Outlier cell identification algorithm

Abstract

The invention discloses an outlier cell recognition algorithm, which comprises the following steps of: 1, obtaining characteristic parameters of a battery in an effective charging process according to all real-time battery charging and discharging data returned by a vehicle; step 2, according to the characteristic parameters of the effective charging process of the battery in the step 1, determining all characteristic parameters of the initial voltage and the end voltage of the battery cell; 3, according to the characteristic parameters of the battery cell initial voltage and the battery cell end voltage in the step 2, determining all the characteristic parameters of the battery cell ohm internal resistance and the battery cell polarization internal resistance; and step 4, according to the characteristic parameters of the battery cell initial voltage, the battery cell end voltage, the battery cell ohm internal resistance and the battery cell polarization internal resistance determined in the step 2 and the step 3, adopting a clustering method to detect the outlier battery cells. The outlier battery cell recognition algorithm based on the multi-dimensional characteristics is high in adaptability, the characteristics of the battery cells can be evaluated and recognized from multiple dimensions, and the calculation accuracy is high.

Description

Technical Field

[0001] The present invention relates to the technical field of balanced charging and discharging of power batteries, and in particular to an outlier cell identification algorithm. Background Art

[0002] At present, with the rapid development of the electric vehicle industry, power batteries are an important part of electric vehicles and the main power source of new energy vehicles. Their performance directly affects the driving experience of the car, and then affects consumers' willingness to buy. The power battery of an electric vehicle is generally a battery pack structure, which is usually composed of hundreds of monomers connected in series and parallel. Due to production errors and the influence of external environmental stress, there will inevitably be inconsistencies in parameters between monomers, and as the vehicle battery is continuously charged and discharged, the consistency difference will gradually increase, and outliers will appear. The performance of the power battery pack is limited by the performance of the "weakest" monomer. Identifying the "weakest" monomer and replacing or removing it can increase the vehicle's driving range.

[0003] In the prior art, when extracting the charging data segment of a battery pack, the attenuation degree of the battery cell during the charging process is often determined by the voltage rise or the amount of charge in the charging data segment, thereby detecting outlier cells. These methods only use a single feature point to characterize the current state of the battery cell, while the sources of inconsistency in a battery pack are multidimensional and complex. A single feature will result in low accuracy and one-sidedness in the detection results.

[0004] To solve this problem, we designed an outlier cell identification algorithm to address this issue. Summary of the invention

[0005] In view of the above technical problems in the prior art, the present invention provides an outlier cell identification algorithm to solve the problem of considering the impact of different factors on balancing from multiple perspectives and angles in a complex and changeable environment, and formulating different balancing control strategies to make better use of battery efficiency and improve battery safety.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An outlier cell identification algorithm, the following steps:

[0008] Step 1: Based on all the real-time battery charging and discharging data sent back by the vehicle, characteristic parameters of the effective charging process of the battery are obtained;

[0009] Step 2: determining characteristic parameters of the starting voltage and the ending voltage of all the cells according to the characteristic parameters of the effective charging process of the battery in step 1;

[0010] Step 3: Determine the characteristic parameters of the ohmic internal resistance and polarization internal resistance of all cells based on the characteristic parameters of the starting voltage and ending voltage of the cells in Step 2;

[0011] Step 4: Detect outlier cells using a clustering method based on the characteristic parameters of the starting voltage, ending voltage, ohmic internal resistance, and polarization internal resistance of the cells determined in Step 2 and Step 3.

[0012] As a further elaboration of the above technical solution:

[0013] In the above technical solution, in Step 1, the characteristic parameters of the effective battery charging process include: determining cell charging data based on the battery charging status and battery current direction recorded from the real-time battery charge and discharge data; determining whether the battery charging data is an effective charging process based on the rest time before battery charging starts, the single-cell voltage at the end of battery charging, and the rest time after battery charging ends.

[0014] In the above technical solution, in Step 2, the voltage at the moment immediately before the start of battery charging in the effective battery charging process obtained in Step 1 is used as the characteristic parameter of the starting voltage of the cell and denoted as characteristic parameter 1, which represents the open-circuit voltage characteristic of the cell in the current state; the voltage at the end of the effective charging process obtained in Step 1 is used as the characteristic parameter of the ending voltage of the cell and denoted as characteristic parameter 2, which represents the dynamic voltage characteristic of the cell.

[0015] In the above technical solution, in Step 3, the difference between the voltage at the first sampling point and the ending voltage after the effective charging process obtained in Step 1 and Step 2 is used as the characteristic parameter of the ohmic internal resistance of the cell and denoted as characteristic parameter 3; the voltage at the tenth sampling point after the effective charging process obtained in Step 1 and Step 2 is used as the characteristic parameter of the polarization internal resistance of the cell and denoted as characteristic parameter 4.

[0016] In the above technical solution, in Step 4, perform normalization processing on characteristic parameter 1, characteristic parameter 2, characteristic parameter 3, and characteristic parameter 4, and use a clustering algorithm and a clustering validity evaluation method to confirm the detection result of the best outlier cells, thereby detecting outlier cells.

[0017] Advantages of the present invention:

[0018] The design of the present invention is reasonable. By analyzing and identifying outlier battery cells in real time based on the big data of vehicle operation, the problematic battery cells can be selected, helping the battery pack to quickly replace the problematic battery cells and improve the overall performance of the battery pack, such as safety and high endurance. The present invention can identify these extremely deviated battery cells, and minimize the impact of the inconsistency of the battery pack through isolation or replacement, so that the performance of the battery pack in terms of safety, service life, etc. is not weakened by the inconsistency as compared with any single cell. The present invention maximally improves the cruising range of the whole vehicle and ensures the safe driving of the whole vehicle. By analyzing and identifying outlier battery cells based on the real-time big data of vehicle operation, it can extract the charging data of all battery cells according to the real-time data transmitted back by the vehicle, determine the effective charging process of the battery cells, extract characteristic parameters based on the data of the effective charging process, and compare and analyze the differences between the battery cells, so as to detect outlier battery cells. It has strong adaptability, easy access to data, no need to match additional empirical parameters, high accuracy in calculating characteristic parameters and strong physical meaning, and can identify outlier battery cells according to statistical theory, thus solving the technical problems urgently needed to be solved in the field of analysis of the attenuation and failure of the cruising range of vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is the flowchart of the operation of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The present invention will be described in detail below in conjunction with specific embodiments and the accompanying drawings.

[0021] Please refer to Figure 1 , this embodiment provides an outlier battery cell identification algorithm, which includes the following steps:

[0022] Step 1: According to all the real-time battery charge and discharge data transmitted back by the vehicle, the characteristic parameters of the effective charging process of the battery are obtained;

[0023] Step 2: According to the characteristic parameters of the effective charging process of the battery in Step 1, the characteristic parameters of the starting voltage and the ending voltage of all battery cells are determined;

[0024] Step 3: According to the characteristic parameters of the starting voltage and the ending voltage of the battery cells in Step 2, the characteristic parameters of the ohmic internal resistance and the polarization internal resistance of all battery cells are determined;

[0025] Step 4: According to the characteristic parameters of the starting voltage, the ending voltage, the ohmic internal resistance and the polarization internal resistance of the battery cells determined in Step 2 and Step 3, the outlier battery cells are detected by using the clustering method.

[0026] As a further improvement of the present invention, in step one, the characteristic parameters of the effective charging process of the battery include: determining the cell charging data based on the battery charging state and the battery current direction recorded according to the real-time battery charge and discharge data; determining whether the above battery charging data is an effective charging process based on the static time before the start of battery charging, the monomer voltage at the end of battery charging, and the static time after the end of battery charging.

[0027] As a further improvement of the present invention, in step two, the voltage at the previous moment when the battery starts charging in the effective charging process of the battery obtained in step one is used as the characteristic parameter of the starting voltage of the cell and denoted as characteristic parameter 1, which represents the open-circuit voltage characteristic of the cell in the current state; the voltage at the moment when the charging ends in the above effective charging process obtained in step one is used as the characteristic parameter of the ending voltage of the cell and denoted as characteristic parameter 2, which represents the dynamic voltage characteristic of the cell.

[0028] As a further improvement of the present invention, in step three, the difference between the voltage at the first sampling point and the ending voltage after the end of the effective charging process obtained in steps one and two is used as the characteristic parameter of the ohmic internal resistance of the cell and denoted as characteristic parameter 3; the voltage at the tenth sampling point after the end of the effective charging process obtained in steps one and two is used as the characteristic parameter of the polarization internal resistance of the cell and denoted as characteristic parameter 4.

[0029] As a further improvement of the present invention, in step four, characteristic parameters 1, 2, 3, and 4 are normalized, and the clustering algorithm and the clustering validity evaluation method are used to confirm the detection result of the best outlier cell, so as to detect the outlier cell.

[0030] In the description of the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", "connection", and "fixation" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication of two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0031] The standard parts used in the present invention can all be purchased from the market. The special-shaped parts can be customized according to the description of the specification and the drawings. The specific connection methods of each part all adopt conventional means such as bolts, rivets, and welding that are mature in the prior art. The machines, parts, and equipment all adopt conventional models in the prior art, and the circuit connection adopts the conventional connection method in the prior art, which will not be elaborated here.

[0032] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art will appreciate that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting the protection scope of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. An outlier cell identification algorithm, characterized in that, The following steps: Step 1: Obtain the characteristic parameters of the effective battery charging process based on all the real-time battery charge and discharge data transmitted back by the vehicle. Step 2: Determine the characteristic parameters of the starting voltage and ending voltage of all the battery cells based on the characteristic parameters of the effective battery charging process in Step 1. Step 3: Determine the characteristic parameters of the ohmic internal resistance and polarization internal resistance of all the battery cells based on the characteristic parameters of the starting voltage and ending voltage of the battery cells in Step 2. Step 4: Detect the outlier battery cells using the clustering method based on the characteristic parameters of the starting voltage, ending voltage, ohmic internal resistance, and polarization internal resistance of the battery cells determined in Step 2 and Step 3.

2. The outlier cell identification algorithm according to claim 1, wherein In Step 1, the characteristic parameters of the effective battery charging process include: determining the cell charging data based on the battery charging state and battery current direction recorded from the real-time battery charge and discharge data; determining whether the battery charging data is an effective charging process based on the static time before the battery charging starts, the cell voltage at the end of the battery charging, and the static time after the battery charging ends.

3. The outlier cell identification algorithm according to claim 2, wherein In Step 2, the voltage at the moment immediately before the battery starts charging in the effective battery charging process obtained in Step 1 is used as the characteristic parameter of the starting voltage of the battery cell and denoted as characteristic parameter 1, which represents the open-circuit voltage characteristic of the battery cell under the current state; the voltage at the moment when the effective charging process ends obtained in Step 1 is used as the characteristic parameter of the ending voltage of the battery cell and denoted as characteristic parameter 2, which represents the dynamic voltage characteristic of the battery cell.

4. The outlier cell identification algorithm according to claim 3, wherein In Step 3, the difference between the voltage at the first sampling point after the effective charging process ends and the ending voltage obtained in Step 1 and Step 2 is used as the characteristic parameter of the ohmic internal resistance of the battery cell and denoted as characteristic parameter 3; the voltage at the tenth sampling point after the effective charging process ends obtained in Step 1 and Step 2 is used as the characteristic parameter of the polarization internal resistance of the battery cell and denoted as characteristic parameter 4.

5. The outlier cell identification algorithm according to claim 4, wherein In Step 4, perform normalization processing on characteristic parameter 1, characteristic parameter 2, characteristic parameter 3, and characteristic parameter 4, and use the clustering algorithm and clustering validity evaluation method to confirm the detection result of the best outlier battery cells, thereby detecting the outlier battery cells.

Images