Battery capacity offline detection method, system and device for electric energy cycle

Abstract

The present application relates to the technical field of battery capacity offline detection, and specifically to a battery capacity offline detection method, system and device for electric energy cycling, the method comprising: collecting load current data, voltage data and battery temperature data; dividing a sliding window; obtaining a current anomaly coefficient at each acquisition moment; obtaining a current fluctuation coefficient at each acquisition moment; obtaining a load electric parameter variation coefficient at each acquisition moment; obtaining a temperature confidence coefficient for each electric energy cycling process, and then obtaining a loss accumulation coefficient for each electric energy cycling process; and then obtaining a battery capacity correction factor at each acquisition moment of each electric energy cycling process; obtaining the available capacity of a battery pack in each electric energy cycling process based on a function of the battery capacity correction factor and the load current changing with time. The present application improves the accuracy of battery capacity detection of battery packs for electric energy cycling.

Description

Technical Field

[0001] The present application relates to the technical field of battery capacity offline detection, and in particular to a battery capacity offline detection method, system and device for electric energy cycling. Background Art

[0002] With the acceleration of informatization, the "information revolution" has set off a wave of enthusiasm in various fields. As a key infrastructure in the information age, the campus-type Internet data center (IDC) carries the tasks of storing, processing and transmitting massive amounts of data. Its stable operation is crucial to ensuring the smooth development of the digital economy. In the power supply system of IDC, the battery pack, as a core component of the uninterruptible power supply system, plays the role of emergency backup power supply. Therefore, the performance and capacity status of the battery pack are directly related to the power supply reliability and business continuity of the IDC.

[0003] Battery capacity detection is a core part of battery maintenance and management, and is crucial for timely identification of potential risks such as battery aging and capacity attenuation. Therefore, implementing efficient and safe battery capacity detection is of great significance to ensure the long-term stable operation of campus-type IDCs. In existing battery maintenance systems, when performing battery pack capacity detection, the DC power of the battery pack is converted into AC power for equipment power supply through offline charging and discharging, and the battery capacity performance of the battery pack is determined by recording the discharge energy when the battery pack reaches the discharge end condition. However, in the actual application of this detection method, due to changes in equipment load, the detection results of the battery capacity of the battery pack are usually less accurate. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a battery capacity offline detection method, system and device for electric energy cycling. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a battery capacity offline detection method for electric energy cycling, the method comprising the following steps:

[0006] Collect the load current data, voltage data and battery temperature data of the offline battery pack at each collection time during each power cycle;

[0007] Taking each collection moment as the center, a window of preset size is used as a sliding window at each collection moment;

[0008] Based on the number of load currents in the sliding window at each acquisition moment that are greater than the preset standard data and the difference between the load current and the preset standard current, the current anomaly coefficient at each acquisition moment is obtained; based on the discrete amplitude and information amount of the load current in the sliding window, the current fluctuation coefficient at each acquisition moment is obtained; based on the current anomaly coefficient, the current fluctuation coefficient, and the difference between the voltage data at adjacent acquisition moments, the load electrical parameter variation coefficient at each acquisition moment is obtained;

[0009] Obtaining a temperature confidence coefficient for each power cycle process based on a difference between the battery temperature data and a predicted value of the battery temperature data; obtaining a loss accumulation coefficient for each power cycle process based on the temperature confidence coefficient and a distance between temperature data in any two power cycle processes;

[0010] Obtaining the battery capacity correction factor at each collection moment of each power cycle process based on the load electrical parameter variation coefficient, loss accumulation coefficient and battery temperature data;

[0011] The available capacity of the battery pack during each power cycle is obtained based on the battery capacity correction factor and the change function of the load current over time.

[0012] Furthermore, the method for obtaining the current anomaly coefficient is:

[0013] For the sliding window of each collection moment, the ratio of the number of collection moments in the sliding window at which the load current data is greater than the preset standard current to the number of all collection moments in the sliding window is calculated as the first ratio of each collection moment;

[0014] For each collection moment, the difference between the load current data at the collection moment and the preset standard current is calculated, and the calculation result of an exponential function with a natural constant as the base and the difference as the exponent is calculated, and the product of the first ratio at each collection moment and the calculation result of the exponential function is used as the current anomaly coefficient at each collection moment.

[0015] Furthermore, the method for obtaining the current fluctuation coefficient is:

[0016] For each acquisition moment, the range difference and information entropy of the load current at all acquisition moments within the sliding window of the acquisition moment are calculated, and the product of the range difference and the information entropy is used as the current fluctuation coefficient at each acquisition moment.

[0017] Furthermore, the method for obtaining the load electrical parameter variation coefficient is:

[0018] The load electrical parameter variation coefficient ; In the formula, For the The load electrical parameter variation coefficient at each acquisition moment; Indicates The current anomaly coefficient at each acquisition moment, For the The current fluctuation coefficient at each acquisition moment, e is a natural constant, , Respectively represent The collection time, The voltage data at each acquisition moment.

[0019] Further, the method for obtaining the temperature confidence coefficient is:

[0020] Fitting the temperature data in each power cycle process to obtain the time series curve of each power cycle process; using a detrending algorithm to detrend the time series curve of each power cycle process to obtain the temperature fluctuation curve of each power cycle process;

[0021] The data points at each acquisition time in the temperature fluctuation curve of each power cycle process are used as inputs of the prediction algorithm to obtain the predicted temperature values ​​at each acquisition time of the next adjacent power cycle process;

[0022] For each power cycle process, the absolute value of the difference between the predicted temperature value at each collection moment in the power cycle process and the battery temperature data is calculated, and the sum of all the absolute values ​​of the difference in the power cycle process is used as the temperature confidence coefficient of each power cycle process.

[0023] Furthermore, the calculation formula of the loss accumulation coefficient is: ; In the formula, For the The cumulative loss coefficient of the electric energy cycle; It is The temperature confidence coefficient of the temperature data during the power cycle, , They are sequence Temperature fluctuation curve of temperature data during the power cycle, Indicates Before the power cycle The number of electrical energy cycles, is the DTW distance.

[0024] Furthermore, the method for obtaining the battery capacity correction factor is as follows: for each electric energy cycle process, a normalized value of the cumulative loss coefficient of the electric energy cycle process is calculated, the product of the normalized value and the battery temperature data at each collection moment in the electric energy cycle process is calculated as a first product, a calculation result of an exponential function with a natural constant as the base and the first product as the exponent is calculated, and the product of the calculation result and the load electrical parameter variation coefficient at each collection moment in the electric energy cycle process is used as the battery capacity correction factor at each collection moment in each electric energy cycle process.

[0025] Furthermore, the method for obtaining the available capacity is: ; In the formula, The available capacity of the battery pack in each power cycle process; Indicates the discharge time of the electric energy cycle process, It represents the function of the load current changing with time during each power cycle. Indicates the first Battery capacity correction factor at each acquisition moment; Represents the normalization function.

[0026] In the second aspect, an embodiment of the present application provides an offline battery capacity detection device for electric energy cycling, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements any one of the steps of the offline battery capacity detection method for electric energy cycling when executing the computer program.

[0027] In a third aspect, an embodiment of the present application provides a battery capacity offline detection system for electric energy cycling, the system comprising:

[0028] A data acquisition module is used to collect load current data, voltage data and battery temperature data of an offline battery pack at each acquisition moment during each power cycle;

[0029] A battery capacity correction factor acquisition module is used to take each acquisition moment as the center and use a window of a preset size as a sliding window at each acquisition moment; based on the number of load currents in the sliding window at each acquisition moment that are greater than the preset standard data and the difference between the load current and the preset standard current, obtain the current anomaly coefficient at each acquisition moment; based on the discrete amplitude and information amount of the load current in the sliding window, obtain the current fluctuation coefficient at each acquisition moment; based on the current anomaly coefficient, the current fluctuation coefficient, and the difference between the voltage data at adjacent acquisition moments, obtain the load electrical parameter variation coefficient at each acquisition moment; based on the difference between the battery temperature data and the predicted value of the battery temperature data, obtain the temperature confidence coefficient of each electric energy cycle process; based on the temperature confidence coefficient and the distance between the temperature data in any two electric energy cycle processes, obtain the loss accumulation coefficient of each electric energy cycle process; based on the load electrical parameter variation coefficient, the loss accumulation coefficient and the battery temperature data, obtain the battery capacity correction factor at each acquisition moment of each electric energy cycle process;

[0030] The battery capacity offline detection module is used to obtain the available capacity of the battery pack during each power cycle based on the battery capacity correction factor and the change function of the load current over time.

[0031] This application has at least the following beneficial effects:

[0032] This application performs the following analysis by collecting load current, voltage, and battery temperature data during battery pack capacity detection;

[0033] Firstly, the load electric parameter variation coefficient is constructed by analyzing the changes of load current and voltage during capacity detection of the battery pack. The beneficial effect is that the load electric parameter variation coefficient takes into account the difference between the load current and the preset standard current during the battery capacity detection process of the battery pack and the changes of the load current and battery pack voltage caused by the sudden addition and removal of the equipment load. According to the load electric parameter variation coefficient, the negative impact of the fluctuation of load current on the utilization rate of the active material inside the battery can be better reflected;

[0034] Secondly, a battery capacity correction factor is constructed based on the battery temperature change during the battery pack capacity detection process and combined with the load electrical parameter variation coefficient. The beneficial effect is that the battery capacity correction factor takes into account the internal battery power loss caused by the magnetic field and internal resistance during the battery pack capacity detection process. The battery power loss for power cycling is evaluated by the difference between the battery temperature data when there is no power cycling and the battery temperature data when there is power cycling, and the temperature difference of the battery pack is combined with the load electrical parameter variation coefficient. According to the battery capacity correction factor, the actual utilization efficiency of the internal battery power during the power cycling process can be reflected. Therefore, using the battery capacity correction factor to correct the load current change function during the battery pack capacity detection can more accurately detect the battery capacity of the battery pack for power cycling. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A flowchart of the steps of an offline detection method of battery capacity for electric energy cycling provided in one embodiment of the present application;

[0037] Figure 2 A schematic diagram of a temperature fluctuation curve provided for an embodiment of the present application. DETAILED DESCRIPTION

[0038] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, describes in detail the specific implementation methods, structures, features and effects of the battery capacity offline detection method, system and device for electric energy cycling proposed in the present application. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0039] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0040] The specific scheme of the battery capacity offline detection method, system and device for electric energy cycling provided by the present application is described in detail below with reference to the accompanying drawings.

[0041] See also Figure 1 , which shows a flowchart of a method for offline detection of battery capacity for electric energy cycling provided by an embodiment of the present application, the method comprising the following steps:

[0042] Step S001, obtaining voltage, current, and temperature data of the Internet data center battery pack during discharge, and preprocessing the obtained data to obtain the working status data of the battery pack during the preprocessed battery pack capacity detection process.

[0043] The capacity of the battery pack of the Internet data center (IDC) is not only related to the discharge energy of the battery when the battery pack reaches the discharge cut-off condition, but also closely related to the load state and working environment when the battery pack capacity is detected. This application uses current sensors, voltage sensors, and temperature sensors to obtain the battery load current of the offline battery pack during capacity detection. ,Voltage and battery temperature Data, load current refers to the output DC current provided by the battery pack when it is discharged, and the data sampling interval is The total amount of data collected during the battery pack capacity detection process is .

[0044] In order to prevent data loss in the collected data, the missing data is preprocessed using mean filling technology, which is a well-known technology and its specific implementation process is not described in detail in this application. Thus, the working status data of the battery pack during the preprocessed battery pack capacity detection process is obtained.

[0045] Step S002: Based on the pre-processed working status data of the battery pack, the data characteristics during the power cycle of the battery pack are analyzed, and the battery discharge efficiency correction coefficient is constructed according to the characteristics between the data.

[0046] The purpose of battery capacity detection is to accurately evaluate the actual capacity of the battery pack of the Internet data center, so as to understand the performance of the battery pack and perform corresponding maintenance and management. The traditional battery capacity detection method, taking the electronic load meter as an example, the battery provides a constant load current, detects the battery current, voltage and time, and determines the battery capacity performance by calculating the battery discharge energy when the battery reaches the discharge cut-off condition.

[0047] However, the battery capacity detection for power cycle first converts the DC power provided by the battery into AC power and inputs it into the power grid to continue to power the equipment to achieve power recycling. Then, the battery capacity performance is determined by calculating the discharge energy when the battery reaches the discharge cut-off condition. During the battery capacity detection process, a constant load current is a key factor in ensuring the reliability of the test results. However, the battery capacity detection method for power cycle has a large fluctuation in the load current provided by the battery pack because the types of tasks run by Internet data centers are too complex and the amount of tasks is not fixed. The frequent fluctuations in load current make it impossible for the detection process to maintain a stable discharge environment, which makes it difficult for the test results to accurately reflect the true capacity of the battery.

[0048] Furthermore, the load current of the battery pack will change with the change in the degree of power demand of the power grid load. Therefore, when the operating task volume of the Internet data center suddenly increases or decreases, the load current provided by the battery pack will fluctuate. This fluctuation will change the chemical reaction rate inside the battery, thereby affecting the discharge performance of the battery, and thus making the final battery capacity detection result less accurate. For example, when the operating task volume of the Internet data center suddenly increases, the battery pack needs to provide more load current, and the diffusion rate of the solution near the battery plate may not be as fast as the generation rate of the reactants under high current, thereby reducing the utilization rate of the active substances inside the battery. On the contrary, when the load current of the battery is small, the diffusion effect of the reactants inside the battery is better, and the battery active substances react more fully, thereby making full use of the battery active substances. At the same time, due to the large fluctuation of the load current of the battery pack, the voltage change of the battery pack will change greatly. When the load current of the battery pack increases, the battery pack voltage will drop rapidly to adapt to the change of the load current of the battery pack.

[0049] Furthermore, the greater the positive difference between the actual load current of the battery pack and the preset standard current, the lower the utilization rate of the battery active material; it should be noted that when the actual load current of the battery pack is higher than the preset standard current, the difference between the two is called the positive difference.

[0050] Furthermore, the load current fluctuation of the battery pack will have a further negative impact on the utilization rate of the active material of the battery pack. Furthermore, for the load current at each acquisition moment, a sliding window is divided with each acquisition moment as the center, and the information entropy of all load currents in the sliding window is calculated. In this embodiment, the width of the sliding window is set to 21. The implementer can select other windows according to actual conditions, and repeatedly fill the edge data for the data with less than 21 in the sliding window.

[0051] Furthermore, for each collection moment, the ratio of the number of collection moments in the sliding window of the collection moment at which the load current is higher than the preset standard current to all collection moments in the sliding window is calculated as the first ratio. The greater the proportion of the load current higher than the preset standard current, the lower the utilization rate of the active material in the battery pack.

[0052] Further, according to the above analysis, based on the difference between the first ratio, the load current and the preset standard current, the current abnormality coefficient at each acquisition moment is calculated, and the calculation formula is: ; In the formula, For the The current anomaly coefficient at each acquisition moment; Indicated in The first ratio of the acquisition time, e is a natural constant, Indicates the battery pack is performing capacity testing. The load current at each acquisition moment, Indicates the preset standard current of the battery pack when performing capacity detection. In this embodiment The value is 250A.

[0053] It should be noted that when the battery pack is testing its capacity, the greater the positive difference between the load current and the preset standard current, The larger the value of , and the longer the battery load current exceeds the preset standard current within the sliding window of the current acquisition moment, The larger the value of, the greater the load current of the calculated battery during the energy cycle. Current anomaly coefficient at the time of acquisition The bigger.

[0054] Furthermore, based on the discrete amplitude and information amount of the load current in the sliding window at each acquisition moment, the current fluctuation coefficient at each acquisition moment is calculated, and the calculation formula is: ; In the formula, For the Current fluctuation coefficient at each acquisition moment; , Respectively represent The range and information entropy of the load current data in the sliding window at the time of acquisition.

[0055] It should be noted that, in the sliding window at the current acquisition time, when the battery pack load suddenly increases or decreases, the battery load current will fluctuate greatly accordingly, thus making the load current in the sliding window at the current acquisition time extremely different. And information entropy Increases, at this time, the current fluctuation coefficient at each acquisition moment increases; conversely, the current fluctuation coefficient at each acquisition moment decreases.

[0056] Furthermore, based on the current anomaly coefficient, the current fluctuation coefficient, and the difference between the voltage data at adjacent acquisition times, the load current variation coefficient at each acquisition time is obtained, and the calculation formula is: ; In the formula, For the The load electrical parameter variation coefficient at each acquisition moment; Indicates The current anomaly coefficient at each acquisition moment, For the The current fluctuation coefficient at each acquisition moment, e is a natural constant, , Respectively represent The collection time, The voltage data at each acquisition moment.

[0057] It should be noted that the greater the load current of the battery pack, the faster the voltage of the battery pack drops. The larger the value, the higher the battery pack calculated. The greater the load electrical parameter variation coefficient at each acquisition moment, This indicates that the greater the negative impact of the fluctuation of the load current on the utilization rate of the active materials inside the battery during the power cycle, the lower the actual detected battery capacity; conversely, the smaller the load electrical parameter variation coefficient, the lower the actual detected battery capacity.

[0058] Furthermore, when conducting battery capacity detection for electric energy circulation, a factor that cannot be ignored is the energy loss during the electric energy conversion process. Specifically, when the direct current of the battery is converted into alternating current, a certain amount of energy loss will occur. On the one hand, as the current changes, the magnetic field of the transformer core will change, thereby generating eddy currents and molecular friction of the core itself. In this process, part of the electric energy will be converted into heat energy, causing the battery's electric energy loss; in addition, in the circuit, the presence of resistance will also cause part of the electric energy to be converted into heat energy, thereby further increasing the battery's electric energy loss. Although the electric energy loss in a short period of time can be ignored, in the battery capacity detection of the battery pack, the long-term electric energy loss will accumulate, thereby causing the battery temperature to gradually rise. Therefore, the present application filters and corrects the collected battery temperature data, and evaluates the relationship between the battery temperature and the battery discharge performance according to the temperature data fluctuations caused by the current conversion.

[0059] Since it is an energy cycle process, if there is energy loss, then as the cycle progresses, that is, the number of cycles increases, the eddy currents generated and the molecular friction of the iron core itself cause the battery energy loss to gradually increase. Therefore, this application considers using data prediction to evaluate the degree of influence of energy loss on temperature at different collection times during the energy cycle process by predicting the deviation between the predicted value and the actual value at the time of collection.

[0060] According to the above analysis, for each power cycle process, the battery temperature data in the power cycle process is fitted with a polynomial, with the acquisition time as the horizontal coordinate and the battery temperature data as the vertical coordinate, to obtain the time series curve of each power cycle process. Further, the DFA detrending algorithm is used to detrend the time series curve of each power cycle process to obtain the temperature fluctuation curve of each power cycle process. The schematic diagram of the temperature fluctuation curve is shown in the figure. Figure 2 As shown, in Figure 2 In the figure, the solid line is the original temperature data curve, and the dotted line is the temperature data curve after detrending.

[0061] Furthermore, the data points at each acquisition time in the temperature fluctuation curve of each power cycle process are used as the input of the EMA prediction algorithm, and the temperature values ​​at each acquisition time in the temperature fluctuation curve of the temperature data in the adjacent next power cycle process are predicted to obtain the predicted temperature values ​​at each acquisition time in the adjacent next power cycle process; in this embodiment, the smoothing factor of the EMA algorithm is set to 0.9, and the implementer can select other values ​​according to actual conditions. Polynomial fitting, detrending analysis, and EMA algorithm are all commonly used techniques in data analysis, and the specific process will not be repeated.

[0062] Furthermore, for the temperature fluctuation curves of the temperature data during each power cycle, if the power loss accumulated during the current conversion process affects the temperature to varying degrees, then the differences between the temperature fluctuation curves and the deviation between the predicted battery temperature data and the battery temperature data are caused by the power loss accumulated to varying degrees.

[0063] It should be noted that there will hardly be enough power loss during the first power cycle. Therefore, in this embodiment, the temperature data during the first power cycle is not predicted and corrected.

[0064] Based on the above analysis, calculate the The loss accumulation coefficient of the electric energy cycle is calculated as follows: ; In the formula, For the Temperature confidence factor of the power cycle process; , They are During the first power cycle The predicted temperature value and battery temperature data at each collection moment, For the The total number of acquisition moments during an energy cycle.

[0065] It should be noted that The greater the difference between the predicted value and the actual value of the temperature data during the first power cycle, the Temperature confidence factor of temperature data during the power cycle The larger the The greater the degree of power loss during the first power cycle; on the contrary, The smaller the energy loss during the power cycle.

[0066] Furthermore, the loss accumulation coefficient of each power cycle process is calculated to characterize the impact of accumulated power loss on temperature data as the power cycle time increases. The calculation formula is: ; In the formula, For the The cumulative loss coefficient of the electric energy cycle; It is The temperature confidence coefficient of the temperature data during the power cycle, , They are sequence Temperature fluctuation curve of temperature data during the power cycle, Indicates Before the power cycle The number of electrical energy cycles, is the DTW distance. The DTW distance is a well-known technology and will not be described in detail here.

[0067] It should be noted that the greater the DTW distance between the temperature fluctuation curve in the historical power cycle and the temperature fluctuation curve in the current cycle, the The larger the value of is, the greater the impact of the accumulated energy loss on the temperature data will be as the number of energy cycles increases. The cumulative loss factor during the power cycle The larger the value, the more serious the battery energy loss in the current cycle.

[0068] Furthermore, based on the cumulative loss coefficient of the number of energy cycles at each collection moment, the battery temperature data detected when there is energy cycling is corrected to evaluate the energy loss caused by magnetic loss and resistance loss. The battery temperature data and the calculated cumulative loss coefficient during the energy cycle reflect the heat energy loss generated by the internal battery energy when powering the device. The larger the battery temperature and loss coefficient, the greater the energy loss during the battery energy cycle; at the same time, the greater the fluctuation of the load current and the greater the positive difference between the load current and the preset standard current, the larger the calculated load electrical parameter variation coefficient, indicating that the utilization rate of the active materials inside the battery is lower, that is, the actual utilization efficiency of the internal battery energy is lower. Therefore, based on the above analysis and combined with the load electrical parameter variation coefficient Constructing Battery Capacity Correction Factors , the calculation formula is: ; In the formula, Indicates During the first power cycle Battery capacity correction factor at each acquisition moment; Indicates During the power cycle Battery temperature data at each collection moment; It is The cumulative loss coefficient of the electric energy cycle; represents the normalization function; For the During the first power cycle The load electrical parameter variation coefficient at each acquisition moment.

[0069] When the battery pack is in the state of power cycle, the battery pack's battery capacity is tested. Due to the existence of magnetic loss and battery internal resistance, the battery pack's power cannot be fully used to power the device, and part of the power is converted into heat energy, resulting in power loss. Therefore, when the battery pack is facing power cycle, the battery capacity is tested. The greater the difference between the temperature fluctuation curve during the power cycle and the temperature fluctuation curve during the historical power cycle, the greater the loss accumulation coefficient. The larger the battery pack temperature is, the Battery temperature at the time of collection The higher the value, the greater the proportion of electrical energy in the battery converted into heat energy, which in turn indicates that the The first battery energy cycle The more serious the battery pack power loss is at each collection moment. In addition, the greater the load current fluctuation of the battery pack, the greater the calculated load electric parameter variation coefficient, indicating that the utilization rate of the active material inside the battery is lower, and the larger the calculated battery capacity correction factor is. This indicates that the actual utilization efficiency of the internal battery power is lower, that is, the actual detected battery pack capacity is lower.

[0070] Step S003, according to the calculated battery capacity correction factor, combined with the discharge time of the battery pack and the load current, the available capacity of the battery pack in the current state is calculated using the ampere-hour integration method.

[0071] The above calculations yield the battery capacity correction factor at each acquisition moment during the battery pack capacity detection process for power cycling. The battery capacity correction factor takes into account the negative impact of load current changes on the utilization rate of active materials inside the battery and battery temperature changes caused by battery power loss during battery capacity detection for power cycling. The battery capacity correction factor can better reflect the actual utilization efficiency of the battery's internal power during the power cycling process. Therefore, the battery capacity correction factor is used as the weight of the load current during the battery power cycling process, and the ampere-hour integration method is used to calculate the available capacity of the battery pack. , the calculation formula is: ; In the formula, The available capacity of the battery pack in each power cycle process; Indicates the discharge time of the electric energy cycle process, It represents the function of the load current changing with time during each power cycle. Indicates the first Battery capacity correction factor at each acquisition moment; The load current variation function over time and the ampere-hour integration method are well-known technologies and will not be described in detail here.

[0072] So far, the battery capacity of the Internet data center battery pack is obtained.

[0073] Based on the same inventive concept as the above method, an embodiment of the present application also provides a battery capacity offline detection device for electric energy cycling, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned battery capacity offline detection methods for electric energy cycling are implemented.

[0074] Similarly, based on the same inventive concept as the above method, an embodiment of the present application provides a battery capacity offline detection system for electric energy cycling, the system comprising:

[0075] A data acquisition module is used to collect load current data, voltage data and battery temperature data of an offline battery pack at each acquisition moment during each power cycle;

[0076] A battery capacity correction factor acquisition module is used to take each acquisition moment as the center and use a window of a preset size as a sliding window at each acquisition moment; based on the number of load currents in the sliding window at each acquisition moment that are greater than the preset standard data and the difference between the load current and the preset standard current, obtain the current anomaly coefficient at each acquisition moment; based on the discrete amplitude and information amount of the load current in the sliding window, obtain the current fluctuation coefficient at each acquisition moment; based on the current anomaly coefficient, the current fluctuation coefficient, and the difference between the voltage data at adjacent acquisition moments, obtain the load electrical parameter variation coefficient at each acquisition moment; based on the difference between the battery temperature data and the predicted value of the battery temperature data, obtain the temperature confidence coefficient of each electric energy cycle process; based on the temperature confidence coefficient and the distance between the temperature data in any two electric energy cycle processes, obtain the loss accumulation coefficient of each electric energy cycle process; based on the load electrical parameter variation coefficient, the loss accumulation coefficient and the battery temperature data, obtain the battery capacity correction factor at each acquisition moment of each electric energy cycle process;

[0077] The battery capacity offline detection module is used to obtain the available capacity of the battery pack during each power cycle based on the battery capacity correction factor and the change function of the load current over time.

[0078] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0079] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0080] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A battery capacity offline detection method for electric energy cycling, characterized in that: The method comprises the following steps: Collect the load current data, voltage data and battery temperature data of the offline battery pack at each collection time during each power cycle; Taking each collection moment as the center, a window of preset size is used as a sliding window at each collection moment; Based on the number of load currents in the sliding window at each acquisition moment that are greater than the preset standard data and the difference between the load current and the preset standard current, the current anomaly coefficient at each acquisition moment is obtained; based on the discrete amplitude and information amount of the load current in the sliding window, the current fluctuation coefficient at each acquisition moment is obtained; based on the current anomaly coefficient, the current fluctuation coefficient, and the difference between the voltage data at adjacent acquisition moments, the load electrical parameter variation coefficient at each acquisition moment is obtained; Obtaining a temperature confidence coefficient for each power cycle process based on a difference between the battery temperature data and a predicted value of the battery temperature data; obtaining a loss accumulation coefficient for each power cycle process based on the temperature confidence coefficient and a distance between temperature data in any two power cycle processes; Obtaining the battery capacity correction factor at each collection moment of each power cycle process based on the load electrical parameter variation coefficient, loss accumulation coefficient and battery temperature data; The available capacity of the battery pack in each power cycle is obtained based on the battery capacity correction factor and the function of the load current changing with time; the method for obtaining the current abnormality coefficient is: For the sliding window of each collection moment, the ratio of the number of collection moments in the sliding window at which the load current data is greater than the preset standard current to the number of all collection moments in the sliding window is calculated as the first ratio of each collection moment; For each collection moment, the difference between the load current data at the collection moment and the preset standard current is calculated, and the calculation result of an exponential function with a natural constant as the base and the difference as the exponent is calculated, and the product of the first ratio at each collection moment and the calculation result of the exponential function is used as the current abnormality coefficient at each collection moment; the method for obtaining the current fluctuation coefficient is: For each acquisition moment, the range and information entropy of the load current at all acquisition moments within the sliding window of the acquisition moment are calculated, and the product of the range and information entropy is used as the current fluctuation coefficient at each acquisition moment; the method for obtaining the load electric parameter variation coefficient is: The load electrical parameter variation coefficient ; In the formula, For the The load electrical parameter variation coefficient at each acquisition moment; Indicates The current anomaly coefficient at each acquisition moment, For the The current fluctuation coefficient at each acquisition moment, e is a natural constant, , Respectively represent The collection time, The voltage data at the time of acquisition; the method for obtaining the temperature confidence coefficient is: Fitting the temperature data in each power cycle process to obtain the time series curve of each power cycle process; using a detrending algorithm to detrend the time series curve of each power cycle process to obtain the temperature fluctuation curve of each power cycle process; The data points at each acquisition time in the temperature fluctuation curve of each power cycle process are used as inputs of the prediction algorithm to obtain the predicted temperature values ​​at each acquisition time of the next adjacent power cycle process; For each power cycle process, the absolute value of the difference between the predicted temperature value at each acquisition moment in the power cycle process and the battery temperature data is calculated, and the sum of all the absolute values ​​of the difference in the power cycle process is used as the temperature confidence coefficient of each power cycle process; the calculation formula of the loss accumulation coefficient is: ; In the formula, For the The cumulative loss coefficient of the electric energy cycle process; It is The temperature confidence coefficient of the temperature data during the power cycle, , They are sequence Temperature fluctuation curve of temperature data during the power cycle, Indicates Before the power cycle The number of electrical energy cycles, is the DTW distance; the method for obtaining the battery capacity correction factor is: for each electric energy cycle process, the normalized value of the loss accumulation coefficient of the electric energy cycle process is calculated, the product of the normalized value and the battery temperature data at each acquisition time in the electric energy cycle process is calculated as the first product, the calculation result of an exponential function with a natural constant as the base and the first product as the exponent is calculated, and the product of the calculation result and the load electric parameter variation coefficient at each acquisition time in the electric energy cycle process is used as the battery capacity correction factor at each acquisition time in each electric energy cycle process; the method for obtaining the available capacity is: ; In the formula, The available capacity of the battery pack in each power cycle process; Indicates the discharge time of the electric energy cycle process, It represents the function of the load current changing with time during each power cycle. Indicates the first Battery capacity correction factor at each acquisition moment; Represents the normalization function.

2. A battery capacity offline detection device for electric energy cycling, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the battery capacity offline detection method for electric energy cycling as claimed in claim 1 are implemented.

3. A battery capacity offline detection system for electric energy cycling, which implements the battery capacity offline detection method for electric energy cycling as claimed in claim 1, characterized in that: The detection system comprises: A data acquisition module is used to collect load current data, voltage data and battery temperature data of an offline battery pack at each acquisition moment during each power cycle; A battery capacity correction factor acquisition module is used to take each acquisition moment as the center and use a window of a preset size as a sliding window at each acquisition moment; based on the number of load currents in the sliding window at each acquisition moment that are greater than the preset standard data and the difference between the load current and the preset standard current, obtain the current anomaly coefficient at each acquisition moment; based on the discrete amplitude and information amount of the load current in the sliding window, obtain the current fluctuation coefficient at each acquisition moment; based on the current anomaly coefficient, the current fluctuation coefficient, and the difference between the voltage data at adjacent acquisition moments, obtain the load electrical parameter variation coefficient at each acquisition moment; based on the difference between the battery temperature data and the predicted value of the battery temperature data, obtain the temperature confidence coefficient of each electric energy cycle process; based on the temperature confidence coefficient and the distance between the temperature data in any two electric energy cycle processes, obtain the loss accumulation coefficient of each electric energy cycle process; based on the load electrical parameter variation coefficient, the loss accumulation coefficient and the battery temperature data, obtain the battery capacity correction factor at each acquisition moment of each electric energy cycle process; The battery capacity offline detection module is used to obtain the available capacity of the battery pack during each power cycle based on the battery capacity correction factor and the change function of the load current over time.

Images