Industrial Internet of Things load equipment state data processing method and system

Abstract

The invention discloses an industrial Internet of Things load equipment state data processing method and system, and relates to the technical field of data processing, and the method comprises the steps: obtaining a historical power consumption sequence and a historical temperature sequence, and obtaining the current power consumption data; acquiring first temperature field change data and a plurality of first power consumption change data, and acquiring a trend correlation index of the current power consumption data; acquiring historical power consumption data, acquiring second temperature field change data and second power consumption change data within a preset time period before the historical time point, and acquiring a trend correlation index of each piece of historical power consumption data according to the plurality of pieces of second temperature field change data and the plurality of pieces of second power consumption change data; and subtracting the trend association indexes of the historical power consumption data from the trend association indexes of the current power consumption data to obtain index difference values, and if all the index difference values are greater than zero, generating an early warning strategy of the current time point according to the minimum index difference value. The method has the advantages of dynamic trend quantification, historical boundary comparison and hierarchical response.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method and system for processing status data of industrial Internet of Things (IoT) load devices. Background Technology

[0002] In real-time monitoring scenarios of IoT load devices (such as industrial motors, smart power distribution cabinets, etc.), the device status is usually characterized by power consumption and temperature data. Existing methods rely on static threshold alarm mechanisms, which trigger an alarm when the device's current power consumption or temperature exceeds a preset threshold. However, in actual operation, devices often face complex working conditions and environmental interference, and existing technologies lack effective quantification of the state before a single value exceeds the limit.

[0003] Specifically, existing technologies neglect the dynamic correlation of historical data. Devices with the same power consumption may exhibit significantly different temperature field variations due to factors such as deteriorating heat dissipation or sudden load changes, and existing models only compare numerical values ​​rather than trend patterns. Secondly, for scenarios where power consumption is continuously increasing but has not yet exceeded a threshold (such as abnormal heat dissipation systems), existing technologies cannot quantify whether the persistence and acceleration of this increase have broken historical boundaries, leading to delayed warnings. Finally, these problems result in a high false alarm rate and a significant risk of missed alarms when dealing with gradual faults and intermittent anomalies. This is especially true for power IoT devices that require long-term stable operation, as they cannot provide accurate tiered warnings when devices exhibit historically rare trend strengths but whose values ​​are not exceeded—a high-risk state—leading to wasted maintenance resources or unexpected equipment downtime. Therefore, there is an urgent need for a method that can deeply mine the correlation characteristics of historical data and quantify the degree of anomaly in the current state trend to achieve earlier and more reliable device status prediction. Summary of the Invention

[0004] In view of the technical problems described in the background art, the present invention provides a method and system for processing status data of industrial Internet of Things load devices.

[0005] A method for processing status data of an industrial Internet of Things (IoT) load device includes: acquiring the operating data of the IoT load device, and acquiring historical power consumption sequence and historical temperature sequence of the IoT load device based on the operating data, and acquiring the current power consumption data of the IoT load device at the current time point; acquiring multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point based on the historical power consumption sequence and historical temperature sequence, and acquiring trend correlation indicators of the current power consumption data based on the multiple first temperature field change data and multiple first power consumption change data; acquiring multiple historical power consumption data that are the same as the current power consumption data based on the historical power consumption sequence, acquiring the historical time point corresponding to each historical power consumption data, acquiring multiple second temperature field change data and multiple second power consumption change data within a preset time period before the historical time point corresponding to each historical power consumption data, and acquiring trend correlation indicators of each historical power consumption data based on the multiple second temperature field change data and multiple second power consumption change data; subtracting the trend correlation indicators of each historical power consumption data from the trend correlation indicators of the current power consumption data to obtain the indicator difference; if all indicator differences are greater than zero, generating an early warning strategy for the current time point based on the smallest indicator difference.

[0006] Optionally, obtaining multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point based on historical power consumption sequences and historical temperature sequences includes: obtaining multiple consecutive power consumption data within a preset time period before the current time point based on historical power consumption sequences and using them as first power consumption data to be processed; subtracting the preceding power consumption data from the subsequent power consumption data of adjacent data in the multiple first power consumption data to be processed to obtain a first power consumption difference value, and defining the first power consumption difference value as first power consumption change data; obtaining multiple consecutive temperature field data within a preset time period before the current time point based on historical temperature sequences and using them as first temperature field data to be processed; subtracting the preceding temperature field data from the subsequent temperature field data of adjacent data in the multiple first temperature field data to be processed to obtain a first temperature difference value, and defining the first temperature difference value as first temperature field change data.

[0007] Optionally, obtaining the trend correlation index of the current power consumption data based on multiple first temperature field change data and multiple first power consumption change data includes: obtaining standard power consumption data, dividing each first power consumption change data by the standard power consumption data to obtain a first power consumption trend ratio, and summing all the first power consumption trend ratios to obtain a first trend parameter; obtaining standard temperature field data, dividing each first temperature field change data by the standard temperature field data to obtain a first temperature field trend ratio, and summing all the first temperature field trend ratios to obtain a second trend parameter; obtaining the number of first power consumption change data with the same sign as the first trend parameter and recording it as a first quantity, obtaining the number of first temperature field change data with the same sign as the second trend parameter and recording it as a second quantity, and obtaining the trend correlation index of the current power consumption data based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter.

[0008] Optionally, obtaining the trend correlation index of the current power consumption data based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter includes: dividing the first quantity by the sum of the first quantity and the second quantity to obtain a first coefficient; dividing the second quantity by the sum of the first quantity and the second quantity to obtain a second coefficient; multiplying the first coefficient by the first trend parameter and taking the absolute value; multiplying the second coefficient by the second trend parameter and taking the absolute value; and adding the two absolute values ​​to obtain the trend correlation index of the current power consumption data.

[0009] Optionally, acquiring multiple second temperature field change data and multiple second power consumption change data within a preset time period before each historical time point corresponding to each historical power consumption data includes: acquiring multiple consecutive power consumption data within a preset time period before each historical time point based on the historical power consumption sequence and using them as the second data to be processed corresponding to each historical time point; subtracting the previous power consumption data from the next power consumption data of adjacent data in the multiple second data to be processed corresponding to each historical time point to obtain a second power consumption difference, and defining the second power consumption difference as the second power consumption change data; acquiring multiple consecutive temperature field data within a preset time period before each historical time point based on the historical temperature sequence and using them as the second temperature field data to be processed; subtracting the previous temperature field data from the next temperature field data of adjacent data in the multiple second temperature field data to be processed corresponding to each historical time point to obtain a second temperature difference, and defining the second temperature difference as the second temperature field change data.

[0010] Optionally, generating an early warning strategy for the current time point based on the smallest indicator difference includes: setting multiple numerical intervals in descending order, with different numerical intervals corresponding to different early warning strategies; matching numerical intervals based on the smallest indicator difference and obtaining the corresponding early warning strategy.

[0011] An industrial IoT load device status data processing system is also provided. The system includes a management platform, a sensor network platform, and an object platform connected in sequence. The management platform includes: a data acquisition module, used to acquire operating data of the IoT load device, and based on the operating data, acquire historical power consumption sequences and historical temperature sequences of the IoT load device, and acquire current power consumption data of the IoT load device at the current time point; a first historical correlation module, used to acquire multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point based on the historical power consumption sequence and historical temperature sequence, and to acquire the trend correlation of the current power consumption data based on the multiple first temperature field change data and multiple first power consumption change data. The system comprises the following modules: a first module for obtaining trend correlation indicators; a second historical correlation module for obtaining multiple historical power consumption data that are identical to the current power consumption data based on the historical power consumption sequence, obtaining the historical time point corresponding to each historical power consumption data, obtaining multiple second temperature field change data and multiple second power consumption change data within a preset time period before each historical time point, and obtaining trend correlation indicators for each historical power consumption data based on the multiple second temperature field change data and multiple second power consumption change data; and a data early warning processing module for subtracting the trend correlation indicators of each historical power consumption data from the trend correlation indicators of the current power consumption data to obtain the indicator difference. If all indicator differences are greater than zero, an early warning strategy for the current time point is generated based on the smallest indicator difference.

[0012] Optionally, the first historical association module is further configured to: obtain multiple consecutive power consumption data within a preset time period before the current time point based on the historical power consumption sequence and use them as first power consumption data to be processed; subtract the previous power consumption data from the next power consumption data of adjacent data in the multiple first power consumption data to be processed to obtain a first power consumption difference value, and define the first power consumption difference value as first power consumption change data; obtain multiple consecutive temperature field data within a preset time period before the current time point based on the historical temperature sequence and use them as first temperature field data to be processed; subtract the previous temperature field data from the next temperature field data of adjacent data in the multiple first temperature field data to be processed to obtain a first temperature difference value, and define the first temperature difference value as first temperature field change data.

[0013] Optionally, the first historical correlation module is further configured to: acquire standard power consumption data, divide each first power consumption change data by the standard power consumption data to obtain a first power consumption trend ratio, and sum all the first power consumption trend ratios to obtain a first trend parameter; acquire standard temperature field data, divide each first temperature field change data by the standard temperature field data to obtain a first temperature field trend ratio, and sum all the first temperature field trend ratios to obtain a second trend parameter. The number of first power consumption change data with the same sign as the first trend parameter is obtained and recorded as the first quantity. The number of first temperature field change data with the same sign as the second trend parameter is obtained and recorded as the second quantity. The trend correlation index of the current power consumption data is obtained based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter.

[0014] Optionally, the first historical correlation module is further configured to: divide the first quantity by the sum of the first quantity and the second quantity to obtain a first coefficient; divide the second quantity by the sum of the first quantity and the second quantity to obtain a second coefficient; multiply the first coefficient by a first trend parameter, multiply the second coefficient by a second trend parameter, and add the two products together to obtain the trend correlation index of the current power consumption data.

[0015] The beneficial effects of this invention are reflected in: In the entire industrial IoT load device status data processing method, firstly, by establishing a trend correlation index comparison mechanism between the current status and historical power consumption points, the limitations of existing numerical thresholds are overcome. This allows for the identification of high-risk states where power consumption is not exceeded but the trend of change abnormally exceeds historical extremes, reducing the false alarm rate of gradual faults. Secondly, a dynamic weighted fusion of power consumption and temperature change trends is adopted, and the changes within a short time window are normalized and continuously statistically analyzed. This enables the trend correlation index to simultaneously capture the intensity and directional stability of changes, achieving sensitive quantification of complex abnormal patterns (such as load acceleration accompanied by local temperature rise). Thirdly, a graded early warning strategy is set based on the minimum index difference. The response level is intelligently matched using the degree of deviation of similar historical events. This avoids false alarms for occasional fluctuations (such as not triggering early warnings for similar trends that already exist in the past) and allows for emergency intervention when extreme abnormal trends first appear, improving the applicability and efficiency of maintenance resources. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0017] Figure 1 This is a schematic diagram illustrating the steps of the industrial IoT load device status data processing method of the present invention; Figure 2 This is a schematic diagram of a portion of step S2 in the industrial IoT load device status data processing method of the present invention; Figure 3 This is a schematic diagram of another part of step S2 in the industrial Internet of Things load device status data processing method of the present invention; Figure 4This is a schematic diagram of a portion of step S25 in the industrial IoT load device status data processing method of the present invention; Figure 5 This is a schematic diagram of another part of step S3 in the industrial Internet of Things load device status data processing method of the present invention; Figure 6 This is a schematic diagram of another part of step S4 in the industrial Internet of Things load device status data processing method of the present invention; Figure 7 This is a partial flowchart of S1 and S2 in the industrial IoT load device status data processing method of the present invention. Figure 8 This is a partial flowchart of S1 and S3 in the industrial IoT load device status data processing method of the present invention. Figure 9 This is a partial flowchart of steps S2, S3, and S4 in the industrial IoT load device status data processing method of the present invention. Figure 10 This is a schematic diagram of the composition of the industrial Internet of Things load device status data processing system of the present invention; Figure 11 This is a schematic diagram of the optimized industrial Internet of Things (IoT) involved in this invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] like Figure 1 , Figure 7 , Figure 8 and Figure 9As shown, an industrial Internet of Things (IoT) load device status data processing method is provided. In one embodiment, the method includes: S1. Obtain the operating data of the IoT load device, and obtain the historical power consumption sequence and historical temperature sequence of the IoT load device based on the operating data, and obtain the current power consumption data of the IoT load device at the current time point; S2. Based on historical power consumption sequence and historical temperature sequence, obtain multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point, and obtain trend correlation indicators of the current power consumption data based on the multiple first temperature field change data and multiple first power consumption change data. S3. Based on the historical power consumption sequence, obtain multiple historical power consumption data that are the same as the current power consumption data, obtain the historical time point corresponding to each historical power consumption data, obtain multiple second temperature field change data and multiple second power consumption change data within a preset time period before the historical time point corresponding to each historical power consumption data, and obtain the trend correlation index of each historical power consumption data based on the multiple second temperature field change data and multiple second power consumption change data. S4. Subtract the trend correlation indicators of each historical power consumption data from the trend correlation indicators of the current power consumption data and obtain the indicator difference. If all indicator differences are greater than zero, generate the early warning strategy for the current time point based on the smallest indicator difference.

[0022] In this embodiment, it should be noted that in S1, core monitoring data is collected and extracted from the actual operating environment of the IoT load device to ensure reliable data support for subsequent trend analysis. Specifically, S1 first collects the device's operating data in real time through sensors (such as power meters or thermal imaging units) deployed on the device. This data typically includes power consumption values ​​and temperature field distribution information. Power consumption reflects the energy consumption status of the device, while the temperature field characterizes heat dissipation efficiency and thermal stability.

[0023] Subsequently, the data processing module extracts continuous power consumption and temperature record sequences from the historical database, forming historical power consumption sequences and historical temperature sequences. These sequences represent the trajectory of the device's operating status changes over a period of time, such as recording power consumption fluctuations and spatial temperature distribution changes every minute or second, providing a benchmark for historical pattern comparison.

[0024] Simultaneously, S1 also needs to capture the current power consumption data at the current point in time. This instantaneous data serves as a trigger point, aligned with historical sequences, to initiate subsequent trend correlation calculations. The entire function of S1 is to construct a data framework. Historical sequences act as a reference library, covering the device's behavior patterns under various operating conditions (such as normal start-up and shutdown, load fluctuations, or environmental interference); while current power consumption data serves as the sample to be tested, used for real-time matching and analysis. This ensures that subsequent steps (such as the trend correlation index calculation in S2) have complete input, avoiding warning failures due to missing data. It is especially important in complex scenarios where, although power consumption is not exceeded, potential faults are implied, and abnormal trends are identified by combining historical and current data.

[0025] Furthermore, in practical applications of IoT load device monitoring, taking industrial motors as an example, operating data is acquired through built-in sensors, such as real-time acquisition of the power consumption value of the motor windings and the thermal map of the casing (forming a temperature field). First, it is connected to the cloud or edge storage, and the historical power consumption sequence (continuous power value time series) and historical temperature sequence (such as the temperature change sequence of key points in the thermal field) of the past 30 days are retrieved. These sequences cover a variety of load conditions and ambient temperature changes, such as the power consumption-temperature pattern of the motor under full load, no load or vibration interference.

[0026] Next, at the current time point (such as the start of an operation cycle), current power consumption data, such as instantaneous power input values, is collected synchronously. This step implicitly relates to subsequent analysis: although the current power consumption may be at a historically common level (such as rated power), the historical sequence obtained through S1 provides a basis for comparison in S2 and S3. For example, in S3, historical points with the same power consumption will be called to calculate trend correlation indicators; and the current data serves as the entry point to ensure that the entire process is traceable to the actual device status.

[0027] In S2, by analyzing the temporal variation characteristics of the current power consumption data, the degree of trend anomaly of the device in the current state is quantified, thereby establishing a benchmark for subsequent historical comparisons. Specifically, based on the historical power consumption sequence and historical temperature sequence obtained in S1, continuous data change characteristics within a preset time window preceding the current time point are extracted. Within the preset time period (e.g., 5 minutes), all power consumption data points within this window are identified, and a series of power consumption changes, i.e., the first power consumption change data, are generated by calculating the difference between adjacent points. These changes record the direction (increase or decrease) and intensity of power consumption fluctuations in a short period of time. At the same time, the same operation is performed on the temperature field data of the same time window to obtain a series of temperature changes as the first temperature field change data. Temperature field data usually comes from the comprehensive results of multiple spatial monitoring points, so its changes reflect the overall thermal dynamic characteristics of the device (such as uneven heat dissipation or local overheating). This method of calculating the difference between adjacent points can effectively isolate the influence of numerical magnitude and focus on the continuous characteristics of the trend itself. For example, if multiple consecutive power consumption changes are all positive and the magnitude increases, it indicates that the device power consumption is in an accelerating upward phase.

[0028] Furthermore, S2 synthesizes trend correlation indicators based on the aforementioned change data. First, each first power consumption change is normalized relative to the standard power consumption to eliminate the influence of inherent differences in device characteristics. All normalization results are then summed to form a first trend parameter, which comprehensively reflects the overall trend direction and intensity accumulation of power consumption changes within a preset time period. Simultaneously, a similar normalization and summation operation is performed on the first temperature field change to generate a second trend parameter characterizing the temperature trend.

[0029] Secondly, the system counts the number of power consumption changes consistent with the first trend parameter and the number of temperature changes consistent with the second trend parameter. These two values ​​reflect the continuity and directional stability of the trend, respectively. Finally, the first and second trend parameters are weighted and synthesized (the weights are dynamically determined by the proportion of consistent signs in both trends) to generate a trend correlation index for the current power consumption data. For example, in a scenario of deteriorating heat dissipation, even if the current power consumption is not exceeded, a continuously positive temperature change with high sign consistency will significantly increase the weight of the second trend parameter, thus reinforcing abnormal temperature fluctuations in the trend correlation index. This index includes both trend strength information (reflected by the accumulated parameter value) and trend stability information (reflected by the number of consistent signs), providing a quantitative basis for S4 to determine the strength of historically rare trends.

[0030] Furthermore, taking the monitoring scenario of a smart power distribution cabinet as an example, after S1 acquires the current power consumption data (such as the power value corresponding to a certain phase current), S2 analyzes the change characteristics within a previously preset time window (such as 10 sampling periods). First, it extracts 10 consecutive power values ​​from the historical power consumption sequence for that window and calculates the difference between adjacent points to obtain 9 first power consumption change data. For example, if it is observed that the power difference is positive for 4 consecutive times and the difference range increases from 0.1kW to 0.3kW, it indicates that the current load is continuously accelerating. At the same time, it extracts 9 temperature change quantities within the same time window from the temperature field sequence: if the cooling fan of the power distribution cabinet fails, causing a local temperature rise to intensify, it is possible that the temperature difference at monitoring point 1 exceeds 0.8°C for 5 consecutive times and is positive, while the change range at monitoring point 2 is small or even negative. At this time, the first temperature field change data shows a spatially uneven distribution characteristic.

[0031] Continuing with the example above, in the index synthesis stage, assume that the calculated first trend parameter (cumulative power consumption trend strength) is positive, and 7 out of 9 power consumption changes have the same sign; the second trend parameter (cumulative temperature trend strength) is also positive, but due to conflicting temperature monitoring point data (e.g., some locations experience severe temperature rise while others remain normal), only 5 temperature changes have the same sign. In this case, the first trend parameter is assigned a weight of approximately 0.58 (7 / (7+5)), and the second trend parameter is assigned a weight of approximately 0.42 (5 / (7+5)), resulting in a weighted index for the final trend correlation. In this scenario, while the inconsistency in temperature changes weakens its contribution to the overall index, the strong continuity of the accelerating power consumption increase still significantly boosts the index value. It is worth noting that if only a single numerical threshold is relied upon, the current power consumption may be within a historically safe range, but the index generated by S2, through a dynamic weighting mechanism, has quantified the combined abnormal trend of uneven heat dissipation accompanied by accelerated load.

[0032] In S3, a trend benchmark library of historical power consumption states is constructed to provide a reference for quantifying the degree of anomaly in the current trend. First, based on the historical power consumption sequence obtained in S1, all historical power consumption points that are exactly equal to the current power consumption data are selected. These points represent the moments when the device has historically reached the same power consumption level. For each matched historical power consumption point, a preset time period (consistent with the window length in S2) prior to its corresponding time point is traced back, and continuous power consumption and temperature field data within that window are extracted.

[0033] Subsequently, the processing logic of S2 is reproduced: the difference between adjacent power consumption data within the historical time window is calculated to generate second power consumption change data, and the adjacent difference between temperature field data is calculated simultaneously to generate second temperature field change data. Essentially, this process applies the analysis method of S2 for the current time point in parallel to each historical point with the same power consumption, thereby reconstructing the trajectory of the device's preceding state changes each time the same power consumption was reached historically. In this way, S3 establishes a trend similarity library covering the same power consumption points under different times and operating conditions, ensuring that subsequent comparisons are not affected by absolute power consumption values, and revealing the deviation between the recent state change trend of the device and its historical norm at the same power consumption endpoint.

[0034] Furthermore, S3 calculates independent trend correlation indicators for each historical point with the same power consumption. Its algorithm is identical to S2, but the data source differs: First, the second power consumption change data is normalized and accumulated to generate the first trend parameter for that point; the second temperature field change data is normalized and accumulated to generate the second trend parameter. Second, the number of trend parameters with consistent signs is counted, and the trend correlation indicators for that historical point are weighted and synthesized accordingly. The historical set of these indicators constitutes a key reference dimension—if the device exhibited a strong abnormal trend during a historical event with the same power consumption (such as a sharp increase in temperature before a heat dissipation failure), its indicators will be significantly higher; conversely, the indicators will be lower under stable operating conditions. It is worth noting that S3 achieves dual stripping through retrospective analysis, stripping away both the influence of current values ​​(comparing only points with the same power consumption) and time dimension differences (uniformly focusing on trend features within a preset window), thereby transforming the detection of abnormal device status into a pure trend pattern comparison problem.

[0035] Furthermore, taking a smart power distribution cabinet as an example, assuming that the current power consumption obtained by S1 is 5.0kW, S3 will scan the historical records to find all times when the power consumption is the same as 5.0kW (e.g., 20 times in the past 30 days). For the Nth historical occurrence point (e.g., the stable state after a load switch), the window data of the previous 10 sampling periods is extracted: if a cooling fan briefly stopped before this historical point, the calculated second power consumption change data within this window may show a smooth fluctuation (because the power has stabilized at 5.0kW), but the second temperature field change data shows a continuous positive jump (e.g., the temperature rise difference at monitoring point 1 continuously exceeds 1℃). At this time, when synthesizing the trend correlation index according to the S2 logic, due to the strong positive continuity of the temperature change data, the weight of the temperature trend parameter increases significantly, generating a high trend index value. Conversely, for another power consumption point when the ambient temperature is lower, its trend index may be lower due to the smooth temperature change.

[0036] Furthermore, suppose a historical power consumption point (5.0kW) occurred during the rainy season, and dust accumulation on the radiator caused abnormally high temperature trend parameters within the window, with values ​​exceeding the normal range. However, this was not identified at the time because it did not exceed the temperature threshold. S3 provides a refined comparison benchmark for the current state by reconstructing the trend indicators of such historical events. If the current distribution cabinet is experiencing a similar temperature rise trend at 5.0kW due to blocked heat dissipation grilles, S3 will capture this historical similarity; and if the current trend indicators far exceed all historical power consumption point indicators, it will trigger a high-risk warning from S4. This mechanism is particularly suitable for identifying states where the values ​​are normal but the trends suddenly become abnormal. For example, if the current power consumption is stable at 5.0kW, and S2 detects an accelerated temperature rise and a record high indicator value, even if the absolute temperature does not exceed the limit, it can issue a maintenance signal before the heat dissipation completely fails.

[0037] In S4, by comparing the current trend correlation index with the trend index at historical power consumption points, it is identified whether the device is in an unprecedented abnormal trend state. First, for each historical power consumption point obtained in S3, the trend correlation index calculated is subtracted from the current trend correlation index generated in S2, generating a set of index differences. These differences quantify the extent to which the current trend strength exceeds the trend strength of each historical power consumption state. If any index difference is zero or negative, it indicates that the current trend strength does not exceed that historical event (a similar or stronger trend state already existed in the past), and no special warning is required. Only when all index differences are positive does it mean that the current trend strength has broken the historical record for the first time, triggering a warning response.

[0038] Secondly, after confirming the need for an alert, focus on the smallest indicator difference (i.e., the difference between the current trend indicator and the closest historical trend indicator). This difference represents the smallest relative distance between the current state and the historical norm. Since a smaller minimum difference indicates a closer abnormal state in the past, and a larger minimum difference indicates a greater degree of deviation from the past, the corresponding level of alert strategy needs to be matched according to the predefined range of the minimum difference.

[0039] Furthermore, the closer the minimum indicator difference is to zero, the more similar the current trend is to the strongest historical anomaly, in which case a lower-level warning may be issued (such as prompting operations and maintenance to pay attention); conversely, the larger the minimum indicator difference, the more the current state has far exceeded historical experience, thus triggering a higher-level warning (such as requiring immediate shutdown for inspection). This warning mode based on dynamic division of historical boundaries effectively avoids misjudgments caused by a single threshold, ensuring both sensitivity to gradual faults (such as continuous deterioration of heat dissipation) and the identification of sudden trend anomalies (such as a sudden surge in load).

[0040] Furthermore, taking industrial motor monitoring as an example, suppose S3 filters out 10 historical state points with the same power consumption as the current state, and S4 calculates the difference between the current trend correlation index and the index of these 10 historical points. If 3 of the differences are negative (indicating that the trend of these historical points is stronger), 5 differences are less than 0.5 (the current trend is slightly stronger than the historical trend), and 2 differences are between 0.5 and 1.0, then no warning is triggered because of the existence of negative differences—this indicates that although the current load has an upward trend, similar or more severe fluctuations have occurred many times in the past when the same power consumption was used.

[0041] If all differences are positive, for example, the minimum difference is 0.2 (the current trend indicator is 0.2 higher than the closest historical abnormal trend indicator) and the maximum difference is 2.0, a preset strategy will be matched based on the minimum difference of 0.2: assuming the preset range is 0-0.3 corresponding to "Level 1 Warning" (generating a maintenance prompt), 0.3-0.6 corresponding to "Level 2 Warning" (sending a real-time alarm), and greater than 0.6 corresponding to "Level 3 Warning" (forced shutdown). In this case, the minimum difference of 0.2 falls into the Level 1 range, and a maintenance work order is generated suggesting checking the cooling system. This judgment is based on the fact that although the current trend has broken through historical extremes, historical records show that a very close abnormal state (difference of 0.2) occurred in the past, suggesting that the risk is still within a controllable range.

[0042] In another scenario, if all differences are positive and the minimum difference reaches 0.8 (for example, a sharp deterioration in current heat dissipation conditions leading to an accelerated temperature rise, with a trend strength far exceeding all historical records for the same power consumption), it will match the level three warning interval and immediately send a shutdown command. This response is based on the fact that the larger the minimum difference, the more severe the disconnect between the current state and historical experience, and the more drastically the risk of equipment failure increases.

[0043] In summary, the overall industrial IoT load device status data processing method firstly establishes a trend correlation index comparison mechanism between the current status and historical power consumption points, overcoming the limitations of existing numerical thresholds. This mechanism can identify high-risk states where power consumption is within limits but the trend of change abnormally exceeds historical extremes, reducing the false alarm rate of gradual faults. Secondly, it adopts dynamic weighted fusion of power consumption and temperature change trends, and normalizes and continuously statistically analyzes the changes within a short window, enabling the trend correlation index to simultaneously capture the intensity and directional stability of changes, achieving sensitive quantification of complex abnormal patterns (such as load acceleration accompanied by local temperature rise). Thirdly, it sets a graded early warning strategy based on the minimum index difference, and intelligently matches the response level using the deviation degree of similar historical events. This avoids false alarms for occasional fluctuations (such as not triggering early warnings for similar trends that already exist in the past), and enables emergency intervention when extreme abnormal trends first appear, improving the applicability and efficiency of maintenance resources.

[0044] like Figure 2As shown, in one embodiment, S2, obtaining multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point based on historical power consumption sequence and historical temperature sequence, includes: S21. Based on the historical power consumption sequence, obtain multiple consecutive power consumption data within a preset time period before the current time point and use them as the first power consumption data to be processed. Subtract the previous power consumption data from the next power consumption data of adjacent data in the multiple first power consumption data to be processed to obtain the first power consumption difference value, and define the first power consumption difference value as the first power consumption change data. S22. Based on the historical temperature sequence, obtain multiple consecutive temperature field data within a preset time period before the current time point and use them as the first temperature field data to be processed. Subtract the previous temperature field data from the next temperature field data of adjacent data in the multiple first temperature field data to be processed to obtain the first temperature difference value, and define the first temperature difference value as the first temperature field change data.

[0045] In this embodiment, it should be noted that in S21, the power consumption change characteristics within a short window before the current time are extracted to provide basic input for trend quantification. First, based on the historical power consumption sequence obtained in S1, a preset time period (e.g., 5 minutes) before the current time point is located, and all continuously recorded power consumption data points within this window are extracted to form the first power consumption dataset to be processed.

[0046] Subsequently, sequential difference calculation is performed on two adjacent data points in the sequence, that is, the power consumption value of the next time step is subtracted from the power consumption value of the previous time step to generate a series of first power consumption difference data. These differences directly characterize the instantaneous change direction (positive value is increase, negative value is decrease) and magnitude (absolute value reflects the degree of change) of power consumption within adjacent sampling intervals, forming the first power consumption change data set.

[0047] For example, in the operation of an industrial motor, if the window contains 10 power sampling points, S21 will calculate 9 consecutive differences. If they show continuously increasing positive values ​​(such as +0.1kW, +0.2kW, +0.3kW), it implies that the load is accelerating. This change pattern is the core basis for subsequent judgment of abnormal trends.

[0048] In S22, which runs parallel to S21, the feature extraction of temperature field changes is performed. First, based on the historical temperature sequence in S1, continuous temperature field data points within the same time window as S21 are extracted (each data point typically integrates temperature readings from multiple spatial locations and takes the average value) to form the first temperature field dataset to be processed. Then, frame-by-frame differencing is performed on the temperature field data at adjacent time points, subtracting the value of the previous time point from the temperature field value at the next time point to generate the first temperature difference sequence.

[0049] like Figure 3As shown, in one embodiment, the trend correlation index for obtaining the current power consumption data based on multiple first temperature field change data and multiple first power consumption change data in S2 includes: S23. Obtain standard power consumption data, divide each first power consumption change data by the standard power consumption data to obtain the first power consumption trend ratio, and add all the first power consumption trend ratios to obtain the first trend parameter. S24. Obtain standard temperature field data, divide each first temperature field change data by the standard temperature field data to obtain the first temperature field trend ratio, and add all the first temperature field trend ratios to obtain the second trend parameter. S25. Obtain the number of first power consumption change data with the same sign as the first trend parameter and record it as the first quantity; obtain the number of first temperature field change data with the same sign as the second trend parameter and record it as the second quantity; and obtain the trend correlation index of the current power consumption data based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter.

[0050] In this embodiment, it should be noted that in S23, the interference of individual device differences on trend analysis is eliminated, and comparability is improved through normalization processing. First, preset standard power consumption data (such as device rated power or historical average power consumption; where the rated power value directly used is suitable for scenarios with stable loads; the long-term moving average of the historical power consumption sequence of the device under normal operating conditions is calculated, suitable for scenarios with frequent load fluctuations) is called. The first power consumption change data generated in S21 is divided by this standard value one by one to obtain the first power consumption trend ratio set. This ratio value converts the absolute change amount into the relative change intensity (for example, a +0.2kW change accounts for 2% in a rated 10kW device and 4% in a rated 5kW device), realizing horizontal comparability of devices of different specifications.

[0051] Subsequently, all proportional values ​​are algebraically summed to generate the first trend parameter—positive values ​​indicate an upward trend in overall power consumption within the window, while negative values ​​indicate a downward trend. The absolute value of the sum directly reflects the strength of the accumulated trend. For example, if the motor experiences a sudden increase in load while operating under light load, the accumulated trend parameter may reach +8%, while it will be close to 0% during stable operation.

[0052] In S24, the normalization logic of S23 is replicated, but applied to the temperature field dimension. First, standard temperature field data is introduced (this can be the upper limit of temperature thresholds for multiple core monitoring points in the equipment safety specifications, such as a 120°C limit for motor windings and a 90°C limit for bearings, weighted and averaged to generate a standard value; it can also calculate the average value of temperature data at various locations under historical normal operating conditions, highlighting the consistency of the heat dissipation path, such as the comprehensive value when the temperature difference between the top and bottom of the distribution cabinet does not exceed 5°C). Each first temperature field change data generated in S22 is converted into a first temperature field trend proportion (e.g., a 1°C temperature rise accounts for 10% of the standard temperature rise threshold of 10°C). The cumulative value of all proportions constitutes the second trend parameter, which effectively coordinates the thermal dynamic characteristics under different ambient temperatures.

[0053] For example, in high-temperature environments during summer, the temperature rise rate caused by the same heat dissipation failure may be lower than in winter (due to higher base temperatures in summer), but seasonal interference can be eliminated through normalization. In scenarios with poor heat dissipation in distribution cabinets, if there is a continuous local temperature rise, the second trend parameter may reach +15%, which is far higher than the normal range of ±2%, indicating obvious thermal anomaly characteristics.

[0054] In S25, a comprehensive trend correlation index is generated through multi-dimensional feature fusion. It first identifies the number of first power consumption change data points (i.e., trend continuity) that are in the same direction as the first trend parameter: if the parameter is positive and most changes are positive, it indicates that power consumption is steadily increasing; conversely, if the signs are chaotic, the trend reliability decreases. Simultaneously, the number of first temperature field change data points with the same sign as the second trend parameter is counted, reflecting the stability of the temperature trend. Based on this, the number of signs consistent with power consumption and temperature is converted into weighting coefficients, and the two trend parameters are weighted and fused accordingly—the dimension with a higher sign consistency rate receives a higher weight.

[0055] For example, if the power consumption trend parameter of a motor is +6% and 8 / 10 of the changes are in the same sign (strong continuity), and the temperature trend parameter is +5% but only 4 / 10 are in the same sign (weak continuity), then the power consumption contribution in the final index is higher (approximately 8 / 12 = 67%).

[0056] like Figure 4 As shown, in one embodiment, the trend correlation index for obtaining the current power consumption data based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter in S25 includes: S251. Divide the first quantity by the sum of the first and second quantities to obtain the first coefficient; S252. Divide the second quantity by the sum of the first and second quantities to obtain the second coefficient; S253. Multiply the first coefficient by the first trend parameter and take the absolute value, multiply the second coefficient by the second trend parameter and take the absolute value, and add the two absolute values ​​to obtain the trend correlation index of the current power consumption data.

[0057] In this embodiment, it should be noted that in S251, the quantitative calculation of weight allocation is performed, focusing on the trend confidence of the power consumption dimension. Specifically, the first quantity (the amount of change data with the same sign as the power consumption trend parameter) is divided by the sum of the first quantity and the second quantity. For example, if there are 10 power consumption change data in a certain window, of which 8 have the same sign as the first trend parameter, and 7 of the temperature change data have the same sign as the second trend parameter, then the first quantity is 8, and the sum is 8+7=15, with a first coefficient of approximately 0.53. This coefficient dynamically reflects the confidence level of the power consumption trend—the higher the coefficient, the stronger the consistency of the power consumption change direction, the more stable the trend, and the greater its contribution weight in the final index.

[0058] In S252, following the same logic as S251, a weighting coefficient is calculated for the temperature dimension. The second quantity (the number of temperature change data points conforming to the second trend parameter sign) is divided by the sum of the first and second quantities. Using the previous example data, the second quantity is 7, the sum is 15, and the second coefficient is approximately 0.47. This coefficient strengthens the weighting of the temperature field's contribution through statistical validity: if multiple monitoring points cause temperature data signs to be dispersed (e.g., some locations experience temperature rises while others experience temperature drops), the second coefficient decreases, weakening the noise data; conversely, if all monitoring points simultaneously show a sharp temperature rise (consistent signs), the second coefficient significantly increases, ensuring that temperature anomalies are fully characterized.

[0059] In S253, the absolute value of the product of the first coefficient and the first trend parameter is taken, and the absolute value of the product of the second coefficient and the second trend parameter is taken. The two absolute values ​​are then added together. This design provides dual protection: on the one hand, it avoids the offsetting effect caused by the superposition of positive and negative trends (such as a power consumption increase of +5% and a temperature decrease of -3%), ensuring that the overall strength of the abnormal state is not underestimated; on the other hand, it enhances the sensitivity of the indicators to deviations from the normal state by using absolute values.

[0060] For example, when the motor load suddenly increases, the absolute value of the product of the first trend parameter +6% and the first coefficient 0.6 is 3.6. The temperature parameter temporarily lags behind due to thermal inertia (-1%), but its weight is low (coefficient 0.4) and contributes 0.4 after taking the absolute value. The final index value is 4.0, objectively reflecting the intensity of the load change. This index has both directional sensitivity (through symbolic statistics) and intensity quantification capability (through the accumulation of absolute values).

[0061] like Figure 5As shown, in one embodiment, S3 involves acquiring multiple second temperature field change data and multiple second power consumption change data within a preset time period prior to each historical power consumption data point, including: S31. Based on the historical power consumption sequence, obtain multiple consecutive power consumption data within a preset time period before each historical time point and use them as the second data to be processed corresponding to each historical time point. Subtract the previous power consumption data from the next power consumption data of the adjacent data in the multiple second data to be processed corresponding to each historical time point to obtain the second power consumption difference, and define the second power consumption difference as the second power consumption change data. S32. Based on the historical temperature sequence, obtain multiple consecutive temperature field data within a preset time period before each historical time point and use them as the second temperature field data to be processed. Subtract the previous temperature field data from the next temperature field data of the adjacent data in the multiple second temperature field data to be processed corresponding to each historical time point to obtain the second temperature difference value, and define the second temperature difference value as the second temperature field change data.

[0062] In this embodiment, it should be noted that S31 and S32 calculate independent trend correlation indicators for each historical power consumption point. The algorithm is completely consistent with S21 and S22, but the data source is different. In S31, specifically based on each historical time point (the moment with the same power consumption value as the current value) selected by S3, continuous power consumption records within a preset time window before that point are extracted from its historical power consumption sequence (the window length is consistent with that of S21).

[0063] The window data is then subjected to the same differential processing as in S21: adjacent power consumption data points in the sequence are subtracted one by one to generate a series of second power consumption differences. These differences record the instantaneous change trajectory of the device power consumption within a short window before the occurrence of the same power consumption state in history (such as fluctuations in the growth rate or maintenance of a steady state).

[0064] In S32, the temperature field change characteristics of historical points with the same power consumption are extracted synchronously. For each historical time point locked in S3, the temperature field data sequence (average value of multiple monitoring points) within the same leading window is extracted from the historical temperature sequence, and the calculation method of S22 is completely reproduced—the second temperature difference set is generated frame by frame by difference.

[0065] like Figure 6 As shown, in one implementation, the early warning strategy for the current time point generated in S4 based on the minimum index difference includes: S41. Set multiple numerical ranges in descending order, with different numerical ranges corresponding to different early warning strategies; S42. Match the numerical range based on the smallest index difference and obtain the corresponding early warning strategy.

[0066] In this embodiment, it should be noted that in S41, the refined early warning classification strategy is based on the interval mapping of the minimum indicator difference. It predefines a set of continuous numerical intervals covering all possible differences (such as 0-0.3, 0.3-0.6, and above 0.6), and arranges them in descending order of the interval endpoints (from largest to smallest corresponding to high risk to low risk), with each interval associated with differentiated response measures.

[0067] For example, the settings could be: a range Δ ≥ 0.6 corresponds to an "immediate shutdown and maintenance" strategy, 0.3 ≤ Δ < 0.6 corresponds to "real-time alarm and diagnosis," and 0 < Δ < 0.3 corresponds to "generating a maintenance work order." A larger difference indicates that the current state deviates further from the historical safety boundary, requiring more urgent intervention. It is worth noting that the range division should be combined with the equipment failure model: for heat-sensitive equipment, the low-risk range can be compressed, while for equipment tolerant of load fluctuations, the threshold can be appropriately relaxed.

[0068] In S42, after confirming that all indicator differences are positive in S4 (the current trend strength has reached a historical high), the minimum indicator difference (i.e., the degree of deviation from the historical closest abnormal trend) is located and matched with the preset range in S41. For example, in the monitoring of the power distribution cabinet, if the minimum difference falls into the 0-0.3 range, the "work order prompt maintenance of heat dissipation" strategy is automatically invoked; if the difference exceeds the 0.6 range, the "power cut-off" command is triggered.

[0069] This mechanism effectively coordinates risk urgency with resource utilization: when the difference is close to zero (e.g., 0.2), it indicates that there are still similar historical cases in the current state, so a low-intensity warning is adopted to avoid over-response; while a significant difference (e.g., 0.8) indicates that the historical experience boundary has been exceeded, and the failure chain must be stopped immediately.

[0070] like Figure 10 As shown, an industrial IoT load device status data processing system is also provided. The system includes a management platform, a sensor network platform, and an object platform that are sequentially connected in communication. The management platform includes: The data acquisition module is used to acquire the operating data of IoT load devices, and based on the operating data, acquire the historical power consumption sequence and historical temperature sequence of IoT load devices, and acquire the current power consumption data of IoT load devices at the current time point. The first historical correlation module is used to obtain multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point based on the historical power consumption sequence and the historical temperature sequence, and to obtain the trend correlation index of the current power consumption data based on the multiple first temperature field change data and the multiple first power consumption change data. The second historical association module is used to obtain multiple historical power consumption data that are the same as the current power consumption data based on the historical power consumption sequence, obtain the historical time point corresponding to each historical power consumption data, obtain multiple second temperature field change data and multiple second power consumption change data within a preset time period before the historical time point corresponding to each historical power consumption data, and obtain the trend association index of each historical power consumption data based on the multiple second temperature field change data and multiple second power consumption change data. The data early warning processing module is used to subtract the trend correlation indicators of each historical power consumption data from the trend correlation indicators of the current power consumption data and obtain the indicator difference. If the difference of all indicators is greater than zero, an early warning strategy for the current time point is generated based on the smallest indicator difference.

[0071] In one embodiment, the first historical association module is further configured to: obtain multiple consecutive power consumption data within a preset time period before the current time point based on the historical power consumption sequence and use them as first power consumption data to be processed; subtract the previous power consumption data from the next power consumption data of adjacent data in the multiple first power consumption data to be processed to obtain a first power consumption difference value, and define the first power consumption difference value as first power consumption change data; obtain multiple consecutive temperature field data within a preset time period before the current time point based on the historical temperature sequence and use them as first temperature field data to be processed; subtract the previous temperature field data from the next temperature field data of adjacent data in the multiple first temperature field data to be processed to obtain a first temperature difference value, and define the first temperature difference value as first temperature field change data.

[0072] In one embodiment, the first historical correlation module is further configured to: acquire standard power consumption data, divide each first power consumption change data by the standard power consumption data to obtain a first power consumption trend ratio, and sum all the first power consumption trend ratios to obtain a first trend parameter; acquire standard temperature field data, divide each first temperature field change data by the standard temperature field data to obtain a first temperature field trend ratio, and sum all the first temperature field trend ratios to obtain a second trend parameter; acquire the number of first power consumption change data with the same sign as the first trend parameter and record it as a first quantity, acquire the number of first temperature field change data with the same sign as the second trend parameter and record it as a second quantity, and acquire the trend correlation index of the current power consumption data based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter.

[0073] In one implementation, the first historical correlation module is further configured to: divide the first quantity by the sum of the first quantity and the second quantity to obtain a first coefficient; divide the second quantity by the sum of the first quantity and the second quantity to obtain a second coefficient; multiply the first coefficient by a first trend parameter, multiply the second coefficient by a second trend parameter, and add the two products together to obtain a trend correlation index for the current power consumption data.

[0074] In this embodiment, it should be noted that the specific method of performing the operation in the above-mentioned industrial IoT load device status data processing system has been described in detail in the embodiments of the relevant industrial IoT load device status data processing method, and will not be elaborated here.

[0075] It should also be noted that the entire industrial IoT load device status data processing system can be applied to the optimized industrial IoT. Figure 10 This is a schematic diagram of the composition of the industrial Internet of Things (IoT) load device status data processing system of the present invention. Figure 11 This is a schematic diagram illustrating the optimized industrial Internet of Things (IIoT) involved in this invention. (See diagram below.) Figure 10 and Figure 11 As shown, the optimized Industrial Internet of Things (IIoT) includes a user platform, a service platform, a management platform, a sensor network platform, and an object platform that establish communication in sequence. The user platform is configured to provide front-end services to users; users obtain the necessary perception service information through the user platform, process the perception service information, and transform it into user perception information; users analyze the user perception information and make corresponding decisions based on their own wishes, and transform the user perception information into user control information through the corresponding information system and send it to the service platform, thereby demonstrating the user's corresponding service needs and wishes.

[0076] The physical entities of the user platform include various user terminals, such as mobile phones, computers, and dedicated terminals, which provide user services through integration with user information system software.

[0077] The service platform is configured as an API server or other server used to establish communication between the management platform and the user platform to achieve corresponding functions; the physical entity of the service platform includes various servers.

[0078] The management platform is configured to perform at least one of the following: device operation status monitoring and management, data monitoring and management, device parameter management, and lifecycle management; the management platform is the overall operation platform for the Internet of Things, which may include various management sub-platforms, with different management sub-platforms performing different management tasks; the physical entities of the management platform include various servers.

[0079] The sensor network platform is configured to perform at least one of the following functions: network management, command management, device status management, data protocol management, data parsing, data classification, data transmission monitoring, and data transmission security management. The sensor network platform provides functions such as data communication, transmission, parsing, identification, and classification, avoiding the direct aggregation of data from various object platforms onto the management platform, which would otherwise result in data redundancy and low data processing efficiency. The physical entities of the object platforms include various gateways, edge computing devices, etc.

[0080] The object platform is configured to perform specific production control, detection, measurement and other production tasks; the physical entities of the production objects include various production equipment, sensors and so on.

[0081] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0082] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0083] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for processing status data of industrial Internet of Things (IoT) load devices, characterized in that, include: Obtain the operating data of the IoT load device, and based on the operating data, obtain the historical power consumption sequence and historical temperature sequence of the IoT load device, and obtain the current power consumption data of the IoT load device at the current time point; Based on historical power consumption sequences and historical temperature sequences, obtain multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point, and obtain trend correlation indicators of the current power consumption data based on the multiple first temperature field change data and multiple first power consumption change data. Based on the historical power consumption sequence, obtain multiple historical power consumption data that are the same as the current power consumption data, obtain the historical time point corresponding to each historical power consumption data, obtain multiple second temperature field change data and multiple second power consumption change data within a preset time period before the historical time point corresponding to each historical power consumption data, and obtain the trend correlation index of each historical power consumption data based on the multiple second temperature field change data and multiple second power consumption change data. Subtract the trend correlation indicators of each historical power consumption data from the trend correlation indicators of the current power consumption data to obtain the indicator difference. If all indicator differences are greater than zero, then generate the early warning strategy for the current time point based on the smallest indicator difference.

2. The industrial IoT load device status data processing method according to claim 1, characterized in that, The process of obtaining multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point based on historical power consumption sequence and historical temperature sequence includes: Based on the historical power consumption sequence, multiple consecutive power consumption data within a preset time period before the current time point are obtained and used as the first power consumption data to be processed. The power consumption data of adjacent data in the multiple first power consumption data to be processed is subtracted from the previous power consumption data to obtain the first power consumption difference, and the first power consumption difference is defined as the first power consumption change data. Based on historical temperature sequences, multiple consecutive temperature field data within a preset time period before the current time point are obtained and used as the first temperature field data to be processed. The previous temperature field data is subtracted from the next temperature field data of adjacent data in the multiple first temperature field data to be processed to obtain the first temperature difference value, and the first temperature difference value is defined as the first temperature field change data.

3. The industrial IoT load device status data processing method according to claim 1 or 2, characterized in that, The trend correlation index for obtaining the current power consumption data based on multiple first temperature field change data and multiple first power consumption change data includes: Obtain standard power consumption data, divide each first power consumption change data by the standard power consumption data to obtain the first power consumption trend ratio, and add all the first power consumption trend ratios to obtain the first trend parameter; Obtain standard temperature field data, divide each first temperature field change data by the standard temperature field data to obtain the first temperature field trend ratio, and add all the first temperature field trend ratios to obtain the second trend parameter. The number of first power consumption change data with the same sign as the first trend parameter is obtained and recorded as the first quantity. The number of first temperature field change data with the same sign as the second trend parameter is obtained and recorded as the second quantity. The trend correlation index of the current power consumption data is obtained based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter.

4. The industrial IoT load device status data processing method according to claim 3, characterized in that, The trend correlation index for obtaining the current power consumption data based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter includes: Divide the first quantity by the sum of the first and second quantities to obtain the first coefficient; Divide the second quantity by the sum of the first and second quantities to obtain the second coefficient; Multiply the first coefficient by the first trend parameter and take the absolute value, multiply the second coefficient by the second trend parameter and take the absolute value, and add the two absolute values ​​to obtain the trend correlation index of the current power consumption data.

5. The industrial IoT load device status data processing method according to claim 1, characterized in that, The acquisition of multiple second temperature field change data and multiple second power consumption change data within a preset time period before each historical power consumption data point includes: Based on the historical power consumption sequence, multiple consecutive power consumption data within a preset time period before each historical time point are obtained and used as the second data to be processed corresponding to each historical time point. The power consumption data of the next adjacent data in the multiple second data to be processed corresponding to each historical time point is subtracted from the previous power consumption data to obtain the second power consumption difference, and the second power consumption difference is defined as the second power consumption change data. Based on historical temperature sequences, multiple consecutive temperature field data within a preset time period before each historical time point are obtained and used as the second temperature field data to be processed. The next temperature field data of adjacent data in the multiple second temperature field data to be processed corresponding to each historical time point is subtracted from the previous temperature field data to obtain the second temperature difference value, and the second temperature difference value is defined as the second temperature field change data.

6. The industrial IoT load device status data processing method according to claim 1, characterized in that, The early warning strategy for generating the current time point based on the minimum index difference includes: Multiple numerical ranges are set in descending order, with different numerical ranges corresponding to different early warning strategies; Match numerical ranges based on the smallest index difference and obtain the corresponding early warning strategy.

7. An industrial Internet of Things (IoT) load device status data processing system, characterized in that, The system includes a management platform, a sensor network platform, and an object platform that are sequentially connected in communication. The management platform includes: The data acquisition module is used to acquire the operating data of IoT load devices, and based on the operating data, acquire the historical power consumption sequence and historical temperature sequence of IoT load devices, and acquire the current power consumption data of IoT load devices at the current time point. The first historical correlation module is used to obtain multiple first temperature field change data and multiple first power consumption change data within a preset time period before the current time point based on the historical power consumption sequence and the historical temperature sequence, and to obtain the trend correlation index of the current power consumption data based on the multiple first temperature field change data and the multiple first power consumption change data. The second historical association module is used to obtain multiple historical power consumption data that are the same as the current power consumption data based on the historical power consumption sequence, obtain the historical time point corresponding to each historical power consumption data, obtain multiple second temperature field change data and multiple second power consumption change data within a preset time period before the historical time point corresponding to each historical power consumption data, and obtain the trend association index of each historical power consumption data based on the multiple second temperature field change data and multiple second power consumption change data. The data early warning processing module is used to subtract the trend correlation indicators of each historical power consumption data from the trend correlation indicators of the current power consumption data and obtain the indicator difference. If the difference of all indicators is greater than zero, an early warning strategy for the current time point is generated based on the smallest indicator difference.

8. The industrial IoT load device status data processing system according to claim 7, characterized in that, The first historical association module is also used for: Based on the historical power consumption sequence, multiple consecutive power consumption data within a preset time period before the current time point are obtained and used as the first power consumption data to be processed. The power consumption data of adjacent data in the multiple first power consumption data to be processed is subtracted from the previous power consumption data to obtain the first power consumption difference, and the first power consumption difference is defined as the first power consumption change data. Based on historical temperature sequences, multiple consecutive temperature field data within a preset time period before the current time point are obtained and used as the first temperature field data to be processed. The previous temperature field data is subtracted from the next temperature field data of adjacent data in the multiple first temperature field data to be processed to obtain the first temperature difference value, and the first temperature difference value is defined as the first temperature field change data.

9. The industrial IoT load device status data processing system according to claim 7, characterized in that, The first historical association module is also used for: Obtain standard power consumption data, divide each first power consumption change data by the standard power consumption data to obtain the first power consumption trend ratio, and add all the first power consumption trend ratios to obtain the first trend parameter; Obtain standard temperature field data, divide each first temperature field change data by the standard temperature field data to obtain the first temperature field trend ratio, and add all the first temperature field trend ratios to obtain the second trend parameter. The number of first power consumption change data with the same sign as the first trend parameter is obtained and recorded as the first quantity. The number of first temperature field change data with the same sign as the second trend parameter is obtained and recorded as the second quantity. The trend correlation index of the current power consumption data is obtained based on the first quantity, the second quantity, the first trend parameter, and the second trend parameter.

10. The industrial IoT load device status data processing system according to claim 7, characterized in that, The first historical association module is also used for: Divide the first quantity by the sum of the first and second quantities to obtain the first coefficient; Divide the second quantity by the sum of the first and second quantities to obtain the second coefficient; Multiply the first coefficient by the first trend parameter, multiply the second coefficient by the second trend parameter, and add the two products together to obtain the trend correlation index of the current power consumption data.

Images