An apparatus anomaly degree index ranking method, device and computer storage medium

Abstract

This application discloses a method, apparatus, and computer storage medium for ranking equipment anomaly severity indices, relating to the field of intelligent operation and maintenance technology. The method includes: statistically analyzing the frequency and processing time of anomalies in N types of parameters for all abnormal devices in a historical anomaly event database; determining dynamic weight coefficients for each type of parameter based on the anomaly frequency and processing time; determining the equipment anomaly severity index for each target abnormal device in the abnormal device set based on real-time collected abnormal device data and the dynamic weight coefficients for each type of parameter; and ranking all target abnormal devices in the abnormal device set according to the equipment anomaly severity index of each target abnormal device to generate an equipment anomaly ranking list. This application solves the technical problem of traditional methods, which rely solely on a single parameter or static threshold to determine anomalies and cannot quantify the differences in the impact of multiple parameters on faults, leading to a subjective judgment of anomaly priority.

Description

Technical Field

[0001] This application relates to the field of intelligent operation and maintenance technology, and more specifically, to a method, apparatus, and computer storage medium for ranking equipment anomaly severity index. Background Technology

[0002] In the operation and maintenance management of data center cooling equipment, real-time monitoring of equipment operating status and anomaly response are key aspects to ensure system stability. Existing technologies typically use a single parameter (such as outlet air temperature) or fixed threshold rules for anomaly detection. When multiple operating parameters fluctuate simultaneously, the lack of quantitative analysis capabilities to assess the differences in the impact of various parameters leads to high false alarm rates and frequent missed alarms.

[0003] In addition, existing systems generally lack a dynamic priority ranking mechanism based on historical maintenance behavior when multiple devices experience concurrent anomalies. The allocation of maintenance resources often relies on the experience and judgment of maintenance personnel, which can easily lead to problems such as delayed response to high-impact anomalies and excessive intervention for low-impact anomalies, affecting overall maintenance efficiency and equipment availability.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a method, apparatus, and computer storage medium for ranking equipment anomaly severity index, which at least solves the technical problem in traditional methods that rely solely on a single parameter or static threshold to determine anomalies and cannot quantify the differences in the impact of multiple parameters on faults, resulting in a subjective judgment of anomaly priority.

[0006] According to one aspect of the embodiments of this application, a method for ranking equipment anomaly severity index is provided, comprising: statistically analyzing the frequency and duration of anomalies of N types of parameters of all abnormal equipment in a historical anomaly event database, wherein the N types of parameters include at least air outlet temperature, vibration intensity, energy consumption, and noise decibels; the frequency of anomalies represents the total number of times each type of parameter has anomalies in the historical anomaly event database; the duration of anomalies represents the total duration of each type of parameter in the historical anomaly event database after maintenance and restoration to normal operation; determining dynamic weight coefficients for each type of parameter based on the frequency of anomalies and the duration of anomalies; determining the equipment anomaly severity index of each target abnormal equipment in the abnormal equipment set based on the real-time collected abnormal equipment set and the dynamic weight coefficients of each type of parameter; and ranking all target abnormal equipment in the abnormal equipment set according to the equipment anomaly severity index of each target abnormal equipment to generate an equipment anomaly ranking list.

[0007] Optionally, before calculating the frequency and duration of abnormalities of the N types of parameters of all abnormal devices in the historical abnormal event database, the method further includes: obtaining the preset abnormal thresholds corresponding to the N types of parameters; comparing the N types of parameters of each abnormal device in the historical abnormal event database with the corresponding preset abnormal thresholds, and determining the abnormal status of each type of parameter based on the comparison results.

[0008] Optionally, based on the anomaly frequency and anomaly handling time, the dynamic weight coefficient of each type of parameter is determined, including: calculating the ratio of the anomaly frequency of each type of parameter to the total frequency of anomalies of all parameters in the historical anomaly event database to obtain the first data corresponding to each type of parameter; calculating the ratio of the anomaly handling time of each type of parameter to the total handling time of anomalies of all parameters in the historical anomaly event database to obtain the second data corresponding to each type of parameter; and multiplying the first data and the second data of each type of parameter to obtain the dynamic weight coefficient of that type of parameter.

[0009] Optionally, before determining the device anomaly index of each target anomaly device in the anomaly device set based on the real-time collected set of anomaly devices and the dynamic weight coefficients of each type of parameter, the method further includes: real-time collection of device numbers and N types of parameters from multiple devices; denoising and standardization of the collected N types of parameters to obtain a running dataset with device numbers; calculating the device running index of each device based on the relevant parameters of each device in the running dataset, wherein the relevant parameters include at least the usage years data, the usage years threshold, the N types of parameters, and the preset anomaly thresholds of the N types of parameters for each device; obtaining a historical running index dataset and determining the device running threshold based on the historical running index dataset; comparing the device running index of each device with the device running threshold, and determining the device status of each device based on the comparison result.

[0010] Optionally, based on the relevant parameters of each device in the running dataset, the device operation index of each device is calculated, including: obtaining the service life, preset service life threshold, and preset abnormal thresholds corresponding to N types of parameters for each device; calculating the ratio of the service life of each device to the preset service life threshold to obtain the first ratio of each device; calculating the ratio of each type of parameter of each device to the corresponding preset abnormal threshold to obtain the second set of ratios for each device, wherein the second set of ratios for each device includes at least the ratio of the outlet air temperature to the preset temperature threshold, the ratio of the vibration intensity to the preset vibration intensity threshold, the ratio of the energy consumption to the preset energy consumption threshold, and the ratio of the noise decibel to the preset noise decibel threshold; and calculating the product of all ratios in the first and second set of ratios for each device to obtain the device operation index of each device.

[0011] Optionally, the device operation index of each device is compared with the device operation threshold, and the device status of each device is determined based on the comparison result, including: when the device operation index of the i-th device is greater than or equal to the device operation threshold, the i-th device is determined to be a target abnormal device, where i is an integer greater than or equal to 1; when the device operation index of the i-th device is less than the device operation threshold, the i-th device is determined to be a normal device.

[0012] Optionally, the equipment anomaly severity index ranking method further includes: integrating multiple devices identified as target anomalies into an anomaly device set.

[0013] Optionally, based on the real-time collected set of abnormal devices and the dynamic weighting coefficients of each type of parameter, the device abnormality index of each target abnormal device in the set of abnormal devices is determined, including: calculating the absolute value of the difference between each type of parameter of each target abnormal device in the set of abnormal devices and the corresponding preset abnormal threshold, to obtain the deviation of each type of parameter; calculating the ratio of the deviation of each type of parameter to the corresponding preset abnormal threshold, to obtain the standard deviation value of each type of parameter; multiplying the standard deviation value of each type of parameter by the corresponding dynamic weighting coefficient, to obtain the weighted deviation value of each type of parameter; and summing all the weighted deviation values ​​of the same target abnormal device to obtain the device abnormality index of the target abnormal device.

[0014] According to another aspect of the embodiments of this application, a device for ranking equipment anomaly severity index is also provided, comprising: a statistics unit, used to count the frequency of anomalies and the duration of handling anomalies of N types of parameters of all abnormal equipment in a historical anomaly event database, wherein the N types of parameters include at least air outlet temperature, vibration intensity, energy consumption, and noise decibels, the frequency of anomalies represents the total number of times each type of parameter has anomalies in the historical anomaly event database, and the duration of handling anomalies represents the total handling time for each type of parameter to be restored to normal operation after maintenance in the historical anomaly event database; a first determination unit, used to determine the dynamic weight coefficient of each type of parameter according to the frequency of anomalies and the duration of handling anomalies; a second determination unit, used to determine the equipment anomaly severity index of each target abnormal equipment in the abnormal equipment set based on the real-time collected abnormal equipment set and the dynamic weight coefficient of each type of parameter; and a ranking unit, used to rank all target abnormal equipment in the abnormal equipment set according to the equipment anomaly severity index of each target abnormal equipment, and generate an equipment anomaly ranking list.

[0015] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, which stores a computer program, wherein when the computer program is executed, it causes the device where the computer-readable storage medium is located to execute the above-described device anomaly index ranking method.

[0016] According to another aspect of the embodiments of this application, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors cause the one or more processors to perform the above-described device anomaly index ranking method.

[0017] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program or instructions, which, when executed by a processor, implement the above-described device anomaly index ranking method.

[0018] In this application, the equipment anomaly severity index ranking method first statistically analyzes the frequency and processing time of anomalies in N types of parameters of all abnormal equipment in the historical anomaly event database. The N types of parameters include at least outlet air temperature, vibration intensity, energy consumption, and noise decibels. The frequency of anomalies represents the total number of times each type of parameter occurs in the historical anomaly event database, and the processing time represents the total processing time for each type of parameter to be restored to normal operation after maintenance in the historical anomaly event database. Based on the frequency and processing time, dynamic weight coefficients for each type of parameter are determined. Based on the real-time collected abnormal equipment set and the dynamic weight coefficients for each type of parameter, an equipment anomaly severity index is determined for each target abnormal equipment in the abnormal equipment set. Based on the equipment anomaly severity index of each target abnormal equipment, all target abnormal equipment in the abnormal equipment set are ranked to generate an equipment anomaly ranking list.

[0019] In this embodiment, the abnormal frequency and processing time of each parameter in the historical abnormal event database are statistically analyzed. By normalizing the product of abnormal frequency and processing time, the dynamic weight coefficient of each type of parameter is calculated. This achieves the goal of quantifying the actual impact of different types of parameters on the abnormal state of equipment in historical faults. This realizes the technical effect of adjusting the weight of various parameters based on historical operation and maintenance data, and solves the technical problem in traditional methods that rely on a single parameter or static threshold to judge abnormalities and cannot quantify the differences in the impact of multiple parameters on faults, resulting in a more subjective judgment of abnormality priority. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0021] Figure 1 This is a flowchart of an optional equipment anomaly degree index ranking method according to an embodiment of this application;

[0022] Figure 2This is a schematic diagram of an optional device anomaly index sorting device according to an embodiment of this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] According to an embodiment of this application, a method embodiment of a device anomaly degree index ranking method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0026] According to the embodiments of this application, a device anomaly degree index ranking system (hereinafter referred to as the system) can be used as the execution subject of the device anomaly degree index ranking method of this application embodiment. The device anomaly degree index ranking system can be a software system or an embedded system combining software and hardware. Of course, the method execution subject in the embodiments of this application can also be other forms of execution subject, such as devices or equipment. Those skilled in the art should know that this application does not particularly limit the specific form of the method execution subject.

[0027] Figure 1 This is a flowchart of an optional device anomaly degree index ranking method according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0028] Step S101: Calculate the frequency and duration of abnormality of N types of parameters of all abnormal devices in the historical abnormal event database. The N types of parameters include at least air outlet temperature, vibration intensity, energy consumption and noise decibels. The frequency of abnormality represents the total number of times each type of parameter has an abnormality in the historical abnormal event database. The duration of abnormality handling represents the total handling time for each type of parameter to be restored to normal operation after maintenance in the historical abnormal event database.

[0029] Optionally, the historical abnormal event database refers to a database of processed abnormal events of refrigeration equipment that has been accumulated in the system over a long period of time. Each record contains structured information such as the time of occurrence of the abnormality, the number of the equipment involved, the type of abnormal parameter, the duration of the abnormality, the maintenance operation record, and the recovery time.

[0030] Optionally, the system extracts all records from the historical anomaly event database and aggregates them according to parameter type. It iterates through all related events for each parameter type, counts the total number of times that parameter type is marked as an anomaly, and obtains the anomaly frequency. At the same time, it calculates the time difference from alarm to recovery for each event, and sums them up to obtain the anomaly handling time for that parameter type.

[0031] Optionally, this step quantifies the error frequency and repair difficulty of parameters in real-world operation and maintenance scenarios, abandoning subjective pre-set weights and constructing a parameter impact assessment basis based on actual operation and maintenance costs. High frequency indicates that the parameter is prone to triggering anomalies, and long processing time indicates that the anomaly consequences are serious and the repair costs are high. Together, they constitute an objective measure of parameter risk exposure, providing real and traceable data support for subsequent weight calculations.

[0032] Step S102: Determine the dynamic weight coefficient of each type of parameter based on the frequency of anomalies and the duration of anomaly handling.

[0033] Optionally, the system calculates the product of the anomaly frequency and the anomaly handling time for each type of parameter to obtain the comprehensive impact value of that parameter. Then, the comprehensive impact values ​​of all N types of parameters are summed to obtain a total, and the comprehensive impact value of each type of parameter is divided by the total to obtain the normalized dynamic weight coefficient.

[0034] Optionally, the dynamic weighting coefficient is a calculated value for each type of parameter, used to characterize the relative importance of that parameter to the abnormal state of the equipment among all N types of parameters. This coefficient is not a fixed value, but is dynamically generated after normalization based on the product of historical anomaly frequency and anomaly handling time. It reflects the comprehensive influence of the parameter on the health status of the equipment in history, so that the system's judgment on "which type of parameter has a greater impact from anomalies" no longer relies on subjective experience, but is driven by historical costs, thereby improving the objectivity and adaptability of anomaly assessment.

[0035] Step S103: Based on the real-time collected abnormal device set and the dynamic weight coefficient of each type of parameter, determine the device abnormality index of each target abnormal device in the abnormal device set.

[0036] Optionally, the abnormal device set refers to the set of all refrigeration devices in the system that are currently identified as abnormal. The basis for determining the abnormality of the devices is whether the real-time sensor data exceeds the preset threshold. It is unrelated to historical data and is only used to identify which devices are currently in an abnormal state.

[0037] Optionally, the equipment anomaly severity index refers to a comprehensive score calculated for a specific abnormal device based on the current abnormal parameter category of the device and combined with a determined dynamic weight coefficient. This score reflects the severity of the device's anomaly at the current moment. The higher the value, the greater the overall threat the abnormal state poses to the system operation.

[0038] Optionally, for each device in the abnormal device set, the system first identifies which parameters of each device are currently in an abnormal state. For example, device A: abnormal air outlet temperature, abnormal vibration intensity; device B: abnormal energy consumption, abnormal noise level. Then, the dynamic weight coefficients corresponding to each type of abnormal parameter of the device are added together to calculate the device's abnormality index.

[0039] Optionally, this step implements a breadth-first assessment of anomalous devices. Even if a device has only one abnormal parameter, if that parameter has a high weight, the device's anomalousness index may still be higher than that of devices with multiple low-weight parameters that are abnormal. This allows the system to identify devices with few anomalies but out-of-control key parameters, reducing the misleading judgment that multiple small anomalies equal large anomalies in traditional methods. This provides a quantitative basis that truly reflects the nature of the risk for subsequent ranking.

[0040] Step S104: Based on the equipment anomaly degree index of each target abnormal device, sort all target abnormal devices in the abnormal device set to generate an equipment anomaly sorting list.

[0041] Optionally, the equipment anomaly sorting list refers to a list of equipment arranged from high to low according to the equipment anomaly severity index. Each item in the list includes the equipment number and the corresponding index, which serves as the sole basis for dispatching maintenance work orders.

[0042] Optionally, the system sorts the device anomaly severity index of all devices in the abnormal device set from largest to smallest, forming an ordered list, and automatically pushes it to the operation and maintenance terminal without manual intervention in sorting.

[0043] Optionally, this step enables precise and automated decision-making regarding maintenance priorities. Traditionally, maintenance personnel often rely on experience to determine the most urgent issue when faced with multiple malfunctioning devices, easily overlooking devices with high-weighted abnormal parameters. This application uses index-based sorting to ensure that devices with simultaneously abnormal outlet air temperature and vibration intensity are always prioritized, even if their outlet air temperature exceeds the standard by only 1°C. This prioritizes devices with simultaneously abnormal energy consumption and noise levels but lower total weight, reducing subjective judgment errors and helping to allocate limited maintenance resources to high-risk devices that truly impact system stability, thereby improving fault response efficiency and system availability.

[0044] Through the above steps, the system integrates multiple types of sensors to collect real-time data on various parameters of the refrigeration equipment, including outlet air temperature, vibration intensity, energy consumption, and noise levels. This data is then processed through noise reduction and standardization to form a unified dataset, overcoming the limitations of traditional methods that rely on single parameters. This not only improves the comprehensiveness of monitoring but also reduces the probability of false alarms and missed alarms through data fusion, thus contributing to the stability of equipment operation. Secondly, a dynamic weighting coefficient calculation mechanism based on historical anomaly data is introduced. This mechanism quantitatively analyzes the differences in the contribution of anomaly frequency and processing time to various data types, allowing the system to adaptively adjust the sensitivity of different parameters to anomaly states. This further enhances the accuracy and intelligence of diagnosis, contributing to the reliability and timeliness of anomaly identification. Furthermore, by constructing an equipment anomaly severity index model, combining weighting coefficients with real-time data deviations, the system achieves a quantitative ranking of equipment anomaly severity. This allows maintenance personnel to prioritize equipment with high anomaly severity based on the electronic maintenance list, optimizing the allocation efficiency of maintenance resources, shortening fault response time, and reducing maintenance delays caused by unclear priorities in traditional methods.

[0045] In an optional embodiment, before calculating the frequency and duration of abnormalities of the N types of parameters of all abnormal devices in the historical abnormal event database, the method further includes: obtaining preset abnormal thresholds corresponding to the N types of parameters; comparing the N types of parameters of each abnormal device in the historical abnormal event database with the corresponding preset abnormal thresholds, and determining the abnormal state of each type of parameter based on the comparison results.

[0046] Optionally, the preset abnormal threshold refers to the numerical limit set in advance for each type of parameter within the normal operating range of the equipment, used to determine whether the parameter is abnormal.

[0047] Optionally, the abnormal status is used to determine whether the measured value of a certain type of parameter of a certain device at a certain time exceeds or falls below the corresponding preset abnormal threshold. If it exceeds, the parameter is determined to be in an abnormal state in that event; if it does not exceed, it is in a normal state. The abnormal status is the basic unit constituting the historical abnormal event database and is a prerequisite for subsequent statistics on abnormal frequency and processing time.

[0048] Optionally, before proceeding to the steps of calculating the frequency and duration of anomalies, the system first loads a preset anomaly threshold table from the configuration database. This table contains static thresholds for four types of parameters. Subsequently, the system verifies each record in the historical anomaly event database item by item to determine whether the measured value of each type of parameter in each record exceeds the corresponding preset anomaly threshold.

[0049] Optionally, the system introduces a preset anomaly threshold as a unified criterion before calculating the frequency and duration of anomalies, thereby transforming raw sensor data into structured, statistically verifiable anomaly event labels. Before threshold comparison, historical data is merely a raw numerical sequence and cannot be directly used for frequency statistics. By performing a binary judgment with the preset anomaly threshold, the system transforms ambiguous numerical fluctuations into clear event attributes, facilitating frequency statistics.

[0050] In one optional embodiment, the dynamic weight coefficient of each type of parameter is determined based on the anomaly frequency and the anomaly handling time, including: calculating the ratio of the anomaly frequency of each type of parameter to the total frequency of anomalies of all parameters in the historical anomaly event database to obtain the first data corresponding to each type of parameter; calculating the ratio of the anomaly handling time of each type of parameter to the total handling time of anomalies of all parameters in the historical anomaly event database to obtain the second data corresponding to each type of parameter; and multiplying the first data and the second data of each type of parameter to obtain the dynamic weight coefficient of that type of parameter.

[0051] Optionally, the first data refers to the proportion of the abnormal frequency of a certain type of parameter to the total abnormal frequency of all parameters in the historical abnormal event database. This value represents the relative frequency of the parameter occurring abnormally in all abnormal events.

[0052] Optionally, the second data refers to the proportion of the anomaly handling time of a certain type of parameter to the total anomaly handling time of all parameters in the historical anomaly event database. This value represents the average repair cost weight caused by the anomaly of that parameter.

[0053] Optionally, the dynamic weighting coefficient refers to the normalized weight value derived from the product of the first and second data mentioned above, used to quantify the comprehensive impact of this parameter on the overall abnormal state of the equipment in historical operation and maintenance. This coefficient is not fixed or subjectively set, but is dynamically calculated entirely from historical operation and maintenance behavior data. The larger the value, the more significant the comprehensive impact of this parameter in terms of both high frequency of occurrence and high repair cost.

[0054] Optionally, the formula for calculating the dynamic weighting coefficient is as shown in formula (1):

[0055] in, For the first The total frequency of anomalies in historical refrigeration equipment failures, (This parameter is not translated as it is not part of the main text.) For the first The total processing time for class parameters to return to normal after repair and maintenance when refrigeration equipment malfunctioned in the past. This represents the total frequency of malfunctions that have occurred in the refrigeration equipment throughout history. This represents the total processing time for a historical refrigeration unit to be repaired and restored to normal operation after experiencing a malfunction.

[0056] Optionally, the system employs a product model of frequency and duration to achieve a two-dimensional coupled assessment of the impact weights of various parameters: the frequency proportion reflects the parameter's susceptibility, characterizing how frequently it occurs in historical anomalies and revealing the sensitivity and degradation tendency of equipment components; the duration proportion reflects the parameter's severity, reflecting the time cost and resource investment required for each anomaly repair, revealing the parameter's substantial impact on system operation recovery. The product mechanism inherently possesses a non-linear reinforcement effect; the weight of a parameter only increases when it is frequently triggered and time-consuming to repair, thus identifying the core risk points that truly constitute high maintenance costs. For example, air outlet temperature is prone to exceeding limits and repairs are slow, resulting in the largest product and the highest weight; while energy consumption, though occasional, is repaired quickly, resulting in the smallest product and the lowest weight. Compared to traditional, crude methods such as simple averaging or maximum value normalization, which may misjudge noise as more important than vibration based solely on frequency, or underestimate high-frequency rapid repair temperature anomalies based solely on duration, this method achieves intelligent weight allocation by multiplicative coupling, using real operation and maintenance costs as the yardstick to measure the value of anomalies. This allows the system's decision-making logic to truly return to the essence of operation and maintenance, focusing not on where alarms are triggered, but on where repairs are slowest and most inefficient.

[0057] In an optional embodiment, before determining the device anomaly index of each target abnormal device in the abnormal device set based on the real-time collected abnormal device set and the dynamic weight coefficients of each type of parameter, the method further includes: real-time collection of device numbers and N types of parameters from multiple devices; denoising and standardizing the collected N types of parameters to obtain a running dataset with device numbers; calculating the device running index of each device based on the relevant parameters of each device in the running dataset, wherein the relevant parameters include at least the usage years data, the usage years threshold, the N types of parameters, and the preset anomaly thresholds of the N types of parameters for each device; obtaining a historical running index dataset and determining the device running threshold based on the historical running index dataset; comparing the device running index of each device with the device running threshold, and determining the device status of each device based on the comparison result.

[0058] Optionally, the equipment number is a unique identifier assigned to each cooling device in the computer room to distinguish the operating data of different devices, which helps to make the subsequent data ownership clear and traceable.

[0059] Optionally, the equipment operation index is a comprehensive evaluation value calculated using a specific formula based on the equipment's current operating parameters, its service life, and preset reference thresholds. It reflects the overall operational health of the equipment. A lower equipment operation index indicates that it is closer to the normal operating boundary; a higher index indicates that it deviates further from the normal state.

[0060] Optionally, the system extracts the device operation index of all devices identified as abnormal from the historical abnormal event database at the time of the abnormality, and takes the minimum value as the device operation threshold. For example, if the historical operation index dataset is {1.05, 1.18, 0.99, 1.21, 1.01}, and the minimum value is 0.99, then the device operation threshold is 0.99.

[0061] Optionally, the system can address the pain points of traditional methods, such as rampant false alarms and wasted resources, by introducing a comprehensive health assessment mechanism that integrates multiple parameters and age on a device-by-device basis.

[0062] Optionally, the system constructs an equipment operation index—a comprehensive health assessment indicator integrating multiple parameters (outlet air temperature, vibration intensity, energy consumption, noise) and equipment service life—to achieve a global and adaptive initial screening of the refrigeration equipment's operating status. This reduces the false alarm problems easily caused by traditional single-parameter threshold alarms. Even if a parameter slightly exceeds the standard, if other indicators are normal and the equipment is relatively new, it can still be judged as safe. Simultaneously, the equipment operation threshold is dynamically determined by the minimum operation index of historical real abnormal events, reducing the subjectivity and lag of manual settings and helping the judgment standard always align with the actual operation and maintenance environment. More importantly, by introducing a service life / standard service life factor, the system can identify performance degradation caused by long-term aging, allowing for the timely detection of faulty equipment with slight parameter fluctuations but nearing its lifespan limit. Ultimately, only a small number of devices with comprehensive risks exceeding the threshold are selected as target abnormal devices for subsequent high-level analysis, reducing system computational load and resource consumption, and achieving an intelligent operation and maintenance closed loop of accurate identification and focused processing.

[0063] In one optional embodiment, based on the relevant parameters of each device in the running dataset, the device operation index of each device is calculated, including: obtaining the service life, preset service life threshold, and preset abnormal thresholds corresponding to N types of parameters for each device; calculating the ratio of the service life of each device to the preset service life threshold to obtain a first ratio for each device; calculating the ratio of each type of parameter of each device to the corresponding preset abnormal threshold to obtain a second set of ratios for each device, wherein the second set of ratios for each device includes at least the ratio of the outlet air temperature to a preset temperature threshold, the ratio of vibration intensity to a preset vibration intensity threshold, the ratio of energy consumption to a preset energy consumption threshold, and the ratio of noise decibels to a preset noise decibel threshold for each device; and calculating the product of all ratios in the first and second set of ratios for each device to obtain the device operation index for each device.

[0064] Optionally, the first ratio refers to the ratio of the actual service life of the equipment to the preset service life threshold. This ratio is used to quantify the performance degradation trend of the equipment due to the extended service life. The closer the ratio is to or exceeds 1, the closer the equipment is to or has exceeded its service life, and the higher the risk.

[0065] Optionally, the second ratio set refers to a set of values ​​obtained by performing a ratio calculation on each type of parameter value currently collected by the device and its corresponding preset anomaly threshold. Each ratio reflects the degree of deviation of the current state of the parameter from the design baseline.

[0066] Optionally, the equipment operation index is a comprehensive index obtained by multiplying the first ratio by all ratios in the second set of ratios, used to characterize the degree of deviation of the overall operating status of a piece of equipment from its ideal healthy state. The larger the index, the more the equipment deviates from its normal operating range.

[0067] Optionally, the formula for calculating the equipment operating index is formula (2):

[0068] in, The outlet air temperature of the refrigeration equipment. The preset temperature for the refrigeration equipment, For the vibration intensity of refrigeration equipment, This is the preset lower limit value for the vibration intensity during normal operation of the refrigeration equipment. This is the preset upper limit of vibration intensity for normal operation of the refrigeration equipment. For the energy consumption of refrigeration equipment, The preset standard energy consumption value for refrigeration equipment. The noise level of the refrigeration equipment in decibels. This is the preset lower limit for the noise level in decibels during normal operation of the refrigeration equipment. This is the preset upper limit of the noise level in decibels for the refrigeration equipment during normal operation. The service life of refrigeration equipment. This refers to the preset standard service life of the refrigeration equipment.

[0069] In one optional embodiment, the device operation index of each device is compared with the device operation threshold, and the device status of each device is determined based on the comparison result, including: when the device operation index of the i-th device is greater than or equal to the device operation threshold, the i-th device is determined to be a target abnormal device, where i is an integer greater than or equal to 1; when the device operation index of the i-th device is less than the device operation threshold, the i-th device is determined to be a normal device.

[0070] In an optional embodiment, the device anomaly severity index ranking method further includes: integrating multiple devices identified as target anomalous devices into an anomalous device set.

[0071] Optionally, the target abnormal equipment refers to the refrigeration equipment identified as being in an abnormal state through the initial screening mechanism of equipment operation index in the above steps. The judgment criterion is: when the equipment operation index of the refrigeration equipment is greater than or equal to a preset equipment operation threshold, which is the minimum value of the historical abnormal equipment operation index, the refrigeration equipment is determined to be the target abnormal equipment. Each target abnormal equipment possesses the dual risk characteristics of aging service life and deviation of multiple parameters, thus qualifying for in-depth evaluation.

[0072] Optionally, the abnormal equipment set is a structured collection of all refrigeration equipment identified as the target abnormal equipment. The abnormal equipment set is not a simple list, but a unified data container containing metadata such as equipment number, real-time parameters, equipment operating index, and dynamic weighting coefficients required for subsequent calculations. Its core function is to centrally manage equipment objects that need further sorting, serving as the sole input source for subsequent calculations of the equipment abnormality index and the generation of the electronic maintenance list.

[0073] In one optional embodiment, based on the real-time collected set of abnormal devices and the dynamic weighting coefficients of each type of parameter, the device abnormality index of each target abnormal device in the set of abnormal devices is determined, including: calculating the absolute value of the difference between each type of parameter of each target abnormal device in the set of abnormal devices and the corresponding preset abnormal threshold, to obtain the deviation of each type of parameter; calculating the ratio of the deviation of each type of parameter to the corresponding preset abnormal threshold, to obtain the standard deviation value of each type of parameter; multiplying the standard deviation value of each type of parameter by the corresponding dynamic weighting coefficient, to obtain the weighted deviation value of each type of parameter; and summing all the weighted deviation values ​​of the same target abnormal device to obtain the device abnormality index of the target abnormal device.

[0074] Optionally, the deviation refers to the absolute difference between the real-time measured value of a certain parameter of a device and the corresponding preset abnormal threshold, used to quantify the physical extent to which the parameter deviates from the normal range. The deviation is an absolute value, reflecting only the degree of exceedance, without distinguishing between being too high or too low.

[0075] Optionally, the standard deviation value refers to the ratio of the deviation amount to the preset abnormal threshold, which is used to reduce the interference of different parameter units and magnitudes and achieve the uniformity of comparability.

[0076] Optionally, the weighted deviation value refers to the product of the standard deviation value of a certain type of parameter and the dynamic weight coefficient, representing the weighted contribution of that parameter to the overall severity of equipment anomalies. The larger the weighted deviation value, the more significant the deviation of the parameter is, and the more important it is in terms of operation and maintenance costs.

[0077] Optionally, the equipment anomaly severity index refers to the sum of weighted deviations of all N types of parameters for a target malfunctioning equipment, and is the final quantitative score of the overall severity of the equipment anomaly. The higher the index value, the more severe the equipment anomaly, and the higher the maintenance priority should be. It is the sole basis for generating the equipment anomaly ranking list.

[0078] Optionally, the formula for calculating the equipment anomaly index is formula (3):

[0079] (3)

[0080] in, The weighting coefficients for the outlet air temperature data are: These are the weighting coefficients for the vibration intensity data. These are the weighting coefficients for energy consumption data. The weighting coefficients for the noise decibel data are... The outlet air temperature of the refrigeration equipment. The preset temperature for the refrigeration equipment, For the vibration intensity of refrigeration equipment, This is the preset lower limit value for the vibration intensity during normal operation of the refrigeration equipment. This is the preset upper limit of vibration intensity for normal operation of the refrigeration equipment. For the energy consumption of refrigeration equipment, The preset standard energy consumption value for refrigeration equipment. The noise level of the refrigeration equipment in decibels. This is the preset lower limit for the noise level in decibels during normal operation of the refrigeration equipment. This is the preset upper limit of the noise level in decibels for the refrigeration equipment during normal operation. The service life of refrigeration equipment. This refers to the preset standard service life of the refrigeration equipment.

[0081] Optionally, the system constructs a method for quantifying equipment anomaly severity in industrial-grade operation and maintenance scenarios by calculating deviation, standard deviation, weighted deviation, and total index through a four-step progressive model. This method combines physical interpretability with adaptive weighting. Specifically, replacing absolute values ​​with relative deviation reduces unit interference and enables fair assessment across parameters; replacing fixed weights with dynamic weights allows the assessment logic to continuously evolve with historical operation and maintenance data, eliminating the need for manual weighting based on experience; replacing maximum value priority with weighted summation reduces the focus on a single anomaly parameter while ignoring overall system risk; and outputting a single index enables a seamless conversion from complex, multi-dimensional data to operable maintenance priorities.

[0082] Figure 2 This is a schematic diagram of an optional equipment anomaly degree index sorting device according to an embodiment of this application. According to another aspect of an embodiment of this application, an equipment anomaly degree index sorting device is also provided, including: a statistics unit 201, a first determination unit 202, a second determination unit 203, and a sorting unit 204.

[0083] The system includes: a statistics unit 201, used to count the frequency and duration of anomalies in N types of parameters of all abnormal devices in the historical anomaly event database; where N types of parameters include at least outlet air temperature, vibration intensity, energy consumption, and noise decibels; anomaly frequency represents the total number of times each type of parameter has anomalies in the historical anomaly event database; and anomaly handling duration represents the total handling time for each type of parameter to be restored to normal operation after maintenance in the historical anomaly event database; a first determination unit 202, used to determine the dynamic weight coefficient of each type of parameter based on the anomaly frequency and anomaly handling duration; a second determination unit 203, used to determine the equipment anomaly severity index of each target abnormal device in the abnormal device set based on the real-time collected abnormal device set and the dynamic weight coefficient of each type of parameter; and a sorting unit 204, used to sort all target abnormal devices in the abnormal device set according to the equipment anomaly severity index of each target abnormal device, generating an equipment anomaly sorting list.

[0084] Optionally, the equipment anomaly degree index ranking device further includes: a first acquisition unit, used to acquire preset anomaly thresholds corresponding to N types of parameters respectively; and a third determination unit, used to compare the N types of parameters of each abnormal device in the historical anomaly event database with the corresponding preset anomaly thresholds respectively, and determine the anomaly state of each type of parameter based on the comparison results.

[0085] Optionally, the first determining unit 202 includes: a first calculation subunit, used to calculate the ratio of the abnormal frequency of each type of parameter to the total frequency of abnormal occurrences of all parameters in the historical abnormal event database, to obtain first data corresponding to each type of parameter; a second calculation subunit, used to calculate the ratio of the abnormal processing time of each type of parameter to the total processing time of abnormal occurrences of all parameters in the historical abnormal event database, to obtain second data corresponding to each type of parameter; and a third calculation subunit, used to multiply the first data and the second data of each type of parameter to obtain the dynamic weight coefficient of that type of parameter.

[0086] Optionally, the equipment anomaly index ranking device further includes: a data acquisition unit for real-time acquisition of equipment numbers and N types of parameters from multiple devices; a processing unit for denoising and standardizing the acquired N types of parameters to obtain a running dataset with equipment numbers; a calculation unit for calculating the equipment running index of each device based on the relevant parameters of each device in the running dataset, wherein the relevant parameters include at least the usage years data, the usage years threshold, the N types of parameters, and the preset anomaly thresholds of the N types of parameters for each device; a second acquisition unit for acquiring a historical running index dataset and determining the equipment running threshold based on the historical running index dataset; and a fourth determination unit for comparing the equipment running index of each device with the equipment running threshold and determining the equipment status of each device based on the comparison result.

[0087] Optionally, the calculation unit further includes: an acquisition subunit for acquiring the service life, preset service life threshold, and preset abnormal thresholds corresponding to N types of parameters for each device; a first calculation subunit for calculating the ratio of the service life of each device to the preset service life threshold to obtain a first ratio for each device; a second calculation subunit for calculating the ratio of each type of parameter of each device to the corresponding preset abnormal threshold to obtain a second set of ratios for each device, wherein the second set of ratios for each device includes at least the ratio of the outlet air temperature to a preset temperature threshold, the ratio of vibration intensity to a preset vibration intensity threshold, the ratio of energy consumption to a preset energy consumption threshold, and the ratio of noise decibels to a preset noise decibel threshold for each device; and a third calculation subunit for calculating the product of all ratios in the first and second set of ratios for each device to obtain the device operation index for each device.

[0088] Optionally, the fourth determining unit further includes: a first determining subunit, used to determine the i-th device as a target abnormal device when the device operation index of the i-th device is greater than or equal to the device operation threshold, where i is an integer greater than or equal to 1; and a second determining subunit, used to determine the i-th device as a normal device when the device operation index of the i-th device is less than the device operation threshold.

[0089] Optionally, the equipment anomaly degree index ranking device further includes: integrating multiple devices identified as target abnormal devices into an abnormal device set.

[0090] Optionally, the second determining unit 203 includes: a first calculation subunit, used to calculate the absolute value of the difference between each type of parameter of each target abnormal device in the abnormal device set and the corresponding preset abnormal threshold, to obtain the deviation of each type of parameter; a second calculation subunit, used to calculate the ratio of the deviation of each type of parameter to the corresponding preset abnormal threshold, to obtain the standard deviation value of each type of parameter; a third calculation subunit, used to multiply the standard deviation value of each type of parameter by the corresponding dynamic weight coefficient, to obtain the weighted deviation value of each type of parameter; and a fourth calculation subunit, used to sum all the weighted deviation values ​​of the same target abnormal device to obtain the device abnormality index of the target abnormal device.

[0091] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, which stores a computer program, wherein when the computer program is executed, it causes the device where the computer-readable storage medium is located to execute the above-described device anomaly index ranking method.

[0092] According to another aspect of the embodiments of this application, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors cause the one or more processors to perform the above-described device anomaly index ranking method.

[0093] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program or instructions, which, when executed by a processor, implement the above-described device anomaly index ranking method.

[0094] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0095] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0096] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0097] 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 units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0098] Furthermore, the functional units in the various embodiments of this application 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0099] If the integrated unit 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 application, in essence, or the part that contributes to the prior art, or all or 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 described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0100] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method of ranking a device anomaly severity index, the method comprising: include: The frequency of anomalies and the duration of handling anomalies for all N types of abnormal equipment in the historical anomaly event database are statistically analyzed. The N types of parameters include at least air outlet temperature, vibration intensity, energy consumption, and noise decibels. The frequency of anomalies represents the total number of times each type of parameter has anomalies in the historical anomaly event database, and the duration of handling anomalies represents the total handling time for each type of parameter to be restored to normal operation after maintenance in the historical anomaly event database. Based on the anomaly frequency and the anomaly handling duration, the dynamic weight coefficients for each type of parameter are determined, including: Calculate the ratio of the abnormal frequency of each type of parameter to the total frequency of abnormal occurrences of all parameters in the historical abnormal event database to obtain the first data corresponding to each type of parameter; Calculate the ratio of the anomaly handling time for each type of parameter to the total handling time for all parameters in the historical anomaly event database to obtain the second data corresponding to each type of parameter; Multiply the first and second data of each type of parameter to obtain the dynamic weight coefficient of that type of parameter; Based on the real-time collected abnormal device set and the dynamic weight coefficient of each type of parameter, the device abnormality index of each target abnormal device in the abnormal device set is determined. Based on the equipment anomaly severity index of each target abnormal device, all target abnormal devices in the abnormal device set are sorted to generate an equipment anomaly sorting list; The method further includes: calculating the device operation index of each device based on the relevant parameters of each device in the running dataset. The device operation index is a comprehensive evaluation value calculated by a specific formula based on the current operating parameters of the device and the device's own service life, combined with a preset reference threshold, to reflect the overall operational health of the device. This includes: obtaining the service life of each device, a preset service life threshold, and a preset abnormal threshold corresponding to the N types of parameters. Calculate the ratio of the service life of each device to a preset service life threshold to obtain the first ratio for each device; Calculate the ratio of each type of parameter of each device to the corresponding preset abnormal threshold to obtain a second set of ratios for each device. The second set of ratios for each device includes at least the ratio of the outlet air temperature of each device to a preset temperature threshold, the ratio of the vibration intensity to a preset vibration intensity threshold, the ratio of the energy consumption to a preset energy consumption threshold, and the ratio of the noise decibels to a preset noise decibel threshold. The device operation index of each device is obtained by multiplying the first ratio and the product of all ratios in the second ratio set for each device.

2. The method of claim 1, wherein, Before calculating the frequency and duration of anomalies in the N types of parameters of all abnormal devices in the historical anomaly event database, the method further includes: Obtain the preset abnormal thresholds corresponding to the N types of parameters respectively; The N types of parameters for each abnormal device in the historical abnormal event database are compared with the corresponding preset abnormal threshold, and the abnormal state of each type of parameter is determined based on the comparison results.

3. The method of claim 1, wherein, Before determining the device anomaly severity index for each target anomaly device in the set of anomaly devices based on the real-time collected set of anomaly devices and the dynamic weighting coefficients of each type of parameter, the method further includes: Real-time acquisition of device IDs and the N types of parameters from multiple devices; The collected N types of parameters are denoised and standardized to obtain a running dataset with device numbers; Based on the relevant parameters of each device in the running dataset, the device running index of each device is calculated, wherein the relevant parameters include at least the usage years data of each device, the usage years threshold, the N types of parameters, and the preset abnormal threshold of the N types of parameters; Obtain a historical operating index dataset and determine the device operating threshold based on the historical operating index dataset; The device operation index of each device is compared with the device operation threshold, and the device status of each device is determined based on the comparison result.

4. The method according to claim 3, characterized in that, The device operation index of each device is compared with the device operation threshold, and the device status of each device is determined based on the comparison result, including: When the device operation index of the i-th device is greater than or equal to the device operation threshold, the i-th device is determined to be a target abnormal device, where i is an integer greater than or equal to 1; When the device operation index of the i-th device is less than the device operation threshold, the i-th device is determined to be a normal device.

5. The method of claim 4, wherein, The method further includes: integrating multiple devices identified as target abnormal devices into the abnormal device set.

6. The method of claim 1, wherein, Based on the real-time collected set of abnormal devices and the dynamic weighting coefficients of each type of parameter, the device abnormality index of each target abnormal device in the set of abnormal devices is determined, including: Calculate the absolute value of the difference between each type of parameter of each target abnormal device in the abnormal device set and the corresponding preset abnormal threshold to obtain the deviation of each type of parameter; Calculate the ratio of the deviation of each type of parameter to the corresponding preset abnormal threshold to obtain the standard deviation value of each type of parameter; The standard deviation of each parameter is multiplied by the corresponding dynamic weight coefficient to obtain the weighted deviation of each parameter. The weighted deviation values ​​of all the abnormal devices for the same target are summed to obtain the device abnormality index.

7. An apparatus for ranking a degree of abnormality index, characterized by comprising: include: The statistics unit is used to count the frequency and duration of abnormality of N types of parameters of all abnormal devices in the historical abnormal event database. The N types of parameters include at least air outlet temperature, vibration intensity, energy consumption and noise decibels. The frequency of abnormality represents the total number of times each type of parameter has an abnormality in the historical abnormal event database. The duration of abnormality handling represents the total handling time for each type of parameter to be restored to normal operation after maintenance in the historical abnormal event database. The first determining unit is used to determine the dynamic weight coefficient of each type of parameter according to the anomaly frequency and the anomaly handling duration. The first determining unit includes: a first calculation subunit, used to calculate the ratio of the anomaly frequency of each type of parameter to the total frequency of anomalies of all parameters in the historical anomaly event database, to obtain first data corresponding to each type of parameter; a second calculation subunit, used to calculate the ratio of the anomaly processing time of each type of parameter to the total processing time of anomalies of all parameters in the historical anomaly event database, to obtain second data corresponding to each type of parameter; and a third calculation subunit, used to multiply the first data and the second data of each type of parameter to obtain the dynamic weight coefficient of that type of parameter. The second determining unit is used to determine the device abnormality index of each target abnormal device in the abnormal device set based on the real-time collected abnormal device set and the dynamic weight coefficient of each type of parameter. The sorting unit is used to sort all the target abnormal devices in the abnormal device set according to the device abnormality degree index of each target abnormal device, and generate a device abnormality sorting list; The device further includes: a calculation unit, used to calculate the device operation index of each device based on relevant parameters of each device in the operational dataset, wherein the device operation index is a comprehensive evaluation value calculated by a specific formula based on the current operating parameters of the device and its own service life, combined with a preset reference threshold, to reflect the overall operational health of the device, including: The acquisition subunit is used to acquire the service life, preset service life threshold, and preset abnormal threshold corresponding to the N types of parameters for each device. The first calculation subunit is used to calculate the ratio of the service life of each device to a preset service life threshold, and obtain the first ratio of each device; The second calculation subunit is used to calculate the ratio of each type of parameter of each device to the corresponding preset abnormal threshold, and obtain the second ratio set of each device. The second ratio set of each device includes at least the ratio of the outlet air temperature of each device to the preset temperature threshold, the ratio of the vibration intensity to the preset vibration intensity threshold, the ratio of the energy consumption to the preset energy consumption threshold, and the ratio of the noise decibel to the preset noise decibel threshold. The third calculation subunit is used to calculate the product of all ratios in the first and second ratio sets for each device to obtain the device operation index for each device.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed, the device on which the computer-readable storage medium is located performs the device anomaly index ranking method according to any one of claims 1 to 6.

9. An electronic device, comprising: It includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to perform the device anomaly index ranking method according to any one of claims 1 to 6.

10. A computer program product, characterised in that, Includes a computer program or instructions that, when executed by a processor, implement the device anomaly index ranking method according to any one of claims 1 to 6.

Images