown in fig.3 (b). Table-I shows the number of steps to be shifted in each column for 5X5 PV array.

6X6 PV array:

Consider a 6X6 PV array as shown in fig.4 (a). Here m=n=6 and square root of m truncated to an integer (f) is 2. The value of p is equal to zero. Therefore step-l I is performed. All the six columns are divided into two groups of each having three columns (m/f). The modules in first three columns of Ay are reconfigured as Bkj where k=i+ (j-1)f and for modules in the next three columns the value of k=i+G-1)f+1 as shown in fig.4(b).

Table-ll shows the modules to be shifted in each column for 6X6 PV array.

The performance of the scheme is studied by varying the shading level under different shading conditions. The shading level refers to the irradiance received by a shaded module. The shading conditions are defined based on the number of shaded strings and shaded modules per string. The four shading conditions are short wide (SW), short narrow (SN), long wide (LW) and long narrow (LN) [20]. To evaluate and compare the performance of the different schemes under partial shaded conditions, a loss indicator is specified.

Loss indicator = p, - p2 (1)

where Pi is the sum of the maximum power generated (taken individually) by modules in the array in a given insolation and temperature condition and P2 is the total power generated by the array under the given insolation and temperature.

Table I Table II
Illustration for a 5 x 5 array Illustration for a 6 x 6 array

RESULTS AND DISCUSSIONS
The simulations are carried out using MATLAB/Simulink environment. A comparison has been carried out between the proposed scheme and Electrical Array Reconfiguration (EAR) for different shading patterns. In addition, all the four configurations namely SP, TCT, BL and proposed configuration are also compared for their losses under partial shading conditions. Here a 5X5 PV array is considered for study. The specifications of the cell at standard test conditions are shown in Table III. To validate the proposed configuration, diagonal shading and non-continuous shading are also considered in addition to short wide (SW), short narrow (SN), long wide (LW) and long narrow (LN) shading patterns. The shading conditions are defined based on the number of shaded columns (width) and shaded modules per column (length).

TABLE III

PV SPECIFICATIONS AT STC 1000 W/m2,25 °C

i) Short and Narrow shading conditions:
Here all the panels receive a uniform insolation of 1000 W/m2 except four panels at bottom right corner receive 600 W/m2. The shading patterns for both the schemes are described in fig.5 (a) and fig.5 (b). The reconnection of panels with EAR scheme is shown in fig.5(c). The current generated in row 1 with EAR algorithm to get optimized power with minimal reallocations is expressed as

IR1 = K11 I11 + K12 I12 + K13-I13 +K14I14 + K15I15 (2)

where kn= Gn/G0=0.6 where Gn is the solar irradiance of the panel numbered 11 and Go is the maximum solar irradiance and In is the current generated by the panel. Assuming the current generated by each panel under STC as lm,

IRi = 5xIm (3)

All the panels in row 2 receive 1000W/m2 except one panel which receives an insolation of 600W/m2. The current generated by the panel is expressed as

IR2-4xIm+lx0.6xIm (4)

Since all the rows 2, 3, 4, and 5 have similar shading pattern, therefore

lR2=lR3=lR4=lR5=4-6Im (5)

Neglecting the small variations in voltage across each row, the voltage of the array Va = 5 Vm, if none of the panels are bypassed and Va = 4 Vm + Vd, where Vd is the voltage across the diode if a single row is bypassed. As Vd « Va, Vd is neglected. The approximate power generated by the array if the shaded panels in the array are bypassed is given by

Pa=IaVa=(5Jm)(Vm) = 5ImVm (6)

The approximate power produced by the array if no panel in array is bypassed

Pa=IaVa=(4.6Im)(5Vm) = 23ImVm (7)

To verify the theoretical results, simulations are carried out in MATLAB/Simulink environment. It is observed that the maximum power generated by the array is 8.91 W and the global peak (GP) occurs at 13.71 V. To validate the proposed fixed arrangement of modules, the array is subjected to the same shading pattern (Fig.5 (b)) and its shade dispersion is shown in fig.5 (d). The current in each row is calculated as below

IR1 =IR2 =IR3 = IR4 =4xlm+0.6xlm

IR5=5Im (8)

The approximate power generated by the whole array if the shaded panels in the array are bypassed is given by

Pa=I.V.=(51m)(Vm) = 5ImVm (9)

The approximate power produced by the array if no panel in the array is bypassed is given by

Pa =IaVa =(4.61m)(5Vm) = 23ImVm (10)

It is observed that the proposed configuration is also generating the same maximum power as generated in Electrical Array Reconfiguration (EAR) i.e. 8.91 W at voltage of 13.71 V.

ii) Long and wide Shading:

In contrast to short and narrow shading, here all the panels are shaded with an insolation of 600 W/m2 except four panels at top left receive an insolation of 1000 W/m2 as shown in fig.6(a). The reconnection of panels with EAR algorithm is shown in fig.6(c). The current generated, in rows 1, 2, 3 and 4 is same since four out of five panels in each row are shaded with EAR algorithm, is

lRl=lR2 = lR3=lR4=4x0.6Im+Im (11)
The current generated in row 5 is given by
IR5 = 5x0.6xIm (12)

The approximate power generated by the array if no panel in the array is bypassed is given by
Pa=IaVa=(3.4Im)(5Vm) = 17ImVm (13)

The same shading pattern is applied to the array with proposed arrangement (fig.6 (b)) and its shade dispersion is shown in fig.6 (d). The approximate power generated in the array
Pa=IaVa=(3.4Im)(5Vm) = 17ImVm (14)

Therefore, the power produced in both cases is same and the maximum power generated by the array is 6.19 W at a voltage of 14.07V.

iii) Short and wide shading:
In this case, bottom two rows are subjected to shading with an insolation of 600W/m2 and the remaining all panels receive an insolation of 1000W/m2 as shown in fig.7 (a) and 7(b). The reconnection of panels with EAR algorithm is shown in fig. 7(c) and shade dispersion in proposed method is shown in fig.7 (d).

The current generated in each row with EAR algorithm in which two modules out of five modules are shaded is

IRl=lR2=lR3 = lR4 = lR5=2x()-6Im+3Im (15)

Therefore, the approximate power generated in the array is
Pa =IaVa = (4.2Im)(5Vm) = 21ImVm (16)

On the other hand, the power generated with proposed arrangement is

Pa =IaVa = (4.2Im)(5Vm) = 21ImVm (17)

Therefore, the approximate power produced in both cases is same and the maximum power generated by the array is 8.05 W at a voltage of 13.65V.

iv) Long and Narrow shading:

The six panels at right bottom are shaded with 600W/m2 and the remaining panels receive an isolation of 1000W/m2 as shown in fig.8 (a) and fig.8 (b). The reconnection of panels with EAR configuration is shown in fig. 8(c) and shade dispersion with proposed fixed configuration is shown in fig.8 (d). The current generated in each row with EAR algorithm is,

IR1 = !R2 = XR3 = !R4 = °-6Im + 4Im

and IR5 = 2x0.6im+3im (18)

Therefore the approximate power generated in the array is,
Pa =IaVa =(4.2Im)(5Vm) = 21ImVm (19)

On the other hand, approximate power generated in the array with proposed arrangement is equal to
P. =I8V8 =(4.2Im)(5Vm) = 21ImVm (20)

Therefore, the power produced in both cases is same and the maximum power generated by the array is 8.48 W at a voltage of 13.94V.

v) Diagonal Shading:
Here the panels in diagonal are shaded with an insolation of 600 W/m2 and the remaining all panels receive an insolation of 1000W/m2 in both EAR configuration and proposed arrangement as shown in fig 9(a) and 9(b) respectively. The reconnection of panels with EAR scheme is shown in fig. 9(c) and shade dispersion with proposed fixed configuration is shown in fig.9 (d). The current generated by each row with EAR scheme is

lRl=lR2 = lR3 = IR4 = lR5=0.6Im+4Im (21)

Therefore, the approximate power generated in the array is

Pa = IaVa = (4.6Im)(5 Vm) = 23ImVm (22)

On the other hand, approximate power generated in the array with proposed arrangement is equal to
Pa =IaVa =(4.6Im)(5Vm) = 23ImVm (23)

Therefore, the power produced in both cases is same and the maximum power generated by the array is 8.83 W at a voltage of 13.58V.

vi) Non continuous shading:

In this case, non continuous shading is considered for EAR scheme and proposed scheme as shown in fig.10 (a) and 10(b) respectively. The reconnection of panels with EAR scheme (fig. 10 (c)) and shade dispersion in proposed scheme (fig. 10 (d)) generates same current and power. The current generated in row with EAR algorithm is lRi = lR4 = lR5 = 2x0.6lm+3lm and

IR2=IR3 = 3x0.6Im+2Im (24)

Therefore, the approximate power generated in the array is
P. =I.VB =(3.8Im)(5Vm) = 19ImVm (25)

On the other hand, approximate power generated in the array with proposed arrangement is equal to
Pa =IaVa =(3.8Im)(5Vm) = 191mVm (26)

Therefore, the power produced in both cases is same and the maximum power generated by the array is 7.57 W at a voltage of 13.76V. It is found that the proposed scheme yields same power as produced with EAR scheme in most of the non continuous shading conditions. Table-IV shows the power generated in the PV array for different configurations against different shading patterns when the shaded module insolation is 600 W/m2 and unshaded module insolation is 1000 W/m2. It is observed that proposed scheme and EAR scheme generate maximum power under all shading conditions. However, in some occasional situations with non continuous shading pattern, the EAR scheme is likely to yield marginally more power output than the proposed scheme.

TABLE-IV
Power generated (mW) in a 5X5 PV array for different configurations against different shading patterns (Shaded panel insolation is 600W/m2 and unsahaded panel insolation is 1000W/m2) Sum of individual
Thus arranging the modules according to the proposed method produces the same power as produced with EAR configuration. Moreover, the proposed configuration does not require switching matrix i.e. large number of switches and sensors like EAR scheme. The same technique can be extended to larger size PV arrays to reduce the mismatch losses due to shading.

As per the invention, there is a system for rearranging panels of a series parallel (SP) array A arranged in m x n PV matrix having 'n' series panel modules in each row with 'm' parallel groups as columns. Each panel is represented as Ay wherein T is the row index and 'j' is the column index.

It includes a relocation arrangement means which arranges each of panels Ay based on a rearrangement rule for relocating the panels of array A on the basis of value of 'p' into another array B wherein each panel in array B is represented as Bkj wherein 'p' is (m mod f) and f is square root of 'm' truncated to an integer.

In another aspect the rearrangement rule is k = i + (j-1)f when the value of 'p' is not equal to zero and when value of 'k' is less than or equal to 'm'.

In another aspect k = [i +(j-1)f]-m when the value of 'p' is not equal to zero and when value of 'k' is greater than 'm'.

In another aspect when 'p' is equal zero then the array is divided into plurality of 'x' groups wherein each group having (m/f) columns wherein k= i +Q'-1)f for the first group, wherein k = [i +(j-1 )f] +1 for the second group and wherein k = [i +G-1)f]+(x-1) for the xth group.

The invention also discloses a method of rearranging panels of a series parallel (SP) array A arranged in m x n PV matrix having 'n' series panel modules in each row with 'm' parallel groups as columns. Each panel is represented as Ay wherein T is the row index and 'j' is the column index comprising of rearranging each of the panels Ay based on a rearrangement rule for relocating the panels of A on the basis of value of P into another array B wherein each parallel in array B is represented as Bkj wherein 'p' is (m mod f) and f is square root of 'm' truncated to an integer.

In another aspect k = i + (j-1)f when the value of 'p' is not equal to zero and when value of 'k' is less than or equal to 'm'.

In another aspect k = [i +(j-1)f]-m when the value of 'p' is not equal to zero and when value of 'k' is greater than 'm'.

In another aspect 'p' is equal zero then the array is divided into plurality of 'x' groups wherein each group having (m/f) columns wherein k= i +(j-1 )f for the first group , wherein k = [i +G-1)f] +1 for the second group and wherein k = [i +G-1)f]+(x-1) for the Xth group.

This invention has identified the impact of interconnection schemes on reducing the losses under different shading conditions. The proposed structure facilitates to distribute the effect of shading over the array thereby reducing the losses due to partial shading. The performance of the system was investigated for different shading conditions and simulation results show that the power extracted under partial shading conditions is more in the proposed configuration compared to other configurations. It is also found that the EAR scheme and proposed scheme yield same power generation under different shading patterns.
Moreover, implementation of proposed arrangement is very simple compared to the EAR scheme.

There is therefore provided in accordance with the invention an efficient and effective manner of providing a fixed configuration method to arrange the modules so as to enhance the PV power generation under partial shaded conditions, which specifically reduces the occurrence of shading in the modules of same row thereby maximizing the output power from the array.

One skilled in the art will understand that the embodiment of the present invention as shown in the drawings and described above is exemplary only and not intended to be limiting.

It will thus be seen that the objects of the invention have been fully and effectively accomplished few embodiments have been shown and described for the purposes of illustrating the functional and structural principles of the present invention and is subject to change without departure from such principles. Therefore this invention includes all modifications encompassed within the sport and scope of the claims.

WE CLAIM:

1. A system for rearranging panels of a series parallel (SP) array A arranged in m x n PV matrix having 'n' series panel modules in each row with 'm' parallel groups as columns wherein each panel is represented as Ay wherein T is the row index and 'j' is the column
index comprising of:

a relocation arrangement means which arranges each of panels Ay based on a rearrangement rule for relocating the panels of array A on the basis of value of 'p' into another array B wherein each panel in array B is represented as Bkj wherein 'p' is (m mod f) and f is square root of 'm' truncated to an integer.

2. The system as claimed in claim 1 wherein the rearrangement rule is k = i + (j-1)f when the value of 'p' is not equal to zero and when value of 'k' is less than or equal to 'm'.

3. The system as claimed in claim 1 wherein k = [i +G-1)f]-m when the value of 'p' is not equal to zero and when value of 'k' is greater than 'm'.

4. The system as claimed in claim 1 wherein 'p' is equal zero then the array is divided into plurality of 'x' groups wherein each group having (m/f) columns

wherein k= i +(j-1 )f for the first group wherein k = [i +0-1 )f] +1 for the second group wherein k = [i +(j-1)f]+(x-1) for the Xth group.

5. A method of rearranging panels of a series parallel (SP) array A arranged in m x n PV matrix having 'n' series panel modules in each row with 'm' parallel groups as columns wherein each panel is represented as Ay wherein T is the row index and 'j' is the column index comprising of rearranging each of the panels Ay based on a rearrangement rule for relocating the panels of A on the basis of value of 'p' into another array B wherein each parallel in array B is represented as Bkj wherein 'p' is (m mod f) and f is square root of 'm' truncated to an integer.

6. The method as claimed in 5 wherein k = i + (j-1)f when the value of 'p' is not equal to zero and when value of 'k' is less than or equal to 'm'.

7. The method as claimed in 5 wherein k = [i +(j-1)f]-m when the value of 'p' is not equal to zero and when value of 'k' is greater than 'm'.

8. The method as claimed in 5 wherein 'p' is equal zero then the array is divided into plurality of 'x' groups wherein each group having (m/f) columns

wherein k= i +(j-1 )f for the first group wherein k = [i +(j-1 )f] +1 for the second group wherein k = [i +(j-1)f]+(x-1) for the xth group