The Industrial Internet of Things (IIoT) promises to deliver a step change in efficiency and a leap towards autonomy for industrial automation and other sectors, such as security and surveillance and building management. The prospect of self-monitoring, self-managing factories and manufacturing processes is no longer beyond the distant horizon. The ability to remotely identify, monitor, and control every individual device on a manufacturing process network with minimal or no human intervention offers opportunities that were beyond comprehension just a decade ago – even in the eyes of engineers working at the very forefront of industrial manufacturing technology. At the core of achieving the IIoT’s true potential will be the effective interplay and connection of sensing, computing, and control technologies in robust, energy-efficient implementations.
Successful IIoT design and management can mean continuously optimized efficiency, reduced operating costs, and more resilient self-learning processes. At the fulcrum of this are a number of semiconductor technologies on offer and in development at some of the world’s top electronic component manufacturers.
The key drivers in the IIoT are more measurement of an expanding number of parameters, fast and extensive analysis of and reactions to data, and accelerated enhancements to processes.
Vision-based sensing and energy harvesting
Sensors are at the heart of the IIoT, gathering more and more data. But to measure more, we need to sense more – more parameters, more accurately, and more often. Overlaying software on established technologies can bring incremental gains, but to make the necessary advances we need to add more parameters with accuracy. Each additional parameter can make the system smarter.
Sensor technology to measure “traditional” parameters, such as temperature, light, position, level, humidity, and pressure, continues to advance. But despite becoming smaller, more cost-effective, and often embedded, each sensor is dedicated and, hence, limited in its functionality and versatility. Vision-based sensing enables a new paradigm. Corollary to the concept of “a picture is worth a thousand words,” once a machine can “see,” much more is possible much faster.
With vision sensing, programmability brings flexibility, enabling a single vision system to sense missing or misplaced components and other process variables. The trend towards vision-based sensing – both still and video – will make future systems more intelligent, flexible, and ultimately more valuable.
Current advances in machine vision, in many situations, can provide better sight than the human eye. Coupled with the reduction in risks associated with human error, it enables a new degree of adaptability and speed for reconfigurable production processes. This could be anything from optimization based on real-time events to the reconfiguration of a system that may allow multiple products to be made on a single line. They only require a simple change from one control program to the next and the system is ready to run. This has obvious cost, time, and labor benefits and reduces the risks associated with error.
However, the distributed nature of the IIoT and the requirement to place multiple sensors at the various points of measurement presents another challenge: the reliable delivery of power.
Successful sensors, especially those utilized in IIoT designs, have four basic attributes: they need to be self-powered, collect data, broadcast their status, and have the ability to connect.Wireless sensors with the ability to harvest energy, i.e. self-powered, are essential if the IIoT is to advance and realize its potential. ON Semiconductor’s energy harvesting, wirelessSmart Passive Sensor family, as shown in Figure 1, provides an example of the technology that can meet these requirements. These devices can be used to sense in the most hard to access, space constrained places with no direct power source.
[Figure 1 | ON Semiconductor’s Smart Passive Sensor family are wireless and self-powered]
Being able to harvest power will allow sensors to function autonomously.
Instant access to data
Sensors gather data, and then post-processing and analytics generate valuable information, along with the ability to control our factories and processes.
The processing of large, sustained flows of data associated with real-time sensing for the IIoT is reliant on the cloud. This analysis requires platforms that can store large data sets across distributed clusters; often combining and processing data from many geographically dispersed sources.
Among the many benefits the cloud brings is the removal of barriers in global organizations. For example, a problem in a Chinese smart factory can trigger an almost instantaneous process improvement in a similar factory located somewhere else in the world.
Advances in secure communications and authentication allow mobile devices to connect into these networks, bringing opportunities for instantaneous access to information. Flexible access can then drive value into the business and its relationships with the outside world. In fact, this IT/Operational Technology (OT) integration, as shown in Figure 2, has the potential to bring the greatest benefits to a broad range of industrial-automation applications.
[Figure 2 | IT/OT integration]
Reacting to analyzed data
Many industrial control systems are very sophisticated and require accurate or careful, often rapid, positioning. This could mean anything from the speed and direction of a cooling fan, to a motor or servo that adjusts a valve position, or a stepper motor for linear or angular positioning in a precision task.
Alongside the rapid development of sensing for the IIoT, theactuators and their controllers that provide the physical reactions to gathered, analyzed data are seeing similar advancements. Discrete, component based solutions for motor control are being superseded by advanced integrated power modules that are smaller and more straightforward to implement.
Combining sensing, connectivity, and actuation
Integrated manufacturing processes require a broad range of sensing technologies to be combined with connectivity and actuation, which creates design and expertise challenges. Fully integrated hardware and software development platforms that combine these elements are crucial to help speed and ease the customization of specific functions for adoption into end products. Modularity makes these platforms extensible to newIoT/IIoT functions and devices that are based on new advances, allowing more rapid adoption. Open-source support is also important, since a broad ecosystem and interoperability are crucial for the IoT’s success.
Vision sensing is a good example. From a hardware perspective, this requires video processing capability to implement, as well as image-processing software to interpret the data stream. Similarly, data gathered from energy harvesting wireless sensors must be able to be moved to the cloud (see Figure 3).
Sensors, data processing, and actuators are all significant building blocks of an IIoT application. However, without a means to communicate, share data, and transmit, receive and execute instructions, the IIoT cannot function.
Considering the IIoT’s unique requirements, not all standards and protocols are satisfactory. The technologies suitable for smartphone PANs or single-supplier standards are unlikely to be successful. Instead, it’s important for IIoT platforms to demonstrate flexibility by supporting a broad array of standards, including Thread, SIGFOX, EnOcean, M-BUS, KNX,ZigBee, and proprietary protocols. The adoption of a software-defined radio approach allows a single platform to support multiple protocols. ZigBee and Thread are complementary, and the alliance of the industry organizations behind these protocols should also drive broad adoption within smart home applications.
Thread is an IP-based (IPv6) networking protocol built on open standards for low-power 802.15.4 mesh networks that can easily and securely connect hundreds of devices to each other and directly to the cloud. Security and interoperability are two of Thread’s key value-added capabilities.
Conversely, SIGFOX enables wide-area networks that provide relatively low bandwidth communication with fixed or mobile smart objects or sensors that are deployed over a large area. Example applications for this protocol include the nationwide tracking of shipping containers or vehicles, and communication with geographically spread assets such as smart-city equipment or oil pumps and pipelines.
Courtesy of www.embedded-computing.com