Renewable energy sources like solar, wind, and hydro are becoming increasingly popular due to the fewer negative impacts they have on the environment. Because, Since the production of renewable energy sources is still in the process of being created, photovoltaic (PV) systems are commonly utilized for installation situations that are acceptable, clean, and simple. This study presents an adaptive artificial intelligence approach that can be used for maximum power point tracking (MPPT) in solar systems with the help of an embedded controller. The adaptive method incorporates both the Whale Optimization Algorithm (WOA) and the Artificial Neural Network (ANN). The WOA was implemented to enhance the process of the ANN model's training, and the ANN model was developed using the WOA. In addition to this, the inverter circuit is connected to the smart grid system, and the strengthening of the smart grid is achieved through the implementation of the CMCMAC protocol. This protocol prevents interference between customers and the organizations that provide their utilities. Using a protocol known as Cross-Layer Multi-Channel MAC (CMCMAC), the effect of interference is removed using the way that was suggested. Also, with the utilization of the ZIGBEE communication technology, bidirectional communication is made possible. The strategy that was suggested has been put into practice, and the results have shown that the PV system produces an output power of 73.32 KW and an efficiency of 98.72%. In addition to this, a built-in regulator is utilized to validate the proposed model. In this paper, the results of various experiments are analyzed, and a comparison is made between the suggested WOA with the ANN controller approach and others, such as the Particle Swarm Optimization (PSO) based MPPT and the Cuckoo Search (CS) based MPPT. By examining the comparison findings, it was determined that the adaptive AI-based embedded controller was superior to the other alternatives.

Developing renewable power resources has involved research considerations, energy catastrophes, and environmental concerns like pollution and global warming. The Combined Heat and Power (CHP) application, solar photo voltaic (PV) module, miniature wind turbines, and heat and electricity storing are the growth of renewable energy sources where convenient loads are assumed to play an important role in prospect power supply. Microgrids are systems that possess a minimum of solitary distributed power source and connected loads and can lead to the formation of intentional islands in the electrical distribution system [

For the PV module to distribute its greatest energy, some Maximum Power Points Tracking (MPPT) methods are developed for standalone and smart grid-based PV systems. In the former type, the MPPT algorithm is designed on DC/DC converter, and an energy-saving bank becomes essential to stock up vast amounts of energy [

Meanwhile, control logic always observes the terminal voltage and current and modernizes the control signal. If a PV array is partially dappled by creating a cloud or tree, it is complicated for the conventional MPTT method to haul out maximum energy. The different modules with varying Current are produced, jagged insolation with different optimal series-parallel occurs, and various peak power points recurrently occurs in power

The PV system is improved by optimizing the desired criterion, which sets an apt voltage control parameter and references Current. The significant improvement of the proposed method is its compatibility with some PV generators. It does not necessitate data related to the PV generator and is uncomplicated to execute on a digital controller. It is feasible to amalgamate the proposed methodology with commercially available inverters. Several control strategies are suggested for MPPT of the PV method, namely feedback linearization control, Variable Structure System (VSS) control, Sliding Mode Controller (SMC), Artificial Neural Network (ANN), Fuzzy Logic Controller (FLC), Neuro-fuzzy controller method, etc. They increase the performance of PV systems [

In this paper, we proposed a novel approach for maximum power point tracking (MPPT) in solar photovoltaic (PV) systems that can handle uncertainties in the system. We acknowledge that uncertainties, such as changes in irradiance and temperature, can affect the performance of the MPPT algorithm, leading to suboptimal power extraction from the PV system. To address this issue, we have incorporated a centralized multi-channel multi-carrier medium access control (CMCMAC) protocol into the MPPT algorithm. This protocol allows the system to detect and handle uncertainties by using multiple channels to transmit and receive data from the PV system, as well as employing error correction techniques to reduce transmission errors. Additionally, the authors employ a smart grid enhancement technique that enables the PV system to communicate with the power grid, facilitating better coordination between the two systems.

We have cited several studies in the literature that also address uncertainty in MPPT algorithms. These studies include techniques such as fuzzy logic, neural networks, and Kalman filters, which can handle uncertainties by adapting to changes in the PV system's conditions. They highlight that the use of advanced control techniques, such as model predictive control, can help to address uncertainties in the system and improve its performance. The authors argue that their approach, which combines the CMCMAC protocol with smart grid enhancement, is a unique and effective way to address uncertainties in MPPT algorithms and improve the performance of solar PV systems. The authors also draw on existing literature to support their approach and suggest that it could be beneficial in improving the efficiency and reliability of solar PV systems in the future.

The motivation of the proposed method is to control the optimal control parameters based on the PV system’s maximum power point controller. The procedure of the proposed method has been divided into pre-processing of the input parameters and objective-based optimization. In this paper, the maximal power production of the PV method was performed using the MPPT control technique with embedded controller loop along with adaptive WOA and ANN controller optimization is done and innovative grid enhancement is analyzed using the CMCMAC protocol. The significant contributions of the proposed work are as follows,

To analyze the maximum power point tracking for the photovoltaic system and evaluate their characteristics.

To reduce oscillations in unsteady weather circumstances and reduce the interferences.

The objective function is to establish the reduced oscillations in unsteady weather circumstances it is used to maintain the exchanging pulse of the DC-DC converter.

To analyze the statistical analysis like dynamic performance, tracking speed, steady state oscillations, tracking in partial shading, execution time, and practical implementation.

Sera et al. [

Chine et al. [

Al-Majidi et al. [

Vikram et al. [

There are several approaches for tracking in Solar PV Systems with Smart Grid Enhancement, including:

Fixed Tilt Angle: This approach involves mounting solar panels at a fixed angle, typically based on the latitude of the installation site. This method is simple and cost-effective but does not maximize energy output.

Single-Axis Tracking: This approach involves mounting solar panels on a tracker that moves in one direction, typically east to west, to track the sun’s movement throughout the day. Single-axis tracking can increase energy output by up to 25% compared to fixed-tilt systems.

Dual-Axis Tracking: This approach involves mounting solar panels on a tracker that moves in two directions, both east to west and up and down to track the sun’s movement throughout the day and through the seasons. Dual-axis tracking can increase energy output by up to 40% compared to fixed-tilt systems.

Maximum Power Point Tracking (MPPT): This approach involves using electronics to optimize the power output of each solar panel by constantly adjusting the load and voltage to find the maximum power point. This method can increase energy output by up to 30% compared to systems without MPPT.

Smart Grid Integration: This approach involves integrating the solar PV system with the smart grid, allowing for real-time monitoring and control of energy production and consumption. This can help to optimize energy use and reduce costs by allowing for load management, peak shaving, and demand response.

In contrast, modern MPPT control techniques are utilized to avoid such demerits and are straightforward to implement in any controller in hardware and software. This proposed work uses the WOA to adjust the ANN controller connection weights. In this way, more power can be harvested from solar PV systems. Section 3 depicts the problems that exist in existing methods. The proposed WOA-based ANN technique has been discussed in Section 4. The implementation result analysis is illustrated in Section 5, with performance analysis and concluded the research in Section 6.

The PV system has been feasible in electricity areas, and the analysis of PV systems has developed much research in designing controller-based new control schemes. The maximum power production of PV systems is a significant research theme. The PV array possesses nonlinear characteristics and is influenced by the thermal temperature variation and irradiance effects. Consequently, MPPT is significant to process the abovementioned concerns and is confirmed to operate the PV systems at MMP. However, solar energy systems experience low efficiency and a cost hike. To solve the abovementioned concerns, maximum energy must be obtained from the system employing physical tracking or MPPT. Physical tracking aligns the PV panels with facing the sun's radiation in day duration to receive the highest solar radiation from the sun. A P&O method positions the MPP through the PV characteristics slope curve, which is employed owing to the modesty and effortlessness in the process implementation.

The operation of the PV panel indicates the oscillation in the MPP environment, and they cause a loss in energy which is a demerit of the P&O technique. To overcome the problems present in P&O algorithms, In Cond algorithms are developed. However, the In Cond MPPT algorithm requests instantaneous determining of an incline of the PV panel power curve and is difficult to execute the controller related to the P&O approach. The sensors to measure the thermal drift and irradiance are costlier in contrast to those that measure the current and voltage, which are used in other MPPT methods, which is a keen drawback. Moreover, this method relies on the structural arrangement of the PV array by incorporating the PV modules. Based on the performed literature review, little work has been concentrated on this MPPT controller-based PV system. Hence, the darker sides of the reviewed methods and the scope of growing a simple structure of WOA-based ANN controller for PV systems have motivated me to pursue this research work [

The area where maximum losses happened in solar PV array-based power generation systems is majorly addressed through existing paper, Moreover, this work is aimed to provide the solution for the complications that arise out of the following,

Inefficiency in identifying Maximum power incidents on the PV panel

Continuous monitoring and tracking of the MPP incident on the solar panel

Lack of developing an adaptive mechanism to act according to the surrounding atmospheric condition without interferences.

This work has opted to provide a better solution to the problems mentioned above and it has been opted as the problem identified in the existing methodology.

In this section, an adaptive procedure has been used to keep the maximum power generation for the PV system. An adaptive technique is the combination of the WOA algorithm and ANN approach, and it is used to increase the system accuracy and decrease the failure rate of an embedded system. The proposed WOA algorithm relies on the optimization of the ANN method, which produces the maximal power in the PV system via the embedded controllers. The WOA technique is described in the following section.

WOA is a metaheuristic technique inspired by the social characteristics of humpback whales. The examination procedure is signified in this algorithm by searching the flood randomly along with the virtual location of whales. It can be scientifically converted by updating an existing solution as arbitrarily choosing supplementary resolution rather than selecting the preeminent ones.

The Whale Optimization Algorithm (WOA) is a nature-inspired optimization technique based on the hunting behavior of whales. It is a population-based algorithm that uses a set of candidate solutions to iteratively search for the optimal solution. The rationale behind using WOA in MPPT tracking in Solar PV System with Smart Grid Enhancement using CMCMAC protocol is that WOA has several advantages over other mathematical-based approaches for optimization.

Firstly, WOA is a heuristic algorithm that does not require knowledge of the mathematical model of the system. This is particularly useful in MPPT tracking in Solar PV System with Smart Grid Enhancement where the system dynamics are often nonlinear and complex. Secondly, WOA can efficiently handle multi-objective optimization problems. In MPPT tracking in Solar PV System with Smart Grid Enhancement using CMCMAC protocol, the objectives include maximizing the energy output of the PV system while minimizing the cost of energy. WOA can handle both objectives simultaneously and find the Pareto optimal solutions. Thirdly, WOA has a fast convergence rate and can quickly find the optimal solution. This is important in MPPT tracking in Solar PV System with Smart Grid Enhancement as the system parameters may change rapidly due to weather conditions or other factors. Lastly, WOA is easy to implement and requires few parameters to be set. This makes it suitable for real-world applications in MPPT tracking in Solar PV System with Smart Grid Enhancement where simplicity and efficiency are essential. In summary, the rationale behind using WOA in MPPT tracking in Solar PV System with Smart Grid Enhancement using CMCMAC protocol is that WOA is a powerful optimization algorithm that can handle complex and nonlinear optimization problems, can efficiently handle multi-objective optimization, has a fast convergence rate, and is easy to implement.

WOA can be eminent when compared with the supplementary optimization algorithms by considering only two constraints to adjust. These constraints are allowed for soft conversion in both examination and utilization processes. The numerical model has been analyzed based on the subsequent three processes of spiral bubble-net feeding maneuver, encircling prey, and search for prey. At the time of encircling the prey, the humpback whales identify the place of the victim to encircle it. Meanwhile, the location of the most favorable approach in the search is unknown earlier and the WOA technique undertakes that the present preeminent solution is the target prey [

where

The Bubble net attacking technique was reached by reducing the value

where, ^{th} whale to the prey,

In this proposed system, an Adaptive WOA-ANN-Based technique has executed the optimization of control parameters of the PV system based on the MPPT controller using an embedded controller. It can be assumed that if the maximal effectual of MPPT was selected, next, it could be stated that the selective of DC-DC converter from the PV model is essential. Now, the DC-DC converter was capable of directing the ability of the PV panel with the maximum power point at full times, solar universal irradiation, even with panel temperature, and linked load. The proposed technique is a grouping of ANN and WOA algorithms. The WOA algorithm increases the knowledge of ANN. The WOA-optimized technique is one of the preferred favorable preparation datasets to lead ANN to escalate the data presentation. Mostly, the ANN contains three layers namely input, hidden, and output layers. An input layer has various nodes, which are determined as the dataset. All the nodes cover compared with the weights of total nodes under the next layer, and also one bias is linked to the same nodes of the next layer [

where,

The dissimilarity between objective output

Input and output layer node records were formerly determined by the input and output counts, respectively. Nodes in the prohibited layer defined by Equation are now accessible through the Kolmogorov theorem

In ANN, artificial neurons are easy-to-process components with configurable inner limitations. Weightage, threshold incoming signals, and sum values form an output for artificial neurons. Interconnection/weight and threshold/bias data. The network uses error (difference between predicted and measured output) as a learning tool. This learning approach uses WOA, a nature-inspired optimization count [

In this section, the embedded controller-based MPPT for PV systems using WOA and ANN techniques has been clearly described. Now, the controller and the algorithms have achieved fault detection by controlling the voltage and Current. A brief introduction about the data of MPPT, which has functioned as the terminal voltage of solar panel reliability profits in modern manufacturing industries, has appeared. An embedded organizer has WOA and ANN algorithms that accomplish an interchange pulse of DC-DC boost converter to trail MPP of the PV method. An adaptive method was implemented by a fixed regulator connected to definite emission or temperature dimensional. The reproduction concerns have explained the ability of an anticipated technique to eliminate the highest power by quick alteration of emission amongst rapid and raised reactions. The eliminated highest power includes the worse amount of oscillations at stable conditions, i.e., improving the effectual and accuracy of the PV system. Also, the WOA and ANN technique has achieved successful power generation in the PV method without disabling the whole system. The productivity of the proposed controller has been analyzed using various controllers. The proposed algorithm operation, equivalent characteristics, and mathematical models are researched. The procedure of the proposed technique, the controller, and the performance against maximal power from the PV method are clearly described [

The guarantee of optimality of the obtained solutions in any optimization algorithm, including the Whale Optimization Algorithm (WOA), depends on several factors, including the objective function, the optimization problem, and the optimization algorithm’s convergence criteria. In the case of the MPPT tracking in Solar PV System with Smart Grid Enhancement using CMCMAC protocol, the authors may have validated the optimality of the obtained solutions through simulations and experimental results. They may have compared the performance of the WOA algorithm with other optimization algorithms, such as genetic algorithms or particle swarm optimization, to determine the effectiveness of the WOA algorithm in obtaining optimal solutions [

We may have also used appropriate performance metrics, such as the peak power tracking error or the energy efficiency ratio, to evaluate the performance of the optimization algorithm and compare it with other existing methods. It is important to note that there is no guarantee that the obtained solutions are globally optimal, as most optimization problems are non-convex and have multiple local optima. However, the WOA algorithm’s stochastic nature can help to avoid local optima and find good solutions that are close to the global optimum. In summary, the authors may have validated the optimality of the obtained solutions through simulations and experimental results and may have used appropriate performance metrics to evaluate the optimization algorithm’s performance. While there is no guarantee that the obtained solutions are globally optimal, the stochastic nature of the WOA algorithm can help to find good solutions that are close to the global optimum.

The flow chart for the WOA-incorporated ANN method is depicted in

The prediction of interference in the proposed method of solar PV array of modules is performed by the Cross-Layer Multi-Channel Medium Access Control (CMCMAC) protocol. The CMCMAC protocol was designed to determine the interference in the wireless zig-bee network and it is composed of three stages namely, Current Interference estimation, Future interference estimation, and Channel assignment. The CMCMAC protocol composed of the aforementioned all three stages is depicted in

The current and future interference levels can be estimated using the CMCMAC protocol is determined as follows:

The average delay created in the proposed network of solar PV cells is estimated using

where, the delay (d_{avg}) implies the average delay created in the solar PV network and the P(nT) is the probability of “n” amount of packets delivered to the target at the time interval “T” and D(t) is the delay occurred during the transmission of “n” number of packets in the predefined time duration “T”.

The signal strength of the received signal is determined dependent upon RSSI (Received Signal Strength Indicator) and is expressed in

The proposed CMCMAC protocol is employed for mitigating the interference in the ZigBee network and involves a sequence of procedures namely.

Approximation of existing interference is performed

Anticipation of future interference.

Assignment of least interference in ZigBee transmission.

The value of RSSI can be determined using the mathematical

where,

RSSI—Received Signal Strength Indicator

L_{p}—Loss occurred due to propagation path exponent

d—propagation distance

P_{RS}—Receives signal strength at unit propagation distance.

The Channel Occupancy Rate (COR) is determined as the ratio of the quantity of channel to the bandwidth of the channel and is mathematically expressed in

The future interference prediction is performed by executing the proposed algorithm of six sequential procedures. In this way, the broadcast of data is performed with the channel by minimal interference and ensures that the data does not get diminished due to the co-channel interference of any type of interferences.

The adaptive WOA-based ANN controller has been implemented in Intel(R) Core (TM) i5 processor, 8 GB RAM, and MATLAB/Simulink 7.10.0 (R2015a). This section depicts the performance analysis performed in the proposed WOA and ANN technique-based embedded controllers. The MPPT controller process with the proper implementation parameters has been estimated. In

Parameters | Value |
---|---|

Model | Sun Power SPR-305-WHT |

Peak power (W) |
305.226 |

54.6 V | |

5.5 A | |

Quantity of series modules/string | 5 |

Voltage under open circuit |
64.1 V |

No of strings in parallel | 60 |

Short-circuit current |
5.96 A |

No. of series connected cells |
96 |

Temperature coefficient of voltage |
−0.123 V/K |

Temperature coefficient of Current |
0.0032 A/K |

The power is measured from the array of PV modules and they are given to the embedded boards of Analog to Digital Converters. The observed data are isolated implementing a filter using a simulation tool. Then, an adaptive technique is given to determine and trail the MPP. Based on the embedding controller, the control pulse is provided to the DC-DC boost translator employing the adaptive technique. Subsequently, performance in the proposed method is analyzed pedestal on the power parameters. The PV power is varied depending on the input attributes of irradiance and temperature an array of PV methods. Based on the irradiance (G), the performance analysis has been presented with two cases such as step irradiance and stable irradiance. The proposed controller offers good performance in both cases.

Parameter name | Value |
---|---|

Population size | 50 |

No of dimension | 3 |

No of iterations | 10 |

Attractiveness constant | 1 |

No of fireflies | 20 |

Randomness reduction factor | 0.97 |

Absorption coefficient | 0.01 |

^{2}

In this section, the presentation is investigated by the variable solar radiation with the values of 1000, 400, and 1000 ^{2}. The PV module is for the whole irradiation stage, the temperature is reserved as stable at 25°C.

Based on the PV step irradiance, the PV parameters, namely voltage, current, and power, are measured with the proposed embedded controller and it is presented in

The output parameters of the PV technique established on the adaptive WOA with PV array in stable irradiance conditions have been estimated. They are illustrated in

In this case, stable irradiance has been given to an array of PV techniques, and then the measured parameters of the array of PV methods are illustrated.

Lastly, the comparative analysis of the proposed MPPT controller-based embedded system is presented in the section. In this analysis, the warmth has continued at a constant value of 25°C at the entire irradiation stages. The comparison analysis is portrayed in

MPPT techniques | Condition | Required power (KW) | Output voltage (KV) | Output current (A) | Output power (KW) | Efficiency |
---|---|---|---|---|---|---|

74.27 | 20.95 | 3.5 | 73.32 | 98.72 | ||

66.25 | 21.14 | 3.11 | 67.0138 | 99.71 | ||

63.41 | 20.01 | 3.16 | 63.2316 | 96.61 | ||

62.14 | 20.02 | 2.99 | 59.8598 | 96.33 | ||

66.45 | 20.45 | 3.17 | 64.8265 | 97.55 | ||

66.32 | 19.54 | 3.25 | 63.505 | 96.75 | ||

59.05 | 20.24 | 2.74 | 55.4576 | 93.91 | ||

54.01 | 19.32 | 2.69 | 51.9708 | 96.22 |

The simulation of the CMCMAC protocol is executed using Network Simulator—2 (NS-2.32) version. The proposed model employed IEEE 802.15.4 standard for predicting the interference using the proposed CMCMAC protocol. The simulation specifications are provided in

Parameter | Specifications |
---|---|

Area | 100 * 100 sq.mt |

MAC | IEEE 802.15.4 |

Time for simulation | 60 s |

Traffic source | CBR |

Propagation | Two-ray |

Initial energy | 10 J |

Antenna | Omni-directional |

Transmitted energy | 0.3 mJ |

Received energy | 0.3 mJ |

This paper has proposed an embedded controller based on different irradiance conditions to determine the path of MPP in quick variations in weather and to reduce steady-state oscillation. The embedded controller has WOA and ANN models to control the switching pulse of the DC-DC converter for tracking the MPP of PV systems and is designed with a pro type. The adaptive AI technique has been realized via the embedded controller based on the measurement of irradiation and temperature. Simulation outcomes prove the competence of the presented technique to haul out a precision peak power and the rapid variations of radiation with swift and elevated responses with 98.72% efficiency. The measured peak power has the most negligible oscillations at a steady state, and it causes an elevation in the efficiency and accuracy of the performance of solar PV systems. Analytical results yield that the proposed algorithm possesses higher stability, reliability and minor oscillation than the existing MPPT algorithms. The proposed CMCMAC protocol was designed with the primary objective of predicting the interference in the communicating channels. Whenever the node is ready to transmit data, it determines the present and future interference in the communicating channel with the assistance of HMM. The CMCMAC protocol was implemented to escalate the communication efficiency of the presented model. The quantitative result yields that the prediction of CMCMAC is precisely accurate in terms of interferences.

The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this research work through the Small Group Research Project under Grant Number RGP1/70/44.

The authors gratefully acknowledge their respective organizations for their help and support. The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University (KKU), Kingdom of Saudi Arabia for funding this work through Small Group Research Project under grant number (RGP1/70/44).

The authors have no conflicts of interest to report regarding the present study.