Charge-Controller Optimization on Lead-Acid Battery in Solar PV Systems: Temperature Effects and Efficiency Improvement

. Long-term cell performance is very sensitive to the cell operating temperature, and cell storage capacity can degrade quickly, if the temperature is not maintained within a narrow range (25–50 °C) during charging and discharging of (solar) batteries [1]. Efforts are recently being dedicated to developing models that seek to provide insights into that issue. However, not all models consider the operation of the photovoltaic (PV) battery storage system with regard to battery optimization and temperature effects. The present work provides a controllable algorithm to help charge controllers provide exact amount of PV electricity (charge equalization) to batteries with temperature compensation included, and a proposed charging and discharging schedules of the battery storage. The temperature compensated duty cycle for charging is modelled and the pulse-width modulation (PWM) signal is programmed to change with temperature following this duty cycle model. This research work is based on the optimization of solar battery storage where the micro controller-based charge controller enhances battery life by monitoring the temperature and controlling charging voltages and float charging voltages for specific temperatures. A buck converter was simulated in Proteus, and then realized in the laboratory. The duty cycle of the buck converter was adjusted with temperature. Results collected from lab experiments were plotted on MatLab and it shows homogeneity with calculated results. Moreover, battery-charging currents, battery direct current (DC) disconnect and battery switching for charging and discharging were performed for the converter. Future work is to extend this study to large-scale solar photovoltaic systems in order to overcome the operation limits encountered.


Introduction
Cameroon is a tropical country where the amount of sunlight is commonly more than sufficient to meet energy demands. The direct transformation of solar energy into electricity is highly appealing due to its rather low environmental and health hazards. The output of photovoltaics (PV) dependents on sun radiation which changes over the course of the day and is also affected by other more unpredictable phenomena such as clouds. [3]. Temporal mismatches between solar photovoltaic (PV) system output and residential electricity demand are one of the primary challenges to wide-scale residential PV deployment [4]. PV output often exceeds residential electric loads during the day but falls short of demand in the late afternoon and evening when residential load tends to increase [4]. This can be solved using energy storage systems. One of the key parts of the technical and financial viability of offgrid PV systems is the selection of an adequate storage system [5]. Operating as an energy buffer storage batteries are essential for a steady and stable energy supply from * Corresponding author: a.fopah-lele@ubuea.cm renewable energy sources such as PV. However, they can also be the source of system failures and contribute significantly to both the initial and operation costs [2]. In the category of rechargeable batteries, lead-acid batteries are most commonly used as they have the advantage in terms of specific energy and is therefore preferred in compact designs [3]. A charge controller is necessary for the charging of these batteries. The charging mode consists of two phases, constant current and constant voltage. The constant current takes 20-30 % of the charging time and accounts to 70-80% of the battery level. Whereas the constant voltage phase which takes 70 to 80% of the charging time, accounts for 20-30% of the total battery level. The constant voltage phase contributes little towards the capacity but accelerates the capacity degradation. The discharge rates should be kept at a low level because the life cycle depend proportionally on the discharge rates [3]. In particular, long-term cell performance is very sensitive to the cell operating temperature, and cell storage capacity can degrade quickly if the temperature is not maintained within a narrow range (25-50°C) during charging and discharging [2]. Batteries experience a wide range of operational conditions in PV applications, including varying rates of charge and discharge, frequency and depth of discharges, temperature fluctuations, and the methods and limits of charge regulation [6]. These variables make batteries wear out faster.

Solar Charge Controller
A charge controller limits the rate at which electric charge is added to or drawn from electric batteries [7,8,9]. Charging a battery with a solar system sustainably is a demanding challenge. In the "old days," simple on-off regulators were used to limit battery outgassing when a solar panel produced excess energy. However, with time it became more and more clear that these simple devices are often insufficient for stable long-term battery operation. The history for on-off regulators includes early battery failures, increasing load and disconnects, leading to growing user dissatisfaction [10,11].

Types of solar charge controller
Over the years, trials have been made to efficiently charge batteries in solar PV systems. This has led to the development of charge controllers from the least complex to the most complex and today evolution is still ongoing. There are three different types of solar charge controllers [12], they are Simple 1 and 2 stage control PWM (pulse width modulated) MPPT (maximum power point tracking) A simple 1 and 2 stage control includes a shunt regulator and a series regulator. These are on/off type controllers. The regulator allows the current from the PV array to flow to the battery until disconnect voltage is reached, at which time the solar array is shorted or disconnected preventing any further current to flow to the battery. Without any charge current, the battery voltage will drop slowly (due to self-discharge) until reconnect voltage is reached at which time the regulator will allow the current to flow to the battery again. This cycle repeats as long as the battery is being used.
Pulse Width Modulation (PWM) is the most effective means to achieve constant voltage battery charging by switching the solar system controller's power devices. When in PWM regulation, the current from the solar array tapers according to the battery's condition and recharging needs. PWM solar chargers use technology similar to other modern high quality battery chargers. When a battery voltage reaches the regulation setpoint, the PWM algorithm slowly reduces the charging current to avoid heating and gassing of the battery, yet the charging continues to return the maximum amount of energy to the battery in the shortest time. The result is a higher charging efficiency, rapid recharging, and a healthy battery at full capacity [10].
MPPT stands for Maximum Power Point Tracking, which stands for the method used to regulate charge. MPPT charge controllers use this method of charging, which essentially finds out at any given condition, what is the maximum operating point for the panels current and voltage. With this method, MPPT controllers are actually 94-99% efficient. MPPT controllers have two special features about them that will be mentioned in the MPPT Charge Controller Sizing section. One is that they can accept a high input voltage and step this voltage down to match your battery bank voltage for a correct charge. Two is that even though they lower the voltage, they are able to recover any potential lost power via a boost current, which increase the amperage to make up for the lost voltage [13].

Why solar charge controller
Typically, a common battery charger is required to control the current to the battery with an optimal rate, and to cease charging when the battery is fully charged. However, for battery charging with solar energy, the input of the solar battery charger is uncontrollable owing to the impact of changing weather and day cycle conditions on the PV system. For this reason, conventional battery chargers fail to charge batteries efficiently and safely when using solar energy. A battery charged by solar energy without the optimal input control of a converter was described [13]. It is shown that battery charging with an unregulated voltage or current is a contributing factor in battery damage, and reduces the battery life cycle. The most important function of a solar battery-management system is the solar battery charger, which consists of a DC/DC converter application, and provides the interface between solar energy and the battery, leading to optimal energy transfer. Hence, the DC/DC converter is critical for regulating and transferring a suitable charging voltage and current according to battery specifications [14].

Battery Optimization
Battery optimization techniques are used to increase battery lifespan and efficiency by controlling the factors that shorten battery lifespan. In a PV battery system, the battery cells are commonly less durable components. Since they are relatively expensive, to be cost effective, there is a need to optimize the battery utilization in particular to lengthen their lifespans. Factors such as charge and discharge rate, depth of discharge, battery temperature and over charging affect the lifespan of batteries [14]. Since the cost of replacing batteries poses a concern, there is therefore a need to optimize battery usage to reduce the overall cost of a solar battery system. expectancy of a battery can be severely shortened by excessive temperatures [29]. The optimum operating temperature for the lead-acid battery is 25 °C (77 °F) [26]. It is important to note that increasing the temperature by 10 °C (18 °F) will result in roughly a 50% reduction in battery service life. On the other hand, reducing the temperature by the same amount can reduce the capacity of the battery by roughly 25% [30,31].

Battery temperature compensation
Temperature compensation during battery charging is commonly achieved by regulating the voltage. This adjustment is a charging feature that helps to ensure that a battery is appropriately charged taking battery temperature into account. Specifically, cold batteries require a higher charging voltage in order to push current into the battery plates and electrolyte, and warmer batteries require a lower charging voltage to eliminate potential damage to valve regulated lead acid (VRLA) cells and reduce unnecessary gassing if flooded cells are used [15]. Using standard target voltages to charge a battery that is colder than approximately 25 °C (77 °F) will probably result in an undercharged battery, which will deliver lower performance, shorter cycle and higher life cycle cost. Applying standard target voltages to a battery that is hotter than 25°C may result in an overcharging of the battery. This condition could lead to excessive outgassing of potentially dangerous amounts of hydrogen, increased battery maintenance in the form of more frequent watering, reduced battery life due to thermal stress, or the drying out of VRLA battery cells [15].

Relating battery temperature to charging voltage
Most of the temperature effects are related to chemical reactions of materials used in the batteries. Regarding chemical reactions, the relationship between the rate of chemical reactions and reaction temperature follows Arrhenius equation, and temperature variation can lead to the change of electrochemical reaction rate in batteries. Besides chemical reactions, the ionic conductivities within electrodes and electrolytes are also affected by temperature. For example, the ionic conductivity of lead salts-based electrolytes decreases at low temperatures. Taking such effects into account, to meet the expectation of a 20-year service life, as suggested by the United States Advanced Battery Consortium (USABC), is very challenging. In the following sections, we will discuss both the low temperature effects and the high temperature effects on lead-acid batteries. The relationship between battery temperature and cycle as well as float charging voltage is shown in Table 1.
For n cell battery, this equation is given by Figure 1 depicts the variation of float charging voltage as a function of temperature [22]. This clearly shows that the optimum operation temperature is at room temperature . The slope of the central section of the line has a value of -5 mV/°C, the battery temperature coefficient of most lead-acid batteries.

Relating battery temperature to duty cycle
Duty cycle is the amount of time a digital signal is active relative to the period of the signal. The amount of time the charging "signal" is active can be modelled to compensate battery temperature during charging. The duty cycle is given by: Considering the above equation of the charge voltage compensation, the compensated duty cycle D ' is given by = −∆ * * (4) This gives = −∆ * * (5) And substituting the values of n and α for a 12V, 6 cell lead-acid battery, we get = . − ∆ * * * (6) Therefor the PWM signal's duty cycle is programmed to change as the temperature changes. This will effectively correct charging voltages based on temperatures.

Optimizing battery life by using two battery banks
In order to reduce the frequency of loading and unloading and prevent battery usage during charging, two battery banks system has been adopted. When one battery bank is charging, the other is available for use (discharge). Switching is controlled by the relays and the control prevents usage or charging of both banks simultaneously. Figure 2 shows the battery bank selection circuit for both charging and discharging.

Charge Controller Algorithm Design and Programming
This section covers program flow charts, operational truth tables, algorithm design and programming of the charge controller using micro C. Figure 3 shows the flow chart of the functioning of the charge controller. Vref is the reference input voltage. The solar panel voltage is measured and the DC-DC converters are turned on based on if this voltage is at least equal to the reference voltage. Once the DC-DC converter is turned on, the temperature of the battery is constantly being monitored and normal; charging is applied if this temperature is within the safe operation range. Otherwise, the temperature compensating charging is applied using the modified duty cycle. During the charging process, the microcontroller is constantly checking the battery voltage and if this voltage is just equal to the float charging voltage for the given battery temperature, charging is stopped.

Battery bank charging and usage selection
There are two battery banks, Bank 1 and Bank 2. The preference for battery charging is Bank 1. Bank 2 only charges if Bank 1 is full. Table 2 shows the battery Banks charging preference.    Table 4a) and 4b) show the operational truth table for the charge controller. Vs is the supplied voltage, TBank 1 and TBank 2 are the measured temperatures of banks one and two respectively. In the truth table, 0 stands for LOW, not used, not up to the level, or out of the limit and 1 stands for HIGH, within limit or used. 1* is used here to indicate charging with temperature compensation.

PWM Signal Generation for Buck Converter
Our choice of microcontroller is PIC16F887 and the PWM signal is realized using CCP1 and CCP2 modules found on pin 17 and 16, respectively. Our program uses pin 17. The PWM signal was generated at 500 kHz and 85% duty cycle for charging within normal temperature range.

Simulated and Experimental results and discussions
This section presents the different results of simulations and lab results obtained throughout this research.

Simulation of buck converter activity without temperature compensation
The buck converter was built with PIC16F887 as the PWM signal generator and simulated in Proteus. Figure  4 presents the simulated result of the buck converter simulated in Proteus with a duty cycle of 85% without temperature compensation (basic operation). The output voltage gave 14.6 V from the 18 V input voltage. This is in line with the expected 14.5 V from calculations.

Simulation results of charging voltages and duty cycles at varying temperatures (with temperature compensation)
The various charging voltages and duty cycles were simulated in MATLAB for various temperatures and the results are presented in Table 5. From this MATLAB result, the duty cycle and float voltage of the controller falls as temperature increases, which agrees with the recommendations to lower the charging voltage when temperature rises. This helps to avoid gassing of the battery.

Charging voltage at various duty cycles
The charge controller was tested in the lab. The duty cycle was varied, and various values of float charging voltages were recorded. The results are presented as a plot in Figure 6. In contrast to the simulations, that provide a strictly linear dependency between duty cycle and float charging voltage, the experimental data show a distinct curvature at lower duty cycle (0.80-0.84). The results indicate that the simulation model merely provides a reasonable estimation.

Conclusion
This research work aims for the optimization of battery storage in solar photovoltaic systems. A buck converter was simulated in Proteus and then realized in the laboratory. Also, temperature control was incorporated in the algorithm and the duty cycle of the buck converter was adjusted with temperature. This gave the desired charging voltages for different temperatures. Battery level measurement was used for continuous monitoring of the battery for charging and usage decisions. The results collected from lab experiments were plotted in MATLAB and it shows homogeneity with calculated results. Lastly, battery charging currents, battery DC disconnect and battery switching for charging and discharging were realized for the converter. We have proposed a working algorithm for the optimization of the battery control in solar photovoltaic systems where the micro controller-based charge controller enhanced battery life by monitoring the temperature and controlling the charging voltages to suit float charging voltages for the specific temperatures. Two battery banks were used and decisions were made on loading and uploading based on battery state of charge and battery temperature. The work was successful and the proposed model is cost effective and feasible.