Font Size: a A A

Artificial Heart Pump Control And Performance Evaluation

Posted on:2020-06-23Degree:MasterType:Thesis
Country:ChinaCandidate:Q G MengFull Text:PDF
GTID:2370330590481620Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
With the improvement of living standards,people are suffering from more and more cardiovascular diseases.Among the cardiovascular diseases,heart failure is the most serious.However,when heart failure is quite serious,the most common method for treating heart disease is heart transplantation.Or install a heart pump.Because of the large number of people waiting to be transplanted,and the corresponding portable living heart is too small to meet the requirements of the number of patients,various types of heart pumps have developed rapidly.The artificial heart pump used in this paper is the newly developed split artificial heart pump.The blood pump of this type of heart pump is installed on the human aorta.If the heart pump is controlled according to the traditional control method,it will not reach.The expected effect,the traditional method,does not fully consider the patient in daily life,with the change of state,the physiological parameters are also different,the rotational speed of the heart pump can not adapt well to human physiological requirements.In the future,research and development combined with human physiological parameters,to meet different human conditions,to adapt to human physiological requirements,is the main problem in the study of artificial heart pumps,in order to ensure the normal operation of the heart pump,meet the control requirements of the heart pump,so that the heart pump To achieve the best level of assistants,this paper has the following three control strategies for the above problems:1)Establishment of a circulatory system and a cardiac pump parameter model.The first is the mathematical modeling of the blood circulation system.In this paper,the flow characteristics in the circulatory system are the same as those in the circuit.Some devices in the circuit are used to model the human circulatory system structure,such as the left and right ventricles,arteries,veins,aorta,and peripheral resistance.Vascular compliance,inductance represents blood flow inertia,resistance represents resistance,can better help understand and master the rules of each part of the circulatory system,and facilitate the subsequent design of the control system,providing a reliable theoretical basis.2)In the complex environment of the circulatory system,there are many process parameters that affect the expected value of the blood pump controller speed.If theobtained data is not processed,using these data directly as the input variables of the optimization model will result in too many input variables of the model.In turn,it will affect the output of the optimization model.For this reason,the principal component is used to reduce the dimensionality of the parameters,simplifying the optimization model and improving the optimization accuracy.On this basis,the first network optimization model is to optimize the whitening parameters of the conventional gray neural network(GNN)with improved PSO,to ensure the accuracy of the optimized network,and to select three influencing factors for the optimal network of heart pump speed.As an input sample,the actual rotational speed is used as an output sample,and the improved PSO-GNN and GNN models are established,and the degree of influence of various factors is obtained.The experimental results show that the improved PSO-GNN model not only improves the optimization accuracy,improves the performance,but also has good stability.It shows that the improved PSO-GNN algorithm proposed in this paper is effective and feasible.The second network optimization model is an improved particle swarm optimization(IPSO)optimization support vector machine(SVM)method that takes into account the complex factors in the circulatory system.Principal Component Analysis(PCA)was used to analyze the influencing factors of the circulatory system.The improved particle swarm optimization algorithm was used to optimize the support vector machine to obtain the model parameters,and the expected value of the blood pump speed was optimized.The results show that the improved particle swarm optimization(PSO)algorithm based on principal component analysis has better approximation ability and optimization precision in the blood pump speed optimization network,which indicates that the algorithm can be effective and feasible.3)In order to better evaluate the performance of the heart pump,this paper selects the historical data of the changes of the parameters of the parameters of the heart pump and the installed heart pump to establish the evaluation criteria of the principal component analysis,and reduces the dimensionality of the seven parameters.For the three principal components,the analysis shows that the larger the comprehensive index value,the better the heart pump performance,the better the cardiac pump assist level,and then compare with the optimized speed expectation value and the variation of each factor.
Keywords/Search Tags:artificial heart pump, principal component analysis, grey neural network, support vector machine, performance evaluation
PDF Full Text Request
Related items