Abstract : This paper introduces the structure and principle of an inverter power supply that can be paralleled, and uses Motorola's DSP56F805 digital signal processor as the control core, and gives the hardware and software design scheme. The experimental results show that the control of this system has achieved good results.
Keywords : inverter; parallel connection; digital signal processor
1 Introduction
The rapid development of information technology has placed higher and higher requirements on the capacity, performance and reliability of power supply systems, and has also promoted the continuous deepening of research on power electronics technology. It is recognized that the realization of large-capacity power supply by connecting multiple modules in parallel is one of the important directions for the development of power conversion technology today.
Parallel technology can also be used to achieve large-capacity inverter power supply. Since inverter power supplies often use new fully controlled power switching devices to form unit modules, the capacity of a single inverter power module is very limited due to the capacity of power switching devices. By connecting multiple modules in parallel to expand the capacity, not only can the advantages of new fully controlled power switching devices be fully utilized, the system volume can be reduced, and the noise can be reduced, but also the dynamic response speed of the system and the versatility of the inverter can be improved.
1.1 Principle of inverter power
parallel operation The parallel operation between AC power supplies is much more complicated than the parallel operation of DC power supplies. Due to the sinusoidal wave output, the following problems must be solved:
1) When two or more units are put into parallel operation, the frequency, phase, and amplitude of each other and the system must be consistent or less than the allowable error before they can be put into operation, otherwise it will cause system instability or circulation between the inverter units;
2) During the parallel operation, the output of each inverter unit must be consistent, otherwise, the accumulation of slight frequency differences will cause periodic changes in the system output amplitude and waveform distortion, and the different phases will make the output amplitude unstable;
3) High current sharing requirements, current sharing includes active and reactive current sharing, that is, the average distribution of power includes the average distribution of active power and reactive power;
4) Fault protection In addition to the internal fault protection of the unit, when the current sharing or synchronization is abnormal, the corresponding faulty inverter unit must be removed to ensure the stability of the system.
The key to solving the above problems is to solve the current sharing problem. In view of this, the active and reactive parallel control method is adopted.
This control method actually realizes parallel power deviation control. When the parallel inverter units have inconsistent output active power or reactive power, the phase and amplitude of the inverter unit output voltage are adjusted by detecting the active power or reactive power deviation value of the unit to ensure that the active power and reactive power output of each inverter unit are equal to achieve the purpose of current sharing. Figure 1 is a network model of two inverter units connected in parallel to supply power to the load. The output active power P1 and reactive power Q1 of inverter unit 1 are :
300)this.width=300" border=0 ncpop="Click to view the original image in a new window">
300)this.width=300" border=0 ncpop="Click to view the original image in a new window">
From equations (1) to (4), it can be seen that the size of active power mainly depends on the power angle δ, and the size of reactive power mainly depends on the output amplitudes E1 and E2 of the inverter unit . Therefore , the size of the output active power can be adjusted by adjusting the power angle δ, and the size of the reactive power can be adjusted by adjusting the amplitude of the inverter unit output voltage, so that the current of each output power module can be achieved.
1.2 Digital control of inverter power paralleling
The early microprocessors had limited computing speed and usually only had functions such as the generation of a given sine wave, controlling the switching of the inverter power supply, and realizing protection display. The control of the inverter, the core of the inverter power supply, still required the participation of analog circuits. With the emergence of DSP dedicated to motor control and the development of control theory, the control technology of inverter power supply has developed in the direction of full digitalization.
The inverter power supply adopts digital control, which has the following obvious advantages:
1) Each inverter unit module running in parallel adopts full digital control, which is easy to better carry out current sharing control and communication between modules, or implement complex current sharing control algorithms in the modules, so as to realize a high reliability and high redundancy inverter unit parallel operation system;
2) It is easy to adopt advanced control methods and intelligent control strategies, making the inverter power supply more intelligent and more perfect in performance;
3) Flexible control, convenient maintenance, good system consistency and low cost.
The control strategies of sinusoidal inverter power supply include PLD control, deadbeat control, fuzzy control, etc. For the design of high-performance inverter power supply, fuzzy controller has the following advantages:
1) The precise mathematical model of the controlled object is not required in the design process of fuzzy controller, and fuzzy controller has strong robustness and adaptability;
2) It only takes a small amount of processor time to search the fuzzy control table, so a higher sampling rate can be used to compensate for the deviation between fuzzy rules and actual experience.
2 System Overview
2.1 System Features
1) Based on DSP56F805 full digital design, fewer control components, high reliability and high stability;
2) High reliability SPWM design;
3) Adopt CAN bus technology, easy to install in parallel;
4) Can realize N+1 inverter unit parallel expansion;
5) Each inverter unit works independently, 100% current sharing;
6) Adopt unique control principle, "current sharing imbalance" ≤2%;
7) Can be hot-swapped with power on, easy to operate and maintain;
8) High output voltage accuracy, 220 (1±1%) V;
9) High output frequency accuracy, 50±0.001Hz;
10) Full LCD digital display, measurement, menu control operation, convenient for online real-time monitoring of system status;
11) Intelligent control, RS-232 standard interface, can easily realize local and remote centralized monitoring and management;
12) Full protection functions, including DC input polarity reverse connection protection, DC input voltage over-high and under-low protection, output voltage over-high protection, overload protection, short circuit protection, overheat protection, etc.
2.2 System Overview
2.2.1 Inverter Power Supply Parallel System
This inverter power supply is based on DA2000HP (2000VA) inverter unit, equipped with monitor, static switch, etc., to form a complete inverter power supply parallel system. The working principle block diagram of DA-HP inverter power supply parallel system is shown in Figure 2.
300)this.width=300" border=0>
When the system is working, each inverter unit DA2000HP will first perform self-test. When the input voltage, temperature and hardware are detected to be normal, synchronization and phase lock will be performed, and finally the inverter unit will send out AC voltage. When the input voltage is detected to be too low or too high, the temperature is too high or the hardware fails, the inverter unit will stop outputting. When the inverter unit is working normally, the voltage, current, phase and other parameters of the system are detected in real time through the CAN bus, and control is carried out in time to achieve the same phase and current sharing between the inverter units. At the same time, the CAN bus and the synchronous bus are connected in parallel to send the system parameters and status to the monitor in real time, and the monitor sends the system parameters and status to the microcomputer through the RS?232 interface. During the working process, if the inverter unit detects a fault, it will quickly perform "offline" processing, that is, the faulty inverter unit is separated from the system to ensure the stability of the system, and at the same time, an alarm signal and information are issued to remind the user to handle it in time.
The function of the static switch in the system is to ensure the uninterrupted power supply of the user's load. When the inverter system fails, the static switch will connect the mains in time, disconnect the inverter system, and the load will be powered by the mains.
2.2.2 Inverter unit
The DA2000HP inverter unit adopts DSP chip DSP56F805 and advanced digital signal processing (DSP) technology to realize digital control and management of the inverter unit's conversion, control, feedback, measurement, display, communication, etc. At the same time, advanced software technology is used to control and protect key circuits, minimize the number of components of the whole machine, reduce unstable factors caused by temperature, aging and other problems, and improve the stability and reliability of the inverter unit. The working principle
block diagram of the DA2000HP inverter unit is shown in Figure 3.
The main conversion circuit of the DA2000HP inverter unit adopts a high-reliability single-ended high-frequency power conversion circuit with a conversion frequency of 64kHz. The DC input is sent to the single-ended high-frequency power conversion circuit through the input filter, input circuit breaker, and input contactor. After conversion, the transformer secondary outputs a high-voltage sinusoidal modulated waveform. The high-frequency component of the high-voltage sinusoidal modulated waveform is filtered out by a high-frequency filter to obtain a 100Hz half-bridge sine wave. The 100Hz half-bridge sine wave is converted by a 50Hz IGBT inverter bridge to obtain a 50Hz220V pure sine wave. Finally, the 50Hz220V pure sine wave is sent to the load through the output contactor, output circuit breaker, and output filter.
In order to improve the reliability and load applicability of the inverter unit, a compensator and a loss device are added before the 50Hz IGBT inverter bridge.
300)this.width=300" border=0>
The algorithm used by the DA2000HP inverter unit is a fuzzy control algorithm, which takes the voltage error and current as input fuzzy variables to realize the fuzzy control of the inverter unit.
2.3 Introduction to DSP56F805
The digital signal processor DSP56F805 developed by Motorola has 16-bit high-speed fixed-point computing capabilities, and has both the flexible control functions and rich peripherals of a single-chip microcomputer (MCU) and the high-speed computing capabilities of a DSP. It is very suitable for power control, motor control, industrial control, instrument manufacturing and other fields. This type of digital signal processing chip has the following advantages:
1) Very high processing speed
- single instruction execution cycle is 25ns (when the operating frequency is 80MHz), that is, 40M instructions can be executed per second;
- Single cycle 16×16 parallel multiplication and accumulation;
2) Unique parallel structure
- Adopting Harvard structure, the storage units of program area and data area are separated, and the efficient 16-bit DSP56800DSP core;
- 3 internal address buses and 1 external address bus;
- 4 internal data buses and 1 external data bus;
3) Flexible programming
- Has a programming method similar to that of a single-chip microcomputer;
- Supports high-level C language programming;
- Easy to develop, flexible EVM board and rich SDK software package;
4) Highly integrated internal resources
- On-chip integrated flash memory (Flash) and RAM, including 31.5K×16-bit program Flash, 512×16-bit program RAM, 4K×16-bit data Flash, 2K×16-bit data RAM, 2K×16-bit boot Flash;
——2 independent PWM modules, each PWM module has 6 independently programmable PWM output pins, 3 current sensing sampling pins and 4 fault detection input pins, supporting center-aligned PWM and edge-aligned PWM working modes; ——2 12-
bit ADC modules that can work simultaneously, each ADC module contains 4 input pins, and the ADC module can work synchronously with the PWM module;
——14 independent input and output ports, 18 multiplexed input and output ports;
——1 CAN2.0 module;
——2 asynchronous serial ports (SCI) and 1 synchronous serial port (SPI);
——2 differential decoders;
——4 groups of counting timers;
——Built-in COP module to facilitate the completion of the watchdog function;
——2 external interrupt sources;
——Programmable PLL clock;
——JTAG/OnCE interface for easy debugging and production.
3 System Hardware
The system hardware circuit includes the main control unit, A/D circuit, PWM circuit, parallel and synchronization circuit, detection, control and display circuit, JTAG/OnCE circuit, RS-232, clock and power supply circuit, etc. The main control chip uses a DSP56F805 digital signal processor in a 144?pin LQFP package. The specific circuit is shown in Figure 4.
3.1 The hardware of the main control unit
is centered on the DSP56F805, making full use of its built-in functions such as A/D, PWM, internal Flash, and CAN to simplify the design.
When the system works normally, the PWMA0~PWMA1 pins output a pair of SPWM waveforms, which drive the single-ended conversion circuit power tube (MOSFET) through the isolation and drive circuit, and then step up the voltage through the main transformer. The secondary obtains a high-voltage SPWM sine modulation waveform, and obtains a pure 100Hz half-bridge sine wave through L and C filtering. The PWMA2~PWMA3 pins output a pair of PWM waveforms, which drive the power tube (IGBT) through the isolation and drive circuit to obtain a 50Hz220V pure sine wave. PWMA4 is used as a D/A converter, which is filtered into a DC signal and drives the loss device through the isolation and drive circuit. PWMB0~PWMB2 are used as output ports. According to the reactive power, appropriate capacitors are selected to drive the compensator through the isolation and drive circuit. The A/D circuit constantly detects parameters such as input voltage, output voltage, output current, and internal temperature. When one or more parameters exceed the software setting value, the DSP immediately shuts off the SPWM signal and sends out an alarm signal. In addition, FAULTA0 is used as an output overcurrent sampler. Once the FAULTA0 voltage exceeds the threshold, the DSP immediately shuts off the PWM output.
300)this.width=300" border=0>
300)this.width=300" border=0>
3.2A/D circuit
The ADC module of DSP56F805 has the following features:
1) 12-bit accuracy;
2) Simultaneous or continuous sampling mode;
3) In the simultaneous sampling mode, the conversion time of 8 channels is 26.5ADC clock cycles, that is, 26.5×0.2μs=5.3μs;
4) The ADC conversion can be triggered by the internal synchronization signal of PWM or the timer or the external signal.
In order to improve the conversion speed, this system adopts the simultaneous sampling mode and triggers the A/D conversion by the internal synchronization signal of PWMA. The pairing of the two ADC modules is as follows:
AN0 (100Hz current sampling) - AN4 (100Hz voltage sampling);
AN1 (output AC current sampling) - AN5 (output AC voltage sampling);
AN2 (input DC voltage sampling) - AN6 (absorption tube current sampling);
AN3 (temperature sampling) - AN7 (reference voltage 1.25V).
Since the ADC samples both DC and AC quantities, the two different quantities need to be processed separately.
The DC quantity (input DC voltage, temperature and reference voltage 1.25V) adopts the general digital filtering processing method, and the expression is as shown in formula (5).
300)this.width=300" border=0>
Wherein: X is the A/D sampling result; X(n-1) is the sampling result of the n-1th time;
X(n) is the sampling result of the nth time;
K1 and K2 are correction coefficients. The AC component includes 100Hz voltage, current, output voltage, current, and absorption tube current. Samples are taken 160 times in one cycle (10ms), and the relevant power values are calculated according to formulas (6) to (8). 300 )this.width=300" border=0> Wherein: S is the apparent power; P is the active power; Q is the reactive power; Ks and Kp are correction coefficients . 3.3PWM circuit DSP56F805PWM module has the following main features: 1) 3 sets of complementary PWM pairs or 6 independent PWM; 2) Adjustable dead zone; 3) Half-cycle reload capability; 4) 20mA output drive capability. The two PWM modules of this system work as follows: PWMA0, PWMA1 (SPWM0, SPWM1) work in complementary PWM pairs to generate SPWM modulation waves, with a carrier of 64kHz and a modulation wave of 100Hz; PWMA2, PWMA3 (PWM0, PWM1) work in software-controlled I/O to generate 50Hz square wave signals. Convert 100Hz half-wave to 50Hz full-wave; PWMA4 (D/A) works in D/A to generate a correction signal to adjust the absorption power tube according to the temperature or loss inside the machine; PWMB0, PWMB1, PWMB2 (F0~F2) work in software-controlled I/O to control the compensation circuit. 3.4 Parallel and Synchronous Circuit The synchronous circuit is completed by PA0 and PA1, where PA0 is the input pin to detect the 50Hz synchronization signal of the external (other inverter units), and PA1 is the output pin to send out the 50Hz synchronization signal of the local unit. When the system is powered on, the local unit first detects whether there is an external synchronization signal. If there is, the local unit tracks the external signal and sends out a synchronization signal. If not, it works on the synchronization signal of the local unit. Parallel operation is completed by CAN. The CAN module is responsible for collecting the status values of other inverter units (voltage, current, frequency, active power, reactive power, etc.) and sending its own status values. 3.5 Detection, control and display circuit 1) PD2 is set as input port, and the power supply starts only when the S1 switch is closed; 2) PD6 and PD7 are set as input ports, respectively detecting the input contactor status and the output circuit breaker status. The inverter unit works only when both are normal; 3) PB0~PB7, PD0, PD1, PE2 are LCD display control circuits, of which PE2 is the input port, which is the display menu button S2, PD0 and PD1 are the output ports, controlling the RS and E of the LCD, and PB0~PB7 are the output ports, sending signals to the LCD data ports DB0~DB7; since a 16×2-bit character LCD module is used, the threshold voltage of the LCD is 300)this.width=300" border=0> , which is in line with the logic of the DSP chip, so the DSP56F805 can directly drive the LCD without level conversion; 4) PD3~PD5, PE4~PE7 are output ports, respectively controlling related indicator lights and relays. 3.6 JTAG/OnCE circuit DSP56F805 provides JTAG/OnCE circuit, which can be convenient for users to write programs into the Flash memory on the chip, and also convenient for users to program, modify and upgrade software online. 3.7 RS-232, clock and power supply circuit DSP56F805 has two sets of SCI. This system uses SCI0 as RS232 interface. If RS232 is used as a communication port to connect to the PC when used alone, if it is used in parallel, this port is not used, and the main monitor is responsible for communicating with the PC. DSP56F805 has a PLL phase-locked loop clock unit, which can easily change the DSP clock through software programming. The main circuit of DSP56F805 is powered by +3.3V. In order to prevent noise interference from affecting the accuracy of A/D conversion, A/D uses an independent power supply system. If the external digital circuit has a +5V power supply system, the level conversion must be performed when necessary with the DSP interface.
To be continued
提升卡
变色卡
千斤顶
京公网安备 11010802033920号
