Direct-current battery discharge timely power-off method and device

Abstract

The invention discloses a direct-current battery discharge timely power-off method and device, and the method comprises the steps: monitoring key parameters of a battery in real time, and collecting the data of the key parameters; evaluating the discharge state of the battery according to the key parameter data, and judging whether an evaluation result reaches a preset discharge threshold value or not; if the evaluation result reaches a preset discharge threshold value, carrying out power-off operation on the battery; historical discharge data of the battery is collected, the model is used for training, the optimal power-off threshold value is predicted through the trained model according to real-time monitoring data, and the power-off threshold value to be executed is dynamically adjusted according to the optimal power-off threshold value. By monitoring key parameters such as the voltage, the current and the temperature of the battery in real time, the discharging state of the battery is intelligently evaluated, and the power-off operation is automatically executed when the preset threshold value is reached. In addition, a machine learning algorithm is adopted to continuously optimize a power-off threshold value, the intelligentization and self-adaptive control capability of the system are improved, the battery is effectively prevented from being over-discharged, and the service life of the battery is prolonged.

Description

Technical Field

[0001] The present application relates to the field of battery applications, and in particular to a method and device for timely power-off of a DC battery during discharge. Background Art

[0002] Currently, in the field of DC battery applications, battery overdischarge is a common problem, which can cause battery damage and shorten the life of the battery. In the existing technology, although there are some monitoring and power-off protection measures, they often lack intelligence and adaptive capabilities and cannot cope well with different usage environments and changes in battery status. For example, traditional monitoring systems cannot dynamically adjust the power-off threshold according to the actual discharge status of the battery, thereby failing to protect the battery to the maximum extent. Summary of the invention

[0003] The present application provides a method and device for timely power-off of a DC battery during discharge.

[0004] According to a first aspect of the present application, the present application provides a method for timely power-off of a DC battery during discharge, comprising: Real-time monitoring of key parameters of the battery and collection of key parameter data, wherein the key parameters include voltage, current and temperature; Evaluating the discharge state of the battery according to the key parameter data, and determining whether the evaluation result reaches a preset discharge threshold; If the evaluation result reaches a preset discharge threshold, the battery is powered off; Collect historical discharge data of the battery, use the model for training, use the trained model to predict the optimal power-off threshold using the real-time monitoring data, and dynamically adjust the power-off threshold to be executed based on the optimal power-off threshold.

[0005] In the above method, the step of evaluating the discharge state of the battery according to the key parameter data and determining whether the evaluation result reaches a preset threshold value includes: Synchronizing the timestamps of the key parameter data; Extracting a first key feature of the synchronized key parameter data, wherein the first key feature includes a discharge voltage, a discharge current and a temperature change rate; Analyze the key features using a trained battery discharge state evaluation model to evaluate the discharge state of the battery; The evaluation result is compared with a preset threshold to determine whether the evaluation result reaches the preset threshold.

[0006] In the above method, if the evaluation result reaches a preset threshold, the battery is powered off, including: If the evaluation result reaches a preset threshold, a power-off instruction is issued to stop discharging the battery; Verify whether the power-off status is successful. If the power-off is successful, record the operation log. If the power-off fails, activate the emergency plan.

[0007] The above method, wherein the historical discharge data of the battery is collected, the model is used for training, the optimal power-off threshold is predicted by the trained model using the real-time monitoring data, and the power-off threshold to be executed is dynamically adjusted according to the optimal power-off threshold, comprises: Collecting historical discharge data of the battery and preprocessing the historical discharge data, wherein the historical discharge data includes voltage, temperature and discharge duration; Extracting a second key feature of the preprocessed historical discharge data, wherein the second key feature includes a voltage change rate and a temperature change rate; The LSTM model is trained using the second key feature data, and the optimal power-off threshold is predicted by the trained LSTM model according to the key parameter data monitored in real time; The power-off threshold to be executed is dynamically adjusted according to the optimal power-off threshold.

[0008] In the above method, the preprocessing of the historical discharge data comprises: Cleaning the historical discharge data; normalizing the cleaned historical discharge data; Among them, the normalized formula is:

[0009] is the normalized value, is the original data value, and are the minimum and maximum values ​​in the data set, respectively.

[0010] According to a second aspect of the present application, the present application provides a DC battery discharge timely power-off device, comprising: A monitoring module is used to monitor key parameters of the battery in real time and collect key parameter data, wherein the key parameters include voltage, current and temperature; An evaluation module, used to evaluate the discharge state of the battery according to the key parameter data, and determine whether the evaluation result reaches a preset discharge threshold; A processing module, configured to cut off power to the battery when the evaluation result reaches a preset discharge threshold; The optimization module is used to collect historical discharge data of the battery, use the model for training, predict the optimal power-off threshold value through the trained model using the real-time monitoring data, and dynamically adjust the power-off threshold value to be executed according to the optimal power-off threshold value.

[0011] The above device, the evaluation module, comprises: A synchronization unit, used for synchronizing the timestamps of the key parameter data; A first extraction unit, used to extract a first key feature of the synchronized key parameter data, wherein the first key feature includes a discharge voltage, a discharge current and a temperature change rate; An evaluation unit, configured to analyze the key features using a trained battery discharge state evaluation model to evaluate the discharge state of the battery; The judging unit is used to compare the evaluation result with a preset threshold value to judge whether the evaluation result reaches the preset threshold value.

[0012] The above device, the processing module, includes: a processing unit, configured to issue a power-off instruction to stop discharging the battery when the evaluation result reaches a preset threshold; The verification unit is used to verify whether the power-off status is successful. If the power-off is successful, the operation log is recorded. If the power-off fails, the emergency plan is activated.

[0013] The above device, the optimization module, includes: A collecting unit, used to collect historical discharge data of the battery and pre-process the historical discharge data, wherein the historical discharge data includes voltage, temperature and discharge duration; A second extraction unit, used to extract a second key feature of the pre-processed historical discharge data, wherein the second key feature includes a voltage change rate and a temperature change rate; A prediction unit, used to train an LSTM model using the second key feature data, and predict an optimal power-off threshold through the trained LSTM model according to the key parameter data monitored in real time; The optimization unit is used to dynamically adjust the power-off threshold to be executed according to the optimal power-off threshold.

[0014] The above-mentioned device, the collection unit is also used to clean the historical discharge data; normalizing the cleaned historical discharge data; Among them, the normalized formula is:

[0015] is the normalized value, is the original data value, and are the minimum and maximum values ​​in the data set, respectively.

[0016] Due to the adoption of the above technical solution, the beneficial effects of this application are: The embodiment of the present application provides a method for timely power-off of DC battery discharge, including: real-time monitoring of key parameters of the battery, collecting key parameter data, key parameters include voltage, current and temperature; evaluating the discharge state of the battery according to the key parameter data, and judging whether the evaluation result reaches the preset discharge threshold; if the evaluation result reaches the preset discharge threshold, powering off the battery; collecting historical discharge data of the battery, using a model for training, using the real-time monitoring data to predict the optimal power-off threshold through the trained model, and dynamically adjusting the power-off threshold to be executed according to the optimal power-off threshold. By real-time monitoring of key parameters such as battery voltage, current, and temperature, the discharge state of the battery is intelligently evaluated, and the power-off operation is automatically executed when the preset threshold is reached. In addition, the system uses a machine learning algorithm to continuously optimize the power-off threshold, improve the intelligence and adaptive control capabilities of the system, thereby effectively preventing over-discharge of the battery and extending the battery life. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a method provided in an embodiment of the present application in one implementation manner; Figure 2 A flowchart of another implementation of the method provided in an embodiment of the present application; Figure 3 A flowchart of another embodiment of the method provided in the present application; Figure 4 A flowchart of a method provided in an embodiment of the present application in another implementation manner; Figure 5 A schematic diagram of a program module of an apparatus provided in an embodiment of the present application in one implementation manner; Figure 6 A schematic diagram of a program module of the device provided in an embodiment of the present application in another implementation manner. DETAILED DESCRIPTION

[0018] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present application better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present application are not shown or described in the specification, this is to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0019] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0020] The serial numbers assigned to the components in this article, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning.

[0021] In addition, the technical features and technical solutions described herein may be combined in any suitable manner in one or more embodiments. It is easy for a person skilled in the art to understand that the steps or operation sequence of the methods related to the embodiments provided herein may also be changed. Therefore, any sequence in the drawings and embodiments is only for illustrative purposes and does not imply a requirement to follow a certain sequence unless it is explicitly stated that a certain sequence is required.

[0022] like Figure 1 As shown, a method for timely powering off a DC battery during discharge provided in an embodiment of the present application, one implementation manner of which includes the following steps: Step 101: monitor key parameters of the battery in real time and collect key parameter data, where the key parameters include voltage, current and temperature.

[0023] The key parameter data can be collected through voltage sensors, current sensors and temperature sensors. The analog signals output by each sensor are sent to the signal conditioning circuit for necessary signal processing. The data acquisition card samples the processed signal at a set sampling rate and converts the analog signal into a digital signal. After the monitoring software receives the digital signal, it applies a digital filtering algorithm to process the data and displays the key parameter data in real time, such as voltage, current and temperature values.

[0024] Step 102: Evaluate the discharge state of the battery according to the key parameter data, and determine whether the evaluation result reaches a preset discharge threshold.

[0025] The discharge status assessment is to use monitoring data and data analysis algorithms to accurately assess the discharge status of the battery, thereby providing a reliable decision-making basis for powering off the battery, ensuring that protective measures are taken in time when the battery is close to an over-discharge state, thereby extending the battery life and ensuring safe use.

[0026] like Figure 2 As shown, step 102 may include the following steps: Step 1021: Synchronize the timestamps of key parameter data.

[0027] The timestamps of the received voltage data, current data, and temperature data are synchronized to ensure the accuracy of data analysis.

[0028] Step 1022: extracting the first key feature of the synchronized key parameter data, where the first key feature includes discharge voltage, discharge current and temperature change rate.

[0029] The first key feature is extracted from the synchronized data. The first key feature includes but is not limited to the discharge voltage, discharge current, temperature change rate, etc. of the battery, which can reflect the discharge state of the battery. Taking the discharge voltage as an example, the first key feature can be specifically expressed by the following formula:

[0030] in, represents the first key feature set, Indicates the discharge voltage, represents the discharge current, Indicates the rate of temperature change.

[0031] Step 1023: Use the trained battery discharge state evaluation model to analyze key features and evaluate the battery discharge state.

[0032] A machine learning algorithm, such as support vector machine (SVM), random forest (RF) or neural network, is used to train the extracted first key features and build a battery discharge state assessment model. The model can estimate the battery discharge state based on real-time data, and the discharge state may include healthy, warning and dangerous.

[0033] Step 1024: Compare the evaluation result with a preset threshold to determine whether the evaluation result reaches the preset threshold.

[0034] According to the specifications and experimental data provided by the battery manufacturer, the thresholds corresponding to different discharge states are set. For example, when the discharge voltage of the battery is lower than a certain value, the battery will be judged to be in a dangerous state.

[0035] Step 103: If the evaluation result reaches a preset discharge threshold, the battery is powered off.

[0036] When it is determined that the battery reaches or exceeds the preset discharge threshold, the power-off operation is automatically triggered to prevent the battery from being over-discharged and causing permanent damage.

[0037] like Figure 3 As shown, step 103 may include the following steps: Step 1031: If the evaluation result reaches a preset threshold, a power-off instruction is issued to stop discharging the battery.

[0038] The estimation result of the battery discharge state evaluation model is continuously received, and the estimation result is compared with a preset discharge threshold value.

[0039]

[0040] in, is the real-time discharge status of the battery. is the preset discharge threshold.

[0041] Step 1032: Verify whether the power-off status is successful. If the power-off is successful, record the operation log. If the power-off fails, activate the emergency plan.

[0042] After the power-off operation is executed, the embodiment of the present application verifies whether the power-off state is successfully implemented through the feedback circuit to ensure battery safety. If the power-off is successful, the operation log is recorded; if the power-off fails, the emergency plan is activated, such as repeated power-off attempts or alarm prompts.

[0043] Step 104: Collect historical discharge data of the battery, use the model for training, use the trained model to predict the optimal power-off threshold using the real-time monitoring data, and dynamically adjust the power-off threshold to be executed according to the optimal power-off threshold.

[0044] The embodiment of the present application adopts a machine learning algorithm, based on the battery's usage history data and real-time monitoring data, to dynamically adjust and optimize the power-off threshold to improve the adaptive ability and intelligence level, and ensure that the power-off threshold can be flexibly adjusted according to the actual state of the battery and changes in the external environment, thereby ensuring the battery safety while maximizing the battery life. Through the above implementation, it is possible to ensure that the battery's power-off threshold is always in the best state, which not only protects the battery, but also improves the intelligence level of the system.

[0045] like Figure 4 As shown, step 104 may include the following steps: Step 1041: Collect historical discharge data of the battery and pre-process the historical discharge data, the historical discharge data including voltage, temperature and discharge duration. In one implementation, the historical discharge data may also include current.

[0046] In the embodiment of the present application, preprocessing the historical discharge data may specifically include: Clean the historical discharge data, such as removing outliers; Normalize the historical discharge data after cleaning; Among them, the normalized formula is:

[0047] is the normalized value, is the original data value, and are the minimum and maximum values ​​in the data set, respectively.

[0048] Step 1042: extract the second key features of the pre-processed historical discharge data, the second key features including the voltage change rate and the temperature change rate, for training the machine learning model. The extracted second key features may include:

[0049] in, is the rate of change of voltage, is the rate of temperature change, and are the changes in voltage and temperature, respectively. is the time interval.

[0050] Step 1043: Use the second key feature data to train an LSTM (Long Short-Term Memory) model, and predict the optimal power-off threshold through the trained LSTM model according to the key parameter data monitored in real time.

[0051] The extracted second key feature data is used to train the machine learning model. This application adopts the LSTM algorithm because it is good at processing time series data. The LSTM model learns the relationship between the battery discharge state and the power-off threshold by adjusting the network weights.

[0052] Step 1044: Dynamically adjust the power-off threshold to be executed according to the optimal power-off threshold.

[0053] The power-off threshold is adjusted dynamically based on the prediction results and the real-time status of the battery. The model is continuously optimized through feedback from actual applications to improve prediction accuracy and system adaptability.

[0054] like Figure 5 As shown, a DC battery discharge timely power-off device provided in an embodiment of the present application, one implementation manner thereof, includes a monitoring module 510 , an evaluation module 520 , a processing module 530 and an optimization module 540 .

[0055] The monitoring module 510 is used to monitor the key parameters of the battery in real time and collect key parameter data. The key parameters include voltage, current and temperature.

[0056] The key parameter data can be collected through voltage sensors, current sensors and temperature sensors. The analog signals output by each sensor are sent to the signal conditioning circuit for necessary signal processing. The data acquisition card samples the processed signal at a set sampling rate and converts the analog signal into a digital signal. After the monitoring software receives the digital signal, it applies a digital filtering algorithm to process the data and displays the key parameter data in real time, such as voltage, current and temperature values.

[0057] The evaluation module 520 is used to evaluate the discharge state of the battery according to the key parameter data, and determine whether the evaluation result reaches a preset discharge threshold.

[0058] The discharge status assessment is to use monitoring data and data analysis algorithms to accurately assess the discharge status of the battery, thereby providing a reliable decision-making basis for powering off the battery, ensuring that protective measures are taken in time when the battery is close to an over-discharge state, thereby extending the battery life and ensuring safe use.

[0059] The processing module 530 is used to cut off the power supply to the battery when the evaluation result reaches a preset discharge threshold.

[0060] When it is determined that the battery reaches or exceeds the preset discharge threshold, the power-off operation is automatically triggered to prevent the battery from being over-discharged and causing permanent damage.

[0061] The optimization module 540 is used to collect historical discharge data of the battery, use the model for training, predict the optimal power-off threshold value through the trained model using the real-time monitoring data, and dynamically adjust the power-off threshold value to be executed according to the optimal power-off threshold value.

[0062] The embodiment of the present application adopts a machine learning algorithm, based on the battery's usage history data and real-time monitoring data, to dynamically adjust and optimize the power-off threshold to improve the adaptive ability and intelligence level, and ensure that the power-off threshold can be flexibly adjusted according to the actual state of the battery and changes in the external environment, thereby ensuring the battery safety while maximizing the battery life. Through the above implementation, it is possible to ensure that the battery's power-off threshold is always in the best state, which not only protects the battery, but also improves the intelligence level of the system.

[0063] like Figure 6 As shown, another implementation manner of the DC battery discharge timely power-off device provided in the embodiment of the present application includes a monitoring module 610, an evaluation module 620, a processing module 630 and an optimization module 640.

[0064] The monitoring module 610 is used to monitor the key parameters of the battery in real time and collect key parameter data. The key parameters include voltage, current and temperature.

[0065] The evaluation module 620 is used to evaluate the discharge state of the battery according to the key parameter data, and determine whether the evaluation result reaches a preset discharge threshold.

[0066] In one implementation, the evaluation module 620 may include a synchronization unit 621 , a first extraction unit 622 , an evaluation unit 623 , and a judgment unit 624 .

[0067] The synchronization unit 621 is used to synchronize the timestamps of the key parameter data.

[0068] The timestamps of the received voltage data, current data, and temperature data are synchronized to ensure the accuracy of data analysis.

[0069] The first extraction unit 622 is used to extract the first key feature of the synchronized key parameter data, where the first key feature includes discharge voltage, discharge current and temperature change rate.

[0070] The first key feature is extracted from the synchronized data. The first key feature includes but is not limited to the discharge voltage, discharge current, temperature change rate, etc. of the battery, which can reflect the discharge state of the battery. Taking the discharge voltage as an example, the first key feature can be specifically expressed by the following formula:

[0071] in, represents the first key feature set, Indicates the discharge voltage, represents the discharge current, Indicates the rate of temperature change.

[0072] The evaluation unit 623 is used to analyze the key features using the trained battery discharge state evaluation model to evaluate the discharge state of the battery.

[0073] A machine learning algorithm, such as support vector machine (SVM), random forest (RF) or neural network, is used to train the extracted first key features and build a battery discharge state assessment model. The model can estimate the battery discharge state based on real-time data, and the discharge state may include healthy, warning and dangerous.

[0074] The judging unit 624 is used to compare the evaluation result with a preset threshold value to judge whether the evaluation result reaches the preset threshold value.

[0075] According to the specifications and experimental data provided by the battery manufacturer, the thresholds corresponding to different discharge states are set. For example, when the discharge voltage of the battery is lower than a certain value, the battery will be judged to be in a dangerous state.

[0076] The processing module 630 is used to cut off the power supply to the battery when the evaluation result reaches a preset discharge threshold.

[0077] In one implementation, the processing module 630 provided in the embodiment of the present application may include a processing unit 631 and a verification unit 632 .

[0078] The processing unit 631 is used to issue a power-off instruction to stop discharging the battery when the evaluation result reaches a preset threshold.

[0079] The estimation result of the battery discharge state evaluation model is continuously received, and the estimation result is compared with a preset discharge threshold value.

[0080]

[0081] in, is the real-time discharge status of the battery. is the preset discharge threshold.

[0082] The verification unit 632 is used to verify whether the power-off state is successful. If the power-off is successful, the operation log is recorded. If the power-off fails, the emergency plan is activated.

[0083] After the power-off operation is executed, the embodiment of the present application verifies whether the power-off state is successfully implemented through the feedback circuit to ensure battery safety. If the power-off is successful, the operation log is recorded; if the power-off fails, the emergency plan is activated, such as repeated power-off attempts or alarm prompts.

[0084] The optimization module 640 is used to collect historical discharge data of the battery, use the model for training, predict the optimal power-off threshold value through the trained model using the real-time monitoring data, and dynamically adjust the power-off threshold value to be executed according to the optimal power-off threshold value.

[0085] In one implementation, the optimization module 640 may include a collection unit 641 , a second extraction unit 642 , a prediction unit 643 , and an optimization unit 644 .

[0086] The collecting unit 641 is used to collect the historical discharge data of the battery and pre-process the historical discharge data, where the historical discharge data includes voltage, temperature and discharge duration.

[0087] In one implementation, the collection unit 641 may also be used to clean the historical discharge data and perform normalization processing on the cleaned historical discharge data.

[0088] Among them, the normalized formula is:

[0089] is the normalized value, is the original data value, and are the minimum and maximum values ​​in the data set, respectively.

[0090] The second extraction unit 642 is used to extract the second key features of the pre-processed historical discharge data, the second key features including the voltage change rate and the temperature change rate, which are used to train the machine learning model. The extracted second key features may include:

[0091] in, is the rate of change of voltage, is the rate of temperature change, and are the changes in voltage and temperature, respectively. is the time interval.

[0092] The prediction unit 643 is used to train the LSTM model using the second key feature data, and predict the optimal power-off threshold through the trained LSTM model according to the key parameter data monitored in real time.

[0093] The extracted second key feature data is used to train the machine learning model. This application adopts the LSTM algorithm because it is good at processing time series data. The LSTM model learns the relationship between the battery discharge state and the power-off threshold by adjusting the network weights.

[0094] The optimization unit 644 is used to dynamically adjust the power-off threshold to be executed according to the optimal power-off threshold.

[0095] The power-off threshold is adjusted dynamically based on the prediction results and the real-time status of the battery. The model is continuously optimized through feedback from actual applications to improve prediction accuracy and system adaptability.

[0096] Those skilled in the art will appreciate that all or part of the steps of the various methods in the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which may include: a read-only memory, a random access memory, a disk or an optical disk, etc.

[0097] The above contents are further detailed descriptions of the present application in combination with specific implementation methods, and it cannot be determined that the specific implementation of the present application is limited to these descriptions. For ordinary technicians in the technical field to which the present application belongs, several simple deductions or substitutions can be made without departing from the concept of the present application.

Claims

1. A method for timely power-off of DC battery discharge, characterized in that: include: Real-time monitoring of key parameters of the battery and collection of key parameter data, wherein the key parameters include voltage, current and temperature; Evaluating the discharge state of the battery according to the key parameter data, and determining whether the evaluation result reaches a preset discharge threshold; If the evaluation result reaches a preset discharge threshold, the battery is powered off; Collect historical discharge data of the battery, use the model for training, use the trained model to predict the optimal power-off threshold using the real-time monitoring data, and dynamically adjust the power-off threshold to be executed based on the optimal power-off threshold.

2. The method according to claim 1, characterized in that The step of evaluating the discharge state of the battery according to the key parameter data and determining whether the evaluation result reaches a preset threshold value includes: Synchronizing the timestamps of the key parameter data; Extracting a first key feature of the synchronized key parameter data, wherein the first key feature includes a discharge voltage, a discharge current and a temperature change rate; Analyze the key features using a trained battery discharge state evaluation model to evaluate the discharge state of the battery; The evaluation result is compared with a preset threshold to determine whether the evaluation result reaches the preset threshold.

3. The method according to claim 1, characterized in that If the evaluation result reaches a preset threshold, the battery is powered off, including: If the evaluation result reaches a preset threshold, a power-off instruction is issued to stop discharging the battery; Verify whether the power-off status is successful. If the power-off is successful, record the operation log. If the power-off fails, activate the emergency plan.

4. The method according to claim 1, characterized in that The collecting of historical discharge data of the battery, using the model for training, using the trained model to predict the optimal power-off threshold of the real-time monitoring data, and dynamically adjusting the power-off threshold to be executed according to the optimal power-off threshold, includes: Collecting historical discharge data of the battery and preprocessing the historical discharge data, wherein the historical discharge data includes voltage, temperature and discharge duration; Extracting a second key feature of the preprocessed historical discharge data, wherein the second key feature includes a voltage change rate and a temperature change rate; The LSTM model is trained using the second key feature data, and the optimal power-off threshold is predicted by the trained LSTM model according to the key parameter data monitored in real time; The power-off threshold to be executed is dynamically adjusted according to the optimal power-off threshold.

5. The method according to claim 4, characterized in that The preprocessing of the historical discharge data includes: Cleaning the historical discharge data; normalizing the cleaned historical discharge data; Among them, the normalized formula is: is the normalized value, is the original data value, and are the minimum and maximum values ​​in the data set, respectively.

6. A DC battery discharge timely power-off device, characterized in that: include: A monitoring module is used to monitor key parameters of the battery in real time and collect key parameter data, wherein the key parameters include voltage, current and temperature; An evaluation module, used to evaluate the discharge state of the battery according to the key parameter data, and determine whether the evaluation result reaches a preset discharge threshold; A processing module, configured to cut off power to the battery when the evaluation result reaches a preset discharge threshold; The optimization module is used to collect historical discharge data of the battery, use the model for training, predict the optimal power-off threshold value through the trained model using the real-time monitoring data, and dynamically adjust the power-off threshold value to be executed according to the optimal power-off threshold value.

7. The device according to claim 6, characterized in that The evaluation module comprises: A synchronization unit, used for synchronizing the timestamps of the key parameter data; A first extraction unit, used to extract a first key feature of the synchronized key parameter data, wherein the first key feature includes a discharge voltage, a discharge current and a temperature change rate; An evaluation unit, configured to analyze the key features using a trained battery discharge state evaluation model to evaluate the discharge state of the battery; The judging unit is used to compare the evaluation result with a preset threshold value to judge whether the evaluation result reaches the preset threshold value.

8. The device according to claim 6, characterized in that The processing module comprises: a processing unit, configured to issue a power-off instruction to stop discharging the battery when the evaluation result reaches a preset threshold; The verification unit is used to verify whether the power-off status is successful. If the power-off is successful, the operation log is recorded. If the power-off fails, the emergency plan is activated.

9. The device according to claim 6, characterized in that The optimization module comprises: A collecting unit, used to collect historical discharge data of the battery and pre-process the historical discharge data, wherein the historical discharge data includes voltage, temperature and discharge duration; A second extraction unit, used to extract a second key feature of the pre-processed historical discharge data, wherein the second key feature includes a voltage change rate and a temperature change rate; A prediction unit, used to train an LSTM model using the second key feature data, and predict an optimal power-off threshold through the trained LSTM model according to the key parameter data monitored in real time; The optimization unit is used to dynamically adjust the power-off threshold to be executed according to the optimal power-off threshold.

10. The device according to claim 9, characterized in that The collection unit is further used to clean the historical discharge data; normalizing the cleaned historical discharge data; Among them, the normalized formula is: is the normalized value, is the original data value, and are the minimum and maximum values ​​in the data set, respectively.

Images