Abstract
As environmental pollution and energy shortages become increasingly severe, finding measures to address these issues and promote harmonious development among the environment, economy, and energy is a pressing concern. Solar energy, as an inexhaustible renewable energy source, has gained significant attention, and photovoltaic (PV) power generation has become a focal point. However, as of 2023, the conversion efficiency of photovoltaic cells has only reached a maximum of 35.5%, which is relatively low. The efficiency of PV power output is influenced by factors such as maximum power point tracking (MPPT), inverter efficiency, solar radiation, the tilt angle of PV modules, module efficiency, and matching losses. Among these, improving MPPT control technology is the most cost-effective and impactful method to enhance tracking efficiency. Due to the difficulty in determining the global maximum power point (GMPP) under changing environmental conditions such as temperature and light intensity, this paper focuses on MPPT control technology to improve the output efficiency of PV systems and accurately capture the GMPP under varying conditions.
1. Introduction
1.1 Research Background and Significance
The phenomenon of global warming is intensifying, and the depletion of non-renewable energy sources has led to a growing focus on clean and renewable energy generation. Governments worldwide are advocating for the use of solar, hydro, and wind energy to address the impending exhaustion of non-renewable resources. Solar energy, in particular, has the potential to replace fossil fuels like oil and natural gas, reducing carbon emissions and mitigating pollution, which is beneficial for human health and environmental protection. Over the past four decades, China has made significant progress in solar and wind energy. However, despite the widespread integration of solar and wind power into the grid, they still lag behind coal power in terms of efficiency.
Under the “dual carbon” goals of carbon peak and carbon neutrality, China is expected to increase its carbon reduction efforts. PV power generation is a strategic emerging industry that is crucial for advancing energy system reform and improving the natural environment. PV systems are safe, reliable, noise-free, non-depleting, and emission-free, making them a clean energy source that has garnered global attention. For a PV system to operate at its maximum power point (MPP), it must track the MPP, which is influenced by factors such as solar radiation, ambient temperature, shading, and dirt. Therefore, improving MPPT control technology is essential to maximize the conversion of solar energy into electricity under varying climatic conditions.
1.2 Current Research Status at Home and Abroad
The research on MPPT control technology for PV systems integrated into active distribution networks has been a hot topic. PV cells are susceptible to environmental changes, making it difficult to maximize solar energy utilization. Current research on MPPT control technology can be divided into three categories: traditional tracking techniques and their improvements, intelligent optimization algorithms, and hybrid algorithms combining two or more techniques.
Traditional MPPT techniques, such as the Perturb and Observe (P&O) method and the Incremental Conductance (INC) method, are well-established and effective under uniform lighting conditions. However, under partial shading conditions, where the PV array exhibits multiple peaks in its power-voltage (P-V) curve, these traditional methods often fail to track the global maximum power point (GMPP) and may get stuck in local maxima.
Recent studies have proposed various improvements to traditional MPPT techniques. For instance, some researchers have developed adaptive variable step-size INC methods to enhance tracking speed and accuracy. Others have combined intelligent algorithms like Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO) with traditional methods to improve tracking performance under partial shading conditions.
1.3 Main Research Content of This Paper
This paper aims to improve the MPPT control strategy for PV systems operating under partial shading conditions. The main contributions are as follows:
- Analysis of PV System Output Characteristics: The paper analyzes the output characteristics of PV arrays under uniform and shaded conditions, establishing the relationship between these characteristics and the GMPP. It also reviews the principles of the INC and P&O methods and identifies the limitations of current MPPT control algorithms.
- Improvement of GWO and INC Algorithms: The paper proposes an improved GWO algorithm by incorporating the Lévy flight function to enhance its global and local search balance. Additionally, it introduces a modified INC algorithm that divides the MPP into two regions and adjusts the step size using different factors. The improved INC algorithm is shown to reduce tracking time compared to traditional P&O and INC methods.
- Proposal of a Hybrid GWO-INC Algorithm: The paper combines the improved GWO and INC algorithms to create a hybrid GWO-INC control algorithm. This hybrid algorithm uses the improved GWO for global search to approximate the GMPP and then switches to the improved INC for precise tracking. The proposed algorithm is validated through simulations and experiments, demonstrating its effectiveness in tracking the GMPP under varying environmental conditions.
2. Photovoltaic Power Generation System and MPPT Control Technology
2.1 Photovoltaic Power Generation System
PV systems can be classified into two types based on their relationship with the power grid: standalone PV systems and grid-connected PV systems.
- Standalone PV Systems: These systems consist of PV arrays, battery banks, charge controllers, DC/DC converters, and loads. They are typically used in remote areas where grid connection is not feasible.
- Grid-Connected PV Systems: These systems include PV arrays, DC/DC converters, inverters, and grid connection components. They are designed to feed excess power into the grid.
2.2 Photovoltaic Cell Principles and Models
The operation of a PV cell is based on the photovoltaic effect, where light energy is converted into electrical energy. The equivalent circuit of a PV cell includes a current source, a diode, and series and shunt resistances. The output current of the PV cell can be described by the following equation:
I=Iph−I0(exp(V+IRsnVT)−1)−V+IRsRshI=Iph−I0(exp(nVTV+IRs)−1)−RshV+IRs
Where:
- IphIph is the photocurrent,
- I0I0 is the reverse saturation current,
- VV is the output voltage,
- RsRs is the series resistance,
- RshRsh is the shunt resistance,
- nn is the ideality factor,
- VTVT is the thermal voltage.
2.3 Output Characteristics of PV Arrays
The output characteristics of PV arrays are influenced by factors such as solar radiation, temperature, and shading. Under uniform conditions, the P-V curve of a PV array exhibits a single peak, while under partial shading, multiple peaks may appear. The presence of multiple peaks makes it challenging to track the GMPP using traditional MPPT methods.
2.4 MPPT Control Technology
MPPT control technology aims to maximize the power output of a PV system by continuously adjusting the operating point of the PV array. The most common MPPT methods include:
- Perturb and Observe (P&O): This method perturbs the operating voltage of the PV array and observes the change in power output. If the power increases, the perturbation continues in the same direction; otherwise, it reverses.
- Incremental Conductance (INC): This method uses the incremental conductance of the PV array to determine the direction of the perturbation. It is more accurate than P&O but requires more computational resources.
3. Improved GWO-INC Fusion Algorithm
3.1 Grey Wolf Optimization (GWO) Algorithm
The GWO algorithm is inspired by the social hierarchy and hunting behavior of grey wolves. The algorithm simulates the leadership hierarchy and hunting mechanism of grey wolves to solve optimization problems. The social hierarchy of grey wolves consists of four levels: alpha (α), beta (β), delta (δ), and omega (ω). The hunting process involves three main steps: searching for prey, encircling prey, and attacking prey.
The mathematical model of GWO is as follows:
D⃗=∣C⃗⋅X⃗p(n)−X⃗(n)∣D=∣C⋅Xp(n)−X(n)∣
X⃗(n+1)=X⃗p(n)−A⃗⋅D⃗X(n+1)=Xp(n)−A⋅D
Where:
- D⃗D is the distance between the wolf and the prey,
- X⃗p(n)Xp(n) is the position of the prey,
- X⃗(n)X(n) is the position of the wolf,
- A⃗A and C⃗C are coefficient vectors.
3.2 Improved GWO Algorithm
To enhance the global search capability of the GWO algorithm, the Lévy flight function is introduced. The Lévy flight is a random walk that alternates between short and long steps, allowing the algorithm to escape local optima and explore the search space more effectively. The position update equation with Lévy flight is given by:
X⃗(n+1)=X⃗(n)+M⊕Levy(β)X(n+1)=X(n)+M⊕Levy(β)
Where:
- MM is the step size control factor,
- Levy(β)Levy(β) is the Lévy random path.
3.3 Improved INC Algorithm
The traditional INC method has a fixed step size, which can lead to oscillations around the MPP and slow tracking speed. To address this, the paper proposes a variable step-size INC method that divides the MPP into two regions and adjusts the step size based on the slope of the P-V curve. The step size is adjusted using different factors on either side of the MPP, improving tracking speed and accuracy.
3.4 Hybrid GWO-INC Algorithm
The hybrid GWO-INC algorithm combines the global search capability of the improved GWO with the precise tracking ability of the improved INC. The algorithm first uses the improved GWO to approximate the GMPP and then switches to the improved INC for fine-tuning. This approach reduces tracking time and improves tracking accuracy, especially under partial shading conditions.
4. Simulation and Analysis
4.1 Simulation Model
The simulation model is built using MATLAB/Simulink and includes a PV array, a DC/DC boost converter, an MPPT controller, and a load. The PV array consists of five series-connected PV modules, and the boost converter is controlled by the MPPT algorithm to maximize power output.
4.2 Simulation Results
The simulation results show that the proposed hybrid GWO-INC algorithm outperforms traditional P&O, INC, and GWO methods in terms of tracking speed and accuracy. Under uniform conditions, the hybrid algorithm achieves a tracking efficiency of 99.72%, compared to 99.26% for GWO and 99.35% for INC. Under partial shading conditions, the hybrid algorithm achieves a tracking efficiency of 99.95%, compared to 97.86% for GWO and 99.80% for INC.
5. Experimental Testing and Case Application
5.1 Experimental Platform
The experimental platform consists of a PV simulator, a DSP controller, a boost converter, and a load. The PV simulator emulates the behavior of a PV array under different environmental conditions, and the DSP controller implements the MPPT algorithm.
5.2 Experimental Results
The experimental results confirm the simulation findings, showing that the hybrid GWO-INC algorithm achieves higher tracking efficiency and faster convergence compared to traditional methods. Under dynamic shading conditions, the hybrid algorithm maintains a tracking efficiency of over 99%, while traditional methods struggle to track the GMPP accurately.
5.3 Case Application
The proposed hybrid GWO-INC algorithm is applied to a real-world PV system at the Wanzhou Vocational Education Center in Chongqing. The system’s tracking efficiency improves by 2.7% compared to the original system without the MPPT control algorithm. The annual power generation increases by 81,843.12 kWh, resulting in significant economic and environmental benefits.
6. Conclusion and Future Work
6.1 Conclusion
This paper proposes an improved GWO-INC algorithm for MPPT control in PV systems. The algorithm combines the global search capability of GWO with the precise tracking ability of INC, resulting in faster convergence and higher tracking accuracy. The proposed algorithm is validated through simulations and experiments, demonstrating its effectiveness under varying environmental conditions.
6.2 Future Work
Future research will focus on implementing the proposed algorithm on a full-scale hardware platform and testing it under real-world conditions. Additionally, the algorithm will be integrated into grid-connected PV systems to evaluate its performance in a more complex environment. Further improvements will also be explored, such as incorporating machine learning techniques to enhance the algorithm’s adaptability to changing environmental conditions.
Tables
| Parameter | Description | Value |
|---|---|---|
| IphIph | Photocurrent | – |
| I0I0 | Reverse saturation current | – |
| VV | Output voltage | – |
| RsRs | Series resistance | – |
| RshRsh | Shunt resistance | – |
| nn | Ideality factor | – |
| VTVT | Thermal voltage | – |
| Algorithm | Tracking Time (s) | Tracking Efficiency (%) |
|---|---|---|
| GWO | 0.086 | 99.26 |
| INC | 0.090 | 99.35 |
| ICS-IP&O | 0.080 | 99.60 |
| Improved GWO-INC | 0.077 | 99.72 |