Wind-solar inverter grid-connected system based on MSP430 and FPGA

In order to alleviate the energy problem, on the basis of being fully compatible with the existing power supply system, the system uses wind and solar energy to replenish electricity, and comes with the functions of quickly detecting islanding effects, fast grid connection and disconnection. The power circuit part of the system adopts a full-bridge topology for inversion, and the digital control system adopts an MCU + FPGA architecture. The frequency doubling of the external grid power is completed by all hardware, and then the FPGA dynamically adjusts the system output phase so that the output and the external grid power are in phase. The MCU completes the maximum power point tracking (MPPT) of the solar panel, the undervoltage detection of the power generation end, and the islanding effect detection. In view of the characteristics of the strong power of the power system and combined with the popular Internet of Things technology, the system has humanized the design of wireless detection functions, and users can understand the current system status through mobile phones, computers or handheld terminals. The creative design of this system can be used for multi-energy parallel power generation in power plants, and is also suitable for home use, allowing families to transform from electricity users to micro power plants, thereby greatly alleviating energy problems.

In the past two hundred years, human beings have used coal, oil and natural gas as energy, which has increased productivity by nearly 200 times. However, fossil energy is gradually depleting, and pollution is also serious. As energy problems become increasingly prominent, finding new green energy has become an urgent issue. Among the recognized green energy, solar energy and wind energy are the easiest to obtain and use efficiently.

This article focuses on solar energy and wind energy, and designs a simulation device for wind-solar grid-connected power generation. It can convert the DC voltage of solar or wind generators into AC power, detect the frequency and phase of external AC power, and dynamically adjust the waveform of its own AC power to make it have the same frequency and phase as the external power grid. The internal resistance of the generator was taken into consideration when designing the device. During the test, a 60 V DC regulated power supply was used to simulate an ideal solar panel or wind turbine, and a 30 Ω power resistor was connected in series at the power input stage to simulate the internal resistance of the power generation part.

The device is small in size, low in cost, easy to mass produce, has a friendly human-interface, and is equipped with input voltage monitoring, output overcurrent monitoring, real-time dynamic phase monitoring and other monitoring settings, which also makes the device have good safety performance. It can be widely used with a little modification.

1. Proposal demonstration

1.1 Main power circuit topology

Solution 1: Full-bridge inverter.

The full bridge is composed of 4 power switch tubes, which are divided into 2 groups, of which Q1 and Q4 are one group, and Q2 and Q3 are another group. The two groups are switched on and off alternately, and the output AC square wave voltage is passed through the LC low-pass filter to obtain the AC sinusoidal output voltage (see Figure 1). The output filter capacitor voltage of the full-bridge inverter can be measured continuously. The output of this circuit can obtain a good waveform after LC filtering.

Wind and solar inverter

Option 2: Dual Boost DC/AC single-stage conversion circuit topology.

The structure consists of two symmetrical Boost DC/DC conversion circuits with bidirectional current flow (see Figure 2). The load R is connected across the two capacitors, and the output of AC power frequency voltage is achieved on the load through the bidirectional flow of current on both sides. Switches M1 to M4 are all controllable switches composed of MOSFET and diodes, and energy can flow in both directions. Since the circuit works in a completely symmetrical state, it is particularly sensitive to the selection of L1 and L2. If they are asymmetrical, the output waveform will be distorted.

In the second scheme, the positive and negative half axes of the sine wave are completed by two filtering circuits, so it is difficult to complete the waveform distortion. However, in the first scheme, the same inductor is used for filtering, and the sine wave distortion after filtering is very small. Therefore, the first scheme is adopted.

1.2 Sine wave generation scheme

Solution 1: Use a dedicated SPWM chip to achieve inversion.

The peripheral circuit of the current SPWM dedicated chip is simple and easy to implement. However, it is difficult to complete the phase tracking and adjustment of the mains in this system. Therefore, this solution is not adopted.

Solution 2: Use FPGA to generate SPWM waveform.

The advantage of this solution is that it is easy to accurately and conveniently control the phase and amplitude of the output sine wave, and the peripheral circuit is simpler, more flexible and convenient. Compared with the first solution, it is more optimized, so this solution is selected.

1.3 Overall system design architecture

To summarize the above-selected solutions, the structure based on digital circuits and simple analog circuits is selected here. The high integration, high accuracy, high cost performance and high stability of digital circuits are fully combined with the characteristics of analog high power, which better realizes the design requirements. In addition, the wireless monitoring function is expanded, which more realistically reflects the actual application environment of this design and presents a more humanized design. The overall solution is shown in Figure 3.

2 Main circuit electrical appliance selection and parameter calculation

The main circuit of the system consists of a DC-AC converter circuit and a shaping and measuring circuit for the input/output waveform. In order to reduce losses and prevent reverse breakdown, the main switch tube is IRFB52N15 (rated current 60A, withstand voltage 150V, on-resistance 32MΩ). The inverter circuit using SPWM control contains a large number of high-frequency harmonics in the output SPWM wave. In addition to the dead zone set to prevent the upper and lower bridge arms from passing through, the switching time and the difference in power device parameters, the output voltage can only contain certain low-order harmonics. In order to ensure that the waveform distortion is as low as possible, an output filter must be used. The full bridge uses LC filtering, in which the inductive reactance XL=ωL=2πfL, and the capacitive reactance XC=1／(ωC)=1／(2πfC). Let ωL=1／(ωC) to obtain the corresponding cutoff frequency d.jpg. Let the fundamental wave of the inverter output voltage be f0, the lowest harmonic frequency fk, f0 电感对基波信号的阻抗很小，电容对基波的分流很小，即电感对基波信号的阻抗很小，电容对基波的分流很，即滤波器允许基波信号通过，而ωkl> >1/(ωkC), the inductor has a large impedance to the harmonic signal, and the capacitor has a large shunt to the harmonic signal, that is, the filter does not allow the harmonic signal to pass through the load. Generally, the filter cutoff frequency fc=(3~5)f0. In order to avoid excessive amplification of a certain harmonic, fc=4.5f0=1800 Hz is taken. The output power and output voltage of the inverter are used to obtain the load impedance RL. The nominal characteristic impedance of the filter is R=(0.5~0.8)RL, then Lf=R/(4πfC), Cf=Lf/R2=1/(2πfCR). In the actual circuit, L is 200 μH and C=470μF.

3 Control and Algorithm Design

The MCU of this system is MSP430 . The MSP430 series is an ultra-low power 16-bit single-chip microcomputer launched by TI. It has high cost performance, powerful functions, fast operation, and its working current is less than 1mA. It also has multiple low-power modes. This solution uses MSP430F2618 as the main control chip to monitor input current and voltage, protect against overcurrent and undervoltage, and recover after troubleshooting; sample output voltage and voltage tracking maximum power; and display current system status and output related data.

3.1 Maximum Power Tracking Algorithm

The maximum power point tracking algorithm can be roughly summarized into six types according to the judgment principle and implementation method: constant voltage and its improved algorithm, constant current and its improved algorithm, perturbation observation method, incremental conductance method, fuzzy logic control algorithm and neural network control algorithm.

The perturbation observation method is a relatively simple, practical and easy-to-implement method. Its idea is to periodically add a disturbance △V to the output voltage of the power supply, measure the output current and voltage of the power supply, and compare the output power P(t) at the sampling moment with the output power P(t-1) at the previous sampling moment; if P(t)>P(t-1), then add a disturbance in the same direction in the next cycle, otherwise change the direction of the disturbance, so as to gradually approach the maximum power point. However, the setting of the tracking step cannot take into account both the tracking accuracy and the response speed. Oscillating operation near the maximum power point will result in a certain power loss.

3.2 FPGA-based phase tracking

The sine fundamental wave signal of the SPWM signal generated in this system is the address inside the FPGA accumulated by 1 bit each time, and then the ROM storing the sine table in the FPGA is queried. The external reference sine signal and the sine waveform generated by the system itself are input into the FPGA after being shaped by the comparator. The new signal is obtained after passing through the XOR gate inside the FPGA. The new signal is high, indicating that there is still a phase difference between the two signals. At this time, the address accumulator inside the FPGA increases by 2 bits, that is, the phase of the sine wave generated by itself increases by a quantized value, until the result of the XOR of the two signals is completely low. Due to the high-speed operation of the FPGA, the entire phase tracking can be completed within two cycles, which can meet the requirements of market applications.

4 Conclusion

The system is based on MCU-FPGA and realizes the wind-solar inverter grid-connected system. The system makes full use of the calculation accuracy of the digital system to control the phase difference between the inverter waveform and the external grid power within 2°, and through maximum power tracking, the power generation efficiency of the solar panel or wind turbine is maximized. The system is low-cost, small in size, and user-friendly, which is convenient for direct mass market use in the future.