Systems and methods for detecting and identifying arcing

A method comprising:

• determining a model probability density including
• measuring a current in a conductor when inducing arcing in the conductor,
• converting the current to a proportional voltage,
• converting the current to a digitized current at a sample frequency,
• converting the proportional voltage to a digitized voltage at the sample frequency,
• determining a waveform of a cycle of a primary load current based on the digitized current,
• determining at least one noise signal according to the determined waveform of the cycle of the primary load current and the digitized current,
• setting a time window within a cycle of the digitized voltage by setting a start time and a stop time of a time interval based on voltage amplitude at the start time and the stop time being within a predetermined value of the peak voltage of the waveform of the cycle of the digitized voltage,
• determining a probability density of the at least one noise signal according to the set time window, and
• determining the model probability density at a voltage magnitude as a reference for a positive arc detection at the voltage magnitude;
• detecting and identifying arcing including
• converting a first current from a load to a first proportional voltage,
• converting the first current to a first digitized current at a first sample frequency,
• converting the first proportional voltage to a first digitized voltage at the first sample frequency,
• determining a first waveform of a cycle of a primary load current based on the first digitized current,
• determining at least one first noise signal according to the determined first waveform of the cycle of the primary load current and the first digitized current,
• setting a first time window within a cycle of the first digitized voltage by setting a first start time and a first stop time of a first time interval based on first voltage amplitude at the first start time and the first stop time being within a first predetermined value of the peak voltage of the first waveform of the cycle of the first digitized voltage,
• determining a first probability density of the at least one first noise signal according to the first set time window, and
• comparing the probability density of the at least one first noise signal with the model probability density;
• and
• generating an output indicative of a positive arc detection based on the comparing when the first time window is set based on values of the first digitized voltage indicative of the arc.