Design and implementation of DC DVR device based on supercapacitor energy storage

With the development of science and technology and the expansion of industrial scale, the electricity consumption of various sectors of the economy is increasing. More and more users are using high-tech equipment with good performance and high efficiency but sensitive to changes in power supply characteristics, such as robots, automated production lines, precision CNC machine tools, high-precision measuring instruments and computer information management systems. These systems and equipment are very sensitive to various interferences from the power grid. Any power quality problem may cause significant economic losses and bring adverse social impacts. Among the user complaints about power quality problems, more than 90% involve voltage sags; statistical data and case studies show that most of the factors causing abnormal operation or power outages of electrical equipment are also caused by voltage sags. Therefore, this paper mainly studies the problem of voltage sag control. For sensitive load equipment with rectifier inverter structure, a new idea of ​​using supercapacitors to control voltage sags is proposed.

1 Voltage sag

Voltage sag refers to an event in which the supply voltage suddenly drops in a short period of time. The International Electrotechnical Commission (IEC) defines voltage sag as a voltage drop to 90% to 1% of the rated value, while the Institute of Electrical and Electronics Engineers (IEEE) defines it as a drop to 90% to 10% of the rated value, with a typical duration of 0.5 to 30 cycles. Severe voltage sags will cause electrical equipment to stop working or cause the quality of the products produced to deteriorate, and the severity of the consequences varies depending on the characteristics of the electrical equipment.

The control of voltage sag is a complex project. Usually, auxiliary equipment is set up to enable the main equipment load to withstand frequent voltage sags. The supercapacitor voltage sag suppression device studied in this paper is such an auxiliary device. At present, the voltage sag control devices studied at home and abroad mainly include dynamic voltage restorers (DVRs) and uninterruptible power supplies (UPS) of AC systems. For devices with DC bus, if UPS compensation equipment is installed, the system has a low cost performance due to the characteristics of short UPS service life, small discharge current and long charging time; if AC system DVR and other devices are installed, the system main circuit has two inverter circuits, which not only reduces the system efficiency but also increases the cost. For equipment with rectifier inverter structure, we have developed a DC DVR device based on supercapacitor energy storage, combining a bidirectional half-bridge DC-DC converter with a supercapacitor, and controlling the charging and discharging of the supercapacitor through a dual closed-loop method. When the system voltage sags, the DC bus voltage supporting the sensitive load is used to achieve the purpose of voltage sag control (Figure 1).

　　Figure 1 Main circuit of voltage sag control system

2 Supercapacitor Energy Storage

Supercapacitors, also known as electrochemical capacitors, are a new type of large-capacity energy storage element that utilizes the double-layer principle, uses new materials and new processes, has performance between capacitors and batteries, has a large capacitance density, and has excellent pulse charge and discharge performance. The commonly used double-layer capacitor structure is shown in Figure 2. The two inactive porous plates suspended in the electrolyte are electrodes. The positive plate attracts negative ions in the electrolyte, and the negative plate attracts positive ions in the electrolyte, thus forming a double-layer capacitor on the surface of the two electrodes. Its capacity is related to factors such as the surface area of ​​the electrodes and the distance between the plates.

　　Figure 2 Structure of a double-layer capacitor

Unlike conventional capacitors used for energy storage, supercapacitors can reach farad or even kilofarad levels. They have both the high energy density characteristics of rechargeable batteries and the high power density characteristics of capacitors. They are efficient, practical, and green energy storage devices. Table 1 shows the performance comparison of supercapacitors, energy storage capacitors, and batteries. Compared with ordinary capacitors and batteries, supercapacitors are not only pollution-free, maintenance-free, and have obvious environmental benefits, but also have the following advantages:

(1) High power density.

The power density of supercapacitors can reach about 10 kW/kg, which is ten to a hundred times that of batteries. They can release hundreds to thousands of amperes of current in a short period of time, making them very suitable for situations where high power needs to be output in a short period of time.

(2) Fast charging speed.

Supercapacitor charging and discharging is a physical process of double-layer charging and discharging or a fast and reversible electrochemical process on the surface of electrode materials. It can be charged with high current and completed within tens of seconds to minutes. Under the current technical level, it takes several hours to charge a battery, and even with fast charging, it takes tens of minutes.

(3) Long service life.

The electrochemical reaction that occurs during the charging and discharging process of supercapacitors is highly reversible. The theoretical number of charge and discharge cycles is infinite, and the actual number can reach 100,000 times, which is 10 to 100 times longer than the life of a battery.

(4) Excellent low temperature performance.

Most of the charge transfer that occurs during the charging and discharging process of supercapacitors occurs on the surface of the electrode active material, so the amount of capacity decay as the temperature decreases is very small; while the capacity decay of batteries at low temperatures can be as high as 70%.

Power quality problems often have the characteristics of high occurrence rate and short duration, so the use of supercapacitors as energy storage devices for rapid compensation is an ideal technical solution.

　　Table 1 Performance comparison of three electrochemical energy storage components

3 Main circuit and working principle of bidirectional DC-DC converter

The main circuit structure of the bidirectional DC-DC converter is shown in Figure 3. By controlling switches T1 and T2, the purpose of bidirectional DC boost and buck is achieved. In boost operation, T2 is activated and T1 is turned off, and the converter works in the Boost state; when T1 is activated and T2 is turned off, the converter works in the Buck state to achieve the buck function.

　　Figure 3 Bidirectional DC-DC converter main circuit

3.1 Boost Mode

The switch T2 is in constant pulse width modulation mode, and the equivalent circuit of the bidirectional DC-DC converter main circuit in Boost mode is shown in Figure 4. When T2 is turned on (Figure 4 (a)), the power supply v2 charges the inductor L, and the electrical energy is converted into magnetic energy and stored in L, while the capacitor C2 supplies power to v1; when T2 is turned off (Figure 4 (b)), the inductor L releases magnetic energy to supply power to v1. The energy storage effect of the inductor L can pump up the voltage, and after voltage stabilization by the capacitor C2, the output voltage can be higher than the input voltage.

　　Figure 4 Equivalent circuit in Boost mode

3.2 Buck Mode

The switch T1 is in constant pulse width modulation mode, and the equivalent circuit of the main circuit of the bidirectional DC-DC converter in Buck mode is shown in Figure 5. When T1 is turned on (Figure 5 (a)), v1 charges v2 through the inductor L, and part of the electrical energy is converted into magnetic energy and stored in L; when T1 is turned off (Figure 5 (b)), the magnetic energy stored in the inductor L is converted into electrical energy and charges v2 through the diode. The current flow direction in Buck mode is opposite to that in Boost mode.

　　Figure 5 Equivalent circuit in Buck mode

4 Supercapacitor charging and discharging control strategy

According to the characteristics of supercapacitors, this paper proposes a time-sharing control strategy of constant current charging and double closed-loop discharging.

4.1 Supercapacitor charging control

The DC bus operates within the normal voltage range. When the voltage of the supercapacitor array is lower than the rated operating voltage, the supercapacitor is charged. The charging control block diagram is shown in Figure 6. By comparing the hysteresis loop of the actual charging current with the reference charging current and limiting the maximum switching frequency, a signal is generated to control constant current charging. Constant current charging is beneficial to the protection of the energy storage device and has a faster dynamic response.

　　Figure 6 Supercapacitor charging control block diagram

4.2 Supercapacitor discharge control

The supercapacitor discharge control system adopts a dual closed-loop structure with a voltage outer loop and a current inner loop (Figure 7). The voltage loop is used to calculate the voltage deviation, and then the current loop reference value is calculated; the current loop obtains the appropriate compensation current based on the reference value, and the compensation value is obtained through transfer function transformation. In Figure 7:

Vref is the given voltage control quantity, Kv is the voltage feedback amplification factor, Ki is the current feedback amplification factor, Gvd is the control voltage of the S domain, Gid is the control current of the S domain, is the duty cycle disturbance, and is the high-voltage side output voltage disturbance.

　　Figure 7: Double closed-loop control structure block diagram.

For the Boost mode working state, the state equation can be obtained using the state space averaging method:

In the formula: v1--output voltage on the high-voltage side; v2--input voltage on the low-voltage side; α--time coefficient, equivalent to duty cycle, α = ton÷(toff+ton); iL--inductor current; R--current limiting resistor; L--charge and discharge inductance; C--supercapacitor capacity; r1--capacitor internal resistance.

Applying a small signal disturbance to the state equation gives an instantaneous value:

Where: V1 is the steady-state value of the high-voltage side output voltage; V2 is the steady-state value of the low-voltage side input voltage; iL^ is the inductor current disturbance; v2^ is the low-voltage side input voltage disturbance; D is the static duty cycle; d is the dynamic duty cycle.

Substituting equation (2) into equation (1), we get the steady-state equation:

By interfering with the state space average equation, the transfer functions of the control voltage (Equation (4)) and the control current (Equation (5)) in the S domain can be obtained:

　　Where: D′=1-D.

The small signal transfer function of the disturbance voltage and current in the S domain is as follows:

5 Simulation study

In order to verify the parameters and control strategy, 200 2.7 V/2 700 F double-layer capacitors were selected and connected in series to form a supercapacitor array, and a simulation experiment was carried out using Matlab/Simulink software (Figure 8).

　　Figure 8 Simulation model structure diagram

The system uses a resistive load, and the parameters are as follows: system phase voltage E=220 V; supercapacitor array capacitance CS=13.5 F, r=0.2Ω, charge and discharge inductance L=1 mH, operating voltage range 300~530 V, maximum output power 4 kW; simulation running time is 10 s. When the DC bus operating voltage is normal and the supercapacitor voltage is lower than the operating voltage, the bus charges the supercapacitor (Figure 9); when the DC bus voltage is lower than the lower limit of the system operating voltage, the supercapacitor discharges (Figure 10).

The power supply voltage of the device is 380 V, and the DC bus voltage experiences a voltage sag of 80% at 1 second. The simulation waveforms of the bus voltage before and after the supercapacitor voltage sag suppression device is connected to the DC bus are shown in Figures 11 and 12.

　　Figure 9 Supercapacitor charging control diagram

　　Figure 10 Supercapacitor discharge control block diagram

　　Figure 11 Waveform when no suppression device is added and the DC bus voltage drops by 80%

　　Figure 12 Waveform when suppression device is added and voltage drops by 80%

At the moment of 1 second, a voltage sag of 20% occurs on the DC bus. The simulation waveforms of the bus voltage before and after the supercapacitor voltage sag suppression device is connected to the DC bus are shown in FIGS. 13 and 14 .

　　Figure 13. Waveform when voltage drops by 20% without suppression device.

　　Figure 14: Waveform when suppression device is added and voltage drops by 20%.

The voltage sags in the above simulations are all three-phase voltage sags. When single-phase and two-phase sags occur, the effective value of the voltage on the DC bus is lower than that of the three-phase voltage. Therefore, this article does not conduct simulation introduction.

6 Experimental verification

The experimental design is to use the changes in the DC bus voltage without switching and switching the suppression device when a voltage sag occurs as a set of controls to verify the feasibility of the device. The supercapacitor uses a laboratory supercapacitor module, which is composed of 200 2.7 V/2 700 F double-layer capacitors connected in series; the load is a 7.5 kW electric furnace, and the experimental circuit structure is shown in Figure 15.

　　Figure 15 Experimental circuit

By simulating disturbance, the DC bus voltage drops by 80%, and the voltage drops from 510 V to 200 V (Figure 16). Figure 17 shows the DC bus voltage waveform after the supercapacitor voltage sag suppression device is connected in parallel to the DC bus.

　　Figure 16: Waveform when there is no suppression device and the DC bus voltage temporarily drops by 80%.

　　Figure 17: Waveform when suppression device is added and voltage drops by 80%.

Figure 18 shows the DC bus voltage waveform when a 20% voltage sag occurs (i.e., the DC bus drops from 510 V to about 400 V). After the supercapacitor voltage sag suppression device is connected in parallel, the DC bus voltage is well supported, and its voltage waveform is shown in Figure 19.

　　Figure 18: Waveform when voltage drops by 20% without suppression device.

　　Figure 19: Waveform when suppression device is added and voltage drops by 20%.

It can be seen from the above two groups of comparative experiments that when a voltage sag occurs on the DC bus, after the supercapacitor voltage sag suppression device is incorporated, the sag suppression effect is very obvious, the waveform is relatively stable, the response time is about 10 ms and there is no large fluctuation, which proves that the device can effectively suppress the voltage sag of the DC bus.

7 Conclusion

For equipment with a rectifier-inverter structure (i.e., with a DC bus), this paper proposes to use a bidirectional half-bridge DC-DC structure combined with a supercapacitor to control the voltage sag problem, studies its PWM control method, and proposes a dual closed-loop time-sharing control strategy for charging constant current and discharging combined with the characteristics of supercapacitor charging and discharging current. The response speed and suppression accuracy of the algorithm are verified through simulation; combined with the simulation results, an experimental circuit is built, and the performance of the device is verified (without considering the change of AC load voltage and the performance of bidirectional DC-DC at high power). As a new type of voltage sag control device based on power electronics technology, the supercapacitor voltage sag suppression device has a very broad application prospect.