Data memory is Industry 40 One of the most pressing challenges faced. However, in view of Industry 40 and related memory solutions continue to evolve, and some memory-based trends are becoming apparent.
Globally, the convergence of modern digital technologies with industrial facilities is rapidly evolving, but not without its challenges. with Industry 4One of the main issues related to 0 is how to handle the dramatically increased data volume and throughput in a reliable, error-free manner.
Industry 40 The combination of massive amounts of data, extensive process automation, powerful artificial intelligence, and a change in the location of data analysis is changing the way many industries around the world do business.
I. Industry 40
Industry 40 is the universal label of the Fourth Industrial Revolution and includes several hallmarks such as the development and widespread adoption of process automation, extensive data collection, machine-to-machine communication, and enhanced data analytics. It also includes localized (machine-level) AI and edge decision-making instead of centralized decision-making.
2. Memory: Towards Industry 40 Challenges
Data is Industry 40 The most critical aspect, because the retrieval, processing, communication, storage, and availability of data are essential for Industry 4The implementation of 0 is critical. For example, consider how sensors generate massive amounts of data that need to be recorded and processed in real time. Sensor data interacts with three types of memory in the system through modern industrial controllers and field devices: flash, extended RAM, and data-logging RAM. On the road to Industry 40, memory is one of the challenges that needs to be solved.
3. Memory requirements for real-world applications
Many industrial systems use a variety of external memories for different functions. Common embedded systems on the factory floor include programmable logic controllers (PLCs), servo motor drives, CNC machines, and industrial robots. These systems typically include the following building blocks (see Figure 2), which include an application processor, multiple IO peripherals, wired and wireless connectivity, sensors and transmitters, and three different types of memory.
Figure 2Block diagram of an industrial system.
Industrial systems, such as PLCs, use external flash memory devices to store boots**, often with capacities up to 256 MB. These memories need to support features such as extremely fast read performance and execution-in-place (XIP) so that the system can execute directly from NOR flash without having to copy it to RAM.
Data logging RAM enables real-time IO and sensor data to be written at extremely fast speeds, while also capturing mission-critical data immediately in the event of a power outage. Most industrial systems must continuously and accurately record data over a 15-year service life while operating in harsh environments. Industry 40 relies on continuous data in order to prevent failures before they occur and minimize system downtime – a process known as sexual maintenance. RAM solutions for this type of memory need to reliably and accurately perform millions of reads and writes over the lifetime of the product. Also, keep in mind that harsh environments include extreme temperatures, which often pose a challenge for certain types of memory.
Internal processors, such as microcontrollers or FPGAs, may not have enough internal SRAM to execute complex algorithms and temporarily store system runtime data. Modern industrial systems use expansion RAM to augment limited internal RAM as high-speed scratch memory to run user applications seamlessly. An industrial robot or machine vision system consists of a display that uses extended RAM to buffer image frames for the graphics display. They also require optimized external memory interface IP and low-pin-count solutions to reduce design complexity and optimize system power consumption.
Fourth, the flash challenge
NOR flash technology is responsible for storing boots** and data. As Industry 40 technology, it must be highly reliable, excellent security and confidentiality, and meet stringent industry requirements. Another challenge for NOR flash memory is its ability to operate in the extreme temperature environments common to many industrial systems.
5. Challenges in scaling RAM
Embedded designs that perform RAM-intensive applications require a large amount of temporary storage for data buffering, often relying on DRAM as extended memory. DRAM is scalable and high-performance, but it consumes more power, increases design complexity, and occupies a larger PCB area.
6. Challenges in data logging RAM
Figure 3 shows the massive amount of data generated every second from numerous sensors, devices, and processes in a smart manufacturing environment. Therefore, an important aspect of data logging memory is the immediate capture and maintenance of data backups in the event of a power outage. The most prevalent data logging solution in traditional industrial automation systems is battery-supported SRAM (BBSRAM), which combines traditional low-power SRAM with a battery and additional power supply monitoring circuits** to form a non-volatile memory.
Figure 3 The vast amount of data generated by machine and system sensors must be stored and processed.
However, battery maintenance and replacement, bit errors caused by electromagnetic devices or cosmic radiation, and design complexity due to the addition of supervisory circuitry are all potential problems. The use of batteries in the design can lead to problems with reliability; As a result, engineers sought to eliminate battery disposal, improve RoHS compliance, and reduce battery maintenance.
7. Future memory trends
In view of Industry 40 and related memory solutions continue to evolve, and some memory-based trends are becoming apparent.
8. Biased towards higher performance
Modern industrial systems will require high-performance RAM with no latency. As the amount of input data increases, higher random access is required for better performance. In addition, these visits must be instantaneous (at bus speeds) and must be achieved without page access overhead and limited durability.
9. Prefer higher density
As more and more sensors are integrated into industrial systems, more and more remote IOs must be connected to the controller, and the data generated will also increase substantially. This data must be stored locally for processing and maintenance to minimize system downtime. As the amount of edge sensor data increases, higher density of non-volatile memory is required to record this data in real time. Similarly, more complex industrial systems require higher density NOR flash memory to store** and system data.
10. Higher reliability
Industry 40 requires robust operation in challenging environmental conditions. Reliable data logging is required to ensure seamless recovery after a power outage and to limit downtime for industrial systems. In addition, electromagnetic interference (EMI) from the operation of motors on the factory floor can affect the content of the memory used in industrial systems. Infineon's F-RAM and NVSRAM memories are immune to magnetic field interference and are among the most reliable data logging memories in the industry.
With the integration of security features in next-generation systems, manufacturers are requiring hardware secure elements to be incorporated into storage memory to protect against cyber threats and tampering.