Electrochemical energy storage remaining life estimation method and system

Abstract

The present application relates to the technical field of electrical performance monitoring, in particular to a method and system for estimating the residual life of electrochemical energy storage. In the present application, by applying multi-stage voltage step excitation and capturing the maximum offset of instantaneous current, high sensitivity characteristics reflecting the internal electrochemical dynamic characteristics of the energy storage unit can be obtained, avoiding the limitations of traditional methods relying on capacity attenuation, a lagging indicator for evaluation, realizing early insight into performance changes. At the same time, by collecting the ambient temperature in real time and correcting the dynamic response characteristics using a preset correction coefficient, the interference of working temperature fluctuations on performance evaluation is effectively removed, ensuring the consistency and comparability of the characteristic data under different environmental conditions. Further, by constructing a performance characteristic evolution offset atlas and specifically identifying the consistent and directional change sequence, the irreversible degradation trend caused by internal aging is accurately locked, and the influence of random noise and short-term fluctuations is filtered.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electrical performance monitoring, and in particular to an electrochemical energy storage residual life estimation method and system. BACKGROUND

[0002] The technical field of electrical performance monitoring refers to the detection and monitoring technology of electrical performance in various types of electrical energy storage and conversion systems, which is widely used in batteries, electrochemical energy storage devices and other devices related to power transmission and storage.

[0003] Among them, the electrochemical energy storage residual life estimation method refers to predicting the residual service life of the energy storage system by modeling the capacity decay law during the charging and discharging process of the battery or energy storage system, combined with the known battery aging law.

[0004] The existing technology mainly relies on the analysis and application of the law of battery capacity decay, which has inherent defects. The aging law relied on is usually based on standard and constant working conditions in the laboratory, while the energy storage system in actual operation often faces complex temperature changes and irregular load shocks. Directly applying the aging law based on laboratory conditions to the variable field environment will cause significant deviations in the evaluation results. For example, an energy storage system running in a low-temperature environment will temporarily reduce its available capacity. The existing technology may misjudge this non-permanent performance decline as structural aging, thus triggering premature scrap warning and causing unnecessary economic losses. In addition, capacity decay itself is a slow cumulative result, and as an evaluation indicator, it has obvious hysteresis. When the system detects a significant decrease in capacity, significant and irreversible material aging may have occurred inside the energy storage cell, which makes it difficult for the existing technology to provide early warning and cannot provide sufficient advance for maintenance decisions. Moreover, this method tends to use universal aging laws, ignoring individual differences between energy storage cells of the same batch due to manufacturing differences and different running histories, resulting in insufficient accuracy in predicting the life of individual cells. SUMMARY

[0005] The purpose of the present application is to solve the shortcomings in the prior art and propose an electrochemical energy storage residual life estimation method and system.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme: an electrochemical energy storage residual life estimation method, comprising the following steps:

[0007] S1: Apply an externally controlled multi-level voltage step excitation to the electrochemical energy storage cell to be tested, and collect the terminal voltage and current data of the electrochemical energy storage cell in real time during the step excitation process, construct a voltage-current response sequence, extract temperature-sensitive features from the sequence, and construct an electrochemical performance response set;

[0008] S2: Collecting temperature information of the current electrochemical energy storage cell, error correcting temperature sensitive features in the electrochemical performance response set, and generating a temperature corrected performance response set;

[0009] S3: Constructing a feature sequence of the current operation cycle according to the temperature corrected performance response set of the electrochemical energy storage cell in time sequence, comparing the same type of temperature sensitive features of the electrochemical energy storage cell in the previous operation cycle, judging the shift trend of each temperature sensitive feature, and constructing a performance shift feature map;

[0010] S4: Identifying the shift trend sequence in which the change direction of the temperature sensitive features in the performance shift feature map remains consistent, and constructing a stage degradation index sequence;

[0011] S5: Calculating the remaining life estimation value based on the stage degradation index sequence and the total operation time of the electrochemical energy storage cell, and obtaining the electrochemical energy storage remaining life estimation result.

[0012] As a further scheme of the present application, the voltage current response sequence includes voltage change rate and current maximum deviation, the temperature corrected performance response set includes corrected voltage change rate and corrected current maximum deviation, the performance shift feature map includes voltage change rate shift trend and current maximum deviation shift trend, the stage degradation index sequence is specifically trend consistent interval, feature item identification and time sequence index, and the electrochemical energy storage remaining life estimation result includes life change trend, current remaining life value and estimation time node.

[0013] As a further scheme of the present application, the obtaining step of the electrochemical performance response set is specifically:

[0014] S111: A multi-stage voltage step excitation is applied to the electrochemical energy storage cell to be tested, the electrochemical energy storage cell is a lithium iron phosphate battery, and the voltage and current data of the electrochemical energy storage cell are collected in real time during the step excitation process to construct a voltage current response sequence;

[0015] S112: The voltage step occurrence section in the voltage current response sequence is identified, the voltage change rate in each voltage step occurrence section is calculated, the maximum deviation of the current in the corresponding section is obtained synchronously, the voltage change rate and the current maximum deviation of all sections are collected as temperature sensitive features, and dynamic response feature data is obtained;

[0016] S113: call the voltage change rate and the current maximum offset of all segments in the dynamic response feature data, integrate the two temperature sensitive features into a structured data set, combine the structured data sets obtained under multiple excitations, and construct an electrochemical performance response set.

[0017] As a further scheme of the present application, the step of obtaining the temperature-corrected performance response set specifically comprises:

[0018] S211: acquire the ambient temperature of the electrochemical energy storage cell in the response process through a temperature sensor, record and integrate the temperature data collected at each time point, and establish ambient temperature information;

[0019] S212: according to the temperature interval matching rule, compare the temperature data in the ambient temperature information with the preset multiple temperature intervals one by one, lock the unique temperature interval, extract the corresponding preset correction coefficient from the temperature interval, and obtain the matching correction coefficient value;

[0020] S213: error correction is performed on the temperature sensitive features in the electrochemical performance response set, the voltage change rate and the current maximum offset are called, and the matching correction coefficient value is used for correction processing, the current maximum offset after temperature and voltage platform effect correction is calculated, and the temperature-corrected performance response set is generated.

[0021] As a further scheme of the present application, the step of obtaining the performance offset feature map specifically comprises:

[0022] S311: arrange the temperature-corrected performance response set of the electrochemical energy storage cell in time sequence, and establish a current running period feature sequence;

[0023] S312: call the same temperature sensitive features of the previous running period of the electrochemical energy storage cell, and perform difference and change rate comparison of the same index points for the current running period feature sequence, judge the offset trend of each temperature sensitive feature, and obtain feature offset trend information;

[0024] S313: combine the offset trend of each temperature sensitive feature extracted from the feature offset trend information, and construct a performance offset feature map.

[0025] As a further scheme of the present application, the step of obtaining the stage degradation index sequence specifically comprises:

[0026] S411: call the offset trend information of each temperature sensitive feature in the performance offset feature map, judge the direction of the time sequence data of each feature, filter the sequence paragraphs with unchanged change direction in continuous periods, record the association of the temperature sensitive feature type, construct a consistent offset sequence;

[0027] S412: For each sequence paragraph in the consistency offset sequence, extract the temperature-sensitive feature type, the duration of the sequence paragraph trend, and the offset amplitude value, integrate the three pieces of information into an independent structured index item, collect all the index items of the sequence paragraphs to obtain the degradation trend quantification data;

[0028] S413: Based on the degradation trend quantification data, call all the index items in the collection, arrange and summarize according to the time sequence identifier attached to each index item, and construct a phased degradation index sequence.

[0029] As a further scheme of the present application, the process of directionally judging the time series data of each feature is specifically:

[0030] The temperature-sensitive feature value of the current running period is subjected to difference operation with the same temperature-sensitive feature value of the previous running period, and a positive offset threshold and a negative offset threshold are set;

[0031] If the difference operation result is greater than the positive offset threshold, the change direction of the temperature-sensitive feature is judged as positive offset;

[0032] If the difference operation result is less than the negative offset threshold, the change direction of the temperature-sensitive feature is judged as negative offset;

[0033] If the difference operation result is between the negative offset threshold and the positive offset threshold, the change direction of the temperature-sensitive feature is judged as no offset.

[0034] As a further scheme of the present application, the acquisition step of the electrochemical energy storage remaining life estimation result is specifically:

[0035] S511: Based on the phased degradation index sequence of the electrochemical energy storage cell and the obtained total running time data, the two are time-point paired, the change amplitude of the index in each stage specified time is calculated, and the phased index change amplitude information is obtained;

[0036] S512: The phased index change amplitude information is constructed into a change trend curve, an extrapolation operation is performed on the degradation of the future change trend curve, and is calibrated in combination with the current total running time data to generate a life trend extension result;

[0037] S513: Calculate the time span required for the life trend extension result to extend from the current time point to the preset life termination threshold, take the time span value as the estimation value of the remaining life of the electrochemical energy storage cell, and obtain the electrochemical energy storage remaining life estimation result.

[0038] As a further scheme of the present application, the process of performing extrapolation operation on the recession condition of the future change trend curve is specifically:

[0039] A latest continuous data segment in time is intercepted from the change trend curve;

[0040] A recession prediction curve is established by performing curve fitting on the latest continuous data segment in time in a polynomial fitting manner.

[0041] An electrochemical energy storage remaining life estimation system for performing the electrochemical energy storage remaining life estimation method described above, the system comprising:

[0042] A voltage response acquisition module, which acquires the terminal voltage and current data of the electrochemical energy storage unit in real time during the step excitation process by applying an externally controlled multi-stage voltage step excitation to the electrochemical energy storage unit to be tested, constructs a voltage current response sequence, extracts temperature sensitive features from the sequence, and constructs an electrochemical performance response set;

[0043] A temperature correction processing module, which acquires the temperature information of the current electrochemical energy storage unit to correct the temperature sensitive features in the electrochemical performance response set for errors and generate a temperature corrected performance response set;

[0044] A performance offset analysis module, which constructs a feature sequence of the current operation cycle in time sequence according to the temperature corrected performance response set of the electrochemical energy storage unit, compares the same type of temperature sensitive features of the previous operation cycle of the electrochemical energy storage unit, judges the offset trend of each temperature sensitive feature, and constructs a performance offset feature map;

[0045] A recession trend identification module, which identifies the offset trend sequence in which the change directions of the temperature sensitive features in the performance offset feature map remain consistent and constructs a stage recession index sequence;

[0046] A remaining life estimation module, which calculates a remaining life estimation value based on the stage recession index sequence and the total operation time of the electrochemical energy storage unit to obtain an electrochemical energy storage remaining life estimation result.

[0047] Compared with the prior art, the present application has the following advantages and positive effects:

[0048] In the present application, by applying multi-stage voltage step excitation and capturing the maximum deviation of instantaneous current, high sensitivity characteristics reflecting the internal electrochemical dynamic characteristics of the energy storage unit can be obtained, avoiding the limitations of traditional methods relying on capacity attenuation, a lagging indicator, for evaluation, realizing early insight into performance changes, at the same time, by collecting the ambient temperature in real time and using the preset correction coefficient to correct the dynamic response characteristics, the interference of working temperature fluctuation on performance evaluation is effectively removed, ensuring the consistency and comparability of the characteristic data under different environmental conditions, and then, by constructing the deviation spectrum of the performance characteristics over time, and specially identifying the consistent change sequence, the irreversible degradation trend caused by internal aging is accurately locked, and the influence of random noise and short-term fluctuations is filtered, and finally, based on the latest stage of the degradation index change amplitude data, polynomial fitting and extrapolation are carried out, and a degradation prediction curve that can be dynamically adjusted according to the recent actual aging rate of the energy storage unit is established. This method avoids the use of general aging rules and greatly improves the individualization and accuracy of the remaining life estimation. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 The working flowchart of the present application is shown in the figure;

[0050] Figure 2 The flowchart of step S1 of the present application is shown in the figure;

[0051] Figure 3 The flowchart of step S2 of the present application is shown in the figure;

[0052] Figure 4 The flowchart of step S3 of the present application is shown in the figure;

[0053] Figure 5 The flowchart of step S4 of the present application is shown in the figure;

[0054] Figure 6 The flowchart of step S5 of the present application is shown in the figure. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0056] Please refer to Figure 1 The present application provides a technical scheme: an electrochemical energy storage remaining life estimation method, comprising the following steps:

[0057] S1: by applying an external controlled multi-stage voltage step excitation at both ends of the to-be-tested electrochemical energy storage cell, collecting the terminal voltage and current data of the electrochemical energy storage cell in real time during the step excitation process, constructing a voltage-current response sequence, extracting temperature-sensitive features from the sequence, and constructing an electrochemical performance response set;

[0058] S2: collecting the temperature information of the current electrochemical energy storage cell, correcting the temperature-sensitive features in the electrochemical performance response set, and generating a temperature-corrected performance response set;

[0059] S3: constructing a feature sequence of the current operation cycle according to the temperature-corrected performance response set of the electrochemical energy storage cell, comparing the same temperature-sensitive features of the previous operation cycle of the electrochemical energy storage cell, judging the shift trend of each temperature-sensitive feature, and constructing a performance shift feature map;

[0060] S4: identifying the shift trend sequence in which the change direction of the temperature-sensitive features in the performance shift feature map remains consistent, and constructing a stage degradation index sequence;

[0061] S5: calculating the remaining life estimate value based on the stage degradation index sequence and the total operation time of the electrochemical energy storage cell, and obtaining the electrochemical energy storage remaining life estimation result;

[0062] The voltage-current response sequence includes voltage change rate and current maximum deviation, the temperature-corrected performance response set includes corrected voltage change rate and corrected current maximum deviation, the performance shift feature map includes voltage change rate shift trend and current maximum deviation shift trend, the stage degradation index sequence specifically includes consistent trend interval, feature item identifier, and time sequence index, and the electrochemical energy storage remaining life estimation result includes life change trend, current remaining life value, and estimation time node.

[0063] Please refer to Figure 2 , the acquisition step of the electrochemical performance response set is specifically:

[0064] S111: by applying an external controlled multi-stage voltage step excitation at both ends of the to-be-tested electrochemical energy storage cell, the electrochemical energy storage cell is a lithium iron phosphate battery, collecting the terminal voltage and current data of the electrochemical energy storage cell in real time during the step excitation process, and constructing a voltage-current response sequence;

[0065] A lithium iron phosphate battery with a rated capacity of 100 ampere-hours and a nominal voltage of 3.2 volts was operated by applying externally controlled multi-step voltage step excitation across the battery. The excitation signal was generated by a programmable DC power supply with an output voltage accuracy of 0.001 volts. First, the lithium iron phosphate battery was left to rest in an environment of 25 degrees Celsius, and the open circuit voltage of the lithium iron phosphate battery was continuously monitored. When the fluctuation of the open circuit voltage was less than 0.002 volts within 10 consecutive minutes, it was determined that the open circuit voltage of the lithium iron phosphate battery had stabilized, and the recorded stable voltage was 3.200 volts. The excitation process began by applying a constant voltage of 3.250 volts to the lithium iron phosphate battery through the programmable DC power supply, and maintained for 30 seconds. During this 30-second constant voltage period, the terminal voltage and current of the lithium iron phosphate battery were synchronously recorded by a Hall effect current sensor with a sampling frequency of 1000 hertz in series with the lithium iron phosphate battery, and a voltage sensor with a sampling frequency of 1000 hertz in parallel with the lithium iron phosphate battery. Every 0.001 seconds, the data acquisition device collected and recorded a voltage value and a current value. For example, the terminal voltage recorded at 10.001 seconds was 3.250 volts, and the current was 0.512 amperes; the terminal voltage recorded at 10.002 seconds was 3.250 volts, and the current was 0.511 amperes. After completing the 30-second constant voltage phase, the externally controlled programmable DC power supply linearly raised the output voltage from 3.250 volts to 3.300 volts within 0.01 seconds, completing the first voltage step. Then, 30 seconds were maintained at 3.300 volts, and voltage and current data were continuously collected. The process was repeated, and the voltage was stepped to 3.350 volts, 3.400 volts, 3.450 volts, and 3.500 volts, respectively, each voltage platform being maintained for 30 seconds. The entire multi-step voltage step excitation process lasted about 2.5 minutes, and finally a collection of hundreds of thousands of data points containing time stamps, voltage readings, and current readings arranged in chronological order was obtained, constructing a voltage-current response sequence.

[0066] S112: In the voltage-current response sequence, identify the section where the voltage step occurs, calculate the voltage change rate in each voltage step occurrence section, and simultaneously obtain the maximum deviation of the current in the corresponding section, collect the voltage change rate and the maximum deviation of the current of all sections as temperature-sensitive features, and obtain dynamic response feature data;

[0067] In the voltage-current response sequence, the voltage data is scanned point by point to identify the section where the voltage changes rapidly. The specific identification process is as follows: starting from the first data point of the voltage-current response sequence, the voltage difference between the current sampling point and the next adjacent sampling point is calculated, and the voltage difference is divided by the sampling time interval of 0.001 seconds to obtain the instantaneous voltage change rate. A voltage change rate threshold is set in advance to define the start and end of the voltage step. The setting of the voltage change rate threshold is based on the continuous voltage monitoring of a brand new, healthy lithium iron phosphate battery of the same model for 1 hour in a completely static state. The normal voltage fluctuation range of the lithium iron phosphate battery caused by internal electrochemical reactions without external excitation is recorded. The maximum voltage fluctuation rate monitored in the experiment is 0.05 volts per second. To ensure the accuracy of the identification, the voltage change rate threshold is set to 0.1 volts per second, which is twice the maximum fluctuation rate observed. During the scanning process, when the calculated instantaneous voltage change rate continuously exceeds 0.1 volts per second, the timestamp of the first data point that exceeds the threshold is marked as the start of the voltage step occurrence section; when the instantaneous voltage change rate falls below 0.1 volts per second, the timestamp of the first data point that falls below the threshold is marked as the end of the voltage step occurrence section. For example, in the process of jumping from 3.250 volts to 3.300 volts, the voltage changes by 0.050 volts in 0.01 seconds, and the average voltage change rate in this section is 0.050 volts divided by 0.01 seconds, which is 5 volts per second. After identifying such a voltage step occurrence section, the corresponding current data section is locked in synchronization with the timestamp. In this current section, all current values are traversed from beginning to end to search for and record the maximum current value. For example, in the above step section, the current rises from the stable value of 0.4 amperes before the step to the peak value of 8.5 amperes, and then falls. Therefore, the maximum offset of the current is the difference between the peak current of 8.5 amperes found and the stable current of 0.4 amperes just before the step, and the calculation result is 8.1 amperes. The calculated voltage change rate (5 volts per second) and the obtained maximum current offset (8.1 amperes) are stored as a data pair. The above section identification, rate calculation, and offset acquisition process is repeated for all identified voltage step sections (from 3.250 volts to 3.300 volts, from 3.300 volts to 3.350 volts, etc.) in the entire excitation process. The voltage change rates of all five sections and the corresponding current maximum offsets are collected as a set of temperature-sensitive features, and the dynamic response feature data is obtained.

[0068] S113: Call the voltage change rate and current maximum offset of all sections in the dynamic response feature data, integrate the two temperature-sensitive features into a structured data set, merge the structured data sets obtained under multiple excitations, and construct an electrochemical performance response set;

[0069] Call dynamic response feature data, dynamic response feature data contains all voltage step segment voltage rate of change and current maximum offset. With a complete five-level voltage step excitation as an example, the data pair obtained is: [(first segment voltage rate of change, first segment current maximum offset), (second segment voltage rate of change, second segment current maximum offset), …, (fifth segment voltage rate of change, fifth segment current maximum offset)]. The specific values are: [(5 volts per second, 8.1 amps), (5 volts per second, 7.9 amps), (5 volts per second, 7.7 amps), (5 volts per second, 7.5 amps), (5 volts per second, 7.3 amps)]. The voltage rate of change and the current maximum offset, two temperature sensitive features, are integrated into a structured data set, which is labeled with the timestamp of the current excitation operation, for example, "3:40 on September 29th". In order to construct a response set that can reflect the long-term changes of lithium iron phosphate battery performance, such excitation operations are repeated at a fixed 4-hour interval. In the next excitation cycle, i.e. "7:40 on September 29th", the steps of S111 and S112 are executed again to obtain a new set of dynamic response feature data, for example: [(5 volts per second, 8.09 amps), (5 volts per second, 7.89 amps), (5 volts per second, 7.69 amps), (5 volts per second, 7.49 amps), (5 volts per second, 7.29 amps)]. Then, the newly obtained structured data set is merged with all previous data sets in chronological order. The merging operation is specifically to append the new structured data set to the end of the list storing all historical sets, and to construct the electrochemical performance response set.

[0070] Please refer to Figure 3 , the temperature corrected performance response set acquisition step is specifically:

[0071] S211: Obtain the environmental temperature of the electrochemical energy storage unit during the response process through the temperature sensor, record and integrate the temperature data collected at each time point, and establish the environmental temperature information;

[0072] The surface temperature of the lithium iron phosphate battery during the response process is obtained by a PT100 type temperature sensor with an accuracy of 0.1 degrees Celsius. To ensure the accuracy of temperature measurement, the temperature sensor is tightly attached to the center of the electrochemical energy storage cell shell using heat-conducting silicone grease, which can best reflect the overall temperature of the electrochemical energy storage cell. The sampling frequency of the temperature sensor is set to be consistent with the sampling frequency of the voltage and current sensors, both being 1000 Hz. During the entire process of the multi-stage voltage step excitation of S111, the temperature sensor synchronously performs data acquisition. Each collected temperature data point is attached with a time stamp accurate to milliseconds, ensuring accurate correspondence in time with the voltage and current data. For example, at the time point "September 29, 3:40:15.123", in addition to the voltage and current values, the temperature value is also collected as 25.2 degrees Celsius. During the entire 2.5 minute excitation process, more than 150,000 temperature data points will be collected. These temperature data with time stamps are recorded and integrated to form a temperature time sequence parallel to the voltage and current response sequence. All temperature readings in a single excitation period are arithmetically averaged to calculate a representative ambient temperature, which is to smooth out minor temperature fluctuations that may occur in a short period of time and obtain a temperature value that represents the stable working condition during the entire excitation process. For example, the average temperature calculated by adding the values of 150,000 temperature data points and dividing by the total number of data points is 25.3 degrees Celsius. The representative ambient temperature of each excitation period and its corresponding excitation period starting time stamp are stored to establish the ambient temperature information.

[0073] S212: According to the temperature interval matching rule, the temperature data in the ambient temperature information is compared with the preset multiple temperature intervals one by one, the unique temperature interval to which it belongs is locked, and the corresponding preset correction coefficient is extracted from the temperature interval to obtain the matching correction coefficient value;

[0074] The matching according to the preset temperature intervals is based on the calibration experimental data that the lithium iron phosphate battery has significant differences in electrochemical characteristics at different temperatures. The specific process of the calibration experiment is as follows: a brand new lithium iron phosphate battery of the same model as the battery to be tested is placed in a high-precision temperature control box, and the ambient temperature is set to 0 degrees Celsius, 10 degrees Celsius, 25 degrees Celsius, and 40 degrees Celsius in turn. At each temperature point, first, the battery is kept constant for 2 hours to ensure that the internal temperature of the battery is uniform and stable, and then the steps of S111 and S112 are executed to obtain the dynamic response characteristic data of the battery at the corresponding temperature. The maximum current offset measured at 25 degrees Celsius is taken as the reference (at this time the correction coefficient is defined as 1.000). Assuming that the maximum current offset measured at 25 degrees Celsius is 8.1 A under a voltage step of 3.250 V to 3.300 V, and the value measured at 10 degrees Celsius is 6.5 A. The correction coefficient at 10 degrees Celsius is calculated as the reference value 8.1 A divided by the measured value 6.5 A, which is about 1.246. In this way, a mapping relationship between temperature intervals and preset correction coefficients is established. For example, temperature interval one: -10.0 degrees Celsius to 5.0 degrees Celsius, correction coefficient 1.450; temperature interval two: 5.1 degrees Celsius to 15.0 degrees Celsius, correction coefficient 1.246; temperature interval three: 15.1 degrees Celsius to 30.0 degrees Celsius, correction coefficient 1.000; temperature interval four: 30.1 degrees Celsius to 45.0 degrees Celsius, correction coefficient 0.885. The temperature data in the environmental temperature information established by S211, such as 25.3 degrees Celsius obtained above, is compared with these preset temperature intervals one by one. The comparison process is as follows: first, determine whether 25.3 is in interval one, no; then determine whether it is in interval two, no; then determine whether it is in interval three (15.1<=25.3<=30.0), yes. Therefore, the only temperature interval to which it belongs is temperature interval three. Then, the corresponding preset correction coefficient 1.000 in temperature interval three is extracted to obtain the matching correction coefficient value.

[0075] S213: Error correction is performed on the temperature-sensitive characteristics in the electrochemical performance response set by calling the voltage change rate and the maximum current offset, and using the matching correction coefficient value for correction processing to calculate the maximum current offset after temperature and voltage platform effect correction, and generate a temperature-corrected performance response set;

[0076] The error correction is performed for the temperature-sensitive feature of the electrochemical performance response set. A structured data set of the electrochemical performance response set is called, such as [ (5 volts per second, 8.1 amps), (5 volts per second, 7.9 amps), (5 volts per second, 7.7 amps), (5 volts per second, 7.5 amps), (5 volts per second, 7.3 amps) ] obtained at an average temperature of 25.3 degrees Celsius. At the same time, the matching correction coefficient value 1.000 obtained by calling S212 is obtained. The correction process is only performed for the temperature-sensitive feature of the maximum current deviation, and no correction is performed for the voltage change rate. The reason for not correcting the voltage change rate is that the voltage change rate is precisely controlled by an external excitation device (a programmable direct current power supply), which is an input signal applied to the battery, and its value is not affected by the state of the battery itself and the ambient temperature. The correction calculation formula is: wherein the explanation of each letter is as follows: : represents the maximum current deviation after temperature and voltage plateau effect correction. The subscript corr means "corrected". This value is the final result of the calculation, and is intended to reflect the equivalent performance of the lithium iron phosphate battery at a standard temperature of 25 degrees Celsius and a reference voltage plateau, with units of amps (A). : represents the original maximum current deviation actually measured in S112. The subscript raw means "raw". This value is the direct measurement data without any processing, which will be affected by the current actual ambient temperature and the voltage plateau, with units of amps (A). : represents a specified level in a multi-level voltage step excitation. In the current scenario, the excitation process contains five voltage step segments from 3.250 volts to 3.500 volts, so the value of n is an integer from 1 to 5, to distinguish the response characteristics at different voltage steps. : represents the temperature correction coefficient. The subscript T represents "temperature". This coefficient is a dimensionless value that quantifies the effect of ambient temperature on the internal impedance of the lithium iron phosphate battery, and then corrects the current response. The coefficient value is obtained according to the mapping relationship between the temperature interval and the correction coefficient established in S212. : represents the preset temperature interval in S212. According to the average ambient temperature calculated in S211, it is matched to a unique temperature interval. For example, when the temperature is 12.0 degrees Celsius, it is matched to temperature interval two, so m is 2. : represents the temperature correction weight coefficient. Subscript T stands for "Temperature". This coefficient is a dimensionless value, which is used to adjust the proportion of the temperature correction term in the whole correction formula. The principle followed in setting this weight is: the greater the influence of a physical effect on the measurement result, the greater the corresponding weight coefficient should be. Through the calibration experiment in S212, it is found that when the ambient temperature changes from 25 degrees Celsius to 10 degrees Celsius, the change range of the maximum current deviation is much larger than that between different voltage platforms (such as 3.300 volts and 3.400 volts) at constant temperature. This shows that temperature is the main factor affecting this characteristic parameter, so it is assigned a weight of up to 0.9 to ensure that the temperature correction effect dominates in the total correction. : represents the voltage correction coefficient. Subscript V stands for "Voltage". This coefficient is a dimensionless value, which is used to eliminate the influence of different voltage step platforms on the maximum current deviation. Because even at constant temperature, the polarization effect of lithium iron phosphate battery at different states of charge (approximately reflected by different voltage platforms) also exists. : represents the voltage correction weight coefficient. Subscript V stands for "Voltage". This coefficient is a dimensionless value, which is used to adjust the proportion of the voltage correction term in the whole correction formula. Its setting follows the same principle as . The calibration experiment data shows that the influence of voltage platform on is secondary and small. In order to reflect this fact and ensure that the sum of the total weights is 1.0 ( ), the voltage correction weight is set to 0.1. This makes the correction of the voltage platform as a fine tuning to supplement the correction result mainly determined by the temperature.

[0077] Suppose the average temperature measured at another time point is 12.0 degrees Celsius, which falls into temperature interval two (5.1 degrees Celsius to 15.0 degrees Celsius), and its matching temperature correction coefficient is 1.246. The original maximum current deviation of the first voltage step section (n = 1, from 3.250 volts to 3.300 volts) measured at that time is 6.5 amperes. The voltage correction coefficient corresponding to this voltage step section is preset to 1.02. The parameters are brought in for calculation: amperes. By performing this correction process on the current maximum deviation of all data points in the electrochemical performance response set, the original data affected by the actual working temperature and voltage platform are all converted into equivalent performance data under the calibration temperature (25 degrees Celsius) and the reference voltage platform, generating the temperature-corrected performance response set.

[0078] Please refer toFigure 4 The obtaining step of the performance deviation feature map is specifically as follows:

[0079] S311: Arranging the temperature-corrected performance response set of the electrochemical energy storage unit in time sequence to establish a current running cycle feature sequence;

[0080] The temperature-corrected performance response set is strictly arranged according to the original recording of the excitation operation time sequence. The temperature-corrected performance response set contains a structured data set obtained after each excitation operation, which has eliminated the temperature influence. Each data set in the temperature-corrected performance response set is associated with a unique time stamp. For example, the data record starts from “3:40 on September 29th”, and the subsequent record points are “7:40 on September 29th”, “11:40 on September 29th”, etc., and the time interval is 4 hours. These data sets are organized into a sequence in the order of time stamp from far to near. Each item in this sequence represents the dynamic performance of the lithium iron phosphate battery at a specific time under standard temperature (25 degrees Celsius). This complete and ordered sequence is the basis for subsequent analysis of the performance degradation trend of the lithium iron phosphate battery, and establishes the current running cycle feature sequence. The current running cycle feature sequence is represented as an ordered list in the data structure, and each element in the list is a data object containing a time stamp and five corrected current maximum deviation values.

[0081] S312: Retrieving the same temperature-sensitive features of the previous running cycle of the electrochemical energy storage unit, and performing a difference and change rate comparison of the same index points for the current running cycle feature sequence to judge the deviation trend of each temperature-sensitive feature and obtain the feature deviation trend information;

[0082] The temperature-sensitive characteristic of the same kind in the previous operation cycle of the electrochemical energy storage cell is called. For each data point in the characteristic sequence of the current operation cycle (except the first data point), the difference value and the change rate of the index point are compared. The comparison process is as follows: select the data point in the characteristic sequence of the current operation cycle located at "7:40 on September 29", and the maximum current offset after temperature correction corresponding to the first voltage step in the data point is 8.09 A. Call the data point immediately before in time in the characteristic sequence of the current operation cycle, that is, the data point at "3:40 on September 29", and find the corresponding value at the same index position (the first voltage step) is 8.10 A. First, calculate the difference value, subtract the value of the previous cycle from the value of the current cycle: 8.09 A minus 8.10 A, the difference value is -0.01 A. Then calculate the change rate, divide the calculated difference value by the value of the previous cycle: -0.01 A divided by 8.10 A, the result is about -0.00123, or -0.123%. This set of calculation results (difference value -0.01 A, change rate -0.123%) collectively describes the offset trend of the specific temperature-sensitive characteristic from the previous cycle to the current cycle. Repeat this operation for all five temperature-sensitive characteristics (i.e. the maximum current offset under five different voltage steps) in the "7:40 on September 29" data point. And, for each data point in the entire characteristic sequence of the current operation cycle (from the second point), perform comparison with the previous data point to judge the offset trend of each temperature-sensitive characteristic, and obtain the characteristic offset trend information.

[0083] S313: Combine the offset trend of each temperature-sensitive characteristic extracted in the characteristic offset trend information to construct a performance offset characteristic map;

[0084] The acquired feature offset trend information is systematically combined. For any data point (except the first one) in the current running period feature sequence, S312 has calculated an offset trend (consisting of a difference value and a change rate) for each temperature-sensitive feature inside it. For example, at the time point of "7:40 on September 29", the offset trend information obtained for the maximum offset amount of the current of the five voltage steps is combined into a record, the content of which is: feature one (offset difference value -0.01 amp, offset change rate -0.123%), feature two (offset difference value -0.01 amp, offset change rate -0.127%), feature three (offset difference value -0.01 amp, offset change rate -0.130%), feature four (offset difference value -0.01 amp, offset change rate -0.133%), and feature five (offset difference value -0.01 amp, offset change rate -0.135%). The five sets of information describing the offset trend are combined as a whole data unit and associated with the time stamp "7:40 on September 29". Such information extraction and combination operations are performed for all time points in the current running period feature sequence (starting from the second time point). Finally, the offset trends of each temperature-sensitive feature extracted at all time points are integrated to form a multi-dimensional, time-sequenced data set. This data set is the performance offset feature map, which has a time sequence structure, and each element of the sequence contains a time stamp and five sets of corresponding offset trend information.

[0085] Referring to Figure 5 The acquisition step of the phased degradation indicator sequence is specifically:

[0086] S411: Call the offset trend information of each temperature-sensitive feature in the performance offset feature map, judge the direction of each feature's time sequence data, filter the sequence segments with unchanged change direction in consecutive periods, associate and record with the type of the temperature-sensitive feature, and construct a consistent offset sequence.

[0087] The process of judging the direction of each feature's time sequence data is specifically:

[0088] Perform a difference operation on the temperature-sensitive feature value of the current running period and the same type of temperature-sensitive feature value of the previous running period, and set a positive offset threshold and a negative offset threshold.

[0089] If the difference operation result is greater than the positive offset threshold, the change direction of the temperature-sensitive feature is judged to be positive offset.

[0090] If the difference operation result is less than the negative offset threshold, the change direction of the temperature-sensitive feature is judged to be negative offset.

[0091] If the difference value is between the negative offset threshold and the positive offset threshold, the change direction of the temperature-sensitive feature is determined as no offset.

[0092] The offset trend information of each temperature-sensitive feature in the performance offset feature map is called. The direction of the time series data of each independent temperature-sensitive feature (e.g., the difference value of the maximum current offset corresponding to the first voltage step) in the performance offset feature map is determined. This determination process relies on the preset positive offset threshold and negative offset threshold. The positive offset threshold and the negative offset threshold are set based on the monitoring of a brand new lithium iron phosphate battery in the initial stage of continuous operation (e.g., the first 100 cycles), and analyzing the natural fluctuation range of its characteristic parameters without significant degradation. Experimental data shows that the standard deviation of the numerical distribution of the maximum current offset difference value calculated in the continuous period is 0.003 A. In order to effectively filter out random fluctuations caused by measurement noise and minor environmental changes, the positive offset threshold is set to be positive three times of the standard deviation, i.e., +0.009 A; the negative offset threshold is set to be negative three times of the standard deviation, i.e., -0.009 A. Now, the direction of the time series data of each feature is determined. Take the difference value calculated in S312, for example, -0.01 A. Compare the difference value -0.01 A with the threshold. The specific determination logic is: if the difference value is greater than the positive offset threshold +0.009 A, the change direction of the temperature-sensitive feature is determined as "positive offset"; if the difference value is less than the negative offset threshold -0.009 A (such as -0.01 A in this example), the change direction of the temperature-sensitive feature is determined as "negative offset"; if the difference value is between the negative offset threshold and the positive offset threshold (including equal to -0.009 A or +0.009 A), the change direction of the temperature-sensitive feature is determined as "no offset". In a plurality of continuous operation periods (e.g., 10 consecutive periods, i.e., 40 hours), if the change direction of a certain feature is always determined as "negative offset", the sequence paragraph with continuous "negative offset" is identified, and is associated with the temperature-sensitive feature type (the maximum current offset corresponding to the first voltage step) to which the sequence paragraph belongs, and a consistent offset sequence is constructed.

[0093] S412: For each sequence paragraph in the consistent offset sequence, the temperature-sensitive feature type to which the sequence paragraph belongs, the duration of the sequence paragraph trend, and the offset amplitude value are extracted, and the three information are integrated into an independent structured index item. The index items of all sequence paragraphs are collected to obtain the degradation trend quantification data.

[0094] An analysis is performed for each sequence paragraph in the consistent offset sequence. Assuming that a sequence paragraph is identified, the information of the sequence paragraph is: the temperature sensitive feature type is "the maximum current offset corresponding to the first voltage step", and the change direction of the sequence paragraph is "negative offset" in the continuous 20 measurement periods from "October 15" to "October 25". First, the temperature sensitive feature type belonging to is extracted, that is, "the maximum current offset corresponding to the first voltage step". Second, the duration of the sequence paragraph trend is extracted, which is obtained by calculating the number of continuous measurement periods contained in the paragraph, that is, 20 periods. Finally, the cumulative offset amplitude value in the paragraph is calculated. This calculation is obtained by accumulating the original difference value calculated each time in the 20 periods (for example, -0.010 A, -0.012 A, -0.011 A, etc., all of which are less than -0.009 A). Assuming that the total sum after accumulation is -0.230 A. The three pieces of information: feature type (the maximum current offset corresponding to the first voltage step), duration (20 periods), and offset amplitude (-0.230 A) are integrated into an independent structured index item. Repeat the information extraction and integration process for all identified consistent offset sequence paragraphs, and collect the index items generated by all sequence paragraphs to obtain the degradation trend quantification data.

[0095] S413: Based on the degradation trend quantification data, call all index items in the collection, arrange and summarize according to the time sequence identifier attached to each index item, and construct a phased degradation index sequence;

[0096] Based on the obtained degradation trend quantification data, all structured index items in the degradation trend quantification data collection are called. Each structured index item records the complete information of a degradation event, and can be assigned a clear time sequence identifier according to the end time point of its corresponding sequence paragraph. For example, the end time of the sequence paragraph corresponding to the first index item is "October 25", and the end time of the sequence paragraph corresponding to the second index item is "November 10". According to the time sequence identifier attached to each index item, arrange and summarize all index items in the collection from early to late. The specific operation of arranging and summarizing is to create a new ordered list, traverse all index items, and insert the index items into the new list according to the time sequence identifier of the index items. This sorted and summarized collection forms a new time sequence, and each element in the sequence is a quantified and phased degradation event. This sequence clearly depicts the specific characteristics, duration and severity of the performance degradation of the lithium iron phosphate battery at different stages since it was put into use, and a phased degradation index sequence is constructed.

[0097] Please refer to Figure 6 , the acquisition step of the electrochemical energy storage remaining life estimation result is specifically:

[0098] S511: Based on the stage degradation index sequence of the electrochemical energy storage cell and the obtained total running time data, the two are time-point paired, the change amplitude of the index in each stage specified time is calculated, and the stage index change amplitude information is obtained;

[0099] Based on the stage degradation index sequence of the electrochemical energy storage cell and the obtained total running time data, the stage degradation index sequence and the total running time data are accurately paired at the time point. The total running time data is a timing record that continues to accumulate from the start of the lithium iron phosphate battery. For example, on "October 25", the total running time is 1200 hours; on "November 10", the total running time is 1500 hours. Take an index item from the stage degradation index sequence, the offset amplitude recorded by the index item is -0.230 ampere, and the corresponding sequence paragraph starts at the total running time of 800 hours and ends at the total running time of 1200 hours. Calculate the index change amplitude in this stage specified time, and the calculation method is to divide the offset amplitude by the running time of the stage. The specific calculation process is: -0.230 ampere divided by (1200 hours minus 800 hours), that is, -0.230 ampere divided by 400 hours, and the change amplitude obtained is -0.000575 ampere per hour. Each index item in the stage degradation index sequence is paired and calculated, and a series of index change amplitude values in different total running time stages are obtained, which together constitute the stage index change amplitude information.

[0100] S512: Construct the stage index change amplitude information into a change trend curve, perform extrapolation operation on the degradation of the future change trend curve, and calibrate it with the current total running time data to generate the life trend extension result;

[0101] Among them, the process of performing extrapolation operation on the degradation of the future change trend curve is specifically:

[0102] Cut the latest continuous data segment in time from the change trend curve;

[0103] Adopting a polynomial fitting method to curve fit the latest continuous data segment in time, a degradation prediction curve is established;

[0104] The phase indicator change amplitude information is constructed into a change trend curve composed of a series of discrete points, with the total running time as the horizontal coordinate and the indicator change amplitude as the vertical coordinate. The latest continuous data segment in the curve, such as the last 30 data points, is subjected to extrapolation operation. The extrapolation operation adopts a polynomial fitting method, specifically, a second-order polynomial is established to describe the change of the decay rate. The fitting process is to find a second-order polynomial such that the mean square error between the polynomial curve and the 30 latest continuous data points is minimized. Through the least squares method algorithm, the three coefficients of the polynomial can be determined, thereby establishing the decay prediction curve. After generating the decay prediction curve, the starting point of the decay prediction curve is aligned and calibrated with the current latest total running time data point to ensure that the starting value of the prediction curve is exactly the same as the last actual measurement value, thereby generating the life trend extension result. The life trend extension result is a curve extending from the current time to the future, predicting how the performance decay rate of the lithium iron phosphate battery continues to change with the running time.

[0105] S513: Calculate the time span required for the life trend extension result to extend from the current time point to the preset life termination threshold, take the time span value as the estimated value of the remaining life of the electrochemical energy storage unit, and obtain the electrochemical energy storage remaining life estimation result;

[0106] The time span required for the life trend extension result to extend from the current time point to the preset life termination threshold is calculated. The setting of the preset life termination threshold is based on the complete cycle life test experiment conducted on the same type of lithium iron phosphate battery. The battery is repeatedly charged and discharged under standard conditions until its actual available capacity decays to 80% of the factory rated capacity. At the time when the capacity decays to 80%, the steps of S111 and S112 are performed to measure the value of the "first voltage step corresponding current maximum deviation" (after temperature correction) as 4.86 A, while the value of this battery in the brand new state is 8.10 A. Therefore, 4.86 A is set as the life termination threshold of this feature. The calculation formula of the remaining life estimation value is: , where each letter is explained as follows: : represents the estimated value of the remaining life of the electrochemical energy storage unit. This is the final target that needs to be calculated, indicating the length of time that the lithium iron phosphate battery can continue to work until it reaches the life termination threshold from the current time, with the unit being hours (h). : represents the total running time at the predicted life end. This is a predicted value, indicating the total working time accumulated from the start of use of the lithium iron phosphate battery to the time when its performance decays to the preset life termination threshold, with the unit being hours (h). : represents the current total running time. This is an actual recorded value, indicating the total working time accumulated from the start of use of the lithium iron phosphate battery to the time when the life prediction calculation is performed, with the unit being hours (h).

[0107] The intersection point of the decay prediction curve established in S512 and the life termination threshold is obtained by solving the decay prediction curve, which is a second-order polynomial, in the general form of . The life termination time is solved by setting , so that the following calculation formula is obtained:

[0108] wherein the explanation of each letter is as follows: : In the quadratic polynomial model, the predicted maximum current deviation value at the total running time T, unit: ampere (A). : The independent variable in the quadratic polynomial model, representing the total running time, unit: hour (h). : Represents the quadratic coefficient of the quadratic polynomial. The coefficient determines the curvature of the decay curve, reflecting the "acceleration" of performance degradation. It is not pre-set, but calculated by the least squares method in S512 to fit the latest stage index change amplitude data points obtained in S511. Its value is completely determined by the actual performance degradation data of the measured battery in the near future. Unit: ampere / hour² (A / h²). : Represents the linear coefficient of the quadratic polynomial. The coefficient mainly determines the linear trend of the decay curve, which can be understood as the "initial speed" of the decay. Like , the value is also calculated by least squares fitting of actual measurement data, rather than artificially set. It reflects the main rate of recent decay. Unit: ampere / hour (A / h). : Represents the constant term of the quadratic polynomial. The coefficient is the intercept of the fitting curve at T=0 (the starting point of the fitting interval). It is also a result of least squares fitting, used to ensure the mathematical integrity of the fitting curve, and together with and determines the overall shape of the curve. Unit: ampere (A). : Represents the preset life termination threshold. The value is a specific current maximum deviation value, when the performance indicator of the lithium iron phosphate battery decays to this value, it is determined that its life is terminated. The threshold is determined by full-life experiments on benchmark batteries, unit: ampere (A).

[0109] Let the current total running time be 3000 hours. By fitting the latest 30 data points with a second-order polynomial, the decay prediction curve coefficients are: ampere / hour², ampere / hour, ampere. The preset life termination threshold 4.86 ampere. The total running time at the end of life is calculated : hours. The predicted total running time calculated above is set as 5500 hours. Then the remaining life is calculated : hours. The time span value of 2500 hours is taken as an estimate of the remaining life of the electrochemical energy storage cell, and the electrochemical energy storage remaining life estimation result is obtained.

[0110] An electrochemical energy storage remaining life estimation system for performing the electrochemical energy storage remaining life estimation method described above, the system comprising:

[0111] A voltage response acquisition module, which acquires the terminal voltage and current data of the electrochemical energy storage cell in real time during the step excitation process by applying an externally controlled multi-stage voltage step excitation to the two ends of the electrochemical energy storage cell to be tested, constructs a voltage-current response sequence, extracts temperature-sensitive features from the sequence, and constructs an electrochemical performance response set;

[0112] A temperature correction processing module, which acquires the temperature information of the current electrochemical energy storage cell and corrects the errors of the temperature-sensitive features in the electrochemical performance response set to generate a temperature-corrected performance response set;

[0113] A performance offset analysis module, which constructs a feature sequence of the current operating cycle in time sequence according to the temperature-corrected performance response set of the electrochemical energy storage cell, compares it with the same temperature-sensitive features of the previous operating cycle of the electrochemical energy storage cell, judges the offset trend of each temperature-sensitive feature, and constructs a performance offset feature map;

[0114] A degradation trend identification module, which identifies the offset trend sequence in which the change directions of the temperature-sensitive features in the performance offset feature map remain consistent, and constructs a stage degradation index sequence;

[0115] A remaining life estimation module, which calculates the remaining life estimation value based on the stage degradation index sequence of the electrochemical energy storage cell and the total running time, and obtains the electrochemical energy storage remaining life estimation result.

[0116] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any person skilled in the art can modify or change the above disclosed technical content to obtain equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments within the technical solution content of the present application still falls within the protection scope of the present application.​

Claims

1. A method of electrochemical energy storage remaining life estimation, characterized by, The method comprises the following steps: S1: by applying an external controlled multi-stage voltage step excitation to both ends of the to-be-tested electrochemical energy storage cell, collecting the terminal voltage and current data of the electrochemical energy storage cell in real time during the step excitation process, constructing a voltage-current response sequence, extracting temperature-sensitive features from the sequence, and constructing an electrochemical performance response set; S2: collecting the temperature information of the current electrochemical energy storage cell, correcting the temperature-sensitive features in the electrochemical performance response set, and generating a temperature-corrected performance response set; S3: constructing a feature sequence of the current operation cycle in chronological order according to the temperature-corrected performance response set of the electrochemical energy storage cell, comparing the same temperature-sensitive features of the previous operation cycle of the electrochemical energy storage cell, judging the shift trend of each temperature-sensitive feature, and constructing a performance shift feature map; S4: identifying the shift trend sequence in which the change directions of the temperature-sensitive features in the performance shift feature map are consistent, and constructing a stage degradation index sequence; S5: calculating a remaining life estimation value based on the stage degradation index sequence and the total operation time of the electrochemical energy storage cell, and obtaining an electrochemical energy storage remaining life estimation result.

2. The electrochemical energy storage remaining life estimation method of claim 1, wherein, The voltage-current response sequence includes voltage change rate and current maximum deviation, the temperature-corrected performance response set includes corrected voltage change rate and corrected current maximum deviation, the performance shift feature map includes voltage change rate shift trend and current maximum deviation shift trend, the stage degradation index sequence specifically includes trend consistent interval, feature item identifier, and time sequence index, and the electrochemical energy storage remaining life estimation result includes life change trend, current remaining life value, and estimation time node.

3. The electrochemical energy storage remaining life estimation method of claim 1, wherein, The acquisition step of the electrochemical performance response set is specifically: S111: by applying an external controlled multi-stage voltage step excitation to both ends of the to-be-tested electrochemical energy storage cell, the electrochemical energy storage cell is a lithium iron phosphate battery, collecting the terminal voltage and current data of the electrochemical energy storage cell in real time during the step excitation process, and constructing a voltage-current response sequence; S112: identifying the section where the voltage step occurs in the voltage-current response sequence, calculating the voltage change rate in each voltage step occurrence section, synchronously acquiring the maximum deviation of the current in the corresponding section, collecting the voltage change rate and the current maximum deviation of all sections as temperature-sensitive features, and obtaining dynamic response feature data; S113: calling the voltage change rate and the current maximum deviation of all sections in the dynamic response feature data, integrating the two temperature-sensitive features into a structured data set, merging the structured data sets obtained under multiple excitations, and constructing an electrochemical performance response set.

4. The electrochemical energy storage remaining life estimation method of claim 3, wherein, The acquisition step of the temperature-corrected performance response set is specifically: S211: acquiring the environmental temperature of the electrochemical energy storage cell during the response process through a temperature sensor, recording and integrating the temperature data collected at each time point, and establishing environmental temperature information; S212: According to the temperature interval matching rule, the temperature data in the environmental temperature information is compared with the preset multiple temperature intervals one by one, the unique temperature interval is locked, and the corresponding preset correction coefficient is extracted from the temperature interval to obtain the matching correction coefficient value; S213: The error correction is performed on the temperature sensitive characteristics in the electrochemical performance response set, the voltage change rate and the current maximum offset are called, and the matching correction coefficient value is used for correction processing, the current maximum offset after temperature and voltage platform effect correction is calculated, and the temperature corrected performance response set is generated.

5. The electrochemical energy storage remaining life estimation method of claim 4, wherein, For the correction processing using the matching correction coefficient value, the formula is: ； calculating the maximum current excursion corrected for temperature and voltage plateau effects ; wherein, represents a measured original current maximum deviation, represents a specified level in a multi-level voltage step excitation, represents a temperature correction coefficient, represents a preset temperature interval, represents a temperature correction weight coefficient, represents a voltage correction coefficient, : represents a voltage correction weight coefficient.

6. The electrochemical energy storage remaining life estimation method of claim 4, wherein, The performance offset characteristic map is obtained by: S311: The temperature corrected performance response set of the electrochemical energy storage unit is arranged in time sequence to establish the current running period characteristic sequence; S312: The same temperature sensitive characteristics of the previous running period of the electrochemical energy storage unit are called, and the difference and change rate of the same index point are compared for the current running period characteristic sequence, the offset trend of each temperature sensitive characteristic is judged, and the characteristic offset trend information is obtained; S313: The offset trend of each temperature sensitive characteristic extracted from the characteristic offset trend information is combined to construct a performance offset characteristic map.

7. The electrochemical energy storage remaining life estimation method of claim 6, wherein, The acquisition steps of the stage degradation index sequence are specifically: S411: The offset trend information of each temperature sensitive characteristic in the performance offset characteristic map is called, the directionality of the time sequence data of each characteristic is judged, the sequence paragraphs with unchanged change direction in the continuous period are screened, the temperature sensitive characteristic type is associated and recorded, and a consistent offset sequence is constructed; S412: For each sequence paragraph in the consistent offset sequence, the temperature sensitive characteristic type, the duration of the sequence paragraph trend and the offset amplitude value are extracted, the three information is integrated into an independent structured index item, the index items of all sequence paragraphs are collected, and the degradation trend quantitative data is obtained; S413: Based on the degradation trend quantitative data, all index items in the collection are called, and are arranged and summarized according to the time sequence identifier attached to each index item, and a stage degradation index sequence is constructed.

8. The electrochemical energy storage remaining life estimation method of claim 7, wherein, The acquisition steps of the electrochemical energy storage remaining life estimation result are specifically: S511: Based on the stage degradation index sequence and the obtained total running time data of the electrochemical energy storage unit, the two are time point paired, the change amplitude of the index in each stage specified time is calculated, and the stage index change amplitude information is obtained; S512: The stage index change amplitude information is constructed into a change trend curve, the degradation of the future change trend curve is extrapolated, and is calibrated combined with the current total running time data to generate a life trend extension result; S513: The time span required for the life trend extension result to extend from the current time point to the preset life termination threshold is calculated, the time span value is taken as the estimation value of the remaining life of the electrochemical energy storage unit, and the electrochemical energy storage remaining life estimation result is obtained.

9. The electrochemical energy storage remaining life estimation method of claim 8, wherein, The process of performing extrapolation operation on the recession of the future change trend curve is specifically: Cutting the latest continuous data segment in time from the change trend curve; Using polynomial fitting to fit the latest continuous data segment in time to establish a recession prediction curve.

10. An electrochemical energy storage remaining life estimation system, comprising: The electrochemical energy storage remaining life estimation method according to any one of claims 1-9, the system comprises: A voltage response acquisition module, which acquires the terminal voltage and current data of the electrochemical energy storage cell in real time during the step excitation process by applying an externally controlled multi-level voltage step excitation to the electrochemical energy storage cell to be tested, constructs a voltage-current response sequence, extracts temperature-sensitive features from the sequence, and constructs an electrochemical performance response set; A temperature correction processing module, which acquires the temperature information of the current electrochemical energy storage cell, corrects the temperature-sensitive features in the electrochemical performance response set for errors, and generates a temperature-corrected performance response set; A performance offset analysis module, which constructs a feature sequence of the current operating cycle in time sequence according to the temperature-corrected performance response set of the electrochemical energy storage cell, compares the same temperature-sensitive features of the previous operating cycle of the electrochemical energy storage cell, judges the offset trend of each temperature-sensitive feature, and constructs a performance offset feature map; A recession trend identification module, which identifies the offset trend sequence in which the change direction of the temperature-sensitive features in the performance offset feature map remains consistent, and constructs a stage recession index sequence; A remaining life estimation module, which calculates the remaining life estimation value based on the stage recession index sequence and the total operating time of the electrochemical energy storage cell, and obtains the electrochemical energy storage remaining life estimation result.

Images