In the intelligent age, sensors, as the source of data collection, are almost everywhere. Situational awareness required by the front end of intelligence starts from sensors. Whether it is intelligent manufacturing, smart city, smart medical care, intelligent equipment and big data analysis, no matter how huge the intelligent system is, it must start with sensors.
Sensors can sense the information and state of the measured objects, and can transform various physical quantities, chemical quantities and biomass in nature into measurable electrical signals. Therefore, as the entrance of receiving and transmitting massive data, sensors are the foundation of the Internet.
Of course, sensors have also experienced many rounds of technical changes. Nowadays, sensors are developing toward to intelligence, thinking, analysis and diagnosis. As an increasingly intelligent microsystem, sensors are becoming more independent and have the ability of self-correction. If the algorithm can be placed not only in machines and equipment, but also in the smallest sensing unit-sensor, then embedded artificial intelligence will vigorously promote the development of the intelligent era.
Sensor and Intelligent Sensor
Reviewing the development history of sensors, it can be divided into three generations of sensors: the first generation is structural sensors, which use the change of structural parameters to sense and transform signals. The second generation is a solid sensor, which is composed of solid elements such as semiconductors, dielectrics and magnetic materials, and is made by using some characteristics of materials. The third generations are today's intelligent sensor, which is made up of microcomputer and detection technology.
Specifically, intelligent sensor integrates sensing as chip, communication chip, microprocessor, driver and software algorithm, and its advantage lies in high complexity information processing ability and high integration information transmission ability.
Ordinary sensors have only a single function of perception-output, and can't be judged in time after failure, which has increasingly restricted the development of information technology and automation technology, and can no longer meet people's differentiated needs. Intelligent sensors have advantages that ordinary sensors can't.
The biggest feature and advantage of intelligent sensor is "intelligence". On the one hand, intelligent sensors have the functions of automatic calibration and automatic compensation. The intelligent sensor can remove the zero by automatic zero gas, compare with the standard reference benchmark in real time, automatically calibrate the whole system, correct the system errors such as nonlinearity, collect a large amount of data in real time for analysis and processing, eliminate the influence of accidental errors, and ensure the high accuracy of the intelligent sensor.
The intelligent sensor can also automatically compensate the drift of system characteristics caused by the change of working conditions and environmental parameters, such as the drift of zero point and sensitivity caused by the fluctuation of environmental temperature and system power supply voltage, so as to ensure the high reliability and stability of the sensor. At the same time, because the intelligent sensor has the functions of data storage and information processing, the noise in the input data can be removed and useful data can be automatically extracted through digital filtering and other related analysis and processing; Through data fusion and neural network technology, the influence of cross sensitivity in multi-parameter state can be eliminated, and dynamic calibration can be realized.
In addition to recalibrating and compensating the original sensing data, the intelligent sensor also supports feature extraction of the original data, and realizes a more complex edge computing mode based on the data features, thus realizing the intelligent decision processing capability at the service level. For example, the intelligent sensor of mechanical equipment condition monitoring scene can identify abnormal equipment faults on the edge side, and trigger data reporting and more intensive data acquisition mode immediately after identifying equipment faults.
On the other hand, intelligent sensors can automatically diagnose themselves. Ordinary sensors need to be inspected and calibrated regularly to ensure that they have enough accuracy in normal use. These tasks generally require that the sensors be disassembled from the use site and sent to the laboratory or inspection department. If the on-line measurement sensor is abnormal, it can't be diagnosed in time. When the intelligent sensor is used, the intelligent sensor can start the self-diagnosis function when the power supply is turned on, and carry out self-inspection and diagnostic test to determine whether the component is faulty or not. According to the service time, the intelligent sensor can be corrected online to ensure stable and reliable operation.
Intelligent sensors also have the functions of judgment, analysis and processing. For example, for the intelligent sensor of mechanical equipment condition monitoring scene, the intelligent sensor can automatically increase the acquisition frequency when the equipment starts to achieve faster state change response speed; When the equipment is shut down, the intelligent sensor can automatically enter the low power sleep mode to prolong the battery life.
Future Development of Intelligent Sensors
However, the parameters that one sensor can measure are limited, so it has become a new trend that multiple sensors are integrated together, and various combined sensors are coming one after another, which ushered in a wide development space in the field of intelligent manufacturing. Sensors such as temperature + humidity, pressure + flow, vibration + acceleration + deceleration have become the most widely used combination. Mixed sensor fleet realizes "one device for multiple purposes" of multi-parameter detection, and forms closed-loop automatic application by detecting various parameters.
On the other hand, intelligent sensors are still developing towards fusion sensors. For example, intelligent sensors are accelerating the development of unmanned cars, and the multi-functional integration of sensors will take advantage of the advantages of different sensors to provide data analysis and control capabilities, thus having embedded intelligence. This also has important applications in military affairs. American F35 fighter has been connecting and analyzing multi-domain data. The core is to use fusion sensors to realize high-speed analysis of multi-dimensional data, and can use data from different platforms, whether it is sensor data at sea, air, sea and land, and process a variety of heterogeneous data concurrently.
Sensors, communications and computers are called the three pillars of modern information technology, and the role of sensors cannot be overemphasized. Today, with the new intelligent sensors becoming more and more intelligent, more and more intelligent sensors have been used in various applications. With the advent of ChatGPT, future sensors will even produce cognitive attributes. It can be said that sensors are the facial features of automatic systems, and intelligent sensors will also lead mankind to the digital future.
The intelligent era is accelerating. Smart sensors can be expected in the future-China exportsemi.com
