Equipment early warning method and device based on business data association, electronic equipment and storage medium

Abstract

The invention provides an equipment early warning method and device based on business data association, electronic equipment and a storage medium. The method comprises the following steps: acquiring real-time operation data and current business scene data of target equipment; screening out historical effective operation data matched with the current service scene from the historical data, and constructing an operation baseline model and a rate change model according to the historical effective operation data; comparing the real-time operation data with the operation baseline model to obtain a first deviation result; calculating a real-time change rate of the target equipment according to the real-time operation data, and comparing the real-time change rate with a rate change model to obtain a second deviation result; and generating target equipment early warning information according to the first deviation result and the second deviation result. According to the method and the device, the association relationship between the business data and the equipment operation is established, and the historical data analysis based on business filtering and the data change rate analysis based on business association are comprehensively judged, so that the accuracy and the reliability of equipment early warning are improved.

Description

Technical Field

[0001] This application relates to the field of intelligent operation and maintenance technology, and in particular to a device and apparatus for early warning of equipment based on business data correlation, as well as electronic equipment and storage media. Background Technology

[0002] In the field of intelligent operation and maintenance of buildings or industrial facilities, equipment early warning systems are a key technology for ensuring business continuity and avoiding major losses. The core task of such systems is to identify early signs of operational abnormalities before serious alarms or malfunctions occur, so that operation and maintenance personnel can intervene in a timely manner.

[0003] In related technologies, early warning systems often establish predictive models of equipment operating data, issuing warnings when real-time monitoring data deviates from these models. However, such methods have significant technical drawbacks in practical applications: First, they are not linked to business operations, making it impossible to distinguish whether fluctuations in equipment parameters are caused by actual faults or normal changes in business scenarios. This leads to an inability to make accurate predictions when the equipment's user changes, resulting in numerous false alarms. Second, the predictive models of these methods are relatively simplistic, relying on monitoring the absolute values ​​of parameters. This dependence on a single dimension makes predictions inaccurate and unreliable when facing complex, multi-factor-influenced early-stage faults. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application provides a device and apparatus for equipment early warning based on business data correlation, an electronic device, and a storage medium. This application improves the accuracy and reliability of equipment early warning by establishing a correlation between business data and equipment operation, and on this basis, comprehensively judging by combining historical data analysis based on business filtering and data change rate analysis based on business correlation.

[0005] In a first aspect, embodiments of this application provide a device early warning method based on business data association, including: Acquire real-time operational data of the target device and current business scenario data of the target device; Filter out historical valid operating data that matches the current business scenario from historical data, and construct an operating baseline model and a rate change model based on the historical valid operating data; The real-time operating data is compared with the operating baseline model to obtain the first deviation result; The real-time change rate of the target device is calculated based on the real-time operating data, and the real-time change rate is compared with the rate change model to obtain a second deviation result; Based on the first deviation result and the second deviation result, a warning message for the target device is generated.

[0006] In some embodiments, the step of filtering historical valid operational data that matches the current business scenario from historical data, and constructing an operational baseline model and a rate change model based on the historical valid operational data, includes: Based on the current business scenario data, historical business scenarios are determined from the historical data, and the historical data corresponding to the historical business scenarios are determined as the historical valid operating data; Statistical analysis is performed on the historical valid operating data to determine the normal numerical range of the historical valid operating data, thereby obtaining the operating baseline model; The historical rate of change is calculated based on the historical valid operating data, and statistical analysis is performed on the historical rate of change to determine the normal rate range of the historical rate of change, which serves as the rate change model.

[0007] In some embodiments, the current business scenario data includes at least one of time-related impact data, weather-related impact data, seasonal-related impact data, personnel-related impact data, enterprise-related impact data, vehicle-related impact data, and temporary activity-related impact data. The step of determining historical business scenarios from the historical data based on the current business scenario data, and determining the historical data corresponding to the historical business scenarios as the historical valid operational data, includes: The current business scenario data is combined and quantized to obtain the current business scenario feature vector; Construct corresponding historical business scenario feature vectors for the historical data at each historical moment; Calculate the multidimensional similarity between the current business scenario feature vector and the feature vectors of each of the historical business scenarios; Based on a preset matching threshold, target historical business scenario feature vectors are selected from each of the historical business scenario feature vectors, and the historical moment corresponding to the target historical business scenario feature vector is determined as the historical business scenario. Extract the historical operational data corresponding to the historical business scenarios as the historical valid operational data.

[0008] In some embodiments, the step of performing statistical analysis on the historical valid operating data to determine the normal numerical range of the historical valid operating data and obtain the operating baseline model includes: Calculate the mean and standard deviation of the historical valid operating data; Based on the mean and the standard deviation, and in conjunction with a preset multiple, the upper and lower limits of the normal value range are determined; wherein, the upper limit is obtained by adding the product of the standard deviation and the preset multiple to the mean, and the lower limit is obtained by subtracting the product of the standard deviation and the preset multiple from the mean; The numerical range defined by the lower limit and the upper limit is determined as the normal numerical range and used as the operating baseline model.

[0009] In some embodiments, the step of calculating the historical rate of change based on the historical valid operating data, performing statistical analysis on the historical rate of change, and determining the normal rate range of the historical rate of change as the rate change model includes: Calculate the mean and standard deviation of the historical rate of change; Based on the mean and standard deviation of the historical rates of change, and in conjunction with a preset rate multiple, the upper and lower limits of the normal rate range are determined; wherein, the upper limit is obtained by adding the product of the mean and standard deviation of the historical rates of change and the preset rate multiple to the mean of the historical rates of change, and the lower limit is obtained by subtracting the product of the standard deviation and the preset rate multiple from the mean of the historical rates of change; The rate range defined by the lower rate limit and the upper rate limit is determined as the normal rate range and used as the rate change model.

[0010] In some embodiments, the step of calculating the real-time change rate of the target device based on the real-time operating data, comparing the real-time change rate with the rate change model, and obtaining a second deviation result includes: The real-time rate of change is obtained by performing differential calculations on multiple data sampling points of the real-time running data within a predetermined time window; Determine whether the real-time rate of change is within the normal rate range defined by the lower limit and the upper limit. If the real-time rate of change is outside the normal rate range, then the second deviation result is determined to be abnormal; if the real-time rate of change is within the normal rate range, then the second deviation result is determined to be normal.

[0011] In some embodiments, generating target device warning information based on the first deviation result and the second deviation result includes: If at least one of the first deviation result is abnormal or the second deviation result is abnormal, it is determined that the target device has abnormal signs; If the target device is found to exhibit abnormal signs, an early warning message for the target device is generated and sent.

[0012] Secondly, embodiments of this application provide a device early warning apparatus based on business data association, comprising: The acquisition module is used to acquire the real-time operating data of the target device and the current business scenario data of the target device; The filtering module is used to filter out historical valid operating data that matches the current business scenario from historical data, and to construct an operating baseline model and a rate change model based on the historical valid operating data; The first comparison module is used to compare the real-time running data with the running baseline model to obtain a first deviation result; The second comparison module is used to calculate the real-time change rate of the target device from the real-time operating data, compare the real-time change rate with the rate change model, and obtain a second deviation result. The generation module is used to generate early warning information for the target device based on the first deviation result and the second deviation result.

[0013] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the device early warning method based on business data association as described in any one of the embodiments of the first aspect of this application.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing a program that is executed by a processor to implement the device early warning method based on business data association as described in any one of the embodiments of the first aspect of this application.

[0015] The device early warning method based on business data association according to the embodiments of this application has at least the following beneficial effects: The device early warning method based on business data association according to the embodiments of this application includes: acquiring real-time operating data of the target device and current business scenario data of the target device; filtering out historically valid operating data that matches the current business scenario from historical data, and constructing an operating baseline model and a rate change model based on the historically valid operating data; comparing the real-time operating data with the operating baseline model to obtain a first deviation result; calculating the real-time change rate of the target device based on the real-time operating data, comparing the real-time change rate with the rate change model to obtain a second deviation result; and generating early warning information for the target device based on the first deviation result and the second deviation result.

[0016] This application first acquires real-time operational data from the target device and incorporates current business scenario data, providing clear and realistic data input for subsequent refined, scenario-aware early warning judgments, thus forming a reliable analysis foundation. Then, based on the current business scenario data, matching historical valid operational data is selected from historical data. Based on this, an operational baseline model for static numerical comparison and a rate change model for dynamic trend comparison are constructed, overcoming the shortcomings of related technologies where early warning models are not linked to business operations, failing to distinguish between real faults and normal business fluctuations, leading to numerous false alarms when business scenarios change. Next, a first deviation result is obtained by comparing real-time operational data with the operational baseline model, and a second deviation result is obtained by comparing the calculated real-time change rate with the rate change model, achieving a multi-dimensional quantitative assessment of the device status. This multi-dimensional analysis method overcomes the shortcomings of related technologies that rely solely on single parameter absolute value monitoring and are inaccurate in predicting complex early faults. Finally, the first and second deviation results are comprehensively judged, and the final early warning information is generated based on this. The resulting early warning result comprehensively reflects the true operational health of the device under specific business scenarios. This application establishes a correlation between business data and equipment operation, and on this basis, comprehensively judges the analysis of historical data based on business filtering and the analysis of data change rate based on business correlation, thereby improving the accuracy and reliability of equipment early warning.

[0017] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0018] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 A flowchart illustrating an optional device early warning method based on business data association, provided for an embodiment of this application; Figure 2 A flowchart of another optional device early warning method based on business data association provided for embodiments of this application; Figure 3 A flowchart of another optional device early warning method based on business data association provided for embodiments of this application; Figure 4 A flowchart of another optional device early warning method based on business data association provided for embodiments of this application; Figure 5 A flowchart of another optional device early warning method based on business data association provided for embodiments of this application; Figure 6 A flowchart of another optional device early warning method based on business data association provided for embodiments of this application; Figure 7 A flowchart of another optional device early warning method based on business data association provided for embodiments of this application; Figure 8 A schematic diagram of a device early warning device based on business data association provided in an embodiment of this application; Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0019] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0020] In the description of this application, "several" means one or more, "more than" means two or more, "greater than," "less than," "exceeding," etc. are understood to exclude the stated number, while "above," "below," "within," etc. are understood to include the stated number. Where "first" or "second" is mentioned, it is only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the sequential relationship of the indicated technical features.

[0021] In the description of this application, it should be understood that the orientation descriptions, such as up, down, left, right, front, and back, are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0022] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0023] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setting," "installation," and "connection" should be interpreted broadly. Those skilled in the art can reasonably determine the specific meaning of the above terms in this application based on the specific content of the technical solution. Furthermore, the identification of specific steps in the following text does not imply a limitation on the order of steps or execution logic. The execution order and logic between each step should be understood and inferred from the content described in the embodiments.

[0024] In the field of intelligent operation and maintenance of buildings or industrial facilities, equipment early warning systems are a key technology for ensuring business continuity and avoiding major losses. The core task of such systems is to identify early signs of operational abnormalities before serious alarms or malfunctions occur, so that operation and maintenance personnel can intervene in a timely manner.

[0025] In related technologies, early warning systems often establish predictive models of equipment operating data, issuing warnings when real-time monitoring data deviates from these models. However, such methods have significant technical drawbacks in practical applications: First, they are not linked to business operations, making it impossible to distinguish whether fluctuations in equipment parameters are caused by actual faults or normal changes in business scenarios. This leads to an inability to make accurate predictions when the equipment's user changes, resulting in numerous false alarms. Second, the predictive models of these methods are relatively simplistic, relying on monitoring the absolute values ​​of parameters. This dependence on a single dimension makes predictions inaccurate and unreliable when facing complex, multi-factor-influenced early-stage faults.

[0026] Based on this, this application first acquires real-time operational data of the target device and incorporates current business scenario data, providing clear and actual business condition-compliant data input for subsequent refined and scenario-aware early warning judgments, thus forming a reliable analysis foundation. Then, based on the current business scenario data, matching historical valid operational data is selected from historical data. Based on this, an operational baseline model for static numerical comparison and a rate change model for dynamic trend comparison are constructed, overcoming the shortcomings of related technologies where early warning models are not linked to business operations, cannot distinguish between real faults and normal business fluctuations, and thus generate numerous false alarms when business scenarios change. Next, by comparing real-time operational data with the operational baseline model to obtain the first deviation result, and by comparing the calculated real-time change rate with the rate change model to obtain the second deviation result, a multi-dimensional quantitative assessment of the device status is achieved. This multi-dimensional analysis method overcomes the shortcomings of related technologies that rely solely on the absolute value monitoring of a single parameter and are inaccurate in predicting complex early faults. Finally, the first and second deviation results are comprehensively evaluated, and the final warning information is generated based on this. The resulting warning result can comprehensively reflect the actual operational health of the equipment under specific business scenarios. This application improves the accuracy and reliability of equipment warnings by establishing a correlation between business data and equipment operation, and on this basis, comprehensively evaluating historical data analysis based on business filtering and data change rate analysis based on business correlation.

[0027] Please see Figure 1 The present invention provides a device early warning method based on business data association, which may include, but is not limited to, the following steps 101 to 105: Step 101: Obtain the real-time operating data of the target device and the current business scenario data of the target device.

[0028] Step 102: Select historical valid operating data that matches the current business scenario from historical data, and build an operating baseline model and a rate change model based on the historical valid operating data.

[0029] Step 103: Compare the real-time running data with the running baseline model to obtain the first deviation result.

[0030] Step 104: Calculate the real-time change rate of the target device based on the real-time operating data, compare the real-time change rate with the rate change model, and obtain the second deviation result.

[0031] Step 105: Generate early warning information for the target equipment based on the first deviation result and the second deviation result.

[0032] Steps 101 to 105, as illustrated in this embodiment, firstly, real-time operating data of the target device is acquired, and current business scenario data of the target device is introduced. This provides clear and actual business condition-compliant data input for subsequent refined and scenario-aware early warning judgments, forming a foundation for reliable analysis. Then, based on the current business scenario data, matching historical valid operating data is selected from historical data. Based on this, an operating baseline model for static numerical comparison and a rate change model for dynamic trend comparison are constructed, overcoming the shortcomings of related technologies where early warning models are not linked to business operations, cannot distinguish between real faults and normal business fluctuations, and thus generate a large number of false alarms when business scenarios change. Next, a first deviation result is obtained by comparing real-time operating data with the operating baseline model, and a second deviation result is obtained by comparing the calculated real-time change rate with the rate change model, achieving a multi-dimensional quantitative assessment of the device status. This multi-dimensional analysis method overcomes the shortcomings of related technologies that rely solely on the absolute value monitoring of a single parameter and are inaccurate in predicting complex early faults. Finally, the first and second deviation results are comprehensively evaluated, and the final warning information is generated based on this. The resulting warning result can comprehensively reflect the actual operational health of the equipment under specific business scenarios. This application improves the accuracy and reliability of equipment warnings by establishing a correlation between business data and equipment operation, and on this basis, comprehensively evaluating historical data analysis based on business filtering and data change rate analysis based on business correlation.

[0033] In step 101 of some embodiments, the data acquisition layer needs to reliably acquire two types of data synchronously. The first type is real-time operating data of the target device, which is usually high-frequency time-series data from underlying sensors or device controllers, such as air conditioner current, compressor vibration frequency, and air outlet temperature. This data reflects the physical operating status of the device. The second type is current business scenario data. Business scenario data refers to factors other than the device itself and comes from various sources, including outdoor ambient temperature, building occupancy density, current time, and whether there are special activities (such as meetings) in progress. The synchronous acquisition of these two types of data is the basis for subsequent correlation analysis.

[0034] In step 102 of some embodiments, a computationally intensive data preprocessing and modeling task is performed. First, scenario matching and filtering are performed on massive amounts of historical data based on the current business scenario data obtained in step 101 (e.g., "weekdays, morning peak, outdoor high temperature"). The goal of this filtering process is to construct a homogeneous historical effective operational dataset, that is, to retain only historical data generated under similar business scenarios and remove all data from irrelevant scenarios (such as nighttime, holidays). This step is the core of eliminating business interference. Subsequently, based on this scenario-specific dataset, two models are constructed in parallel: a baseline model, which uses statistical analysis to quantify the normal static range of equipment operating parameters under this specific scenario; and a rate change model, which quantifies the normal dynamic rate of change of parameters under this specific scenario by analyzing the time series derivatives of the data.

[0035] Please see Figure 2 In some embodiments, step 102 may include, but is not limited to, steps 201 to 203.

[0036] Step 201: Based on the current business scenario data, determine the historical business scenarios from the historical data, and determine the historical data corresponding to the historical business scenarios as the historical valid operating data.

[0037] Step 202: Perform statistical analysis on the historical valid operating data to determine the normal numerical range of the historical valid operating data and obtain the operating baseline model.

[0038] Step 203: Calculate the historical rate of change based on historical valid operating data, perform statistical analysis on the historical rate of change, and determine the normal rate range of the historical rate of change as the rate of change model.

[0039] In step 201 of some embodiments, the current business scenario data is first parsed, such as a commercial complex under a specific scenario of "3 PM on a weekday, outdoor temperature 30°C, and moderate foot traffic." Next, using these business scenario characteristics as indexes, all matching time periods for historical business scenarios are retrieved and identified from a massive historical database. The core of this process is to extract and aggregate the historical operating data of the equipment corresponding to these matching historical time periods (e.g., chiller power consumption, stress, etc.) to form a data subset. This data subset, filtered by business scenarios, is determined as historically valid operating data. Historically valid operating data represents the equipment's past normal performance under similar business loads, while data from all non-similar business scenarios (e.g., "early morning, low temperature, no one around") are excluded, ensuring that the data foundation for subsequent modeling is homogeneous and relevant.

[0040] Please see Figure 3In some embodiments, the current business scenario data includes at least one of time-related impact data, weather-related impact data, seasonal impact data, personnel-related impact data, enterprise-related impact data, vehicle-related impact data, and temporary activity-related impact data. Step 201 may include, but is not limited to, steps 301 to 305.

[0041] Step 301: Combine and quantize the data of the current business scenario to obtain the feature vector of the current business scenario.

[0042] Step 302: Construct corresponding historical business scenario feature vectors for historical data at each historical moment.

[0043] Step 303: Calculate the multidimensional similarity between the feature vector of the current business scenario and the feature vectors of each historical business scenario.

[0044] Step 304: Based on a preset matching threshold, select the target historical business scenario feature vector from the feature vectors of each historical business scenario, and determine the historical moment corresponding to the target historical business scenario feature vector as the historical business scenario.

[0045] Step 305: Extract the historical operational data corresponding to the historical business scenarios as historical valid operational data.

[0046] In step 301 of some embodiments, the current business scenario data is combined and quantized. This business scenario data may specifically include one or more of the following: time-related data (e.g., weekdays, Saturdays, public holidays), weather-related data (e.g., sunny days, rainy days, typhoons), seasonal-related data (e.g., spring, summer), personnel-related data (e.g., number of people in a building), enterprise-related data (e.g., new enterprise entry), vehicle-related data (e.g., number of vehicles in a building), and temporary activity-related data (e.g., fire drills). This combined quantization process converts these heterogeneous data of different types and units into a unified mathematical representation. For example, categorical data (e.g., "weekdays" or "summer") can be processed using one-hot encoding, while numerical data (e.g., "number of people in a building") can be processed using normalization. All these quantified values ​​are combined into a high-dimensional vector, namely the current business scenario feature vector.

[0047] In step 302 of some embodiments, a quantization process consistent with step 301 is performed on the historical data of each historical moment in the historical database. That is, for each historical time point, various business scenario data corresponding to that time are extracted (e.g., it was a "workday", the weather was "rainy", the number of people was "300", etc.), and the same combination quantization method is applied to construct their respective historical business scenario feature vectors. After this step is completed, a large feature vector library corresponding one-to-one with historical moments is formed. This library is a digital and vectorized archive of all historical business scenarios (such as time, weather, personnel, activities, etc.) that have affected the operation of the equipment.

[0048] In step 303 of some embodiments, the current business scenario feature vector obtained in step 301 is compared with the feature vectors of each historical business scenario in the vector library generated in step 302, and their similarity is calculated one by one. Multidimensional similarity calculation evaluates the degree of closeness between the business scenarios represented by two vectors in a high-dimensional feature space. As mentioned earlier, different times, weather, seasons, number of people, enterprises, vehicles, or temporary activities will all have different impacts on the operational requirements of the equipment. Therefore, the purpose of this similarity calculation is to quantify the similarity between the business scenarios of any two historical moments and the current business scenario after comprehensively considering all these influencing factors.

[0049] In step 304 of some embodiments, the similarity scores calculated in step 303 are filtered based on a preset matching threshold (e.g., a similarity score greater than 0.95). All historical business scenario feature vectors with similarity scores higher than this threshold are filtered out and identified as target historical business scenario feature vectors, indicating that their corresponding historical operating conditions (combining factors such as time, weather, and personnel) are highly consistent with the current operating conditions. Subsequently, the historical times corresponding to these target historical business scenario feature vectors are extracted (e.g., "9:00 AM on July 5, 2024", "9:05 AM on October 28, 2025", etc.), and the set of these times is determined as the historical business scenario.

[0050] In step 305 of some embodiments, the original historical database is indexed using the historical business scenarios determined in step 304. Next, historical operational data corresponding to these specific moments (i.e., the physical operating parameters of the device itself, such as power consumption, current, temperature, etc.) are extracted. This filtered set of device operational data is ultimately determined as valid historical operational data. This dataset has eliminated interference from all irrelevant historical data related to business scenarios (such as different times, weather, personnel, activities, etc.) and will be used for modeling in subsequent steps 202 and 203.

[0051] Through steps 301 to 305 above, this application provides a complete technical solution for quantifying, matching, and filtering abstract, multi-dimensional "business scenarios." The core technical effect of this solution lies in its ability to uniformly quantify various data such as time impact, weather impact, seasonal impact, personnel impact, enterprise impact, vehicle impact, and other impacts into feature vectors. By utilizing multi-dimensional similarity calculation and matching threshold filtering, the subjective judgment of whether business scenarios are similar is transformed into an objective and quantifiable mathematical problem. This ensures that the extracted historical effective operational data possesses extremely high scenario relevance, thus providing a reliable data foundation for constructing high-precision operational baseline models and rate change models—a necessary prerequisite for achieving high-accuracy early warning.

[0052] In step 202 of some embodiments, statistical analysis is performed on the historical effective operating dataset obtained in step 201. The purpose of this analysis is to quantify the steady-state operating characteristics of the device under the specific business scenario. For example, by analyzing the distribution of historical effective power consumption data points, its core central tendency and dispersion are determined. Based on this analysis, a normal numerical range can be defined, such as a percentile-based interval or a confidence interval based on the mean and standard deviation. This range accurately describes the reasonable fluctuation range of the device's operating parameters under the current business scenario. This normal numerical range is solidified into a data structure or mathematical boundary, i.e., an operating baseline model, as a benchmark used in subsequent step 103 to determine whether the real-time absolute value deviates.

[0053] Please see Figure 4 In some embodiments, step 202 may include, but is not limited to, steps 401 to 403.

[0054] Step 401: Calculate the mean and standard deviation of the historical valid running data.

[0055] Step 402: Based on the mean and standard deviation, and in conjunction with a preset multiple, determine the upper and lower limits of the normal numerical range.

[0056] Step 403: The numerical range defined by the lower limit and the upper limit is determined as the normal numerical range and used as the running baseline model.

[0057] In step 401 of some embodiments, the historical valid running dataset determined in step 201 will be... Perform statistical calculations, where This represents the number of data points in the dataset. Specifically, it calculates the mean of this dataset. The mean This dataset characterizes the central tendency of the device's operating parameters under specific business scenarios. The standard deviation of this dataset is also calculated. The standard deviation Used to quantify operating parameters around this mean. The degree of dispersion of fluctuations.

[0058] In step 402 of some embodiments, the mean calculated in step 401 is used. with standard deviation Furthermore, a preset multiple k (e.g., k=3) is introduced to construct a dynamic statistical boundary. This preset multiple k is a configurable engineering parameter, and its value directly determines the sensitivity of the early warning model. Specifically, by using the mean... Add standard deviation The product of the product with the preset multiple k yields the upper limit of the normal numerical range. The calculation formula is as follows: At the same time, by using the mean Subtracting this product from the middle yields the lower limit of the normal numerical range. The calculation formula is as follows: .

[0059] In step 403 of some embodiments, the closed numerical range jointly defined by the lower limit and the upper limit determined in step 402 is determined as the normal numerical range under the current business scenario. This numerical range itself is constructed as the operating baseline model required in step 202, quantifying the reasonable steady-state range of the target device's operating parameters under this specific scenario, for use in subsequent real-time deviation comparison (step 103).

[0060] Through steps 401 to 403 above, this application embodiment provides a specific technical implementation for constructing a statistically significant operational baseline model. First, the mean and standard deviation are used to describe the central tendency and dispersion characteristics of historical effective operational data, and a preset multiple is used to define the normal numerical range. In this way, the operational baseline model is learned and constructed from the distribution of the data itself in an objective and quantifiable manner, so that the model can closely fit the real distribution characteristics of the data in a specific business scenario, providing a high-confidence evaluation benchmark for the static deviation judgment in the subsequent step 103.

[0061] In step 203 of some embodiments, the rate of change of the historical effective operational dataset is analyzed. First, the historical rate of change between time-series data points in the dataset is calculated, for example, by calculating the first difference between adjacent data points to obtain the change per minute. This step converts the original numerical dataset into a rate dataset. Next, statistical analysis is performed on this historical rate of change dataset to quantify the dynamic change characteristics of the device in the business scenario. This analysis determines a normal rate range, which describes the reasonable rate of change of device parameters when responding normally to business fluctuations. This normal rate range constitutes the rate change model, which will serve as the benchmark for determining whether the real-time change trend is abnormal in the subsequent step 104.

[0062] Please see Figure 5 In some embodiments, step 203 may include, but is not limited to, steps 501 to 503.

[0063] Step 501: Calculate the mean and standard deviation of the historical rate of change.

[0064] Step 502: Based on the mean and standard deviation of historical change rates, and in conjunction with a preset rate multiple, determine the upper and lower limits of the normal rate range.

[0065] Step 503: The rate range defined by the lower and upper rate limits is determined as the normal rate range and used as the rate change model.

[0066] In step 501 of some embodiments, a dataset of historical rates of change is first calculated based on historical valid operational data (i.e., the product of step 201). ,in This represents the amount of change in historical data at adjacent time points. Next, this historical rate of change dataset... Perform statistical calculations. Specifically, calculate the mean of this rate dataset. The mean This represents the average trend of changes in the operating parameters of the equipment under specific business scenarios. Simultaneously, the standard deviation of this rate dataset is calculated. The standard deviation Used to quantify the volatility of these rates of change themselves.

[0067] In step 502 of some embodiments, the average historical rate of change calculated in step 501 is used. Standard deviation of historical rate of change And introduce a preset rate multiple. This is used to construct a dynamic rate boundary. This preset rate multiple... This is a configurable parameter used to define the tolerance for the rate of change of device parameters. Specifically, it is defined by the average of historical rates of change. Add the standard deviation of the historical rate of change Compared to the preset rate multiple The product of these values ​​yields the upper limit of the normal speed range. The calculation formula is as follows: At the same time, by analyzing the average historical rate of change... Subtracting this product from the middle yields the lower limit of the normal speed range. The calculation formula is as follows: .

[0068] In step 503 of some embodiments, the closed rate range jointly defined by the lower and upper rate limits determined in step 502 is defined as the normal rate range under the current business scenario. This rate range itself is constructed as the rate change model required in step 203. This model quantifies the reasonable dynamic change range of the target device's operating parameters under this specific scenario, for use in subsequent real-time rate deviation comparison (step 104).

[0069] Through steps 501 to 503 described above, this application embodiment provides a specific technical implementation for constructing a rate change model. The dynamic characteristics (i.e., historical change rates) of historical valid operating data are used as the analysis object, and the mean and standard deviation are used to describe the central tendency and fluctuation range of these rates. By combining the normal rate range defined by a preset rate multiple, the expected limits of normal changes in equipment parameters under specific business scenarios are objectively learned from the data. The rate change model constructed by this method is another dimension orthogonal to the operating baseline model, providing a high-confidence evaluation benchmark for judging dynamic trend deviations in the subsequent step 104.

[0070] Through steps 201 to 203 described above, this embodiment of the application constructs a dynamic and context-aware early warning benchmark. Step 201, with its business scenario matching and filtering, ensures that the analyzed data is highly relevant, enabling the subsequently built model to adapt to the current actual business conditions and resolving the technical problem of model inaccuracy caused by normal business fluctuations. Steps 202 and 203, based on this relevant data, model from both static numerical and dynamic trend dimensions, jointly ensuring that the operational baseline model and rate change model upon which subsequent early warning judgments (steps 103 and 104) are based are accurate and comprehensive, laying the data model foundation for achieving highly accurate early warnings.

[0071] In step 103 of some embodiments, the real-time operating data obtained in step 101 (e.g., the instantaneous power consumption of an electrical appliance is 120kW) is compared in real time with the operating baseline model constructed in step 102 for the current business scenario (e.g., the normal power consumption range in this scenario is defined as [90kW, 110kW]). The purpose of this step is to determine whether the current absolute operating state of the equipment has deviated from its normal range under similar historical scenarios. If the real-time data point falls outside the normal range defined by the baseline model, the first deviation result is determined to be abnormal. This verification is good at capturing significant, large deviations, such as sudden component failure or instantaneous overload caused by incorrect modification of setting parameters.

[0072] In step 104 of some embodiments, a real-time rate of change is calculated based on continuous real-time operating data (e.g., power consumption over the past 5 minutes) within a preset time window. This rate characterizes the current dynamic trend of parameter change. Subsequently, this real-time rate of change is compared with a rate change model constructed in step 102 (e.g., a defined normal rate range of 5 kW per minute). If the real-time rate of change exceeds the rate range defined by the model, a second deviation result is marked as abnormal. This step is used to detect early signs of faults where the values ​​have not yet exceeded limits, but the trend of change is abnormal, such as rapid increases or fluctuations in parameters.

[0073] Please see Figure 6 In some embodiments, step 104 may include, but is not limited to, steps 601 to 602.

[0074] Step 601: The real-time rate of change is obtained by performing differential calculation on multiple data sampling points of the real-time running data within a predetermined time window.

[0075] Step 602: Determine whether the real-time rate of change is within the normal rate range defined by the lower and upper rate limits. If the real-time rate of change is outside the normal rate range, the second deviation result is determined to be abnormal. If the real-time rate of change is within the normal rate range, the second deviation result is determined to be normal.

[0076] In step 601 of some embodiments, to obtain the real-time rate of change of the target device, differential calculation is performed on real-time operating data within a predetermined time window (e.g., the past 60 seconds). This predetermined time window defines the time span of the data points used to calculate the rate. Differential calculation is a method to quantify the rate of change of time-series data. For example, by obtaining the value of the current data sampling point and the value of the previous data sampling point, calculating the difference between the two, and dividing by the time interval between the two sampling points, the real-time rate of change is obtained. This rate value characterizes the instantaneous change trend of the device's operating parameters at the current moment.

[0077] In step 602 of some embodiments, a logical judgment is performed to evaluate whether the real-time change rate obtained in step 601 is within the normal dynamic range. The judgment logic is as follows: if the real-time change rate is greater than or equal to the lower limit of the rate and less than or equal to the upper limit of the rate, that is, the real-time change rate is within the normal rate range, then the dynamic trend of the device is determined to be normal, and the second deviation result is determined to be normal. Conversely, if the real-time change rate is less than the lower limit of the rate or greater than the upper limit of the rate, that is, the real-time change rate is outside the normal rate range, then the dynamic trend of the device is determined to be out of control or abnormal, and the second deviation result is determined to be abnormal.

[0078] Through steps 601 to 602 described above, this embodiment of the application extracts key instantaneous dynamic features, namely the real-time rate of change, from the original real-time operating data using differential calculation. Then, it determines whether the real-time rate of change falls within the normal rate range defined by the rate change model. The combination of these two steps enables this method to capture early fault symptoms where the values ​​have not yet exceeded limits, but the rate of change is already abnormal, thereby improving the sensitivity and timeliness of the entire early warning scheme.

[0079] In step 105 of some embodiments, the first deviation result generated in step 103 and the second deviation result generated in step 104 are comprehensively evaluated. By analyzing the combined state of these two deviation results, it is ultimately determined whether there is a potential risk to the target device. If a risk is determined to exist, a warning message for the target device will be generated. This message may specifically manifest as an alarm notification pushed to maintenance personnel, an automatically generated pending work order, or a highlight on a monitoring screen to trigger timely preventive maintenance.

[0080] Please see Figure 7 In some embodiments, step 105 may include, but is not limited to, steps 701 to 702.

[0081] Step 701: If the first deviation result is abnormal or the second deviation result is abnormal, determine that the target device has abnormal signs.

[0082] Step 702: If it is determined that there are abnormal signs in the target device, generate and send a warning message for the target device.

[0083] In step 701 of some embodiments, the first deviation result (i.e., the static numerical deviation state generated in step 103) and the second deviation result (i.e., the dynamic rate deviation state generated in step 104) are evaluated simultaneously. The "OR" gate logic can be used for judgment, that is, as long as at least one of the two situations, the first deviation result is abnormal or the second deviation result is abnormal, it is determined that there are abnormal signs in the target device. This logic setting ensures that no abnormality in any dimension is missed, thereby maximizing the sensitivity of the warning.

[0084] In step 702 of some embodiments, if an abnormality is detected in the target device, an early warning response will be triggered. This response includes generating a target device early warning message, which is a structured data notification that may include the device identifier, the time of the abnormality, the type of abnormality (e.g., whether it is a static or dynamic deviation), and relevant real-time operational data and current business scenario data as evidence. Subsequently, this target device early warning message is sent, for example, via application push notification, email, or API call, to the operation and maintenance management platform or relevant responsible persons to initiate subsequent investigation or maintenance processes.

[0085] Please see Figure 8 This application also provides a device early warning apparatus based on business data association, which can implement the above-mentioned device early warning method based on business data association, including: The acquisition module is used to acquire real-time operating data and current business scenario data of the target device. The filtering module is used to filter out historical valid operating data that matches the current business scenario from historical data, and to build an operating baseline model and a rate change model based on the historical valid operating data; The first comparison module is used to compare the real-time running data with the running baseline model to obtain the first deviation result; The second comparison module is used to calculate the real-time change rate of the target device from the real-time running data, compare the real-time change rate with the rate change model, and obtain the second deviation result. The generation module is used to generate early warning information for the target device based on the first deviation result and the second deviation result.

[0086] The device early warning method based on business data association according to the embodiments of this application includes: acquiring real-time operating data of the target device and current business scenario data of the target device; filtering out historically valid operating data that matches the current business scenario from historical data, and constructing an operating baseline model and a rate change model based on the historically valid operating data; comparing the real-time operating data with the operating baseline model to obtain a first deviation result; calculating the real-time change rate of the target device based on the real-time operating data, comparing the real-time change rate with the rate change model to obtain a second deviation result; and generating early warning information for the target device based on the first deviation result and the second deviation result.

[0087] This application first acquires real-time operational data from the target device and incorporates current business scenario data, providing clear and realistic data input for subsequent refined, scenario-aware early warning judgments, thus forming a reliable analysis foundation. Then, based on the current business scenario data, matching historical valid operational data is selected from historical data. Based on this, an operational baseline model for static numerical comparison and a rate change model for dynamic trend comparison are constructed, overcoming the shortcomings of related technologies where early warning models are not linked to business operations, failing to distinguish between real faults and normal business fluctuations, leading to numerous false alarms when business scenarios change. Next, a first deviation result is obtained by comparing real-time operational data with the operational baseline model, and a second deviation result is obtained by comparing the calculated real-time change rate with the rate change model, achieving a multi-dimensional quantitative assessment of the device status. This multi-dimensional analysis method overcomes the shortcomings of related technologies that rely solely on single parameter absolute value monitoring and are inaccurate in predicting complex early faults. Finally, the first and second deviation results are comprehensively judged, and the final early warning information is generated based on this. The resulting early warning result comprehensively reflects the true operational health of the device under specific business scenarios. This application establishes a correlation between business data and equipment operation, and on this basis, comprehensively judges the analysis of historical data based on business filtering and the analysis of data change rate based on business correlation, thereby improving the accuracy and reliability of equipment early warning.

[0088] Reference Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the device early warning method based on business data association according to the embodiments of this application. The input / output interface 903 is used to implement information input and output; The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, Wi-Fi, Bluetooth, etc.). Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904); The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0089] This application also provides a computer program product, which includes a computer program. The processor of a computer device reads and executes the computer program, causing the computer device to perform the aforementioned device early warning method based on business data association.

[0090] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in this disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “including,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes 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 apparatuses.

[0091] It should be understood that in this disclosure, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0092] It should be understood that in the description of the embodiments of this application, "multiple" means two or more, "greater than", "less than", "exceeding" etc. are understood to exclude the number itself, and "above", "below", "within" etc. are understood to include the number itself.

[0093] In the several embodiments provided in this disclosure, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and 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 through some interfaces, indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0094] 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.

[0095] Furthermore, the functional units in the various embodiments of this disclosure 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.

[0096] 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 disclosure, 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 of the various embodiments of this disclosure. 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.

[0097] It should also be understood that the various implementation methods provided in this application can be combined arbitrarily to achieve different technical effects.

[0098] The above is a detailed description of the embodiments of this disclosure. However, this disclosure is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this disclosure. All such equivalent modifications or substitutions are included within the scope defined by the claims of this disclosure.

Claims

1. A device early warning method based on business data association, characterized in that, include: Acquire real-time operational data of the target device and current business scenario data of the target device; Filter out historical valid operating data that matches the current business scenario from historical data, and construct an operating baseline model and a rate change model based on the historical valid operating data; The real-time operating data is compared with the operating baseline model to obtain the first deviation result; The real-time change rate of the target device is calculated based on the real-time operating data, and the real-time change rate is compared with the rate change model to obtain a second deviation result; Based on the first deviation result and the second deviation result, a warning message for the target device is generated.

2. The method of claim 1, wherein, The step of filtering historical valid operational data that matches the current business scenario from historical data, and constructing an operational baseline model and a rate change model based on the historical valid operational data, includes: Based on the current business scenario data, historical business scenarios are determined from the historical data, and the historical data corresponding to the historical business scenarios are determined as the historical valid operating data; Statistical analysis is performed on the historical valid operating data to determine the normal numerical range of the historical valid operating data, and the operating baseline model is obtained. The historical rate of change is calculated based on the historical valid operating data, and statistical analysis is performed on the historical rate of change to determine the normal rate range of the historical rate of change, which serves as the rate change model.

3. The device early warning method based on business data association according to claim 2, characterized in that, The current business scenario data includes at least one of the following: time-related impact data, weather-related impact data, seasonal impact data, personnel-related impact data, enterprise-related impact data, vehicle-related impact data, and temporary event-related impact data. The step of determining historical business scenarios from the historical data based on the current business scenario data, and identifying the historical data corresponding to the historical business scenarios as the historical valid operational data, includes: The current business scenario data is combined and quantized to obtain the current business scenario feature vector; Construct corresponding historical business scenario feature vectors for the historical data at each historical moment; Calculate the multidimensional similarity between the current business scenario feature vector and the feature vectors of each of the historical business scenarios; Based on a preset matching threshold, target historical business scenario feature vectors are selected from each of the historical business scenario feature vectors, and the historical moment corresponding to the target historical business scenario feature vector is determined as the historical business scenario. Extract the historical operational data corresponding to the historical business scenarios as the historical valid operational data.

4. The device early warning method based on business data association according to claim 2, characterized in that, The step of statistically analyzing the historical valid operating data to determine the normal numerical range of the historical valid operating data and obtaining the operating baseline model includes: Calculate the mean and standard deviation of the historical valid operating data; Based on the mean and the standard deviation, and in conjunction with a preset multiple, the upper and lower limits of the normal value range are determined; wherein, the upper limit is obtained by adding the product of the standard deviation and the preset multiple to the mean, and the lower limit is obtained by subtracting the product of the standard deviation and the preset multiple from the mean; The numerical range defined by the lower limit and the upper limit is determined as the normal numerical range and used as the operating baseline model.

5. The device early warning method based on business data association according to claim 2, characterized in that, The step of calculating the historical rate of change based on the historical valid operating data, performing statistical analysis on the historical rate of change, and determining the normal rate range of the historical rate of change as the rate change model includes: Calculate the mean and standard deviation of the historical rate of change; Based on the mean and standard deviation of the historical rates of change, and in conjunction with a preset rate multiple, the upper and lower limits of the normal rate range are determined; wherein, the upper limit is obtained by adding the product of the mean and standard deviation of the historical rates of change and the preset rate multiple to the mean of the historical rates of change, and the lower limit is obtained by subtracting the product of the standard deviation and the preset rate multiple from the mean of the historical rates of change; The rate range defined by the lower rate limit and the upper rate limit is determined as the normal rate range and used as the rate change model.

6. The device early warning method based on business data association according to claim 5, characterized in that, The step of calculating the real-time change rate of the target device based on the real-time operating data, comparing the real-time change rate with the rate change model, and obtaining a second deviation result includes: The real-time rate of change is obtained by performing differential calculations on multiple data sampling points of the real-time running data within a predetermined time window; Determine whether the real-time rate of change is within the normal rate range defined by the lower limit and the upper limit. If the real-time rate of change is outside the normal rate range, then the second deviation result is determined to be abnormal; if the real-time rate of change is within the normal rate range, then the second deviation result is determined to be normal.

7. The device early warning method based on business data association according to claim 1, characterized in that, The step of generating target device early warning information based on the first deviation result and the second deviation result includes: If at least one of the first deviation result is abnormal or the second deviation result is abnormal, it is determined that the target device has abnormal signs; If the target device is found to exhibit abnormal signs, an early warning message for the target device is generated and sent.

8. A device early warning system based on business data association, characterized in that, include: The acquisition module is used to acquire the real-time operating data of the target device and the current business scenario data of the target device; The filtering module is used to filter out historical valid operating data that matches the current business scenario from historical data, and to construct an operating baseline model and a rate change model based on the historical valid operating data; The first comparison module is used to compare the real-time running data with the running baseline model to obtain a first deviation result; The second comparison module is used to calculate the real-time change rate of the target device from the real-time operating data, compare the real-time change rate with the rate change model, and obtain a second deviation result. The generation module is used to generate early warning information for the target device based on the first deviation result and the second deviation result.

9. An electronic device, characterized in that, include: The device warning method based on business data association as described in any one of claims 1 to 7 is provided in the memory and the processor.

10. A computer-readable storage medium, characterized in that, The storage medium stores a program that is executed by a processor to implement the device early warning method based on business data association as described in any one of claims 1 to 7.

Images