Water Quality Monitoring System Based on Improved Kalman Filter Algorithm

For the processing of data in water quality detection in fishing grounds, a water quality monitoring system based on cloud edge integration is established. For the noise problem encountered in data collection, an improved Kalman filter data fusion algorithm based on sparrow search algorithm is proposed, and a mathematical model in the filtering process is established, making the system highly robust. This system can accurately obtain environmental data without supervision and upload it to the cloud platform, making contributions to improving the digitalization level and breeding efficiency of fishing grounds and the sustainable development of fisheries.

I. Introduction

Compared with the past, modern aquaculture mode has realized the basic fully automated aquaculture mode, and the water quality monitoring system in automated aquaculture plays a key role in automated aquaculture. Improving the accuracy of data transmission of water quality monitoring system is crucial for water quality monitoring. In the process of observation and transmission of experimental observation data, due to the influence of the environment and the accuracy of the measurement device, the final measurement result will deviate from the actual value. In order to make the result closer to the real value, the filtering algorithm can be used to eliminate the white noise generated in the measurement process and data transmission process. Although the traditional Kalman filter algorithm[1]–[3] can remove a certain amount of noise, the artificial determination of the initial estimation value and the initial estimation error will lead to the instability of the algorithm accuracy. Therefore, scholars have been deepening the improvement of Kalman filter, such as extended Kalman filter, state error Kalman filter, complementary Kalman filter, etc. In this paper, swarm intelligence algorithm is introduced to improve the Kalman filter algorithm. Through iterative optimization, the initial estimation value and initial estimation error with the minimum error are automatically obtained, so that the error of the Kalman filter algorithm is minimized. Swarm intelligence algorithm [5]–[7] was proposed in the last century by scholars to simulate the behavior of low intelligent swarm organisms in nature to complete high intelligent swarm activities through cooperation. For example, particle swarm optimization algorithm is a classical algorithm that simulates the information transmission behavior of birds in nature when foraging. Due to the excellent performance and wide range of applications of search algorithms, many new algorithms and improved classical algorithms have been proposed continuously, such as sparrow search algorithm, cuckoo algorithm, etc. There is little research on the combination of swarm intelligence algorithm and Kalman filter. This paper will study the combination of swarm intelligence algorithm and Kalman filter algorithm.