Application Research of DSP Technology in Numerical Control Technology

A design scheme based on DSP is presented.

In addition to increasing the manufacturing industry as a basic industry of the whole industry, the level of manufacturing is an important indicator to measure the degree of industrial development of a country. Vigorously developing advanced manufacturing technology has become the most important technology strategy of all countries in the world. one. CNC technology is the foundation of advanced manufacturing technologies such as flexible manufacturing systems, computer integrated manufacturing systems, and more. CNC machine tools are at the foundation and core position in the modern manufacturing system. DSP has data throughput of up to tens of MTPS and instruction cycle as short as tens of nanoseconds. It is very suitable for high-speed data acquisition systems and real-time control systems with large data volumes. Therefore, DSP is very important in high-performance CNC systems. Value.

1DSP and its characteristics Digital signal processing (DSP) theory was born in the mid-1960s. In the early days of digital signal processing technology development, people only carried out algorithm research and processing system simulation and simulation on a general-purpose digital computer. In recent years, with the development of LSI technology, some large foreign companies such as Texas Instruments and AD have pushed the microprocessor-type digital signal processor (DSP) of the fi type, and some added the capacity of the on-chip RAM and the off-chip search. In addition to the address capability, the number and speed of the serial/parallel port are also increased, and the counting-timer, ADC, DAC, etc. are added, and the time for processing one instruction is increased to several tens of nanoseconds, and the data throughput capacity is up to 80 MTPS or more, which is very suitable for High-speed data acquisition system and real-time control system with large data volume. DSPs in foreign countries have been widely used in communications, remote sensing, voice and key processing, electronic measurement, automatic control and mode i identification. The reason why DSP has such a wide range of applications is mainly determined by its characteristics.

1.1 Harvard structure general-purpose microprocessor, using Feng. The Neumann structure, program instructions and data, shares a single storage space and a single address and data bus. In order to further improve the operation speed to meet the requirements of the real-time digital signal processing algorithm, the current DSP adopts a different structure from the general-purpose microprocessor, that is, gives up Feng. The Neumann structure is based on the Harvard structure. The so-called Harvard structure separates the program instructions from the data storage space, each with its own address and data bus. This allows processing instructions and data to operate in parallel, greatly improving processing efficiency.

1.2 Pipeline technology DSP mostly uses pipeline technology. Pipeline operations allow operations such as fetching, decoding, and execution to be performed in an overlapping manner, and instructions can be executed in a single machine cycle, greatly increasing processing speed. The Harvard architecture of DSP provides convenience for the use of flowing water technology.

1.3 Fast Computing Capabilities The DSP chip has a dedicated hardware multiplier that can be multiplied and added once in one instruction cycle.

1.4 High-speed data transmission capability Most of the new DSPs are equipped with a separate DMA bus and its controller for parallel data transfer without affecting or substantially affecting the DSP processing speed. This provides convenience for series and parallel connection between DSPs.

1.5 Good simulation development technology In order to facilitate user design and debugging, many DSPs set the TAG emulation interface and high-level language compiler on the chip. DSP also has simulation software, which can simulate program running, interrupt, timing, etc., which has great convenience and is very suitable for the needs of numerical control development.

The early DSP was mainly used as the operation of the control algorithm. Now the control function is also realized by the DSP. In the digital control system, the DSP can handle all the work.

For example, the DSP improves the noise and inaccurate signals by processing the input and feedback signals, which improves the operation of all sensors.

DSP can also improve system control performance in real-time applications of advanced algorithms. Many control algorithms including adaptive, multivariate optimization, learning, self-correcting neural network, genetic algorithms, and fuzzy logic all require DSP speed and performance. to fulfill. For many systems, some system parameters must be estimated before or during normal operation, and the DSP has sufficient capability to perform identification and parameter estimation while processing other tasks.

Many motor digital control systems include power signal conditioning and power factor correction. The system that controls the motor usually uses the PWM method to control the energy converter. The DSP has the function of generating PWM. The fast DSP can use advanced algorithms to improve the utilization of electric energy, and has fault diagnosis and protection functions.

Therefore, in the DSP-based CNC system, the DSP can complete complex interpolation calculation programs and speed control programs, especially in high-speed and high-precision multi-axis CNC machining, which requires the CNC system to feed back the axes in a very short time. The processing of position signals and the application of DSP to the development of high-performance CNC systems is a good strategy. At the same time, real-time monitoring and dynamic compensation of tool wear can be realized through the program, which effectively improves the performance and accuracy of the numerical control system. In addition, the DSP can also implement non-control functions, including communication with the host computer, digital filtering and data bus control protocols (such as SCSI).

2 DSP-based CNC system design The CNC system based on DSP on PC platform can enable users to continuously upgrade software and hardware with little cost, keep up with the development of CNC technology in a certain period of time, more importantly: it The advantages of fast DSP processing and easy-to-open design of the PC platform will surely become the mainstream of CNC technology development.

Since TI, AD and other companies have introduced DSP chips for motor control, and their performance has been continuously improved, in addition to directly using PMAC and other products, it is a lot of users to choose the appropriate DSP chip to design the CNC system according to specific needs and applications. It can reduce costs and be more targeted. At present, there are more and more researches on the design of DSP-based CNC systems in China, and they have begun to be applied in practice.

The design principle of the DSP-based CNC system is shown in Figure 1. The system adopts a dual CPU structure, and the host (industrial PC) and the DSP are directly connected through the bus, and both the host and the DSP can read and write memory, thereby achieving the purpose of communication. At the same time, the host CPU can directly read and write I/O signals and read feedback signals, and its hardware structure is as shown in 圄1. The DSP controller provides advanced computation and real-time control using a simple and unique computer architecture. The controller uses a 50MHz TMS320C40 chip to achieve high-speed operation. The DSP loops into the servo loop operation of the axis and processes the related I/O functions.

The main CPU and DSP can directly access the data bus for fast communication. The DSP board does not require the PC to be in communication at all times. The computer only needs to read or write the data bus when there is information transfer or command operation. And any motion commands can be temporarily stored in the DSP's memory for execution.

The scheme uses more large-scale integrated circuits (such as 8154, 8255, etc.) as peripheral devices. In order to further improve the stability and reliability of the system, DSP can also be used as the system control microprocessor, combined with large-scale programmable logic. Devices (CPLD or FPGA) are designed, because large-scale programmable logic devices can be defined as logic devices such as counters, comparators, and buffers, etc., and can integrate a large number of logic functions into a single IC, reducing system requirements. The number of components required, while providing greater flexibility in design. The structure of the system is shown in Figure 2, where the programmable logic device FPGA implements data buffering, position comparison, pulse generation, and I/O interface functions, while the DSP accesses the FPGA through the data bus and reads the buffer area. The data, the execution position and speed control algorithm, and the communication function with the host.

The status quo and key technologies of construction machinery automation Guo Hui (Hangzhou Zheda Chenguang Technology Co., Ltd., Hangzhou 310013, Zhejiang Province has a vision.

With the advancement of social economy, technology and the continuous improvement of human living standards, it is required to develop construction in a safe, efficient, labor-saving, high-quality and comfortable direction. Although traditional construction machinery can realize mechanized construction, to meet the above requirements to a certain extent, However, it is still manually operated. The labor intensity of the workers is relatively poor, the labor productivity is low, and the construction quality and construction safety depend to a large extent on various human factors. The construction industry belongs to the bitter, dirty and dangerous industries, which is second only to the extractive industry. The number of casualties caused by accidents is also high. Based on this, the call for automation of construction machinery is very strong.

1 Construction Machinery Automation Status The development and research work of construction machinery automation began in the 1950s. In 1978, Sakamoto developed the world's first construction robot, and the development pace accelerated after the 1980s. At present, the developed countries in the West and the Harbin Institute of Technology in China and the Shenyang Institute of Automation of the Chinese Academy of Sciences have achieved certain results in the development and research of construction machinery automation. The development and research work of the two countries is at a leading level.

The numerical control system is developing towards open, high-speed, high-precision, intelligent and networked. The digital signal processor (DSP) has powerful computing power, good I/O operation function, high processing speed, and Good real-time control performance, has a wide range of application prospects and advantages in the CNC system. It is foreseeable that the development of high-performance and high-precision CNC systems using DSP is the mainstream trend of CNC technology in the future.

(Finish)

Mask Making Machine