Battery replacing cabinet system applied to electric bicycle

Abstract

The invention belongs to the technical field of battery replacement, and provides a battery replacement cabinet system applied to an electric bicycle, which comprises the following steps: analyzing the fluctuation condition of temperature data and establishing a dynamic threshold compensation mechanism by using the temperature data collected by a temperature sensing monitoring unit and a composite monitoring unit; the method comprises the following steps: performing time sequence construction on temperature data collected by a temperature sensing monitoring unit and a composite monitoring unit, analyzing the time sequence to obtain an intrinsic mode function component, converting the intrinsic mode function component to obtain a Hilbert spectrum, distinguishing a low-frequency intrinsic mode function component and a high-frequency intrinsic mode function component according to the Hilbert spectrum, and calculating the intrinsic mode function component according to the low-frequency intrinsic mode function component and the high-frequency intrinsic mode function component. Integrating the low-frequency intrinsic mode function and the high-frequency intrinsic mode function to obtain a temperature change time domain envelope line; analyzing the slope of the smooth rising stage of the temperature change time domain envelope, judging whether the charging temperature change of the battery is normal or not in real time, and if not, generating a pre-alarm signal; and based on the generated pre-alarm signal, analyzing a Pearson's correlation coefficient between the monitoring units, and positioning and adjusting a temperature anomaly region.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery replacement, and specifically relates to a battery replacement cabinet system for electric bicycles. Background Art

[0002] In the field of electric bicycle battery replacement, the widespread use of lithium batteries has greatly improved the performance and convenience of vehicles, but it has also brought a series of safety risks, making lithium battery safety technology a key focus of research and application;

[0003] Currently, lithium battery security technology has made some progress in temperature monitoring and fire warning. However, there are still significant technical gaps and deficiencies in the key link of analyzing temperature fluctuations when the battery swap cabinet door is opened and closed.

[0004] Existing lithium battery security technologies primarily focus on monitoring battery temperature during normal conditions, such as charging and storage. For example, some systems install temperature sensors at key locations in the battery pack to collect real-time temperature data and trigger alarms based on preset temperature thresholds. While this type of technology can detect abnormal battery temperature rises to a certain extent, it overlooks the impact of common operations such as opening and closing the battery swap cabinet door on temperature monitoring.

[0005] During daily operation, battery swap cabinets frequently open and close their doors. Each time the door is opened, cold or hot air from the outside environment rushes in, causing the temperature inside the cabinet to fluctuate dramatically. When the door is closed, the temperature inside the cabinet gradually returns to normal. Existing security systems do not conduct in-depth analysis of this process, making it impossible to effectively distinguish between normal temperature fluctuations caused by door opening and closing and temperature changes caused by abnormal battery temperature rise when processing temperature monitoring data. This can easily lead to false alarms, disrupting the normal work of operation and maintenance personnel and increasing unnecessary manpower and material costs.

[0006] To this end, the present invention provides a battery-exchanging cabinet system for an electric bicycle. Summary of the Invention

[0007] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0008] The technical solution adopted by the present invention to solve its technical problem is:

[0009] The present invention provides a battery-swapping cabinet system for electric bicycles, comprising:

[0010] Temperature fluctuation monitoring module: Utilizes the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit to analyze temperature data fluctuations and establish a dynamic threshold compensation mechanism;

[0011] Envelope construction module: constructs a time series of the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit, analyzes the time series to obtain the intrinsic mode function components, converts the intrinsic mode function components to obtain the Hilbert spectrum, distinguishes the low-frequency intrinsic mode function components from the high-frequency intrinsic mode function components based on the Hilbert spectrum, and integrates the low-frequency intrinsic mode function and the high-frequency intrinsic mode function to obtain the temperature change time domain envelope;

[0012] Abnormal temperature judgment module: Analyzes the slope of the temperature change time domain envelope in the smooth rising phase, and judges in real time whether the battery charging temperature change is normal. If it is abnormal, it generates a pre-alarm signal;

[0013] Abnormal temperature positioning module: Based on the generation of pre-alarm signals, the Pearson correlation coefficient between monitoring units is analyzed to locate the abnormal temperature area.

[0014] As a further improvement of the present invention: the process of establishing the dynamic threshold compensation mechanism is as follows:

[0015] Calculate the temperature gradient in real time based on any temperature sensing monitoring unit or composite monitoring unit, i.e., the temperature change per unit time; compare the calculated temperature gradient with the temperature gradient threshold;

[0016] If the temperature gradient is greater than or equal to the temperature gradient threshold, the battery swap cabinet system determines it as a door opening disturbance, activates the buffer period, and suppresses the normal alarm logic of the battery swap cabinet system;

[0017] During the buffer period, the battery swap cabinet system will temporarily increase the alarm threshold dynamically according to the preset temperature value.

[0018] As a further improvement of the present invention: the specific process of constructing the time series is as follows:

[0019] For the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit, a time series is constructed with time as the horizontal axis.

[0020] As a further improvement of the present invention, the specific process of obtaining the intrinsic mode function components is as follows:

[0021] Obtain the original temperature time series, find all the maximum and minimum points in the original temperature time series, use the cubic spline interpolation method to fit the maximum and minimum points respectively to obtain the upper envelope and lower envelope; calculate the mean of the upper and lower envelopes to obtain the mean envelope; subtract the mean envelope from the original temperature time series to calculate the residual component;

[0022] Analyze the calculated residual components and the conditions of the eigenmode functions;

[0023] If the residual component does not meet the conditions of the intrinsic mode function, the residual component is regarded as a new temperature time series, and the process of determining the extreme points, constructing the upper and lower envelopes, calculating the mean envelope and subtracting them is repeated until the residual component meets the conditions of the intrinsic mode function and the intrinsic mode function component is obtained.

[0024] As a further improvement of the present invention: the specific process of obtaining the Hilbert spectrum is:

[0025] Among the multiple intrinsic mode function components obtained by decomposition, the components closely related to the temperature change trend are selected for Hilbert transform, and the time domain signal is converted to the frequency domain to obtain the corresponding Hilbert spectrum.

[0026] As a further improvement of the present invention: the specific process of obtaining the temperature-varying time-domain envelope is as follows:

[0027] After Hilbert-Huang transform decomposition, the intrinsic mode function components and residual components are obtained;

[0028] The residual component, low-frequency intrinsic mode function component and medium-frequency intrinsic mode function component after decomposition are selected and superimposed to obtain the temperature-varying time domain envelope.

[0029] As a further improvement of the present invention, the specific process of determining whether the battery charging temperature change is normal is as follows:

[0030] The smooth rising phase of the temperature change time domain envelope is selected as the analysis object, and the temperature change during battery charging when the battery swap cabinet door is closed is analyzed in real time to determine whether it is normal.

[0031] If the slope of the real-time smooth rising stage is greater than the slope of the normal smooth rising stage, it means that the battery has abnormal temperature change during the charging process, and a pre-alarm signal is generated;

[0032] If the slope of the real-time smooth rising stage is less than or equal to the slope of the normal smooth rising stage, it means that the temperature change of the battery during charging is normal, and the thermal runaway phenomenon of the battery swap cabinet system is monitored in real time according to the normal alarm logic.

[0033] As a further improvement of the present invention: the Pearson correlation coefficient between the analysis and monitoring units is specifically:

[0034] For normal heat convection in the power exchange cabinet, the temperature changes between the composite monitoring units and the temperature sensing monitoring units usually have a strong correlation, and the Pearson correlation coefficient is high;

[0035] When an abnormal temperature rise occurs in a certain area of ​​the battery swap cabinet, the correlation between the temperature data of the composite monitoring unit or temperature sensing monitoring unit in that area and other composite monitoring units or temperature sensing monitoring units will be significantly reduced, and the Pearson correlation coefficient will become smaller.

[0036] As a further improvement of the present invention, the specific process of locating the abnormal temperature area is as follows:

[0037] When an abnormal temperature rise occurs in a certain area of ​​the battery swap cabinet, the temperature data of the composite monitoring unit or temperature sensing monitoring unit in this area will be significantly different from the temperature data of other composite monitoring units or temperature sensing monitoring units. By analyzing the Pearson correlation coefficient matrix, the temperature abnormality area can be quickly located.

[0038] As a further improvement of the present invention, the following modules are also included:

[0039] Adjustment control module: Based on the temperature abnormality area, identify whether the temperature abnormality area is within the temperature fluctuation radiation range when the battery swap cabinet is closed. If so, the battery swap cabinet system automatically adjusts the speed of closing the battery swap cabinet to avoid false alarms.

[0040] The beneficial effects of the present invention are as follows:

[0041] 1. The temperature fluctuation detection module collects temperature data from the edge and core areas of the battery swap cabinet in real time when the door is opened, calculates the temperature gradient and compares it with the threshold; when it is determined to be a door opening disturbance, the buffer period is activated to dynamically increase the alarm threshold, effectively avoiding false alarms caused by sudden temperature changes when the door is opened and closed; this enables the alarm system to more accurately reflect the actual temperature anomalies of the battery, reduce unnecessary alarm interference, improve operation and maintenance efficiency, and reduce manpower and material costs. The envelope construction module uses technologies such as the Hilbert-Huang transform to construct the temperature data into a time series, decompose the intrinsic mode function components, obtain the Hilbert spectrum, and then construct the temperature change time domain envelope; this process can intuitively display the overall temperature trend and fluctuation range during the battery charging process, providing a clear and accurate data basis for subsequent abnormality judgment, making the monitoring of battery temperature changes more accurate.

[0042] 2. The abnormality judgment module selects the smooth rising phase of the temperature change time domain envelope and compares the slope of the real-time envelope in this phase with the historical normal data. If the real-time slope is greater than the normal slope, a pre-alarm signal can be generated in a timely manner to indicate possible battery charging temperature anomalies, buying time to pre-emptively address potential safety hazards and reducing the risk of accidents such as battery thermal runaway. The abnormal temperature positioning module, based on the pre-alarm signal, analyzes the Pearson correlation coefficient between each monitoring unit and utilizes the characteristics of high correlation during normal thermal convection and reduced correlation during abnormal temperature rise to quickly and accurately locate the temperature anomaly area. At the same time, the module can also eliminate false alarms caused by single sensor failures or short-term external interference, further improving the accuracy of abnormality judgment and helping operation and maintenance personnel to take quick measures to prevent the expansion of the fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below with reference to the accompanying drawings.

[0044] Figure 1 This is a system module diagram of an electric bicycle battery swapping cabinet system applied in the present invention;

[0045] Figure 2 This is a flowchart of the steps of a method for using an electric bicycle battery swap cabinet in the present invention;

[0046] Figure 3 is a connection diagram of Example 1 of the present invention;

[0047] Figure 4 This is a schematic diagram of the normal alarm logic of Example 2 of the present invention. DETAILED DESCRIPTION

[0048] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0049] Example 1

[0050] The electric bicycle battery swapping cabinet system described in Example 1 of the present invention includes:

[0051] Hardware layout module: layout of composite monitoring unit, fire host, fire pipeline, main control unit, sound and light alarm;

[0052] The connection relationship topology diagram of the composite monitoring unit, fire host, main control unit, and sound and light alarm is as follows: Figure 3 As shown;

[0053] The composite monitoring unit is equipped with a corresponding number of composite monitoring units according to the number of single batteries in the battery swap cabinet; the composite monitoring unit uploads the real-time monitored data to the main control unit;

[0054] The composite monitoring unit is arranged directly above the single battery and can accurately and quickly determine the fire signal of the battery swap cabinet. The composite monitoring unit is linked with the fire host to deliver the fire extinguishing agent to the surface of the single battery through the fire pipeline, effectively suppressing the heat diffusion of the single battery inside the battery swap cabinet.

[0055] The composite monitoring unit includes: CO concentration monitoring, VOC concentration monitoring and temperature monitoring, which can monitor in real time the temperature and smoke changes, characteristic gases, and early signs of thermal runaway caused by internal short circuit, overcharging and external short circuit in each single cell; when the fire information reaches the corresponding alarm level, the sound and light alarm connected to the main control unit will issue an alarm message of the corresponding level;

[0056] The alarm levels include: level one alarm, level two alarm, and level three alarm;

[0057] When a level 1 alarm occurs, the main control unit uploads the alarm information to the BMS of the battery swap cabinet, and the sound and light alarm does not start;

[0058] When a secondary alarm occurs, the main control unit outputs a warning dry contact, and the sound and light alarm is activated;

[0059] When a level 3 alarm occurs, the fire host outputs a sprinkler dry contact. At this time, the power exchange cabinet system enters the automatic sprinkler mode and starts the fire host to spray fire extinguishing agent to extinguish the fire.

[0060] A special new type of heat insulation material is arranged between the single cells, which can further isolate the spread of fire between the single cells;

[0061] The fire extinguishing agent is installed in the fire host, and the fire extinguishing agent includes but is not limited to: heptafluoropropane, perfluorohexanone, and water; the fire suppressant is pressurized by nitrogen in the fire host to achieve a spraying effect;

[0062] The fire protection pipeline is connected to the fire protection host and the atomizing nozzle;

[0063] The atomizing nozzle is arranged directly above the single battery;

[0064] The main control unit is connected to the fire host, the composite monitoring unit and the BMS of the battery exchange cabinet via a CAN bus;

[0065] The technical solution of the embodiment of the present invention is as follows: a corresponding number of composite monitoring units are equipped according to the number of single batteries in the electric bicycle battery exchange cabinet. The composite monitoring units are arranged directly above the single batteries and upload real-time monitoring data to the main control unit. The composite monitoring units can monitor in real time the early thermal runaway characteristics such as temperature, smoke changes, characteristic gases, electrolyte leakage, etc. caused by internal short circuit, overcharging, and external short circuit in the single batteries. When the third-level alarm is triggered, the fire host outputs a spray dry contact to the outside, the battery exchange cabinet system enters the automatic spray mode, and the fire host is activated to spray the fire extinguishing agent to extinguish the fire. Special new thermal insulation materials are arranged between the single batteries to prevent the spread of fire.

[0066] Example 2

[0067] like Figure 1 As shown, based on Example 1, an electric bicycle battery swapping cabinet system according to an embodiment of the present invention includes the following modules:

[0068] Temperature fluctuation monitoring module: Utilizes the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit to analyze temperature data fluctuations and establish a dynamic threshold compensation mechanism;

[0069] A temperature monitoring unit is placed on the door of the battery swap cabinet to collect real-time temperature data changes in the edge area when the door is opened;

[0070] A composite monitoring unit is arranged in the single battery compartment area of ​​the battery swap cabinet to collect real-time temperature data changes in the core area when the battery swap cabinet door is opened;

[0071] Real-time analysis of dual temperature data from the single battery compartment area and the edge area of ​​the battery swap cabinet when the cabinet door is open. Based on the analysis results, a dynamic threshold compensation mechanism is established to eliminate false alarms caused by large temperature fluctuations when the cabinet door is opened.

[0072] Specifically:

[0073] Based on any temperature sensing monitoring unit or composite monitoring unit, the temperature gradient G, i.e. the temperature change per unit time, is calculated in real time; by setting a second-level sliding window, the temperature data within the sliding window is processed;

[0074] For example, assuming that the temperature changes from T1 to T2 within 3 seconds, the temperature gradient ;

[0075] Calculate the temperature gradient G each time the battery swap cabinet door is opened; compare the calculated temperature gradient G with the temperature gradient threshold;

[0076] If the temperature gradient G is greater than or equal to the temperature gradient threshold, the battery swap cabinet system determines it as a door opening disturbance, activates the buffer period, and suppresses the normal alarm logic of the battery swap cabinet system;

[0077] It should be noted that during the buffer period, the battery swap cabinet system will temporarily increase the alarm threshold according to the preset temperature value to avoid false alarms caused by sudden temperature changes when the battery swap cabinet door is opened;

[0078] For example, the alarm threshold is increased by 5°C on the basis of the original threshold during the buffer period before the alarm is triggered;

[0079] During the buffer period, the battery swap cabinet system continuously monitors the temperature data of the core area and the edge area;

[0080] If the temperature has returned to the normal temperature gradient fluctuation range at the end of the buffer period, the buffer period will be exited and the normal alarm logic will be restored;

[0081] If the temperature continues to be higher than the dynamic alarm threshold after the buffer period, a temperature signal will be generated, prompting staff to inspect the battery swap cabinet to check whether there is a safety hazard of battery overheating;

[0082] like Figure 4As shown, the normal alarm logic includes: level one alarm, level two alarm, level three alarm and the alarm conditions corresponding to the level one alarm, level two alarm and level three alarm;

[0083] Envelope construction module: constructs a time series of the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit, analyzes the time series to obtain the intrinsic mode function components, converts the intrinsic mode function components to obtain the Hilbert spectrum, distinguishes the low-frequency intrinsic mode function components from the high-frequency intrinsic mode function components based on the Hilbert spectrum, and integrates the low-frequency intrinsic mode function and the high-frequency intrinsic mode function to obtain the temperature change time domain envelope;

[0084] Specifically:

[0085] For the temperature data collected by the temperature monitoring unit and the composite monitoring unit, a time series is constructed with time as the horizontal axis;

[0086] First, the temperature time series is processed using the Hilbert-Huang transform to decompose the intrinsic mode function (IMF);

[0087] By analyzing the intrinsic mode function, the time domain envelope of temperature change is extracted, and the temperature change time domain envelope is obtained through superposition processing. The temperature change time domain envelope can intuitively show the overall trend and fluctuation range of temperature change over time;

[0088] Specifically:

[0089] Obtain the original temperature time series. First, find all the maximum and minimum points in the original temperature time series. Use the cubic spline interpolation method to fit the maximum and minimum points respectively to obtain the upper and lower envelopes.

[0090] It should be noted that the upper envelope connects all the maximum points, and the lower envelope connects all the minimum points, ensuring that the temperature time series is always between the upper and lower envelopes;

[0091] Calculate the mean of the upper envelope and the lower envelope to obtain the mean envelope;

[0092] The residual component is calculated by subtracting the mean envelope from the original temperature time series;

[0093] Analyze the calculated residual components and the conditions of the eigenmode functions;

[0094] If the residual component does not meet the conditions of the intrinsic mode function, the residual component is regarded as a new temperature time series, and the process of determining the extreme points, constructing the upper and lower envelopes, calculating the mean envelope and subtracting them is repeated until the residual component meets the conditions of the intrinsic mode function and the intrinsic mode function component is obtained;

[0095] Through multiple iterations, the original temperature time series is decomposed into multiple intrinsic mode function components and a residual component; each intrinsic mode function component represents the fluctuation characteristics of different scales in the original signal, and the residual component reflects the overall trend of the signal;

[0096] It should be noted that the conditions of the intrinsic mode function are specifically: that is, the residual component still has the number of zero-crossing points and the number of extreme points that are not 0 or 1, and the local mean is 0;

[0097] The intrinsic mode function components include five intrinsic mode function components: intrinsic mode function component 1 (IMF1), intrinsic mode function component 2 (IMF2), intrinsic mode function component 3 (IMF3), intrinsic mode function component 4 (IMF4), and intrinsic mode function component 5 (IMF5);

[0098] Among the multiple intrinsic mode function components obtained by decomposition, the components closely related to the temperature change trend are selected for Hilbert transform, and the time domain signal is converted to the frequency domain to obtain the corresponding Hilbert spectrum;

[0099] The Hilbert spectrum can clearly show the frequency components and amplitude changes of each intrinsic mode function component at different times, providing a basis for further analysis of temperature change characteristics;

[0100] Specifically:

[0101] For example, the low-frequency intrinsic mode function component can reflect the overall trend of temperature change, while the high-frequency intrinsic mode function component reflects the short-term fluctuation and noise of temperature;

[0102] By superimposing the low-frequency intrinsic mode function components and the high-frequency intrinsic mode function components, the temperature-varying time-domain envelope is constructed;

[0103] The temperature change time domain envelope is constructed as follows:

[0104] For example, taking the battery charging process when the battery swap cabinet is closed as an example, it is assumed that five intrinsic mode function components and one residual component are obtained through Hilbert-Huang transform decomposition;

[0105] Among them, IMF1-IMF2 are high-frequency components, mainly reflecting environmental noise and small temperature fluctuations; IMF3-IMF4 are medium-frequency components, reflecting normal temperature fluctuations during battery charging;

[0106] IMF5 and residual components are low-frequency components, reflecting the overall trend of temperature change;

[0107] The temperature variation time domain envelope is obtained by superimposing IMF3-IMF5 and the residual component;

[0108] The temperature change time domain envelope can clearly show the process in which the battery temperature gradually rises from the initial value during charging, reaches a stable charging temperature and remains there for a period of time, and finally slowly decreases after charging is completed;

[0109] Abnormal judgment module: By analyzing the slope of the smooth rising phase of the temperature change time domain envelope, it can judge in real time whether the battery charging temperature change when the battery swap cabinet is closed is normal;

[0110] By analyzing the slope characteristics of the temperature change time domain envelope and comparing it with the historical normal temperature change time domain envelope, we can determine whether the temperature change during battery charging when the battery swap cabinet is closed conforms to normal rules;

[0111] It should be noted that, under normal circumstances, the time domain envelope of the battery charging process has a specific shape, such as a smooth rising phase, a stable platform phase, and a slow falling phase;

[0112] In the present invention, the smooth rising phase of the temperature change time domain envelope is selected as the analysis object to analyze in real time whether the temperature change during battery charging when the battery swap cabinet is closed is normal;

[0113] Compare and analyze the slope of the real-time smooth rising phase in the real-time temperature change time domain envelope during the battery charging process when the battery swap cabinet is closed with the slope of the normal smooth rising phase in the normal temperature change time domain envelope;

[0114] It should be noted that the normal temperature variation time domain envelope refers to the normal temperature variation time domain envelope template library for different types of batteries and different environmental conditions, which is constructed based on the temperature data of the battery charging process when the battery swap cabinet is closed in the past. Each template contains the key features of temperature rise rate, stable temperature range, and charging time. The real-time temperature variation time domain envelope refers to the use of Hilbert-Huang transform to process the temperature time series collected in real time during the battery charging process when the battery swap cabinet is closed, decompose the intrinsic mode function (IMF), and superimpose the intrinsic mode function components to generate a real-time temperature variation time domain envelope, which intuitively presents the temperature change trend during the battery charging process.

[0115] If the slope of the real-time smooth rising stage is greater than the slope of the normal smooth rising stage, it means that the battery has abnormal temperature change during the charging process, and a pre-alarm signal is generated;

[0116] If the slope of the real-time smooth rising stage is less than or equal to the slope of the normal smooth rising stage, it means that the battery temperature changes normally during charging. Then the thermal runaway phenomenon of the battery swap cabinet system will be monitored in real time according to the normal alarm logic;

[0117] Abnormal temperature location module: Based on the generation of pre-alarm signals, it locates the abnormal temperature area by analyzing the Pearson correlation coefficient between each monitoring unit;

[0118] For normal heat convection in the power exchange cabinet, the temperature changes between the composite monitoring units and the temperature sensing monitoring units usually have a strong correlation, and the Pearson correlation coefficient is high;

[0119] When an abnormal temperature rise occurs in a certain area of ​​the battery swap cabinet, the correlation between the temperature data of the composite monitoring unit or temperature sensing monitoring unit in this area and other composite monitoring units or temperature sensing monitoring units will be significantly reduced, and the Pearson correlation coefficient will become smaller;

[0120] When an abnormal temperature rise occurs in a certain area of ​​the power exchange cabinet, the temperature data of the composite monitoring unit or temperature sensing monitoring unit in that area will be significantly different from the temperature data of other composite monitoring units or temperature sensing monitoring units. By analyzing the Pearson correlation coefficient matrix, the abnormal temperature area can be quickly located, and false alarms caused by single sensor failure or temporary external interference can be eliminated.

[0121] Adjustment control module: Based on the temperature abnormality area, it identifies whether the temperature abnormality area is within the temperature fluctuation radiation range when the battery swap cabinet is closed. If so, the battery swap cabinet system automatically adjusts the speed of closing the battery swap cabinet to avoid false alarms;

[0122] Specifically, a large amount of historical speed data of the battery swap cabinet when the door is closed, and temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit when the battery swap cabinet is closed are obtained;

[0123] The collected speed and temperature data are cleaned to remove outliers and duplicate data; the collected speed and temperature data are integrated into a training set; the training set is divided into a simulation training set with a ratio of 80% and a test training set with a ratio of 20%;

[0124] Use the divided simulation training set and test training set for random forest model training;

[0125] The specific process of random forest model training is as follows:

[0126] The random forest model randomly selects samples and features from the training set to construct multiple decision trees. Each decision tree is trained based on the selected samples and features, learning the relationship between the speed data, temperature data, and radiation range when the battery swap cabinet door is closed.

[0127] During the training process, the random forest model continuously adjusts the node splitting conditions of the decision tree to minimize the prediction error;

[0128] If the model evaluation results are not ideal, optimize the model by adjusting model parameters, increasing the amount of data, or using feature engineering;

[0129] When tuning parameters, try different parameter combinations, train the model again, and evaluate the performance;

[0130] Increasing the amount of data means collecting more speed data when the battery swap cabinet door is closed, and temperature data collected by the temperature monitoring unit and the composite monitoring unit when the battery swap cabinet door is closed, further enriching the learning samples of the model;

[0131] The evaluated and optimized random forest model is recorded as the temperature fluctuation prediction model;

[0132] Use the temperature fluctuation prediction model to identify whether the temperature abnormality area is within the temperature fluctuation radiation range when the door is closed;

[0133] The specific identification process is:

[0134] When the speed of the new battery swap cabinet when closing the door is obtained, the speed of the battery swap cabinet when closing the door is input into the temperature fluctuation prediction model, and the temperature fluctuation prediction model can predict the corresponding temperature radiation range;

[0135] Compare the temperature radiation range with the positional relationship of the composite monitoring unit or the temperature sensing monitoring unit;

[0136] If the position of the composite monitoring unit or the temperature sensing monitoring unit is within the temperature radiation range, it means that the pre-alarm signal is generated by the temperature fluctuation when the battery swap cabinet door is closed. In this case, the pre-alarm signal is shielded, and the battery swap cabinet system automatically adjusts the speed of closing the battery swap cabinet door according to the temperature fluctuation prediction model to avoid false alarms.

[0137] If the position of the composite monitoring unit or the temperature sensing monitoring unit is not within the temperature fluctuation radiation range, it means that the generation of the pre-alarm signal is due to the temperature fluctuation caused by abnormal charging of the single battery. Then the normal alarm logic is restored, and the battery swap cabinet system monitors the thermal runaway phenomenon of the battery swap cabinet system in real time according to the normal alarm logic;

[0138] The technical solution of the embodiment of the present invention is as follows: using a temperature sensing monitoring unit arranged on the battery swap cabinet door and a composite monitoring unit in the single battery compartment area of ​​the cabinet to collect temperature data of the edge and core area when the door is opened in real time; by calculating the temperature gradient, when the gradient is greater than the threshold, it is determined to be a door opening disturbance, activating a buffer period to dynamically increase the alarm threshold to avoid false alarms caused by sudden temperature changes caused by opening and closing the door; after the buffer period ends, it is determined whether to restore normal alarm logic or trigger an inspection prompt based on the temperature situation;

[0139] A time series is constructed for the temperature data collected by the temperature sensing and composite monitoring units. The intrinsic mode function (IMF) is decomposed using the Hilbert-Huang transform. After multiple iterations to ensure that the residual component meets the conditions, multiple IMF components and residual components are obtained. The Hilbert transform is performed on the relevant IMF components to obtain the Hilbert spectrum, distinguish between high-frequency and low-frequency components, and superimpose to construct the temperature change time domain envelope, which can intuitively present the temperature change trend.

[0140] The smooth rising phase of the temperature variation time-domain envelope is selected, and the slope of the real-time envelope in this phase is compared with a normal template library built based on historical data. If the real-time slope is greater than the normal slope, the battery charging temperature change is determined to be abnormal and a pre-alarm signal is generated. If it is less than or equal to the normal slope, thermal runaway monitoring is performed according to the normal alarm logic.

[0141] After receiving the pre-alarm signal, a matrix is ​​constructed by calculating the Pearson correlation coefficient between each monitoring unit. Taking advantage of the fact that the correlation between each unit is high during normal thermal convection and the correlation between local units is significantly reduced during abnormal temperature rise, the matrix is ​​analyzed to locate the temperature anomaly area and eliminate false alarms caused by sensor failure or external interference.

[0142] Identify whether the temperature abnormality area is within the temperature fluctuation radiation range when the battery swap cabinet is closed. If so, the battery swap cabinet system automatically adjusts the speed of closing the cabinet door to avoid false alarms.

[0143] Example 3

[0144] Based on Example 1 and Example 2, a method for using an electric bicycle battery swap cabinet according to an embodiment of the present invention includes the following steps:

[0145] S1: Utilize the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit to analyze the temperature data fluctuation and establish a dynamic threshold compensation mechanism;

[0146] S2: Construct a time series of the temperature data collected by the temperature monitoring unit and the composite monitoring unit, analyze the time series to obtain the intrinsic mode function components, convert the intrinsic mode function components to obtain the Hilbert spectrum, distinguish the low-frequency intrinsic mode function components from the high-frequency intrinsic mode function components based on the Hilbert spectrum, and integrate the low-frequency intrinsic mode function and the high-frequency intrinsic mode function to obtain the temperature change time domain envelope;

[0147] S3: Analyze the slope of the temperature change time domain envelope in the smooth rising phase to determine in real time whether the battery charging temperature change is normal. If it is abnormal, generate a pre-alarm signal;

[0148] S4: Based on the generated pre-alarm signal, the Pearson correlation coefficient between monitoring units is analyzed to locate the temperature abnormality area;

[0149] S5: Based on the temperature abnormality area, identify whether the temperature abnormality area is within the temperature fluctuation radiation range when the battery swap cabinet is closed. If so, the battery swap cabinet system automatically adjusts the speed of closing the battery swap cabinet to avoid false alarms.

[0150] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A battery-swapping cabinet system for electric bicycles, characterized by: include: Temperature fluctuation monitoring module: Utilizes the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit to analyze temperature data fluctuations and establish a dynamic threshold compensation mechanism; Envelope construction module: constructs a time series of the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit, analyzes the time series to obtain the intrinsic mode function components, converts the intrinsic mode function components to obtain the Hilbert spectrum, distinguishes the low-frequency intrinsic mode function components from the high-frequency intrinsic mode function components based on the Hilbert spectrum, and integrates the low-frequency intrinsic mode function and the high-frequency intrinsic mode function to obtain the temperature change time domain envelope; Abnormal temperature judgment module: Analyzes the slope of the temperature change time domain envelope in the smooth rising phase, and judges in real time whether the battery charging temperature change is normal. If it is abnormal, it generates a pre-alarm signal; Abnormal temperature positioning module: Based on the generation of pre-alarm signals, the Pearson correlation coefficient between monitoring units is analyzed to locate the abnormal temperature area.

2. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: The process of establishing the dynamic threshold compensation mechanism is as follows: Calculate the temperature gradient in real time based on any temperature sensing monitoring unit or composite monitoring unit, i.e., the temperature change per unit time; compare the calculated temperature gradient with the temperature gradient threshold; If the temperature gradient is greater than or equal to the temperature gradient threshold, the battery swap cabinet system determines it as a door opening disturbance, activates the buffer period, and suppresses the normal alarm logic of the battery swap cabinet system; During the buffer period, the battery swap cabinet system will temporarily increase the alarm threshold dynamically according to the preset temperature value.

3. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: The specific process of constructing the time series is as follows: For the temperature data collected by the temperature sensing monitoring unit and the composite monitoring unit, a time series is constructed with time as the horizontal axis.

4. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: The specific process of obtaining the intrinsic mode function components is: Obtain the original temperature time series, find all the maximum and minimum points in the original temperature time series, use the cubic spline interpolation method to fit the maximum and minimum points respectively to obtain the upper envelope and lower envelope; calculate the mean of the upper and lower envelopes to obtain the mean envelope; subtract the mean envelope from the original temperature time series to calculate the residual component; Analyze the calculated residual components and the conditions of the eigenmode functions; If the residual component does not meet the conditions of the intrinsic mode function, the residual component is regarded as a new temperature time series, and the process of determining the extreme points, constructing the upper and lower envelopes, calculating the mean envelope and subtracting them is repeated until the residual component meets the conditions of the intrinsic mode function and the intrinsic mode function component is obtained.

5. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: The specific process of obtaining the Hilbert spectrum is as follows: Among the multiple intrinsic mode function components obtained by decomposition, the components closely related to the temperature change trend are selected for Hilbert transform, and the time domain signal is converted to the frequency domain to obtain the corresponding Hilbert spectrum.

6. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: The specific process of obtaining the temperature-varying time-domain envelope is as follows: After Hilbert-Huang transform decomposition, the intrinsic mode function components and residual components are obtained; The residual component, low-frequency intrinsic mode function component and medium-frequency intrinsic mode function component after decomposition are selected and superimposed to obtain the temperature-varying time domain envelope.

7. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: The specific process of determining whether the battery charging temperature change is normal is as follows: The smooth rising phase of the temperature change time domain envelope is selected as the analysis object, and the temperature change during battery charging when the battery swap cabinet door is closed is analyzed in real time to determine whether it is normal. If the slope of the real-time smooth rising stage is greater than the slope of the normal smooth rising stage, it means that the battery has abnormal temperature change during the charging process, and a pre-alarm signal is generated; If the slope of the real-time smooth rising stage is less than or equal to the slope of the normal smooth rising stage, it means that the temperature change of the battery during charging is normal, and the thermal runaway phenomenon of the battery swap cabinet system is monitored in real time according to the normal alarm logic.

8. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: The Pearson correlation coefficient between the analysis and monitoring units is specifically: For normal heat convection in the power exchange cabinet, the temperature changes between the composite monitoring units and the temperature sensing monitoring units usually have a strong correlation, and the Pearson correlation coefficient is high; When an abnormal temperature rise occurs in a certain area of ​​the battery swap cabinet, the correlation between the temperature data of the composite monitoring unit or temperature sensing monitoring unit in that area and other composite monitoring units or temperature sensing monitoring units will be significantly reduced, and the Pearson correlation coefficient will become smaller.

9. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: The specific process of locating the temperature abnormality area is as follows: When an abnormal temperature rise occurs in a certain area of ​​the battery swap cabinet, the temperature data of the composite monitoring unit or temperature sensing monitoring unit in this area will be significantly different from the temperature data of other composite monitoring units or temperature sensing monitoring units. By analyzing the Pearson correlation coefficient matrix, the temperature abnormality area can be quickly located.

10. The battery-swap cabinet system for electric bicycles according to claim 1, characterized in that: Also includes the following modules: Adjustment control module: Based on the temperature abnormality area, identify whether the temperature abnormality area is within the temperature fluctuation radiation range when the battery swap cabinet is closed. If so, the battery swap cabinet system automatically adjusts the speed of closing the battery swap cabinet to avoid false alarms.

Images