Battery cabin fan failure early warning method, device, equipment, medium and program product

Abstract

This application provides a method, apparatus, device, medium, and program product for early warning of battery compartment fan failure. The method includes: acquiring the physical locations of multiple fans within the battery compartment, and the time-series operating data of the fans; the fans are used to dissipate heat from the battery compartment; based on the physical locations and the time-series operating data, determining the degree of deviation between the fan and the characteristics of a fan group using a spatiotemporal clustering algorithm; the abnormal probability of the fan is positively correlated with the degree of deviation; based on the time-series operating data, determining the initial abnormality judgment result of the fan using a time-series analysis model; and providing a failure warning for the fan based on the degree of deviation and the initial abnormality judgment result. This method is used to improve the accuracy of early warning of battery compartment fan failure.

Description

Technical Field

[0001] This application relates to the field of energy storage systems, and in particular to a method, device, equipment, medium, and program product for early warning of battery compartment fan failure. Background Technology

[0002] During the operation of an energy storage system, temperature management within the battery compartment is a core element in ensuring the system's safety and stability. As the actuator of the thermal management system, the operating status of the fan directly affects heat dissipation efficiency. If the fan malfunctions, it may cause an abnormal rise in the battery compartment temperature, potentially leading to thermal runaway.

[0003] Current technologies primarily rely on temperature monitoring for fan fault diagnosis, deploying numerous temperature sensors within the battery compartment and combining these sensors with temperature fluctuation characteristics to determine if the fan is malfunctioning. However, this method is susceptible to ambient temperature fluctuations, leading to missed or incorrect diagnoses.

[0004] Therefore, improving the accuracy of battery compartment fan failure early warning is an urgent problem to be solved. Summary of the Invention

[0005] This application provides a method, apparatus, device, medium, and program product for early warning of battery compartment fan failure, in order to improve the accuracy of early warning of battery compartment fan failure.

[0006] In a first aspect, embodiments of this application provide a method for early warning of battery compartment fan failure, including:

[0007] The physical locations of multiple fans within the battery compartment, and the timing operation data of the fans, are obtained; the fans are used to dissipate heat from the battery compartment.

[0008] Based on the physical location and the time-series operation data, a spatiotemporal clustering algorithm is used to determine the degree of deviation between the fan and the characteristics of the fan group; the abnormal probability of the fan is positively correlated with the degree of deviation.

[0009] Based on the aforementioned time-series operation data, the initial anomaly judgment result of the fan is determined through a time-series analysis model;

[0010] Based on the degree of deviation and the initial anomaly judgment result, a fault warning is issued for the fan.

[0011] In one possible implementation, the time-series operating data of the fan includes: the fan's operating time; the physical location is represented by two-dimensional coordinates; and based on the physical location, the time-series operating data is used to determine the degree of deviation between the fan and the characteristics of the fan group through a spatiotemporal clustering algorithm, including:

[0012] Based on the physical location and the runtime, construct the three-dimensional features of the fan;

[0013] Based on the three-dimensional features of the multiple fans, an adaptive parameter search algorithm is used to determine the cluster radius and the threshold number of fans included in each cluster.

[0014] The spatiotemporal clustering algorithm determines the characteristics of the fan group based on the cluster radius and the fan number threshold. The fan group characteristics include the cluster center of the fan group. The clustering constraints of the spatiotemporal clustering algorithm include at least one of the following: physical spacing constraints between the multiple fans, runtime difference constraints between the multiple fans, and synchronization rate constraints on whether the multiple fans are turned on or off synchronously.

[0015] Based on the physical location, the runtime, and the cluster center, the target Mahalanobis distance between the fan and the cluster center is obtained. The target Mahalanobis distance is used to characterize the degree of deviation between the fan and the characteristics of the fan group.

[0016] In one possible implementation, obtaining the target Mahalanobis distance between the fan and the cluster center based on the physical location, the runtime, and the cluster center includes:

[0017] Based on the physical location and the physical location of the cluster center, the first Mahalanobis distance between the fan and the cluster center is obtained;

[0018] Based on the runtime and the runtime of the cluster center, the second Mahalanobis distance between the fan and the cluster center is obtained;

[0019] The target Mahalanobis distance is obtained based on the first Mahalanobis distance and the second Mahalanobis distance.

[0020] In one possible implementation, the time-series operational data includes: the fan's start / stop signal and the fan's runtime; the time-series analysis model includes: a sliding window dynamic time warping (DTW) model and a long short-term memory network-cumulative sum (LSTM-CUSUM) model; the step of determining the initial anomaly judgment result of the fan based on the time-series operational data and the time-series analysis model includes:

[0021] Based on the start and stop signals of the fan, the degree of irregularity in the start and stop of the fan is determined by the DTW model.

[0022] Based on the fan's runtime, the cumulative residual sum between the predicted runtime and the actual runtime is determined using the LSTM-CUSUM model.

[0023] Based on the judgment result of the irregularity of start-stop and the cumulative sum of residuals, the initial abnormality judgment result of the fan is obtained.

[0024] In one possible implementation, the step of providing a fault warning for the fan based on the degree of deviation and the initial anomaly judgment result includes:

[0025] Based on the degree of deviation and the weighted sum of the initial anomaly judgment results, the health index of the fan is determined;

[0026] If the health index is less than the first threshold, a level one fault warning signal is output.

[0027] If the health index is greater than or equal to the first threshold and less than the second threshold, a level two fault warning signal is output.

[0028] If the health index is greater than or equal to the second threshold, a level 3 fault warning signal is output; the probability of fan failure is positively correlated with the level of the fault warning signal.

[0029] In one possible implementation, the fault warning is used to characterize a fan malfunction. After issuing a fault warning to the fan based on the degree of deviation and the initial anomaly judgment result, the method further includes:

[0030] If the running time of the fan increases over multiple consecutive historical running time units, and the fluctuation rate of the running time over the multiple consecutive historical running time units is less than or equal to a preset fluctuation rate, the fault type of the fan is determined to be a mechanical fault.

[0031] If the fan's runtime decrease rate during the historical unit operating time is greater than or equal to a preset decrease rate, or if the fan's runtime is zero, the fault type of the fan is determined to be an electrical fault.

[0032] If the running time of the fan in multiple consecutive historical unit running times all have a target difference from the preset time, the fault type of the fan is determined to be a control fault.

[0033] Secondly, embodiments of this application provide a battery compartment fan malfunction early warning device, comprising:

[0034] The acquisition module is used to acquire the physical locations of multiple fans inside the battery compartment, as well as the timing operation data of the fans; the fans are used to dissipate heat from the battery compartment.

[0035] The first determining module is used to determine the degree of deviation between the fan and the characteristics of the fan group based on the physical location and the time-series operation data, using a spatiotemporal clustering algorithm; the abnormal probability of the fan is positively correlated with the degree of deviation.

[0036] The second determining module is used to determine the initial abnormality judgment result of the fan based on the time-series operation data and through a time-series analysis model.

[0037] The early warning module is used to provide a fault warning for the fan based on the degree of deviation and the initial anomaly judgment result.

[0038] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0039] The memory stores computer-executed instructions;

[0040] The processor executes computer execution instructions stored in the memory, causing the processor to perform the method described in any of the first aspects above.

[0041] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in any of the first aspects above.

[0042] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the first aspects above.

[0043] This application provides a battery compartment fan fault early warning method, device, equipment, medium, and program product. It can acquire the physical location and temporal operating data of multiple fans within the battery compartment, analyze the deviation of each fan from the group's characteristics using a spatiotemporal clustering algorithm, and determine the initial abnormal state of the fans using a time-series analysis model. Combining the deviation degree and the initial abnormality judgment results, it provides early warning of fan faults. Multi-dimensional data analysis reduces false alarms and improves the accuracy of fan fault early warning. Furthermore, this method uses relevant fan data instead of temperature monitoring in existing technologies, reducing the interference of ambient temperature on fault early warning and eliminating the need to deploy numerous temperature sensors, thus lowering hardware investment and maintenance costs. Attached Figure Description

[0044] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0045] Figure 1A flowchart illustrating a battery compartment fan failure early warning method provided in this application embodiment;

[0046] Figure 2 A flowchart illustrating a method for determining the degree of fan deviation provided in an embodiment of this application;

[0047] Figure 3 A flowchart illustrating a specific battery compartment fan failure early warning method provided in this application embodiment;

[0048] Figure 4 A schematic diagram of a battery compartment fan failure early warning device provided in this application;

[0049] Figure 5 This is a schematic diagram of the structure of an electronic device provided in this application.

[0050] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0051] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0052] In this application, the term "comprising" and its variations can refer to non-limiting inclusion; the term "or" and its variations can refer to "and / or". The terms "first", "second", etc., in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. In this application, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0053] During the operation of an energy storage system, temperature management within the battery compartment is a core element in ensuring the system's safety and stability. As the execution unit of the thermal management system, the operating status of the fan directly affects heat dissipation efficiency. If the fan malfunctions, such as due to mechanical wear, electrical short circuits, or abnormal control signals, it may cause an abnormal rise in the battery compartment temperature, potentially leading to thermal runaway.

[0054] In existing technologies, battery compartment fan fault early warning mainly relies on temperature monitoring to diagnose fan faults. This involves deploying numerous temperature sensors within the battery compartment to collect real-time temperature data. This data, combined with temperature fluctuation characteristics such as average temperature and temperature gradient, is used to determine if the fan is malfunctioning. For example, abnormal fan operation can be identified by comparing the deviation of the actual temperature from a preset threshold or analyzing abrupt changes in the temperature curve.

[0055] However, this method has several problems. First, it is susceptible to factors such as ambient temperature, battery charging and discharging status, and thermal management system control strategies, which can cause temperature fluctuations under normal operating conditions to be misjudged as faults. For example, increased cell internal resistance in low-temperature environments may cause localized heating, but the overall temperature may not exceed the threshold, thus missing potential faults. Second, this method requires the deployment of a large number of temperature sensors, resulting in high hardware costs. Third, the reliability of this method is limited by the accuracy of the temperature sensors, and there is a risk of single-point failure. Once a temperature sensor fails, it may directly cause false alarms or missed alarms, and may also cause the thermal management system to fail, preventing timely alarm triggering.

[0056] Therefore, this application provides a method, apparatus, device, medium, and program product for early warning of fan failure in a battery compartment. It can acquire the physical location and temporal operating data of multiple fans within the battery compartment, analyze the degree of deviation between the fans and the group's characteristics using a spatiotemporal clustering algorithm, and determine the initial abnormal state of the fans using a time-series analysis model. Combining the deviation degree with the initial abnormality judgment results, it provides early warning of fan failure. Through multi-dimensional data analysis, it reduces false alarms and improves the accuracy of fan failure early warning. Simultaneously, this method uses relevant fan data to replace temperature monitoring in existing technologies, reducing the interference of ambient temperature on failure early warning and eliminating the need to deploy a large number of temperature sensors, thus reducing hardware investment and maintenance costs.

[0057] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0058] It should be noted that the executing entity of this application can be any electronic device with processing capabilities, such as a user terminal or a server, for example, a computer.

[0059] Figure 1 This is a flowchart illustrating a battery compartment fan failure early warning method provided in an embodiment of this application. Figure 1 As shown, the method includes:

[0060] S101. Obtain the physical location of multiple fans in the battery compartment, as well as the timing data of the fan operation; the fans are used to dissipate heat from the battery compartment.

[0061] Optionally, the physical location of the fan can be its actual spatial coordinates or relative position information within the battery compartment space. The timing operation data of the fan can be the operating status information of the fan at different points in time, which may include, for example, any one or more of the following: fan runtime, fan start / stop signal, fan speed, etc.

[0062] Optionally, the physical locations of multiple fans within the battery compartment can be pre-stored in an electronic device. The electronic device can use a high-precision timer within the battery compartment, such as a timer accurate to the second, to record timestamps of the operating status information, thereby constructing the fan timing data. In one embodiment, the electronic device can store the fan timing data in a timing database according to a uniform format.

[0063] S102. Based on physical location and time-series operation data, the degree of deviation between the fan and the characteristics of the fan group is determined by a spatiotemporal clustering algorithm; the abnormal probability of the fan is positively correlated with the degree of deviation.

[0064] Optionally, a spatiotemporal clustering algorithm can combine two dimensions—space (i.e., physical location) and time (i.e., sequential operational data)—to group multiple fans into multiple clusters. This spatiotemporal clustering algorithm can classify fans with similar spatial locations and temporal operational characteristics into one category.

[0065] Optionally, the fan group characteristics can be the common physical location distribution characteristics and temporal operation characteristics exhibited by the fans within each cluster after the fans are grouped using a spatiotemporal clustering algorithm. For example, the fans in a certain cluster may be concentrated in a certain area of ​​the battery compartment, and their operating times may be similar.

[0066] The degree of deviation between a fan and its fan group characteristics is used to measure the magnitude of the difference between the characteristics of an individual fan and the characteristics of its fan group (i.e., cluster). For example, a fan is considered to have a high degree of deviation if its physical location is far from the concentration area of ​​other fans in its cluster, and / or its time-series operating data (such as runtime) is significantly different from that of other fans in the cluster.

[0067] The probability of a fan malfunctioning indicates the likelihood of it exhibiting an abnormal behavior. This probability is positively correlated with the degree of deviation; the greater the deviation of a fan from the characteristics of the fan group, the higher the probability of it malfunctioning; conversely, the smaller the deviation, the lower the probability of it malfunctioning.

[0068] S103. Based on the timing operation data, determine the initial abnormal judgment result of the fan through the timing analysis model.

[0069] Optionally, the time-series analysis model can be a mathematical model and algorithm used to analyze time-series operational data, capable of capturing trends, periodicity, and other characteristics in the time-series data. For example, the time-series analysis model can use methods such as moving averages or exponential smoothing to perform trend analysis on fan runtime data, observing the changing trend of runtime over time. If the fan runtime continuously increases over a period of time, and the increase exceeds the normal range, it indicates that the fan is malfunctioning, leading to reduced heat dissipation efficiency, requiring a longer runtime to achieve the same heat dissipation effect.

[0070] Optionally, the initial anomaly judgment result can be presented in numerical form after digital processing. The larger the value, the more severe the fan anomaly is determined by the time-series analysis model based on the time-series operating data.

[0071] S104. Based on the degree of deviation and the initial anomaly judgment results, issue a fault warning for the fan.

[0072] Optionally, the electronic device can digitally process the degree of deviation and the initial anomaly judgment result. By determining the weighted sum of the degree of deviation and the initial anomaly judgment result, and combining it with a pre-set fault warning threshold, a fault warning is issued for the fan when the weighted sum exceeds the fault warning threshold.

[0073] Electronic devices can also digitally process the degree of deviation and the initial anomaly judgment result, and issue a fault warning to the fan when the degree of deviation exceeds a preset deviation threshold and / or the initial anomaly judgment result exceeds a preset threshold.

[0074] This application embodiment can acquire the physical location and time-series operating data of multiple fans within the battery compartment. It utilizes a spatiotemporal clustering algorithm to analyze the degree of deviation between the fans and the group's characteristics, and a time-series analysis model to determine the initial abnormal state of the fans. Combining the deviation degree and the initial abnormality judgment results, it provides early warning of fan failures. Through multi-dimensional data analysis, it reduces false alarms and improves the accuracy of fan failure warnings. Furthermore, this method uses relevant fan data instead of temperature monitoring in existing technologies, reducing the interference of ambient temperature on fault warnings and eliminating the need to deploy numerous temperature sensors, thus lowering hardware investment and maintenance costs.

[0075] The following section uses the time-series operating data of fans, including their runtime, as an example to explain in detail how to determine the degree of deviation between the characteristics of a fan and the characteristics of a fan group based on its physical location and time-series operating data through a spatiotemporal clustering algorithm. Figure 2 This is a flowchart illustrating a method for determining the degree of fan deviation provided in an embodiment of this application, as shown below. Figure 2 As shown, S102 above includes:

[0076] S201. Based on physical location and runtime, construct the three-dimensional features of the fan.

[0077] The physical location of a fan can be represented by two-dimensional coordinates, for example, by (x, y) coordinates. Optionally, the electronic device can use the physical location represented by two-dimensional coordinates as two dimensions in three-dimensional space, and the runtime as a third dimension, for example, using z to represent the runtime. Each fan has a corresponding point in three-dimensional space, and the coordinates (x, y, z) of this point can characterize the fan's physical location and runtime.

[0078] S202. Based on the three-dimensional features of multiple fans, an adaptive parameter search algorithm is used to determine the cluster radius and the threshold number of fans included in each cluster.

[0079] Optionally, the adaptive parameter search algorithm can dynamically change any one or more parameters such as the search step size and direction based on the distribution and changes of the three-dimensional features of multiple fans, thereby determining the cluster radius and the threshold number of fans included in each cluster.

[0080] The cluster radius determines which fans will be grouped into the same cluster based on their 3D feature points. Taking a cluster center as an example, 3D feature points of fans whose distance from the cluster center is less than or equal to the cluster radius will be assigned to that cluster. The size of the cluster radius directly affects the clustering result; a radius that is too large may lead to fans with different characteristics being incorrectly grouped into one category, while a radius that is too small may separate fans that should belong to the same category.

[0081] The threshold for the number of fans included in each cluster can be the minimum number of fans included in each cluster. If this threshold is set too low, the clusters may become too scattered, with each cluster containing very few fans, making it impossible to effectively extract groups of fans with common characteristics. If the threshold is set too high, some fans with significantly different characteristics may be forcibly grouped into one category, affecting the accuracy of clustering.

[0082] S203. Using a spatiotemporal clustering algorithm, based on the cluster radius and the fan number threshold, determine the characteristics of the fan group; the fan group characteristics include: the cluster center of the fan group; the clustering constraints of the spatiotemporal clustering algorithm include at least one of the following: physical spacing constraints between multiple fans, runtime difference constraints between multiple fans, and synchronization rate constraints on whether multiple fans are turned on or off synchronously.

[0083] Optionally, the cluster center of a fan group can be a point representing the average level or typical characteristics of the three-dimensional features of all fans in the cluster. For example, in the three-dimensional feature space of multiple fans (constructed based on physical location and runtime), the coordinates of the cluster center can be obtained by calculating the average of the corresponding dimensional coordinates of all fans within the cluster.

[0084] Optionally, the physical spacing constraint between multiple fans can be that the distance between fans in the same cluster in physical space cannot exceed a certain set value. Through this physical spacing constraint, it can be ensured that the fans in the same group have a certain spatial correlation, such as being in the same area or adjacent positions, facing similar environmental conditions and heat dissipation requirements.

[0085] Runtime difference constraints among multiple fans can specify that the runtime difference of fans within a cluster cannot exceed a certain range. This runtime difference constraint allows fans within the same group to have similar workloads and usage frequencies.

[0086] The synchronization rate constraint, which determines whether multiple fans turn on or off synchronously, can be a condition for measuring the degree of synchronization between the on and off times of fans within the same cluster. By setting a synchronization rate threshold, the proportion of fans within a cluster that turn on or off synchronously is required to reach a certain level.

[0087] S204. Based on physical location, runtime, and cluster center, the target Mahalanobis distance between the fan and the cluster center is obtained. The target Mahalanobis distance is used to characterize the degree of deviation between the fan and the fan group characteristics.

[0088] Alternatively, the electronic device may first merge the physical location and runtime to obtain the merged result, and then obtain the target Mahalanobis distance between the fan and the cluster center based on the merged result and the cluster center.

[0089] In one implementation, the electronic device may first obtain a first Mahalanobis distance between the fan and the cluster center based on the physical location and the physical location of the cluster center, and then obtain a second Mahalanobis distance between the fan and the cluster center based on the runtime and the runtime of the cluster center.

[0090] Optionally, the calculation process for obtaining the first and second Mahalanobis distances can refer to existing technologies, and will not be elaborated here. Ultimately, the electronic device can obtain the target Mahalanobis distance between the fan and the cluster center based on the first and second Mahalanobis distances.

[0091] Optionally, the electronic device can obtain the target Mahalanobis distance between the fan and the cluster center by performing a weighted calculation on the first Mahalanobis distance and the second Mahalanobis distance.

[0092] This application embodiment calculates the Mahalanobis distance between the fan and the cluster center in terms of physical location and runtime, and combines these two distances to obtain the target Mahalanobis distance. This quantifies the degree of deviation between the fan and the cluster characteristics, which helps to more accurately assess the abnormal state of the fan and improve the reliability of fault early warning.

[0093] This application's embodiments construct three-dimensional features of fans, namely physical location and operating time, and use an adaptive parameter search algorithm to determine clustering parameters. Then, a spatiotemporal clustering algorithm is used to determine the fan group characteristics, accurately classifying fans with similar locations and operating conditions. This method can more accurately identify fans that deviate significantly from the group characteristics, thereby identifying fans with a higher probability of anomalies and improving the sensitivity and accuracy of fan fault detection.

[0094] The above example, using the fan's time-series operating data including its runtime, illustrates how to determine the degree of deviation between a fan and the characteristics of a fan group using a spatiotemporal clustering algorithm. The following example, using the fan's time-series operating data including its start / stop signals and runtime, details how to determine the initial anomaly judgment result of a fan based on the time-series operating data and a time-series analysis model.

[0095] In one implementation, the time series analysis model may include a sliding window dynamic time warping (DTW) model and a long short-term memory network-cumulative sum (LSTM-CUSUM) model.

[0096] First, electronic devices can determine the degree of irregularity in fan start-stop based on the fan's start-stop signal using the DTW model.

[0097] Optionally, the fan start / stop signal is an electrical signal or data sequence used to indicate the fan's on and off states. For example, using a digital signal, a high level (such as 1) can indicate that the fan is on, and a low level (such as 0) can indicate that the fan is off.

[0098] Optionally, the DTW model can analyze the start-stop signal time series of the fan. For example, the electronic device can first build a reference template using the DTW model. This reference template represents the start-stop signal time series of the fan under normal operating conditions. This reference template can be stored in the electronic device in advance.

[0099] Secondly, in the DTW model, a 1-hour window length and a 10-minute step size can be set. The electronic device captures the fan's start-stop signal in 1-hour window intervals to obtain the fan's start-stop signal time series. The DTW distance between this fan's start-stop signal time series and the baseline template is calculated. The analysis window is continuously moved (every 10 minutes), and the DTW distance between the runtime sequence within each window and the baseline template is calculated. When the DTW distance calculated in a certain window is significantly greater than that in other windows, it means that there is a large difference between the fan's runtime sequence and the baseline template within that time period, indicating a transient anomaly, i.e., irregular fan start-stop.

[0100] Optionally, the degree of fan start-stop irregularity can be positively correlated with the DTW distance value; that is, the larger the DTW distance value, the higher the degree of fan start-stop irregularity; the smaller the DTW distance value, the lower the degree of fan start-stop irregularity. To more intuitively and accurately quantify the degree of fan start-stop irregularity, specific numbers can be used to represent it.

[0101] Secondly, electronic devices can determine the cumulative residual sum between the predicted and actual runtime values ​​of the fan based on the fan's runtime using an LSTM-CUSUM model.

[0102] Optionally, electronic devices can learn the time-dependent characteristics of normal operating time through an LSTM network, output a predicted value of the operating time, and then calculate the residual between the predicted and actual values ​​of the operating time to reflect the difference between the predicted and actual values. The residual sequence output by the LSTM network serves as the input to a CUSUM detector. By accumulating and analyzing the residuals, the CUSUM detector can capture the gradual change in operating time. When the accumulated deviation exceeds a threshold, the electronic device can trigger a progressive fault warning.

[0103] After determining the start-stop irregularity assessment result and the cumulative residual sum, the electronic device can obtain the initial anomaly assessment result for the fan. For example, the electronic device can normalize the start-stop irregularity assessment result and the cumulative residual sum, and then perform a weighted summation of the two. Based on this weighted sum, the initial anomaly assessment result is determined.

[0104] For example, as described above, the initial anomaly judgment result can be presented in numerical form, which can be positively correlated with the weighted sum. The larger the weighted sum, the larger the value of the initial anomaly judgment result, that is, the more severe the fan anomaly is determined by the time-series analysis model based on the time-series operating data.

[0105] This application's embodiments capture the irregularities of start / stop signals using a DTW model to detect transient anomalies in the fan, and detect long-term trends in runtime using an LSTM-CUSUM model to detect progressive fan failures. By combining transient and progressive fan failures, a comprehensive assessment of the fan's operating status can be achieved, improving the accuracy of initial anomaly detection results.

[0106] Using both methods together allows for a comprehensive assessment of the fan's operating status, improving the accuracy of initial anomaly detection. The technical advantage lies in providing a more comprehensive means of fan fault detection, helping to promptly identify and address various types of fan malfunctions.

[0107] After the electronic device determines the degree of deviation between the fan and the characteristics of the fan group, as well as the initial anomaly judgment result of the fan, the electronic device can provide a fault warning for the fan. In one embodiment, the electronic device can determine the fan's health index (HI) based on the degree of deviation and the weighted sum of the initial anomaly judgment results.

[0108] Optionally, the fan's health index can be a numerical indicator reflecting the fan's health condition. For example, assuming the weight of the deviation degree is 0.6 and the weight of the initial anomaly judgment result is 0.4, the health index can be calculated using the following formula:

[0109]

[0110] Electronic devices can set a first threshold and a second threshold. By comparing the health index with the relationship between the first threshold and the second threshold, the level of the fault warning signal can be determined. The probability of fan failure can be positively correlated with the level of the fault warning signal, that is, the higher the level of the fault warning signal, the greater the probability of fan failure.

[0111] Taking a first threshold of 60 and a second threshold of 80 as an example, an electronic device can output a level-one fault warning signal when the health index is less than the first threshold, i.e., HI < 60. For example, this level-one fault warning signal can be a low-risk warning signal.

[0112] Electronic devices can output a level-two fault warning signal when the health index is greater than or equal to a first threshold and less than a second threshold, i.e., 60 ≤ HI < 80. For example, this level-two fault warning signal can be a medium-risk warning signal.

[0113] Electronic devices can output a three-level fault warning signal when the health index is greater than or equal to a second threshold, i.e., HI≥80. For example, this three-level fault warning signal can be a high-risk warning signal.

[0114] The embodiments of this application can determine the health index of the fan by weighted calculation of the deviation degree and the initial anomaly judgment result, and output different levels of fault warning signals according to different ranges of the health index. This can intuitively reflect the degree of fan failure risk, enabling maintenance personnel to quickly locate and handle high-risk fans, and improve the timeliness and effectiveness of fault warning.

[0115] After the electronic device issues a fault warning for the fan, embodiments of this application can further determine the fault type based on the fan's runtime. In one implementation, the electronic device can use a decision tree model to determine that the fan's fault type is a mechanical fault if the fan's runtime increases over multiple consecutive historical running time units and the fluctuation rate of the runtime over multiple consecutive historical running time units is less than or equal to a preset fluctuation rate.

[0116] Optionally, the historical unit of operation time can be a unit of time period obtained by dividing the operation time according to a certain time interval. For example, a day can be used as a historical unit of operation time, or it can be divided into intervals of hours, minutes, etc. This historical unit of operation time is used to record and analyze the operation of the fan in different time periods. Volatility can represent the degree of change in the operation time in different historical unit of operation time periods. Mechanical failure can be a type of failure in which the fan cannot operate normally or its operating performance deteriorates due to damage, wear, loosening, deformation, etc. of mechanical parts.

[0117] The fan's runtime increases over multiple consecutive historical operating units, indicating that the fan's runtime in each subsequent historical operating unit is longer than that in the previous one. For example, if the fan's runtime increases by 2 hours, 3 hours, and 4 hours for three consecutive days, this could indicate a problem with the fan's mechanical components, causing the fan to require more time to achieve the desired operating effect, or a mechanical failure preventing the fan from completing its task within the normal timeframe and continuing to operate.

[0118] If the fluctuation rate of the fan's runtime over multiple consecutive historical operating periods is less than or equal to the preset fluctuation rate, it indicates that the fan's runtime remains relatively stable over these periods, without significant fluctuations. In other words, while the fan's runtime increases, the magnitude and rhythm of this increase are relatively regular and controllable, remaining within the preset fluctuation range. This further rules out the possibility of random changes in runtime due to external factors (such as unstable power supply or significant changes in ambient temperature), and instead suggests that the changes are caused by problems with the fan's own mechanical structure.

[0119] If the fan's runtime decrease rate over a historical unit of operating time is greater than or equal to a preset decrease rate, or if the fan's runtime is zero, the electronic equipment can determine that the fan's fault type is an electrical fault.

[0120] Optionally, the runtime reduction rate can characterize the percentage decrease in runtime of the later historical unit of runtime compared to the runtime of the earlier historical unit of runtime between two adjacent historical units of runtime. Electrical faults can be types of faults that prevent the fan from operating normally or cause abnormal operation due to problems with the electrical system, such as any one or more of the following: short circuit or open circuit in the motor windings, poor contact in the power supply line, or damage to the control switch.

[0121] If the fan's runtime decrease rate within a historical unit of operating time is greater than or equal to the preset decrease rate, it indicates that there is a problem with certain components in the electrical system, causing the motor to be unable to obtain sufficient power or the control circuit to malfunction, preventing the fan from operating continuously for the normal time, thus resulting in a significant decrease in runtime.

[0122] A fan's runtime of zero means that the fan did not operate at all within a certain historical unit of operation time. This could indicate a serious problem with the electrical system, such as a complete power outage, a completely damaged motor that cannot start, or a control circuit malfunction that prevents the fan from receiving a start signal, among other things. When the runtime is zero, it means that the fan's electrical system is unable to bring the fan into operation.

[0123] If the fan's operating time differs from the preset time in multiple consecutive historical operating time periods, the fan's fault type is determined to be a control fault.

[0124] Optionally, the target difference can be the allowable range of difference between a pre-set running time and a preset duration. A control fault can be a problem in the fan's control system that prevents the fan from operating as preset. The control system includes components such as control circuits, timers, sensors, and remote controls. Control faults may manifest as one or more of the following: the fan cannot start, cannot stop, the running time is uncontrolled, or the speed regulation fails.

[0125] The running time of a fan is regulated and controlled by its control system. If the running time of multiple consecutive historical units deviates from the preset time and exceeds the target difference, it indicates that there is a problem in a certain part of the control system. For example, a control circuit failure may prevent the accurate reception and execution of preset running instructions, a timer malfunction may prevent the running time from being counted correctly, or a sensor malfunction may prevent the accurate perception of environmental information, thus affecting the running time.

[0126] Based on the characteristics of fan operating time variation, this application embodiment subdivides fan failure types into mechanical failures, electrical failures, and control failures. This helps maintenance personnel to more accurately understand the causes of fan failures, thereby taking targeted maintenance measures, improving the targeting and efficiency of fan failure maintenance, and reducing unnecessary maintenance work and costs.

[0127] Figure 3 A flowchart illustrating a specific battery compartment fan failure early warning method provided in this application embodiment is shown below. Figure 3 As shown, the electronic device may include a data acquisition layer, an intelligent analysis layer, and a decision output layer. The data acquisition layer is responsible for data acquisition, data preprocessing, and building a time-series database. The intelligent analysis layer can first perform group collaborative analysis on the time-series data in the time-series database, that is, determine the degree of deviation between fans and fan group characteristics through spatiotemporal clustering algorithms, and then determine the initial anomaly judgment result of the fans through time-series pattern mining, that is, through time-series analysis models, and output a decision. The decision output layer can calculate the Health Index (HI), and perform fault classification and graded alarm warnings.

[0128] The above are the method embodiments provided in this application. The apparatus provided in this application will be described below.

[0129] Figure 4 This application provides a structural schematic diagram of a battery compartment fan failure early warning device, as shown below. Figure 4 As shown, the battery compartment fan failure early warning device 400 provided in this embodiment includes: an acquisition module 401, a first determination module 402, a second determination module 403, and an early warning module 404. Optionally, the battery compartment fan failure early warning device may further include a processing module 405.

[0130] The acquisition module 401 is used to acquire the physical location of multiple fans in the battery compartment, as well as the timing operation data of the fans; the fans are used to dissipate heat from the battery compartment.

[0131] The first determining module 402 is used to determine the degree of deviation between the fan and the characteristics of the fan group based on the physical location and the time-series operation data through a spatiotemporal clustering algorithm; the abnormal probability of the fan is positively correlated with the degree of deviation.

[0132] The second determining module 403 is used to determine the initial abnormal judgment result of the fan based on the timing operation data and through the timing analysis model.

[0133] The early warning module 404 is used to provide early warning of fan failure based on the degree of deviation and the initial anomaly judgment result.

[0134] Optionally, the fan's timing operation data includes: fan runtime. The physical location is represented by two-dimensional coordinates. The first determining module 402 is specifically used to construct three-dimensional features of the fan based on its physical location and runtime. Based on the three-dimensional features of multiple fans, an adaptive parameter search algorithm is used to determine the cluster radius and a threshold number of fans included in each cluster. A spatiotemporal clustering algorithm is used to determine fan group characteristics based on the cluster radius and the fan number threshold. The fan group characteristics include the cluster center of the fan group. The clustering constraints of the spatiotemporal clustering algorithm include at least one of the following: physical distance constraints between multiple fans, runtime difference constraints between multiple fans, and synchronization rate constraints regarding whether multiple fans are synchronously turned on or off. Based on the physical location, runtime, and cluster center, the target Mahalanobis distance between the fan and the cluster center is obtained. The target Mahalanobis distance is used to characterize the degree of deviation between the fan and the fan group characteristics.

[0135] For example, the first determining module 402 is specifically used to: obtain a first Mahalanobis distance between the fan and the cluster center based on the physical location and the physical location of the cluster center; obtain a second Mahalanobis distance between the fan and the cluster center based on the runtime and the runtime of the cluster center; and obtain a target Mahalanobis distance based on the first and second Mahalanobis distances.

[0136] Optionally, the time-series operational data includes: fan start / stop signals and fan runtime; the time-series analysis model includes: a sliding window dynamic time warping (DTW) model and a long short-term memory network-cumulative sum (LSTM-CUSUM) model; the second determination module 403 is specifically used to: determine the degree of fan start / stop irregularity judgment based on the fan start / stop signals using the DTW model; determine the cumulative residual sum between the predicted value and the actual value of the runtime using the LSTM-CUSUM model based on the fan runtime; and obtain the initial anomaly judgment result for the fan based on the degree of start / stop irregularity judgment result and the cumulative residual sum.

[0137] Optionally, the early warning module 404 is specifically used to determine the fan's health index based on the degree of deviation and a weighted sum of the initial anomaly judgment results. If the health index is less than a first threshold, a level one fault warning signal is output. If the health index is greater than or equal to the first threshold and less than a second threshold, a level two fault warning signal is output. If the health index is greater than or equal to the second threshold, a level three fault warning signal is output; the probability of fan failure is positively correlated with the level of the fault warning signal.

[0138] Optionally, the fault warning is used to characterize a fan malfunction. In the warning module 404, specifically after issuing a fault warning for the fan based on the degree of deviation and the initial anomaly judgment result, the processing module 405 determines the fan fault type as a mechanical fault if the fan's runtime increases over multiple consecutive historical operating time units, and the fluctuation rate of the runtime over these multiple consecutive historical operating time units is less than or equal to a preset fluctuation rate. If the fan's runtime decrease rate over a historical operating time unit is greater than or equal to a preset decrease rate, or if the fan's runtime is zero, the fan fault type is determined to be an electrical fault. If the fan's runtime over multiple consecutive historical operating time units has a target difference from a preset duration, the fan fault type is determined to be a control fault.

[0139] The battery compartment fan failure early warning device provided in this embodiment can execute the method provided in any of the above method embodiments. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0140] Figure 5 This is a schematic diagram of the structure of an electronic device provided in this application. Figure 5 As shown, the electronic device 500 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 500 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0141] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0142] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0143] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0144] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0145] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0146] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0147] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0148] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0149] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0150] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0151] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0152] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0153] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0154] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0155] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for early warning of battery compartment fan failure, characterized in that, The method includes: The physical locations of multiple fans within the battery compartment, and the timing operation data of the fans, are obtained; the fans are used to dissipate heat from the battery compartment. Based on the physical location and the time-series operation data, a spatiotemporal clustering algorithm is used to determine the degree of deviation between the fan and the characteristics of the fan group; the abnormal probability of the fan is positively correlated with the degree of deviation. Based on the aforementioned time-series operation data, the initial anomaly judgment result of the fan is determined through a time-series analysis model; Based on the degree of deviation and the initial anomaly judgment result, a fault warning is issued for the fan.

2. The method according to claim 1, characterized in that, The time-series operation data of the fan includes: the fan's runtime, the physical location represented by two-dimensional coordinates, and the determination of the degree of deviation between the fan and the fan group characteristics based on the physical location and the time-series operation data using a spatiotemporal clustering algorithm, including: Based on the physical location and the runtime, construct the three-dimensional features of the fan; Based on the three-dimensional features of the multiple fans, an adaptive parameter search algorithm is used to determine the cluster radius and the threshold number of fans included in each cluster. The spatiotemporal clustering algorithm determines the characteristics of the fan group based on the cluster radius and the fan number threshold. The fan group characteristics include the cluster center of the fan group. The clustering constraints of the spatiotemporal clustering algorithm include at least one of the following: physical spacing constraints between the multiple fans, runtime difference constraints between the multiple fans, and synchronization rate constraints on whether the multiple fans are turned on or off synchronously. Based on the physical location, the runtime, and the cluster center, the target Mahalanobis distance between the fan and the cluster center is obtained. The target Mahalanobis distance is used to characterize the degree of deviation between the fan and the characteristics of the fan group.

3. The method according to claim 2, characterized in that, The step of obtaining the target Mahalanobis distance between the fan and the cluster center based on the physical location, the runtime, and the cluster center includes: Based on the physical location and the physical location of the cluster center, the first Mahalanobis distance between the fan and the cluster center is obtained; Based on the runtime and the runtime of the cluster center, the second Mahalanobis distance between the fan and the cluster center is obtained; The target Mahalanobis distance is obtained based on the first Mahalanobis distance and the second Mahalanobis distance.

4. The method according to any one of claims 1-3, characterized in that, The time-series operational data includes: the fan's start / stop signals and the fan's runtime; the time-series analysis model includes: a sliding window dynamic time warping (DTW) model and a long short-term memory network-cumulative sum (LSTM-CUSUM) model; the determination of the initial anomaly judgment result of the fan based on the time-series operational data and the time-series analysis model includes: Based on the start and stop signals of the fan, the degree of irregularity in the start and stop of the fan is determined by the DTW model. Based on the fan's runtime, the cumulative residual sum between the predicted runtime and the actual runtime is determined using the LSTM-CUSUM model. Based on the judgment result of the irregularity of start-stop and the cumulative sum of residuals, the initial abnormality judgment result of the fan is obtained.

5. The method according to any one of claims 1-3, characterized in that, The method of providing a fault warning for the fan based on the degree of deviation and the initial anomaly judgment result includes: Based on the degree of deviation and the weighted sum of the initial anomaly judgment results, the health index of the fan is determined; If the health index is less than the first threshold, a level one fault warning signal is output. If the health index is greater than or equal to the first threshold and less than the second threshold, a level two fault warning signal is output. If the health index is greater than or equal to the second threshold, a level 3 fault warning signal is output; the probability of fan failure is positively correlated with the level of the fault warning signal.

6. The method according to any one of claims 1-3, characterized in that, The fault warning is used to characterize a fan malfunction. After issuing a fault warning for the fan based on the degree of deviation and the initial anomaly judgment result, the method further includes: If the running time of the fan increases over multiple consecutive historical running time units, and the fluctuation rate of the running time over the multiple consecutive historical running time units is less than or equal to a preset fluctuation rate, the fault type of the fan is determined to be a mechanical fault. If the fan's runtime decrease rate during the historical unit operating time is greater than or equal to a preset decrease rate, or if the fan's runtime is zero, the fault type of the fan is determined to be an electrical fault. If the running time of the fan in multiple consecutive historical unit running times all have a target difference from the preset time, the fault type of the fan is determined to be a control fault.

7. A battery compartment fan malfunction early warning device, characterized in that, include: The acquisition module is used to acquire the physical location of multiple fans in the battery compartment, as well as the timing operation data of the fans; The fan is used to dissipate heat from the battery compartment; The first determining module is used to determine the degree of deviation between the fan and the characteristics of the fan group based on the physical location and the time-series operation data through a spatiotemporal clustering algorithm. The probability of the fan malfunctioning is positively correlated with the degree of deviation. The second determining module is used to determine the initial abnormality judgment result of the fan based on the time-series operation data and through a time-series analysis model. The early warning module is used to provide a fault warning for the fan based on the degree of deviation and the initial anomaly judgment result.

8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-6.

Images