Heat management method suitable for monorail crane lithium battery fast charging module

Abstract

This invention belongs to the field of power battery thermal management and control technology, specifically relating to a thermal management method for lithium battery fast-charging modules applicable to monorail cranes. The method includes: synchronously reading cell thermal data and system operating condition data; calculating the temperature rise rate sequence; obtaining extreme thermal data and heat generation benchmark data; outputting dynamic driving force data; calculating runaway probability data; obtaining performance boundary data; outputting target compensation data; calculating final driving data; and generating fan control commands and thermal management ledger records. This invention solves the technical problems of existing technologies that rely on global average temperature and absolute temperature thresholds for passive triggering, failing to consider discrete heating within the battery pack and the lag in electrothermal physical conduction, resulting in local extreme temperature rises being masked by the average, severe lag in cooling system response, and missed detection of thermal runaway.

Description

Technical Field

[0001] This invention relates to the field of power battery thermal management and control technology. More specifically, this invention relates to a thermal management method suitable for fast-charging modules of lithium batteries in monorail cranes. Background Technology

[0002] With the continuous development of new energy and special transportation equipment, the fast charging module of monorail cranes, as a key node in the core energy storage system, directly affects the thermal management quality of the equipment during continuous high-intensity operation and the safety of the power battery. During high-load fast charging operations, the battery pack typically contains multiple individual cells with different physical locations and heat dissipation conditions. These cells need to undergo short-term continuous injection of high-rate current. Due to the differences in polarization resistance of individual cells and the delay in thermal conduction, the battery pack is highly susceptible to thermal runaway risks such as extreme localized heating of individual cells and extremely uneven spatial distribution of heat.

[0003] In related technologies, for example, Chinese patent application CN108777336B discloses a lithium battery pack thermal management system. This patent collects global temperature data of the battery pack and compares it with an absolute temperature threshold to trigger a cooling mechanism and map the fan speed. This approach can achieve basic temperature anomaly screening and full-process adjustment under simple slow charging conditions where the battery rate is low or heat generation is gradual. However, analysis of the performance of existing technologies under the complex conditions of high-rate current fast charging of monorail cranes reveals that this algorithm architecture, which relies on global absolute temperature mapping and passive static triggering, has the drawback of ignoring the discrete heat generation of the battery space and the lag in electrothermal physical conduction.

[0004] Under high-rate current injection during fast charging, the physical heating of individual battery cells exhibits extreme unevenness. Due to varying internal heat dissipation conditions and internal resistance distributions, severe temperature rises can easily occur in localized central regions. Furthermore, the confined space within the battery pack and the frequent conduction delays and localized heat shielding between various phase change materials and the outer casing result in a severely lagging and diluted numerical representation of the global temperature in the temperature sensor network. Existing technologies often extract the spatially highly discrete individual cell temperatures uniformly and perform global mean fitting when processing temperature characteristics. The algorithm cannot distinguish between dangerous cells with severe localized temperature rises and cells that appear safe due to rapid heat dissipation at the edges. This mathematical model assigns higher weights to cells with gentler edges when calculating the actual overall thermal state of the battery, thus masking the true extreme heating characteristics. Because the control algorithm relies on this averaging masking effect, as long as the algorithm adheres to the judgment benchmark of global aggregation and dimensionality reduction, conventional threshold parameter fine-tuning cannot detect thermodynamic abrupt changes in individual cells within the averaged data. This prevents the system from locking down extreme overheating in local cells in advance, ultimately causing the active defense system to compromise with delayed passive cooling, compromising the safety and reliability of the monorail crane fast charging module. To address the technical problem of existing technologies relying on global averaging, which masks local extreme temperature rises and leads to severe lag in cooling system response and missed detection of thermal runaway, a solution is needed. Summary of the Invention

[0005] To address the technical problems of existing technologies that rely on global average temperature and absolute temperature thresholds for passive triggering, failing to consider discrete heating within the battery pack and the lag in electrothermal physical conduction, resulting in local extreme temperature rises being masked by the average, severe lag in cooling system response, and missed detection of thermal runaway, this invention provides a thermal management method suitable for lithium battery fast charging modules in monorail cranes, including:

[0006] S1: By activating the underlying sensor network and locking the global clock, synchronously read the cell thermal data and system operating condition data; S2: Based on the dynamic relationship of temperature change over time, and combining the real-time battery temperature sequence and the initial battery temperature sequence in the cell thermal data, calculate the temperature rise rate sequence; based on the Joule law heat source driving relationship, and combining the bus charging current and the battery corrected internal resistance in the system operating condition data, obtain extreme thermal data and heat generation benchmark data; based on the temperature rise rate and Joule thermal cross-validation characteristics, and combining the extreme thermal data and heat generation benchmark data, output dynamic driving force data; S3 S4: Based on the scaling characteristics of deviation from the safety baseline, combined with dynamic driving force data and heat generation baseline data, calculate the runaway probability data; pre-acquire the fan full-load speed and fan base speed, perform physical work envelope margin feature calculation on both to obtain performance boundary data; perform scalar multiplication operation on performance boundary data and runaway probability data to output target compensation data; S5: Based on the principle of base speed feedforward and dynamic compensation superposition, combined with fan base speed and target compensation data, calculate the final driving data; transform the final driving data to generate fan control commands and thermal management ledger records.

[0007] This invention addresses the technical shortcomings of existing monorail crane lithium battery fast charging module thermal management technology, which relies on global average temperature and absolute temperature threshold for passive triggering, without considering discrete heating within the battery pack and the lag in electrothermal physical conduction. This results in local extreme temperature rises being masked by the average, severe lag in cooling system response, and missed detection of thermal runaway. This invention synchronously collects temperature and electrical condition data from multiple independent nodes within the battery pack, preserving the discrete heating characteristics of the internal space. It captures the dynamic thermal excitation intensity by calculating the temperature rise rate sequence of each cell using first-order difference calculation. It extracts extreme thermal data through global extremum optimization to break the mean mask, and combines this with Joule's law to calculate heat generation benchmark data, achieving cross-verification between temperature appearance and heat source. It outputs dynamic driving force data through nonlinear fusion of temperature rise rate and basic Joule heat, amplifying the real hidden dangers of dual deterioration of heat source and temperature appearance. It calculates runaway probability data through normalization calculation of dangerous heat drop and risk smoothing scale, replacing hard thresholds to achieve smooth risk assessment. It combines fan performance boundary data with runaway probability mapping to output target compensation data, superimposing dynamic compensation on the base speed and generating final driving data through saturation limiting. Finally, it encodes this into fan control commands for execution and generates thermal management ledger records with serial numbers and timestamps.

[0008] Preferably, synchronously reading cell thermal data and system operating condition data includes: The bus listening module of the battery management system is activated to establish the physical topology boundary of the battery pack. Each independent cell node is traversed and read online. The absolute surface temperature of each node in the previous acquisition cycle is recorded as the initial battery temperature, and a battery initial temperature sequence is generated; the current acquisition cycle's temperature is simultaneously acquired online. The real-time absolute surface temperature of each independent cell node is recorded as the real-time battery temperature, and a real-time battery temperature sequence is generated. Then, the local high-precision hardware clock is accessed to parse the time difference between two valid acquisition actions, generate the temperature sampling duration, and package the initial battery temperature, real-time battery temperature, and temperature sampling duration into cell thermal data. The bus charging current and battery corrected internal resistance are pre-acquired and packaged into system operating condition data.

[0009] This invention activates the battery management system bus listening module, establishes independent cell nodes based on the physical topology of the battery pack, reads the absolute surface temperature of each node in the previous cycle and the current cycle respectively, generates the initial temperature sequence and the real-time temperature sequence of the battery, analyzes the time difference between the two acquisitions to generate the temperature sampling duration, and packages it into cell thermal data; at the same time, it extracts the bus charging current and the battery corrected internal resistance and packages them into system operating condition data.

[0010] Preferably, calculating the temperature rise rate sequence includes: The real-time battery temperature sequence and the initial battery temperature sequence are extracted from the cell thermal data. In the arithmetic logic unit, a vector difference operation is performed along the spatial dimension on the corresponding element positions of these two sequences to obtain the result containing… A discrete temperature difference sequence of battery cells is obtained; the temperature sampling duration of the preceding analysis is extracted; each element in the above battery cell temperature difference sequence is divided by the temperature sampling duration to perform algebraic division normalization operation; and then strictly reconstructed according to the original grid coordinate mapping to obtain the solution. The cell temperature rise rate sequence is calculated and the results are encapsulated as a temperature rise rate sequence.

[0011] This invention obtains a cell temperature difference sequence by subtracting the real-time battery temperature sequence from the initial battery temperature sequence element by element along the spatial dimension. Then, each temperature difference element is divided by the temperature sampling time to perform a normalization operation, and the result is reconstructed into a temperature rise rate sequence based on the original grid coordinates. This method converts static absolute temperature into a dynamic temperature rise rate index through first-order difference, directly characterizing the intensity of heat accumulation per unit time in each cell. It can capture early thermal runaway risks where the absolute temperature has not yet exceeded the limit but the temperature rise has already deteriorated rapidly, overcoming the hysteresis limitation of traditional threshold control.

[0012] Preferably, obtaining extreme thermodynamic data and heat production baseline data includes: Historical steady-state currents under fault-free conditions are extracted, and their heat generation fluctuation data are statistically analyzed to extract the variance of heat generation fluctuations. A temperature rise rate sequence is extracted, and a global traversal of the extreme value optimization based on a comparison register is performed in the numerical space to lock and extract the scalar element with the largest value in the sequence as the extreme thermal data. The bus charging current and the battery corrected internal resistance are parsed from the system operating condition data. The squared bus charging current is multiplied by the battery corrected internal resistance to calculate the base heat generation power. The squared historical steady-state current is multiplied by the battery corrected internal resistance to generate the steady-state reference heat generation power. The square root of the heat generation fluctuation variance is extracted to obtain the risk smoothing scale. The base heat generation power, the steady-state reference heat generation power, and the risk smoothing scale are packaged into the heat generation reference data.

[0013] This invention extracts the heat generation fluctuation data from the fault-free historical steady-state current to statistically calculate the heat generation fluctuation variance; performs a global traversal of the temperature rise rate sequence to find the extreme value and extracts the maximum value as the extreme value thermal data; calculates the basic heat generation power by squaring the bus charging current and multiplying it by the battery corrected internal resistance; generates the steady-state reference heat generation power by squaring the historical steady-state current and multiplying it by the nominal internal resistance; and obtains the risk smoothing scale by taking the square root of the heat generation fluctuation variance. The three are packaged into the heat generation reference data.

[0014] Preferably, the output dynamic driving force data includes: Extract the extreme thermal data and heat production baseline data generated in the previous step, parse out the basic heating power in the heat production baseline data, divide the extreme thermal data by the preset baseline temperature rise rate to obtain the dimensionless temperature rise ratio coefficient, multiply by the preset scaling factor and add 1, and then multiply by the basic heating power to obtain the dynamic thermal driving force scalar, which is denoted as dynamic driving force data.

[0015] Preferably, calculating the runaway probability data includes: Extract the heat generation baseline data and analyze the steady-state baseline heat generation power and risk smoothing scale; perform algebraic subtraction operation between the dynamic driving force data and the steady-state baseline heat generation power to obtain the dangerous heat drop; perform proportional division scaling operation on the dangerous heat drop based on the risk smoothing scale, and truncate the final calculation result to the convergence interval of 0 to 1 to calculate the dimensionless runaway risk probability coefficient, which is denoted as the runaway probability data.

[0016] This invention extracts the steady-state baseline heating power and risk smoothing scale from the heat generation baseline data, subtracts the steady-state baseline heating power from the dynamic driving force data to obtain the dangerous heat drop, and then performs a proportional division scaling operation on the dangerous heat drop based on the risk smoothing scale. The result is truncated and limited to the interval between 0 and 1 to obtain the runaway risk probability coefficient, i.e., the runaway probability data.

[0017] Preferably, obtaining performance boundary data includes: Further read the configuration nameplate parameters of the heat dissipation drive control board, extract the fan full-load speed and fan base speed, and obtain the battery pack serial number through external handshake; in the instruction generator, subtract the fan base speed from the fan full-load speed, and the difference is the available speed range scalar. The available speed range scalar and the battery pack serial number are encapsulated into performance boundary data.

[0018] Preferably, the output target compensation data includes: The available speed range scalar is parsed from the performance boundary data, and the previously generated runaway probability data is extracted; the available speed range scalar and the runaway probability data are multiplied by a scalar to calculate the target compensation data.

[0019] Preferably, calculating the final driving data includes: Perform an algebraic addition operation on the fan base speed and the target compensation data. When the sum of the two is less than the reconstruction saturation limit threshold of the full-load speed, output the sum directly. When the sum of the two exceeds the reconstruction saturation limit threshold, output according to the reconstruction saturation limit threshold of the full-load speed, and record it as the final drive data.

[0020] Preferably, generating fan control commands and thermal management log records includes: The final drive data is encoded using the microcomputer PWM protocol, transforming it into a duty cycle instruction sequence that can be directly recognized by the underlying heat dissipation module to drive the underlying heat dissipation module, generate fan control commands, and send them out for execution; the battery pack serial number obtained through the preceding external handshake is extracted, and a length of [missing information] is [missing information]. The battery pack serial number and duty cycle instruction sequence are treated as independent fields. The data clusters are packaged and encapsulated in strict accordance with the timestamp time sequence format to generate a thermal management ledger record with the current time sequence stamp and write it into the database.

[0021] The beneficial effects of this invention are as follows: Based on existing battery management systems and cooling fan hardware, no additional temperature sensors or cooling components are required. The algorithm can be directly integrated into existing monorail crane battery management systems, resulting in low modification costs. It is applicable to lithium battery fast-charging modules for various special equipment such as mining monorail cranes and underground transport vehicles. The multi-cell independent extreme value identification and temperature rise rate early warning mechanism effectively solves the problem of missed detection of local thermal runaway hazards by traditional global average control, increasing the early detection lead time for thermal runaway and reducing the risk of safety accidents caused by lithium battery thermal runaway. The runaway probability normalization and on-demand precise heat dissipation mechanism avoids frequent fan start-stop and excessive heat dissipation caused by traditional threshold control, reducing the energy consumption of the cooling system and extending the fan's lifespan. The persistent recording function of thermal management ledgers provides traceable data for safety audits and full life-cycle maintenance of special equipment, meeting the stringent safety supervision requirements of mining equipment. This solution provides a highly reliable and easy-to-implement technical solution for the intelligent thermal management of lithium battery fast-charging modules in special transportation equipment, helping to improve the operational safety and maintenance efficiency of new energy special equipment. Attached Figure Description

[0022] Figure 1 This schematically illustrates a flowchart of a thermal management method for a lithium battery fast-charging module for a monorail crane according to the present invention. Figure 2 This diagram schematically illustrates the time-series comparison of temperature and fan speed during the fast charging process in this invention. Figure 3 A bar chart illustrating the performance comparison of different thermal management methods in this invention is shown. Detailed Implementation

[0023] This invention discloses a thermal management method applicable to fast charging modules of lithium batteries for monorail cranes, referring to... Figure 1 This includes steps S1-S4: S1: Synchronously read cell thermal data and system operating data.

[0024] It's important to note that in the harsh environment of continuous high-load operation of monorail cranes, battery packs are often located in confined spaces with limited heat dissipation. One issue is the slow, localized temperature rise due to proximity to the motor heat source, characterized by a relatively high static temperature but a slow overall temperature increase. Simultaneously, there's the risk of sudden thermal runaway caused by micro-short circuits or high-current polarization within the cells, characterized by initially inconspicuous absolute temperature increases but extremely rapid transient temperature rises. Both of these scenarios coexist in complex fast-charging batches. Without a unified and thorough benchmark for assessing this phenomenon, the thermal management system faces a severe decision-making dilemma, unable to determine which should be prioritized for emergency cooling. From the perspective of battery thermodynamics and physical spatial distribution, the potential risk of thermal runaway and explosion from sudden, localized temperature rises, masked by averaging, is far greater than that of a gradual, global temperature increase. Therefore, it is necessary to delve into the individual physical nodes at the lowest level, taking the original surface temperature of each cell and the underlying bus electrical excitation as an independent spatiotemporal anchor point, so as to extract a set of underlying state indicators that can truly reflect the characteristics of microscopic anisotropic distribution, so that any evaluation parameters extracted later can be directly related to the real physical source.

[0025] Specifically, by activating the underlying sensor network and locking the global clock, the thermal data of the battery cells and the system operating condition data are read synchronously, including: The bus listening module of the battery management system is activated to establish the physical topology boundary of the battery pack. Each independent cell node is traversed and read online. The absolute surface temperature of each node in the previous acquisition cycle is recorded as the initial battery temperature. These initial battery temperatures are then cascaded and assembled strictly according to the spatial grid order of their physical numbers to generate... A one-dimensional sequence, denoted as the battery initial temperature sequence; synchronous online acquisition of the current acquisition cycle. The real-time absolute surface temperature of each independent cell node, denoted as the real-time battery temperature, is recombined according to the same grid coordinate mapping to generate... The system obtains the real-time battery temperature sequence; then accesses the local high-precision hardware clock, analyzes the time difference between two valid acquisition actions, generates the temperature sampling duration, and packages the initial battery temperature, real-time battery temperature, and temperature sampling duration into cell thermal data; it extracts the bus charging current representing the current bus energy injection intensity through the underlying protocol unpacking, and extracts the factory-calibrated battery corrected internal resistance from the read-only memory; and packages the bus charging current and battery corrected internal resistance into system operating condition data.

[0026] It should be noted that a multi-channel synchronous dynamic acquisition algorithm is used in this invention, wherein the total number of established independent cell nodes is... The range of values ​​is to Preferred ,like Excessively large values ​​will cause congestion in the underlying bus communication and crashes in the microcontroller's computing power; excessively small values ​​will create blind spots in physical space coverage, leading to missed detections of extreme temperature rise abrupt changes. The initial battery temperature refers to the individual cell temperature sequence of the previous cycle, representing the historical thermodynamic state. Obtaining the initial battery temperature aims to provide a baseline for differential calculations. The real-time battery temperature refers to the absolute temperature sequence of the current cycle, representing transient thermodynamic phenomena. Obtaining the real-time battery temperature aims to capture the latest operating conditions. The temperature sampling duration refers to the time difference between two acquisitions, representing the time granularity of the data. Constructing the temperature sampling duration aims to achieve time-domain normalization. The underlying sampling period threshold range involved here is set as follows: to Preferred If this value is too large, extreme temperature rise mutations will be diluted and missed by the time dimension; if it is too small, it will cause bus communication congestion and microcontroller computing power collapse. In this embodiment, the battery pack to be managed includes 128 monitored cell nodes. Each cell node corresponds one-to-one with a physical cell in the battery pack. Multiple cells are not sampled together, so the total number of independent cell nodes is set to 128. For other battery pack specifications, the total number of independent cell nodes can be determined according to the actual number of monitored cells, preferably set between 64 and 256. Setting the total number of independent cell nodes to 128 can cover all cell locations inside the battery pack, avoiding monitoring blind spots caused by local abnormal cells, and can also control the amount of temperature data collected in a single session within the real-time processing capacity of the underlying communication bus and controller. When the number of cell nodes exceeds 256, it is easy to increase communication occupancy and data traversal latency; when the number of cell nodes is less than the actual number of cells in the battery pack, it will reduce the spatial identification accuracy of local abnormal temperature rises. The battery's internal resistance is corrected by looking up a table based on the current average cell temperature. The temperature-internal resistance mapping table is obtained from the battery's factory calibration and covers the entire operating temperature range of -20℃ to 60℃. Each 1℃ corresponds to an internal resistance value, ensuring that the accuracy of Joule heating calculations is not affected by temperature changes.

[0027] Thus, we obtained the thermal data of the battery cells and the operating data of the system.

[0028] S2: Output dynamic driving force data.

[0029] It's important to note that in the transient surge scenario of high-rate fast charging for monorail cranes, the interference of ambient temperature and the battery's own self-heating effect are often deeply intertwined. One factor is the overall high base temperature caused by years of operation in high-temperature mines, characterized by large absolute temperature values ​​but minimal increments. Simultaneously, another factor is the sudden change in local polarization resistance caused by high-rate charging, characterized by an initial temperature within the normal range but an exponential increase in heat accumulation per unit time. These two states coexist in various stages of the charging cycle. If only absolute temperature thresholds are used as a metric, the cooling system will face serious misjudgment difficulties, easily overreacting to overall high temperatures while missing localized fatal mutations. From the perspective of heat transfer dynamics, the potential thermal runaway direction of the transient accumulation rate of heat in a small space is far greater than that of the static temperature appearance. Therefore, it is necessary to use the time span as a normalized weighting factor to perform dynamic derivative transformation on the discrete temperature difference, thereby extracting a first-order dynamic characteristic index that can accurately characterize the intensity of internal thermal excitation, so that the extracted rate parameter is directly related to the actual degree of heat accumulation danger.

[0030] Specifically, based on the dynamic relationship of temperature over time, and combining the real-time battery temperature sequence and the initial battery temperature sequence from the cell thermal data, a temperature rise rate sequence is calculated, including: The real-time battery temperature sequence and the initial battery temperature sequence are extracted from the cell thermal data. In the arithmetic logic unit, a vector difference operation is performed along the spatial dimension on the corresponding element positions of these two sequences to obtain the result containing… A discrete temperature difference sequence of battery cells is obtained; the temperature sampling duration of the preceding analysis is extracted; each element in the above battery cell temperature difference sequence is divided by the temperature sampling duration to perform algebraic division normalization operation; and then strictly reconstructed according to the original grid coordinate mapping to obtain the solution. The cell temperature rise rate sequence is calculated and the results are encapsulated as a temperature rise rate sequence.

[0031] It should be noted that static absolute temperature cannot reflect the intensity of heat generation during high-rate charging. This step specifically employs a linear spatial first-order difference estimation algorithm. Here, the cell temperature difference sequence refers to the absolute increment of the individual cell temperature, representing the pure magnitude of heat accumulation. Obtaining the cell temperature difference sequence aims to eliminate the appearance of static ambient temperature. The temperature rise rate refers to the temperature rise value per unit time, representing the true heating intensity. Constructing the cell temperature rise rate aims to eliminate time interference caused by different sampling periods. The effective difference calculation accuracy retained within the threshold range is... to Preferred If the precision is too high, it will cause the loss of early features of slight temperature rise; if it is too low, it will easily misjudge the sensor noise floor as a temperature transition.

[0032] Thus, a temperature rise rate sequence for characterizing the actual heating temperature was obtained.

[0033] It should be noted that in complex monorail crane operation networks, data fluctuations in multiphase physical fields are often highly deceptive. One phenomenon is a single-point temperature jump caused by transient electromagnetic interference from sensors, characterized by an abrupt temperature rise rate but lacking underlying electrical power support. Simultaneously, another phenomenon is deep thermal runaway caused by actual cell aging, characterized not only by an extreme temperature rise rate throughout the sequence but also by a massive injection of energy into the bus. These two phenomena coexist in noisy industrial fast-charging environments. Without a comprehensive measurement standard encompassing both electrothermal and electrical dimensions, the algorithm system will face extremely high false alarm and false negative rates, unable to determine which should trigger substantial heat dissipation intervention. From the perspective of the Joule heating effect, the potential irreversible thermal damage impact of a localized temperature rise supported by a real high-current physical source is far greater than that of passive temperature fluctuations. Therefore, it is necessary to use the basic heat generation power of the physical source as a verification factor, and combine it with the benchmark tolerance of historical steady-state heat generation to trace the source of the extracted extreme thermal phenomena. This will allow for the extraction of a set of benchmark comparison indices that can comprehensively reflect the current trend of heat deterioration and the historical health baseline, so that the extracted benchmark is directly related to the real source driving force.

[0034] Preferably, based on the Joule law heat source driving relationship, and combined with the bus charging current and battery corrected internal resistance in the system operating condition data, extreme thermodynamic data and heat generation baseline data are obtained, including: Access the steady-state charging history log in the underlying memory, extract the historical steady-state current under fault-free conditions, and statistically analyze its heat generation fluctuation data to extract the variance of heat generation fluctuation; extract the preceding temperature rise rate data and parse it. The cell temperature rise rate sequence is analyzed, and a global traversal extreme value search based on a comparison register is performed in the numerical space to lock and extract the scalar element with the largest value in the sequence as the extreme thermal data. The bus charging current and the battery corrected internal resistance are parsed from the system operating data. The square of the bus charging current is multiplied by the battery corrected internal resistance to calculate the basic heating power. The square of the historical steady-state current is extracted and multiplied by the battery corrected internal resistance to calculate the steady-state reference heating power. The square root of the variance of heat generation fluctuation is extracted to obtain the risk smoothing scale. The basic heating power, the steady-state reference heating power and the risk smoothing scale are packaged into the heat generation reference data.

[0035] It should be noted that this invention employs a multivariate dynamic optimization algorithm and a Joule physics deduction mechanism. Extreme thermal data refers to the most intense temperature rise across the entire envelope, characterizing the deterioration trend of the most dangerous local unit. Obtaining extreme thermal data aims to break through the masking effect of mean values. The basic heating power refers to the heating power of the electrical source, characterizing the system's intrinsic thermal driving force. Constructing the basic heating power aims to establish a correlation between temperature results and the heating source. The quantization overflow cutoff threshold range for the current square operation involved here is... to Preferred If the value is too large, it will increase the risk of memory overflow in the computing module; if it is too small, it will cause the high-rate charging characteristics to be forcibly clipped.

[0036] Thus, extreme thermodynamic data and heat production baseline data for evaluating the intrinsic heat intensity of the underlying physics were obtained.

[0037] It should be noted that in the high-density energy storage architecture of monorail cranes, there is a nonlinear spatiotemporal misalignment between heat generation and manifestation. One issue is the localized residual heat accumulation caused by temporary blockage of external cooling ducts, characterized by a high temperature rise rate but extremely low current injection. Simultaneously, another issue arises from the uncontrolled internal chemical reaction directly triggered by high-rate fast charging, characterized by an extremely high temperature rise rate and a massive simultaneous burst of base heat generation power. These two situations coexist in the dynamic operating conditions of the equipment. If simple arithmetic superposition or a single-dimensional threshold is used to address this phenomenon, the control system will face the dilemma of fragmented response logic, making it impossible to accurately determine which is the true combined threat to battery pack lifespan. From the perspective of multi-physics field coupling evolution, the potential chain reaction destructive force of the dual synchronous deterioration of the heat source and thermal manifestation is far greater than that of a single-dimensional anomaly. Therefore, it is necessary to use the base heat generation power as a nonlinear amplification factor to perform feature cross-fusion of extreme thermal data, thereby extracting a dynamic driving index that quantifies the comprehensive risk level, directly linking this extracted index to the actual potential energy of the thermal runaway.

[0038] Preferably, based on the cross-validation characteristics of temperature rise rate and Joule heat, combined with extreme thermodynamic data and heat production baseline data, dynamic driving force data is output, including: Extract the extreme thermal data and heat production baseline data generated in the previous step, parse out the basic heating power in the heat production baseline data, divide the extreme thermal data by the preset baseline temperature rise rate to obtain the dimensionless temperature rise ratio coefficient, multiply by the preset scaling factor and add 1, and then multiply by the basic heating power to obtain the dynamic thermal driving force scalar, which is denoted as dynamic driving force data.

[0039] It should be noted that dynamic thermal driving force refers to the comprehensive physical intensity resulting from the mutual mapping and recombination of extreme temperature rise and Joule heat, characterizing the dual deterioration trend from the heat source to the thermal manifestation. Constructing dynamic thermal driving force aims to achieve a nonlinear amplification mapping of potential hazards; the threshold range of the preset scaling factor here is... to Preferred If the value is too large, it will amplify the thermal fluctuations under normal operating conditions, leading to false triggering; if it is too small, it will weaken the sensitivity of the cross-validation system to extreme and dangerous situations. The preset reference temperature rise rate is the average temperature rise rate of the cell under fault-free steady-state charging conditions, with a value ranging from 0.1℃ / s to 0.5℃ / s. The preset reference temperature rise rate is preferably 0.3℃ / s, calibrated at the battery factory under a 25℃ environment and a 0.5C standard charging condition, representing the normal temperature rise level of a healthy battery during steady-state charging.

[0040] It should be noted that when the extreme temperature rise rate is less than or equal to 0, it indicates that the battery pack is in a cooling or constant-temperature state with no risk of heat accumulation. At this time, the temperature rise rate factor is set to 0, the dynamic driving force is equal to the steady-state baseline heating power, the runaway probability is set to 0, and the fan maintains its base speed to avoid ineffective speed regulation. The variance of heat generation fluctuation is calculated using the most recent 100 hours of fault-free steady-state charging data, with the variance of heating power statistically analyzed at a sampling frequency of 1Hz. This covers the complete charge-discharge cycle and typical operating temperature range, ensuring that the risk smoothing scale conforms to the actual operating fluctuation characteristics of the battery pack.

[0041] Thus, dynamic driving force data for assessing the potential for localized extreme heating were obtained.

[0042] S3: Output target compensation data.

[0043] It should be noted that in the actual deployment of fast charging modules, the absolute energy range in a physical sense often spans a vast range and is difficult for microprocessors to directly control. A slight heat drop occurring in an extremely cold environment, though small in absolute value, is abnormal relative to the smoothness scale of that environment; simultaneously, normal fluctuations in a high-temperature, full-load environment, while having a large absolute drop, are still within the reasonable envelope of that operating condition. These two situations coexist in operating batches from different seasons or aging stages. Without a unified dimensionless normalization standard, hardware drive units will face a decision-making dilemma regarding control parameters exceeding limits, unable to determine what absolute energy deviation should be mapped to what level of execution intervention. From the perspective of probabilistic mapping in control engineering, the relative scaling ratio crossing the dynamic baseline has a far greater potential instructional guidance significance than isolated physical absolute values. Therefore, it is necessary to use the risk smoothing scale of historical fluctuations as a normalization weighting factor to truncate and scale dangerous heat drops exceeding the safety threshold, thereby extracting a standardized risk probability index that can assess the deterioration trend. This extracted probability index is then directly correlated with the urgency of the actual required heat dissipation intervention.

[0044] Specifically, based on the scaling characteristics of deviations beyond the safety baseline, and combining dynamic driving force data with heat generation baseline data, runaway probability data is calculated, including: Extract the heat generation baseline data and analyze the steady-state baseline heat generation power and risk smoothing scale; perform algebraic subtraction operation between the dynamic driving force data and the steady-state baseline heat generation power to obtain the dangerous heat drop; perform proportional division scaling operation on the dangerous heat drop based on the risk smoothing scale, and truncate the final calculation result to the convergence interval of 0 to 1 to calculate the dimensionless runaway risk probability coefficient, which is denoted as the runaway probability data.

[0045] It should be noted that the energy parameters in the physical dimension need to be normalized into probability weight coefficients that are easily identifiable and executable by the control system; in this invention, this can be specifically achieved using nonlinear interval mapping and a Sigmoid normalization approximation algorithm. Specifically, the dangerous heat drop is the positive excess of the dynamic driving force relative to the steady-state baseline heating power, representing the physical span of deviation from the normal safety baseline. Obtaining the dangerous heat drop aims to filter out interference from steady-state heat generation. The runaway risk probability refers to a dimensionless coefficient normalized to the zero-to-one interval, representing the exact probability level of the system moving towards thermal runaway. Constructing the runaway risk probability aims to avoid abrupt start-stop interventions caused by hard absolute values. The probability cutoff lower limit threshold range involved here is fixed at [insert range here]. to Preferred When the initial risk ratio is not higher than When the normalized runaway risk coefficient is set to zero; when the initial risk ratio is higher than zero, the normalized runaway risk coefficient is set to zero. At that time, the remaining risk is recalculated to a range of 0 to 1. (Settings) The risk dead zone threshold at that time can filter out minor thermal fluctuations and measurement noise, avoiding frequent fan speed adjustments. The risk dead zone threshold is higher than... When this happens, it can easily reduce the sensitivity of early anomaly detection; without setting a risk dead zone, the fan speed can easily be repeatedly adjusted due to temperature fluctuations within the normal range. The specific conversion rule for the probability of runaway is as follows: when the ratio of the dangerous heat drop to the risk smoothing scale is less than or equal to 0.05, the probability of runaway is 0; when the ratio is greater than 0.05 and less than 1.05, the probability of runaway is... When the ratio is greater than or equal to 1.05, the probability of loss of control is taken as 1, and the calculation result is rounded to two decimal places.

[0046] For example, if the dangerous heat drop is 50W, and there is a scenario where the smoothing scale is 100W, then the risk ratio is 0.5, falling within the range of 0.05 to 1.05, indicating a probability of runaway. Scenario 2: Smoothing scale = 40W, then the risk ratio... The probability of loss of control is 1, and the above calculation results are rounded to two decimal places.

[0047] Thus, runaway probability data were obtained to characterize the exact probability level of the system moving toward thermal runaway.

[0048] It should be noted that, at the mechanical execution level of a monorail crane's cooling system, the physical efficiency of the fan motor is not an ideal model with infinite extension. A fan idling under low load conditions is characterized by a very large range of remaining available speeds and ample cooling capacity allocation space. Simultaneously, another fan, having already been operating under high load, is characterized by an extremely high base speed and a mere millimeter's margin from its full load limit. These two mechanical states coexist in long-cycle continuous charging tasks. Without a clear understanding of the actual work boundaries of the current hardware, the software-level control algorithm will face the dilemma of instruction overflow. Blindly issuing over-limit instructions may lead to motor stalling or PWM module burnout. This invention uses the fan's full load capacity as a hard upper limit factor, performing differential stripping on the base speed that currently ensures the basic cycle, thereby extracting a boundary index that truly represents the range of remaining available cooling capacity. This extracted boundary is directly related to the physical feasibility of all subsequent dynamic allocations.

[0049] Preferably, the full-load fan speed and the base fan speed are obtained in advance, and the physical work envelope margin feature is calculated on both to obtain performance boundary data, including: Further read the configuration nameplate parameters of the heat dissipation drive control board, extract the fan full-load speed and fan base speed, and obtain the battery pack serial number through external handshake; in the instruction generator, subtract the fan base speed from the fan full-load speed, and the difference is the available speed range scalar. The available speed range scalar and the battery pack serial number are encapsulated into performance boundary data.

[0050] It should be noted that this embodiment uses a DC brushless cooling fan with a rated full-load speed of 6000 RPM. Therefore, the fan's full-load speed is set to 6000 RPM; this speed is the maximum continuous operating speed specified on the fan's nameplate, used to limit the physical upper limit of the fan drive commands. Based on the calibration results of the minimum airflow required to maintain basic ventilation of the battery pack under normal fast charging conditions, the fan's base speed is set to 1800 RPM, approximately 30% of the fan's full-load speed. This base speed is higher than the fan's stable start-up speed, which can prevent the fan from stalling at low speeds and maintain basic heat dissipation of the battery pack when no abnormal temperature rise occurs, while retaining a large dynamic speed adjustment range. The battery pack serial number is set to 32 bytes in length to record the manufacturer code, battery model, production date, production batch, and device serial number; 32 bytes provide sufficient space for unique device identification without significantly increasing the database field length and persistent storage overhead.

[0051] Thus, performance boundary data representing the remaining available performance span of the heat dissipation equipment was obtained.

[0052] It's important to note that in the process of translating the probability of runaway at the software layer into physical actions at the hardware layer, the accuracy of the linear mapping directly determines the effectiveness of the defense system. A high-probability, high-risk thermal runaway hazard requires immediate matching with a significant increase in rotational speed when hardware margins are sufficient; simultaneously, a lower-probability, slight fluctuation requires only an extremely small adjustment in rotational speed under the same hardware margins. These two dynamic matching requirements coexist in every real-time control clock cycle. Without a proportionally scaled gain matching standard, the actuator will face either slow response or excessive noise, making it impossible to accurately inject energy as needed. From the perspective of full-process negative feedback regulation, targeted physical compensation strictly aligned with the risk probability has a potential energy utilization rate and cooling effect far exceeding that of rigid control with fixed steps. Therefore, the runaway risk probability needs to be used as a core gain allocation factor, and a nonlinear mapping transformation should be performed on the calculated available rotational speed range to extract a dynamic compensation index that accurately represents the additional heat dissipation intensity, directly linking this extracted dynamic parameter to the actual thermal countermeasure requirements.

[0053] Preferably, a scalar multiplication operation is performed on the performance boundary data and the runaway probability data to output target compensation data, including: The available speed range scalar is parsed from the performance boundary data, and the previously generated runaway probability data is extracted; the available speed range scalar and the runaway probability data are multiplied by a scalar to calculate the target compensation data.

[0054] It should be noted that projecting intangible probability onto a tangible mechanical performance span is the core bridge for the transition from software algorithms to hardware control; in this invention, a proportional gain mapping algorithm can be specifically employed. Here, the dynamic compensation speed refers to the specific physical adjustment amount allocated based on the current level of danger, characterizing the system's counterbalancing work intensity against sudden local extreme heating. Constructing the dynamic compensation speed aims to achieve on-demand dynamic injection of cooling capacity; the mapping response hysteresis filtering time threshold involved here is... to Preferred If the value is too large, it will weaken the effectiveness of feedforward defense under high current fast charging; if it is too small, it will cause the fan speed to fluctuate frequently and violently with electrical noise.

[0055] Thus, target compensation data for characterizing the additional heat dissipation intensity was obtained.

[0056] S4: Generate fan control commands and thermal management log records.

[0057] It should be noted that there are two types of fan speeds: a base fan speed determined by the environmental steady-state baseline, characterized by providing continuous weak circulation to ensure basic operating conditions; and a dynamic compensation speed calculated by an algorithm in an emergency, characterized by providing transient large-scale cooling to suppress sudden thermal runaway. These two sources of cooling coexist in the system's global thermal balance equations. Without a scientific algebraic reconstruction standard, the cooling module will face an output dilemma of mutual interplay between the base and emergency cooling states, potentially leading to frequent and violent oscillations between the two states. This invention uses the dynamic compensation speed as an emergency superposition factor to safely synthesize the base fan speed, thereby reconstructing a drive index that can characterize the global absolute control quantity, directly linking the extracted target parameter to the actual steady-state operation of the physical actuator.

[0058] Specifically, based on the principle of superposition of base speed feedforward and dynamic compensation, and combining the fan base speed and target compensation data, the final drive data is calculated, including: Perform an algebraic addition operation on the fan base speed and the target compensation data. When the sum of the two is less than the reconstruction saturation limit threshold of the full-load speed, output the sum directly. When the sum of the two exceeds the reconstruction saturation limit threshold, output according to the reconstruction saturation limit threshold of the full-load speed, and record it as the final drive data.

[0059] It should be noted that the target fan speed refers to the final absolute reconfiguration parameter sent to the drive module, representing the heat dissipation solution under the combined effects of the current physical environment and potential electrothermal hazards. Constructing the target fan speed aims to establish a globally reliable physical control index; the reconfiguration saturation limit threshold range involved here is set to be no higher than the full-load fan speed. to Preferred If the value is too large, long-term full load can easily lead to premature wear of the fan bearing; if it is too small, the cooling capacity cannot be fully released under extreme operating conditions.

[0060] Thus, the final driving data for establishing global reliable physical control indicators was obtained.

[0061] It's important to note that in the final execution and recording stage of the entire thermal management method, simply issuing commands blindly in one direction is insufficient to meet the stringent auditing requirements of specialized equipment. A high-speed cooling action that successfully intercepts thermal runaway cannot drive the hardware without underlying protocol coding; its characteristic is merely virtual numbers existing only at the software level. Simultaneously, another cooling event, though executed, is not recorded, and during subsequent troubleshooting, it is characterized by isolated data lacking timing proof and sequence number anchoring. These two deficiencies are common in traditional, extensive control systems. Without a dual-track standard of end-to-end physical drive and persistent archiving, the system's entire lifecycle maintenance will face a traceability dilemma with no verifiable evidence, making it impossible to assess the long-term reliability of the algorithm. From the perspective of digital twins for industrial IoT devices, the potential operational value of an end-to-end action with dual characteristics of microcomputer level conversion and asynchronous database writing far exceeds that of a single cooling event. Therefore, the target fan speed needs to be used as a core load factor, its duty cycle protocol needs to be encoded and persistently packaged in combination with the battery pack serial number, so as to generate a set of thermal management ledger index that can directly drive physical devices, and make the extracted ledger record directly associated with the safety traceability of the entire life cycle of the equipment.

[0062] Preferably, the final drive data is converted to generate fan control commands and thermal management ledger records, including: The final drive data is encoded using the microcomputer PWM protocol, transforming it into a duty cycle instruction sequence that can be directly recognized by the underlying heat dissipation module to drive the underlying heat dissipation module, generate fan control commands, and send them out for execution; the battery pack serial number obtained through the preceding external handshake is extracted, and a length of [missing information] is [missing information]. The battery pack serial number and duty cycle instruction sequence are treated as independent fields. The data clusters are packaged and encapsulated in strict accordance with the timestamp time sequence format to generate a thermal management ledger record with the current time sequence stamp and write it into the database.

[0063] It should be noted that the ultimate goal of the algorithm is to form the entire process of the physical link; in this invention, it can specifically employ a PWM microcomputer encoding and asynchronous database writing algorithm. Specifically, the byte sequence length of the battery pack serial number... The value range is set to to Preferred ,like Values ​​that are too long will invalidate and occupy persistent sector resources in memory; values ​​that are too short will fail to uniquely identify a large number of devices throughout their lifecycle, leading to ledger overwrite conflicts. Fan control commands refer to duty cycle level waveforms recognizable by the underlying hardware, representing the actual implementation of the control algorithm on the physical medium. Obtaining fan control commands aims to drive IGBT or MOSFET components to perform heat dissipation actions. Thermal management ledger records refer to relational data clusters containing timestamps, serial numbers, and action parameters, representing the control history of the entire system lifecycle. Constructing thermal management ledger records aims to provide an immutable data foundation for subsequent equipment maintenance and safety traceability. The database persistent synchronization cycle threshold range involved here is... to Preferred If the value is too long, the record of high-risk actions at the end of life may be lost when the system suddenly loses power; if it is too short, the lifespan of the storage medium will be rapidly exhausted due to frequent erasing and writing.

[0064] It should be noted that the complete process of this method can be illustrated by a typical working condition of fast charging of lithium batteries for a mining monorail crane: the battery pack has a total of 128 cell nodes, the temperature sampling period is 20ms, the maximum value of the initial temperature sequence in a certain period is 35.20℃, the maximum value of the real-time temperature sequence is 35.24℃, and the extreme temperature rise rate is... Bus charging current 200A, nominal internal resistance 0.5mΩ, basic heat dissipation power Historical steady-state current 100A, steady-state reference heating power Heating power standard deviation 2W; preset reference temperature rise rate 0.3℃ / s, scaling factor 0.5, temperature rise ratio Dynamic driving force Dangerous heat drop Risk ratio The probability of loss of control after exceeding the limit cutoff is 1; the fan operates at full load at 6000 rpm, with a base speed of 1800 rpm and a usable speed range of 4200 rpm. The target compensation speed is... Final speed After triggering 98% saturation limiting, it outputs 5880rpm, encoded as PWM duty cycle 98% and sent down, and is also written to the thermal management log.

[0065] It should be noted that, Figure 2This graph compares the temperature and fan speed during fast charging. The horizontal axis represents time, covering the entire process of temperature rise and heat dissipation control during fast charging. The upper vertical axis represents fan speed, and the lower vertical axis represents cell temperature. The smooth curve represents the thermal management method of this invention, the stepped transition curve represents the existing global temperature threshold mapping method, and the vertical dashed line represents the moment when the existing method triggers full-speed heat dissipation. This invention dynamically adjusts the fan speed in advance based on the temperature rise rate and the probability of runaway, with the speed smoothly increasing with the temperature rise, resulting in a lower peak cell temperature and faster cooling. The existing method maintains a low speed before the temperature reaches a fixed threshold, and jumps directly to full speed after triggering, resulting in a delayed response and a higher peak cell temperature and delayed cooling. This graph intuitively presents the core advantages of this invention: advance warning and dynamic smooth speed adjustment, solving the problem of delayed response in traditional threshold control. Figure 3 A bar chart comparing the performance indicators of different thermal management methods is presented. The horizontal axis represents three evaluation indicators: local extreme heat generation recognition rate, cooling response delay, and thermal runaway false alarm rate. Recognition rate and false alarm rate are expressed in percentage (%), and response delay is expressed in milliseconds (ms). Dark bars represent existing global temperature threshold mapping methods, while the diagonally lined bars represent the thermal management method of this invention. Regarding the local extreme heat generation recognition rate, the existing method achieves 76%, while this invention achieves 91%. For the cooling response delay, the existing method achieves 620ms, while this invention achieves 505ms. For the thermal runaway false alarm rate, the existing method achieves 8.5%, while this invention achieves 6.9%. This invention demonstrates a higher recognition rate, shorter response delay, and lower false alarm rate, reflecting the technical effectiveness of its extreme temperature rise recognition and probabilistic dynamic control. This allows for early detection of potential local thermal runaway hazards, improving the safety of the monorail crane's lithium battery fast charging process.

[0066] This completes the thermal management method applicable to the fast charging module of lithium battery for monorail cranes.

[0067] While various embodiments of the invention have been shown and described in this specification, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention.

Claims

1. A thermal management method for lithium battery fast charging modules of monorail cranes, characterized in that, include: S1: By activating the underlying sensor network and locking the global clock, synchronously read the cell thermal data and system operating condition data; S2: Based on the dynamic relationship of temperature change over time, and combining the real-time battery temperature sequence and the initial battery temperature sequence in the cell thermal data, calculate the temperature rise rate sequence; based on the Joule law heat source driving relationship, and combining the bus charging current and the battery corrected internal resistance in the system operating data, obtain extreme thermal data and heat generation benchmark data; based on the temperature rise rate and Joule thermal cross-validation characteristics, and combining the extreme thermal data and heat generation benchmark data, output dynamic driving force data; S3: Based on the scaling characteristics of the deviation from the safety baseline, combined with dynamic driving force data and heat generation benchmark data, calculate the runaway probability data; pre-obtain the fan full-load speed and fan base speed, perform physical work envelope margin feature calculation on both, and obtain performance boundary data; Perform scalar multiplication on the performance boundary data and the runaway probability data, and output the target compensation data; S4: Based on the principle of superposition of basic speed feedforward and dynamic compensation, and combined with the fan's basic speed and target compensation data, calculate the final drive data; The final drive data is converted to generate fan control commands and thermal management ledger records.

2. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 1, characterized in that, The synchronous reading of cell thermal data and system operating condition data includes: The bus listening module of the battery management system is activated to establish the physical topology boundary of the battery pack. Each independent cell node is traversed and read online. The absolute surface temperature of each node in the previous acquisition cycle is recorded as the initial battery temperature, and a battery initial temperature sequence is generated; the current acquisition cycle's temperature is simultaneously acquired online. The real-time absolute surface temperature of each independent cell node is recorded as the real-time battery temperature, and a real-time battery temperature sequence is generated. Then, the local high-precision hardware clock is accessed to parse the time difference between two valid acquisition actions, generate the temperature sampling duration, and package the initial battery temperature, real-time battery temperature, and temperature sampling duration into cell thermal data. The bus charging current and battery corrected internal resistance are pre-acquired and packaged into system operating condition data.

3. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 2, characterized in that, The calculated temperature rise rate sequence includes: The real-time battery temperature sequence and the initial battery temperature sequence are extracted from the cell thermal data. In the arithmetic logic unit, a vector difference operation is performed along the spatial dimension on the corresponding element positions of these two sequences to obtain the result containing… A discrete temperature difference sequence of battery cells is obtained; the temperature sampling duration of the preceding analysis is extracted; each element in the above battery cell temperature difference sequence is divided by the temperature sampling duration to perform algebraic division normalization operation; and then strictly reconstructed according to the original grid coordinate mapping to obtain the solution. The cell temperature rise rate sequence is calculated and the results are encapsulated as a temperature rise rate sequence.

4. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 1, characterized in that, The acquisition of extreme thermal data and heat production baseline data includes: Historical steady-state currents under fault-free conditions are extracted, and their heat generation fluctuation data are statistically analyzed to extract the variance of heat generation fluctuations. A temperature rise rate sequence is extracted, and a global traversal of the extreme value optimization based on a comparison register is performed in the numerical space to lock and extract the scalar element with the largest value in the sequence as the extreme thermal data. The bus charging current and the battery corrected internal resistance are parsed from the system operating condition data. The squared bus charging current is multiplied by the battery corrected internal resistance to calculate the base heat generation power. The squared historical steady-state current is multiplied by the battery corrected internal resistance to generate the steady-state reference heat generation power. The square root of the heat generation fluctuation variance is extracted to obtain the risk smoothing scale. The base heat generation power, the steady-state reference heat generation power, and the risk smoothing scale are packaged into the heat generation reference data.

5. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 1, characterized in that, The output dynamic driving force data includes: Extract the extreme thermal data and heat production baseline data generated in the previous step, parse out the basic heating power in the heat production baseline data, divide the extreme thermal data by the preset baseline temperature rise rate to obtain the dimensionless temperature rise ratio coefficient, multiply by the preset scaling factor and add 1, and then multiply by the basic heating power to obtain the dynamic thermal driving force scalar, which is denoted as dynamic driving force data.

6. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 1, characterized in that, The data used to calculate the probability of runaway includes: Extract the heat generation baseline data and analyze the steady-state baseline heat generation power and risk smoothing scale; perform algebraic subtraction operation between the dynamic driving force data and the steady-state baseline heat generation power to obtain the dangerous heat drop; perform proportional division scaling operation on the dangerous heat drop based on the risk smoothing scale, and truncate the final calculation result to the convergence interval of 0 to 1 to calculate the dimensionless runaway risk probability coefficient, which is denoted as the runaway probability data.

7. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 1, characterized in that, The obtained performance boundary data includes: Further read the configuration nameplate parameters of the heat dissipation drive control board, extract the fan full-load speed and fan base speed, and obtain the battery pack serial number through external handshake; in the instruction generator, subtract the fan base speed from the fan full-load speed, and the difference is the available speed range scalar. The available speed range scalar and the battery pack serial number are encapsulated into performance boundary data.

8. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 1, characterized in that, The output target compensation data includes: The available speed range scalar is parsed from the performance boundary data, and the previously generated runaway probability data is extracted; the available speed range scalar and the runaway probability data are multiplied by a scalar to calculate the target compensation data.

9. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 1, characterized in that, The final driving data for the calculation includes: Perform an algebraic addition operation on the fan base speed and the target compensation data. When the sum of the two is less than the reconstruction saturation limit threshold of the full-load speed, output the sum directly. When the sum of the two exceeds the reconstruction saturation limit threshold, output according to the reconstruction saturation limit threshold of the full-load speed, and record it as the final drive data.

10. The thermal management method for a lithium battery fast charging module for a monorail crane according to claim 7, characterized in that, The generation of fan control commands and thermal management ledger records includes: The final drive data is encoded using the microcomputer PWM protocol, transforming it into a duty cycle instruction sequence that can be directly recognized by the underlying heat dissipation module to drive the underlying heat dissipation module, generate fan control commands, and send them out for execution; the battery pack serial number obtained through the preceding external handshake is extracted, and a length of [missing information] is [missing information]. The battery pack serial number and duty cycle instruction sequence are treated as independent fields. The data clusters are packaged and encapsulated in strict accordance with the timestamp time sequence format to generate a thermal management ledger record with the current time sequence stamp and write it into the database.

Images