Research on the application of the hottest dual pr

Research on the application of dual processors in substation monitoring

Abstract: This paper analyzes the development requirements of substation monitoring technology, and takes engineering application as the starting point, preliminarily discusses the needs of substation monitoring unit for dsp+mcu dual processor mode; The technical background of DSP and MCU is introduced; The data communication mode between DSP and MCU is described; An implementation example of high-performance substation monitoring unit based on dsp+mcu dual processor mode is presented; It also points out the development trend of DSP and MCU towards integration

key words: substation monitoring; DSP； Data communication; Embedded Ethernet

0 introduction

with the continuous improvement of substation automation, higher requirements have been put forward for the design of monitoring unit for the purpose of on-site monitoring and control: in addition to the large increase in the amount of monitoring and control, the real-time data acquisition, the rapidity of control operation, the diversity of analysis means, the complexity of analysis algorithm The flexibility of communication mode has put forward more stringent requirements. In the face of big data processing capacity, complex analysis algorithms and advanced communication methods, the hardware platform constructed with a single DSP or MCU processor is obviously inadequate. Adopting dsp+mcu dual processor mode, using the strong data processing ability of DSP chip to realize real-time data acquisition, analysis and calculation, plus a MCU with strong communication processing ability to meet the requirements of the field for communication technology, is a solution that has the best of both worlds

1 technical background

1.1 substation monitoring

the main task of substation monitoring is to collect the analog quantities and state quantities representing the operation state of the power system in the power plant or substation, transmit these analog quantities and state quantities to the dispatching center, and execute the control and regulation commands issued by the dispatching center. The monitoring unit is the front i/o module of the whole monitoring system, which is directly connected with the primary equipment on site. It is the core of data acquisition, processing and control operation. Therefore, its performance directly determines the performance index of the whole system. The early monitoring unit is composed of some discrete components, which can process little information and have extremely simple functions. With the application of microprocessor technology, the monitoring unit has also developed into an embedded system with single chip microcomputer as the core. However, in the case of less data acquisition and processing, low accuracy requirements, and less strict requirements on the real-time and progressiveness of communication, it can be achieved by using a single-chip low bit 8-bit or 16 bit MCU

with the development of substation automation technology and the improvement of analysis algorithm and analysis means, more and more users, especially some special users (such as high-voltage, ultra-high voltage substations, electrified railway traction substations, large enterprise substations, etc.) have higher and higher requirements for the design of substation monitoring, such as being able to analyze high-order harmonics in real time, calculate sequence components to analyze power quality It can carry out logic locking of control operation, which involves high-speed data sampling and processing; In terms of communication, the requirements for real-time, reliability and progressiveness of communication are becoming higher and higher. Embedded Ethernet technology and dual fieldbus technology have also been adopted in substation monitoring units. In this context, using dsp+mcu dual processor mode can better solve the contradiction between complex data processing and advanced communication technology

1.2 characteristics of DSP chip

dsp, also known as digital signal processor, is a microprocessor especially suitable for digital signal processing operations. Its main application is to realize various digital signal algorithm processing in real time and quickly. Its main technical characteristics:

a. the hardware adopts multi bus Harvard structure, which improves the data processing ability and speed

b. instruction execution adopts flow operation, which has high instruction execution speed

c. DSP generally includes multiple processing units, such as arithmetic logic operation unit, auxiliary register operation unit, accumulator and hardware multiplier, which can calculate simultaneously in an instruction cycle

d. there is an independent DMA bus controller inside, which can realize the parallel work of program execution and data transmission

e. it has multi processor interface, which provides a technical basis for using embedded subsystems to realize large and complex embedded systems

therefore, DSP is particularly suitable for the occasions of large capacity and high complex data processing. In substation monitoring, DSP chip is the core of data acquisition, calculation and processing of monitoring unit

1.3 function positioning and selection of MCU

mcu is embedded microprocessor. With the development of semiconductor technology and the improvement of system design level, MCU continues to produce new changes and progress. The performance gap between MCU and microcomputer is getting smaller and smaller, and the integration is getting higher and higher, and the power consumption is getting lower and lower

in the MCU family, it is really a headache for designers to make the best choice. According to the role of DSP and MCU in substation monitoring unit, MCU focuses on processing communication tasks. For occasions where dual processor mode is adopted and DSP chip has undertaken most of the data calculation and processing, the selection of MCU should be considered from the following aspects:

a. the communication mode to be adopted by the system

the selection of MCU depends on the communication mode adopted by the monitoring unit to a certain extent. If the system adopts can bus interconnection, a MCU chip with CAN bus interface can be selected; If embedded Ethernet technology is adopted, MCU chip with Ethernet interface can be selected. Although the required communication mode can be realized by expanding the protocol chip, it will greatly increase the software and hardware overhead of the system

b. whether the resources of MCU can meet the requirements and whether the peripheral interface is convenient

according to the functional positioning of the monitoring unit, review the resources of MCU: including bus bits (8 bits, 16 bits, 32 bits), main frequency, instruction cycle, addressing space, interrupt system, timer, peripheral interface, etc., so as to judge whether it can meet the design needs and leave appropriate margin for system upgrading

c. whether the development tools are complete and the programming is convenient

the quality of the development environment directly affects the product development cycle. Whether there is the support of embedded real-time multitasking operating system, whether it supports high-level language programming, and whether it can carry out real-time simulation and debugging on the target board are all factors that need to be considered

d. whether the price is reasonable

finally, it depends on whether the price of MCU chip is within the allowable range, whether the supply of goods is normal in the short term, whether the price will fluctuate greatly, etc

2 data interaction between DSP and MCU

in the intelligent unit with DSP + MCU dual processors as the core, the data communication between them is an important link in the system design, which directly affects the efficiency and reliability of data transmission. There are three common data interaction methods:

2.1 using dual port RAM

using dual port RAM to solve the problem of data communication between dual processors is the most commonly used method, Its implementation block diagram is shown in Figure 1:

dual port RAM has two sets of independent data, address and control buses, which can interface with DSP and MCU respectively. Through this public access space outside the chip, MCU and DSP can easily interact with each other

advantages: it is convenient to realize; Good real-time performance

disadvantages: increased hardware cost; It takes up limited space of printed board; If the control signal (especially the busy signal of two tests) is not processed well, it is easy to reduce the reliability of data transmission

2.2 adopt serial communication

at present, most MCU and DSP chips have strong serial communication ability, and can work at a high communication rate. Using serial interface is also a good choice. Serial communication can adopt synchronous, asynchronous, SPI and other methods. Figure 2 shows the schematic diagram of communication between MCU and DSP through spi[1]:

MCU is the main device of SPI, which controls the initiative of communication and generates chip selection and clock signals; DSP is the slave device of SPI, which passively receives the host command and puts the information required by the host on the sin pin of the host; In this master-slave structure, the slave has no active voice

advantages: simple interface, only four signal lines are needed

disadvantages: poor real-time performance; Occupy processor resources, such as the processors on both sides do not have a dedicated communication coprocessor or enough FIFO, which will frequently generate receive and send interrupts in the communication process; A certain communication protocol needs to be loaded, so the software implementation is more complex

2.3 HPI [2]

many DSPs of tms320c5000 and TMS320C6000 series of TI company contain HPI (host interface), which is used to interface with main equipment or main processor. The external host is the master of HPI. It can directly access the storage space of DSP through HPI, including the memory image register

hpi is mainly composed of the following five parts: HPI memory (daram), which is mainly used to transmit data with the host; HPI address register, which stores the address of the host currently addressing HPI storage unit; HPI data latch is used to store the data that the host is about to read and write; HPI other regions stable control deposit the influencer that China prohibits the import of recycled materials to foreign countries, which is used for the control operation between DSP or mainframe; HPI control logic is used to process interface signals between HPI and host

when DSP exchanges information with the host, HPI is a peripheral device of the host. DSP and host can operate HPI in the common addressing mode. Both host and DSP can address HPI memory. The addressing of asynchronous host can be resynchronized within HPI. If the cycle of DSP conflicts with that of the host, the host has addressing priority and the DSP waits for a cycle

Figure 3 shows the block diagram of HPI interface between MCU and DSP: MCU acts as the host of HPI. It is worth noting that DSP can interrupt MCU in the way of hard interrupt through the operation of control register to realize the rapid response of MCU to DSP. Similarly, MCU can also apply for the rapid response of DSP in the way of internal interrupt through the operation of control register

3 implementation of hardware platform of monitoring unit based on dsp+mcu

in view of the needs of some users for high-performance monitoring unit, the following is the design block diagram of dual processor monitoring unit hardware platform based on TI's high-performance DSP chip TMS320VC5402 and Motorola's high-performance MCU chip mcf5272[3], as shown in Figure 4:

3.1 introduction of main chips:

in this example, The analog-to-digital converter adopts ADS8364 chip newly launched by TI company: 16 bit accuracy, 6-channel synchronous sampling, sampling rate of 250kHz, parallel interface. This ad chip has simple interface with the processor, convenient control, and can simultaneously collect all the voltage and current of a line, which provides conditions for subsequent advanced analysis and calculation

the digital signal processor selects a 16 bit fixed-point DSP chip TMS320VC5402 from the tms320c5000 series of TI company in the United States. It has been widely used in the fields of communication and information systems, signal and signal processing and automatic control. In addition to the characteristics of other DSP chips, it also has an 8-bit HPI interface, and the instruction execution speed is up to 100mips. Its relationship with M