Research And Implementation Of Wideband High Resolution Frequency Synthesizer

Lithium ion batteries are widely used in various fields due to their superior performance.But when the battery capacity drops to a certain stage,the battery life will reach the threshold value,and the battery can not continue to use.Therefore,the prediction of the remaining useful life(rul)of the battery has become a hot issue of current research.At present,there are mainly two methods for the prediction of lithium-ion rul: model-based method and data-driven method.Because of its complex modeling and many parameters,the development of model-based prediction method is limited.However,in most data-driven algorithms,there are some problems,such as low algorithm accuracy,single parameter selection and no dynamic change,and lack of prediction results under high and low temperature conditions.Therefore,this paper proposes a battery RUL prediction framework based on the entropy weight approach to ideal solution TOPSIS and improved particle swarm optimization PSO(1)Firstly,the working principle and degradation mechanism of lithium-ion battery are studied.After analyzing and observing the changes of parameters in the degradation process,it is found that the trend of specific parameters and the trend of battery capacity degradation are highly correlated.Finally,the number of charge and discharge cycles,temperature,internal resistance and open circuit voltage are selected as the parameters of the battery capacity degradation model.(2)In order to predict the rul of battery more accurately,this paper uses the improved particle swarm optimization algorithm to extrapolate the model to predict the rul of battery and carries out the relevant comparative experiments.The experimental results show that the improved PSO algorithm has the advantages of small fitting error,high fitting degree and high precision compared with the standard PSO algorithm and other related algorithms.Therefore,this paper selects the improved particle swarm optimization algorithm as the algorithm of battery rul prediction.Through the data of the first 40 charging and discharging cycles of the battery,the model parameters are obtained by training the model,and then the prediction results are obtained by model extrapolation.The final prediction results show that the improved particle swarm optimization algorithm has high prediction accuracy.(3)In view of the factors such as insufficient data and large data fluctuation in different working conditions,this paper uses TOPSIS method based on entropy weight to dynamically select the parameters of battery degradation model,and filters the battery data through moving average filter MAF algorithm to remove the noise in the data.Finally,TOPSIS method is used to select parameters under different working conditions and build a battery degradation model.Then the filtered data and improved particle swarm optimization algorithm are used to predict the rul of battery.(4)In order to verify the validity of the prediction framework and method,we use the battery data of NASA pcoe research center to carry out relevant comparative experiments.The experimental results show that:(1)compared with the prediction results without TOPSIS and MAF filtering,the prediction accuracy after TOPSIS and MAF filtering is improved by 17.9%.(2)The prediction error rate of this method is 0 under high temperature condition and 2.1% under low temperature condition,which is more accurate.(3)Compared with the existing data-driven rul prediction method,the prediction method proposed in this paper has better prediction accuracy under the data conditions of this paper.