Image acquisition system based on solar power supply

    As the energy crisis intensifies, the use of new energy has become a long-term development strategy for all countries. Photovoltaic energy, with its renewable and pollution-free advantages, is recognized as one of the effective energy sources to solve the energy shortage in the future [1-2]. In situations where power supply is insufficient and image monitoring is required, using solar energy to power monitoring equipment not only reduces the cost of setting up the power grid, but also improves energy utilization [3].
    In recent years, various MPPT control algorithms have emerged in the maximum power point tracking (MPPT) strategy of solar charging. References [4-6] mention that the commonly used constant voltage method CV (Constant Voltage) and perturbation and observation method P&O (Perturbation and Observation) have the advantages of simple control ideas and easy implementation, but they also have their own shortcomings.
    In view of the shortcomings of the perturbation observation method, which causes power oscillation and misjudgment due to rapid disturbances when the external environment or load changes suddenly, this paper proposes an MPPT control method combining the fixed voltage method with the perturbation observation method [7].
    In the image acquisition part, due to the advantages of CMOS image sensors such as low price, light weight and low power consumption, the CMOS color image sensor chip OV7620 developed by Omnivision of the United States is used for image acquisition design [8], and an image acquisition design scheme based on solar power supply is given.
1 System Architecture
    The structural block diagram of the image acquisition system based on solar photovoltaic power supply is shown in Figure 1. The system mainly includes solar panels, lightning protection circuits, voltage/current acquisition circuits, BUCK charging circuits, drive circuits, batteries, AVR microcontrollers, OV7620 camera circuits and serial port circuits.
    This paper focuses on the implementation of the MPPT control algorithm in the BUCK charging circuit and the implementation of the image acquisition system based on solar photovoltaic charging. The core controller AVR of the system adopts the high-performance, low-power ATmega16A microcontroller.

2 System Hardware Design
2.1 Photovoltaic Battery Charging Control
    The AVR microcontroller first performs A/D conversion on the collected charging voltage and current signals of the solar panel, and the obtained digital signals are processed by the controller using the MPPT algorithm.
    The switch tube and freewheeling diode in the BUCK circuit are designed to have similar structures, and a driver chip with internal dead zone generation is used to drive and amplify the signal.
    To prevent transient high voltage from damaging the battery, two power buffer chips MBR20100 with high surge capability are added before the output voltage is connected to the battery. Figure 2 is the hardware schematic diagram of the BUCK charging circuit of the photovoltaic panel to the battery.

2.2 Image acquisition circuit design
    The OV7620 camera performs anti-aliasing filtering, amplification, A/D conversion, and image windowing on the captured analog image, thereby converting it into digital image data. The controller acquires digital image data according to the timing of image acquisition through several synchronization signals.
    In order to minimize the ground interference between the analog circuit and the digital circuit, magnetic beads are used to isolate the analog ground and the digital ground in the design. The schematic diagram of the image acquisition and main controller hardware circuit is shown in Figure 3.
3 System software design
3.1 System software flow
    The system software flow chart is shown in Figure 4.

    The system is started by the host computer command and collects the battery panel voltage data and digital image data in real time. When the battery panel voltage is detected to be greater than 10 V, the battery panel voltage is monitored in real time after a delay of 3 minutes; if the battery panel voltage is still greater than 10 V after 3 minutes, it enters the charging state. When the battery voltage is less than 14.7 V, the current is started to collect the voltage/current signal and adjust the output pulse width signal according to the signal through the maximum power tracking algorithm combining the fixed voltage method and the two-way perturbation observation method; when the battery voltage is detected to be greater than 14.7 V, it enters the floating charge state to prevent overcharging.
    The initialization of the camera uses the three-phase write cycle of the I2C protocol on the AVR two-wire serial interface TWI to configure the function register of 0V7620. By configuring the corresponding registers, OV7620 can work in different modes. For example, to set OV7620 to 16 bit data format, 320×240 output format and continuous scan mode, the following three functions are required:
    I2C_WritePhase(0x42,0x13,0x00);
    I2C_WritePhase(0x42,0x14,0x24);
    I2C_WritePhase(0x42,0x28,0x20).
    The process of writing registers is mainly three-phase transmission, first the write enable instruction 0x42, then the address of the target register, and finally the data to be written.
    The acquisition of image data is completed according to the coordinated action of the synchronization signals VSYNC, HREF, and PCLK.
    The host computer uses VC++ software to write the image acquisition interface, sends the start command through the serial port and starts to receive 320×240 image data.
3.2 MPPT control strategy
    When the output power of the solar panel reaches the maximum value, the corresponding voltage is the maximum power point output voltage. The fixed voltage method is to stabilize the output voltage at a fixed value. The output power of the solar panel will change with the change of light intensity and temperature, so the control accuracy of this method is low. The
    perturbation observation method is to determine whether to increase or decrease the output voltage by comparing the output power of the solar panel this time with the last time, thereby realizing MPPT. It has a disadvantage that in the process of power tracking, factors such as changes in light intensity and rapid disturbances may cause power oscillation and malfunction, so that the MPPT control cannot achieve the expected effect.
    To overcome the above shortcomings, this paper proposes a new MPPT control method. The system first uses the fixed voltage method to lock the output power of the solar panel to the vicinity of the maximum power point (output voltage within 30 V ~ 38 V) to ensure the rapidity of tracking. On this basis, the bidirectional perturbation observation method with a small step size is used to track the maximum power point of the solar panel, further improving the utilization efficiency of the photovoltaic array.
    Different from the general MPPT algorithm, when the external environment or load changes suddenly, the MPPT control is realized by the fixed voltage method, thus effectively avoiding the false operation caused by external factors. The perturbation observation method mainly optimizes the steady-state characteristics near the maximum power point. The perturbation step size can be much smaller than the step size of the general perturbation observation method, and gradually decreases in the direction close to the maximum power point. This can not only improve the control accuracy, but also effectively reduce the power oscillation phenomenon of the system near the maximum power point. The flow chart of the MPPT algorithm used in this paper is shown in Figure 5.
    In the bidirectional perturbation observation method, the system collects the voltage and current data of the solar panel in real time, calculates the power value (Pi or Pj) after bidirectional perturbation (PWM+△PWM or PWM-△PWM), and compares it with the power value Pnow before the perturbation. The next perturbation direction is determined according to the comparison result: when Pi>Pnow, continue to perturb according to PWM+△PWM; when Pj>Pnow, continue to perturb according to PWM-△PWM; when Pnow>Pi and Pnow>Pj, adjust the perturbation step size △PWM=0.5△PWM. When the perturbation step size is smaller than the small amount ε set by the system, it means that the maximum power point of the solar panel has been found.
4 Experimental results
    According to the design method of the solar photovoltaic power supply image acquisition system, the system

    From the collected waveform images, the driving waveforms of the two MOS tubes are stable and well complemented. The alternating conduction of the two tubes can well complete the charging strategy set by the controller. When the output power of the solar panel reaches the maximum, the duty cycle no longer changes. At this time, the tested charging voltage is 30.5 V, the charging current is 1.92 A, the battery input voltage is 14.67 V, the input current is 3.64 A, and the efficiency reaches 91.18%. Therefore, the MPPT control algorithm adopted by the system is a feasible control strategy.
    This paper designs a solar power image acquisition system for image monitoring. The system adopts wired transmission in the transmission mode, which is suitable for short-distance transmission occasions. In the future, the SPI interface can be used to try remote image wireless transmission on this system. A new MPPT control strategy is proposed in the photovoltaic charging strategy, which greatly improves the stability and accuracy of photovoltaic power generation. Experiments have proved that the system works stably and reliably.
References
[1] MOUSAZADEH H, KEYHANI A, JAVADI A, et al. A review of principle and sun-tracking methods for maximizing solar systems output [J]. Renewable and Sustainable Energy Reviews, 2009 (13): 1800-1818.
[2] TO LED O OM, FILHO DO, DINIZ ASA C. Distributed photovoltaic generation and energy storage systems [J]. Renewable and Sustainable Energy Reviews, 2010, 4 (1): 506-511.
[3] ZHANG Qing. Research on camera monitoring system for wind-solar hybrid power supply [J]. Technology and Application, 2011 (8): 53-56.
[4] SALAS V, OLIAS E, BARRADO A, et al. Review of the maximum power point tracking algorithms for stand-alone photovoltaic system [J]. Solar Energy Materials & Solar Cells, 2006 (90): 1555-1578.
[5] Li Jing, Dou Wei, Xu Zhengguo, et al. Research on maximum power point tracking algorithm in photovoltaic power generation system [J]. Acta Energiae Solaris Sinica, 2007, 28(3): 268-273.
[6] Cui Yan, Cai Binghuang, Li Dayong, et al. Comparative study of MPPT control algorithm for solar photovoltaic system [J]. Acta Energiae Solaris Sinica, 2006, 27(6): 535-539.
[7] Xiong Yuansheng, Yu Li, Xu Jianming. Application of fixed voltage method combined with disturbance observation method in maximum power point tracking control of photovoltaic power generation [J]. Electric Power Automation Equipment, 2009, 29(6): 85-88.
[8] Lei Feilin, Liang Zhiyi. Design of acquisition system based on CMOS sensor OV7620 [J]. Electronic Measurement Technology, 2008, 31(12): 110-112.