The applications for PV solar systems are increasing, which need to develop the ingredients and methods used to harness this power source. PV system efficiency, the intensity of source irradiation, and storage methods are the main aspects that disturb the effectiveness of the collection process. The efficiency of a PV is limited by materials used in photovoltaic manufacturing. It is practically hard to do significant enhancements in the cell's performance, hence controlling the global gathering procedure's effectiveness. Therefore, the most available method to improve the solar power system's performance is the rise of irradiation intensity
The total amount of sunlight is mostly existing, and it does have the potential to meet a considerable amount of power demand around the world. Over time the use of Solar panels is spreading particularly throughout too many of the rural areas but what remains as a back flaw is these panels' ability to harness that sunlight with proper efficiency and use it properly. Solar energy (SE) is the optimum solution that can provide griddles power and completely clean pollution and health hazards. Even after having too many solar energy features, it is still not as widely used as possible. The central aspect of concern remains in terms of its efficiency and another big question about its operation during uniform and nonuniform radiation conditions and conditions of partial shading
Due to Yemen's current situation, the power grid has stopped working, and no energy projects have been added to the system. Besides, the increasing demand for electrical power increase because of an increase in population. Moreover, fossil fuels always hard to find, and fuel shortages frequently happened. Therefore, Yemeni people depend on renewable resources as primary energy resources. However, there are no rivers in Yemen, and wind speed cannot produce efficient power in most areas. So, solar energy has been considered the dominant and optimal power source. Solar energy is not only used for homes but also primarily used for farm water pumps. Yemen has excellent solar resources potential, and the solar system is an area where there is much progress. The apparent disadvantage of solar systems is the general low efficiency. These facts make the study of introducing more efficient solar systems in Yemen, and providing background for power engineers is significant.
Yemen belongs to the global sunbelt with average sunshine 911 hour/day and the peak sun hour (PSH) reach 67 hours with annual direct average energy density 16002200 kWh/m^{2}, which makes the less cost of power than any other country with less PSH. This geographical position encourages Yemeni people to use solar energy applications. The generation of electricity from solar energy using PV has low efficiency. Nevertheless, to improve the PV utilization, the system should operate around the MPP by proper MPPT technique.
All of the matters mentioned above, the design of an MPPT solar charge controller is essential to increase the PV solar system's efficiency output. Besides, force the PV array to operate as much as possible at the peak level of power. These MPPT techniques have also been proposed to regulate charging rates depending on the battery's charge level to allow charging closer to the battery's maximum capacity and supervise battery temperature to prohibit overheating. Moreover, the MPPT methods are the best decision for colder conditions; PV array voltage can be more significant than battery voltage;
Cuckoo Search is an optimization method that is motivated by the natural parasitic reproduction strategy of cuckoo birds. This method is similar to the genetic algorithm (GA) and particle swarm optimization depending on the population algorithm; it also has some similarity to the harmony search in the selection procedure. The randomization is made much more efficient by Levy flight, which gives faster convergence. Also, the number of tuning parameters (two parameters) in this method is less than that in GA and PSO (three parameters and more). In addition, the characteristic of the Cuckoo search does not depend on giving the samples initial values
Recently, several studies on another soft computing method called the cuckoo Search Algorithm (CSA) have attracted significant attention
Furthermore, to prove its feasibility, we evaluated three mature MPPT methods, MPSO, MP&O, and ANN. Several performance tests were executed, i.e., its operation during uniform and nonuniform radiation conditions, partial shading conditions, and changing loads. Finally, the prospect of CSA as the MPPT algorithm in a photovoltaic system is discussed and suggested.
The P–V characteristics curve shows a nonlinear, timevarying maximum power point (MPP) issue because of the persistent variation in the weather condition. Firstly, the temperature and solar irradiance and secondly PSCs, which is the most important reason behind MPP's problems in the PV curve features. Therefore, many different MPPT techniques have been proposed to guarantee that the PV arrays are feeding the load demand with the always required maximum power point (MPP) and used in integration with the power converter (DCDC converter and inverter). These MPPT techniques are reported in the literature and classified into either conventional MPPT or soft computing methods. The existing techniques vary in simplicity, accuracy, time response, popularity, cost, and other technical aspects
In
To solve the problems mentioned above, artificial intelligence techniques or soft computing techniques have attracted much interest during previous periods, and the MPPT controller's combination enhances its performance to a great extent. Several MPPT controllers based on soft computing algorithms are reviewed in the literature to extract maximum solar power efficiently. Relatively, few papers use the artificial neural network (ANN) to track the solar system's global peak under different shadow conditions. Such as the Artificial Neural Network in
In
Moreover, a Genetic algorithm is applied to improve the result of FLC. In
Although this soft computing (SC) technologies are flexible, they are usually more complex and slower than traditional methods. For example, artificial neural networks require precise and extended training cycles to produce accurate results. Also, artificial neural networks need to be implemented by expensive microprocessors because of its computationally intensive nature. On the other hand, FLC shows excellent convergence speed, but its performance depends on the programmer's experience and understanding of specific PV modules and system installation's environmental conditions. Other algorithms such as GA and ACO are also used, but they are mainly used as the optimizer of traditional MPPT; this method is often called hybrid MPPT. Also, in recent years, evolutionary algorithms such as multiverse optimization (MVO)
The primary function of photovoltaic systems is to generate energy straight from the sunshine falling on it. Therefore, the PV system is considered a static electricity generator. The central part of the PV system behind generating the electricity from sunlight is the PV cells. The PV cells are probably composed together to form panels or arrays. There are a voltage and current in the terminals of photovoltaic panels that may directly feed small loads like lighting systems and DC motors. Photovoltaic systems come in a range of sizes and outputs appropriate for various applications. They are lightweight, allowing for easy and safe transportation. These converters may adjust the voltage and current at the load, control the power flow in gridconnected systems, and fundamentally track the GMPP of the photovoltaic systems
PV solar cells are the maincompartment of solar power generation. The PN junction causes to generate solar photovoltaic effect and convert solar energy into electricity. The output power amount of just one single photovoltaic cell is small. The desired power can be achieved in actual applications once forming the photovoltaic array either by string or parallel connection. The output attributes of photovoltaic arrays have characteristics such as nonlinearity and timevariation, which are easily affected by the incident light irradiation intensity, battery body temperature, load conditions, and parasitic impedance, resulting in that the solar energy utilization rate cannot always be maximized. When the solar array is not uniformly illuminated, such as shadows of surrounding buildings, leaves, clouds, etc., the shielded portions cannot be irradiated with regular sunlight, resulting in a smaller current than other nonshaded photovoltaic cells. The current, which becomes a load in the circuit, consumes power, causing the overall output energy to decrease, even causing damage to the photovoltaic cells. At this point, the solar array's output features are more complex, and there may be more than one peak point. Therefore, to better control the photovoltaic power generation system, it is excellent to study the photovoltaic array's output features under the condition of partial shading. This section will set up a mathematical prototype for photovoltaic arrays under the condition of partial shading, combining actual shading conditions with existing photovoltaic cell models.
Partial shading is considered one of the main reasons which affect PV curve characteristics. Once the PV system is exposed to abnormal environmental circumstances, particularly the partially shaded case, the nonlinear characteristics of PV curves will get more complicated with several peak maximum power points. Partial shading conditions occur due to many reasons, such as an object covering some parts of the PV arrays for some time, dirt causing a portion of the PV arrays, the clouds covering the sun lights, etc.
The effects of partial shading lead to an increase in the problems in the area of MPPT. When a PV system composes a large number of PV arrays, it is tough to guarantee that each panel receives a similar radiation level. Some of these panels will be exposed to some hard factors such as dust, clouds, trees, etc. will, in practice, be subjected to different amounts of insolation
The required voltage and current rating can be satisfied by constructing the PV array using various series and parallel modules. The PV module's protection from the hotspot problem's adverse effects can be satisfied by connecting each module with a bypass diode, as shown in
In order to mathematically establish modeling of the PV array under partial shading, the photovoltaic cell is referred to as (the type of cell construction) a monomer in this paper, as shown in
The shading can be caused by different factors such as surrounding buildings, leaves, clouds, etc. However, the PV cell dimension in the commonly used photovoltaic array is 125mm*125mm. Therefore, the probability of three different intensity lights appearing when a single cell is blocked is tiny. In this paper, there is a maximum of two kinds of light intensity on a single photovoltaic cell, and based on this modeling, most shading conditions can be included.
Once some of the monomers in the PV array are blocked, resulting in nonuniform illumination, the shaded cells act as a load and generate some heat, which may cause a hotspot effect. This reverse leakage occurs when the two ends are subjected to negative pressure to a certain extent, breaking down and damaging the cell. Bypass diodes are used to avoid the hot spot effect and increase the output power of the PV array under partial shading conditions. When the PV array operates under uniform irradiation conditions, all PV cells usually work, and the bypass diode will be in the reverse cutoff state. When a PV cell is shaded, its photon current is reduced, and the voltage drop is negative. At this time, the bypass diode conducts to be like a shortcircuit to prevent it from being reversed by the reverse leakage current and increasing the overall output power. The connection of blocking diodes ensures each substring direction's direction and prevents each substring from affecting each other due to different output voltages. The structure of the PV array with the connection of bypass diodes and blocking diodes is shown in
In this case, Normal equation cannot be implemented entirely, and the mathematical model of the PV array needs to be modified and reestablished. In this paper, the rate of the optical shading E is introduced to indicate the degree of light shading and the area of the PV array as follows:
Where
where
As shown in
When
When
The mathematical model corresponding to this case can be expressed as the following segment function:
The output characteristics of the equivalent model of the PV circuit, which is established by the above Equation, are shown in
The output current of each PV module, which consists of Ns cells under partially shaded condition, is
When
Where
The block diagram in
The duty cycle D in (6) varies between 0 and 1. That means the voltage ratio is greater than 1, which causes the converter to be used as a stepup converter. The relationship between voltage ratio and duty cycle D is shown in
A PV array has a nonlinear feature, and its output power relies primarily on the radiation level and the operating temperature. Furthermore, a PV panel's output power is a function of its terminal voltage under the same temperature and irradiance. For all specific PV panels, there is only one value for the terminal voltage that corresponds to the maximum output power. Moreover, to get that voltage, there is a procedure known as maximum power point tracking MPPT. MPPT methods of a PV array could be obtained in two steps, either a single step or a double step. In the case of a single step, a DC/AC converter is utilized. While in the case of doublestep, a DC/DC and DC/AC converters are utilized. According to Thevenin's theory, when the value of the power supply’s output impedance equals the load, the output will reach the maximum value. Therefore, it is necessary to add an impedance transformation circuit in the MPPT control system, making the power supply and load reach the state of impedance matching and adjust the output power
The tracking and finding the new altered maximum power point (MPP) in its matching PV solar system curve under any changing radiation level and ambient temperature is the primary duty of maximum power point tracking methods (MPPT). Moreover, it is utilized to catching and given the maximum power from the PV panels and moving that power to the load. A DC/DC (boost/buck) converter performs as linking the PV modules and the load. The MPPT is altering the duty cycle to preserve the transfer power from the PV array to the load at the maximum power point
Currently, ANN has been enormously advanced in our scientist life, especially in the practical disciplines or the theoretical disciplines, to give us solutions for many complicated tasks. ANN contains three layers: the input layer and the output layer, and only these two layers can connect to the perimeter outside the network. In addition to these two layers, the network contains at least one layer called Hidden Layer because it is not connected to the network's outer perimeter and is only related to the layer that precedes it
In this paper, the training data which have been collected for training ANN are PV system voltage Vpv, PV system current Ipv, and duty cycle D as seen in
The proposed modification can eliminate the problems of the traditional algorithm. The modified algorithm proposed that the search space is limited to 10% of the power curve, which shortens the response time and reduces the steadystate oscillation,
Firstly, the voltage V1 and V2 are measured to discover the MPP area. The solar panel's working point is limited to 10% area 2 of the power curve, and then Perturbation and observation are started. MPP is realized and maintained in less perturbation. In uniform weather conditions, it will sustain at the maximum power point. The change of irradiance will find a new local maximum in the same way as the constant irradiance and remain unchanged
Regions 
Starting (% of Voc) 
Ending (% of Voc) 
Total area (% of Voc) 
Region 1 
0 
70 
70 
Region 2 
70 
80 
10 
Region 3 
80 
100 
20 
Particle swarm optimization (PSO) is a search method based on an improved population randomization algorithm. It can reserve the population of each particle to represent the candidate solution. The position expresses the ideal solution to the problem in the optimal particle space. The particle's velocity vector determines the direction and velocity value of the particle. Each particle obeys the current optimal particle and searches for the solution region's optimal solution according to its own flight experience
In the typical particle swarm optimization algorithm, the "particle" part refers to the population members with small mass and small volume affected by speed or acceleration and with the best performance. Each particle in the swarm represents a solution in a highdimensional space. The four vectors are: 1) current position; 2) unique optimal position of all particles after accelerating relative to the old position; 3) particle velocity; and 4) global optimal position of all particles originated from its neighborhood so far. Each particle adjusts its position
The first modification is the velocity step function
The second modification is updating the
We can consider narrowing the search area to find that the optimal value is another modification to the search process. It needs to adjust and modify the restrictions of control variables in each iteration. Therefore, for
The last modification is that the speed step or step size is based on the maximum/minimum value x of the controlled variables
This section shows how optimization deals with the difficulties of photovoltaic systems. The complex control variables, objective functions and system constraints of the photovoltaic system are defined. The above optimization techniques are now implemented on the MPPT controller of the PV system under consideration, which operates under the PSC.
In recent years, many evolutionary algorithms based on natural metaheuristics algorithms have been improved in optimization. These algorithms usually work in suitable search areas based on a random search, which depend on or optimize complex problems. However, there will also be many mechanisms to lead the search so that the solution vector will progress with the number of iterations, so the search is not random. The intensive (Development) and diversified (exploration) are the two essential features of these recent metaheuristic algorithms.
In 2009, Yang and Deb were proposed a natural optimization algorithm CSA. This is a new metaheuristic algorithm, which has already been applied in many optimization types of the research area. Also, CSA's inspiration comes from cuckoo chicks' parasitic behavior, which has a robust reproductive ability when laying eggs in the host nest. Therefore, the determination of egglaying and body shape breeding of cuckoo is based on the optimization algorithm. There are two formulations of cuckoo in the optimization process: egg and mature cuckoo. If adult cuckoos lay eggs in their nests and the host bird does not find or kill the eggs, they will grow into mature cuckoos. The cuckoo community's environmental characteristics and migration help the cuckoo gather and find the best habitat for breeding. This optimal residential area is the global maximum of the objective function
Cuckoos are fascinating because they make beautiful sounds, and because of their breeding strategies. Some of the cuckoo’s species, such as Killa and Anne, are laying their eggs in community nests and eliminating other birds' eggs to progress their chances of hatching their eggs. Some cuckoos lay eggs and parasitize only in host bird’s nests. There are three basic types of parasitism: intraspecific parasitism, cooperative propagation, and nest attachment. Several host birds clashed directly with the parasitic cuckoo. If the host bird gets the cuckoo's eggs, immediately, it will throw them away or leave its nest and build a new one in another place. Some cuckoo species have evolved so that the female parasite mimics the color and shape of the eggs of a few selected host species. This increases the possibility of eggs hatching, thus improving their reproductive capacity.
The most critical part of the cuckoo breeding strategy is to find a convenient host nest. In general, looking for a nest is similar to looking for food, and it takes the form of random or quasirandom. Generally speaking, when animals are looking for food, they will choose the direction or track simulated by a specific mathematical function. One of the most common modes is Levy flight. Levy flight can be regarded as a random walk, in which step size has a levy probability distribution. In cuckoo bird search, birds nest search is characterized by Levy flight. Mathematically speaking, Levy flight is a kind of random walk. According to power law, the step size is extracted from Levy distribution as follows:
Where ɭ is the flight length, and λ is the variance. Because 1 < λ < 3 so y has infinite variance.
The algorithm starts with a premier population of cuckoos. The original cuckoo laid its eggs in the host bird’s nest. Therefore, some of the eggs that have more similarity to the host birds' eggs will grow and hatch into mature cuckoos. Other eggs found by the host birds were destroyed. Mature eggs can detect the quality of nests in that area. The more eggs a region has, the more profits it makes. Therefore, the term CSA is going to optimize will be where more eggs survive
To improve the rate of still alive for cuckoo eggs, cuckoo should look for the rightest area to lay eggs. When the remainder of the eggs grows into mature cuckoos, they will make some societies. Every society has its environment area to live and reproduce. The best environment for all societies will be the aim of other societies. Then cuckoos migrate to their best environments. They will live close to the best habitat
Affording to the nest parasitism actions of cuckoo birds, CSA has three idealized rules: 1) Each cuckoo lay one egg at a time and then put it inside a randomly selected nest; 2) the nest with the best quality of getting chance of surviving of the cuckoo will be moved to the following generation; 3) the number of nests available is fixed, and the probability of keeping cuckoo eggs distinguished by host birds is Pa, of which 0 < Pa < 1.
For the maximization problem, the value of the objective function can be directly proportional to the solution's fitness. For simplicity, we can use the symbol: all eggs in a nest characterize a solution, and cuckoo eggs characterize a new solution. The aim is to use new and potentially better solutions (such as cuckoo) instead of less reasonable solutions in the nest. If a cuckoo's egg is detected, the host bird can leave its nest or destroy the cuckoo's egg. On the other hand, if the numeral of nests is fixed, the new nests will be built by Pa's probability.
When a new solution is generated for cuckoo, Levy flight will be performed according to the next expression
Where
where
In order to use CSA to design MPPT, it is necessary to select appropriate variables to search. First, in this case, all samples are defined as the value of photovoltaic voltage, i.e., VI (I = 1,2). n). The total number of samples is defined as (n). Secondly, the step size is expressed in (α). Then the fitness function (J) is the photovoltaic power value of the maximum PowerPoint. Because J depends on PV voltage, J= f(V). Initially, the resulting sample is applied to the PV module, and the power is set to the initial adaptation value. The voltage corresponding to the maximum power count is considered to be the largest sample at present. Then, according to the Levy flight executed in the next calculation
where
here β = 1.5 is the levy times the coefficient (at the designer's option), while u and v are calculated from the usual distribution curve. Measure the specific power of the new voltage sample from the PV module. The maximum power given by the voltage is selected as the new best sample by comparing the power values. In addition to the best sample, other samples are destroyed at random with a probability of Pa. This process stimulates the host bird's behavior of discovering cuckoo eggs and destroying them. Then a new random sample is generated to replace the broken sample. Therefore, all samples' power is measured again, and the current optimal value is selected by calculating J. The iteration continues until all samples reach MPP.
In general, under partial shading conditions, when using the method of Cuckoo search to obtain the global peak of maximum power of PV arrays, the search process has to be made by choosing suitable variables. The output voltage and step size are the two parameters of the Cuckoo search algorithm. If the new sample is more than the old sample, then the new sample's maximum power is selected as the new best sample. If the new sample has a lesser amount than the old sample, the maximum power is kept. The course continues until all samples have reached the MPP
The previous sections show that the global maximum power point's location depends on two factors; the first factor is how shadow shading is distributed on PV arrays, and the second one is how PV panels expose solar irradiation. As a result, the implementation of traditional maximum power point tracking algorithms to catch the global maximum power point's exact location is exceptionally tricky. Therefore, it is necessary to propose a new method with fast and accurate global search capability and be superior to the traditional MPPT algorithm in dynamic stability performance and other aspects.
This section will show all the simulations for the Photovoltaic array, the boost converter, the proposed artificial cuckoo search algorithm comparing to MPSO, MP&O, and ANN methods. All the simulations are done using MATLAB/SIMULINK software. The simulation results and some case studies for the whole proposed system are discussed in this section.
The schematic diagram of the proposed system, which shows the overall PV system connected to the proposed MPPT predictor, is clarified in
For PV array simulation, the simulated PV module has the same parameters as the UPS250 module at standard test condition (STC), as shown in
Parameters Values Maximum Power (P_{m}) 250 W Open – circuit Voltage (V_{oc}) 36.0 V Short – circuit Current (I_{sc}) 8.56 A Maximum Power Voltage (V_{mmp}) 30.2 V Maximum Power Current (I_{mmp}) 8.28 A Number of Series Modules (N_{s}) 4 Number of cells per module N_{cell } 60
The proposed controlling method's effectiveness is investigated by assuming three shading scenarios, and the PV characteristics curves under partially shaded conditions for all scenarios are presented below.
In the first scenario of investigating PSC, the PV system is exposed to three different irradiation levels and PSC (900, 1000, 400, 1000 W/m^{2}) and one constant level of temperature (25℃). As presented in
In the second scenario of investigating PSCs, solar radiation is considered (900, 1000, 1000, 900 W/m^{2}) and one constant temperature (25℃). The PV curve is shown in
In scenario three, the PV system is exposed to three different irradiation levels (900, 500, 1000, 500 W/m^{2}) and one constant level of temperature (25C^{o}). As presented in
In the fourth scenario of investigating PSC, solar radiation is considered (900, 650, 800, 700 W/m^{2}) and one constant temperature (25C^{o}). The PV curve is shown in
Finally, it can be construed that CSA and MPSO algorithms can track the GMPP while the conventional MP&O and ANN always drop to the local peak and the performance of the CSA method is superior to MPSO, MP&O and ANN methods where it converges in a shorter time compared to other designed MPPT controllers. For further study, as it was shown in
As mentioned in
Cases 
Radiations (G) 



PV1 
PV2 
PV3 
PV4 

Case 1 
900 
1000 
400 
1000 
702.136 
91.257 
Case 2 
900 
1000 
1000 
900 
930.93 
121.518 
Case 3 
900 
500 
1000 
500 
542.606 
127.664 
Case 4 
900 
650 
800 
700 
701.811 
125.396 
Scenario Pmax (W) PmaxCSA (W) PmaxMPSO (W) PmaxMP&O (W) PmaxANN (W) Scenario 1 702.136 699.6 699 447.2 416.6 Scenario 2 930.93 928.5 927.1 764.5 717.5 Scenario 3 542.606 534.7 536.2 534.6 472.1 Scenario 4 701.811 694.7 673.2 691.2 666.6
Scenario Vmax (V) VmaxCSA (V) VmaxMPSO (V) VmaxMP&O (V) VmaxANN (V) Scenario 1 91.257 264.5 264.4 211.5 204.1 Scenario 2 121.518 304.7 304.5 276.5 267.9 Scenario 3 127.664 231.2 231.6 231.2 217.3 Scenario 4 125.396 263.1 259.5 263 258.2
Scenario DCSA DMPSO DMP&O DANN Scenario 1 0.6572 0.6576 0.5198 0.3713 Scenario 2 0.5999 0.5963 0.5198 0.3795 Scenario 3 0.4559 0.4277 0.4523 0.2724 Scenario 4 0.5261 0.4834 0.5198 0.4142
The proposed MPPT method（CSA） have the ability to track the actual maximum power, whatever are the shading conditions and scenarios. For all considered shading scenarios, it has been noticed that the proposed method can easily track the actual maximum power for the UPS250 PV array system. As well, under nonuniform irradiation condition and PSCs for all scenarios, it can be showed that the proposed technique shows fast response and good stabilization at the actual maximum Power Point, and the global MPPT can be tracked rapidly with almost 99% efficiency for all investigated scenarios compared to MPSO, MP&O and ANN methods.
In this section, a descriptive statistical analysis is introduced to evaluate, organize, and summarize the proposed algorithms' results. Moreover, sensitivity analysis is introduced to test the performance stability of these algorithms. The worthy statistical metrics for this evaluation are; the Error, Efficiency, Tracking Time, and Wasted Power. These metrics can be estimated as the following:
Scenarios Wasted Power (W) Tracking Time (s) Efficiency (η%) Error (%) Scenario 1 2.536 0.6 99.64 0.3625 Scenario 2 2.43 0.7 99.74 0.2617 Scenario 3 7.906 0.6 98.54 1.4789 Scenario 4 7.111 0.5 99 1.0236
Scenarios 
Wasted Power (W) 
Tracking Time (s) 
Efficiency (η%) 
Error (%) 
Scenario 1 
3.136 
2.7 
99.55 
0.4486 
Scenario 2 
3.83 
2.9 
99.589 
0.4131 
Scenario 3 
6.406 
2.4 
98.82 
1.1947 
Scenario 4 
28.611 
3 
96 
4.25 
Scenarios 
Wasted Power (W) 
Tracking Time (s) 
Efficiency (η%) 
Error (%) 
Scenario 1 
254.94 
0.3 
63.7 
0.4486 
Scenario 2 
166.43 
0.3 
82.122 
0.4131 
Scenario 3 
8.006 
0.5 
98.524 
1.1947 
Scenario 4 
10.611 
0.4 
98.488 
4.25 
Scenarios 
Wasted Power (W) 
Tracking Time (s) 
Efficiency (η%) 
Error (%) 
Scenario 1 
285.536 
0.2 
59.333 
68.54 
Scenario 2 
213.43 
0.6 
77.073 
29.746 
Scenario 3 
70.506 
0.7 
87.006 
14.934 
Scenario 4 
35.211 
0.2 
95 
5.282 
Tracker 
Wasted Power(W) 
Tracking Time (s) 
Efficiency (η%) 
Error (%) 
CSA MPSO 
4.312 14.562 
0.56 1.5 
99.457 98.065 
0.576 1.973 
MP&O ANN 
115.562 126.945 
0.56 0.9 
84.648 83.136 
18.136 20.285 
Evaluated Parameter [6975] [76] [77, 78] [79] [80] [6] InC Proposed CSA GMPP tracking capability Yes Yes Yes Yes Yes Yes No Yes Simplicity Medium Medium Medium Simple Simple Simple Simple Simple Efficiency High High High High High High Low Very High Tracking speed Medium Medium Medium Medium High High High Very High Steadystate oscillation No No No No No No Yes No Initial location dependency Yes No Yes No No No Yes No Reliability Medium Medium Medium Medium High High Low Very High
Based on the mathematical model, both photovoltaic cells and the photovoltaic array under partial shading conditions are established and verified using the Matlab/Simulink environmental platform. The model can adjust different parameters such as solar irradiation, cell temperature, and the shading area, which can reflect the photovoltaic modules' output characteristics under different irradiation, temperature, and shading conditions. In the simulation, the results present that the PV array characteristics have only one peak under uniform irradiation conditions, which can be tracked easily. When the PV arrays are exposed to different solar irradiations, the output characteristics will have multiple peaks. The external characteristics equation of PV arrays needs to be described by piecewise function, making the maximum power point tracking more difficult. After that, the boost converter's principle is introduced. Its reliability is proved through the simulation experiments, which lays a basis for the following photovoltaic system simulation model establishment and maximum power point tracking experiment.
In this study, the maximum power point tracking problem is discussed, and based on the comprehensive analysis of artificial intelligence and heuristic algorithms, a control strategy of maximum power point tracking of PV system under partial shading and nonuniform irradiation conditions based on all main algorithms are proposed and discussed in more details. Firstly, by analyzing the PV solar generation system's characteristics, the control objective of maximum PV energy tracking is realized. Then, there is a collection of training data for the ANN controller (voltage, current, duty cycle) to train the MPPT tracker ANN under partial shading and nonuniform irradiation conditions. Finally, the model is trained to present the efficient output parameter for ANN and other methods, which then is used to be associated with MPPT controller which its output is used to control the DCDC converter to give the reference voltage which can achieve the maximum power point tracking under partial shading and nonuniform irradiation conditions. In this paper, the proposed MPPT method (CSA) system has been modeled and designed to track the maximum power point for the PV system under partial shading conditions. To get the validity of the proposed method performance, the simulation was conducted under different partial shading scenarios. The proposed MPPT control method can track the maximum power successfully for any partial shading condition and scenario. Simulation results demonstrate that the Cuckoo Search Algorithm (CSA) can extract the actual maximum power point rapidly with high efficiency and negligible oscillations around the global point of maximum power under partial shading conditions for various scenarios.
This study was supported by the Hubei Provincial Natural Science Foundation of China (2015CFA010), the Technology Project of State Grid Company “Soft Connection Mechanism and Modeling of Smart Grid Adapting to the Development of Global Energy Interconnection,” and the 111 Projects (B17040).