Design of digital intelligent aerospace power system based on Picosatellite

Micro-satellites, with their high functional density, advanced technical performance, and high flexibility during launch and operation, have gradually become an important development direction in the field of international aerospace technology research. According to the current international satellite classification method, micro-satellites weighing between 0.1 and 1 kg can be called pico-satellites [1]. For micro-satellites represented by pico-satellites, due to their very limited solar panel area and the complex and changing space environment, the satellite power system is required to have the characteristics of high efficiency, high energy density, and autonomous control, which is difficult to achieve with general industrial power supplies.

In view of the characteristics of the pico-satellite power supply system, this paper develops an intelligent and efficient digital power supply system. Its intelligent design is mainly reflected in: real-time acquisition, processing and analysis of important signals such as voltage and current at key nodes of the power supply system through a variety of measurement circuits, so as to keep track of important parameters such as energy input, storage and output of the power supply system and real-time efficiency; on the basis of data acquisition, through the processing of the microcontroller and its control software, reasonable control strategies such as peak power tracking (MPPT) and charge and discharge regulation (BCR/BDR) are adopted to control the working state of the power supply system and track the maximum input power point; according to different space mission requirements and energy interface parameters, the power supply operation experiment is flexibly carried out by adjusting the software; and the serial communication method is used to communicate with the host computer, which provides good conditions for the measurement and control of the satellite power supply system and the storage and transmission of data.

1 Hardware Design of Picosatellite Intelligent Power System

The intelligent power supply system of the pico-satellite is designed based on the topological structure of “solar array – power supply control system – battery pack” [2]. As the core part of the entire power supply system, the power supply control system mainly consists of the following parts: microcontroller unit, primary bus voltage regulation unit (i.e. peak power tracking unit), secondary bus voltage regulation unit (i.e. discharge regulation unit), charging regulation unit, voltage and current signal acquisition unit, signal processing unit, serial communication unit, etc.

The basic working process of the power control system is as follows: according to the preset space environment parameters, the solar array simulator forms the initial input of the power system; the initial input is adjusted by the primary bus voltage regulation unit to form a primary bus voltage of 7.2V~8.4V that matches the working voltage of the battery group, and at the same time completes the tracking and locking of the input peak power; the power supplied to the secondary bus is adjusted by the secondary bus regulator to provide two secondary bus voltages of 5V and 3.3V for the on-board load respectively; the voltage and current signal acquisition unit continuously collects the voltage and current signals of each key node such as the initial input, primary bus, battery group, secondary bus, etc., and sends them to the microcontroller unit for A/D conversion through the voltage follower, the first-order filter circuit and the multi-channel signal selection chip; according to the signals of each key node, the microcontroller sends control signals to the bus regulation units and the charging control unit at each level after further processing and analysis, and transmits data to the host computer through the serial communication unit.

1.1 Microcontroller unit

The microcontroller circuit is based on the ATmega8L microcontroller launched by ATMEL, and is equipped with peripheral devices such as the MAX 397 dual 8-channel analog multiplexer and the MAX 6129 reference voltage source, as shown in Figure 2. The ATmega8L microcontroller is a low-power CMOS 8-bit high-end microcontroller based on AVR RISC, with a high-speed operation processing capability close to 1 MIPS/MHZ. ATmega8L has 23 programmable multi-function I/O ports, eight-channel 10-bit A/D conversion and three-channel 16-bit PWM output functions, so it completes important functions such as 10-bit signal A/D conversion and processing, MPPT algorithm implementation, and 31.25KHz PWM control signal output in the system.

1.2 Primary bus voltage regulation unit (peak power tracking unit)

The primary bus voltage regulation unit circuit is based on the Boost DC/DC voltage conversion circuit, and a primary bus control switch composed of two MOSFETs is added, as shown in Figure 3. The Boost voltage conversion circuit consists of a MOSFET switch tube Q1, freewheeling diodes D3 and D4, an energy storage inductor L2 and a filter capacitor C13. The boost conversion ratio satisfies

M = Vout/Vin = 1/ (1-D) (1)

Since the primary bus output voltage Vout is clamped at the working voltage of the battery pack, that is, a certain value in the range of 7.2V to 8.4V, the input voltage (that is, the output voltage of the solar array) Vin can be adjusted by adjusting the duty cycle D of the PWM control signal issued by the microprocessor unit. On this basis, the peak power tracking (MPPT) algorithm is called to maximize the output power of the solar array.

1.3 Current and voltage signal acquisition unit

The signal acquisition unit is based on the MAX4373F current sensing amplifier and voltage-dividing precision resistors, which collect voltage and current signals at six nodes, including the initial input, primary bus, battery pack, and 5/3.3V secondary bus. The signal is sent to the integrated operational amplifier LM234 for voltage following, and then filtered out the ripple through the first-order RC filter circuit, and finally sent to the MAX397 for A/D conversion.

1.4 Charge Regulator Unit

The battery pack charging regulator is composed of an n-MOSFET and p-MOSFET electronic switch, and the specific structure is the same as the electronic switch on the right side of Figure 3. During the charging process, the MOSFET driver outputs a high-level signal, then the n-MOSFET IRF3205 is turned on, making the G pole voltage of the p-MOSFET IRF4905 approximately 0. At this time, the voltage between the S pole and the G pole of the IRF4905 is positive, making the IRF4905 turned on. When the battery pack reaches the full charge voltage, the microprocessor controls the electronic switch to turn off.

1.5 Secondary bus voltage regulation unit (discharge regulation unit)

Since the output voltage is a specific value, the secondary bus voltage regulation unit uses the MAX649 (5V output) and MAX651 (3.3V output) Buck-type DC/DC step-down conversion control chips. The MAX649 and MAX651 chips convert any primary bus voltage in the range of 4.0V to 16.5V to 3.3V and 5V respectively to meet the energy needs of various subsystems on the satellite. When the output current is in the range of 10mA to 1.5A, the chip power conversion efficiency can reach more than 90%.

The discharge regulator is also composed of an n-MOSFET and p-MOSFET electronic switch combination driven by a microcontroller unit.

1.6 Serial Communication Unit

The serial communication unit circuit is based on the dual-channel serial communication driver chip MAX232 and uses the serial communication standard EIA-RS-232C protocol. MAX232 converts the TTL level signal "logic 1 level +5V, logic 0 level 0V" output by the microcontroller into the host computer RS-232C signal "logic 1 level -5~-15V, logic 0 level +5~+15V".

2 Software and Algorithm Design of Picosatellite Intelligent Power System

2.1 Basic Process of Picosatellite Power System Control Software

The power system control software process is mainly based on the process of "signal inspection → PWM control signal adjustment → system operation parameter transmission → signal inspection again", and adds branch control functions such as charging control and discharging control in the process of "inspection → control → data transmission". The control software adopts a modular design concept and consists of a system initialization module, a multi-channel A/D conversion module, a digital filter module, a data analysis and control module, a serial communication module, etc. [3].

2.2 Conductivity increment MPPT algorithm based on fuzzy control logic

The intelligent power supply system of the pico-satellite mainly relies on the MPPT algorithm in the software to maximize its power. The principle of the MPPT algorithm is that under certain temperature and light intensity conditions, there is a nonlinear relationship between the output voltage and current of the solar array used in the satellite power supply. When the output voltage reaches a specific value Vmp, the product between it and the corresponding current value Imp reaches the maximum value, which is the peak output power point Pmp of the solar array.

At the peak power point, the differential of the output power to the output voltage is

dP/dV = d(VI)/dV = I+V dI/dV = 0 (2)

Further deduction shows that: -dI/dV = I/V (3)

Based on this relationship, a conductance increment MPPT algorithm based on fuzzy control logic is established.

Among them, V(n), V(n-1), I(n), I(n-1) are the output voltage and current values ​​of the solar array at the current moment and the previous moment respectively, D(n), D(n+1) are the duty cycle at the current moment and the next moment respectively, and △D is the duty cycle adjustment step. According to the collected current and voltage signals, the microprocessor unit continuously increases or decreases the PWM signal duty cycle, and uses the Boost voltage conversion circuit to adjust the output voltage of the solar array, so that the working point reaches the peak power point Pmp, and the satellite power system obtains the maximum output power.

Furthermore, fuzzy control logic is introduced on the basis of the basic algorithm to speed up the peak power tracking. The two input variables of the fuzzy logic controller are respectively taken as the current conductivity increment difference e(n) = -dI/dV- I/V and the duty cycle adjustment step △D(n), and the output variable is taken as the duty cycle adjustment step △D(n+1) at the next moment. Then the corresponding membership function and fuzzy rule base are established, which are omitted here. The simulation experiment shows that under the standard space environment conditions (AM0, 25℃), the conductivity increment MPPT algorithm after the introduction of fuzzy control logic reduces the time required for peak power tracking by more than 60%.

3 Conclusion

This paper develops an intelligent aerospace power system based on the characteristics of the pico-satellite power system. The power system uses the ATmega8L microcontroller as the core, collects and processes the signals of each key node of the power system in real time, and uses control strategies such as peak power tracking to control the working state of the system. The simulation experiment shows that the power system has good peak power tracking performance under standard space environment conditions (AM0, 25℃), the maximum input power reaches about 2.75W, and the overall efficiency of the power supply remains above 82%.

The innovation of this paper is as follows: Using ATmega8L single-chip microcomputer as the core controller, intelligent control methods such as real-time acquisition of operating parameters, system autonomous power tracking, charge and discharge regulation, and data communication between upper and lower computers are realized in the aerospace power system; a conductance increment MPPT algorithm based on fuzzy control logic is proposed to quickly track the input peak power of the power supply system.