As we know, Lead-acid battery is difficult to balance many factors such as the accuracy and the on-line testing requirement. The detecting system, as stated in this article, is based on the vibration test procedure, dynamically following the electrochemical process of the Lead-acid Battery, and collects the real-time state parameters for calculation, analysis and judgment. It also quantizes precisely the degradation and chargeability of the battery and therefore self-adapts to the ideal target values. During the test, it has not charged and discharged large current to the lead-acid battery, it only plus a smaller and shorter time of impulse voltage signal on both ends of lead-acid battery, so the battery measured is damage free, and the system energy consumption is small; Using the load compensation technology, it has solved the influence of load on the test results. What's more, the load characteristics are improved at the same time, it realized the online detection. The vibration detection is based on the adaptive fuzzy inference model which has taken various factors into account, concerning the choices of input aspects which may influence the output value. It realized a number of Lead-acid Battery voltage self-adaption and accomplished a variety of high-precise tests. (C) 2014 Elsevier B.V. All rights reserved.