LEE- Kontrol ve Otomasyon Mühendisliği-Yüksek Lisans

Bu koleksiyon için kalıcı URI

http://hdl.handle.net/11527/19360

Gözat

Yazar "Çalışkan, Fikret" ile LEE- Kontrol ve Otomasyon Mühendisliği-Yüksek Lisans'a göz atma
  • Öge
    Optimized power control strategy for a proton exchange membrane fuel cell system
    (Graduate School, 2024-06-10) Sarıçay, Ömer Burak ; Çalışkan, Fikret ; 504211115 ; Control and Automation Engineering
    Global warming has become one of the greatest threats our planet is facing. The primary causes of this threat include the increase in greenhouse gas emissions into the atmosphere as a result of human activities, leading to a strengthening of the greenhouse effect. Factors such as widespread use of fossil fuels, emissions from industrial facilities, agricultural practices, and deforestation contribute to increasing concentrations of greenhouse gases in the atmosphere, causing global warming. To address this issue, increasing the focus on renewable energy sources is of critical importance. Renewable energy sources have the potential to generate energy without harming the environment, thereby reducing greenhouse gas emissions by decreasing the use of fossil fuels. In this context, fuel cell technology presents an innovative approach to clean energy production, offering an effective solution to the problem of global warming. A fuel cell is an energy conversion device that directly converts chemical energy into electrical energy. This technology typically generates electricity through an electrochemical reaction between hydrogen and oxygen, and also can produce useful outputs such as clean water and heat. Fuel cell technology has a wide range of applications, emerging prominently in both stationary and mobile applications. Stationary use ranges from power generation in power plants to industrial applications, while mobile use extends from transportation vehicles to portable electronic devices. Various types of fuel cells exist, such as solid oxide, alkaline, phosphoric acid, and polymer electrolyte membrane fuel cells. Due to advantages such as high energy density compared to internal combustion engines, high efficiency, and lack of carbon emissions, fuel cell is becoming increasingly prominent in the automotive sector. The European Union aims to reduce harmful emissions from internal combustion engines with updated emission standards. Additionally, a ban on the sale of gasoline and diesel vehicles is targeted from 2035 onwards, leading to an increase in the importance of electrification in the automotive sector, with rising sales of electric vehicles observed in recent years. Advances in battery technology, incentives, widespread infrastructure, and changing consumer behavior have contributed to the increasing market share of battery electric vehicles. While battery electric vehicles offer advantages such as carbon emissions-free and silent operation, low maintenance costs due to fewer parts, and cheaper charging costs, they also have some disadvantages as a result of relying solely on batteries as the power source. Challenges include long charging times, short battery lifespan, high cost, high weight due to low specific energy of batteries leading to range limitations, performance degradation in cold weather, and uncontrolled battery fires. In response, the development of hybrid vehicles provides a separate use of the advantages offered by internal combustion engines and batteries, but the long-term effectiveness of this solution is not expected due to the limited lifespan of internal combustion engines. Fuel cell technology presents an alternative solution, potentially replacing internal combustion engines in hybrid vehicles. Fuel cell technology requires a hydrogen-based fuel type and can be stored in a vehicle via hydrogen tanks. As a result, refueling can be done quickly similar to internal combustion engine vehicles, with the only limitation being the size of the tank. With hybrid use, the required battery size decreases, and performance losses can be reduced with efficient energy management. However, fuel cell technology also has some disadvantages that need to be addressed. One of the main challenges in fuel cell technology is the decrease or deterioration of cell performance due to degradation factors. Catalyst degradation, membrane damage, electrode fouling, water management issues, and voltage losses accelerate cell degradation and reduce its lifespan. Optimizing cell design, selecting suitable electrode materials, using high-quality fuel, and controlling operational conditions considering degradation conditions can extend cell life. From control engineering perspective, improving operational conditions is the main theme of this study. One of the important operational conditions that need to be controlled is oxygen concentration. In situations where oxygen concentration is not controlled, degradation is accelerated. For example, there are studies in the literature showing that oxygen deficiency accelerates degradation. In opposite cases like excess oxygen concentration, there is excessive power consumption due to the overworking of the air compressor. Therefore, oxygen concentration needs to be optimally controlled during power control. This thesis aims to prevent instantaneous drops in oxygen concentration and develop a control strategy that ensures the operation of the cell at optimized oxygen concentration. This control strategy includes power control as well as oxygen concentration control. The proposed power control strategy is tested on an open-source proton exchange membrane fuel cell system model from Michigan University. This model integrates compressor, supply and return manifold, humidifier, anode flow supplier, and cell stack models. The stack current and compressor voltage are inputs to the system model, while the oxygen excess ratio and net power are outputs of the system model. The control system adjusts system inputs to operate the system at the desired net power and desired settling time and overshoot ratio. While the stack produces power, the power consumption of the compressor is simulated with the applied compressor voltage. The stack model also calculates the oxygen excess ratio by proportioning the oxygen supplied to the consumed. Since the compressor is the dominant power-consuming component, other power-consuming components have been neglected in net power calculations. The anode flow distributor controls fuel supply based on the difference between the supply manifold and anode pressure. The humidifier regulates the operation of the cell stack at the set humidity. Collectors are modeled by assuming a collected volume and used in thermal conduction and fluid dynamics calculations. Oxygen concentration is controlled through a parameter called the oxygen excess ratio. The ratio of the supplied oxygen to the consumed oxygen during power generation is given as the oxygen excess ratio. Since the oxygen fed into the system cannot be less than the oxygen used, the value of this ratio can theoretically be a minimum of 1. Since the oxygen fed into the system is proportional to the power consumption of the compressor motor, and the oxygen consumed is proportional to the current drawn, and the current drawn is also proportional to the power generated by the cell. The net power obtained from the fuel cell system is obtained by subtracting the power consumed by the compressor from the power generated from the cell. Although the option of high cell current - low operating voltage for compressor seems suitable for maximizing net power, the desired power cannot be obtained due to inadequate oxygen supply. Therefore, compressor voltage should be regulated to provide optimized oxygen excess ratio to maximize the net power. In this study, a look-up table based reference oxygen excess ratio generator is used via curves of experimental data. The oxygen excess ratio that maximizes the desired power is provided to oxygen excess ratio controller as a reference. One of the difficulties in controlling the excess oxygen ratio is the disturbance effect of cell current. With increasing cell current, reactions accelerate, requiring an increased amount of oxygen to be used. However, if the compressor voltage is not increased in parallel with the increase in current, the excess oxygen ratio becomes insufficient instantly. This situation is called oxygen starvation and has an accelerating effect on cell degradation. To prevent oxygen starvation, changes in compressor voltage parallel to changes in cell current must be applied. In this study, a current-dependent static feedforward controller is used. A feedforward compressor voltage term is generated in proportion to the applied current, and it is added to the compressor voltage generated by the feedback controller to obtain the final compressor voltage applied to the compressor. Test results have shown that sudden decreases in oxygen level are avoided thanks to the feedforward controller. The fuel cell power system is a good example of a non-minimum phase system. When the reference oxygen excess ratio is reached, the power is guaranteed to be maximized while improving the transient response of oxygen dynamics leads to a decrease in net power for a moment due to excessive compressor consumption. The classic PI control method has been preferred as the initial configuration for the control of oxygen excess ratio and power. The reason for its frequent preference in industry, its simple mathematical structure and adjustable flexibility with calibration parameters. The fuel cell system exhibits different responses at each operating point due to its complex nonlinear system characteristics. Therefore, rather than analytical methods, tuning of controller parameters is executed experimentally. Non-minimum phase dynamics result infeasible usage of classical PID controllers due to insufficient stability. To have a better settling time performance for oxygen excess ratio, nonlinear fuzzy logic-based PID controller is proposed as an alternative to linear PI controllers. In the final section, fuzzy type 1 and type 2 PID controllers are tested. While better transient results are obtained after experimental tuning of scaling coefficients for the type-1, no improvement is observed with the adjustment of the type-2 controller. Therefore, the use of a type-1 fuzzy logic-based PID controller has yielded the best results in terms of oxygen transient dynamics. An efficiency formula is derived for this study and implemented within the model. It indicates that system have an efficiency range of 45-50%. Additionally, optimality of OER brings 0.6% improvement in efficiency compared to a non-optimal scenario.