Years of incremental embedded processing, power management, and connectivity technology improvements have now brought us near the break-even point where solar- and wind-generated power will be competitive in cost with power from fossil fuels. Not only will these renewable energy sources avoid contributing to atmospheric changes through carbon release, but soon they will be economically advantageous as well. This development, which is being referred to widely as grid parity, won’t happen overnight. Rather, a gradual transition will occur during the next decade or more, as cost advantages improve region by region and utility by utility. Government subsidies will aid in the transition, but cost-reducing advances in semiconductor technology and manufacturing are rapidly overtaking subsidies as the primary growth driver.

In some areas, wind power has already achieved parity – at least, when the wind is blowing. Solar panels will take longer to reach the goal, though they promise to introduce an even more radical change in energy-gathering. Whereas wind farms usually require large investments in grid infrastructure to operate efficiently, solar generation can be used effectively at almost any scale, from a handful of panels on a rooftop to a solar farm spreading over acres. Both forms of energy gathering, but especially solar, will significantly change the grid by introducing widely distributed sources of power generation. Achieving grid parity will accelerate this trend. However, for renewable energy generation to really take off, something besides favorable cost is needed. Utility networks must also become more intelligent, using smart sensors, control, and communication technology in advanced Integrated Circuits (ICs) such as microcontrollers (MCUs) and processors (MPUs) to operate energy-gathering systems and coordinate overall operations.

Distributed power generation requires smart grids

Challenges lie in adding intelligence to larger utility structures to make them operate more reliably and efficiently. Smart grids are intelligent power networks that measure, communicate, and control energy at every stage. Every generator, substation, relay, transformer, meter, machine, and transmission line – everything in the grid and all that is attached to it can potentially be measured and observed to ensure that electricity is delivered effectively where it is needed and used without waste.

In addition to providing efficient, reliable delivery of energy, smart grids will serve as an indispensable infrastructure for the distributed generation of power from renewable sources. In the future, technology must enable the utility grid to manage the bidirectional flow of electricity to customers who both generate and consume power. In addition, because the sun shines brightest and the wind blows consistently in places that are often far from cities and industrial districts, utilities will have to rely on power lines that transport electricity over long distances. The grid must be able to manage demand versus creation in real time over large areas, relying on varied types of decentralized power generation and new capabilities of energy storage.

Figure 1 shows a smart grid with distributed power generation from large solar and wind farms, as well as bidirectional flow from producer-customers of various sizes. Only an intelligent grid can integrate such widespread electricity generation from renewable sources, along with outputs from traditional large generating stations, and deliver power in real time to customers with varying requirements. The economies of scale required to achieve grid parity can only come about through such an extensive deployment of solar panels and wind turbines in decentralized energy generation. Decentralization, in turn, can happen only through smart grids that are enabled by advanced electronic measurement, control and communication.

It would be difficult to overstate the significance of smart grids in the years ahead. By some estimates, the world wastes as much as 40 percent of the power it generates due to tampering, theft, and inefficiencies. Intelligent, efficient grids can avoid much of that waste, saving money and cutting deeply into greenhouse gas emissions.

ICs make grids "smart"

Power grids become intelligent when they integrate electronic circuits that sense, monitor, measure, and provide communication and control. Advanced embedded and analog technology serves as the foundation for a range of equipments used in the grid infrastructure, including protection relays, power quality monitors, circuit breakers, and substation automation systems. Beyond the infrastructure are inverters for solar and wind power, smart meters at commercial and consumer sites, and a potentially limitless number of switches, sensors and monitoring units that feed information into the network. These systems use electronic intelligence to produce power, measure consumption instantaneously, communicate using low-power RF or Power Line Communication (PLC) technologies, and automatically regulate the production of power from different sources and supply it safely to customers.

The specific requirements for these equipments vary widely, according to usage, standards, and regulations. However, the underlying semiconductor technology, especially in grid infrastructure equipment, can be generalized as shown in Figure 2. (Similar functions with somewhat different architectures are used in the IC solutions that enable solar inverters, smart meters, and other connected equipments.)

In the measurement block on the left of the figure, inputs come from external sensors in the form of analog signals such as voltage or current levels. Analog components condition the inputs and convert them into digital signals that feed into the analytics and control block. Here an MCU, MPU, or Digital Signal Processor (DSP) analyzes the digital signal data to determine electricity usage, quality of service, or a potentially dangerous condition. This processor then orders an action by, for example, a circuit breaker or relay, if such an action is necessary. In addition, the processor notifies control nodes in other equipments, and possibly a central control computer, via specialized modules in the bidirectional communication block. The low-voltage power needed to operate all of these functions is supplied through separate power management and isolation circuitry.

Because equipments comply with different standards, IC solutions must include options that allow designers to select the capabilities they need. A high level of analog and digital System-on-Chip (SoC) integration can help keep costs low and support multiple control channels. Channels include several analog and digital inputs at various stages. On one end, with the system connected to the grid, it has to interface to high voltages and currents, which are analog in nature. Several other analog channels include temperature and humidity. Digital channels consist of event indications coming from other devices on the grid, which include several instances of protection relays, fault passage indicators and circuit breakers. Though, in some cases design requirements may dictate the use of separate ICs for high-voltage, high-current, or high-speed functions such as op amps and Analog-to-Digital converters (ADCs). For example, a key requirement may be using the measurement system to isolate the AC mains. This isolation is done in the analog domain, which not an easy problem to solve for IC designers. Therefore, having them separated from the SoC would be the most practical and cost-effective means. In other cases, there will be a strict need to separate measurement and control, because a certified, fail-safe closed measurement system is mandated to prevent interruptions and various forms of tampering.

On-chip communication solutions such as Industrial Ethernet and Industrial Fieldbus can also provide economy and flexibility. Industrial Ethernet and Fieldbus work together in a real-time distributed control system. In a substation, there will be several sensors, relays, breakers, fault indicators at various critical locations. The ability to control and receive the status from all of these end-nodes in real-time requires a robust and optimized scheme. The Fieldbus connection forms a network of these devices. The amount of cabling required for Fieldbus is much lower and will directly contribute to cost savings. The Ethernet communication can still be used for non-real-time tasks, where higher speed is required for communication, but has limitations in comparison to the Fieldbus.

The processor complex at the heart of a solution should be completely programmable and scalable, supporting multiple protocols, such as Ethernet, MODBUS, PROFIBUS, HART, IEC 61850, DLMS (IEC 62056), and others without needing custom-designed logic.

Software frameworks, tools, and reference designs can also help speed and simplify development. Having ready-made hardware and software tools from module vendors speeds time to market. The development becomes even simpler on a robust framework where the user can add or remove features in hardware and software. The robustness is also reinforced when the provided hardware and software has been certified by either standard bodies or test houses.

Smart, embedded power

Today, advanced electronics with these features are increasing the intelligence of power grids, a step that is indispensable in the long-term realization of decentralized energy production from solar panels and wind farms. While the economics of renewable energy deployment is approaching grid parity, IC technology advancements introduce intelligent measurement, communication, and control. Together, these trends are transforming electrical grids worldwide to provide cleaner, more reliable energy.

Markus Staeblein is General Manager, Smart Grid and Energy Solutions at Texas Instruments Incorporated (TI).

Kripa Venkat is Industrial Systems & Solutions Manager, Smart Grid and Energy Solutions at Texas Instruments Incorporated (TI).

Texas Instruments Incorporated (TI) www.ti.com @txinstruments www.ti.com/linkedin opsy.st/TIGoogle