Battery testing and classification

Apparatus for determining the condition/classification of automotive and other batteries in accordance with one or more sensed parameters thereof comprising:

• a) means for applying a transient electrical test parameter to the terminals of a battery to be tested for no more than one second;
• b) analysis means adapted to utilise a characteristic of the battery's reaction to said test parameter as a basis for said determination;
• c) said analysis means being adapted to utilise the battery's reaction to said transient test parameter to provide a measure of the condition of said battery by analysis of the waveform shape or profile of an electrical value representative of the battery's reaction to said test parameter;
• characterised by;
• d) said analysis means not comprising a neural network and being adapted to effect said analysis by means of an algorithm adapted to identify a corresponding one of two or more battery condition classifications, of which said one classification exhibits a correlation with time-defined voltage aspects of the waveform profile of said electrical value.

A method of making battery test apparatus for classification of automotive and other batteries in accordance with one or more sensed parameters thereof, the method comprising:

• a) providing data input coupling means adapted to couple a battery to data classification means;
• b) providing said data classification means to be coupled by said coupling means to said battery so that said battery provides a data source permitting classification of said battery's state;
• c) said classification means being provided comprising a neural network and said neural network being adapted to carry out a classification step on said data accordingly;
• characterised by;
• d) providing said neural network in the form of a self-organising software network;
• e) said method further comprising causing said self-organising software network to undergo (or be based upon software which has undergone) a controlled degree of training thereof based on an input comprising a representative sample of battery data, said representative sample of battery data comprising a plurality of battery parameters and selected from the group including voltage, current, internal resistance, surface charge/capacitance and thermal parameters;
• and
• f) said method further comprising providing said battery test apparatus with test data generation means adapted to subject a given battery on test to test routines to generate test data for said classification means through said coupling means, which test data is related to said representative sample of battery data on which said software network was trained, whereby said test data may be readily classified by said data classification means, the test routines including applying plural transient loads or plural transient charges to a battery in sequence, each transient load or charge having a duration of no more than one second.

Apparatus for classification of automotive and other batteries in accordance with one or more sensed electrical parameters thereof, said apparatus comprising;

• a) data input coupling means adapted to couple a battery to data classification means;
• b) data classification means adapted to be coupled by said coupling means to a battery so that said battery provides a data source permitting classification of said battery's state;
• c) said classification means comprising a neural network adapted to carry out a classification step on said battery data accordingly;
• characterised by;
• d) said neural network comprising a self-organising software network;
• e) said self-organising network having undergone (or being based upon software which has undergone) a controlled degree of training based on data input comprising a representative sample of battery data, said representative sample of battery data comprising a plurality of battery parameters selected from the group including voltage, current, internal resistance, surface charge/capacitance and thermal parameters, said data providing a basis for the classification step required from said self-organising network;
• and
• f) said apparatus further comprising test data generation means adapted to subject a given battery on test to test routines to generate test data for said classification means through said coupling means, which test data is related to said representative sample of battery data on which said software network was trained, whereby said test data may be readily classified by said data classification means, the test routines including applying plural transient loads or plural transient charges to a battery in sequence, each transient load or charge having a duration of no more than one second.

A method for determining the condition/classification of automotive and other batteries in accordance with one or more sensed parameters thereof comprising the steps of:

• a) applying a transient electrical test parameter to the terminals of a battery to be tested for no more than one second;
• b) utilising a characteristic of the battery's reaction to said test parameter as a basis for said determination;
• c) said method further comprising the step of utilising the battery's reaction to said transient test parameter to provide a measure of the condition of said battery by analysis of the waveform shape or profile of an electrical value representative of the battery's reaction to said test parameter;
• characterised by
• d) applying at least one further transient electrical test parameter to said battery to be tested for no more than one second;
• e) said one and said further transient electrical test parameter forming a microcycle of two or more transient charges and/or loads applied to said battery in sequence;
• and
• f) analysing data relating to the battery's reaction to said microcycle by a neural network and/or an algorithm to enable classification of said battery.

Apparatus for determining the condition/classification of automotive and other batteries in accordance with one or more sensed parameters thereof comprising:

• a) means for applying a transient electrical test parameter to the terminals of a battery to be tested for no more than one second;
• b) analysis means adapted to utilise a characteristic of the battery's reaction to said test parameter as a basis for said determination;
• c) said analysis means further being adapted to utilise the battery's reaction to said transient test parameter to provide a measure of the condition of said battery by analysis of the waveform shape or profile of an electrical value representative of the battery's reaction to said test parameter;
• characterised by
• d) means for applying at least one further transient electrical test parameter to said battery to be tested for no more than one second;
• e) said one and said further transient electrical test parameters forming a microcycle of two or more transient charges and/or loads applied to said battery in sequence;
• and
• f) said analysis means being also adapted to analyse data relating to the battery's reaction to said microcycle by a neural network and/or an algorithm to enable classification of said battery.