The Research On Signal Processing Of Turbidity Sensor Based On Kalman Filter

The sensor technology is an important field in the modern science and technology, and has achieved remarkable results in the world. Many control systems such as aerospace, robots, nuclear power stations, industrial production control and other fields, its most basic component is sensor. If sensors in use broke down and had not been timely processed, to say the least, they will influence the production and quality of products, to say the most, they will cause serious accident, causing paralysis in operation of the system. The consequence is unimaginable. Now people have pay great importance attention to sensor fault diagnosis, and studied some methods to sensor fault diagnosis and signal recovery. For sensor fault detection and diagnosis, primary task is to process its output signal, because the output signal includes two parts, measure signal and noise, and measure noise includes fault signal, so our analysis key objects should be output signal processing on sensor.This paper is mainly to research the turbidity sensor output signal processing based on kalman filtering, which is the important branch in research of sensor signal processing. Kalman filtering can not only estimate the status of signal in past or current, but also can estimate the future state, even when the accurate system is unknown, it also can estimate. The paper is using kalman filtering to process observational data, then can get the estimated variables of the system state, using the characteristics to process the output signal of the turbidity sensors of the dishwasher, and through the kalman filtering, reducing the influence of observational noises, getting the accurate estimate state of turbidity sensor.This paper firstly introduces the research background of the subject; Secondly elaborated the conceptioon of turbidity sensor, output parameters, function characteristics and so on, and the important problem solving tools in the the computer application field, Matlab, as the development platform in the subject; In addition, kalman filtering is one of the key in this paper, which provides important theoretical foundation for the research of signal processing of turbidity sensor; Another key is how to establish a suitable turbidity sensor model, and according to the turbidity sensor's functions characteristics, putting forward a technical scheme of the output signal processing of turbidity sensor. Experiments show that the kalman filtering not only can reduce the influence of observational noises, but also can get the accurate estimate status of turbidity sensor. The system basically reaches signal processing technology requirements with practical application value, and has laid a foundation for the subsequent fault detection and diagnosis of turbidity sensor.