A method for health risk assessment of a battery

Abstract

This invention provides a method for assessing battery health risks. During battery operation, it involves collecting and preprocessing multi-dimensional operational parameters, combining internal resistance estimation, capacity decay estimation, and cycle count statistics to obtain core battery health parameters, and then determining the health level according to preset rules. For different levels, corresponding status recording, alarm generation, and data update operations are performed. The advantages of this invention are: it enables dynamic and accurate identification of battery health risks, and the entire process is automatically executed during normal equipment operation, requiring no special testing conditions, significantly improving the timeliness and accuracy of the assessment.

Description

Technical Field

[0001] This invention relates to the field of battery health management technology, and in particular to a method for assessing the health risks of batteries. Background Technology

[0002] With the widespread application of mobile payment terminals, self-service terminals, industrial PDAs, and portable electronic devices, these devices generally rely on rechargeable batteries as their primary or backup power source, and are characterized by long continuous operating times, diverse operating environments, and long maintenance cycles. During long-term use, batteries inevitably experience capacity decay, increased internal resistance, and changes in thermal characteristics. Changes in their health status not only affect the device's range but may also lead to decreased charging and discharging efficiency, abnormal temperature rise, and even safety risks. Therefore, continuous monitoring and risk assessment of battery health status are of great significance. SOC, or State of Charge, is used in electric vehicles, mobile phones, and other electronic devices to measure the current battery charge level, indicating the remaining or charged state of the battery, and is an important indicator for managing battery life and driving range.

[0003] Current battery health risk assessment solutions generally suffer from several problems: battery health assessments rely heavily on static or offline parameters, lacking the ability to continuously perceive operational characteristics and failing to reflect dynamic changes in actual battery operation; key health indicators such as internal resistance and capacity decay are difficult to obtain automatically and reliably during normal equipment operation, typically requiring dedicated testing conditions or a complete charge-discharge cycle; risk identification methods are simplistic, often based on fixed temperature thresholds or single parameters, making it difficult to promptly detect potential risks caused by aging, usage behavior, or load changes; and there is a lack of a continuous health status learning mechanism across cycles and scenarios, making it difficult to adapt to changes throughout the battery's entire lifecycle.

[0004] To address the aforementioned issues, there is an urgent need for a method that can collect and analyze battery operating parameters from multiple dimensions and conduct health risk classification assessments without affecting the normal operation of terminal devices, in order to improve the safety and reliability of battery use and meet the management needs of the entire battery life cycle. Summary of the Invention

[0005] To address the problems in existing technologies, this invention provides a method for assessing the health risks of batteries. By collecting and preprocessing multi-dimensional operating parameters, and integrating core health parameters such as internal resistance, capacity decay, and cycle count, a graded risk assessment is conducted. This method enables dynamic and accurate identification of battery health risks and is executed automatically throughout the normal operation of the equipment, without the need for specialized testing conditions. The timeliness and accuracy of the assessment are significantly improved, solving the technical problems of single battery risk identification methods, inaccurate assessments, and poor timeliness in existing technologies.

[0006] A method for assessing the health risks of batteries according to the present invention includes the following steps: Step 1: While the battery is in operation, continuously collect the battery's voltage, current, temperature, and state of charge (SOC) data. Step 2: Perform sliding window statistical analysis on the collected voltage, current, temperature and SOC data, and filter out abnormal data to obtain preprocessed effective operating data; Step 3: Based on the effective operating data, perform internal resistance IR estimation to obtain the equivalent internal resistance parameter of the battery; at the same time, perform capacity decay estimation to obtain the effective capacity and capacity decay trend parameters of the battery; and perform cycle count statistics to obtain the charge-discharge cycle count parameter of the battery. Step 4: Using the equivalent internal resistance parameter, capacity decay trend parameter, and charge / discharge cycle number parameter as the core health parameters of the battery, and according to the preset health level judgment rules, classify the battery health status into any one of the following levels: Normal Level 1, Warning Level 2, and Critical Level 3. Step 5: Record the current battery health status and corresponding core health parameters. If the battery health status is warning level 2 or critical level 3, generate the corresponding alarm event. Finally, update all the data from this assessment to the battery history database.

[0007] The present invention is further improved in that, in step 1, the battery's operating state includes normal device use, charging and standby states, and data acquisition is performed automatically on a periodic basis, with the acquisition period adaptively adjusted according to the type of terminal device to which the battery is applied.

[0008] The present invention is further improved in that, in step 2, the window duration of the sliding window statistical analysis is 5-30s, and the window sliding step size is 1-5s; the method of filtering abnormal data is to remove data that exceeds the preset voltage, current and temperature threshold range, and at the same time, pulse interference data in the data is removed by mean filtering.

[0009] The present invention is further improved in that, in step 3, the specific process of estimating the internal resistance IR is as follows: after detecting a current change event, record the effective operating data of current and voltage before and after the current change; under the premise of meeting the data stability conditions, calculate the amount of current change and voltage change; calculate the battery equivalent internal resistance based on the ratio of voltage change to current change; and classify the equivalent internal resistance results according to the sampling environment conditions, and only use the equivalent internal resistance results with excellent quality as the effective internal resistance parameter.

[0010] The present invention is further improved in that, in step 3, the specific process of capacity decay estimation is as follows: during the battery charging process, the amount of electricity is accumulated based on current and time; the effective capacity of the battery is calculated in combination with the SOC change range; the effective capacity estimation result is smoothed and numerically constrained to obtain the capacity decay trend parameters, which include the current effective capacity, capacity decay rate and decay rate.

[0011] The present invention is further improved in that, in step 4, the preset health level determination rule is a multi-parameter fusion determination rule, specifically setting an equivalent internal resistance threshold range, a capacity decay rate threshold range, and a charge-discharge cycle number threshold range. When all three parameters are within the corresponding normal threshold range, it is determined to be normal level one; when any parameter exceeds the normal threshold range but does not reach the critical threshold, it is determined to be warning level two; when any parameter reaches the critical threshold or at least two parameters exceed the normal threshold range, it is determined to be critical level three.

[0012] In a further improvement, in step 5, generating the corresponding alarm event includes generating alarm level information matching the health level, battery abnormal parameter information, and current device operating status information; the alarm event is prompted through the local alarm module of the terminal device, and is also sent to the upper-level terminal management system or operation and maintenance system through the external interaction interface.

[0013] In a further improvement, in step 5, the alarm event interaction process is as follows: after the battery health assessment module completes the risk assessment and alarm determination, it generates standardized alarm data and performs local persistent storage; it sends an alarm request to the upper-layer alarm management module through the business alarm interface; the alarm management module records the alarm notification and returns an ACK confirmation signal; after receiving the confirmation signal, the battery health assessment module sends the complete alarm event to the alarm management module; the alarm management module records the alarm information and returns an ACK confirmation signal again, completing the alarm interaction.

[0014] The present invention is further improved in that when the terminal device is a removable multi-battery device, after the system starts up, it first performs a battery initialization scan to identify the inserted battery and check the battery's unique ID in the historical database; if it is an identified battery, it loads the previously saved battery health status and historical parameters and then executes steps 1-5; if it is a new battery, it first initializes the battery data structure and completes battery calibration, and then executes steps 1-5; when a battery is detected to be removed or a new battery is inserted, it switches and saves the corresponding battery's health status data.

[0015] The present invention is further improved by making the battery history database a persistent database, which stores the unique ID of each battery, the first insertion time, the number of charge and discharge cycles, the historical curve of equivalent internal resistance, the capacity decay trend curve, the historical health status history and alarm event records. The data in the database supports reading and updating across device operating cycles.

[0016] The beneficial effects of this invention are as follows: This invention provides a battery health risk assessment method that, through the collection and preprocessing of multi-dimensional operating parameters, integrates core health parameters such as internal resistance, capacity decay, and cycle count for graded risk assessment. This enables dynamic and accurate identification of battery health risks, and the entire process is automatically executed during normal equipment operation, requiring no special testing conditions, significantly improving the timeliness and accuracy of the assessment. Integrating multiple core health parameters such as internal resistance, capacity decay, and charge / discharge cycle count for risk assessment abandons the traditional single-parameter judgment method, enabling a more comprehensive and accurate capture of potential risks caused by battery aging, usage behavior, or load changes, avoiding missed or false judgments. The introduction of a three-level health level judgment mechanism realizes graded assessment and alarm of battery health risks, enabling early identification of potential risks and taking corresponding measures for different risk levels, effectively improving the safety and reliability of battery use. This technology reduces the probability of equipment downtime and accidents caused by battery failure. It incorporates steps such as data preprocessing, internal resistance result quality grading, and capacity estimation smoothing to effectively eliminate interfering data, ensuring the accuracy of assessment parameters and thus improving the reliability of risk assessment results. It supports multi-battery identification and switching, adapting to terminal devices with removable multi-battery configurations. The battery history database is persistently designed, supporting cross-cycle data reading and updating, enabling full lifecycle health management of batteries. Implemented in software, it can be integrated as a module into the system service layer of various terminal devices, without relying on a single hardware model or battery manufacturer. It offers flexible deployment, minimal modifications to existing equipment, and strong applicability, making it widely applicable to various devices powered by rechargeable batteries, such as financial payment terminals, industrial terminals, smart retail terminals, and IoT terminals. This solves the technical problems of existing technologies, such as single battery risk identification methods, inaccurate assessment, and poor timeliness. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for assessing the health risks of batteries according to the present invention. Detailed Implementation

[0018] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0019] Please see Figure 1 The present invention provides a method for assessing the health risks of batteries, comprising the following steps: Step 1: While the battery is in operation, continuously collect data on the battery's voltage, current, temperature, and state of charge (SOC). The battery's operating status includes normal device use, charging, and standby states. Data collection is performed automatically and periodically, with the collection cycle adaptively adjusted according to the type of terminal device used for the battery application.

[0020] Step 2: Perform sliding window statistical analysis on the collected voltage, current, temperature and SOC data, and filter out abnormal data to obtain preprocessed valid operating data. The window duration of the sliding window statistical analysis is 5-30s, and the window sliding step size is 1-5s. The abnormal data is filtered out by removing data that exceeds the preset voltage, current and temperature threshold ranges, and pulse interference data is removed by mean filtering.

[0021] Step 3: Based on the effective operating data, perform internal resistance (IR) estimation to obtain the battery's equivalent internal resistance parameter; simultaneously, perform capacity decay estimation to obtain the battery's effective capacity and capacity decay trend parameters; and perform cycle count statistics to obtain the battery's charge-discharge cycle count parameter. Specifically, the IR estimation process involves recording the effective operating data of current and voltage before and after a current change event is detected; calculating the current and voltage changes while meeting data stability conditions; calculating the battery's equivalent internal resistance based on the ratio of voltage to current changes; and classifying the equivalent internal resistance results according to the sampling environment conditions, using only the best-quality equivalent internal resistance results as the effective internal resistance parameter. The capacity decay estimation process involves accumulating the charge based on current and time during battery charging; calculating the battery's effective capacity by combining the SOC change range; smoothing and numerically constraining the effective capacity estimation results to obtain the capacity decay trend parameters, which include the current effective capacity, capacity decay rate, and decay rate.

[0022] Step 4: Using the equivalent internal resistance parameter, capacity decay trend parameter, and charge / discharge cycle count parameter as core battery health parameters, and based on preset health level determination rules, classify the battery health status into any one of the following levels: Normal Level 1, Warning Level 2, and Critical Level 3. The preset health level determination rules are multi-parameter fusion rules, specifically setting threshold ranges for equivalent internal resistance, capacity decay rate, and charge / discharge cycle count. When all three parameters are within their corresponding normal threshold ranges, it is determined to be Normal Level 1; when any parameter exceeds the normal threshold range but does not reach the critical threshold, it is determined to be Warning Level 2; when any parameter reaches the critical threshold or at least two parameters exceed the normal threshold range, it is determined to be Critical Level 3.

[0023] Step 5: Record the current battery health status and corresponding core health parameters. If the battery health status is Warning Level 2 or Critical Level 3, generate the corresponding alarm event. Finally, update all data from this assessment to the battery historical database. Generating the corresponding alarm event includes generating alarm level information matching the health level, abnormal battery parameter information, and current device operating status information. The alarm event is displayed through the local alarm module of the terminal device and simultaneously sent to the upper-level terminal management system or maintenance system through an external interaction interface. The alarm event interaction process is as follows: After the battery health assessment module completes the risk assessment and alarm determination, it generates standardized alarm data and stores it locally. It then sends an alarm request to the upper-level alarm management module through the business alarm interface. The alarm management module records the alarm notification and returns an ACK confirmation signal. After receiving the confirmation signal, the battery health assessment module sends the complete alarm event to the alarm management module. The alarm management module records the alarm information and returns an ACK confirmation signal again, completing the alarm interaction. When the terminal device is a removable multi-battery device, the system first performs a battery initialization scan after startup, identifies the inserted battery, and checks the battery's unique ID in the historical database. If the battery is already identified, it loads the previously saved battery health status and historical parameters before executing steps 1-5. If it is a new battery, it first initializes the battery data structure and completes battery calibration before executing steps 1-5. When a battery is detected to be removed or a new battery is inserted, the system switches and saves the corresponding battery's health status data. The battery history database is a persistent database, storing information including each battery's unique ID, first insertion time, charge / discharge cycle count, equivalent internal resistance historical curve, capacity decay trend curve, health status history, and alarm event records. The data in the database supports reading and updating across device operating cycles.

[0024] Please see Figure 1As an embodiment of the present invention, this embodiment is applied to the health risk assessment of lithium batteries in handheld financial payment POS machines. The POS machine is a single-battery device with a nominal lithium battery voltage of 3.7V and a nominal capacity of 2000mAh. In this embodiment, the method for assessing the health risk of the battery specifically includes: Step 1, under the full operating state of the POS machine in normal use, charging, and standby, continuously collecting the voltage, current, temperature, and SOC data of the lithium battery with a collection cycle of 2s; Step 2, performing sliding window statistical analysis on the collected data, setting the window duration to 10s and the sliding step size to 2s, while filtering abnormal data: removing data with voltage <2.5V or >4.2V, current <-5A or >2A, and temperature <0℃ or >60℃, and removing pulse interference data through mean filtering to obtain valid operating data; Step 3, performing internal resistance IR estimation based on the valid operating data: after detecting a current change event, recording the current and voltage data before and after the change, and calculating the internal resistance IR based on the data fluctuation amplitude <5% under stable conditions. Calculate the changes in current and voltage, and obtain the equivalent internal resistance by comparing the values. Classify the quality based on the operating temperature of the POS machine (ideally 25±5℃), retaining only the internal resistance parameters of the highest quality. During charging, perform capacity decay estimation: calculate the effective capacity based on the cumulative current and time-based charge, combined with the SOC range of 10%-90%. After smoothing, obtain the current effective capacity, capacity decay rate, and decay rate. Simultaneously, count the number of charge-discharge cycles, with one complete 0-100% charge-discharge cycle counted as one cycle. Step 4: Set health level judgment thresholds: normal equivalent internal resistance threshold < 80mΩ, critical threshold ≥ 150mΩ; normal capacity decay rate threshold < 20%, critical threshold ≥ 40%; normal charge-discharge cycle count threshold < 500 times, critical threshold ≥ 800 times. Substituting internal resistance, capacity decay rate, and cycle count into the rules, the battery health status is determined as Normal Level 1, Warning Level 2, or Critical Level 3; Step 5: Record the battery health status and core parameters. If it is Warning Level 2 or Critical Level 3, generate an alarm event: including alarm level, abnormal parameters, and the current transaction status of the POS machine. The alarm is displayed on the POS machine's local indicator light (yellow light for warning, red light for critical), and simultaneously sent to the merchant operation and maintenance management system via the 4G module; Finally, update the evaluation data to the POS machine's local battery persistence database.

[0025] As a second embodiment of the present invention, this embodiment is applied to the health risk assessment of lithium batteries in industrial PDAs. The industrial PDA is a removable dual-battery device with a nominal lithium battery voltage of 3.7V and a nominal capacity of 4000mAh. The method of the present invention adds a multi-battery adaptation step based on embodiment 1: After the system starts, it performs a battery initialization scan, identifies the inserted battery and checks its unique ID. If it is the used battery 1, it loads its historical health status and parameters and then executes steps 1-5 of embodiment 1. If it detects that battery 1 is removed and battery 2 is inserted, the system immediately saves the current health status of battery 1. After identifying battery 2 as a new battery, it first initializes the data structure of battery 2 and completes voltage and capacity calibration, and then executes steps 1-5. When switching back to battery 1, it loads its saved health status and continues the risk assessment. In this embodiment, the industrial PDA's data acquisition cycle is set to 3 seconds, the sliding window duration is 15 seconds, the step size is 3 seconds, the temperature threshold is -10℃ to 70℃, the normal internal resistance threshold is <100mΩ, the critical threshold is ≥180mΩ, the normal capacity decay rate threshold is <25%, the critical threshold is ≥45%, the normal charge / discharge cycle count threshold is <800 times, and the critical threshold is ≥1200 times. In addition to local notifications, alarm events are also sent to the management terminal at the industrial site via Bluetooth to achieve timely response from on-site maintenance.

[0026] As can be seen from the above, the beneficial effects of the present invention are as follows: The present invention provides a method for assessing the health risks of batteries. By collecting and preprocessing multi-dimensional operating parameters, and integrating core health parameters such as internal resistance, capacity decay, and cycle count for graded risk assessment, it can achieve dynamic and accurate identification of battery health risks. Furthermore, it is automatically executed throughout the normal operation of the equipment, requiring no special testing conditions, thus significantly improving the timeliness and accuracy of the assessment. Integrating multi-dimensional core health parameters such as internal resistance, capacity decay, and charge / discharge cycle count for risk assessment abandons the traditional single-parameter judgment method, enabling a more comprehensive and accurate capture of potential risks caused by battery aging, usage behavior, or load changes, avoiding missed or false judgments. The introduction of a three-level health level judgment mechanism enables graded assessment and alarm of battery health risks, allowing for early identification of potential risks and the implementation of corresponding handling measures for different risk levels, effectively improving the safety of battery use. This technology improves reliability and reduces the probability of equipment downtime and accidents caused by battery failure. It incorporates steps such as data preprocessing, internal resistance result quality grading, and capacity estimation smoothing to effectively eliminate interfering data, ensuring the accuracy of assessment parameters and thus enhancing the reliability of risk assessment results. It supports multi-battery identification and switching, adapting to terminal devices with removable multi-battery configurations. The battery history database is persistently designed, supporting cross-cycle data reading and updating, enabling full lifecycle health management of batteries. Implemented in software, it can be integrated as a module into the system service layer of various terminal devices, without relying on a single hardware model or battery manufacturer. It offers flexible deployment, minimal modifications to existing equipment, and strong applicability, making it widely applicable to various devices powered by rechargeable batteries, such as financial payment terminals, industrial terminals, smart retail terminals, and IoT terminals. This solves the technical problems of existing technologies, such as single battery risk identification methods, inaccurate assessment, and poor timeliness.

[0027] The specific embodiments described above are preferred embodiments of the present invention and are not intended to limit the specific scope of the present invention. The scope of the present invention includes, but is not limited to, these specific embodiments. All equivalent changes made in accordance with the present invention are within the protection scope of the present invention.

Claims

1. A method for assessing the health risks of batteries, characterized in that, Includes the following steps: Step 1: While the battery is in operation, continuously collect the battery's voltage, current, temperature, and state of charge (SOC) data. Step 2: Perform sliding window statistical analysis on the collected voltage, current, temperature and SOC data, and filter out abnormal data to obtain preprocessed effective operating data; Step 3: Based on the effective operating data, perform internal resistance IR estimation to obtain the equivalent internal resistance parameters of the battery; at the same time, perform capacity decay estimation to obtain the effective capacity and capacity decay trend parameters of the battery. It also performs cycle count statistics to obtain the battery's charge-discharge cycle count parameters; Step 4: Using the equivalent internal resistance parameter, capacity decay trend parameter, and charge / discharge cycle number parameter as the core health parameters of the battery, and according to the preset health level judgment rules, classify the battery health status into any one of the following levels: Normal Level 1, Warning Level 2, and Critical Level 3. Step 5: Record the current battery health status and corresponding core health parameters. If the battery health status is warning level 2 or critical level 3, generate the corresponding alarm event. Finally, update all the data from this assessment to the battery history database.

2. The method for assessing the health risks of batteries as described in claim 1, characterized in that: In step 1, the battery's operating status includes normal device use, charging, and standby states. Data acquisition is performed automatically on a periodic basis, and the acquisition period is adaptively adjusted according to the type of terminal device used for the battery application.

3. The method for assessing the health risks of batteries as described in claim 2, characterized in that: In step 2, the window duration of the sliding window statistical analysis is 5-30s, and the window sliding step size is 1-5s. The method of filtering abnormal data is to remove data that exceeds the preset voltage, current, and temperature threshold ranges, and at the same time, pulse interference data in the data is removed by mean filtering.

4. The method for assessing the health risks of batteries as described in claim 3, characterized in that: In step 3, the specific process of estimating the internal resistance IR is as follows: after detecting a current change event, record the effective operating data of current and voltage before and after the current change; under the premise of satisfying the data stability condition, calculate the amount of current change and voltage change. The equivalent internal resistance of the battery is calculated based on the ratio of voltage change to current change; and the quality of the equivalent internal resistance results is graded according to the sampling environment conditions, with only the best equivalent internal resistance results being used as the effective internal resistance parameter.

5. The method for assessing the health risks of batteries as described in claim 4, characterized in that: In step 3, the specific process of capacity decay estimation is as follows: during battery charging, the amount of electricity is accumulated based on current and time; the effective capacity of the battery is calculated in combination with the SOC change range; the effective capacity estimation result is smoothed and numerically constrained to obtain the capacity decay trend parameters, which include the current effective capacity, capacity decay rate, and decay rate.

6. The method for assessing the health risks of batteries as described in claim 5, characterized in that: In step 4, the preset health level determination rule is a multi-parameter fusion determination rule, specifically setting an equivalent internal resistance threshold range, a capacity decay rate threshold range, and a charge-discharge cycle number threshold range. When all three parameters are within their corresponding normal threshold ranges, it is determined to be normal level one; when any parameter exceeds the normal threshold range but does not reach the critical threshold, it is determined to be warning level two; when any parameter reaches the critical threshold or at least two parameters exceed the normal threshold range, it is determined to be critical level three.

7. The method for assessing the health risks of batteries as described in claim 6, characterized in that: In step 5, generating the corresponding alarm event includes generating alarm level information matching the health level, abnormal battery parameter information, and current device operating status information; the alarm event is prompted through the local alarm module of the terminal device, and is also sent to the upper-level terminal management system or operation and maintenance system through the external interaction interface.

8. The method for assessing the health risks of batteries as described in claim 7, characterized in that: In step 5, the interaction process of the alarm event is as follows: after the battery health assessment module completes the risk assessment and alarm determination, it generates standardized alarm data and performs local persistent storage. The alarm request is sent to the upper-level alarm management module through the business alarm interface. The alarm management module records the alarm notification and returns an ACK confirmation signal. After receiving the confirmation signal, the battery health assessment module sends the complete alarm event to the alarm management module. The alarm management module records the alarm information and returns an ACK confirmation signal again to complete the alarm interaction.

9. The method for assessing the health risks of batteries as described in claim 8, characterized in that: When the terminal device is a removable multi-battery device, after the system starts up, it first performs a battery initialization scan to identify the inserted battery and check the battery's unique ID in the historical database. If it is an identified battery, it loads the previously saved battery health status and historical parameters and then executes steps 1-5. If it is a new battery, it first initializes the battery data structure and completes battery calibration, and then executes steps 1-5. When a battery is detected to be removed or a new battery is inserted, it switches and saves the corresponding battery's health status data.

10. The method for assessing the health risks of batteries as described in claim 9, characterized in that: The battery history database is a persistent database that stores information including each battery's unique ID, first insertion time, charge-discharge cycle count, historical equivalent internal resistance curve, capacity decay trend curve, health status history, and alarm event records. The data in the database supports reading and updating across device operating cycles.

Images