Condition monitoring is a key element of the predictive maintenance approach for industrial installations. When it comes to the design of condition monitoring systems, many different requirements need to be taken into account.
The mission profile of the system needs to be explored from the hardware and software perspective. And this, while looking at the need for different levels of integration between all domains. Security and data flow are key elements to take into account. As these now drive the decision process and choice of wired or wireless connectivity. Additionally, the protocols used, if and how data encryption should be implemented, and where first inference should happen. At the hardware level, any condition-monitoring solution needs to be compact, robust and operationally transparent. For this reason, using the most suitable sensors to measure features such as temperature and vibration. And also connected wirelessly to a backend system, offer the simplest and most effective way of implementing condition monitoring.
MEMS Sensors enable affordable, compact condition monitoring
In recent years, important milestones have been reached in reducing the cost of condition-monitoring sensors. For example, piezoelectrical sensors have largely been replaced by micro-electromechanical systems (MEMS) technologies. These sensors are smaller and more robust, but are also more highly integrated, which delivers comparable performance at lower costs. These benefits have helped lower the barriers to adoption of condition monitoring on a much larger scale. As a result of these recent innovations, from suppliers such as STMicroelectronics, it is now possible to develop battery-powered condition-monitoring-sensors. These sensors can measure less than one centimetre on each side and can be easily retrofitted. They also use artificial intelligence to “understand” what is being measured after they have been deployed.
For condition monitoring, one of the most useful properties to observe is movement. Movement can be interpreted in many ways and often it is intentional – such as the rotation of a gear. But it can also be secondary. Such as the vibration caused in a drive shaft due to the load on the gears changing over time. Vibration-monitoring is fundamental to condition monitoring, which is why vibration-monitoring currently accounts for more than 50 % of the market. The reason for this is that vibration is intrinsically correlated to the various faults that a mechanical system can experience.
An industrial sensor with a Machine-Learning Core
ST has developed an inertial sensor with machine-learning capabilities that is architectured for industrial applications. It is part of the smartest family of iNEMO MEMS sensors. Which is recognisable by the letter X at the end of their nomenclature and the presence of a machine-learning core.
The ISM330DHCX is a system-in-package device featuring a high-performance 3D digital accelerometer and 3D digital gyroscope. It has a machine-learning processor built into the sensor itself with up to 16 configurable decision trees. That is designed to run in a highly power-efficient manner and provides accurate results in the shortest possible time. This means, that there is no using of a host microcontroller to run an algorithm and draw conclusions from the available data. Which demands a lot of energy. Because the decision tree in the sensor can run an inductive algorithm at a fraction of the power consumption. As a result, the system can recognise specific vibration or movements and infer results from pre-defined patterns. Typically using less than one hundredth of the MCU power used for the same typical tasks.
Functionality of the system-in-package device ISM330DHCX
The ISM330DHCX leverages the robust and mature-manufacturing processes already used in the production of billions of micromachined accelerometers and gyroscopes. The various sensing elements are manufactured using specialised micromachining processes, while the IC interfaces are developed using CMOS technology. That allows the design of a dedicated circuit that is trimmed to better match the characteristics of the sensing element.
In the ISM330DHCX, the sensing elements of the accelerometer and of the gyroscope are implemented on the same silicon die. Thus guaranteeing superior stability and robustness. The ISM330DHCX has a bias stability of only 3 degrees per hour. And it also offers very high precision with an angular rate capable of reaching 4,000 degrees per second (dps). It operates over a broad temperature range of -40 ºC to +105 ºC. And includes a temperature-compensation system to better maintain its accuracy, even under the harshest conditions. Additionally, the typical zero-rate-level change vs. temperature, is only ±0.005 dps/ºC. This measures the variation of bias as the environment gets hotter or colder.
Getting started quickly
The best way to start testing the new inertial sensors is to get a development board. The ISM330DHCX is on the STEVAL-MKI207V1 and is compatible with the STEVAL-MKI109V3 motherboard, which incorporates an STM32 microcontroller. This is the quickest way to start experimenting with the graphical-user interface they offer and begin working on a prototype. ST even offers finite-state-machine and machine-learning examples, allowing developers to experiment with their scripts to grasp the components’ capabilities.
Discover more about STMicroelectronis: www.st.com.