Research On Electrochemical Impedance Spectroscopy Early Warning Technology For Big Capacity Lead-acid Battery

Nuclear power plants need a big capacity of 4000Ah-grade lead-acid battery.Nuclear-level electrical equipment is classified into nuclear safety level(referred to as class 1E)and extra-level types of equipment.The big capacity lead-acid battery connected to the emergency cooling equipment of the reactor core belongs to class 1E.In nuclear power plants,non-class 1E application of valve-regulated lead-acid(VRLA)battery is still in its infancy in domestic and foreign countries.The class 1E application of VRLA battery is the first time in domestic and foreign countries.The research on the class 1E application of 4000Ah-grade VRLA battery fills the gap in domestic and foreign industries.Electrochemical impedance spectroscopy(EIS)early warning technology is joint research that integrates material electrochemistry and power electronics.The modeling,detection,inverse calculation,and alarm design of the battery EIS are the key technologies.The EIS early warning technology researched and developed in this paper integrates the following contents:Chapter Two studies the EIS modeling technique that is centered in the average switch polarization impedance.This technique verifies a linear internal resistance model at a full state of charge in deep discharge.This model significantly improves the correlation coefficient between fitting internal resistance and battery remaining useful capacity.The average switch polarization impedance upgrades the normal linear average polarization impedance by one order and sets a constant coefficient of the components of the DC directional polarization impedance.Components of the average switch polarization impedance have a one-to-one correspondence with the components of the normal DC switch polarization impedance.The characteristic charge-transfer resistance based on the average switch polarization impedance is the parallel value of charge and discharge charge-transfer resistors,and it is also the discharge charge-transfer resistance at a full state of charge.Based on the characteristic charge-transfer resistance,this paper verifies a linear internal resistance model at a full state of charge in deep discharge.Chapter Three establishes the low-frequency and weak EIS detection technique that is centered in the fast lock-in amplifier.This technique reduces the heat generated by the discharge resistor,reduces the size and weight of the detection device,and realizes the low-frequency and weak EIS detection by the embedded board.The fast lock-in amplifier replaces the conventional low-pass filter and definite Integrator with the linear average definite Integrator and outputs frequency selection result at any moment of non-integer periods based on a short-term stable sampling signal.The discrete formula of the fast lock-in amplifier cancels the frequency variable,and its format is uniform.The phase of the reference signal in the fast lock-in amplifier is globally optimal.This technique also includes the DC pulse discharge method,develops an adaptive optimization algorithm of the phase of the reference signal,and a linear compensation algorithm for EIS.Chapter Four establishes the EIS inverse calculation technique that is centered in the vector objective function and linear interpolation search algorithm.This technique simplifies the initial value preparation and the gradient descent direction search for the inverse calculation of EIS.The root-mean-square errors of online embedded programming are smaller than the offline results of professional software.The vector objective function targets the rotation angle of the initial polarization impedance vector to be zero during the EIS inverse calculation.This equivalently replaces the regular minimization of root-mean-square error.The linear interpolation algorithm simplifies the gradient descent direction search to a one-dimensional search for the double-layer capacitance of the initial polarization impedance.This algorithm replaces the conventional graphic method and evolutionary algorithm.Chapter Five establishes the alarm design technique of battery remaining useful capacity failure that is centered in the dynamic threshold of series resistance.This technique suppresses the false alarms at the early stage of battery aging and the miss alarms at the end of battery aging.The measurement interval of wrong alarm by online embedded programming is consistent with the theoretical design.The dynamic threshold of series resistance proposes a safety margin of a morphology correction factor to represent the interval design of the wrong alarm.This technique,comparing the current series resistance of EIS with its dynamic threshold,directly gives the result of a capacity failure alarm.This technique also develops a constant voltage drop model of internal resistance at the end of deep discharge,a voltage drops model of internal resistance in deep discharge and an interactive reverse calculation method for deep discharge.This paper develops an online detection system of EIS early warning technology by embedded programming and improves the reliability of class 1E operation of big capacity4000Ah-grade lead-acid battery.This paper provides technical support for the establishment of independent innovation capability of class 1E storage battery which is compatible with the active and orderly development plan of nuclear power in China.This guarantees that the research of this paper has significant economic and social benefits.