Research On Intelligent Nursing Bed Defecation Monitoring Method Based On Data Fusion

As China's aging process continues to deepen,the number of disabled elderly people due to the elderly's own diseases and physiological decline is increasing.At the same time,under the situation of relatively shortage of medical resources,the long-term excretion care needs of disabled elderly people are urgently needed to be solved in today's society.Therefore,the increase in nursing pressure and burden has accelerated the application of artificial intelligence technology in the pension industry.It makes the old excretion care with the core of intelligent nursing bed become a new means to solve the ethical problems caused by the traditional nursing of disabled elderly.However,at present,there are many problems in the nursing bed for defecation care,and the perceptual ability is poor.Most of them use a single sensor to monitor the defecation,and the accuracy is not high and it is easy to cause false positives or false negatives.In view of this problem,this paper proposes to use multiple sensor data to improve the recognition accuracy of intelligent nursing bed defecation monitoring through multi-sensor data fusion technology.The main contents are as follows:Firstly,the multi-sensor data fusion technology is systematically studied,and the related fusion algorithm is analyzed.A hierarchical distributed nursing bed defecation monitoring mode is proposed,and a two-level data fusion combining neural network and improved D-S evidence theory is constructed.Secondly,the overall structure of the intelligent nursing bed is briefly introduced.The mechanical structure of the intelligent nursing bed toilet conversion device is analyzed in detail,so that the selection and installation position of the sensor are determined.A data acquisition system based on STM32 is built to collect the information of temperature and humidity sensors and ammonia sensors,and the experimental data are preprocessed,which is the basic preparation for the research of the algorithm.Then the characteristics of BP network,RBF network and Elman network performance are compared and analyzed.Their respective optimization learning algorithms are analyzed,and the three optimized neural networks are simulated and verified by sample data,At the same time,in order to avoid the subjectivity of a single neural network in constructing the basic reliability allocation of decision-level fusion,the L-M BP network,PSO-RBF network and Elman network algorithm are finally determined as the feature-level fusion methods.Finally,the traditional D-S evidence theory will produce the result of paradox when dealing with highly conflicting problems.This paper proposes an evidence synthesis algorithm based on modified cosine similarity and reliability entropy,and builds an optimized model of improved algorithm.Firstly,the similarity between the evidences is corrected by using the modified cosine similarity measure,then the information quantity of the evidence itself is measured by the reliability entropy,and finally the weight of the evidence is redistributed to obtain the final fusion result.Verified by Iris standard data set,the improved algorithm proposed in this paper has higher accuracy,taking into account not only the interaction between evidence,but also the influence of evidence itself.The data fusion algorithm combining neural network and improved D-S evidence theory is applied to the monitoring of defecation in intelligent nursing bed.Through simulation analysis,the accuracy of the proposed algorithm for the excretion state A₁ reaches 0.9825,and the uncertainty accuracy is greatly reduced,which is better than other improved algorithms.The results show that the data fusion model proposed in this paper is effective in the monitoring of defecation in intelligent nursing beds.In this paper,the intelligent nursing bed defecation monitoring algorithm based on data fusion is used to reduce the interference of other results on the final decision,has better identification,increases the reliability of system work,It provides an important theoretical basis for the accurate identification and monitoring of intelligent nursing beds.