ARM Solutions for Smart Grid Industrial Applications

Data loggers are an important part of today's smart grid. There are many factors that affect the design of data loggers. According to the design complexity of the logger, data loggers in residential and industrial applications can be roughly divided into three levels: entry-level, intermediate and advanced. NXP Semiconductors' ARM solutions can meet all three needs and provide the right combination of functions.

Today's power companies widely use data loggers in smart grids to track power consumption, monitor grid functions, and analyze power operations. Data logger circuits are usually placed between electronic meters that measure power usage and the upstream network of the grid. Data loggers collect input data from electronic meters, perform some data processing, and then relay or upload the centralized data to the network.

Figure 1 for entry-level data acquisition

Resource requirements for entry-level data loggers

There are many factors that influence the design of a data logger, including the type of power line (single-phase or multi-phase) and the functional requirements of the utility system (residential or industrial). In general, however, there are three common categories of data loggers: entry-level, mid-level, and advanced. This article will provide an overview of each of these three types of data loggers and discuss the data processing and system resource requirements that need to be considered when selecting a microcontroller for a data logger.

Entry-level data logger

Entry-level data loggers, as the name implies, meet relatively basic system requirements. They typically support single-phase power lines (often used with residential meters) and are used to collect data from automatic meter reading systems (AMR) or new smart meters with digital outputs. The collected data is usually stored in the flash memory of the logger system (built-in or external to the microcontroller itself), and the centralized data is transmitted to the upstream network at a predetermined time through the selected communication interface.

Figure 2 LPC1200 functional block diagram [page]

Entry-level data collectors typically perform a certain amount of preliminary data processing before passing information to the upstream network. For example, by using a small amount of data sampling and time recording, the data collector can report on power usage over a specific period of time, ranging from just a few minutes to a week or a month; data can also be classified and stored according to different time intervals and filtering methods. This helps power companies to analyze power usage trends in detail, with data granularity down to individual users, and can be dynamically adjusted to achieve more reasonable power transmission and distribution. After configuration, data collectors can also monitor the downstream operation of electronic meters. If meter parameters change, the reporting interval exceeds the tolerance, or a fault or abnormal data is detected, the data collector will implement software intelligence, promptly alarm, and provide the maintenance team with the information needed for remote repair.

The transmission methods of smart grids may vary from place to place, so the basic feature set may need to be expanded in response to local regulations. Depending on where the data logger is deployed, RS-485, general packet radio service (GPRS), or power line communication (PLC) can be used for data transmission, and infrared or RS-485 can be used for external control. Instead of customizing the design for each region or market, many developers take a "one size fits all" approach and build systems to support all possible transmission methods that may be used (but not all transmission methods are used at the same time). This approach may bring economies of scale in manufacturing, but it may also place more demands on the microcontroller.

Figure 3 Example of microcontroller configuration for advanced data logger

Figure 1 shows how to configure a microcontroller for an entry-level data logger, and Table 1 lists the general functional requirements for the design. Assuming that the device collects data from multiple UART ports and supports a variety of basic functions, including input acquisition, data storage, communication, and maintenance, the design should include a real-time clock (RTC) for providing time-stamp data, an optional analog-to-digital converter (ADC) for real-time power quality checks, and an optional SPI interface for use with external memory or external device communications (such as a wireless transmission RF module).

The power requirements of the microcontroller itself are not listed in Table 1, but in general, data loggers need to be efficient in their use of power. Utilities do not want to incur additional costs by increasing grid power consumption, and consumers do not want to increase their electricity bills by using new metering features.

Considering these different requirements, 32-bit microcontrollers are usually the best choice. This is because most 8-bit and 16-bit microcontrollers do not have enough power to process data from multiple data sources and often lack the necessary configuration resources to support system operation. On the other hand, because most 32-bit microcontrollers have multiple power modes, designers can usually optimize the system to improve performance and efficiency.

After deciding to use a 32-bit architecture, designers still need to find a solution that provides a reasonable combination of functions. This is a potential problem because most 32-bit microcontrollers have the necessary number of UARTs, but also provide other advanced functions that the system does not need (such as Ethernet, I2S, and LCD display interfaces). NXP's LPC1200 industrial control series (as shown in Figure 2) provides a good solution to this problem. This series uses an ARMCortex-M0 processor, provides up to 128KB of flash memory, and includes other resources that data loggers can use, such as RTC, ADC, and SPI.

The LPC1200 series comes standard with support for two UARTs. In addition, its unique Application Specific Standard Product (ASSP) function enables the system to support two additional hardware UARTs. The ASSP function allows designers to avoid increasing expenditures on high-end equipment while being flexible enough to perform multiple tasks in different applications. For example, its built-in ASSP can also be configured for I2C to DMA transfer, pin pattern matching, or analog data logging. Using ASSP can reduce the CPU load and reduce interruptions to system operation when processing simple information, which can minimize system overhead while customizing the microcontroller functions.

Intermediate and Advanced Data Loggers

Both mid-level and advanced data loggers have more extensive features than entry-level data loggers. The difference between mid-level and advanced data loggers is usually in CPU speed. That is, advanced data loggers generally require faster CPU speeds, which is critical for microcontroller configurations.

Advanced data loggers are typically used in more complex residential settings and three-phase industrial applications. The higher the computing needs, the higher the CPU performance required. A clock speed of 200MHz or more is usually the best choice. Advanced data loggers also have more advanced communication and control features, such as Ethernet and Wi-Fi, LCD interfaces for interactive displays, and USB hosts for local data downloads. These added features require more flash and system memory and require a real-time operating system (RTOS). Figure 3 shows an example functional block diagram.

For such a solution, NXP LPC32x0 series is a good choice. This series uses ARM9CPU core, which runs at up to 266MHz, and uses vector floating point (VFP) coprocessor for advanced arithmetic operations. In addition, it also provides necessary peripherals and interfaces, including 7 UARTs, 1 10/100 Ethernet MAC with dedicated DMA controller, 1 USBOTG with full-speed host and device performance, 1 RTC, and 1 flexible LCD controller that can support STN and TFT panels.

The cost of advanced data loggers increases significantly compared to entry-level products. The configuration in Figure 3 is always a three-chip solution because the microcontroller requires external SDRAM and NAND flash to form sufficient memory resources. If the system does not require such a fast CPU clock frequency, it is often more desirable to choose a microcontroller with sufficient onboard resources, which also meets the requirements of a mid-level solution (see Figure 4).

Compared with advanced solutions, the outstanding advantage of mid-level data loggers is their lower cost. Because fewer components are required, this can save up to $2 to $3 in PCB costs. Simple configurations bring better economy, but performance limitations make mid-level data loggers more suitable as entry-level upgrade products, not alternatives to advanced solutions. Mid-level data loggers are more suitable for industrial applications that can sacrifice system functions for cost; they are also a good choice for residential applications that want to improve system performance, such as Ethernet real-time communication, which can be used to control the user's power on and off, or to report status changes (display device tampering), etc.

Figure 4 Example of microcontroller configuration for an intermediate data logger

NXP LPC1760 series is very suitable for mid-level data collectors. This series uses ARMCortex-M3CPU with a main frequency of up to 100MHz, and includes a maximum of 64KB SRAM and 512KB flash memory. The onboard peripherals and communication interfaces also provide sufficient resources for mid-level data collectors.

Use in combination

For a multi-device and multi-user smart grid, entry-level and mid-level or advanced data collectors can be combined to create a more comprehensive system. For example, in a high-rise residential building, data collection can be refined to the building and floor level. The electronic meter in each apartment can provide data to the entry-level data collector on each floor; at the same time, a series of mid-level or advanced data collectors are used to merge and aggregate the data from each floor to generate information for the entire building. This setup enables layer-based data collection and processing, balancing the workload while achieving a high level of data granularity and manageability.

in conclusion

Data loggers may be a small component in the overall system, but they perform important tasks to improve the intelligence of the power grid. In addition to collecting data from electronic meter batteries, data loggers can also be configured for a variety of practical operations: checking transmission quality, monitoring power usage data, providing event data records, and reporting system faults. Whether it is an entry-level, mid-level or advanced data logger, choosing the right 32-bit microcontroller can simplify the development steps and design a cost-effective solution. When selecting a microcontroller, engineers should consider on-chip resources, as well as other design factors such as device reliability (temperature and humidity range, data retention capability, current fast transient reliability, anti-static, etc.), system-level component integration, distinguishing functions (such as data encryption), and of course price factors. NXP's ARM solutions, including the LPC1200, LPC32x0 and LPC1760 series, provide the best combination of performance features and are ideal for a variety of residential and industrial data logger designs.