Detection of pump cavitation/blockage and seal failure via current signature analysis

A method for monitoring the condition of a pump driven by a motor, comprising the steps of:

• collecting and monitoring a first sample of current data signal relating to the operation of the pump;
• inputting the first sample of current data signal to a neural network,
• collecting and monitoring a second sample of current data signal relating to the operation of the pump;
• and
• inputting the second sample of current data signal to the neural network, wherein any differences between the first signal and the second signal will be generated as a change in condition signal by the neural network, any change of condition signal representing a pump fault condition.

A stand alone decision module adapted to receive a current signal from a machine and facilitate diagnosing the state of the machine by determining if the current signal contains fault data relating to the state of the machine, the decision module comprising:

• a neural network coupleable to a sensor, the neural network adapted to synthesize a change in condition signal from the sampled current data;
• a preprocessing portion coupled to the neural network, the preprocessing portion adapted to condition the current signal prior to inputting the current signal into the neural network;
• and
• a post processing portion coupled to the neural network, the post processing portion adapted to determine whether the change in condition signal is due to a fault condition related to the state of the machine.

A system for diagnosing a plurality of pumps, each driven by a motor, comprising:

• a plurality of electrically coupleable sensors for obtaining and monitoring current data signals from each motor;
• a channel interface coupleable to the plurality of sensors, the channel interface designating a separate channel for each of the plurality of sensors;
• and
• a host computer coupleable to the channel interface, the host computer including a neural network coupleable to each channel of the channel interface, the neural network adapted to detect at least one fault relating to the operation of the plurality of pumps from the current data signals, wherein a processor of the host computer cycles through each of the channels, the processor performing classical signature analysis on each of the plurality of pumps using the current data signal for each respective pump.

A system for monitoring the condition of a pump driven by a motor, comprising:

• electrically coupleable sensor means for detecting at least one current signal from the motor, said at least one current signal containing fault attributes related to the condition of the pump;
• extraction means for extracting the fault attributes from the at least one current signal;
• processing means for determining if the fault attributes signify a fault condition of the pump;
• and
• means for communicating any fault conditions to a system operator.

A system for monitoring the condition of a pump driven by a motor, comprising:

• a sensor electrically coupleable to a power lead of the motor, the sensor adapted to obtain and monitor at least one current signal relating to the operation of the pump;
• a processor couple able to the sensor, the processor adapted to convert the at least one current signal to a frequency spectrum having a plurality of fault attributes related to the condition of the pump and preprocess the fault attributes;
• an unsupervised artificial neural network coupleable to the processor, the artificial neural network being adapted to recognize and detect any changes in the plurality of preprocessed fault attributes and provide a change of condition pattern relating to any changes in the plurality of preprocessed fault attributes;
• and
• a post processor including decision making rules for determining fault conditions of the pump based on the change of condition pattern, the post processor communicating any fault conditions to a system operator.