Groundwater level prediction based on hybrid hierarchy genetic algorithm and RBF neural network

As the traditional non-linear systems generally based on gradient descent optimization method have some shortage in the field of groundwater level prediction, the paper, according to structure, algorithm and shortcoming of the conventional radial basis function neural network （RBF NN）, presented a new improved genetic algorithm （GA）： hybrid hierarchy genetic algorithm （HHGA）. In training RBF NN, the algorithm can automatically determine the structure and parameters of RBF based on the given sample data. Compared with the traditional groundwater level prediction model based on back propagation （BP） or RBF NN, the new prediction model based on HHGA and RBF NN can greatly increase the convergence speed and precision.

Therapeutic observation on superficial needling with different frequencies for intractable facial paralysis

Objective:To obverse the therapeutic effect of superficial needling with different frequencies for intractable facial paralysis.Methods:A total of 120 patients with intractable peripheral facial paralysis were allocated into a superficial needling with high frequency group(150 times/min),a moderate frequency group(100 times/min)and a low frequency group(50 times/min)according to the random number table method.The Toronto facial grading system(TFGS)was used to evaluate facial nerve functions before treatment and after 2 weeks and 4 weeks of treatment respectively.The electromyography(EMG)test of the mandibular branch of facial nerve was used to compare the motor nerve conduction velocity(MCV),sen sory nerve con ducti on velocity(SCV)and mono phasic acti on pote ntial(MAP)among differe nt groups,and was done before treatment and after 4 weeks of treatment.The clinical efficacy was also compared.Results:After 2 weeks and 4 weeks of treatment,the changes of TFGS scores in the three groups all showed statistical significanee(all P<0.05)z and the TFGS score in the low frequency group was substantially higher than that in the other two groups.After treatment,the changes of the MCV and SCV in the three groups all showed statistical significanee(all P<0.05),and the results in the low frequency group were higher than those in the other two groups;the change of MAP in the three groups showed no statistical significance(P>0.05).The total effective rate was 65.0%,80.0%and 95.0%in the high frequency group,moderate frequency group and low frequency group respectively,and the betweervgroup differences showed statistical significanee(P<0.05).Conclusion:Compared with the superficial needling with high and moderate frequencies,superficial needling with low frequency can produce more significant clinical efficacy for intractable facial paralysis.

Remote therapeutic effect of early nerve transposition in treatment of obstetrical brachial plexus palsy

To report a method and remote therapeutic effect of early nerve transposition in treatment of obstetrical brachial plexus palsy. Methods: From May 1995 to August 1996, 12 patients who had no recovery of biceps 3 months after birth were treated with nerve transposition. Eight had neuroma at the upper trunk and 4 had rupture or avulsion of the upper trunk. Mallet test was used to evaluate the results. Results: The follow up of 40 52 months showed that excellent and good recovery in functions was found in 75% of the patients and the excellent rate of phrenic nerve and accessory nerve transposition was 83.3 % and 66.7 % respectively. A complete recovery in shoulder and elbow joint function was in 3 patients and Mallet Ⅳ was in 6 patients. Conclusions: Satisfactory outcome can be obtained by using early nerve transposition in treating obstetrical brachial plexus.

Parameters optimization and nonlinearity analysis of grating eddy current displacement sensor using neural network and genetic algorithm

A grating eddy current displacement sensor(GECDS) can be used in a watertight electronic transducer to realize long range displacement or position measurement with high accuracy in difficult industry conditions.The parameters optimization of the sensor is essential for economic and efficient production.This paper proposes a method to combine an artificial neural network(ANN) and a genetic algorithm(GA) for the sensor parameters optimization.A neural network model is developed to map the complex relationship between design parameters and the nonlinearity error of the GECDS,and then a GA is used in the optimization process to determine the design parameter values,resulting in a desired minimal nonlinearity error of about 0.11%.The calculated nonlinearity error is 0.25%.These results show that the proposed method performs well for the parameters optimization of the GECDS.