Fuel cell thermal management optimization method and system based on temperature distribution data

Abstract

The present invention discloses a fuel cell thermal management optimization method and system based on temperature distribution data, which relates to the field of fuel cell management technology. By analyzing the historical operation records of each fuel cell, power output and temperature change curves are constructed and aligned to form a historical curve control group; the control group is analyzed to identify the temperature stability segment, and the power output segment of the previous preset time is intercepted to construct a power-stable temperature relationship group; the real-time total power demand and the power output curve of each battery are obtained, multiple virtual power adjustments are performed, and an adjustment strategy that meets the demand is generated; the temperature performance of each strategy is evaluated using the historical relationship group, and the strategy with the highest evaluation value is selected to optimize the battery stack; the present invention predicts temperature performance through historical data, dynamically adjusts power distribution, optimizes temperature distribution, and improves fuel cell efficiency and life.

Description

Technical Field

[0001] The present invention relates to the technical field of fuel cell management, and in particular to a fuel cell thermal management optimization method and system based on temperature distribution data. Background Art

[0002] Fuel cells, as efficient and environmentally friendly power generation devices, are widely used in transportation, stationary power plants, and portable electronic devices. During operation, fuel cells generate heat through electrochemical reactions, and the temperature distribution within the cell stack has a significant impact on their performance, efficiency, and durability. Excessively high or uneven temperatures can lead to membrane material degradation, reduced catalyst activity, or thermal stress, while excessively low temperatures can reduce reaction efficiency. Therefore, effective thermal management is critical to maintaining optimal operating conditions. Traditional thermal management methods typically rely on static cooling systems or uniform power distribution strategies, which are difficult to adapt to dynamic changes in power demand and the resulting temperature fluctuations among individual cells. Some existing technologies attempt to monitor temperature in real time and adjust cooling parameters, but lack in-depth analysis of historical operating data and cannot accurately predict stable temperature conditions at a specific power output. Furthermore, these methods generally lack mechanisms to dynamically optimize the power output of individual fuel cells to achieve a balanced temperature distribution within the cell stack while meeting the overall power demand. Summary of the Invention

[0003] The object of the present invention is to provide a method and system capable of optimizing the thermal management of a fuel cell.

[0004] The present invention discloses a fuel cell thermal management optimization method based on temperature distribution data, comprising:

[0005] Step S100, analyzing the historical operation records of each fuel cell in the battery stack, constructing a historical battery power output curve and a historical temperature change curve, and aligning the historical battery power output curve and the historical temperature change curve to form a historical curve control group;

[0006] Step S200: Analyze the historical curve control group to determine a historical stable temperature curve segment in the historical temperature change curve, intercept a historical battery power output curve segment of a preset time length before the historical stable temperature curve segment, and correlate the historical stable temperature curve segment with the historical battery power output curve segment to obtain a historical battery power output-stable temperature relationship group for the fuel cell;

[0007] Step S300: obtaining a real-time total battery power demand curve and a real-time battery power output curve for each fuel cell, performing a plurality of virtual power output adjustments on each fuel cell in the battery stack, each adjustment being based on a standard of just satisfying the real-time total battery power demand curve, and obtaining a plurality of virtual power output adjustment strategies corresponding to the battery stack;

[0008] Step S400 , using the historical battery power output-stable temperature relationship group to evaluate the temperature performance of the virtual power output adjustment strategy, and selecting the virtual power output adjustment strategy with the highest evaluation value to adjust and optimize the battery stack.

[0009] In some embodiments disclosed herein, a method for analyzing a historical curve control group to determine a historical stable temperature curve segment in a historical temperature change curve includes:

[0010] Step S201: A determination time frame is set, and the determination time frame is gradually shifted in the historical temperature change curve. During each shift, a tortuosity parameter of the historical temperature change curve segment in the determination time frame is calculated. If the tortuosity parameter is less than or equal to a preset value, the historical temperature change curve segment in the determination time frame is marked as a historical stable temperature curve segment.

[0011] Among them, the method of calculating the tortuosity parameter includes:

[0012] Step S2011, calculating the average longitudinal value of the historical temperature change curve segment in the judgment time length frame, and constructing a parallel reference line parallel to the horizontal axis with the average longitudinal value as a reference;

[0013] Step S2012: uniformly set a number of comparison points for the parallel reference line, determine the first longitudinal value of the historical temperature change curve segment corresponding to each comparison point and the second longitudinal value of the parallel reference line, calculate the longitudinal value difference between the first longitudinal value and the second longitudinal value, and sort the longitudinal differences according to the order of the comparison points to obtain a longitudinal difference sequence;

[0014] Step S2013: Randomly combine the longitudinal difference quantities in the longitudinal difference quantity sequence in pairs to obtain a number of longitudinal difference quantity groups, calculate the quadratic difference quantities between the longitudinal difference quantities in the longitudinal difference quantity groups, and record the average value of the quadratic difference quantities as the tortuosity parameter.

[0015] In some embodiments disclosed herein, a method for performing a plurality of virtual power output adjustments on each fuel cell in a fuel cell stack includes:

[0016] In step S301, the real-time temperature of each fuel cell is obtained. If the real-time temperature value is greater than or equal to the optimal temperature value, the fuel cell is adjusted to reduce power output. If the real-time temperature value is less than the optimal temperature value, the fuel cell is adjusted to increase power output or the corresponding fuel cell is shut down.

[0017] In some embodiments disclosed herein, a method for evaluating the temperature performance of a virtual power output adjustment strategy using a historical battery power output-stable temperature relationship includes:

[0018] Step S401, determining a virtual battery power output curve of each fuel cell after the virtual power output adjustment strategy is adjusted;

[0019] Step S402 , performing a step-by-step comparison and analysis on the virtual battery power output curve, finding the most suitable one among several historical battery power output-stable temperature relationship groups, and marking the segments on the virtual battery power output curve;

[0020] Step S403: Determine stable temperature values ​​for different segments of the virtual battery power output curve based on the historical battery power output-stable temperature relationship groups marked on the segmented virtual battery power output curve, and determine sub-temperature performance evaluation values ​​for the virtual battery power output curve based on the temperature differences between the stable temperature values ​​and the optimal temperature values ​​for the different segments and their durations;

[0021] Step S404 : determining a temperature performance evaluation value of a virtual power output adjustment strategy based on the sub-temperature performance evaluation values ​​of all virtual battery power output curves.

[0022] In some embodiments disclosed herein, a method for finding the most suitable relationship among several historical battery power output-stable temperature relationship groups and marking the segments on a virtual battery power output curve includes:

[0023] Step S4021: Perform feature analysis on each historical battery power output curve segment in the historical battery power output-to-stable temperature relationship group to determine a historical characteristic parameter group for the historical battery power output curve segment. Using the same feature analysis method, perform a stepwise shift analysis on the virtual battery power output curve to determine a real-time characteristic parameter group in real time.

[0024] In step S4022, the real-time characteristic parameter group is compared with different historical characteristic parameter groups to select the most consistent historical characteristic parameter group. The historical battery power output curve segment corresponding to the historical characteristic parameter group is then overlapped and compared with the corresponding portion of the virtual battery power output curve corresponding to the real-time characteristic parameter group. If the overlap performance meets the preset standard, the corresponding historical battery power output-stable temperature relationship group is determined to be the most suitable historical battery power output-stable temperature relationship group.

[0025] In some embodiments disclosed herein, a method for performing feature analysis on a battery power output curve segment includes:

[0026] Step S40211: setting a feature analysis segment template for a historical battery power output curve segment, wherein the feature analysis segment template includes a plurality of feature analysis segments connected end to end;

[0027] Step S40212: align the feature analysis segment template and the battery power output curve segment, calculate the average power output value corresponding to each feature analysis segment, and determine the horizontal length ratio of the rising segment and the falling segment of the battery power output curve segment corresponding to the feature analysis segment, which is recorded as the rise-fall ratio. Combine the same average power output value and rise-fall ratio into a group to form a feature parameter group, and sort the feature parameter group according to the order of the feature analysis segments to obtain the feature parameter group.

[0028] In some embodiments disclosed herein, a method for performing overlap comparison on curve segments includes:

[0029] Step S40221, calculate the difference area between the two, if the difference area is less than or equal to a preset value, it is determined that the overlap performance meets the preset standard.

[0030] In some embodiments disclosed herein, based on the temperature difference between the stable temperature value and the optimal temperature value in different segments and the duration thereof, an expression for determining the sub-temperature performance evaluation value of the virtual battery power output curve is as follows:

[0031] Step S4031: setting a curve analysis duration, and using the curve analysis duration as the terminal analysis period of the virtual battery power output curve, analyzing the temperature difference and duration of different segments;

[0032] In step S4032, several temperature value difference intervals are set, each temperature value difference interval is set with a corresponding temperature performance parameter, the temperature value difference interval to which the average value of the temperature value difference of different segments belongs is determined, the temperature performance parameter of the segment is determined, and the duration of the segment is determined, the product of the temperature performance parameter and the duration is calculated, and the sub-temperature performance evaluation value is obtained.

[0033] In some embodiments disclosed herein, the expression for calculating the temperature performance evaluation of the virtual power output power adjustment strategy is:

[0034] ;

[0035] Where W is the corresponding value of temperature performance evaluation, is the temperature performance parameter of the i-th segment, is the duration of the i-th segment, and n is the number of segments;

[0036] Among them, based on The time proportion of the curve analysis time T is taken to determine the credibility of W, and based on the credibility, it is determined whether to adopt the temperature performance evaluation of the virtual power output power adjustment strategy.

[0037] In some embodiments disclosed herein, a fuel cell thermal management optimization system based on temperature distribution data includes:

[0038] The first module is used to analyze the historical operation records of each fuel cell in the battery stack, construct a historical battery power output curve and a historical temperature change curve, and align the historical battery power output curve and the historical temperature change curve to form a historical curve control group;

[0039] The second module is used to analyze the historical curve control group, determine the historical stable temperature curve segment in the historical temperature change curve, intercept the historical battery power output curve segment of a preset time length before the historical stable temperature curve segment, and associate the historical stable temperature curve segment with the historical battery power output curve segment to obtain the historical battery power output-stable temperature relationship group of the fuel cell;

[0040] The third module is used to obtain the real-time total battery power demand curve and the real-time battery power output curve of each fuel cell, and perform a number of virtual power output adjustments on each fuel cell in the battery stack, with each adjustment being based on the standard of just meeting the real-time total battery power demand curve, to obtain a number of virtual power output adjustment strategies for the corresponding battery stacks;

[0041] The fourth module is used to evaluate the temperature performance of the virtual power output adjustment strategy using the historical battery power output-stable temperature relationship group, and select the virtual power output adjustment strategy with the highest evaluation value to adjust and optimize the battery stack.

[0042] The present invention discloses a fuel cell thermal management optimization method and system based on temperature distribution data, which relates to the field of fuel cell management technology. By analyzing the historical operation records of each fuel cell, power output and temperature change curves are constructed and aligned to form a historical curve control group; the control group is analyzed to identify the temperature stability segment, and the power output segment of the previous preset time is intercepted to construct a power-stable temperature relationship group; the real-time total power demand and the power output curve of each battery are obtained, multiple virtual power adjustments are made, and an adjustment strategy that meets the demand is generated; the temperature performance of each strategy is evaluated using the historical relationship group, and the strategy with the highest evaluation value is selected to optimize the battery stack; the present invention predicts temperature performance through historical data, dynamically adjusts power distribution, optimizes temperature distribution, and improves fuel cell efficiency and life.

[0043] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a method step diagram of the fuel cell thermal management optimization method based on temperature distribution data disclosed in the present invention. DETAILED DESCRIPTION

[0045] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0046] The following will be combined with the accompanying drawings and specific embodiments to clearly and completely describe the technical solutions of the present invention. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and cannot be understood as limiting the scope of protection of the present invention. Those skilled in the art in this field can make some non-essential improvements and adjustments based on the content of the present invention described below. In the present invention, unless otherwise clearly specified and limited, the technical terms used in the present invention should have the common meanings understood by those skilled in the art of the present invention.

[0047] Example:

[0048] The present invention discloses a fuel cell thermal management optimization method based on temperature distribution data, comprising:

[0049] Step S100 , analyzing the historical operation records of each fuel cell in the battery stack, constructing a historical battery power output curve and a historical temperature change curve, and aligning the historical battery power output curve and the historical temperature change curve to form a historical curve control group.

[0050] Step S100 analyzes the historical operating records of each fuel cell in the fuel cell stack to construct a historical cell power output curve and a historical temperature change curve. These curves are then time-aligned to form a control group of historical curves, laying the foundation for subsequent analysis. Fuel cell operating data includes power output and temperature changes, which reflect the cell's behavioral characteristics under different operating conditions. By extracting these historical records and plotting them as curves, the dynamic relationship between power and temperature can be visually observed. This alignment process ensures that the power and temperature data are synchronized on the timeline, eliminating errors caused by recording deviations. For example, suppose a fuel cell has recorded minute-by-minute power output (e.g., 100W to 500W) and corresponding temperature (e.g., 60°C to 80°C) over the past month of operation. Step S100 organizes these data into two curves, ensuring consistent timestamps, to form a control group. This control group provides a reliable data foundation for subsequent exploration of the stable power-temperature relationship, avoiding interactions that may be overlooked in single-dimensional analysis.

[0051] In step S200, the historical curve control group is analyzed to determine the historical stable temperature curve segment in the historical temperature change curve, and the historical battery power output curve segment of a preset time length before the historical stable temperature curve segment is intercepted, and the historical stable temperature curve segment and the historical battery power output curve segment are correlated to obtain the historical battery power output-stable temperature relationship group of the fuel cell.

[0052] Step S200 analyzes the historical curve control group to identify stable temperature segments within the historical temperature change curves. These segments are then correlated with corresponding historical power output segments to construct a set of historical fuel cell power output-stable temperature relationships. This process aims to identify operating states with stable temperatures, as stable temperatures generally indicate efficient and safe battery operation. By setting a timeframe (e.g., 5 minutes) and calculating a zigzag parameter (a measure of temperature curve fluctuations), segments with minimal temperature variation can be identified.

[0053] Step S300, obtain the real-time total battery power demand curve, and obtain the real-time battery power output curve of each fuel cell, perform several virtual power output adjustments on each fuel cell of the battery stack, and each adjustment is based on the standard of just meeting the real-time total battery power demand curve, to obtain several virtual power output adjustment strategies for the corresponding battery stacks.

[0054] Step S300 utilizes a real-time total battery power demand curve and the real-time power output curves of each fuel cell to generate multiple virtual power adjustments to meet the total power demand. This process simulates how the power output of each battery cell is allocated to optimize overall performance during actual operation. The real-time total power demand may vary depending on the application scenario (such as acceleration in an electric vehicle). For example, a total power demand of 1000W may be required, but a battery stack has five fuel cells, each with different real-time power and temperature. Step S300 generates multiple allocation combinations through virtual adjustments (e.g., adjusting the power of battery 1 from 200W to 250W, battery 2 from 300W to 280W, etc.), ensuring that the total power meets 1000W with each adjustment. The adjustments take real-time temperature into account. For example, if a battery cell's temperature is too high (above the optimal temperature of 80°C), its power output may be preferentially reduced. This virtual strategy provides a diverse set of candidate solutions for subsequent temperature evaluation, ensuring that power requirements are met while exploring temperature optimization options.

[0055] Step S400 , using the historical battery power output-stable temperature relationship group to evaluate the temperature performance of the virtual power output adjustment strategy, and selecting the virtual power output adjustment strategy with the highest evaluation value to adjust and optimize the battery stack.

[0056] The principle of step S400 is to use the historical battery power output ~ stable temperature relationship group to evaluate the temperature performance of the virtual power adjustment strategy generated in step S300, and select the strategy with the highest evaluation value to optimize the battery stack. This process predicts the temperature performance of each battery under each virtual strategy, evaluates whether it is close to the optimal temperature (such as 80°C), and thus selects the optimal strategy. For example, after a virtual strategy is adjusted, the virtual power output curve of battery 1 matches the "300W~70°C" relationship group in the historical data, and battery 2 matches "350W~75°C". By comparing the differences between these predicted temperatures and the optimal temperature (for example, 70°C and 80°C differ by 10°C for 5 minutes), combined with the duration, the sub-temperature performance evaluation value is calculated (such as the smaller the difference and the longer the duration, the higher the evaluation value). The sub-evaluation values ​​of all batteries are aggregated to form the overall temperature performance evaluation value of the strategy.

[0057] In some embodiments disclosed herein, a method for analyzing a historical curve control group to determine a historical stable temperature curve segment in a historical temperature change curve includes:

[0058] In step S201, a judgment time length box is set, and the judgment time length box is gradually moved in the historical temperature change curve. During each movement, the tortuosity parameter of the historical temperature change curve segment in the judgment time length box is calculated. If the tortuosity parameter is less than or equal to the preset value, the historical temperature change curve segment in the judgment time length box is marked and recorded as a historical stable temperature curve segment.

[0059] Among them, the method of calculating the tortuosity parameter includes:

[0060] Step S2011, calculating the average longitudinal value of the historical temperature change curve segment in the judgment time length frame, and constructing a parallel reference line parallel to the horizontal axis with the average longitudinal value as a reference;

[0061] Step S2012: uniformly set a number of comparison points for the parallel reference line, determine the first longitudinal value of the historical temperature change curve segment corresponding to each comparison point and the second longitudinal value of the parallel reference line, calculate the longitudinal value difference between the first longitudinal value and the second longitudinal value, and sort the longitudinal differences according to the order of the comparison points to obtain a longitudinal difference sequence;

[0062] Step S2013: Randomly combine the longitudinal difference quantities in the longitudinal difference quantity sequence in pairs to obtain a number of longitudinal difference quantity groups, calculate the quadratic difference quantities between the longitudinal difference quantities in the longitudinal difference quantity groups, and record the average value of the quadratic difference quantities as the tortuosity parameter.

[0063] In some embodiments disclosed herein, a method for performing a plurality of virtual power output adjustments on each fuel cell in a fuel cell stack includes:

[0064] In step S301, the real-time temperature of each fuel cell is obtained. If the real-time temperature value is greater than or equal to the optimal temperature value, the fuel cell is adjusted to reduce power output. If the real-time temperature value is less than the optimal temperature value, the fuel cell is adjusted to increase power output or the corresponding fuel cell is shut down.

[0065] In some embodiments disclosed herein, a method for evaluating the temperature performance of a virtual power output adjustment strategy using a historical battery power output-stable temperature relationship includes:

[0066] Step S401 : determining a virtual battery power output curve of each fuel cell after the virtual power output adjustment strategy is adjusted.

[0067] Step S401 calculates the adjusted virtual power output curve for each fuel cell based on the virtual power output adjustment strategy generated in step S300, providing baseline data for subsequent temperature performance evaluation. The virtual power adjustment strategy simulates the power output distribution of each fuel cell in the stack to meet the real-time total power demand. Step S401 simulates the adjustment and generates a time-varying curve of the power output of each cell under the adjustment strategy. For example, suppose a strategy requires a total power output of 1000W for a stack consisting of five fuel cells. After adjustment, the power of cell 1 changes from 200W to 250W, cell 2 from 300W to 280W, and so on. By recording the changes in these adjusted power values ​​over time (e.g., once per second), a virtual power output curve is generated. For example, the curve for cell 1 might display [250W, 250W, 260W, ...]. These curves reflect the impact of the adjustment strategy on the operating status of each cell, providing a specific power output pattern for subsequent comparison with historical data.

[0068] Step S402 : performing step-by-step comparison and analysis on the virtual battery power output curve, finding the most suitable one among several historical battery power output-stable temperature relationship groups, and marking the segments on the virtual battery power output curve.

[0069] Step S402 works by progressively comparing the virtual battery power output curve with the power curves in the historical battery power output-stable temperature relationship group to identify the most suitable historical relationship group. The corresponding segments are then marked on the virtual curve to associate the stable temperature information. This process leverages the established power-temperature relationship in historical data to predict the temperature performance under the virtual power output. For example, suppose the virtual power output curve shows a smooth power transition from 250W to 260W over a 5-minute period. By analyzing characteristics (such as average power and trend), the best matching power curve segment is found within the historical relationship group (e.g., 250W-255W corresponds to a stable temperature of 70°C in the historical data). If the matching degree meets a preset standard (e.g., 90% curve overlap), the segment of the virtual curve is marked as corresponding to a stable temperature of 70°C. This segmentation marking process decomposes the virtual curve into multiple sub-segments associated with historical stable temperatures, providing structured data support for subsequent temperature evaluation.

[0070] Step S403 determines the stable temperature values ​​for different segments of the virtual battery power output curve based on the historical battery power output-stable temperature relationship groups marked on the segmented virtual battery power output curve. The sub-temperature performance evaluation values ​​for the virtual battery power output curve are determined based on the temperature difference between the stable temperature values ​​for the different segments and the optimal temperature value, as well as the duration.

[0071] Step S404 : determining a temperature performance evaluation value of a virtual power output adjustment strategy based on the sub-temperature performance evaluation values ​​of all virtual battery power output curves.

[0072] In some embodiments disclosed herein, a method for finding the most suitable relationship among several historical battery power output-stable temperature relationship groups and marking the segments on a virtual battery power output curve includes:

[0073] In step S4021, a characteristic analysis is performed on each historical battery power output curve segment in the historical battery power output-to-stable temperature relationship group to determine a historical characteristic parameter group of the historical battery power output curve segment. Using the same characteristic analysis method, a step-by-step analysis is performed on the virtual battery power output curve to determine a real-time characteristic parameter group in real time.

[0074] In step S4022, the real-time characteristic parameter group is compared with different historical characteristic parameter groups to select the most consistent historical characteristic parameter group. The historical battery power output curve segment corresponding to the historical characteristic parameter group is then overlapped and compared with the corresponding portion of the virtual battery power output curve corresponding to the real-time characteristic parameter group. If the overlap performance meets the preset standard, the corresponding historical battery power output-stable temperature relationship group is determined to be the most suitable historical battery power output-stable temperature relationship group.

[0075] In some embodiments disclosed herein, a method for performing feature analysis on a battery power output curve segment includes:

[0076] In step S40211 , a feature analysis segment template is set for the historical battery power output curve segment. The feature analysis segment template includes a plurality of feature analysis segments connected end to end.

[0077] The principle behind step S40211 is to set a feature analysis segment template for a historical battery power output curve segment. This template consists of several end-to-end feature analysis segments, used to standardize the power curve feature analysis. This template decomposes complex and variable power curve segments into sub-segments with fixed durations or structures, allowing for the extraction of consistent feature parameters and providing a unified framework for subsequent comparisons. For example, to analyze a 5-minute historical power output curve segment, a feature analysis segment template can be set consisting of five 1-minute segments (i.e., [0-1min, 1-2min, 2-3min, 3-4min, 4-5min]), connected end-to-end to cover the entire curve segment. The template design ensures that each segment has a consistent duration, facilitating the capture of localized characteristics of power variations, such as rising, falling, or plateauing trends. This standardized decomposition provides a structured analysis foundation for subsequent feature extraction and is applicable to power curve segments of varying lengths and shapes.

[0078] Step S40212: align the feature analysis segment template and the battery power output curve segment, calculate the average power output value corresponding to each feature analysis segment, and determine the horizontal length ratio of the rising segment and the falling segment of the battery power output curve segment corresponding to the feature analysis segment, which is recorded as the rise-fall ratio. Combine the same average power output value and rise-fall ratio into a group to form a feature parameter group, and sort the feature parameter group according to the order of the feature analysis segments to obtain the feature parameter group.

[0079] Step S40212 aligns the characteristic analysis segment template with the battery power output curve segment, calculates the average power output value and the rise-to-fall ratio (i.e., the ratio of the horizontal length of the rising segment to the falling segment) for each characteristic analysis segment, and combines these two into a characteristic parameter group. These parameters are then sorted by segment order to form a characteristic parameter group, which is used to quantify the characteristics of the curve segment. For example, suppose a 5-minute historical power output curve segment (with the power value sequence [250W, 260W, 260W, 255W, 250W]) is aligned with a template consisting of five 1-minute segments. The average power of each segment is calculated: the average power of the first segment (250W-260W) is 255W, the average power of the second segment (260W) is 260W, and so on. The rise-fall ratio of each segment is also analyzed. For example, in segment 1 (horizontally extending for 1 minute), the power rises with no fall, resulting in a 1:0 ratio. In segment 4 (255W-250W, horizontally extending for 1 minute), the power falls with no rise, resulting in a 0:1 ratio. The average power and rise-fall ratio of each segment are grouped into characteristic parameters, such as [(255W, 1:0), (260W, 0:0), …], which are then sorted by segment order to form characteristic parameter groups. This process quantifies power levels and trends, generating parameter sets that describe the characteristics of the curve segments. This provides an accurate basis for subsequent feature matching with the virtual power curve.

[0080] In some embodiments disclosed herein, a method for performing overlap comparison on curve segments includes:

[0081] Step S40221, calculate the difference area between the two, if the difference area is less than or equal to a preset value, it is determined that the overlap performance meets the preset standard.

[0082] In some embodiments disclosed herein, based on the temperature difference between the stable temperature value and the optimal temperature value in different segments and the duration thereof, an expression for determining the sub-temperature performance evaluation value of the virtual battery power output curve is as follows:

[0083] In step S4031 , a curve analysis duration is set, and the curve analysis duration is used as the terminal analysis period of the virtual battery power output curve to analyze the temperature difference and duration of different segments.

[0084] In step S4032, several temperature value difference intervals are set, each temperature value difference interval is set with a corresponding temperature performance parameter, the temperature value difference interval to which the average value of the temperature value difference of different segments belongs is determined, the temperature performance parameter of the segment is determined, and the duration of the segment is determined, the product of the temperature performance parameter and the duration is calculated, and the sub-temperature performance evaluation value is obtained.

[0085] In some embodiments disclosed herein, the expression for calculating the temperature performance evaluation of the virtual power output power adjustment strategy is:

[0086] .

[0087] Where W is the corresponding value of temperature performance evaluation, is the temperature performance parameter of the i-th segment, is the duration of the i-th segment, and n is the number of segments;

[0088] Among them, based on The time proportion of the curve analysis time T is taken to determine the credibility of W, and based on the credibility, it is determined whether to adopt the temperature performance evaluation of the virtual power output power adjustment strategy.

[0089] In some embodiments disclosed herein, a fuel cell thermal management optimization system based on temperature distribution data includes:

[0090] The first module is used to analyze the historical operation records of each fuel cell in the battery stack, construct a historical battery power output curve and a historical temperature change curve, align the historical battery power output curve and the historical temperature change curve to form a historical curve control group.

[0091] The second module is used to analyze the historical curve control group, determine the historical stable temperature curve segment in the historical temperature change curve, and intercept the historical battery power output curve segment of a preset time length before the historical stable temperature curve segment, and associate the historical stable temperature curve segment with the historical battery power output curve segment to obtain the historical battery power output~stable temperature relationship group of the fuel cell.

[0092] The third module is used to obtain the real-time total battery power demand curve and the real-time battery power output curve of each fuel cell, and to perform several virtual power output adjustments on each fuel cell of the battery stack. Each adjustment is based on the standard of just meeting the real-time total battery power demand curve, and obtains several virtual power output adjustment strategies for the corresponding battery stacks.

[0093] The fourth module is used to evaluate the temperature performance of the virtual power output adjustment strategy using the historical battery power output-stable temperature relationship group, and select the virtual power output adjustment strategy with the highest evaluation value to adjust and optimize the battery stack.

[0094] The present invention discloses a fuel cell thermal management optimization method and system based on temperature distribution data, which relates to the field of fuel cell management technology. First, by analyzing the historical operation records of each fuel cell, power output and temperature change curves are constructed and aligned to form a historical curve control group; the control group is analyzed to identify the temperature stability segment, and the power output segment of the previous preset time is intercepted to construct a power-stable temperature relationship group; the real-time total power demand and the power output curve of each battery are obtained, multiple virtual power adjustments are performed, and an adjustment strategy that meets the demand is generated; the temperature performance of each strategy is evaluated using the historical relationship group, and the strategy with the highest evaluation value is selected to optimize the battery stack; the present invention predicts temperature performance through historical data, dynamically adjusts power distribution, optimizes temperature distribution, and improves fuel cell efficiency and life.

[0095] Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented via hardware or via software combined with a necessary general-purpose hardware platform. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product. This software product can be stored on a non-volatile storage medium (such as a CD-ROM, USB flash drive, or external hard drive) and includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute the methods described in various implementation scenarios of the present invention.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A fuel cell thermal management optimization method based on temperature distribution data, characterized in that: include: Step S100, analyzing the historical operation records of each fuel cell in the battery stack, constructing a historical battery power output curve and a historical temperature change curve, and aligning the historical battery power output curve and the historical temperature change curve to form a historical curve control group; Step S200: Analyze the historical curve control group to determine a historical stable temperature curve segment in the historical temperature change curve, intercept a historical battery power output curve segment of a preset time length before the historical stable temperature curve segment, and correlate the historical stable temperature curve segment with the historical battery power output curve segment to obtain a historical battery power output-stable temperature relationship group for the fuel cell; Step S300: obtaining a real-time total battery power demand curve and a real-time battery power output curve for each fuel cell, performing a plurality of virtual power output adjustments on each fuel cell in the battery stack, each adjustment being based on a standard of just satisfying the real-time total battery power demand curve, and obtaining a plurality of virtual power output adjustment strategies corresponding to the battery stack; Step S400 , using the historical battery power output-stable temperature relationship group to evaluate the temperature performance of the virtual power output adjustment strategy, and selecting the virtual power output adjustment strategy with the highest evaluation value to adjust and optimize the battery stack.

2. The fuel cell thermal management optimization method based on temperature distribution data according to claim 1, characterized in that: The method of analyzing the historical curve control group and determining the historical stable temperature curve segment in the historical temperature change curve includes: Step S201: A determination time frame is set, and the determination time frame is gradually shifted in the historical temperature change curve. During each shift, a tortuosity parameter of the historical temperature change curve segment in the determination time frame is calculated. If the tortuosity parameter is less than or equal to a preset value, the historical temperature change curve segment in the determination time frame is marked as a historical stable temperature curve segment. Among them, the method of calculating the tortuosity parameter includes: Step S2011, calculating the average longitudinal value of the historical temperature change curve segment in the judgment time length frame, and constructing a parallel reference line parallel to the horizontal axis with the average longitudinal value as a reference; Step S2012: uniformly set a number of comparison points for the parallel reference line, determine the first longitudinal value of the historical temperature change curve segment corresponding to each comparison point and the second longitudinal value of the parallel reference line, calculate the longitudinal value difference between the first longitudinal value and the second longitudinal value, and sort the longitudinal differences according to the order of the comparison points to obtain a longitudinal difference sequence; Step S2013: Randomly combine the longitudinal difference quantities in the longitudinal difference quantity sequence in pairs to obtain a number of longitudinal difference quantity groups, calculate the quadratic difference quantities between the longitudinal difference quantities in the longitudinal difference quantity groups, and record the average value of the quadratic difference quantities as the tortuosity parameter.

3. The fuel cell thermal management optimization method based on temperature distribution data according to claim 1, characterized in that: The method of performing a plurality of virtual power output adjustments on each fuel cell of the fuel cell stack includes: In step S301, the real-time temperature of each fuel cell is obtained. If the real-time temperature value is greater than or equal to the optimal temperature value, the fuel cell is adjusted to reduce power output. If the real-time temperature value is less than the optimal temperature value, the fuel cell is adjusted to increase power output or the corresponding fuel cell is shut down.

4. The fuel cell thermal management optimization method based on temperature distribution data according to claim 1, characterized in that: Methods for evaluating the temperature performance of a virtual power output adjustment strategy using a historical battery power output-stable temperature relationship group include: Step S401, determining a virtual battery power output curve of each fuel cell after the virtual power output adjustment strategy is adjusted; Step S402 , performing a step-by-step comparison and analysis on the virtual battery power output curve, finding the most suitable one among several historical battery power output-stable temperature relationship groups, and marking the segments on the virtual battery power output curve; Step S403: Determine stable temperature values ​​for different segments of the virtual battery power output curve based on the historical battery power output-stable temperature relationship groups marked on the segmented virtual battery power output curve, and determine sub-temperature performance evaluation values ​​for the virtual battery power output curve based on the temperature differences between the stable temperature values ​​and the optimal temperature values ​​for the different segments and their durations; Step S404 : determining a temperature performance evaluation value of a virtual power output adjustment strategy based on the sub-temperature performance evaluation values ​​of all virtual battery power output curves.

5. The fuel cell thermal management optimization method based on temperature distribution data according to claim 4, characterized in that: The method of finding the most suitable one among several historical battery power output-stable temperature relationship groups and marking the segments on the virtual battery power output curve includes: Step S4021: Perform feature analysis on each historical battery power output curve segment in the historical battery power output-to-stable temperature relationship group to determine a historical characteristic parameter group for the historical battery power output curve segment. Using the same feature analysis method, perform a stepwise shift analysis on the virtual battery power output curve to determine a real-time characteristic parameter group in real time. In step S4022, the real-time characteristic parameter group is compared with different historical characteristic parameter groups to select the most consistent historical characteristic parameter group. The historical battery power output curve segment corresponding to the historical characteristic parameter group is then overlapped and compared with the corresponding portion of the virtual battery power output curve corresponding to the real-time characteristic parameter group. If the overlap performance meets the preset standard, the corresponding historical battery power output-stable temperature relationship group is determined to be the most suitable historical battery power output-stable temperature relationship group.

6. The fuel cell thermal management optimization method based on temperature distribution data according to claim 5, characterized in that: Methods for performing characteristic analysis on battery power output curve segments include: Step S40211: setting a feature analysis segment template for a historical battery power output curve segment, wherein the feature analysis segment template includes a plurality of feature analysis segments connected end to end; Step S40212: align the feature analysis segment template and the battery power output curve segment, calculate the average power output value corresponding to each feature analysis segment, and determine the horizontal length ratio of the rising segment and the falling segment of the battery power output curve segment corresponding to the feature analysis segment, which is recorded as the rise-fall ratio. Combine the same average power output value and rise-fall ratio into a group to form a feature parameter group, and sort the feature parameter group according to the order of the feature analysis segments to obtain the feature parameter group.

7. The fuel cell thermal management optimization method based on temperature distribution data according to claim 5, characterized in that: Methods for aligning and comparing curve segments include: Step S40221, calculate the difference area between the two, if the difference area is less than or equal to a preset value, it is determined that the overlap performance meets the preset standard.

8. The fuel cell thermal management optimization method based on temperature distribution data according to claim 4, characterized in that: Based on the temperature difference between the stable temperature value and the optimal temperature value of different segments and the duration, the expression for determining the sub-temperature performance evaluation value of the virtual battery power output curve is: Step S4031: setting a curve analysis duration, and using the curve analysis duration as the terminal analysis period of the virtual battery power output curve, analyzing the temperature difference and duration of different segments; In step S4032, several temperature value difference intervals are set, each temperature value difference interval is set with a corresponding temperature performance parameter, the temperature value difference interval to which the average value of the temperature value difference of different segments belongs is determined, the temperature performance parameter of the segment is determined, and the duration of the segment is determined, the product of the temperature performance parameter and the duration is calculated, and the sub-temperature performance evaluation value is obtained.

9. The fuel cell thermal management optimization method based on temperature distribution data according to claim 8, characterized in that: The expression for calculating the temperature performance evaluation of the virtual power output power adjustment strategy is: ； Where W is the corresponding value of temperature performance evaluation, is the temperature performance parameter of the i-th segment, is the duration of the i-th segment, and n is the number of segments; Among them, based on The time proportion of the curve analysis time T is taken to determine the credibility of W, and based on the credibility, it is determined whether to adopt the temperature performance evaluation of the virtual power output power adjustment strategy.

10. A fuel cell thermal management optimization system based on temperature distribution data, characterized in that: include: The first module is used to analyze the historical operation records of each fuel cell in the battery stack, construct a historical battery power output curve and a historical temperature change curve, and align the historical battery power output curve and the historical temperature change curve to form a historical curve control group; The second module is used to analyze the historical curve control group, determine the historical stable temperature curve segment in the historical temperature change curve, intercept the historical battery power output curve segment of a preset time length before the historical stable temperature curve segment, and associate the historical stable temperature curve segment with the historical battery power output curve segment to obtain the historical battery power output-stable temperature relationship group of the fuel cell; The third module is used to obtain the real-time total battery power demand curve and the real-time battery power output curve of each fuel cell, and perform a number of virtual power output adjustments on each fuel cell in the battery stack, with each adjustment being based on the standard of just meeting the real-time total battery power demand curve, to obtain a number of virtual power output adjustment strategies for the corresponding battery stacks; The fourth module is used to evaluate the temperature performance of the virtual power output adjustment strategy using the historical battery power output-stable temperature relationship group, and select the virtual power output adjustment strategy with the highest evaluation value to adjust and optimize the battery stack.

Images