Study On The Application Of Chemistry Metrology In Automotive Potentiometric Titration

With the extensive application of automatic titration potentiometer, titration analysis continued to be used in the fields of petroleum chemistry, food and drink, bio-pharmacy and metal electroplate. However, the analysis precision of titration results are seriously affected by the solution concentration, decomposition constant of acid or base and the stability constant of complex. Under the extreme conditions, such as all the values are very small, the burst spring range is too narrow. Using the methods of chemistry metrology and dealing with various potentiometric titration data of before-and-after titration end point, interfere could be excluded.In this article, the theoretical formulas of titration analysis are used to discuss the situations of alkali standard solution titrate monoprotic weak acid and the mixture of strong and weak acids. Under deposition titration condition, various appropriate experimental points, such as titration concentration, end point volume, balance constant et al. are selected to obtain the fitting parameters and conditions for the use of the least squares binary polynomial fitting.In addition, the advantage of the least squares binary polynomial fitting is discussed through compared the results of chemistry metrology and the least squares binary polynomial fitting. The nerve network method is complex and suitable for the routine titration which has no practice value for the other methods. The further research on the mathematic model should be carried out to find more applicable models. The results showed that the linear potential titration and polynomial fitting were suitable for the titration of narrow burst spring range. The fitting function of small wave transform and polynomial fitting is similar which primarily used to control the high frequency noise. Meanwhile, the varied factors in these two methods are less and prone to implement by programmed.Finally, the titration analysis applications of the least squares binary polynomial fitting are investigated. The influence of various fitting conditions is studied. The results indicated that the favorable analysis results can be obtained through the proper choice of parameters. In addition, the linear regression is used to calculate which fit to the practice in interior microprocessor.