Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Electrical signal parameter measurement system
Reexamination Certificate
2000-07-05
2003-02-25
Hoff, Marc S. (Department: 2857)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Electrical signal parameter measurement system
C702S057000, C702S081000, C702S108000, C702S117000, C702S182000
Reexamination Certificate
active
06526361
ABSTRACT:
This invention relates to a method and apparatus for testing and classification of automotive and other batteries.
In the automotive industry and likewise in other technical areas, there is a need for improvements in systems for battery testing. We have provided an advance in the art in terms of the testing system disclosed in EP 0 762 135 A2 (case 12 reference P52759EP) which discloses a method and apparatus for testing automotive electronic control units and batteries utilising neural networks to effect waveform analysis on a digitised signal. Battery testing is by waveform analysis of the battery current during transient connection of a load. A network learning stage is employed together with a recognition test routine for characteristic waveforms. This approach is based upon software simulation of the waveform analysis which may be carried out visually by a person skilled in the art to distinguish between the current or voltage profiles of various categories of batteries.
References identified in searches made with respect to the subject matter of the present application consist of the following:
U.S. Pat. No. 4,204,153
U.S. Pat. No. 5,469,528
U.S. Pat. No. 5,537,327
WO 96/35522
WO 96/05508
GB 2285317A
GB 2278452A
None of these is any more pertinent than WO 96/35522 which discloses the use of a capacitor discharge pulse as a battery test step to enable identification of battery type by reference to voltage gradient using a voltmeter of oscilloscope or digital signal processing device whereby the battery type may be recognized prior to sorting discarded batteries for recycling purposes.
EPO 772056A discloses a system for checking the charge-discharge cycles of a rechargeable battery utilizing neural networks which are trainable by collecting data during discharge cycles of a battery. The battery is tested by collecting current battery data and using the neural networks for predicting from the current data a time interval by which the battery voltage will reach a predetermined critical threshold level.
U.S. Pat. No. 5,596,260 discloses a battery testing apparatus and method utilizing an algorithm to determine the charge of a battery. Initially, a battery state model is established comprised of a number of discrete charge states ranging from full charge to no charge. The probability of the actual battery's charge for a particular charge state is specified by the battery state model for each of the charge states. Then a discharge curve specifying the battery's voltage as a function of time is determined and calibrated. Then the battery's voltage is periodically measured, and based on the measured voltage and the discharge curve, a voltage probability distribution over the range of battery states is computed and the battery discharge model is updated to produce a discharge model having smaller variance. The charge of the battery corresponding to the mean value of the battery discharge model is then displayed to the user.
FIRST ASPECT
In a first aspect of the present invention we are seeking to provide a further technical advance in battery test systems whereby the approach of simulating the visual analysis carried out by a technical person is replaced by an alternative approach in which the technical attributes of computer systems are employed in a manner which exploits their inherent advantages rather than by seeking to constrain them to simulate a human analytical approach.
Accordingly, we have sought to utilise the capacity of software systems to process and analyse data relating to multiple parameters, in a way which the human brain finds difficult. In accordance with this approach, we realised that the multi-layer perceptron neural networks of our above-identified EP '135 A2 specification had limitations in terms of their requirement for the input of an approximation to the required analytical answer in any case.
In accordance with this new approach to battery analysis, we sought to provide a system in which multiple battery parameters, not all of them necessarily being electric parameters, such as thermal parameters, would be fed into the system and the system would be capable of recognising particular characteristics of the complex data fed in and thereby to effect an efficient classification step, perhaps on the basis of relatively limited data as compared with that required for waveform analysis.
An object of this aspect of the present invention is to provide a method and apparatus for classification of automotive and other batteries in accordance with one or more sensed parameters thereof, offering improvements in relation to one or more of the matters discussed above, or generally.
According to the invention there is provided a method of making battery test apparatus and there is also provided apparatus for classification of automotive and other batteries, as defined in the accompanying claims.
In an embodiment of the invention the system is provided with a neural network in the form of a self-organising network, namely a Kohonen network. In the embodiment, this software network has undergone (or is based upon software which has undergone) a controlled degree of training. That training is based on a data input comprising a representative sample of battery data. Such a representative sample of battery data comprises a plurality of battery parameters selected from the group including voltage, current, internal resistance, surface charge/capacitance, and thermal parameters. This controlled degree of self-training provides a basis for the classification step required from the self-organising network.
In the embodiment also, the method and apparatus provides test data generation means as part of the system and which is adapted to subject a given battery on test to test routines to generate test data for the classification means. This test data is related to the representative sample battery data on which the software network was trained. For example, the software network may be trained on data from a sequence of transient battery loads or transient battery charging routines, these being separated likewise by relatively transient intervals. Where such a routine has been included in the training data, then the test data generation means included in the battery test apparatus is arranged to generate corresponding data which leads to relatively efficient battery classification.
In connection with the use of multiple transient battery loads in the battery test routine, this approach has practical significance in relation to the capacitance aspects of battery construction, namely that the sequence of multiple transient loads is capable of discharging the surface charge of the battery which is present due to the parallel plate construction of the latter and which can otherwise provide misleading data arising from the (for example) 15 volts potential difference theron arising from alternator ripple.
The embodiments of the invention are software-intensive rather than hardware-intensive when compared with previously known battery testing systems.
SECOND ASPECT
In accordance with a second aspect of the invention, there is a need in the automotive and other industries for improved methods and apparatus for the testing and classification of batteries, notably the provision of systems which permit the in-situ testing of batteries using a minimum of hardware and making low power requirements while offering hand-held operational characteristics and relatively rapid test/classification results, and an object of this aspect of the present invention is to provide a method and apparatus offering improvements in relation to one or more of these requirements, or indeed generally.
According to the invention there is provided a method and apparatus for the testing and/or classification of automotive and other batteries as defined in the accompanying claims.
In an embodiment of the invention described below there is provided a method and apparatus in which a battery is subjected to a microcycle test procedure and the result of that procedure is subjected to analysis by a neur
Clifton Jason Mark
Jones Barbara L.
Smith Paul
Hoff Marc S.
Seyfarth Shaw
Snap-on Equipment Limited
Tsai Carol S. W.
LandOfFree
Battery testing and classification does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Battery testing and classification, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Battery testing and classification will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3148817