Nickel Metal Hydride Battery Storage Stability And State Of Charge Prediction

The effect of charge and different storage temperatures on the storagestability of MH-Ni battery was investigated. The structure and morphology ofthe materials were characterized by means of XRD and SEM techniques, andthe impedance properties were studied by EIS technology. At the same time,a mathematical statistics model based on SVM method was built to predictbattery's State Of Charge (SOC).The results are shown as follows:The performance of battery stored at different temperatures got worse afterexperiments. The inner resistance of the battery was elevated by 6 mΩonaverage, and the open circuit voltage of the battery with 75% charge waslowered to 1.284V and 1.264V after being stored at 45℃and 60℃respectively. The voltage rised first and fell later with the increase of the timeat a discharge rate of 5C. The discharge platform and capacity decreased withthe increase of charge. In particular, the discharge capacity decayed worseafter being stored at 60℃, in which the battery with 100% charge reached to15.5%. The ratio of self-discharge increased, the open-circuit voltage andcharge retention lowered. The charge retention of battery with differentcharge was all below 80% after storage. The charge retention was less than60% for battery with 100% charge compared with that storage before, and itheld to 76.63% and 71.67% for battery with 75% charge after being stored at45℃and 60℃, respectively. High temperature has a large influence on thestorage of battery, and the worse performance was observed when the batterywas stored at 60℃. The battery with 50% charge has better performancethan others in that the discharge platform and discharge capacity.By fitting the EIS images of battery that stored at different conditions, it isfound that the ohm inner resistance increased gradually when the chargelowered, but the electrochemical reaction resistance decreased at one time.The XRD patterns and SEM images indicated that the capacity degradationof MH-Ni battery was caused by the defects and disorder in the crystalstructure of active mass in negative and positive electrodes. At the same time,the surface of separator broke and became rough.A model based on SVM method was built to predict battery's SOC. RBF was chosen and cross validation method was used to control accuracy. TheEIS data, the open circuit voltage and inner resistance data were analyzed,80.0% and 97.1% of original grouped cases correctly classified respectively.