A Method for Discovering the Root Cause of Self-Discharge in a Power Battery System Based on a Fault Tree

Abstract

The present invention relates to the field of vehicle battery anomaly diagnosis methods, and specifically relates to a method for exploring the root cause of self-discharge of a power battery system based on a fault tree, including the following steps: Step 1, collect the real vehicle data characteristics of multiple faulty vehicles; Step 2, obtain the fault data characteristics in the real vehicle data characteristics, and establish a self-discharge anomaly fault tree for the fault data characteristics according to the logical relationship of fault generation based on the fault factors mined in the actual production and manufacturing and test verification processes; Step 3, monitor whether the real vehicle data is abnormal. If so, match the root cause information from the self-discharge anomaly fault tree according to the abnormal performance information of the vehicle. The present invention has an accurate optimization and rectification direction for the defects of the power battery system and a complete fault rectification.

Description

Technical Field

[0001] The present invention relates to the field of vehicle battery abnormality diagnosis methods, and in particular to a method for discovering the root cause of self-discharge of a power battery system based on a fault tree. Background Art

[0002] Self-discharge (internal short circuit) is a common failure mode in new energy vehicle battery systems. At present, most fire accidents in new energy vehicles are related to self-discharge (internal short circuit) of the battery system. With the increasing number of new energy vehicles, the exploration of the self-discharge problem of new energy vehicles has become particularly important. Regarding the self-discharge problem, most existing technologies focus on the screening and judgment of self-discharge (internal short circuit) or self-discharge rate. The causes of self-discharge are complex and varied. From battery cells to battery packs, it is difficult to accurately locate and obtain the root cause of the self-discharge problem during production and use. Summary of the invention

[0003] The present invention aims to provide a method for discovering the root cause of self-discharge of a power battery system based on a fault tree, so as to solve the problem that the root cause of self-discharge of the battery system is difficult to obtain.

[0004] The method for discovering the root cause of self-discharge of a power battery system based on a fault tree in this scheme includes the following steps:

[0005] Step 1, collecting real vehicle data features of multiple faulty vehicles;

[0006] Step 2, obtaining fault data features from real vehicle data features, and establishing the fault data features into a self-discharge abnormality fault tree according to the logical relationship of the fault generation based on the actual production and manufacturing and the fault factors mined during the test verification process;

[0007] Step 3: monitor whether the actual vehicle data is abnormal. If so, match the root cause information from the self-discharge abnormality fault tree based on the abnormal performance information of the vehicle.

[0008] The beneficial effects of this program are:

[0009] According to the logical relationship of the fault generation, the fault tree of abnormal self-discharge is established based on the data characteristics of the actual vehicle, which comprehensively covers the root causes of self-discharge (internal short circuit) failure of the battery system. After monitoring the self-discharge of the power battery system of the actual vehicle, the fault tree of abnormal self-discharge is compared and matched to dig out the root cause of self-discharge. The optimization and rectification direction of the defects of the power battery system is accurate and the fault rectification is complete.

[0010] Furthermore, in step 1, it is determined whether there is a self-discharge abnormality based on the manifestation of the actual vehicle data characteristics, and the manifestation is that the single cell voltage gradually deviates from the group. When there is no self-discharge abnormality, step 2 is entered.

[0011] The beneficial effect is: first determine whether the actual vehicle data is self-discharge abnormality, so as to avoid the error factors in the subsequent generation of the self-discharge abnormality fault tree.

[0012] Furthermore, in step 2, chemical self-discharge and physical self-discharge are used as top-level events according to the generation principle of fault data features; according to the causes of chemical self-discharge and physical self-discharge, some fault data features are added to the top-level events as middle-level events; according to the failure causes, some fault data features are added to the middle-level events as bottom-level events.

[0013] The beneficial effect is that the self-discharge fault tree is built layer by layer from the two main failure modes of physical failure and chemical failure, and different fault data features are added to different events to form a self-discharge abnormal fault tree with clear hierarchy.

[0014] Furthermore, in step 2, logical relationship symbols are added between the top-level events, the middle-level events and the bottom-level events according to the logical relationship generated by the fault, and a self-discharge abnormality fault tree is formed by the top-level events, the middle-level events, the bottom-level events and the logical relationship symbols.

[0015] The beneficial effect is: accurately defining the relationship between each event and the fault through the logical relationship symbol, and improving the judgment speed of root cause information matching.

[0016] Furthermore, in step 3, when the actual vehicle data is abnormal, the intermediate layer events are obtained as abnormal performance information by unpacking the battery pack and disassembling the battery cells, and the bottom layer events are matched as root cause information in the self-discharge abnormal fault tree according to the abnormal performance information.

[0017] The beneficial effects are: matching the underlying events as root cause information according to the actual performance of the vehicle abnormality, the process of matching the root cause information is orderly, the matching speed is fast and it is not easy to make errors.

[0018] Furthermore, the method also includes step 4, after matching the root cause information, performing defect improvement on the battery system according to the root cause information.

[0019] The beneficial effect is: in order to ensure that the root cause of the battery system failure is matched and dealt with in time, the defect is improved immediately to avoid the occurrence of similar accidents.

[0020] Further, in step 2, the intermediate layer events under the physical self-discharge fault include the short circuit between the pole and the aluminum shell, anode copper deposition, anode dendrites and diaphragm failure with logical relationship, the bottom layer events under the short circuit fault between the pole and the aluminum shell include the pole burrs and the conductive foreign matter of the insulating ring with logical relationship, the bottom layer events under the anode copper deposition fault include the charge and discharge and reverse charge with logical relationship, the bottom layer events under the anode dendrite fault include the process abnormality of overcurrent, overcharge and intermediate layer events with logical relationship, and the bottom layer events under the process abnormality fault The events include poor alignment, uneven pole piece coating, formation process defects and dissolution of positive pole magnetic material in an OR logical relationship. The diaphragm failure fault also includes the intermediate layer events of diaphragm damage and diaphragm shrinkage in an OR logical relationship. The underlying events under the diaphragm damage fault include the OR logical relationship of foreign electricity and foreign matter, diaphragm defects and stress concentration. The diaphragm shrinkage fault includes the OR logical relationship of outer membrane heat source and intermediate layer event local overtemperature. The underlying events under the local overtemperature fault include the OR logical relationship of local lithium precipitation and local material drop.

[0021] The beneficial effect is that a fault tree for physical self-discharge is established, the fault cause is complete and accurate, and the logical relationship is clear.

[0022] Further, in step 2, the chemical self-discharge fault includes the intermediate events of aluminum shell corrosion, the intermediate event of cathode decomposition, the intermediate event of SEI abnormality and the underlying event of over-discharge copper precipitation in an OR logical relationship; the aluminum shell corrosion fault includes the underlying events of blue film damage and conduction and material falling off in the battery cell in an OR logical relationship; the cathode decomposition fault includes the underlying events of overcharging, high temperature and active material breakage in an OR logical relationship; the SEI abnormality includes the intermediate events of negative electrode SEI abnormality and positive electrode SEI abnormality in an OR logical relationship; the negative electrode SEI abnormality includes the underlying events of high temperature, overvoltage, over-discharge and overcurrent in an OR logical relationship; the positive electrode SEI abnormality includes the underlying events of overtemperature, overvoltage and overcurrent in an OR logical relationship.

[0023] The beneficial effect is: the causes of chemical self-discharge are classified in detail, the logical relationship is clear, and the fault causes are accurate and complete

[0024] Furthermore, in step 3, after the root cause information is matched, the root cause information is displayed in sequence from the bottom event to the top event in a dynamic manner according to the logical relationship formed by the root cause information.

[0025] The beneficial effect is: the matched root causes are displayed intelligently, dynamically and hierarchically, so that the causes of vehicle failures can be understood intuitively and quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flowchart of an embodiment of a method for discovering the root cause of self-discharge of a power battery system based on a fault tree according to the present invention;

[0027] Figure 2 A schematic diagram of a fault tree in an embodiment of a method for discovering root causes of self-discharge of a power battery system based on a fault tree according to the present invention;

[0028] Figure 3 A schematic diagram of cell voltage in an embodiment of a method for discovering the root cause of self-discharge of a power battery system based on a fault tree of the present invention;

[0029] Figure 4 It is an enlarged diagram of the data curve of the accident stage in the embodiment of the method for discovering the root cause of self-discharge of a power battery system based on a fault tree of the present invention;

[0030] Figure 5 It is a schematic block diagram of an embodiment of a method for discovering the root cause of self-discharge of a power battery system based on a fault tree according to the present invention. DETAILED DESCRIPTION

[0031] The following is further explained in detail through specific implementation methods.

[0032] Embodiment 1

[0033] like Figure 1 As shown, the method for discovering the root cause of self-discharge of a power battery system based on a fault tree includes the following steps:

[0034] Step 1, collect the real vehicle data characteristics of multiple faulty pure electric vehicles. The real vehicle data is collected and stored through the vehicle's built-in battery management system (BMS). According to the manifestation of the real vehicle data characteristics, determine whether it is a self-discharge abnormality. The manifestation is that the single cell voltage gradually deviates from the group. When there is no self-discharge abnormality, go to step 2.

[0035] Step 2, obtain the fault data features in the real vehicle data features, according to the actual production and manufacturing, and the fault factors mined in the test verification process, or use the existing discrete convolution wavelet transform signal processing method (DCWT) and point neuron technology to process respectively, extract the fault data features of the self-discharge anomaly of the real vehicle data, and establish the fault data features into a self-discharge anomaly fault tree according to the logical relationship of the fault generation. Specifically, according to the generation principle of the fault data features, chemical self-discharge and physical self-discharge are used as top-level events. According to the causes of chemical self-discharge and physical self-discharge, some fault data features are added to the top-level events as middle-level events. The middle-level events under the physical self-discharge top-level event include short circuit between pole and aluminum shell, copper deposition on anode, anode dendrite, and diaphragm failure. The middle-level events under the chemical self-discharge top-level event include aluminum shell corrosion, over-discharge copper deposition, cathode decomposition, and SEI anomaly. According to the failure cause, some fault data features are added to the middle layer event as the underlying event. The underlying events under the middle layer event of short circuit between the pole and the aluminum shell include pole burr piercing and foreign matter conduction in the insulating ring. The underlying events under the middle layer event of anode copper deposition include over-discharge and over-charge. The underlying events under the middle layer event of anode dendrite include over-current, over-charge and process abnormality. The process abnormality is further subdivided into the following underlying events: poor pole piece alignment, uneven coating, formation process defects, and dissolution of positive magnetic material.

[0036] The intermediate layer events of diaphragm failure include diaphragm damage and diaphragm shrinkage, among which diaphragm damage includes three underlying events: the influence of conductive foreign matter impurities, diaphragm defects and stress concentration; diaphragm shrinkage includes two underlying events: external heat source and local overtemperature, and local overtemperature includes two underlying events: local lithium precipitation and local material drop; aluminum shell corrosion includes two underlying events: blue film damage and conduction and material drop in the battery cell; cathode decomposition includes three underlying events: overcharge, high temperature and active material breakage; SEI abnormality includes two intermediate layer events: negative SEI abnormality and positive SEI abnormality; negative SEI abnormality includes four underlying events: high temperature, overvoltage, overcurrent and overdischarge; positive SEI abnormality includes three underlying events: overtemperature, overvoltage and overcurrent. The fault tree of this embodiment includes 31 underlying events, covering the basic conditions of all self-discharge abnormalities in the battery system, and realizing full guidance of the causes of faulty parts.

[0037] According to the logical relationship of the fault, logical relationship symbols are added between the top-level event, the middle-level event and the bottom-level event. The logical relationship symbols include sequential AND gates and OR gates. The top-level event, the middle-level event, the bottom-level event and the logical relationship symbols form a self-discharge abnormal fault tree. The self-discharge abnormal fault tree is as follows: Figure 2 shown.

[0038] Step 3, monitor whether the actual vehicle data is abnormal. If so, match the root cause information from the self-discharge abnormal fault tree based on the abnormal performance information of the vehicle. When the actual vehicle data is abnormal, obtain the intermediate layer events as the abnormal performance information by unpacking the battery pack and disassembling the battery cell, and match the bottom layer events as the root cause information in the self-discharge abnormal fault tree based on the abnormal performance information.

[0039] Step 4: After matching the root cause information, the battery system is defect-corrected according to the root cause information to overcome the defect of the battery system.

[0040] like Figure 3 and Figure 4 As shown, taking the actual vehicle data of two vehicles as an example, it can be judged that both vehicles have self-discharge fault anomalies from the gradual deviation of the single cell voltages of the two vehicles. Figure 3 The deviation voltage distribution of the vehicle cells gradually increased, indicating that the power battery system had self-discharge. After on-site maintenance, the cell voltage returned to normal, and the vehicle did not experience thermal runaway. Combined with data feature analysis and Figure 5 Process, the vehicle was chemically self-discharging. After analysis of the vehicle data, there was no overcharging phenomenon. The blue film was damaged and the foreign matter was conductive. Finally, it was discovered that the blue film was damaged. After unpacking and maintenance, it returned to normal. Figure 4 One of the battery cells of the vehicle shown showed a small amount of voltage deviation. After a brief voltage deviation, the voltage of the battery cell suddenly dropped, causing thermal runaway. Combined with the fault tree, it can be seen that the vehicle has an internal battery cell abnormality. The battery cell was disassembled and it was finally discovered that the diaphragm failure caused the battery cell to suddenly fail after a brief self-discharge. The internal short circuit of the battery cell caused the thermal runaway event. Based on the fault cause, the problems in the production process of the same batch of battery cells were investigated and improvements were made to avoid similar accidents in the future.

[0041] This embodiment obtains the actual vehicle data generated from the faulty vehicle of the pure electric vehicle, extracts the fault data features from the real vehicle data, and establishes the fault tree of the abnormal self-discharge of the battery cell with the fault data features, which can improve the versatility of the established fault tree for the root cause analysis of faults for different pure electric vehicles. Compared with the existing analysis of electrical connection abnormalities between battery cells and modules, this embodiment targets self-discharge (internal short circuit) faults, and the fault tree can effectively guide the root cause analysis of the faulty vehicle, discover hidden problems from the battery cell to the battery pack, in the production, manufacturing, and use process, and covers more than 90% of the faults that can be generated by the battery of the pure electric vehicle from multiple aspects of the battery cell. This embodiment improves the accuracy of the root cause results of the fault by 80%, thereby solving the root cause of the fault and reducing the occurrence of similar faults.

[0042] Embodiment 2

[0043] The difference from the embodiment is that in step 3, after matching the root cause information, the root cause information is dynamically displayed from the bottom event to the top event in sequence according to the logical relationship formed by the root cause information. The dynamic mode can be displayed by lighting up or changing the eye-catching color of the events one by one.

[0044] The matched root causes are displayed intelligently, dynamically, and hierarchically, which enables users to intuitively and quickly understand the cause of vehicle failures, facilitate rapid fault location, and improve vehicle batteries.

[0045] The above is only an embodiment of the present invention, and the common knowledge such as the known specific structure and characteristics in the scheme is not described in detail here. It should be pointed out that for those skilled in the art, several deformations and improvements can be made without departing from the structure of the present invention, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for discovering the root cause of self-discharge in a power battery system based on a fault tree, characterized in that, it includes the following steps: Step 1, collect the real vehicle data characteristics of multiple faulty vehicles; Step 2, obtain the fault data characteristics in the real vehicle data characteristics, and establish a self-discharge abnormal fault tree for the fault data characteristics according to the logical relationship of fault generation based on the fault factors mined in the actual production and manufacturing and test verification processes. Take chemical self-discharge and physical self-discharge as the top events according to the generation principle of the fault data characteristics; According to the causes of chemical self-discharge and physical self-discharge, add some fault data characteristics to the top events as intermediate events; According to the failure causes, add some fault data characteristics to the intermediate events as bottom events, and add logical relation symbols between the top events, intermediate events and bottom events according to the logical relationship of fault generation. The self-discharge abnormal fault tree is formed by the top events, intermediate events, bottom events and logical relation symbols; The intermediate events under the physical self-discharge fault include the short circuit between the terminal post and the aluminum shell with an AND logic relationship, copper deposition on the anode, anode dendrite and diaphragm failure. The bottom events under the short circuit fault between the terminal post and the aluminum shell include the terminal post burr and conductive foreign matter in the insulating ring with an OR logic relationship. The bottom events under the copper deposition on the anode fault include charge and discharge and reverse charge with an AND logic relationship. The bottom events under the anode dendrite fault include overcurrent, overcharge and process abnormality of the intermediate event with an OR logic relationship. The bottom events under the process abnormality fault include poor alignment, uneven electrode coating, formation process defect and dissolution of positive magnetic substances with an OR logic relationship. The intermediate events under the diaphragm failure fault also include diaphragm damage and diaphragm shrinkage with an OR logic relationship. The bottom events under the diaphragm damage fault include conductive foreign matter, diaphragm self-defect and stress concentration with an OR logic relationship. The bottom events under the diaphragm shrinkage fault include external film heat source and local overheating of the intermediate event with an OR logic relationship. The bottom events under the local overheating fault include local lithium deposition and local material loss with an OR logic relationship; The intermediate events under the chemical self-discharge fault include aluminum shell corrosion with an OR logic relationship, cathode decomposition, SEI abnormality and bottom event over-discharge copper deposition. The bottom events under the aluminum shell corrosion fault include blue film breakage conduction and material loss inside the battery cell with an OR logic relationship. The bottom events under the cathode decomposition fault include overcharge, high temperature and active material breakage with an OR logic relationship. The SEI abnormality includes negative electrode SEI abnormality and positive electrode SEI abnormality with an OR logic relationship. The bottom events under the negative electrode SEI abnormality include high temperature, overvoltage, over-discharge, overcurrent. The bottom events under the positive electrode SEI abnormality include overheating, overvoltage and overcurrent; Step 3, monitor whether the real vehicle data is abnormal. If so, match the root cause information from the self-discharge abnormal fault tree according to the abnormal performance information of the vehicle. When the real vehicle data is abnormal, obtain the intermediate event as the abnormal performance information through battery pack unpacking and cell disassembly, and match the bottom event as the root cause information in the self-discharge abnormal fault tree according to the abnormal performance information.

2. The method for discovering the root cause of self-discharge of a power battery system based on a fault tree according to claim 1, characterized in that: in step 1, it is judged whether there is an abnormal self-discharge according to the manifestation form of the actual vehicle data characteristics, and the manifestation form is that the voltage of a single cell gradually becomes outlier. When there is no abnormal self-discharge, step 2 is entered.

3. The method for discovering the root cause of self-discharge of a power battery system based on a fault tree according to claim 1, characterized in that: it further includes step 4. After matching the root cause information, the battery system is improved for defects according to the root cause information.

4. The method for discovering the root cause of self-discharge of a power battery system based on a fault tree according to any one of claims 1-3, characterized in that: in step 3, after matching the root cause information, it is displayed sequentially from the bottom event to the top event in a dynamic manner according to the logical relationship formed by the root cause information.

Images