Comparison of load resistor and battery on controlling buck boost converter using fuzzy-pi method

This paper discusses about controlling the buck boost converter output voltage remains constant by comparing of loads resistor and battery. Buck boost converter output voltage tends to be unstable when given a disturbance with a variety of loads, for that it needs proper control to adjust the duty cycle. The controls that will be used are a combination of conventional control methods, i.e. Proportional Integral (PI) and intelligent control, i.e. Fuzzy logic. The results show that the combination of PI control and Fuzzy Logic can improve performance, i.e. small errors close to zero and the buck boost converter output voltage remains constant even though the load is varied, i.e. resistors and battery.


Introduction
Based on the range and usefulness of the DC-DC converter there are various kinds, i.e. Buck Converter, Boost Converter, Buck boost converter, Cuk Converter, Sepic Converter and Zetta Converter that works with switching systems. The power supply with this switching system is better than linear systems. In linear power supply the efficiency is very low due to power losses in the transistor, while the switching system if it is assumed is ideal then the converter efficiency can reach up to 100% [1] This paper will discuss the control of the output voltage of DC-DC buck boost converter which serves to decrease and increase the voltage to remain constant when given of loads a resistance or battery. Buck boost converter is a switching converter that can decrease and increase the output voltage by adjusting the duty cycle provided to switching devices such as: Mosfet, Thyristor, IGBT, GTO. To be able to control the converter output voltage, a control method is needed that can increase efficiency, eliminate offset and produce large initial changes. To overcome this problem, a method is used, which is a combination of PI control and Fuzzy Logic that will regulate the duty cycle. The PI method is a conventional control that has been widely used both for research and for applications in industry. While Fuzzy like Proportional Derivatives (PD) is a fuzzy intelligent control whose behavior resembles a PD.
The converter consists of a DC input voltage source (Vs), a controlled switch (S), inductor (L), diode (D) capacitor filter (C) and resistance load (R). The Buck boost converter produces an output voltage greater than or smaller than the input voltage, depending on switching duty (D). When the switch is ON, the diode is reverse biased, the input voltage goes through the inductor so that the inductor voltage (VL) equals the input voltage (Vs) and the inductor current rises linearly so that the energy in the inductor goes up too.  Microcontroller" discussing how to maximize the use of solar panels with electronic circuit buck boost converter that can increase and decrease the output voltage. In this study using PI and PID control methods to improve system performance that can be seen from the response speed and steady state error. By using trial and error, the efficiency gained from buck boost with PI control is 9.5% and from PID control is 8.7%, but PID Control is better than the maximum overshoot (MP) and Settling Time (TS) values smaller from PI control [2].
Maula Nurul Khakam et al in her research entitled "Design and Implementation of Battery Charging Management Systems and Loads on Independent Power Plants Using Synchronous Non-Inverting Buck Boost DC-DC Converter discusses the design, simulation and implementation of synchronous non-inverting buck boost (SNIBB) using four H-bridge switches to adjust the current and multi-stage voltage charging the lead-acid battery, using two modes ie constant current (CC) and constant voltage (CV). When the CC and CV modes of the SNIBB circuit are able to maintain the charging voltage and current according to the set point with the change in input voltage. The battery management test with loads can regulate the charging process and discharging the battery to the load [3] Jumiyatun et.al in his paper discussed the comparison of two methods, i.e. the PI conventional method with Fuzzy Logic -PI to control the buck converter output voltage remains constant. Control using Fuzzy-PI controllers is better for performance and faster response than conventional methods, i.e. PI is shown from the graph of the output voltage response with large rise time difference [4] Jumiyatun et al, in his research entitled " Controlling DC-DC Buck Converter Using Fuzzy-PID with DC motor load" discusses the DC-DC buck converter which functions to decrease the voltage by using switching MOSFET to control the buck converter output voltage constant with the motor load DC. The method used is a combination of conventional control methods, namely Proportional Integral Derivatives (PID) and intelligent controls i.e. Fuzzy logic to improve performance so that a small error approaches zero and the buck converter output voltage remains constant even though the motor load is altered. [5]

Research Method
In designing the buck boost converter in this research, the first step is to determine the appropriate parameter values of the inductor (L), capacitor (C), and duty cycle components which will then be compared when testing using R and battery loads. The method used is a combination of Fuzzy logic and PI controller, so it is expected to improve performance. In Figure.3. shows the diagram system of a dc-dc buck boost converter, input voltage and current from a DC source, where the input voltage is designed higher or lower than the output voltage that will be increased or lowered according to load requirements. The output voltage will be set according to the specified set point and then feed backed and the error obtained is controlled using the fuzzy-PI combined method.

Design of DC-DC Buck Boost Converter [8]
To design a buck boost converter as a first step is to calculate and determine the value of the appropriate component parameters so that the whole system runs well.

PI Controller (Proportional Integral) [10]
Proportional plus integral (PI) controller is a controller whose control action has proportional and integral properties to error signals.
where is proportional reinforcement and is integral time. The parameters and can be determined. So that the proportional plus integral controller transfer function is The proportional plus integral controller block diagram is as follows:

Fuzzy Logic Controller [11]
Fuzzy logic controller is a fuzzy system that is applied as a controller where in this research will be combined with a PI controller to produce better performance. Variable inputs on fuzzy controllers are generally in the form of a difference between the output reference value and the actual output value called error value. Whereas fuzzy logic controller output is command control given to actuator or drive. Fuzzy controller structure consists of fuzzifier, fuzzy rule base, inference engine and defuzzied.  Arrangement of fuzzy rules by using inference assistance to determine the control signal, by using a membership set of seven pieces for each input shown in Table 4. In the design of the inference used is the Mamdani method which has the Equations: Next to get the control action value (u) a defuzzification process will be carried out, the COA (Center of Area) method.

Result and Discussion
Data retrieval and testing on research of buck boost is done during open loop and close loop, that is, before and after the combined control of Fuzzy logic and PI control is installed. The load used is the resistance and battery.

Conclusion
The results comparison of load the resistor and battery for controlling the output voltage of DC-DC buck boost converter by using a combination of Fuzzy-PI method shows that when open loop and close loop, resistor load performance is better than using a battery load. This can be seen from the graph of the output voltage response where the load resistor is more stable with zero error.